Regulating Risks in the European Union: The Co-production of Expert and Executive Power 9781849468794, 9781509912650, 9781509912667

A growing body of EU law and regulation is preoccupied with the protection of EU citizens from health and environmental

167 71 8MB

English Pages [293] Year 2017

Table of contents :
Acknowledgements
Table of Contents
Contributors
Table of Cases
Table of Legislation
1
Regulating Risks in the European Union: The Co-production of Expert and Executive Power
I. Introduction
II. Risk Regulation and Executive Power in the EU
III. Policy-relevant Expertise in the EU, between Epistemic and Political Authority
IV. Co-production as an Analytical Lens
V. The Co-production of EU Expert and Executive Power
VI. Broadening the Debate-Rethinking EU Constitutionalism
Part 1: Theoretical Perspectives
2
Constitutions of Modernity: Science, Risk and Governable Subjects
I. Two Faces of Risk Governance
II. Risk and European Integration: A Co-productionist View
III. Delegation
IV. Harmonisation
V. Subsidiarity
VI. The Constitutional Place of Science in EU Governance
VII. Conclusion
3
Expert Executive Power, Administrative Constitutionalism and Co-production: Why They Matter
I. Introduction
II. Expert Executive Power and Administrative Constitutionalism and the EU
III. Goshawks and Expertise
IV. Conclusion
Part 2: Practices of Co-production
4
Recombinant Regulation: EU Executive Power and Expertise in Responding to Synthetic Biology
I. Introduction
II. Synthetic Biology
III. Non-legislation and the Implementation of Existing Regulation
IV. An Emerging "Problem" and the Role of Technical Working Groups
V. Regulatory Stability and the Deferral of Socio-ethical Deliberation
5
Defining the Invisible: Between Soft Norms and Hard Realities in the European Regulation of Nanotechnologies
I. Introduction
II. The Legitimacy Dimension of Co-production
III. Defining the Invisible: Walking a Tightrope
IV. The Definition"s Impact: Analysing its Two Dimensions
V. Putting the Recommendation"s Legitimacy to the Test
VI. Conclusion
6
Mixing EU Security and Public Health Expertise in the Health Threats Decision
I. Introduction
II. Reconfiguring Public Health: Globalisation and Security
III. The Health Threats Decision: From "Public Health" to "Health Security"
IV. Conclusion
7
Constituting Public Health Surveillance in Twenty-first Century Europe
I. Introduction
II. Public Health Surveillance: A Powerful International Concept
III. The Empirical, Understudied, Diversity of Surveillance
IV. Changing Justifications of Surveillance
V. Europeanisation: Health Security and State Security
VI. Conclusions: Surveying Surveillance
8
Behavioural Expertise and Regulatory Power in Europe
I. Introduction1
II. Behavioural Expertise in EU Policy-making and Regulation
III. The European "Ecosystem of Expertise": Institutional Voids
IV. Combining Epistemic and Political Authority
V. Reframing Regulation
VI. Conclusions
Part 3: Rethinking Constitutionalism: Legitimacy and Accountability
9
Expertise as Justification: The Contested Legitimation of the EU "Risk Administration"
I. Introduction
II. Shifting Sands-Administrative Legitimation Between Control and Deliberation
III. Administrative Legitimation and GMO Authorisations: The Long Tale of Herculex and Amflora
IV. Judicial Responses
V. Conclusions
10
Evolving Conceptions of Science and Legitimacy: Insights from American Administrative Law
I. Introduction
II. Early Case Law on Risk Regulation and the Breakdown of the Classical Account
III. Early Supreme Court Cases on Risk Regulation
IV. State Farm and Chevron
V. Risk Regulation in the Courts after Chevron
VI. A Retreat to the Classical Approach?
VII. Conclusion
11
Accountability and Co-production beyond Courts: The Role of the European Ombudsman
I. Introduction
II. Co-production and Accountability in EU Governance
III. The EO as a Forum for Accountability
IV. Knowledge and Co-production in EO Decisions
V. Conclusions
Bibliography
Index

Recommend Papers

Executive–Legislative (Im)balance in the European Union 9781509930005, 9781509930036, 9781509930029

Ten years after the entry into force of the Lisbon Treaty, has executive predominance in EU-related matters disappeared?

145 36 5MB Read more

The European Union 9780748632251

GBS_insertPreviewButtonPopup('ISBN:9780748632985'); The European Union is a distinctive creation. There have

99 96 3MB Read more

European Union Treaties: Treaty on European Union Treaty on the Functioning of the European Union 9781472561862, 9781849463614, 9781782253327

This is the new English edition of a Commentary on the basic European Treaties which has already been very successfully

197 83 12MB Read more

Rockefeller of New York: Executive Power in the Statehouse 9781501733819

They offer a compelling, rounded view of a controversial chief executive and a vigorous account of the ongoing, dynamic

105 81 59MB Read more

Europe Recast: A History of the European Union: A History of the European Union 9781685850746

An authoritative history of European integration from its modern origins in the 1940s to the challenges of today

130 55 3MB Read more

Federalism in the European Union 9781472566164, 9781849462426

This edited volume aims to reveal the Janus-faced character of federalism in the European Union. Federalism appears in t

205 102 3MB Read more

Product Safety in the European Union

Dr. -lng. Gabor Czitan - Dr. -lng. Attila Gutassy - Dipl. -lng. Ralf Wilde Product Safety in the European Union From the

520 38 12KB Read more

A Glossary of the European Union 9780748631186

This A-Z guide explains the key concepts and institutions linked to the European Union, along with brief biographies of

113 70 548KB Read more

Institutional Design and Voting Power in the European Union [1 ed.] 9780754694960, 9780754677543

Leading global experts in the field of politics and mathematics bring forth key insights on how voting power should be a

140 107 4MB Read more

Democracy, social resources and political power in the European Union 9781526130334

'Democracy, social resources and political power in the European Union' develops a structural constructivist t

102 18 5MB Read more

Regulating Risks in the European Union: The Co-production of Expert and Executive Power
9781849468794, 9781509912650, 9781509912667

Author / Uploaded
Maria Weimer
Anniek de Ruijter (editors)

0 0 0
Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

File loading please wait...

Citation preview

REGULATING RISKS IN THE EUROPEAN UNION A growing body of EU law and regulation is preoccupied with the protection of EU citizens from health and environmental risks. Which chemicals are safe and should be allowed on the market? How should the EU respond to public health emergencies, such as Ebola and other infectious diseases? Regulatory responses to these questions confront deep uncertainty, limited knowledge and societal contestation. In a time where the use of scientific expertise in EU policy-making is particularly contested, this book offers a timely contribution to both the a cademic and policy debate on the role of specialised expertise in EU public decision-making on risk and technology as well as on its intertwinement with executive power. It draws on insights from law, governance, political sciences, and science and technology studies, bringing together leading scholars in this field. Contributions are drawn together by a shared theoretical perspective, namely by their use of co-production as an analytical lens to study the intricate interplay between techno-scientific expertise and EU executive power. By so doing, this collection produces highly original insights into the development of the EU administrative state, as well as into the role of regulatory science in its construction. This book will be useful to scholars, practitioners, and policy-makers working on risk regulation and the role of expertise in public decision-making.

ii

Regulating Risks in the European Union The Co-production of Expert and Executive Power

Edited by

Maria Weimer and Anniek de Ruijter

OXFORD AND PORTLAND, OREGON 2017

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

Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

www.hartpub.co.uk www.bloomsbury.com Published in North America (US and Canada) by Hart Publishing c/o International Specialized Book Services 920 NE 58th Avenue, Suite 300 Portland, OR 97213-3786 USA www.isbs.com HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2017 © The editors and contributors severally 2017 The editors and contributors severally have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2017. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-84946-879-4 ePDF: 978-1-50991-266-7 ePub: 978-1-50991-264-3 Library of Congress Cataloging-in-Publication Data Names: Weimer, Maria, editor. | Ruijter, Anniek de, editor. Title: Regulating risks in the European Union : the co-production of expert and executive power / Edited by Maria Weimer and Anniek de Ruijter. Description: Portland, Oregon : Hart Publishing, 2017. | Includes bibliographical references and index. Identifiers: LCCN 2017021514 (print) | LCCN 2017022552 (ebook) | ISBN 9781509912643 (Epub) | ISBN 9781849468794 (hardback : alk. paper) Subjects: LCSH: Executive power—European Union countries. | Administrative agencies—European Union countries. | Risk management—Law and legislation—European Union countries. | Risk assessment—European Union countries. | Public health laws—European Union countries. | Environmental law—European Union countries. Classification: LCC KJE5602 (ebook) | LCC KJE5602 .R44 2017 (print) | DDC 344.2404/7—dc23 LC record available at https://lccn.loc.gov/2017021514 Typeset by Compuscript Ltd, Shannon To find out more about our authors and books visit www.hartpublishing.co.uk. Here you will find extracts, author information, details of forthcoming events and the option to sign up for our newsletters.

ACKNOWLEDGEMENTS

This book is the result of a workshop held at the premises of the Royal N etherlands Academy of Arts and Sciences (KNAW) in Amsterdam on 15 May 2015. The workshop was co-organised by Maria Weimer, Anniek de Ruijter and Maria Lee, and explored the idiom of co-production developed by Sheila Jasanoff in the context of European Union risk regulation in the field of public health and the environment protection. We would like to express our utmost gratitude to the organisations, institutions and people that enabled us to share our ideas, and to sharpen our thoughts and writing. We want to thank the KNAW for supporting the workshop. We are also grateful to ACCESS Europe, especially to Jonathan Zeitlin, who have supported the project from its very inception, not only financially, but also with advice and ideas. Furthermore, we thank our respective research groups for their help and providing us with a stimulating working environment: the Amsterdam Centre for European Law & Governance (ACELG), the Maastricht Centre for European Law (MCEL), particularly Angela Moisl of the ACELG for her important research support as well as Monica Claes, Ellen Vos and Deirdre Curtin for their encouragement. We also thank our colleagues at the Law and Justice Across Borders programme of the Law Faculty of the University of Amsterdam for providing us with invaluable comments on work in progress. Lastly, we are grateful to Maria Lee for being an important academic mentor during this project as well as an inspirational example of academic work ethic and perseverance. Finally, we express our gratitude to Danya Chaikel for her excellent language editing; and to the wonderful student-assistants, Juliane Dieroff and Lisa Perridon, who helped with all the tables and the bibliography. Last, but not least, we thank our families for co-producing this book with us by supporting our academic work and by patiently enduring all the missed evenings, weekends and general absentmindedness that is part of this work. Anniek de Ruijter and Maria Weimer Amsterdam January 2017

vi

TABLE OF CONTENTS

Acknowledgements��v Contributors�� ix Table of Cases�� xiii Table of Legislation�� xvii

1. Regulating Risks in the European Union: The Co-production of Expert and Executive Power��1 MARIA WEIMER AND ANNIEK DE RUIJTER Part 1: Theoretical Perspectives 2. Constitutions of Modernity: Science, Risk and Governable Subjects��19 SHEILA JASANOFF 3. Expert Executive Power, Administrative Constitutionalism and Co-production: Why They Matter��37 ELIZABETH FISHER Part 2: Practices of Co-production 4. Recombinant Regulation: EU Executive Power and Expertise in Responding to Synthetic Biology��59 ELEN STOKES 5. Defining the Invisible: Between Soft Norms and Hard Realities in the European Regulation of Nanotechnologies��81 TANJA EHNERT 6. Mixing EU Security and Public Health Expertise in the Health Threats Decision��101 ANNIEK DE RUIJTER 7. Constituting Public Health Surveillance in Twenty-first Century Europe��121 SCOTT L GREER 8. Behavioural Expertise and Regulatory Power in Europe��143 HOLGER STRASSHEIM

viii

Table of Contents Part 3: Rethinking Constitutionalism: Legitimacy and Accountability

9. Expertise as Justification: The Contested Legitimation of the EU ‘Risk Administration’��167 MARIA WEIMER AND GAIA PISANI 10. Evolving Conceptions of Science and Legitimacy: Insights from American Administrative Law��193 CHRISTOPHER ANDERSON 11. Accountability and Co-production beyond Courts: The Role of the European Ombudsman��217 MARIA LEE

Bibliography��241 Index��259

CONTRIBUTORS

Dr Christopher Anderson earned a JD from Harvard Law School in 2002, after which he practised law for eight years in the fields of environmental law and appellate litigation. In 2016, he was awarded a PhD in Comparative Public Law by University College London for his thesis entitled Administrative Legitimacy and Risk Regulation in the European Union and the United States. He currently practises law in Philadelphia, Pennsylvania. Anniek de Ruijter is Assistant-Professor European Law at the University of Amsterdam. She is a research fellow at the ACELG (Amsterdam Centre for European Law and Governance) at the Faculty of Law. Her research is focuses on European health law and policy, risk regulation and the interplay between security and public health. Her monograph forthcoming with Oxford University Press is entitled: The Expansion of EU Power in the Field of Human Health. Tanja Ehnert is a lawyer in the Office for Policy Matters, International Affairs and EFSA Focal Point of the German Federal Institute for Risk Assessment in Berlin. She completed a PhD thesis in the area of EU risk regulation of nanotechnologies at Maastricht University in 2015. Tanja was previously a visiting researcher at the European University Institute in Florence, a trainee at the European Commission, DG Environment, and worked in a Brussels-based international law firm in the area of food and chemicals regulation. Liz Fisher is Professor of Environmental Law at Corpus Christi College and the Faculty of Law, University of Oxford. She researches in the areas of comparative environmental law, risk regulation and administrative law. Scott L Greer is Professor of Global Public Health in the Department of Health Management and Policy and Professor of Political Science (by courtesy) at the University of Michigan. He is also Senior Expert Advisor on Health Governance to the European Observatory on Health Systems and Policies. A political scientist he has published on Ebola in the United States (in Health Economics, Policy and Law 2017), communicable disease control in Europe (Journal of Health Politics, Policy and Law 2012), and European public health in general (European Union Public Health Policies, co-edited with Paulette Kurzer, 2013). Sheila Jasanoff is Pforzheimer Professor of Science and Technology Studies at Harvard University’s John F Kennedy School of Government. Her research centers on the engagements of science and technology with law, politics and policy in

x

Contributors

modern democratic societies, with a particular focus on the role of science in cultures of public reason. Her books include The Fifth Branch (1990), Science at the Bar (1995), Designs on Nature (2005), and The Ethics of Invention (2016). Maria Lee is a professor of law at University College London. She teaches and writes widely in environmental law and tort law, including in EU risk regulation. Gaia Pisani graduated in law at LUMSA University in Rome, and holds an LLM in International Commercial Law with distinction from the University of Cardiff. She specialised in food law and biotechnologies, and focused her research on the regulation of GMOs in EU and US. She worked at UN FAO (Rome), TMC Asser Institute (The Hague), Milieu (Bruxelles) and The ONE Campaign (Bruxelles), before moving to Southern Italy, were she currently works as a legal consultant for asylum seekers and refugees in reception centers. Elen Stokes is a Professorial Research Fellow in Energy and Environmental Law at the University of Birmingham. Her research analyses regulatory responses to new and emerging technologies, such as nanotechnology and synthetic biology. Her recent work focuses on the regulation of advanced energy technologies, including hydraulic fracturing (‘fracking’) for unconventional oil and gas. Holger Strassheim is Assistant Professor at the Department for Social Sciences, Humboldt University, Berlin and Visiting Researcher at the WZB Berlin Social Science Center, Germany. His areas of research include public policy and governance, expertise and democracy, comparative public policy analysis (especially welfare, consumer protection, food safety, employment policy, energy policy), organisation studies and social regulation. Previously, he worked as Visiting Fellow at the Royal Netherlands Academy of Arts and Sciences, Amsterdam and the Program on Science, Technology and Society at Harvard Kennedy School, Cambridge (MA). He held positions as Visiting Professor at the University of Darmstadt, the University of Bremen and at the Humboldt University. He is co-speaker of the working group on Policy, Science and Technology at the German Political Science Association, co-editor of Advances in Critical Policy Book Series (Edward Elgar) and co-editor of Critical Policy Studies Journal. Currently, he is co-editing the Handbook of Behavioural Change and Public Policy (Edward Elgar, forthcoming, with Silke Beck). Maria Weimer is assistant professor in EU law and governance at the Law F aculty of the University of Amsterdam and a senior research fellow at the Amsterdam Centre for European Law and Governance (ACELG). She is also a member to the Amsterdam interdisciplinary research programme Law and Justice Across Borders. Before joining ACELG she worked as a post-doctoral researcher at the Department of International and European Law of Maastricht University (2011–2013). She holds a PhD in Law from the European University Institute in Florence. In the past Maria Weimer participated in collaborative EU projects

Contributors

xi

funded under the 7th Framework Programme, such as RECON (Reconstituting Democracy in Europe) and INPROFOOD (Towards Inclusive Research Programming for Sustainable Food Innovations). She has published extensively in the field of EU risk regulation, environmental, health and food safety law, transnational regulation, WTO, and trade and environment; as well as the role of regulatory science and specialised expertise in public regulation.

xii

TABLE OF CASES

EUROPEAN CASES CJEU Case C-142/05 Aklagaren v Mickelsson & Roos [2009] ECR I-4273��55 Case T-74/00 Artegodan v Commission ECLI:EU:T:2002:283��167 Case T-293/08 BASF v Commission ECLI:EU:T:2010:225��190 Case C-308/11 Chemische Fabrik Kreussler & Co GmbH v Sunstar Deutschland GmbH [2012] ECR 0000��92 Case C-427/12 Commission v Parliament and Council ECLI:EU:C:2014:170��222 Case T-334/07 Denka International BV v Commission [2009] ECR II-4205��52 Case C-234/02P European Ombudsman v Lamberts [2004] ECR I-2803��228 Case C-325/91 France v Commission [1993] ECR I-3283��97 Case C-303/90 France v Commission [1991] ECR I-5315��97 Case C-62/14 Gauweiler and Others v Deutscher Bundestag ECLI:EU:C:2015:400��39 Case T-294/03 Gibault v Commission [2005] ECR-SC II-635��230 Case T-240/10 Hungary vs Commission ECLI:EU:T:2013:645��181, 186–90 Case C-60/81 International Business Machines Cooperation v Commission [1981] ECR 2639��97 Case C-439/05 Land Oberosterreich v Commission of the European Communities [2007] ECR I-7141��47 Case 9/56 Meroni v High Authority [1957/8] ECR 133 ��221 Case T-201/04 Microsoft Corp v Commission of the European Communities [2007] ECR II-03601��149 Case T-326/99 Nancy Fern Olivieri v Commission [2003] ECR II-6053��97 Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609��55 Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305��52–53, 221 Case C-151/98 P Pharos v Commission [1999] ECR I-08157��185 Case T-139/07 Pioneer Hi-Bred International v Commission ECLI:EU:T:2009:307��183 Case T-164/10 Pioneer Hi-Bred International v European Commission ECLI:EU:T:2013:503�� 185–86 Case C-199/13 Polyelectrolyte Producers Group and Others v Commission ECLI:EU:C:2014:205��52 Case T-368/11 Polyelectrolyte Producers Group v Commission ECLI:EU:T:2013:53��52 Case C-331/88 R v Minister of Agriculture, Fisheries and Food and Secretary of State for Health, ex parte FEDESA [1990] ECR I-4023��42 Case C-322/88 Salvatore Grimaldi v Fonds des maladies professionnelles [1989] ECR 04407��97 Case 174/82 Sandoz BV [1983] ECR 2445��42

xiv

Table of Cases

Case T-229/04 Sweden v Commission [2007] ECR I-2437��52, 188 Case C-343/09 The Queen, on the application of Afton Chemical Limited v Secretary of State for Transport [2010] ECR I-7027��52 Case C-558/07 The Queen, on the application of SPCM SA CH Erbslöh KG Lake Chemicals and Minerals Ltd Hercules Inc v Secretary of State for the Environment, Food and Rural Affairs [2009] ECR I-5783 ��92 Case C-270/12 UK v Parliament and Council ECLI:EU:C:2014:18��47 Case C-270/12 UK v European Parliament and Council (Short Selling) ECLI:EU:C:2014:18��3, 221 European Ombudsman Case 1321/2011/LP (19 December 2013)��232 Case 1876/2011/LP (19 December 2013)��232 NATIONAL CASES United Kingdom Downs v Secretary of State for Environment Food and Rural Affairs [2008] EWHC 2666 ��47 Secretary of State for Environment, Food and Rural Affairs v Downs [2009] EWCA Civ 664��47 United States of America Accord Gas Appliance Manufacturers Association v Department of Energy, 998 F2d 1041, 1046 (DC Cir 1993)��205 Alliance for Cannabis Therapeutics v Drug Enforcement Administration, 930 F2d 936, 939 (DC Cir 1991)��208 Allied Local and Regional Manufacturers Caucus v EPA, 215 F3d 61, 71 (DC Cir 2000)��208 American Coke and Coal Chemicals Institute v EPA, 452 F3d 930, 945 (DC Cir 2006)��194, 208 American Dental Association v Martin, 984 F2d 823, 827 (7th Cir 1993) (Posner J)��199 American Petroleum Institute v OSHA, 581 F2d 493, 502–03 (5th Cir 1979)�� 196–97 American Textile Workers v Donovan, 452 US 490 (1981)��202 Animal Legal Defense Fund, Inc v Glickman, 204 F3d 229, 234–35 (DC Cir 2000)��208 Aqua Slide ‘N’ Dive Corp v Consumer Products Safety Commission, 569 F2d 831, 838 (5th Cir 1978) �� 195–96 Baltimore Gas and Electric Co v Natural Resources Defense Council, Inc, 462 US 87 (1983)��200–02, 205–06 BCCA Appeal Group v EPA, 355 F3d 817, 833–34 (5th Cir 2003)��208 Bluewater Network v EPA, 370 F3d 1, 21–22 (DC Cir 2004) ��209 Bowen v Georgetown University Hospital, 488 US 204, 208 (1988)��205 Bowman Transportation, Inc v Arkansas-Best Freight System, Inc, 419 US 281, 297–98 (1974)��197 Cactus Corner, LLC v Department of Agriculture, 450 F3d 428, 433 (9th Cir 2008)��199

Table of Cases

xv

Catawba County v EPA, 571 F3d 20, 35 (DC Cir 2009)��208 Cellular Phone Task Force v Federal Communications Commission, 205 F3d 82 (2d Cir 2000)��208 Center for Biological Diversity v EPA, 749 F3d 1079 (DC Cir 2014)��212 Chemical Manufacturers Association v EPA, 28 F3d 1259 (DC Cir 1994)��207 Chevron USA Inc v Natural Resources Defense Council, Inc, 467 US 837 (1984)��204 Coalition for Responsible Regulation, Inc v EPA, 684 F3d 102 (DC Cir 2012)��212 Competitive Enterprise Institute v National Highway Traffic Safety Administration, 956 F2d 321, 323–24 (DC Cir 1992)��205 Daubert v Merrell Dow Pharmaceuticals, Inc, 509 U.S. 579 (1993)��34 Dithiocarbamate Task Force v EPA, 98 F3d 1394, 1404 (DC Cir 1996)��209 Dry Color Manufacturers’ Association v Department of Labor, 486 F2d 98, 105–06 (3d Cir 1973)��195 Edison Electric Institute v EPA, 2 F3d 438 (DC Cir 1993)��207 Ethyl Corp v EPA, 541 F2d 1, 20–29 (DC Cir 1976) (en banc)��194, 210 FDA v Brown & Williamson Tobacco Corp, 529 US 120, 159–61 (2000)��205 Friends of the Boundary Waters Wilderness v Dombeck, 164 F3d 1115, 1127–28 (8th Cir 1999)��208 Gonzalez v Oregon, 546 US 243, 262–64 (2006)��205 Industrial Union Dep’t, AFL-CIO v American Petroleum Institute, 448 US 607 (1980) (Benzene)��197 Industrial Union Department, AFL-CIO v Hodgson, 499 F2d 467, 476 (DC Cir 1974)��194 Lands Council v McNair, 537 F3d 981, 993 (9th Cir 2008) (en banc)��208 Leather Industries v EPA, 40 F3d 392, 408 (DC Cir 1994)��196 Maier v EPA, 114 F3d 1032, 1043 (10th Cir 1997)��208 Marks v United States, 430 US 188, 193 (1977)��198 Massachusetts v EPA, 415 F3d 50 (DC Cir 2005)��210 Massachusetts v EPA, 549 US 497, 533 (2007)��31, 211 Miami-Dade County v EPA, 529 F3d 1049 (11th Cir 2008)��208, 212 Miami-Dade County v EPA, 529 F3d 1049, 1069 (11th Cir 2008)��199 Michigan v EPA, 135 S Ct 2699, 2712–14 (2015) (Thomas J concurring) ��205 Mississippi v EPA, 744 F3d 1334, 1343 (DC Cir 2013)��208 Morris v Nuclear Regulatory Commission, 598 F3d 677, 684 (10th Cir 2010)��208 Motor Vehicle Manufacturers Association v State Farm Mutual Automobile Insurance Co, 463 US 29, 52 (1983)��201 National Cable and Telecommunications Association v Brand X Internet Services, 545 US 967, 1003 (2005)��194 National Maritime Safety Association v OSHA, 649 F3d 743, 751 (DC Cir 2011) ��199 National Truck Equipment Association v National Highway Traffic Safety Administration, 711 F3d 662, 672 (6th Cir 2013)��208 National Wildlife Federation v EPA, 286 F3d 554, 560 (DC Cir 2002)��208 Natural Resources Defense Council v Jackson, 650 F3d 662, 665–66 (7th Cir 2011)��208 Natural Resources Defense Council, Inc v EPA, 84 F2d 1146 (DC Cir 1987) (en banc)��206 Natural Resources Defense Council, Inc v EPA, 902 F2d 962, 973–74 (DC Cir 1990)��202

xvi

Table of Cases

Natural Resources Defense Council, Inc v Gorsuch, 685 F2d 718 (DC Cir 1982) (RB Ginsburg J)��204 Natural Resources Defense Council, Inc v Herrington, 768 F2d 1355, 1391 (DC Cir 1985)��196 Natural Resources Defense Council, Inc v Nuclear Regulatory Commission, 685 F2d 459 (DC Cir 1982)��200 Ohio Valley Environmental Coalition v Arcoma Coal Co., 556 F3d 177, 192 (4th Cir 2009)��208 Pennsylvania Department of Environmental Protection v EPA, 429 F3d 1125, 1130 (DC Cir 2005)��208 Public Citizen Health Research Group v Department of Labor, 557 F3d 165, 183–84 (3d Cir 2009)��208 Public Citizen Health Research Group v Food and Drug Administration, 740 F2d 21, 28 (DC Cir 1984)��194 Schering Corp v Food and Drug Administration, 51 F3d 390, 399 (3d Cir 1995)��208 Sierra Club v EPA, 671 F3d 955, 963–68 (9th Cir 2012)��209 Society of the Plastics Industry, Inc v OSHA, 509 F2d 1301, 1308 (2d Cir 1975)��194, 196, 199 Synthetic Organic Chemical Manufacturers Association v Brennan, 503 F2d 1155, 1158–59 (3d Cir 1974)��194 Troy Corp v Browner, 120 F3d 277 (DC Cir 1997)��208 Upper Blackstone Water Pollution Abatement District v EPA, 609 F3d 9, 22–24 (1st Cir 2012)��208, 212 Utility Air Regulatory Group v EPA, 134 S Ct 2427 (2014)��212 Vermont Yankee Nuclear Power Corp v Natural Resources Defense Council, Inc, 435 US 519 (1978)��200 Wallach v Town of Dryden, 16 NE 3d 1188 (NY 2014)��35 Whitman v American Trucking Associations, 531 US 457, 474–76 (2001)��205

TABLE OF LEGISLATION

EU Treaties Treaty on the Functioning of the European Union (TFEU) Art 43��88 Art 114��40, 42, 88 Art 122��109 Art 153��88 Art 168��88, 110 Art 169(2)(1)��40 Art 191(2)��167 Art 192��40, 43, 88 Art 196��109 Art 214��109 Art 222��109 Art 352��40 Regulations Regulation (EU) 2015/2283 of the European Parliament and of the Council of 25 November 2015 on novel foods [2015] OJ L 327/1��91 Regulation (EU) 528/2012 of the European Parliament and of the Council of 22 May 2012 concerning the making available on the market and use of biocidal products [2012] OJ L 167/1��86, 90 Regulation (EU) 1169/2011 of the European Parliament and of the Council of 25 October 2011 on the provision of food information to consumers [2011] OJ L 304��86, 90 Regulation (EU) 182/2011 of the European Parliament and of the Council of 16 February 2011 laying down the Rules and General Principles Concerning Mechanisms for Control by Member States of the Commission’s Exercise of Implementing Powers [2011] OJ L 55/13��110 Regulation (EU) 1094/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Insurance and Occupational Pensions Authority) [2010] OJ L 331/48��232 Regulation (EC) 1223/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products [2009] OJ L 342/59��85 Regulation (EC) 1924/ of the European Parliament and of the Council of 20 December 2006 on nutrition and health claims made on foods [2006] OJ L 404��149

xviii

Table of Legislation

Regulation (EC) 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency [2006] OJ L 396/1��220 Regulation (EC) 1901/2006 of the European Parliament and of the Council of 12 December 2006 on medicinal products for paediatric use [2006] OJ L 378/1 ��230 Regulation (EC) 851/2004 of the European Parliament and of the Council of 21 April 2004 establishing a European Centre for Disease Prevention and Control [2004] OJ L 142/1��113 Regulation (EC) 261/2004 of the European Parliament and of the Council of 11 February 2004 establishing common rules on compensation and assistance to passengers in the event of denied boarding and of cancellation or long delay of flights, and repealing Regulation (EEC) No 295/91 [2004] OJ L 46/1��228 Regulation (EC) 1829/2003 of the European Parliament and of the Council of 22 September 2003 on genetically modified food and feed [2003] OJ L 268/1��179 Regulation (EC) 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety [2002] OJ L 31/1��28, 51 Regulation (EC) 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents [2001] OJ L 145/43��98, 229 Commission Implementing Regulation (EU) 365/2013 of 22 April 2013 amending Implementing Regulation (EU) No 540/2011 as regards the conditions of approval of the active substance glufosinate [2013] OJ L 111/27��176 Directives Council Directive 70/220/EEC of 20 March 1970 on the approximation of the laws of the Member States on measures to be taken against air pollution by emissions from motor vehicles [1970] OJ L 76/1 ��40 Council Directive 79/409/EEC of 2 April 1979 on the conservation of wild birds [1979] OJ L 103/1 ��40 Council Directive 91/414/EEC of 15 July 1991 concerning the placing of plant protection products on the market [1991] OJ L 230/1��47, 52 European Parliament and Council Directive 2001/18/EC of 12 March 2001 on the deliberate release into the environment of genetically modified organisms [2001] OJ L 106/1 ��67, 74 European Parliament and Council Directive 2009/41/EC of 6 May 2009 on the contained use of genetically modified micro-organisms [2009] OJ L 125/75��67, 74 European Parliament and Council Directive 2010/75/EU of 24 December 2010 on industrial emissions (integrated pollution prevention and control) [2010] OJ L 334/17��221 European Parliament and Council Directive 2011/83/EU of 25 October 2011 on consumer rights [2011] OJ L 304��149 European Parliament and Council Directive 97/7/EC of 20 May 1997 on the protection of consumers in respect of distance contracts [1997] OJ L 144��149

Table of Legislation

xix

Decisions European Parliament and Council Decision 2119/98/EC of 24 September 1998 on Setting up a Network for the Epidemiological Surveillance and Control of Communicable Diseases in the Community [1998] OJ L 268/1��113 European Parliament Decision on the Regulations and General Conditions Governing the Performance of the Ombudsman’s Duties [1994] OJ L 113/25 ��225 Council Decision 1999/468/EC of 28 June 1999 laying down the procedures for the exercise of implementing powers conferred on the Commission [1999] OJ L 184��185 Commission Decision 2000/57/EC of 22 December 1999 on the Early Warning and Response System for the Prevention and Control of Communicable Diseases under Decision No 2119/98/EC of the European Parliament and of the Council [2000] OJ L 21��103 Commission Decision 2008/721/EC of 5 August 2008 on setting up an advisory structure of Scientific Committees and experts in the field of consumer safety, public health and the environment [2008] OJ L 241/21��73 Commission Decision 2009/312/EC of 2 April 2009 amending Decision 2000/96/ EC as regards dedicated surveillance networks for communable diseases [2009] OJ L 91/27�� 113–14 Commission Decision 2010/135/EU of 2 March 2010 concerning the placing on the market, in accordance with Directive 2001/18/EC of the European Parliament and of the Council, of a potato product (Solanumtuberosum L. line EH92-527-1) genetically modified for enhanced content of the amylopectin component of starch [2010] OJ L 53/11��184 Commission Decision C (2015) 5383 final of 7 August 2015 on establishing Scientific Committees in the field of public health, consumer safety and the environment��73 European Parliament and Council Decision 1082/2013/ of 22 October 2013 on Serious Cross-Border Threats to Health and Repealing Decision No 2119/98/EC [2013] OJ L 293/1�� 102–03 Recommendations Commission Recommendation 2011/696/EU of 18 October 2011 on the definition of nanomaterial [2011] OJ L 275/38�� 86, 89, 90, 97 Commission Commission, ‘Better regulation for better results—An EU agenda’ COM (2015) 215 final��6, 67, 93 Commission, ‘Communication from the Commission to the Council on Transitional Prolongation and Extension of the Mandate of the Health Security Committee in View of a Future General Revision of the Structures Dealing with Health Threats at the EU Level’ COM (2006) 699 final�� 115–16 Commission, ‘Communication from the Commission to the Council, The European Parliament and the European Economic and Social Committee and the Committee of Regions, Pandemic (h1N1)’ COM (2009) 481 final��117 Commission, ‘Communication on Consumer Health and Food Safety’ COM (1997) 183 final��47

xx

Table of Legislation

Commission, ‘Communication on the Precautionary Principle’ COM (2000) 1 final��48, 52, 221 Commission, ‘Explanatory Note how the Comments received during the Public Consultation were taken into Account for the final SCENIHR Opinion on the Scientific Basis for the Definition of the Term Nanomaterial’ SANCO C7��95 Commission, ‘Framework for Commission Expert Groups: Horizontal Rules and Public Register’ C(2010) 7649 final��220 Commission, ‘General Report on REACH’, COM (2013) 49 final��91 Commission, ‘Health Security in the European Union and Internationally’ (Commission Staff Working Document) SEC(2009)1622 final��102, 116 Commission, ‘Impact Assessment Guidelines’ SEC(2009) 92�� 94, 98, 151, 161 Commission, ‘Impact Assessment, Defining Criteria for Identifying Endocrine Disruptors in the Context of the Implementation of the Plant Protection Products Regulation and Biocidal Products Regulation—Main Report Accompanying the Communication from the Commission to the European Parliament and the Council on Endocrine Disruptors and the Draft Commission Acts Setting out Scientific Criteria for Their Determination in the Context of the EU Legislation on Plant Protection Products and Biocidal Products’ (Staff Working Document) COM (2016) 350 final��90 Commission, ‘Joint Procurement of Vaccine against Influenza A H1N1’ (Staff Working Document) COM (2009) 1188 final��117 Commission, ‘Proposal for a Council decision concerning the placing on the market for cultivation, in accordance with Directive 2001/18/EC of the European Parliament and of the Council, of a maize product (Zea mays L., line 1507) genetically modified for resistance to certain lepidopteran pests’ COM (2013) 758 final�� 182–83 Commission, ‘Proposal for a Decision of the European Parliament and of the Council on Serious Cross-Border Threats to Health Brussels’ COM (2011) 866 final��102 Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on novel foods’ COM (2013) 894 final��91 Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on medical devices’ COM (2012) 542 final��90 Commission, ‘Proposal for a Regulation of the European Parliament and of the Council amending Directive 2001/18/EC as regards the possibility for the Member States to restrict or prohibit the cultivation of GMOs in their territory’ COM (2010) 375 final��32 Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on cosmetic products’ COM (2008) 49 final��85 Commission, ‘Questions and answers on the Commission Recommendation on the definition of nanomaterial’ MEMO/11/704��86 Commission, ‘Regulatory Aspects of Nanomaterials’ (Communication) COM (2008) 366 final��66, 85 Commission, ‘Regulatory Process for the Authorisation of Antiviral Medicines and Vaccines in the Protection against Pandemic Influenza H1N1 2009, Accompanying Document to the Communication to the Commission to the Council, the European Parliament and the European Economic and Social Committee and the Committee of Regions, Pandemic Influenza H1N12 2009’ (Commission Staff Document) COM (2009) 481-SEC(2009) 119 final��117

Table of Legislation

xxi

Commission, ‘Report on the experience of Member States with Directive 2009/41/EC of the European Parliament and of the Council of 6 May 2009 on the contained use of genetically modified micro-organisms (recast) for the period 2006–2009’ (Commission Working Document) COM (2012) 398 final��69 Commission, ‘Rules of Procedure of the Scientific Committees on Consumer Safety, Health and Environmental Risks and Emerging and Newly Identified Health Risks’ (18 December 2009)��94, 96 Commission, ‘Rules of Procedure: the Scientific Committees on Consumer Safety (SCCA), Health and Environmental Risks (SCHER), and Emerging and Newly Identified Health Risks (SCENIHR)’ (2013)��63, 73 Commission, ‘Second Regulatory Review on Nanomaterials’ COM (2012) 572 final��66, 85 Commission, ‘Standard Rules of Procedures for Committees’ 2011/C 206/06 [2011] OJ C206/11��69 Commission, ‘Statement by Commissioner Dalli on the lack of agreement in the conciliation procedure on the Novel Food Regulation’ MEMO/11/202��86 Commission, ‘Statement by EU Health Commissioner Tonio Borg, on Commission’s decision on GM Pioneer 1507’ MEMO/13/960��184 Commission, ‘Synthetic Biology Bio—Applying Engineering to Biology’ (2005)��75 Commission, ‘Synthetic Biology: A NEST Pathfinder Initiative’ (2007)��78 Commission, ‘The Operating Framework for the European Regulatory Agencies’ COM (2002) 718 final��49 Commission, ‘Towards a European Strategy for Nanotechnology’ COM (2004) 0338 final��84 Commission, ‘Towards a reinforced culture of consultation and dialogue—General principles and minimum standards for consultation of interested parties by the Commission’ COM (2002) 704 final��93 Commission, ‘Vaccination Strategies against Pandemic (H1N1)2009 Accompanying the Communication from the Commission to the European Parliament, the Council, The European Economic and Social Committee and the Committee of the Regions (Pandemic H1N1 2009)’ (Staff Working Document) COM (2009) 481-SEC(2009) 1189 final��117 Commission, ‘White Paper on Food Safety’ COM (1999) 719 final��48 Commission, ‘Working Group on the Establishment of a List of Techniques Falling under the Scope of Directive 2001/18/EC on the Deliberate Release of Genetically Modified Organisms into the Environment and Directive 90/219/EC on the Contained Use of Genetically Modified Micro-Organisms’ ENV B3/AA/ D(2008)��69 European Parliament European Parliament, ‘Legislative Resolution of 28 October 2015 on the proposal for a regulation of the European Parliament and of the Council on novel foods COM (2013) 0894’, P8_TA-PROV(2015)0380��91 European Parliament, ‘Motion for a Resolution on the Commission delegated regulation of 12 December 2013 amending Regulation (EU) No 1169/2011 of the European Parliament and of the Council on the provision of food information to consumers as regards the definition of ‘engineered nanomaterials’, P7_TA(2014)0218��90

xxii

Table of Legislation

European Parliament, ‘Position of the European Parliament adopted at first reading on 24 March 2009 with a view to the adoption of Regulation (EC) No …/2009 of the European Parliament and of the Council on cosmetic products (recast)’ EP-PE_TC1-COD(2008)0035��85 European Parliament, ‘Resolution of 12 March 2013 on the Special Report of the European Ombudsman concerning his inquiry into complaint 2591/2010/GG against the European Commission’, P7_TA(2013)0062��226 European Parliament, ‘Resolution of 24 April 2009 on regulatory aspects of nanomaterials’, P6_TA(2009)03287��85 European Ombudsman Decision in complaint 1047/2013 against the European Commission��230 Decision in complaint 1301/2013 against the ECHA��230 Decision in complaint 1339/2012 against the ECB��229, 236 Decision in complaint 1581/2013 against the Commission (passenger rights)�� 228–30 Decision in complaint 1682/2010 against the European Commission (expert groups)��220, 226, 233 Decision in complaint 1826/2010 against the ECHA��230 Decision in complaint 1875/2011 against the EBA��232 Decision in complaint 1877/2010 against the EMA��230 Decision in complaint 1892/2012 against the European Commission (renewables) ��228, 230 Decision in complaint 1966/2011 against the European Banking Authority (EBA)��232 Decision in complaint 203/2013 against the European Commission (EGE)��234 Decision in complaint 2202/2012 against the European Commission (Ryanair)��230 Decision in complaint 2575/2009 again the EMA ��230 Decision in complaint 297/2013 against the European Commission (ad hoc ethical committee)��232, 235–36 Decision in complaint 346/2013 against EFSA (conflict of interests)��235 Decision in complaint 364/2013 against the European Medical Agency (EMA) (migraine medication)��229 Decision in complaint 51/2011 against the EASA��230 Decision in complaint 622/2012 against EFSA (Test Biotech)��222 Decision in complaint 693/2011 against the EMA��230 Decision in complaints 1874/2011 and 1877/2011 against the European Insurance and Occupational Pensions Authority (EIOPA)��230, 232–34 Decision in OI/11/2011 concerning the European Environment Agency��225 Decision in OI/12/2011 concerning the European Monitoring Centre for Drugs and Drug Addiction ��226 Decision in OI/12/2012 concerning the ECHA��225 Decision in OI/9/2011 concerning the EMA��225 Decision of the European Ombudsman adopting implementing provisions (2002, as amended 2008)��225 Draft Recommendation in complaint 2186/2012 against the ECHA (animal testing)��237 Draft Recommendation in Complaint 2558/2009 against the European Commission (non-human primates)��238 Draft Recommendation of the EO in her Inquiry into Complaint 726/2012/FOR against the EASA��236

Table of Legislation

xxiii

Draft Recommendations in complaint 1151/2008 against the European Commission (biofuels)�� 224, 233, 235–36, 238 Draft Recommendations in complaint 1171/2013 against the European Aviation Safety Agency (EASA) (flight times)��230, 236 Draft Recommendations in complaint 775/2010 against EFSA (revolving doors)�� 235–36 Special Report in 2591/2010 against the European Commission (Vienna airport)��226 Special Report in OI/5/2012 concerning Frontex (being considered by joint committee, 2014/2215(INI) ��226 Economic and Social Committee Economic and Social Committee, ‘Communication from the Commission—Towards a strategic vision of life sciences and biotechnology: consultation document’ (Opinion) [2002] OJ C94/23��77 European Food Safety Authority (EFSA) EFSA, ‘Clarifications of the Scientific Panel on Genetically Modified Organisms following a request from the Commission related to the opinions on insect resistant genetically modified Bt11 (Reference C/F/96/05.10) and 15072 (Reference C/ES/01/01) maize’ (2006) EFSA Journal 4(11):1561��177 EFSA, ‘Opinion of the Scientific Panel on Genetically Modified Organisms on an application (Reference EFSA-GMO-UK-2005-14) for the placing on the market of genetically modified potato EH92-527-1 with altered starch composition, for production of starch and food/feed uses, under Regulation (EC) No 1829/2003 from BASF Plant Science; Question No EFSA-Q-2005-070’ (2006) EFSA Journal 4(2):324��179 EFSA, ‘Opinion of the Scientific Panel on Genetically Modified Organism on a request from the Commission related to the notification (Reference C/ES/01/01) for the placing on the market of insect-tolerant genetically modified maize 1507 for import, feed and industrial processing and cultivation, under Part C of Directive 2001/18/EC from Pioneer Hi-Bred International/Mycogen Seeds, Question No EFSA-Q-2004-072’ (2005) EFSA Journal 3(3):181��176 EFSA, ‘Opinion of the Scientific Panel on Genetically Modified Organisms on the use of antibiotic resistance genes as marker genes in genetically modified plants; Question No EFSA-Q-2003-109’ (2004) EFSA Journal 2(4):48��179 EFSA, ‘Scientific Opinion of the GMO and BIOHAZ Panels on the “Use of antibiotic resistance genes as marker genes in genetically modified plants” and the Scientific Opinion of the GMO Panel on “Consequences of the opinion on the use of antibiotic resistance genes as marker genes in genetically modified plants on previous EFSA assessments of individual GM plants”; EFSA-Q-2009-00589 and EFSA-Q-200900593’ (2009) EFSA Journal 7(6):1108��180 EFSA, ‘Scientific Opinion of the Panel on Genetically Modified Organisms on a request from the European Commission to review scientific studies related to the impact on the environment of the cultivation of maize Bt11 and 1507, Question No EFSA-Q-2008-679’ (2008) EFSA Journal 6(10):851��177 EFSA, ‘Scientific Opinion on the Potential Risks Arising from Nanoscience and Nanotechnologies on Food and Feed Safety’ (2009) EFSA Journal 7(3):958��81

xxiv

Table of Legislation

EFSA, ‘Scientific Opinion supplementing the conclusions of the environmental risk assessment and risk management recommendations on the genetically modified insect resistant maize 1507 for cultivation’ (2012) EFSA Journal 10(11):2934��178 EFSA, ‘Scientific Opinion updating the evaluation of the environmental risk assessment and risk management recommendations on insect resistant genetically modified maize 1507 for cultivation’ (2011) EFSA Journal 9(11):2429��178 EFSA, ‘Scientific Opinion updating the risk assessment conclusions and risk management recommendations on the genetically modified insect resistant maize 1507’ (2012) EFSA Journal 10(10):2933��178 EFSA, ‘Statement of the Scientific Panel on Genetically Modified Organisms on the safe use of the nptII antibiotic resistance marker gene in genetically modified plant’ (2007) EFSA Journal 5(4):742��180 European Medicine Agency European Medicine Agency, ‘Presence of the antibiotic resistance marker gene nptII in GM plants for food and feed uses’, EMEA/CVMP/56937/2007—Final��179 Scientific Committee on Health and Environmental Risks (SCHER) SCHER, SCENIHR and SCCS, ‘Mandate for a Joint Scientific Opinion on Synthetic Biology’ (26 February 2014)��72 SCHER, SCENIHR and SCCS, ‘Opinion on Synthetic Biology I—Definition’ (2014)��63 SCHER, SCENIHR and SCCS, ‘Opinion on Synthetic Biology II: Risk Assessment Methodologies and Safety Aspects’ (2015) ��76 SCHER, SCENIHR and SCCS, ‘Results of Public Consultation on the Scientific Committees’ Preliminary Opinion on Synthetic Biology I—Definition’ (2014)��74 Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR) SCENIHR, ‘Risk Assessment of Products of Nanotechnologies’ (19 January 2009)��81 US Legislation 1946 US Administrative Procedure Act, 5 USC Chapter 5��27 Emission standards for new motor vehicles or new motor vehicle engines, 42 USC § 7521��210 Hazardous air pollutants, 42 USC § 7412 ��206 Pandemic and All-Hazards Preparedness Act (PAPHA), Pub L No 109-417��136 French Legislation Décret n° 2012-232 du 17 février 2012 relatif à la déclaration annuelle des substances à l‘état nanoparticulaire pris en application de l‘article L. 523-4 du code de l‘environnement ��88 Danish Legislation Ministry of Environment and Food of Denmark, ‘Statutory order on a register of mixtures and articles that contain nanomaterials as well as the requirement for producers and importers to report to the register’ no 644 (13 June 2014)��89

Table of Legislation

xxv

WTO WTO Panel Report, European Communities—Measures Affecting the Approval and Marketing of Biotech Products, WT/DS291 ��183 WHO WHO, International Health Regulations 2005 2nd edn (Geneva, WHO Press, 2008)�� 106, 124, 131–32, 135

xxvi

1 Regulating Risks in the European Union: The Co-production of Expert and Executive Power MARIA WEIMER AND ANNIEK DE RUIJTER

I. Introduction This edited volume has come to completion in the end of 2016—the year in which Donald Trump has been elected as the 45th President of the United States and the citizens of the United Kingdom voted to leave the European Union. Following these events the Oxford Dictionary declared the word ‘post-truth’ to be the international word of the year. Post-truth denotes ‘circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief ’,1 and refers to a general perception that we live in a time where ‘the truth has become so devalued that what was once the gold standard of political debate is a worthless currency.’2 In the public debate, experts and expertise have come under attack. One of the most striking remarks in this regard, in the pre-referendum Brexit public debate, was made by Michael Gove, the conservative politician and Brexit campaigner, who stated: ‘I think people in this country have had enough of experts.’3 Gove responded to arguments made by economic experts that the UK leaving the EU would have negative economic consequences. His remark, unwittingly, hints at what in this volume we refer to as co-production, the idea that expertise— generally referring to an authoritative claim to know better than others—when it is used in the context of law or policymaking, is not apolitical, but entangled with 1 N Midgley, ‘Word of the Year 2016 is …’ (English Oxford living dictionaries) https://en. oxforddictionaries.com/word-of-the-year/word-of-the-year-2016. 2 ibid, citing a newspaper commentator. 3 M Deacon, ‘EU Referendum: Who Needs Experts When We’ve Got Michael Gove?’ The Telegraph (London, 6 June 2016) available at www.telegraph.co.uk/news/2016/06/06/eu-referendum-who-needsexperts-when-weve-got-michael-gove/.

2

Maria Weimer and Anniek De Ruijter

the exercise of political power.4 Hence, rejecting expert views in favour of the EU in the pre-referendum Brexit debate was essentially about rejecting EU political rule including its elite use of expertise as an instrument of (political) authority; as ‘power over’ people.5 This collective volume focuses on a specific form of co-production, namely the co-production of EU expert and executive power. In this way, we aim to contribute to the discussion about the legitimacy of the exercise of EU executive power in an ever-growing policy field, namely in EU regulation of health and environmental risks.6 In other words, we focus on the relationship between EU executive power and regulatory science in decision-making on risk and technology. The term regulatory science, as coined by Sheila Jasanoff,7 refers to the particular use of scientific advice in regulation, and should be differentiated from broader notions of research science.8 Furthermore, we are not putting forward a particular ontological agenda regarding science or knowledge in general—rather, we focus on the political use of the ‘claim to know’ in the context of EU risk regulation. Co-production has been an important guiding theme for all contributors to this edited volume. As such we are heavily indebted to Sheila Jasanoff ’s pioneering work on law, science and technology, which is why we have given her thoughts on the relevance of studying co-production in the EU a prominent place in this volume.9 And yet this volume goes beyond paying a tribute to a remarkable scholar. It aims, first, to gather a critical mass of EU scholars who are able to approach the study of EU risk regulation with the same breathtaking birds-eye view as Jasanoff. And second, the ultimate goal is to explicitly re-connect the seemingly narrow discussions about the role of expertise in EU risk regulation with the broader conversation about EU executive power and its legitimacy. In this chapter we introduce the general pinpoints of this conversation. First, we revisit the relationship between risk, regulation and executive power in the EU. S econd we focus on policy-relevant expertise in the EU, its scope and relevance. Third, we link the previous two sections by introducing the notion of co-production, and fourth we present our findings on the way in which expert and executive power are being co-produced in EU risk regulation. Finally, we

4 We do not claim that the experts Gove was referring to were EU experts, but his remark did not distinguish independently produced expert claims from those produced within EU institutions. 5 S Quack, ‘Expertise and Authority in Transnational Governance’ in R Cotterell and M Del Mar (eds), Authority in Transnational Legal Theory (Cheltenham, Edward Elgar, 2016) with further references. 6 A de Ruijter, The Expansion of EU Power in the Field of Human Health (Oxford, Oxford University Press, 2017 forthcoming); M Weimer, Risk Regulation in the Internal Market and its Limits—Lessons from Agricultural Biotechnology (Oxford, Oxford University Press, 2017 forthcoming). 7 See Jasanoff, ch 2 in this volume. 8 See further on the differences between regulatory science and research science, S Jasanoff, Science at the Bar (Cambridge MA, Harvard University Press, 1995) and S Jasanoff, ‘Quality Control and Peer Review in Advisory Science’ in J Lentsch and P Weingart (eds), The Politics of Scientific Advice. Institutional Design for Quality Assurance (Cambridge, Cambridge University Press, 2011) 19–35. 9 See Jasanoff, ch 2 in this volume.

Regulating Risks in the EU

3

c onclude by outlining a future research agenda that re-connects the study of EU risk regulation, expertise and executive power with the broader debates on the future of the EU constitutional order.

II. Risk Regulation and Executive Power in the EU A fundamental principle of Western democratic societies is that essential political questions need to be decided by democratic constituencies represented through parliaments. In late-modern risk societies10 of the twenty-first century such essential political questions often relate to the distribution of economic, environmental and social risks; in other words to the distribution of the negative effects (the ‘bads’ rather than the ‘goods’) of globalisation, economic interdependence and technological progress. Yet, it is precisely in the face of these complex questions that parliaments and democratically elected politicians today seem to be losing their decision-making authority. Instead, technocrats and experts are increasingly called upon to solve pressing societal problems. For instance, the management of the economic and financial risks in post-crisis Europe has taken place through the empowerment of non-majoritarian institutions such as the European Central Bank and new EU regulatory agencies.11 Such massive delegation of powers is legitimated by reference to the complexity of the societal questions at stake, as well as the need for specific professional and technical expertise to resolve them.12 Part and parcel of these developments is the rise of risk regulation. The latter is here understood as the body of EU rules and practices dedicated to the assessment, management and communication of potential risks arising from technological and economic developments. These rules and practices interfere with market and social processes in order to control and avert possible adverse consequences to publicly protected interests. Health and environmental protection is a key policy field in which the rise of risk regulation in the EU can be observed.13 Protecting EU citizens from health and environmental risks has been one of the main arguments, together with the need to ensure the functioning of the EU internal market,

10

U Beck, Risk Society: Towards a New Modernity (London, SAGE, 1992). See eg C Joerges and M Weimer, ‘A Crisis of Executive Managerialism in the European Unon— No Alternative?’ in J Scott, G De Burca and C Kilpatrick (eds), Critical Legal Perspectives on Global Governance (Oxford, Hart Publishing, 2013) 295. 12 See eg Case C-270/12 Short Selling at para 105. 13 G Majone, ‘The European Community between Social Policy and Social Regulation’ (1993) 31 Journal of Common Market Studies 153; G Majone, ‘The Credibility Crisis of Community Regulation’ (2000) 38 Journal of Common Market Studies 273; E Vos, ‘EU Food Safety Regulation in the Aftermath of the BSE Crisis’ (2000) 23 Journal of Consumer Policy 227; M van Asselt, M Everson and E Vos, Trade, Health and the Environment: The European Union Put to the Test (London/New York, Routledge/ Earthscan, 2013). 11

4

Maria Weimer and Anniek De Ruijter

used to justify the expansion of the EU regulatory state.14 Moreover, the legitimacy of massive re-regulation of the internal market at EU level relied since the very beginning on the claim to epistemic (output) legitimacy. In other words, it relied on the promise of scientifically sound and non-majoritarian expert-based regulation to tackle transboundary health and environmental problems so as to increase the welfare of EU citizens; and thereby to compensate for the shortcomings of a European welfare state and (input) democracy.15 Hence, as Hood et al point out: [T]he two ideas of ‘risk society’ and ‘regulatory state’ could, indeed, be linked insofar as risk and safety is often held to be one of the major drivers of contemporary regulatory growth, for example in the development of EU regulation.16

Even more importantly, the expansion of the EU risk regulatory state has gone hand in hand with an empowerment of the EU executive,17 as the contributions to this volume vividly illustrate.18 The assessment and management of risks has been largely delegated to the executive institutions in the EU, particularly the European Commission19—a process intensified in the course of the 1990s through the rise of comitology committees and quasi-regulatory EU agencies.20 While the Commission was originally conceived of as a technocratic body, it has gradually become ‘normalised’ as the EU’s political executive.21 As such, it tends to outsource specialised tasks to outside bodies and an elaborate network of agencies.

14 eg Art 114 (3) TFEU. G Majone, The Rise of Statutory Regulation in Europe (London, Routledge, 1996); G Majone, ‘The Regulatory State and Its Legitimacy Problems’ (1999) 22 West European Politics 1. 15 Initiated by the Commission’s Single Market programme and the transfer to qualified majority voting in EU legislation in the Single European Act. J Weiler, ‘The Transformation of Europe’ (1991) 100 Yale Law Journal 2403. 16 C Hood, H Rothstein and R Baldwin, The Government of Risk: Understanding Risk Regulation Regimes (Oxford, Oxford University Press, 2001) 4. 17 According to Curtin, the executive refers to individuals, ministers and civil servants that ‘control, from day to day, the state’s instruments of coercion, wealth and information.’ D Curtin, Executive Power of the European Union, Law, Practices and the Living Constitution (Oxford, Oxford University Press, 2009) 19. In this contribution we understand the EU executive as encompassing both the political and the administrative executive, but see both as being inherently interlocked, see also Curtin ibid at 66. We focus on the executive functions of the Commission and its satellite institutions while neglecting the executive role of the Council due to the special relevance of the former in the field of EU regulation of public health and environmental risks. 18 See Weimer and Pisani, ch 9, Stokes, ch 4, Ehnert, ch 5, Greer, ch 7, De Ruijter, ch 6, and Lee, ch 11 in this volume. 19 H Hofmann, ‘A European Regulatory Union—The Role of Agencies and Standards’ in P Koutrakos and J Snell (eds), Research Handbook on the EU’s Internal Market (Cheltenham, Edgar Elgar, 2016); M Egeberg, M Martens and J Trondal, ‘The EU’s Subordinated Agency Administration and the Rise of Executive Power at European Level’ in M Bauer and J Trondal (eds), The Palgrave Handbook of the European Administrative System (Basingstoke, Palgrave Macmillan, 2015). 20 E Vos, ‘The Role of Comitology in European Governance’ in D Curtain and R Wessel (eds), Good Governance and the European Union: Reflections on Concepts, Institutions and Substance (Antwerp, Intersentia, 2005); D Curtin, ‘Delegation to EU Non-Majoritan Agencies and Emerging Practices of Public Accountability’ in D Geradin and N Petit, Regulation through Agencies in the EU: A New Paradigm of European Governance (Cheltenham, Edward Elgar, 2005); M Busuioc, European Agencies: Law and Practices of Accountability (Ofxord, Oxford University Press, 2013). 21 Curtin, above n 17, 91–100.

Regulating Risks in the EU

5

These ‘satellite’ institutions and their actors—policy advisors, evidence gatherers, risk assessors, deliberation networks—play a crucial role in the adoption of risk regulation.22 It is in this underbelly of EU regulation that the distinction between the executive and expertise becomes blurred,23 and where we can find the ‘expert executive’ amalgam exercising power, albeit not always in a transparent way.24

III. Policy-relevant Expertise in the EU, between Epistemic and Political Authority The term expertise generally refers to ‘a widely acknowledged source of reliable knowledge, skill, or technique that is accorded status and authority by the peers of a person who holds it and accepted by members of the larger public.’25 As outlined earlier, in this edited volume we focus on a particular type of expertise, namely regulatory or policy-relevant expertise, which is specialised (often scientific) knowledge that is used for regulatory purposes. We consider policy-relevant expertise to be a special type of ‘epistemic authority’.26 Claims to epistemic authority are raised by virtue of, on the one hand, possessing specialised knowledge (ie expertise) that is different from other more commonly shared forms of knowledge and, on the other hand, being a reliable source of information (ie possessing credibility).27 Policy-relevant expertise, however, stands out from other types of epistemic authorities (eg research science, epistemic communities, non-governmental organisations (NGOs)) in that it is being assigned authority by political institutions.28 In that way, it gains political authority as its interpretations and factual assessments are validated by political decisions. Hence, as Holger Strassheim discusses in this volume,29 policy-relevant expertise is conceptualised as a nexus between political and epistemic authority. It is imbued with political authority. An important consequence of the rise of the EU risk regulatory state as described in the previous section has been an u nprecedented

22

Busuioc, above n 20. Vos above n 20; E Vos, ‘The Rise of the Committees’ (1997) 3 European Law Journal 210. 24 Curtin, above n 20; D Curtin and M Egeberg, ‘Tradition and Innovation: Europe’s Accumulated Executive Order’ (2008) 31 West European Politics; Egeberg, Martens and Trondal, above n 19; A Gornitzka and C Holst, ‘The Expert-Executive Nexus in the EU: An Introduction’ (2015) 3 Politics and Governance 1. 25 F Fisher, Democracy & Expertise. Reorienting Policy Inquiry (Oxford, Oxford University Press, 2009) 17. 26 Hence we use the term ‘regulatory science’ as the tool for providing policy-relevant expertise. 27 R Guess, History and Illusion in Politics (Cambridge, Cambridge University Press, 2001) 38; Quack, above n 5; M Zürn, ‘Autorität und Legitimität in der postnationalen Konstellation’ (2012) 27 Leviathan, Sonderheft 41–62, at 51. 28 Zürn, ibid, who calls it epistemic authority with assigned political authority (translation from German). 29 Strassheim, ch 8 in this volume. 23

6

Maria Weimer and Anniek De Ruijter

t ransformation of the EU’s authority to validate policy-relevant knowledge claims, or what Strassheim refers to as the European ‘ecosystem of expertise.’30 To support its expanding governance structures, the EU has been continuously developing new EU-level institutions and procedures for the generation of policy-relevant expertise. This evolution is exemplified by mainly four trends. First, since the 1990s the EU has been establishing ever more EU regulatory agencies (ie agencification) to which certain tasks of information gathering, coordination and expert advice were outsourced.31 Second, the European Commission has institutionalised an expansive system of expert groups advising it on a variety of policy issues.32 Third, the Better Regulation programme has been dominating the Commission’s institutional policy since the beginning of the 2000s. As part of this programme the Commission’s main instrument of ‘knowing’ in policymaking has become the Regulatory Impact Assessment (RIA) and the accompanying consultation of stakeholders.33 More recently, the Commission has also introduced a new EU Science Advice Mechanism,34 which is supposed to provide the Commission with independent scientific advice on specific policy issues. Fourth, the general trend towards more centralised and institutionalised production of policy-relevant knowledge was augmented by the trend towards regulatory cooperation with third countries and hence the transnationalisation of expertise. Transatlantic trade and cooperation agreements, such as the EU-Canada Comprehensive Economic Trade Agreement (CETA) and the Transatlantic Trade and Investment Partnership (TTIP), require cooperation in the field of knowledge production, and to some extent the harmonisation of procedures in this respect.35 The commitment of TTIP, for example, to RIAs and stakeholder consultations certainly fosters the consolidation of these procedures at EU level (and thereby also the consolidation of EU-level expert structures).36 The increase of EU regulatory powers led to the emergence of knowledge infrastructures intended to secure the EU’s claims to governance authority via

30 ibid. The term has originally been coined by J Wilsdon, R Doubleday and J Hynard, ‘Future irections for Scientific Advice in Europe’ in R Doubleday and J Wilsdon (eds), Future Directions for D Scientific Advice in Europe (Cambridge, Centre for Science and Policy, 2015) 8–22. 31 S Krapohl, ‘Risk Regulation in the EU between Interests and Expertise: The Case of BSE’ (2003) 10 Journal of European Public Policy 189; Busuioc, above n 20; Curtin, above n 20. 32 Vos, above n 23; Vos, above n 20; G Chambers, ‘The BSE Crisis and the European Parliament’ in C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999). 33 A Meuwese, Impact Assessment in EU Lawmaking (The Hague, Kluwer Law International, 2008). 34 Thereby replacing the post of the Chief Scientific Advisor previously held by Anne Glover. 35 J Wiener and A Alemanno, ‘The Future of International Regulatory Cooperation: TTIP as a Learning Process Toward a Global Policy Laboratory’ (2015) 78 Law and Contemporary Problems 103. 36 Commission, ‘Better Regulation for Better Results—An EU Agenda’ (Communication) COM (2015) 215 final; Commission, ‘Better Regulation—Delivering Better Results for a stronger Union’ (Communication) COM (2016) 615 final; See commentary in the Special Issue on the Better Regulation Package—‘How much better is Better Regulation?’ (2015) 6 (3) European Journal of Risk Regulation 344–81.

Regulating Risks in the EU

7

claims to EU epistemic authority.37 The language of risk has been a crucial driver of this transformation: the building of the EU regulatory state—able to secure a free internal market while protecting citizens from market risks—necessitated an ever-complex institutional machinery of EU knowledge production. This transformation of EU expert authority has significant ramifications for both the vertical and the horizontal distribution of power in the Union. And yet it has gone almost unnoticed in EU constitutional debates. To be sure, the rise of EU executive power—the broad delegation of rule-making power to executive institutions at EU level—has been the subject of many constitutional preoccupations and critical analysis.38 Yet the epistemic underpinnings of EU executive power remain largely in the shadows, and a niche subject in mainstream EU law scholarship.39 In contrast, the above-described evolution of EU-level infrastructures of expertise has not gone unnoticed in policy and regulatory studies.40 Yet here we often encounter arguments pertaining to so-called ‘objectivist epistemologies’.41 The policy analysis movement, and as its more recent manifestation, the evidence-based policy movement, are guided by the belief that sound science can provide regulation with certain evidence, which will lead to a more rational problem-solving process.42 This rational-instrumental model of evidence-based regulation43 is rooted in the modernist paradigm, and understands science as culture-neutral, universalistic and objective. The underlying understanding of scientific expertise presupposes a linear separation between the processes of risk assessment and that of risk management, and between facts and values. In the words of Sheila Jasanoff: On this traditional view, the task of evaluating the probability and severity of harm properly belongs with experts, who should work in settings as far removed from politics as possible in order to arrive at technical assessments unaffected by parochial interests or values. Only then can the door into political horse-trading open and political authority come into its play to determine which levels of risk are acceptable to concerned interest groups.44

37 See on the relationship between epistemic and governance authority in the transnational context see Quack, above n 5. 38 Curtin, above n 17. 39 With some notable exceptions, see for example C Joerges, ‘The Law’s Problems with the Governance of the Internal Market’ in C Joerges and R Dehousse (eds) Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002) 3. 40 This is reflected in the ‘knowledge turn’ in EU studies, see C Radaelli, ‘The Role of Knowledge in the Policy Process (1995) 2 (2) Journal of European Public Policy 159–83. 41 See H Strassheim, ‘Politics and Policy Expertise: Towards a Political Epistemology’ in F Fischer et al (eds), Handbook of Critical Policy Studies (Cheltenham, Edward Elgar, 2015) 319 at 322 who traces this term back to Majone. 42 See S Nutley et al, ‘Evidence and Policy in Six European countries: Diverse Approaches and Common Challenges’ (2010) 6 (2) Evidence & Policy 131–44; see for a discussion Strassheim, ibid 322. 43 Others have referred to this model as technocratic, linear or deterministic model of risk regulation, see C Waterton and B Wynne, ‘Knowledge and Political Order in the European Environmental Agency’ in S Jasanoff (ed), States of Knowledge The Co-production of Science and Social Order (London, Routledge, 2004) 87; see also Fisher, ch 3 in this volume. 44 See Jasanoff, ch 2 in this volume.

8

Maria Weimer and Anniek De Ruijter

According to this model, epistemic authority and governance authority are not only neatly separated, but rely on fundamentally different legitimation resources. Scientific and expert claims are legitimised by scientific methodology and peer-review,45 while political claims to authority are legitimised by political (democratic) and legal accountability. A substantial body of interdisciplinary research into the practices of expertbased regulation, however, questions the traditional model, including the feasibility of the underlying separation between expertise and politics.46 Recent empirical research shows that EU institutions (above all the Commission) use expertise not only for technocratic-instrumental, but also for political, purposes.47 It shows, for example, that next to technocratic problem-solving, the Commission also uses expert groups to politically substantiate its pre-defined positions vis-à-vis other actors. Moreover, it uses them for political consensus-building to support negotiations among Member States.48 Technocratic and political concerns are therefore intricately entwined in the formation of EU rules and standards. This volume has emerged as a collective effort to de-construct the objectivistinstrumental model of expertise by critically studying the practices of evidencebased policy making, and thereby to support the call for political epistemologies. In the words of Strassheim: [F]rom the perspective of political epistemologies … evidence-based policy rests on political objectivity as a plausible fiction that tends to obscure many preliminary decisions that influence the production of numbers, indicators and standards. By uncovering the process of politicisation and de-politicisation entrenched in evidence-based practices, political epistemologies in contrast contribute to the capability of political communities to critically re-examining normative issues and to redefine problems that have been framed in ways so far unquestioned in public debates.49

IV. Co-production as an Analytical Lens Studying EU risk regulation through the lens of co-production allows us to uncover those entrenched practices as well as hidden exercises of political power 45 M Hesselink, ‘A European Legal Method? On European Private Law and Scientific Method’ (2009) 15 European Law Journal 20. 46 M van Asselt and E Vos, ‘The Precautionary Principle and the Uncertainty Paradox’ (2006) 9 Journal of Risk Research 313; E Vos and M Everson (eds), Uncertain Risks Regulated (London, Routledge-Cavendish, 2012). 47 eg C Boswell, ‘The Political Functions of Expert Knowledge: Knowledge and Legitimation in European Union Immigration Policy’ (2008) 15 (4) Journal of European Public Policy; J Moodie, ‘Resistant to Change? The European Commission and Expert Group Reform’ (2016) 39 (2) West European Politics; J Metz, The European Commission, Expert Groups and The Policy Process, Demystifying Technocratic Governance (Basingstoke, Palgrave Macmillan, 2015). 48 Metz, ibid. 49 Strassheim, above n 41 at 323.

Regulating Risks in the EU

9

in the EU expert-executive space. But, first, we need to further clarify the meaning of the term co-production as used in this volume. To use Jasanoff ’s words, co-production as an analytical frame for research holds ‘the proposition that the ways in which we know and represent the world (both nature and society) are inseparable from the ways in which we choose to live in it.’50 Humans create science and technology in the context of inherently biased and socially constructed institutions. Co-production can help explain stabilised understandings of the natural world (‘how we know what we know’) but it can also help us seeing the dynamics of how science begets meaning in the social, political or legal world and vice versa.51 In the context of the EU, the social and political construction of its political system hinges on the task to govern not only political subjects—Member States and EU citizens—but also subject matter itself. Risk regulation, as one of the core tasks of the EU, is there to benefit EU citizens. However, risk is often taken to be a self-standing natural occurrence and the determination of an existing risk as a matter of science unaffected by politics and values.52 The idiom of co-production challenges these neat divisions. It acknowledges that risks themselves are ‘framed’ and depend on ‘social acceptance’ for their recognition. Further, risk has become a determinant of social order. As Jasanoff rightly observes, risk does not merely ‘act upon’ society exogenously, but it arises from ‘within social arrangements’: The portrayal of risk governance as merely a technocratic exercise designed to perfect the market overlooks the deeply political character of controlling risks. That political dimension comes clearly into view through the lens of co-production.53

V. The Co-production of EU Expert and Executive Power What does co-production tell us about the role of expertise in shaping the exercise of EU executive power and vice versa? Expertise in the EU is intrinsically linked to EU executive power by providing the latter with legitimacy and justification. This is an important finding of this volume: in the field of risk regulation, the EU seeks to transform claims to EU expertise (made within the institutionalised structures of EU expert advice, eg EU expert groups or agencies) into claims to EU executive authority (ie the claim that the EU executive, above all the Commission, should have the power to decide on certain issues potentially at the expense of other institutions or levels of governance). To refer once more to Jasanoff, ‘centralisation and

50

See Jasanoff, ch 2 in this volume. Jasanoff (ed), above n 43; S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005). 52 See Jasanoff, ch 2 in this volume. 53 ibid. 51

10

Maria Weimer and Anniek De Ruijter

presumed completeness of knowledge supply a firmer basis for credible displays of public reason and regulatory control.’54 The use of expertise including the ubiquitous reliance on science-based and evidence-based regulation is therefore not just an attempt to strengthen the rationality of EU risk regulation, but also and above all a way of justifying decisions on risk and technology including their legal and political consequences.55 Given that a vast body of EU regulatory power is executive in nature,56 the need for a technocratic expert-based justification is amplified. This can also be linked back to characterisations of the EU as an apolitical, non-majoritarian ‘regulatory state’57 In other words, the call for expert-based regulation is deeply connected with the existential need to justify the exercise of executive power at EU level. In her contribution, Elizabeth Fisher states that: ‘It is not just the expertise of knowing about risks that is required in risk regulation, but what is also needed is expertise in how to govern those risks.’58 In that sense, expertise is never just instrumental in the way that the EU Better Regulation discourse would like us to believe. ‘It is embedded in executive spaces and transforms executive power.’59 At the same time, the contributions to this volume produce ample examples of transformations in both directions: not only expertise as transforming EU executive power, but also the latter transforming the nature of policy-relevant expertise. What they show are the non-linear relationships between, on the one hand, the production of science when it is embedded in the EU executive sphere, and on the other hand, the production of executive power through the embedding of knowledge as a policy-instrument in the socio-legal practices of the EU executive. As a result, expert ‘advisors’ and the EU executive inevitably merge into an amalgam of governing power. Elen Stokes’ contribution offers a vivid example. Using EU regulation of synthetic biology as a case study, Stokes considers the ways in which expertise both contributes to and is determined by exercises of EU executive authority. The EU is currently contemplating what regulatory framework would be adequate to regulate this new technological development. Does the current EU regulatory regime on ‘conventional’ genetic modification cover synthetic biology, or is the latter so novel that it requires specially designed regulatory strategies and measures? This calls for attempts to define the very subject matter of EU regulation including 54 ibid. 55

See Weimer and Pisani, ch 9 in this volume. above n 21. From the actor perspective: EU governance is dominated by the involvement of executive actors (even in instances in which legislative acts are being enacted), such as the Commission, the Council, and the European Council, comitology committees and EU agencies. From the perspective of acts: EU legislation usually only provides for broad framework norms, which require substantial implementation by executive actors. In EU risk regulation, the majority of legal acts are executive in nature (individual decisions, implementing and delegated acts or other forms of executive soft-law). 57 Majone, ‘The Rise of Statutory Regulation in Europe’ above n 14. 58 See Fisher, ch 3 in this volume. 59 See Fisher, ch 3 in this volume. 56 Curtin,

Regulating Risks in the EU

11

the conceptualisation of synthetic biology as either novel or conventional genetic modification. Stokes reveals the phenomenon of ‘recombinant regulation’, whereby existing legislative measures recombine with newly emerging applications of technology. By defining synthetic biology as a conventional process that should fall under the existent regulation, the Commission, reliant on a technical expert group, de-politicises the technology. Neither legislative intervention nor a broader debate about the distribution of potential risks and benefits is needed. Moreover, the Commission is able to assert control over a new technology. Treating synthetic biology as equivalent to genetic modification helps to maintain regulatory continuity and ‘normalise’ its novelty. … But it also means that democratic deliberation on the broader ethical, legal and social impacts of synthetic biology is sidelined. … Hence, regulatory stability in the present can come at a cost, in this case to the incorporation of multiple viewpoints and collective decisions about the future. … Overall, then, the recombination of synthetic biology with legislation on ‘conventional’ GMOs provides an example of how new, emerging technological opportunities interact with existent legal and policy institutions and practices at EU level, and of how technical expertise and executive power are constituted and reaffirmed along the way.60

The case study by Tanja Ehnert makes similar observations. Ehnert analyses the definition of the term ‘nanomaterial’ for the purposes of its regulation. Particularly it shows how, in order to address the uncertainty as regards to the actual knowledge of what can be described as ‘nanomaterial,’ the executive took deliberate steps to shift the issue from the European Parliament to the European Commission. As a result the definition of nanomaterials became depoliticised and reframed as a mere technical, soft law issue. Following this, the definition of nanomaterials as carved out by the EU executive became the basis for hard law. The role of the EU executive in defining the world of science and technology and its political uses is not limited to new and emerging fields of research. Policies that are highly interwoven with science—and represent a deep involvement of both political actors and scientific assessments—are shaped by the dominant strain of knowledge entrepreneurs within them. Experts engaged in executive rule-making are able to exert public authority. In the words of Strassheim: To recognise the expertise of an individual or collective actor means to attribute to that actor both the competence to validate and justify knowledge claims (‘epistemic authority’) and the capability to make these knowledge claims relevant for collectively ordering and evaluating society (‘political authority’).61

The case studies by Scott Greer and Holger Strassheim both make this point. Strassheim’s case study on behavioural expertise analyses how European public health policy is recast as a science that can ‘nudge’ people into the right behaviour. ‘Smart-regulation’ through behavioural insights in EU public health and the environment is a key example of the ‘epistemic authority’ that is assigned political 60 61

See Stokes, ch 4 in this volume. See Strassheim, ch 8 in this volume.

12

Maria Weimer and Anniek De Ruijter

authority. The behavioural economic insights are particularly attractive in the context of the EU executive. These insights are presented by its proponents as a solution for the lack of science-policy coordination. Indeed, Strassheim in his chapter shows how ‘the behavioural movement manages to be perceived as both scientifically reputable and politically attractive, capable of filling the institutional void in European policy making.’62 In the case of behavioural expertise, rather than using either science or the political and legal basis as a manner to persuade the public of the legitimacy of a policy, behavioural insights persuade on both those counts. Another field of policy where the experts themselves have a key role in carving out the role for the EU executive is in the field of public health surveillance. As explained by Greer public health surveillance in itself has been developed as an ‘order-making concept’ that carves out an important space for the executive to exercise public authority: Public health surveillance, the collection of public health data for public purposes, is at the core of public health and the modern state. It is invasive, expensive, sometimes coercive, justifies all sorts of public actions, and is fraught with all manner of conflicts about personal data, privacy, state power and public action. It is therefore political. Its political nature, however, is easier to deduce than to see in the everyday work, publications, and statements of people involved in an apparently low-profile, scientific, and cumulative public enterprise. In the center of a vortex of concerns about technology, privacy, disease, power, and liberty, we find an oddly calm world of writing about epidemiological best practice that has shaped, among other institutions, the developing European Union engagement in public health.63

Greer shows how the EU’s adoption of a particular understanding of public health surveillance as both a science and a political tool legitimises far-reaching efforts of personal data gathering, and shapes public health as a EU policy field. The de-politicisation of health surveillance as a source of executive power in the field of public health, is amplified by the re-politicisation of public health when its expertise is merged with security expertise, particularly also when it comes to strengthening the role of the EU executive in public health emergencies, as analysed by de Ruijter. Where Greer focuses on the concept of public health surveillance, the chapter by de Ruijter focuses on the empowerment of the EU executive through the securitisation of responses to public health emergencies. As Greer described, with regard to the field of public health surveillance, EU executive power was co-produced through the assumption of a scientific need for a harmonised understanding of public health surveillance. In contrast, De Ruijter shows that before the last decade the role of the EU executive for managing public health threats remained relatively minor. Yet in tandem with the rise of the risk society,64 and especially after 9/11, 62 ibid. 63

Greer, ch 7 in this volume. Beck, above n 10; U Beck, ‘The Terrorist Threat World Risk Society Revisited’ (2002) 19 Theory, Culture & Society 39. 64

Regulating Risks in the EU

13

the scientific understanding of the nature of health risks arising in emergency (public health) situations has changed. Other types of experts became involved, particularly those working in the area of security. As a result, current EU public health regulation in the field of communicable disease outbreaks, bioterrorism and chemical attacks has merged the scientific input of two very distinct types of expertise—public health experts on the one hand and security experts and economists on the other hand. While the former believe that public health risks are manageable and statistically foreseeable, the latter view such risks as unforeseeable disasters or catastrophes that should be addressed, not only in terms of health and life of humans, but also in terms of security, critical infrastructures and defence.

VI. Broadening the Debate—Rethinking EU Constitutionalism As Elizabeth Fisher observes in this volume,65 in recent years there has been a trend in both scholarly and policy discussions to downplay the broader political issues underlying EU risk regulation with the latter becoming ‘kettled’ into a separate discourse from the mainstream debates about the EU. As stated in the beginning, with this volume we aim to overcome this trend and to re-connect the risk regulatory discussion with the broader discussion on EU executive power and its legitimacy. The salience of expertise for the EU’s democratic governance more generally is well demonstrated by this quote from the chapter by Jasanoff: The European Union emerges … as an inversion of the standard paradigm of modern democratic governance. … Supranational governing bodies such as the EU are less subject to elections or other forms of direct democratic control. For such bodies, the relationship of legitimation is reversed: instead of knowledge and experience serving purposes that the governed have placed on the political agenda, the EU governs by setting the policy agenda on the basis of claimed superior knowledge and expertise. Under these circumstances, the epistemic and social authority of science provide the strongest justification for acting on others’ behalf. Thus, in the context of EU risk governance, the perceived rightfulness of regulatory policies derives from advisory committee claims to know for the entire European polity. Risk governance is a site in which European institutions assert their right to govern all European lives—thereby constructing an imaginary of a united Europe—by generating persuasive technical rationales for centralised policy action.66

These far-reaching shifts of power in the EU’s multi-level polity require a re-thinking of the legal and constitutional foundations of EU (risk) regulation. The main challenge of EU democratic governance (and transnational governance more generally) today is to ensure that the use (including the production) of expert 65 66

See ch 3. Ch 2 in this volume.

14

Maria Weimer and Anniek De Ruijter

knowledge in policymaking is able to safeguard the public interest. Debates on EU constitutionalism, however, have so far neglected this challenge. Epistemic authority largely remains a blind spot in EU constitutionalism debates.67 There is currently no (EU) constitutional theory that can account for epistemic authority and expert power in our pluralist knowledge-based society. This volume supports the claim that expertise serves as ‘an alternate reservoir of power, not subject to the modes of legitimation that govern the legislative, executive or judicial branches of a classic, tripartite national government.’68 We do not claim to deliver such a new constitutional theory, but we do hope to provide for some of its important building blocks. Recognising expertise as ‘an alternate reservoir of power’ means that we need to re-think traditional modes of legitimation of EU executive authority in order to be able to normatively evaluate, and perhaps even control this hidden exercise of power. This means re-thinking EU constitutionalism to ensure the expert-executive amalgam is oriented towards the public good. We need to develop concepts allowing holding EU executive power to account for its use of policy-relevant expertise including how it collects and ensures the quality of that expertise. Law and the study of expertise from a legal-constitutional perspective are relevant here in two respects. First, the EU has been developing soft and hard meta-norms governing the use of expertise by EU institutions, and thereby implicitly acknowledging the significance (and perhaps even the politics) of epistemic authority. We can only briefly refer here to Commission guidelines, communications and white papers on expert consultation, founding regulations for EU agencies, EU legislation on access to documents, Commission Decisions establishing new bodies, such as the Regulatory Scrutiny Board overseeing the Commission’s work on RIAs. Moreover, EU courts are confronted on a daily basis with the legal review of scienceand expert-based decision-making. While this presents an enormous challenge for the courts, they have been gradually developing procedural requirements and principles, which shall govern the exercise of epistemic authority in the EU.69 Lastly, as illustrated by Maria Lee’s contribution, recently the European Ombudsman has produced a number of decisions directly relevant to co-production of epistemic and governance authority.70 Can these developments be seen as beginning counter-movements to the traditional rational-instrumental understandings of the role of expertise in EU risk regulation? We believe that the ways in which law enhances or reduces the possibility of democratic destabilisation of regulatory knowledge demands further scrutiny. Law has the capacity to reveal or to further obscure both the contingency of regulatory choices about technological innovations and the possibility of societal change. This edited volume sheds light

67 But see S Smismans, ‘Constitutionalising Expertise in the EU: Anchoring Knowledge in Democracy’ in J Priban (ed), The Self-Constitution of Europe (Farnham, Ashgate, 2016). 68 Jasanoff, ch 2 in this volume. 69 See Weimer and Pisani, ch 9, Anderson, ch 10, and Lee, ch 11, in this volume. 70 See Lee, ch 11 in this volume.

Regulating Risks in the EU

15

on the many ways in which EU law could contribute to either entrenching the technocratic model of risk regulation or to destabilising it thereby facilitating incremental legal and regulatory change. Second, these developments are not in the spotlight of legal academic interest as they take place in the technical underworld of EU executive governance. Yet, as this volume shows, EU constitutional discourse on co-production provides for a critical lens through which to evaluate the quality of the above-described EU meta-norms on the use of expertise in policy making. How shall we judge the quality of such meta-norms in constitutional terms? In the absence of a constitutional theory as benchmark, the risk is that they simply create a legitimating discourse for functional governance captured by special interests.71 Finally, the dominant EU discourse on multi-level (pluralist) constitutionalism has focused on the interaction between the EU and national legal orders, as well as on issues of subsidiarity and national diversity defined along territorial lines. At the same time, the centralisation of epistemic authority (and hence the lack of epistemic subsidiarity)72 in the EU has been neglected. Yet legal and political multilevel conflicts in the EU are increasingly resolved by recourse to expert claims. Hence, what is won in terms of (de-centralised) governance authority can again be lost in terms of (centralised) epistemic authority. The task ahead is to advance a research agenda for studying EU rules and practices of co-production from a constitutional perspective. This would allow us to ‘theorise more completely the role of science in relation to other institutional means of legitimating governmental power.’73 Pursuing this research agenda will help re-imagining EU constitutionalism taking into account the co-production of EU epistemic and governance authority; and thereby help preserving the core function of constitutionalism, namely to shape the domain of politics in the public interest.74

71

Above n 65. See on this term S Jasanoff, ‘Epistemic Subsidiarity—Co-existence, Cosmopolitanism, Constitutionalism (2013) 4 (2) European Journal of Risk Regulation 133–41. 73 See Jasanoff, ch 2 in this volume. 74 Above n 65. 72

16

Part 1

Theoretical Perspectives

18

2 Constitutions of Modernity: Science, Risk and Governable Subjects SHEILA JASANOFF

I. Two Faces of Risk Governance Risk regulation in the European Union (EU) brings together two concepts that were historically thought to operate independently of one another: risk and the EU. Conventionally, risk is believed to lodge in the universe of what the philosopher Ian Hacking calls ‘natural kinds’,1 that is, categories of things that exist in nature, apart from and distinguishable from human purposes. It is defined as the magnitude of an expected harm (disease, disability, death) times the probability of that harm (one in a hundred, thousand or million).2 Both numbers are taken to reflect natural realities, ascertainable in principle by objective, scientific methods. Benzene, for example, causes aplastic anaemia and leukaemia in a fraction of people who are exposed at concentrations above some definable amount: this is considered a scientifically demonstrable fact, independent of the exposed person’s social and cultural affiliations,3 let alone the political preferences of the authorities who manage people’s exposures. Benzene on this view is a physical agent with properties that are not affected by values or politics. Similar observations could be made about most threats to human health, safety and the environment that have given rise to regulatory concern. The EU, on the other hand, is a political formation through and through, a coordinator of Member States, called a superstate by some and an unidentified 1 I Hacking, ‘A Tradition of Natural Kinds’ (1991) 61 (1/2) Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition 109–26. 2 R Snyder, ‘Leukemia and Benzene’ (2012) 9 (8) International Journal of Environmental Research and Public Health 2875–93. 3 Epidemiologists recognise, of course, that individual biological variability may affect the degree to which a specific human body responds to a particular form of hazard. Nevertheless, for regulatory purposes, inter-individual variability is not normally factored into the construction of a dose-response curve for a given population. Classes of at risk subjects may, however, be isolated for analysis, such as specially vulnerable populations or occupational subgroups exposed to different doses of a hazardous agent at work.

20

Sheila Jasanoff

political object by others, but composed of its own ruling institutions such as an executive, a parliament and a court. It is a social construct, made out of agreements and rules, norms and practices that depend on human actors to interpret and enact them. The role of this political union, like that of democratic nation states, is to govern human conduct for the benefit of the governed, with their duly expressed consent. Unlike nation states, however, the EU is led mostly by unelected officials, and its work has been plagued for many years by charges of a democratic deficit. If any proof were needed of the EU’s precarious political standing, and its need for continual re-legitimation by the European people, the crises of the last several years, culminating in Britain’s referendum of 23 June 2016 to leave the EU after 40 years of membership, offered to many observers definitive affirmation. This conventional way of characterising risk and the EU, holding the former to be natural and the latter to be political, suggests that controlling risk in the EU should consist of two fundamentally non-intersecting tasks: on the one hand, subject-matter governance rationalised by appeals to scientific facts; on the other, the governance of political subjects legitimated through appeals to democracy. EU institutions are authorised to regulate risks for the benefit of the European population, but risks themselves, as products of physical and material conditions, arise apart from the will of politicians or policy experts in Brussels or Strasbourg. On this traditional view, the task of evaluating the probability and severity of harm properly belongs with experts, who should work in settings as far removed from politics as possible in order to arrive at technical assessments unaffected by parochial interests or values. Only then can the door into political horse- trading open and political authority come into play to determine which levels of risk are acceptable to concerned interest groups. For EU policy makers, this way of thinking underlies the judgment that—in fields where regulation is fully harmonised—the technical assessment of risk, viewed as fundamentally scientific, should be conducted at the European level, where experts can make technical findings for the entire European populace, while leaving room for variation in Member States’ policy responses under the principle of subsidiarity. Decades of research on the practices of risk regulation have significantly complicated this picture of a strict and workable separation between science and politics.4 The division between kinds in nature and kinds of political response does not hold up so neatly when risk departs from the safe havens of mathematical calculation and becomes a subject of concern for the regulatory state. First, the kinds of risks that states are entrusted to manage only become palpable, indeed measurable, because of prior collective understandings that have designated some harms as worth worrying about and some causes as important to control. This process of selective acknowledgement is known as framing. Not every framing of risk receives equal policy attention, and introducing a new or different framing 4 For an account of the work done by expert advisory bodies to construct and naturalise perceived boundaries between science and politics, see S Jasanoff, The Fifth Branch: Science Advisers as Policymakers (Cambridge MA, Harvard University Press, 1990).

Constitutions of Modernity

21

of an issue into national or transnational agendas of decision making is itself an exercise of power. Out of the welter of things that might arouse collective concern, only some are picked out as deserving of care and attention. Workplace exposure to a hazard like benzene is but one familiar example of such agenda setting, attesting to modern industrial societies’ particular worries about cancer and possibly also to the strength of their labour movements. Only when a case arrives on the policy agenda is science called upon to distil causes from mere speculative correlations. Science’s fact-finding then defines the precise targets for regulatory intervention—either to mitigate harm that is already occurring or to prevent new, foreseeable injuries to communal well-being—but the threshold choice of which problems policy makers should worry about remains fundamentally social and political. Second, the technical assessment of risk is tied at all points to value-laden choices about how to treat uncertainty. Historically, the legitimacy of nation states depended on their ability to identify and control threats to life and property before their subjects suffered serious harm.5 In late modernity, however, governments confront new challenges occasioned by changes in the nature of population-wide risks. In place of familiar and recurrent hazards, like flash floods, automobile accidents or heart attacks that could be measured and mapped through statistics,6 humanity now faces a host of diffuse, interlocking and synergistic hazards whose combined effects cannot easily be foreseen even with the aid of sophisticated, computerised modelling technologies. Nor can many manifest harms—a child’s cancer, a virulent disease outbreak or an unexpectedly severe storm—be traced back with certainty to definite causes, such as exposure to waterborne c hemicals,7 a new virus,8 or the changing climate. Indeed, even climate change, today’s archexample of a scientifically demonstrated, catastrophic threat to humanity, cannot be pinned down with respect to the specific times, locations, severity or distribution of its consequences. And the ‘world risk society’, so dubbed by the late sociologist Ulrich Beck, faces dangers that are even less determinate than climate change—such as rising antimicrobial resistance, global pandemics, devastating crop failures, and crashes of global financial markets.9 Further, the spread of terrorist networks and cybercrime signals that all-too-knowing human agents can take advantage of weak links in modern institutions of knowledge and power to strike people in places—subways, airports, cafés, buses, beach promenades and

5

M Foucault, Discipline and Punish (New York, Vintage, 1979). referred to this statistically driven mode of governance as the dispositif (or apparatus) of security, through which states characterise and control probable events. See M Foucault, Security, Territory, Population: Lectures at the Collège de France 1977–1978 (trans G Burchell, Paris, Gallimard, 2004) 4–5. 7 See, for example, the protracted and inconclusive litigation arising from children’s leukaemia cases in Woburn, Massachusetts in the 1980s. J Harr, A Civil Action (New York, Random House, 1995). 8 The Zika outbreak of 2016 is an illustrative case. It is known to be transmitted by the yellow fever mosquito, but how infection spreads and who will suffer irreparable health damage remain mysterious. 9 U Beck, World at Risk (Cambridge, Polity, 2007). 6 Foucault

22

Sheila Jasanoff

Internet sites—where they had previously felt safe. In this respect, the promise of total control built into many of modernity’s most advanced technological systems becomes itself an invitation to subversion, producing yet another frontier of risk. Third, neither society nor its perceptions hold still in the face of acknowledged concerns. Much has been written in the sociological literature about the implications of the contemporary era of risk for the structure, functioning and selfunderstanding of modern societies. Beck’s early observation that risk, overriding race, class and other factors of social stratification, has become a pervasive determinant of social order, struck a deep chord,10 and an entire academic superstructure has risen since then to explore its implications.11 Beck himself saw the global reordering of risk society as good reason to move away from a focus on the nation state as the primary unit of social science analysis, and he urged recognition for cosmopolitanism as a phenomenon, a normative project and a research methodology. In the resulting body of work, society continually, and reflexively, remakes itself in the light of half-seen, incompletely understood, intangible, but ubiquitous threats, prompting new forms of social solidarity as well as instability. But powerful as these observations are, the very designation ‘risk society’ continues to treat risk as something exogenous, acting upon society rather than arising organically from and within social arrangements. In its focus on certain well-defined risks such as chemicals, nuclear power and climate change, the risk society literature does not fully take on board the insights from constructivist studies of risk s howing that the things society worries about evolve at one and the same time as society itself. Faced with this ongoing double shift in understandings of both nature and society, risk governance becomes not so much a matter of keeping bad (if poorly understood) things under control as a more complex exercise in world-making, in the particular form that I term co-production.12

II. Risk and European Integration: A Co-productionist View On its face, risk governance conforms very well to the traditional model of European integration that views the market as the main instrument of integration and the standardisation of commodities in that market as merely the epiphenomenon that

10

U Beck, Risk Society: Towards a New Modernity (London, Sage Publications, 1992). See, for example, U Beck, A Giddens and S Lash, Reflexive Modernization (London, Polity, 1994); S Lash, B Szerszynski, and B Wynne (eds), Risk, Environment and Modernity: Towards a New Ecology (London, Sage Publications, 2000); B Adam, U Beck, J Van Loon (eds), The Risk Society and Beyond: Critical Issues for Social Theory (London, Sage, 2000). 12 S Jasanoff (ed), States of Knowledge: The Co-Production of Science and Social Order (London, Routledge, 2004) 15–45. 11

Constitutions of Modernity

23

enables the market to function. Integration, after all, could scarcely be meaningful if the market posed unacceptable risks to the people the EU represents. But this portrayal of risk governance as merely a technocratic exercise designed to p erfect the market overlooks the deeply political character of controlling risks. That political dimension comes clearly into view through the lens of co-production. Co-production, as that concept has developed in science and technology studies (STS), is defined as shorthand for the proposition that the ways in which we know and represent the world (both nature and society) are inseparable from the ways in which we choose to live in it. Knowledge and its material embodiments are at once products of social work and constitutive of forms of social life; society cannot function without knowledge any more than knowledge can exist without appropriate social supports.13

This deep entanglement of knowledge, materiality and norms is most apparent in moments when a given social order changes perceptibly, either through competition between clashing visions of the world or through entirely new arrangements coming into view. Co-production, in short, can be observed most clearly at times of emergence, contestation, standardisation, and importation of ideas from one source into other contexts. In any of these transitional phases, the co-produced settlements—what one might call the ‘new natural’—get built into core elements of social order: most importantly, representations, institutions, discourses and identities. Risk governance, often involving the recognition of emergent hazards and new modes of dealing with them, has proved to be a rich site for observing processes of co-production unfolding. As avid students and occasional critics of modernity, STS scholars have paid particular attention to the moments in which promises of expert competence and control are constructed and made to appear persuasive in the context of risk governance. That preference is consistent with Bruno Latour’s influential claim that modernity can best be understood as a stage of human development in which the products of science (the facts of nature) are carved out from their social matrix and turned into stand-alone representations of reality.14 Even STS scholars fall victim to that most basic form of modern boundary work in choosing technoscientific products, claims or contestations as their favoured sites of analysis. Yet, no account of societal encounters with risk would be complete unless a focus on the production of technical claims is balanced by a symmetrical and correspondingly detailed account of how law, humanity’s other great ordering instrument, works along with science to delineate grounds for political response that appear morally rightful as well as epistemically right. Central to the analytic project of co-production is the observation that the two most widely recognised forms of lawfulness—scientific and legal—work together, even in partitioning the hybrid 13 14

ibid, 2–3. B Latour, We Have Never Been Modern (Cambridge MA, Harvard University Press, 1993).

24

Sheila Jasanoff

metaphysical and normative spaces of modernity into domains that are seen as belonging either to science or to law.15 Getting the relationship of science, law and politics right is crucial to the perceived legitimacy of both nation states and international or supranational bodies, including the unique political construct of the EU. Yet, even at the level of national government, the constitutional role of science remains poorly understood and barely theorised. For the supranational organisations that play an ever more prominent role in global governance, theoretical thinking about the role of expertise remains even more impoverished.16 How, then, might one re-conceptualise risk regulation in the European Union in the co-productionist idiom? The constitution of a politically governable space would have to be seen as simultaneous with and inseparable from the constitution of scientifically knowable targets of regulation, which the authorities understand well enough to order with technical rules and standards. To the extent that EU institutions manage to hold these two forms of governability together, they are seen as successful; if they fail, then so does the EU’s claim to be legitimately responsible for the lives and well-being of the subjects it represents. It is in this sense that one can appropriately speak of science as playing a constitutional role in European governance. Getting the science right, or right enough, is an unwritten but unassailable background norm that must be met if the EU is to get its politics right. In the present period of European crisis, it is worth examining the implications of these statements more closely, since they bear on the validity of the entire European project, a project that has foundationally, albeit controversially, relied on technical capacity as the primary basis for asserting Europe-wide political control. But where should one look? Taking risk as the subject matter, the remainder of this paper lays out a conceptual map for how and where one might investigate the co-production of knowledge and governance in the EU. Three themes will organise that discussion: delegation, harmonisation and subsidiarity.

III. Delegation Democratic governance begins, as any student of politics knows, with the act of delegation. The people delegate power to the sovereign. No ruler can presume 15 For a reminder of the close coupling of the two kinds of lawfulness in philosophical thought, one need only recall the words inscribed on Immanuel Kant’s tombstone: ‘Two things fill the mind with ever new and increasing admiration and awe, the more often and steadily we reflect upon them: the starry heavens above me and the moral law within me’. P Guyer, ‘Introduction: the Starry Heavens and the Moral Law’ in P Guyer (ed), The Cambridge Companion to Kant (Cambridge, Cambridge University Press, 1992). Kant’s moral philosophy was deeply influenced by an early exposure to nature and natural law, and interest in science remained a seminal part of his thinking. 16 A relatively rare exception in the field of international legal scholarship is D Kennedy, A World of Struggle: How Power, Law and Expertise Shape Global Political Economy (Princeton, Princeton University Press, 2016).

Constitutions of Modernity

25

to exercise that power legitimately without adhering to some process by which people consent to be governed. Popular elections, in which people directly express their preferences, are the most common device for ascertaining the collective will. The EU, however, is not in this respect a fully representative democracy; only the European Parliament is directly elected by voters in the Member States. European rule is sanctioned by the treaties that define the Union. And, as illustrated by the Brexit vote of June 2016 and its tortured aftermath, even the most fleeting, ambivalent and poorly thought out expression of popular sovereignty in a Member State carries greater moral authority as an expression of democracy than modes of governance cemented, as are the EU’s, through layers of formal treaties, declarations and regulatory enactments. For risk governance in the EU, a second form of delegation has proved to be much more significant. This is epistemic delegation, or the power of a handful of expert actors in a governing institution to know for the many. In one way or another, the delegation of epistemic power is central to the orderly operation of modern states, although it often remains hidden under the same veil of ignorance that for so long protected the institutions of science and technology from social theorising and analysis. Yet, the practices of national governments and their political leaders show many traces of the importance they attach to collective knowing as a mode of governance. In the United States, for example, presidents from both sides of the political spectrum have made a practice of speaking to the National Academy of Sciences (NAS), acknowledging a tacit partnership between science and government. President George HW Bush offered, in effect, an encapsulated theory of delegation when he addressed the NAS in 1990: Science, like any field of endeavor, relies on freedom of inquiry; and one of the h allmarks of that freedom is objectivity. Now more than ever, on issues ranging from climate change to AIDS research to genetic engineering to food additives, government relies on the impartial perspective of science for guidance. And as the frontiers of knowledge are increasingly distant from the understanding of the many, it is ever more important that we can turn to the few for sound, straightforward advice.17

The forty-first US president was no theoretician of politics or of science, and certainly not trained in STS. Yet in underscoring the problematic connection between the power of the few in relation to the needs of the many, he gave sophisticated voice to the American public’s understanding of modernity. That understanding, or civic epistemology, entails, as I have argued elsewhere, treating science as a space apart from politics and as a source of truths uncontaminated by human values and interests.18 If one reads President Bush here as a political theorist, he is asserting, in line with widely held American beliefs, that science is entitled to an 17 President George W Bush, Remarks to the National Academy of Sciences, April 23, 1990, Public Papers of the Presidents of the United States: George Bush, 1990, 544–47, available at www.presidency. ucsb.edu/ws/?pid=18393. 18 S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005).

26

Sheila Jasanoff

autonomous space in which it can enjoy unchecked freedom of inquiry.19 But that freedom must be won through objectivity, the ‘hallmark’ of science’s impartiality. Such a science, almost by definition, speaks for everyone because it is beholden to no one, and it can therefore be turned to for ‘sound, straightforward advice’ on matters of broad public interest. Impartiality, then, becomes the basis of delegation, from the many who do not have the necessary expertise to the few who can supply that essential ingredient for governing well in modernity. The corresponding hallmark in EU governance is the growing reliance on supposedly independent scientific panels providing risk assessments that, in turn, are meant to guide the risk management functions of the European Commission and Parliament. Ironically, the very issues on which President George HW Bush asserted the need for sound scientific advice—climate change, AIDS, genetic engineering and food additives—were some of those on which his son, President George W Bush, and his political allies, most blatantly ignored scientific findings.20 In American constitutional terms, however, one could see the forty-third presidency as an aberration or rupture that was put right by President Barack Obama, beginning with the much-touted promise in his inaugural address: ‘We’ll restore science to its rightful place.’21 This was a re-performance of élite American political culture on multiple levels, not merely in acknowledging the political significance of science but also in affirming, as his predecessor George HW Bush had done, the need for (American) science to comply with exogenous standards of impartiality. In one of his own speeches to the NAS, Obama stated that conviction as follows: And more important than any single study or report, the members of this institution embody what is so necessary for us to continue our scientific advance and to maintain our cutting-edge, and that’s restless curiosity and boundless hope, but also a fidelity to facts and truth, and a willingness to follow where the evidence leads.22

That demand for a commitment to ‘facts and truth’, by following the evidence, is especially noteworthy in a speech that otherwise stressed science’s ties to industry and to the nation’s economic, security and environmental needs. Put in more analytic terms, one sees here a characteristically American, pragmatic embrace of utilitarian, or Mode 2,23 technoscience, while at the same time asserting the Platonic ideal of an interest-free enterprise, wholly committed to scholarly truth-telling. The level of shock and dismay greeting Donald J Trump’s ascension to the presidency on what many saw as a blatantly ‘post-truth’ campaign ironically 19 D Guston, Between Politics and Science: Assuring the Integrity and Productivity of Research (Cambridge, Cambridge University Press, 2000). 20 C Mooney, The Republican War on Science (New York, Basic Books, 2005). 21 M Philips, ‘President Barack Obama’s Inaugural Address’ (The White House, 20 January 2009) www.whitehouse.gov/blog/2009/01/21/president-barack-obamas-inaugural-address. 22 Office of the Press Secretary, Remarks by the President on the 150th Anniversary of the National Academy of Sciences (The White House, 29 April 2013) www.whitehouse.gov/the-pressoffice/2013/04/29/remarks-president-150th-anniversary-national-academy-sciences, emphasis added. 23 M Gibbons, C Limoges, H Nowotny, S Schwartzman, P Scott and M Trow, The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies (London, Sage, 1994).

Constitutions of Modernity

27

underscored the dominant American attachment to the notion of independent and value-free science. Other nations, governed in accordance with their own civic epistemologies, express their preferences for delegating power to knowledge in different forms and venues. Political leaders in most countries do not practice the ritual of addressing the nation’s top science academy, even where there is one, with the regularity one finds in recent US politics, although Tony Blair notably addressed the Royal Society in 2002 with a speech showing greater awareness of science-society relations, and of science’s role in a nation’s ethical dilemmas, than is typical of comparable US pronouncements.24 Nonetheless, if one looks to the practices of governance in modern nation states, one finds a wealth of express and tacit understandings signalling the elements of proper (and by the same token improper) delegation of power of knowledge.25 These commitments exist in the EU as well as in its Member States.26 Examples abound of the ways in which European expert agencies have emerged as sites of co-production, carrying out delegated functions of producing facts that also serve as bearers of normative presumptions. Claire Waterton and Brian Wynne described, for instance, how the European Environment Agency became a site of contestation between contrasting ideas of European integration implicating different visions of the role of expertise: either fully harmonised from above by centralised expert bodies claiming to know for the European polity; or more chaotically from below by highlighting areas of ignorance and inviting more diverse citizen inputs.27 There are good reasons to think that the former vision remains more palatable to EU decision makers than the latter, in part because centralisation and presumed completeness of knowledge supply to unelected officials a firmer basis for credible displays of public reason and regulatory control. A more recent example of how the traditional model of separating science from politics works is the European Food Standards Agency (EFSA). The agency was 24 See, for example, Full text of Tony Blair’s speech on British science Guardian (London, 22 May 2002) www.theguardian.com/politics/2002/may/23/speeches.tonyblair. Among many constitutional and co-productionist overtones in the speech, Blair asserted, ‘We have the right to judge but we also have a right to know.’ 25 An example is the 1946 US Administrative Procedure Act, 5 US Code Ch 5, which mandated that federal agencies must seek public comment on proposed rules, whether through a notice and comment procedure or a more elaborate public hearing. That provision implies that publics know enough, or are ‘knowledge-able’ enough, to evaluate the state’s technical arguments. It is therefore impermissible for government action to be undertaken without subjecting its evidentiary basis to public review. For a fuller exposition, see S Jasanoff, ‘The Politics of Public Reason’ in P Baert and F Rubio (eds), The Politics of Knowledge (Abingdon, Routledge, 2011) 11–32. 26 At the 2016 Euroscience Open Forum (ESOF), held less than a month after Britain’s historic Brexit vote, both Carlos Moedas, European Commissioner for Science, and Jo Johnson, UK Minister for Universities and Science, took pains to assure the audience that the referendum would not compromise Britain’s participation in European and international science. In short, both speakers performed boundary work separating the unified and collaborative world of science from any hint of partisan political division. Personal observation, Manchester, UK, 25 July 2016. 27 C Waterton and B Wynne, ‘In the Eye of the Hurricane: Knowledge and Political Order in the European Environment Agency’ in Jasanoff (ed), above n 12, 87–108.

28

Sheila Jasanoff

formed in 2002 in response to the BSE (‘mad cow’) crisis and other food scares that revealed serious mismatches in communication and emergency response capabilities across Europe.28 This was a moment of emergence, when neither the form nor the functions of the agency-to-be were more than designer blueprints. The EFSA’s founding mandate, however, recognised from the start a convenient, if intellectually questionable,29 separation between risk assessment (science) and risk management (policy). Only the scientific functions were delegated to the new agency. The EFSA’s operating values are designed to further insulate the agency against charges of playing politics: scientific excellence, independence, openness, innovation and cooperation. This self-presentation may preserve the EFSA’s claims to being an impartial scientific organisation, but it does so by pushing behind the stage curtain of closed expert deliberations,30 away from possibly disruptive popular scrutiny, the inescapable politics of producing scientific advice under conditions of uncertainty. The EFSA in this respect plays out the EU’s still dominant mode of self-legitimation, that of asserting policy legitimacy on the back of a claimed expert capacity to know right for the entire European polity.

IV. Harmonisation Standards and standardisation have long been recognised as the bedrock for working markets, especially when markets extend across political and cultural boundaries. In the EU as in the United States and across the globe, a basic function of expert decision making has been to ensure that standards operate smoothly across an entire trading region, ensuring commensurability between things travelling from one place to another. Commerce would become impossibly difficult if trading partners could not count on a metre being a metre, silk being silk, or pasteurised meaning pasteurised in all corners of a given market. Superimposed on demands for simple uniformity in claims concerning weights and measures, or the physical composition of things, are newer expectations that commodities will conform to the same health, safety and environmental standards everywhere. It should be possible, in principle, for a German farmer to buy a weed killer from Holland or for an Italian company to acquire a piece of software from France knowing that these products are as safe and reliable as they would be if manufactured and regulated within the purchaser’s home country. States and international organisations therefore spend considerable resources in harmonising process and

28

EFSA was established under the General Food Law, Reg 178/2002. above n 4; see also D Winickoff, S Jasanoff, L Busch, R Grove-White and B Wynne, ‘Adjudicating the GM Food Wars: Science, Risk, and Democracy in World Trade Law’ (2005) 30 Yale Journal of International Law 81–123. 30 S Hilgartner, Science on Stage: Expert Advice as Public Drama (Stanford CA, Stanford University Press, 2000). 29 Jasanoff,

Constitutions of Modernity

29

product standards across political divisions; a well-standardised market is believed to raise all economic ships in its jurisdiction. Though seen as mundane and technical, standards are anything but apolitical. In specifying the properties of how things ought to be, even along such elementary axes as size, shape, weight or composition, standards inevitably remake the world to advance some values and not others. Safety standards, in particular, always entail associated questions of political significance: safety for whom, against what, and to what extent? Technological design therefore functions as an instrument of inclusion and exclusion. Even something so commonplace and universally recognised as the red-yellow-green sequence of traffic lights was designed without the needs of colour-blind individuals in mind. Driving, for persons who cannot distinguish all three colours, is marginally more hazardous than for those with full colour vision. Society has deemed this risk tolerable, partly by standardising the vertical sequence of colours so that everyone can see which light is on even if the colours blur. Nonetheless, standardising the colour of traffic lights presumed a false uniformity in drivers’ sensory ability to perceive those colours as distinct. Most regulatory decisions pose far graver problems of inclusion and exclusion. Should environmental standards for air and water quality, for example, be set so as to protect the average person or those most vulnerable, such as the elderly, small children, or those with pre-existing health conditions? Standards for common air pollutants, such as ozone and suspended particulate matter, have proved enormously controversial because people cannot agree who will be affected, or how badly, at different levels of protection. Examples like these led the sociologist Lawrence Busch to speak of standards as ‘recipes for reality’.31 Standards, he argues, borrowing language popularised by the British historian EP Thompson,32 encode a moral economy in which considerations of what kinds of exchanges we want, between which kinds of people, and for what purposes, influence the shape of reality. Standards thus perform a dual function in risk governance, regulating the properties of commercial commodities and processes while also expressing collective values about what kinds of exchanges are worth encouraging or preventing, and at what cost. Standard-setting, so seen, is an instrument of co-production, one that stabilises the world into certain forms of allowed and allowable transactions while making others undesirable or downright impossible. When a supranational body such as an EU agency contributes to the harmonisation of regulatory standards, it thus carries out an act of much more than technical coordination. Of necessity, harmonisation has to bring into alignment divergent scientific perspectives on the material make-up of the world and normative ideas about how to live satisfactorily with those things. Recognising a new environmental hazard, for example, involves an exercise in making both ontologies and norms. By acknowledging that the air we breathe contains within it low-level radiation, 31

L Busch, Standards: Recipes for Reality (Cambridge MA, MIT Press, 2011). E Thompson, ‘The Moral Economy of the English Crowd in the Eighteenth Century’ (1971) 50 Past & Present 76–136. 32

30

Sheila Jasanoff

fine particulate matter or greenhouse gases, the regulatory authority expresses a collective judgement that these things matter to political subjects whose desire for protection from those hazards is entitled to respect. At the same time, the regulator becomes responsible for keeping the troublesome entities in check by specifying how they should be defined, characterised, monitored and controlled. Failures along either dimension of regulatory judgement—defining the right subject matter or placating the concerned political subjects—may destabilise harmonisation efforts. Regulatory controversies typically manifest just such double breakdowns. A telling example was Germany’s attempt to ensure that both genetically modified (GM) and non-genetically modified agriculture could co-exist in the country by enacting strict containment standards that would prevent contamination between GM and non-GM crops. It turned out that, even if the plant varieties could be so compartmentalised, bees could not be kept from feeding on both kinds of crops, to the dismay of beekeepers and organic honey manufacturers, who discovered that they could no longer certify their products as GM-free. German beekeepers mobilised to express their displeasure, demonstrating to the state how its technical prescriptions for GM-non-GM coexistence had not only unwittingly erased bees from the analytic frame of risk assessment but also penalised a whole class of previously invisible political subjects—the beekeepers.33 The honeybee contretemps in Germany occurred within a single nation state, showing that regulatory processes can produce unexpected discord between desired conditions in nature and desired forms of political subjectivity even within a relatively homogeneous political culture. The potential for such problems of fit increases dramatically when regulation seeks to harmonise techno-political standards across the boundaries of political systems—potentially giving rise to new claims of subsidiarity.

V. Subsidiarity Risk governance today is rarely the preserve of any single level of government. Responsibilities for identifying, characterising and controlling threats to human well-being are distributed in complex, often unacknowledged ways both within and beyond nation states. Some arrangements are governed by laws, others by treaties, still others only by customary agreement or judicial interpretation, or not at all. To take just one example, the internal administrative procedures of the Intergovernmental Panel on Climate Change (IPCC), co-winner of the 2007 Nobel Peace Prize and surely one of the most influential expert advisory committees ever 33 J Lezaun, ‘Bees, Beekeepers, and Bureaucrats: Parasitism and the Politics of Transgenic Life’ (2011) Environment and Planning D: Society and Space 738–56. See also Case C-442/09 Karl Heinz Bablok and Others v Freistaat Bayern [2011] I-07419.

Constitutions of Modernity

31

created, are largely self-made and not subject to any transparent process or formal international agreement. It is a truism in political thought that multilevel governance functions best when policy development and implementation are delegated to the level at which those tasks can be carried out with greatest efficiency and political buy-in. The US Constitution accomplishes this result by specifying that the federal government can execute only those powers expressly delegated to it. Other powers, including most matters affecting local communities, are supposed to remain the preserve of the states. The federal government does enjoy broad powers to impose uniform regulatory standards throughout the nation under the interstate commerce clause, but even in those areas where federal pre-emption is the norm, space remains for states to tailor policies and standards that go beyond the specifics of federal law. A case in point is the Regional Greenhouse Gas Initiative, a mandatory, marketbased programme to curb greenhouse gases crafted by nine (originally 10) states in the Northeast and Mid-Atlantic regions of the United States.34 Recurrent controversies over the labelling of GM foods display some of the tensions involved in regimes of shared governance. Until 2016, when the federal government began evolving a controversial nationwide GMO labelling provision, state efforts to demand such labelling had been opposed on the ground that they seemed to assert the existence of a risk in defiance of a federal scientific determination of safety. In something of a reversal of roles, states were allowed to challenge federal expertise on climate change in a case that involved inaction at the higher level of government. The Environmental Protection Agency (EPA) under President George W Bush had refused to set standards to curb greenhouse gases despite mounting evidence of the seriousness of climate change. In Massachusetts v EPA,35 a majority of the US Supreme Court held that the state of climate science was robust enough to require EPA at least to explain its refusal to curb emissions under the authority of the Clean Air Act. While not a frontal assault on the EPA’s scientific integrity or competence, the case can be read as setting limits on a national regulatory authority’s power to insist on its own expert judgement in contravention of a developing body of international scientific opinion on the existence and urgency of an issue of domestic regulatory concern.36 The EU’s principle of subsidiarity aims to strike a comparable balance between European and Member State responsibilities, and here too cracks have appeared in the effort to exert top-down control through appeals to scientific universalism. In the context of risk governance, subsidiarity means that aspects of risk management are left to Member States, but typically not the task of risk assessment,

34

Regional Greenhouse Gas Initiative, www.rggi.org/. Massachusetts v EPA, 549 US 497 (2007). 36 In his dissent, Justice Antonin Scalia laid out an opposing vision of how to balance political sovereignty against the epistemic authority of science. In his view, EPA’s discretionary power under US administrative law allowed the agency to conclude that greenhouse gases were not ‘air pollutants’ as defined by the Clean Air Act and hence not subject to control under that statute. 35

32

Sheila Jasanoff

which (as discussed above) tends to be seen as scientific, apolitical, and hence outside the purview of political sovereignty as conventionally understood. Therefore, risk assessment, in harmonised fields, can appropriately be conducted at the EU level for the Union as a whole. Yet, European regulatory practices complicate this separation of the policy territory between the centralised authority of a universally valid science and the decentralised political sovereignty of nation states and their expert bodies. Despite all efforts at harmonisation, differences may arise throughout the technical processes of risk analysis: identifying the category of things to be regulated (eg chemicals, nanomaterials, biological agents), selecting and characterising specific targets (eg phthalates, trichloroethylene, nanoparticles), and forging consensus in the interpretation of uncertain data.37 These processes may lead to differing results even among EU institutions, depending on how each decision-making body construes overarching legal mandates such as the precautionary principle. Observing such differences in the making has provided grounds for applying the subsidiarity principle in new ways in European risk governance. The long-running debate on agricultural GMOs in the EU offers an example of these dynamics at work, although this represents, in many respects, an exceptional case in the history of EU harmonisation. Under Directive 2001/18/EC, legal authorisation of GMOs became, and still firmly remains, the obligation of the centre, but decisions to cultivate or not to cultivate authorised organisms have recently been delegated to the Member States under Directive (EU) 2015/412.38 This compromise limits the sovereignty of EU expertise to safety assessments conducted by EFSA for purposes of authorisation. It does not extend to possible concerns about local issues such as land use or agricultural policy. As far back as 2009, European Commission President José Manuel Barroso cited the principle of subsidiarity in acknowledging the need for greater flexibility in GMO regulation. His comments spurred a proposal to modify the 2001 directive on deliberate release, a proposal that in turn called attention to tensions between the EU’s claim to scientific authority and Member States’ political freedom: In September 2009 the political guidelines for the new Commission set out by President Barroso made reference to the principle of subsidiarity in the GMO area as an example where the balance may not be always right between an EU framework and the need to take account of diversity in an EU of 27 Member States. According to these guidelines, it should be possible to combine a European Union authorisation system for GMOs, based on science, with freedom for Member States to decide whether or not they wish to cultivate GM crops on their territory.39

37 H Boullier and B Laurent, ‘La précaution réglementaire: Un mode européen de gouvernement des objets techniques’ (2015) 3 Politique européenne 30–53. 38 Directive (EU) 2015/412 of the European Parliament and of the Council of 11 March 2015 amending Directive 2001/18/EC as regards the possibility for the Member States to restrict or prohibit the cultivation of genetically modified organisms (GMOs) in their territory, OJ L 68, 1–18. 39 European Commission, ‘Proposal for a Regulation of the European Parliament and of the Council amending Directive 2001/18/EC as regards the possibility for the Member States to restrict or prohibit the cultivation of GMOs in their territory’ COM (2010) 375 final.

Constitutions of Modernity

33

The 2015 directive realising this proposal created a de facto space of epistemic subsidiarity in which the Commission acknowledges that the ‘EU framework’, as originally conceived, did not take full account of Member States’ local knowledge of their own agricultural needs and circumstances.40

VI. The Constitutional Place of Science in EU Governance The European Union emerges in the light of the foregoing discussion almost as an inversion of the standard paradigm of modern democratic governance. In the standard version, nation states derive legitimacy in the first instance through accountability to a designated polity. In turn, that polity owes allegiance to an imagined community (the nation)41 and is tied to its government through rituals of representation, such as direct elections. To maintain public support, however, modern governments increasingly rely not merely on electoral validation but also on instrumental uses of science and technology to defend or enhance their subjects’ security and well-being.42 Supranational governing bodies such as the EU are less subject to elections or other forms of direct democratic control. For such bodies, the relationship of legitimation is reversed: instead of knowledge and experience serving purposes that the governed have placed on the political agenda, the EU governs by setting the policy agenda on the basis of claimed superior knowledge and expertise. Under these circumstances, the epistemic and social authority of science provide the strongest justification for acting on others’ behalf. Thus, in the context of EU risk governance, the perceived rightfulness of regulatory policies derives from advisory committee claims to know for the entire European polity. Risk governance is a site in which European institutions assert their right to govern all European lives—thereby constructing an imaginary of a united Europe43— by generating persuasive technical rationales for centralised policy action. Studying the practices of delegation, harmonisation and subsidiarity in EU risk decision making allows us to theorise more completely the role of science in relation to other institutional means of legitimating governmental power. We note, first of all, that the very idea of science conjures up an aura of impartial objectivity that carries extraordinary force in European as well as national public reasoning 40 At the same time, the EU’s tendency to artificially separate risk assessment and risk management continues to be reflected even in this measure. This is due to subsidiarity being acknowledged only insofar as national decisions do not conflict with EFSA’s evaluation of public health and environmental risks. For a critical discussion, see M Weimer, ‘Risk Regulation and Deliberation in EU Administrative Governance’ (2015) 5 European Law Journal 622; M Lee, ‘GMOs in the Internal Market: New Legislation on National Flexibility’ (2016) 2 Modern Law Review 317. 41 B Anderson, Imagined Communities (New York, Verso, 1983). 42 Y Ezrahi, The Descent of Icarus: Science and the Transformation of Contemporary Democracy (Cambridge MA, Harvard University Press, 1990). 43 For a fuller exposition of this argument, see Jasanoff, above n 18, 68–93.

34

Sheila Jasanoff

in Member States. In the long history of modernity, some ways of understanding what science is and does have come to seem so natural that they operate as unexamined background beliefs to shore up legal and political decision making. These assumptions draw at best inconsistently on knowledge from disciplines such as the sociology or philosophy of science, or STS, although attempts have been made to translate such knowledge into legal practice.44 For instance, the famous US Supreme Court decision on the admissibility of scientific evidence, Daubert v Merrell Dow,45 somewhat incoherently cited both Karl Popper’s notion of falsifiability and my own work on regulatory peer review in setting forth criteria by which judges should determine whether an offer of evidence meets scientific standards. Yet, the authority of science does not place it entirely outside the sphere of politics. Instead, the distinctiveness of any political culture derives in part from the particular ways in which it harnesses science and technology to shore up the state’s legitimacy.46 Second, science backs up political power through principles of delegation to experts that are sometimes explicit but also often tacit. National administrative law is frequently quite precise in laying out principles by which experts should be consulted and the advice they offer be subjected to wider public scrutiny. Political cultures differ in the extent to which they allow the public eye to penetrate into domains of expert advice-giving, but many recognise the need for some degree of transparency and inclusion. In international and global arenas, by contrast, the rules of expert consultation tend to be more often tacit than explicit. For example, it would be difficult to trace the pathways by which the commitment to treating risk assessment as a science, independent of the politics of risk management, entered into EU governance, although the BSE (‘mad cow’) crisis may well have led to its wider adoption.47 By now, however, the notion that risk assessment is an apolitical exercise is not only well entrenched in Europe, but it serves, as we have seen, to justify the mandates of bodies like the EFSA, and ultimately to steer some dimensions of policy judgement away from the full glare of public scrutiny. Third, moving from the conceptual and institutional bases for relying on science to actual policy practices, one can discern various forms of accountability that constrain the conduct of public reasoning. How regulators collect evidence, how they frame and articulate arguments, and whether they are or are not subject 44 See, for example, S Cole, M Lynch, R McNally and K Jordan, Truth Machine: The Contentious istory of DNA Fingerprinting (Chicago, University of Chicago Press, 2008); S Cole, ‘Forensic C H ulture as Epistemic Culture: The Sociology of Forensic Science’ (2013) 44 (1) Studies in History and P hilosophy of Biological and Biomedical Sciences 36–46; S Jasanoff, ‘What Judges Should Know about the Sociology of Science’ (1992) 32 (3) Jurimetrics 345–59. 45 Daubert v Merrell Dow Pharmaceuticals, Inc, 509 US 579 (1993). 46 See, in particular, S Jasanoff, ‘Judgment under Siege: The Three-Body Problem of Expert Legitimacy’ in P Weingart and S Maasen (eds), Democratization of Expertise? Exploring Novel Forms of Scientific Advice in Political Decision-Making, Sociology of the Sciences Yearbook (Dordrecht, Kluwer, 2005) 209–24. 47 The history of the risk assessment-risk management divide in US policy is much more transparent. The source is a 1983 report of the National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academies Press, 1983).

Constitutions of Modernity

35

to public challenge are all relevant to the legitimate working of a political order, though such practices may not be formally codified. Tacit principles often come to light in contested situations, when things have gone wrong and new rules must be put in place. For example, Britain’s ‘mad cow’ crisis led to the dismantling of an entire governmental body, the Ministry of Agriculture, Fisheries and Food, whose liaisons with industry suddenly seemed too close for comfort. More importantly, that episode helped set the stage for a significantly wider debate on public input into risk governance, as a result of which Britain emerged as a leader in current European initiatives on public engagement in technical decisions.48 This proved to be one of those constitutional moments during which rights and expectations of European and Member State citizenship were reconfigured, with wide implications for the ways in which uncertainty would be acknowledged and represented in subsequent risk decision making.

VII. Conclusion In 1990, when President George HW Bush spoke of turning to science for ‘sound, straightforward advice’, the idea of ‘sound science’ was not yet the buzzword that it has since become. In intervening decades, the phrase has taken on a life of its own and is frequently invoked by politicians the world over as the standard to which public reason must be held, much as political power is held accountable to justice. Yet, science’s role vis-à-vis traditional constitutional norms, such as the separation of powers and the rights and obligations of citizenship, has remained largely unexamined in mainstream legal and political thought. Empirical political studies rarely delve into the rules of the game by which political institutions defer to expert judgement; nor does political theory grapple with the conditions under which expert opinion appropriately depoliticises issues or renders political bargaining moot. Yet tacit delegations of power to science and expertise are continually in play at every level of modern government—whether in the European Commission or in EU Member States, at federal and state levels in the United States, or even in municipalities asserting their police powers to keep hazards out.49

48 J Chilvers and M Kearnes (eds), Remaking Participation: Science, Environment and Emergent Publics (Abingdon, Routledge, 2016). 49 A case in point is Wallach v Town of Dryden, 16 NE 3d 1188 (NY 2014), allowing a small town in upstate New York to block hydraulic fracturing (‘hydrofracking’) for shale oil and gas. The New York Court of Appeals, the state’s highest court, upheld the tiny township’s zoning ordinance, overriding federal claims of preemption. The case can be read as a straightforward interpretation of municipal zoning powers, especially in relation to federal land use law. See, for example, ‘Wallach v Town of Dryden’ (2015) 128 (5) Harvard Law Review 1542–51. At the same time, the decision in effect upholds a kind of epistemic subsidiarity, holding that Dryden’s determination of what serves the health and safety needs of its residents can take precedence over the federal government’s attempts to promote national economic welfare—at least when the federal provision does not expressly preempt the locality’s judgments regarding safety.

36

Sheila Jasanoff

Attentiveness to these dimensions of governance reveals the extent to which s cience serves as an alternate reservoir of power, not subject to the modes of legitimation that govern the legislative, executive or judicial branches of a classic, tripartite national government. The presumed autonomy of science, however, is itself a social and political achievement. It rests on choices that society makes about such issues as when to delegate decisions to experts, when to accept technical standards as binding, and when to give up local preferences in the interests of achieving translocal uniformity. The authority of science and expertise is in these ways intimately bound up with understandings of how to achieve just and well-working societies. A closer look at risk governance practices—in particular, delegation, harmonisation and subsidiarity—shows that neither sound science nor social justice operates as a transcendent, self-executing norm, nor independently of one another. Both are achieved by human actors and institutions in accordance with historically and culturally situated conventions that jointly determine what counts as both reasonable and just in a political order. The case studies and detailed research brought together in this collection shed light on that constant co-production of knowledge and norms—of the subject matter to be governed and the subjects to be protected—in contemporary political regimes. These analyses shed light on the myriad ways in which recognising which risks to guard against calls for simultaneous judgements about who should exercise political power, on whose behalf, under what constraints, and to what ends. Risk and the EU, on this account, do not exist in non-intersecting worlds of the natural and the social; rather, each is a vehicle through which the other’s validity is constituted. Demystifying these processes marks an important step forward in the critique of modernity and in the further development of legal and democratic theory.

3 Expert Executive Power, Administrative Constitutionalism and Co-production: Why They Matter ELIZABETH FISHER1

I. Introduction The literature on the co-production of expertise, executive power, and the physical environment is a fabulously rich one.2 The insights it provides are in many ways not new to lawyers and legal scholars who learn from early on that law creates frames for identifying problems and what should be done about them. The co-production literature highlights that these frames are not fixed and will be shaped by cultures.3 For me, co-production makes clear that an essential part of ensuring legitimate governance is recognising that there are choices to be made, and thus a discussion to be had, about how we understand what we want to govern and how we govern it.4 Again, these issues are not necessarily alien to lawyers, particularly administrative lawyers.5 In many ways, the concept of co-production captures the essence of legal reasoning.6 Right from the start of legal education, 1 I would like to thank the editors of this collection and the attendees at the associated workshop for their thoughtful and constructive comments. All errors and omissions remain my own. 2 A good starting point for the literature is S Jasanoff, ‘The Idiom of Co-Production’ in S Jasanoff (ed), States of Knowledge: The Co-Production of Science and Social Order (London, Routledge, 2006). See also Expert Group on Science and Governance, Taking European Knowledge Society Seriously (Brussels, European Commission, 2007). 3 S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005) and S Jasanoff, ‘Epistemic Subsidiarity—Coexistence, Cosmopolitianism, Constitutionalism’ (2013) 4 European Journal of Risk Regulation 133. 4 E Fisher, ‘Chemicals as Regulatory Objects’ (2014) 23 Review of European, Comparative and International Environmental Law 163. 5 The literature is a rich one. C Harlow and R Rawlings, Law and Administration (Cambridge, Cambridge University Press, 2009); G Frug, ‘The Ideology of Bureaucracy in American Law’ (1984) 97 Harvard Law Review 1276; and R Stewart, ‘The Reformation of American Administrative Law’ (1975) 88 Harvard Law Review 1661. 6 S Jasanoff, ‘A New Climate For Society’ (2010) 27 Theory, Culture and Society 233 and E Fisher, ‘Climate Change Litigation, Obsession and Expertise: Reflecting on the Scholarly Response to Massachusetts v EPA’ (2013) 39 Law And Policy 236.

38

Elizabeth Fisher

we are taught that law frames our understanding of the world. Craft the statute or the legal doctrine differently and we craft institutions and the problems we deal with differently as well.7 There are many ways we can think about these interrelationships. Administrative constitutionalism was my attempt to show the importance of co-production to debates about law and administrative governance.8 It was developed from my extensive engagement with the legal doctrine and administrative practice of risk regulation over many years in a number of legal cultures. What that experience taught me was surprising. The first was that while expertise is central to debates about risk regulation, expertise does not exist in a vacuum. Discussions about expertise in the risk regulation field are nearly always discussions about the deployment of expertise by some form of executive. In this context, expertise is never just a tool. It is embedded in executive spaces and transforms executive power. For this reason I use the phrase ‘expert executive power’ in this chapter. It is an unwieldy term, a clunky descriptor that is hard to get your tongue around. But its inelegance reveals why expertise raises questions about the nature of the legal order. The second surprise is that I started my research with the expectation that I was looking for the ‘best model’ of risk regulation. My conclusion, however, was that a sign of a legitimate political and legal order is where no one model of expert executive power dominates. In the legal cultures I studied what I found was not the linear progression towards a utopia but a steady oscillation between two paradigms of administrative constitutionalism—the deliberative-constitutive (DC) and rational instrumental (RI). Rooted in each are different models of expert executive power. These will be explained in more detail below but here it is important to note that the existence of this discourse over administrative constitutionalism is thus a signifier of a healthy and vibrant polity. It is a reminder that accountability stabilises through destabilising,9 specifically through enabling debate over the standards to which we hold decision-makers to account by and showing how those standards relate to how we constitute institutions. Civic solidarity and a demos are fostered not just through a debate about issues but about how those issues are governed and framed. This chapter is about the need to recognise the importance of having debates about administrative constitutionalism (and thus recognising co-production) in regard to expert executive power in the EU. Or to put the matter another way, debates about expert executive power need to be connected more closely to debates about the nature and legitimacy of the EU legal order.

7 Which is why the Chevron doctrine in the US is so significant. See S Shapiro and E Fisher, ‘Chevron and the Legitmacy of Public Administration ’ (2013) 22 William and Mary Bill of Rights Law Journal 465. 8 E Fisher, Risk Regulation and Administrative Constitutionalism (Oxford, Hart Publishing, 2007). 9 E Fisher, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495, 513.

Administrative Constitutionalism

39

The structure of my analysis is as follows. First, I consider in more detail how expert executive power leads to debates about administrative constitutionalism in most legal cultures. This has not really been the case in the EU. Debate only emerged in the 1990s, but by the early 2000s was already dominated by what is described in the introduction to this collection as the ‘epistemic model’ and I would describe as the RI model of administrative constitutionalism. In the second section, I show why this is problematic. My focal point for this analysis is somewhat counterintuitive: Helen Macdonald’s award winning memoir/literary biography, H is for Hawk.10 I use her work to focus on two features of expertise—the ways in which it is gained and the ways in which it is judged. In regard to the former, expertise is something that needs to be fostered over time in an institutional context. The RI paradigm not only bears little relationship to the descriptive landscape of the EU, but it also promotes an idea of expertise as timeless. In regard to the latter, implicit in understandings of expertise are yardsticks by which expertise can be judged and held to account by. Such yardsticks are a primary manifestation of administrative constitutionalism, and it is important to be having a multi-dimensional debate about them. The dominance of the RI paradigm means that this is not occurring despite the fact that expertise has played an important role in European integration. Four caveats are required before starting. First, my focus is on the regulation of health and safety risks (risk regulation) but it needs to be recognised that neither expertise nor administrative constitutionalism are limited to that area. Indeed, as recent events show, the EU’s response to the financial crisis highlights more than anything else in the last decade the need to debate administrative constitutionalism.11 Second, this chapter is a sketch. Much more can and should be said about many of the issues touched on here. Third, I am exploring concepts of co-production as a legal scholar. But as stressed above, what is striking to me about co-production is that it is a concept that sits comfortably with many strands of mainstream legal scholarship.12 Finally, this chapter was written before the EU referendum in the UK, but I sit here editing it in the aftermath of that referendum’s result. As I explain in the conclusion, that result makes the conclusions of this chapter even more important. 10 H Macdonald, H is for Hawk (London, Jonathan Cape, 2014). Or perhaps it is not counterintuitive. MacDonald has a scholarly background in the history and philosophy of science. 11 C 62/14 Gauweiler and Others v Deutscher Bundestag (16 June 2015). The literature is a big one. Note in particular the prevalence of the idea of ‘rethinking’ in it. G Majone, Rethinking the Union of Europe Post-Crisis (Cambridge, Cambridge University Press, 2014); F Scharpf, ‘After the Crash: A Perspective on Multilevel European Democracy’ (2015) 21 European Law Journal 384; J Habermas, The Lure of Technocracy (Cambridge, Polity Press, 2015); S Champeua, C Closa, D Innerarity and MP Maduro (eds), The Future of Europe: Democracy, Legitimacy and Justice After the Euro Crisis (London, Rowan & Littlefield International 2015). 12 E Fisher, ‘Making Sense of the WTO Sanitary and Phyto Sanitary Agreement: An Essay About Scholarly Expertise’ in B Lange, F Haines and D Thomas (eds), Regulatory Transformations: Rethinking Economy-Society Interactions (Oxford, Hart Publishing, 2015).

40

Elizabeth Fisher

II. Expert Executive Power and Administrative Constitutionalism and the EU Expert executive power plays an important role in the regulating of environmental and health risks. The EU is no exception. Inherent in the EU’s evolution has been the development of a sphere of regulatory activity concerned with protecting the environment and human health. The core of this activity is a set of laws brought about through positive harmonisation and can be broadly described as an area of ‘risk regulation’. That sentence immediately oversimplifies, however. The legal basis for such harmonisation varies,13 reflecting the fact that risk regulation is a product of both neo-functional spill-over from market integration and the outcome of explicit decisions to protect the environmental and public health.14 In regard to the latter, ‘risk regulation’ can be seen as part of the ‘normative power’ of the EU.15 Positive harmonisation needs to be seen in light of negative harmonisation that comes about through the operation of the fundamental freedoms.16 Furthermore, most ‘risk regulation’ laws are directives and their legal operation requires a significant role for Member States. It is also the case that such directives are often complemented with regulations and also create processes of ‘technical progress’ in light of new information.17 Before considering the EU risk regulation landscape in more detail, let me make a few general points about expertise in the risk regulation context. That role has two dimensions, neither of which is fixed. First, its role has a practical dimension—specific knowledge, skills and experience are needed to understand risks and how to evaluate them.18 But what exactly that knowledge is, those skills are, and that experience is could be many things, and these are closely connected to what is understood to be the evidentiary basis for decision-making. In risk regulation there are two types of expertise closely intertwined to the point they are almost indistinguishable—the expertise in evaluating risk and the expertise in governing. The close interrelationship between these two types of expertise is a reminder of the distinctive nature of expert executive power. It is not just the 13

eg TFEU, arts 114,192, 352. See also TFEU, art 169(2)(1). the 1970 Vehicle Emissions directive with the 1975 Wild Birds directive: Council Directive 70/220/EEC of 20 March 1970 on the approximation of the laws of the Member States on measures to be taken against air pollution by emissions from motor vehicles [1970] OJ L76/1 and Council Directive 79/409/EEC of 2 April 1979 on the conservation of wild birds [1979] OJ L103/1 (now replaced with a codified directive). 15 D Kelemen, ‘Globalizing European Union Environmental Policy’ (2010) 17 Journal of European Public Policy 335. For a discussion of normative power see E Keene, ‘Social Status, Social Closure and the Idea of Europe as a “Normative Power”’ (2013) 19 European Journal of International Relations 939. 16 See s III below, but for a general overview see K Armstrong, Regulation, Deregulation, Reregulation (London, Kogan Page, 2000). 17 This of course being a cause for the historical rise of comitology. 18 Above n 8 at 20. 14 Compare

Administrative Constitutionalism

41

expertise of knowing about risks that is required in risk regulation, but what is also needed is expertise in how to govern those risks.19 The second dimension of the role of expertise is a normative one. Expertise is a source of authority, and in particular legitimate authority.20 As such, the deployment of expertise is intricately tied to questions about governance and in being so it is perhaps not surprising that expertise is understood in relative terms as connoting one institution or person that has greater knowledge, skills or experience, and thus greater authority than another. Claims to expertise are thus normative claims to ‘governing authority’. These two dimensions manifest themselves in the operation of expertise. Expertise is not a free floating ‘thing’ but is a concept that only gains substantive meaning embedded in institutions (nearly always administrative)21 and social practices that delineate what knowledge, skills and experience count as expertise. This is again why the focus of this chapter is upon expert executive power. That process of delineation will be shaped by normative assumptions in that legal and socio-political culture about legitimate expert governance. Co-production makes clear that those assumptions are not just about the political world, but also the physical world.22 Moreover, the opaque nature of expert discretion means that claims to expert authority in any particular circumstance rely on the nature of the institution for authority and less on specific reasons. All of this explains why in national legal cultures the rise of expert public administration has been accompanied by debates over administrative constitutionalism. As noted above, administrative constitutionalism refers to the way in which public administration is constituted, limited and held to account on the basis of assumptions about its role and nature.23 These debates are legal because law plays an important (albeit not the only) role in constituting, limiting and holding institutions to account. They are also a mixture (not always an easy one) of theory and doctrine and reflect legal and socio-political cultures in any particular jurisdiction. This results in quite different administrative law regimes in different legal cultures. It is also the case that ‘nations differ systematically in their strategies of public reasoning’.24 Thus in the US there has been quite an explicit debate over the role and nature of the administrative state occurring primarily in legal terms.25 The key catalyst points in those debates have been the development of expert administrative capacity

19 This is implicit in Jasanoff ’s important idea of ‘regulatory science’. See S Jasanoff, The Fifth Branch: Science Advisers as Policy Makers (Cambridge MA, Harvard University Press, 1990) 77. 20 D Estlund, Democratic Authority: A Philosophical Framework (Princeton, Princeton University Press, 2008) 3. 21 Above n 8 at 19–21. 22 Jasanoff, above n 2. 23 Above n 8 at 24. 24 Jasanoff, ‘Epistemic Subsidiarity’, above n 3 at 134. 25 S Shapiro, E Fisher and W Wagner, ‘The Enlightenment of Administrative Law: Looking Inside the Agency for Legitimacy’ (2012) 47 Wake Forest Law Review 463.

42

Elizabeth Fisher

during the New Deal and with new social regulation in the 1970s. The Administrative Procedure Act 1946, judicial review doctrine, and the legal structures of institutions and hybrid rulemaking are all products of those debates. The situation in the UK has been very different. The rise of expert public administration in the mid-nineteenth century26 and then again in the 1920s,27 led to a retreat from legal formalism and expertise being disembedded from bureaucratic structures. Debates over expert executive power occurred, but in non-legal forums.28 With that said, and as already noted, these debates can be seen to pivot around two distinctive paradigms of administrative constitutionalism.29 The DC paradigm envisages expert executive power as having substantive, flexible and ongoing discretionary power to address risks. In contrast under the RI paradigm, expert executive power is a tool of the legislative and its discretionary power is limited. I am not interested in this chapter about the competing virtues of these p aradigms. What is important is that in these legal cultures there is not a single model of administrative constitutionalism. Rather there is an ongoing debate about what is ‘good administration’. In light of this experience in other legal cultures, one would expect that the creation of EU regulatory regimes to regulate health and environmental risks, embedded in which is expert authority, would lead to debates over administrative constitutionalism throughout the EU. That has not really been the case and rather EU discourse has lurched through three rough phases of risk regulation. The first, up until the early 1990s, was a period in which there was little discussion of these issues. Any structures that existed were still in the process of developing, were invisible, and relied heavily on Member States for their operation. More importantly, the legitimacy of the EU until 198730 was firmly grounded in Member State consent, and thus there was no perceived need to have a discourse about EU executive power. If there were assumptions about ‘good’ decision-making in this area they were implicit and appeared in ad hoc cases. They also related more to the legitimacy of the Council31 or Member States acting in justified contravention of the fundamental freedoms.32 The second phase was during the 1990s. The introduction of the principle of conferral33 in the Maastricht Treaty made clear that the EU was required to act

26 C Hamlin, Public Health and Social Justice in the Age of Chadwick, Britain 1800–1854 (Cambridge, Cambridge University Press, 1998); R Macleod (ed), Government and Expertise: Specialists, Administrators, and Professionals 1860–1919 (Cambridge, Cambridge University Press, 1988). 27 A Zimmern, ‘Democracy and the Expert’ (1930) 1 Political Quarterly 7. 28 For a brief account see above n 8 at 66–71. 29 For a longer description see ibid, ch 2. 30 The Single European Act introduced Qualified Majority Voting (QMV) for the newly created ‘internal market’ competence, TEC, art 100a (now TFEU, art 114). 31 Case C-331/88 R v Minister of Agriculture, Fisheries and Food and Secretary of State for Health, ex parte FEDESA [1990] ECR I-4023. 32 Case 174/82 Sandoz BV [1983] ECR 2445. 33 Now TFEU, arts 5(1) and (2).

Administrative Constitutionalism

43

within its powers.34 Discussion then began to take on a ‘constitutional’ aspect. There was concern about the ‘demos’35 and the overall legitimacy of the EU.36 The Single European Act and Maastricht saw the introduction of more explicit EU risk regulation competences37 and this area of activity became more intense and the object of study. This was catalysed by controversies both inside (the BSE crisis) and outside (the SPS Beef Hormones dispute) the EU. By the start of the millennium a vigorous scholarly and policy debate was beginning to emerge about the nature of expert executive power in the EU.38 But as I have commented elsewhere,39 in recent years there has been a narrowing of the policy and some of the scholarly imagination in regard to risk regulation in the EU. The starting date of this narrowing is in the early to middle years of the new millennium and this represents the third phase. I should stress I am not making a monolithic statement about the literature and exceptions to what I describe below are easily identifiable.40 During this phase scholarly and policy discussion about EU ‘risk regulation’ has become ‘kettled’ into a separate discourse from the mainstream debates about the EU. And what increasingly frames and dominates that discourse is the assumptions that regulating health and environmental protection is on the basis of a quantitative concept of risk and the linear framework of risk assessment and risk management are taken as givens. I am not the only person to note this.41 Implicit in these frameworks is a very narrow way of understanding environmental and public health problems, and understanding the role and nature of regulatory institutions and expertise. This process of co-production results in promotion of the RI paradigm of administrative constitutionalism.42 That paradigm, like all paradigms of administrative constitutionalism, has created particular ‘songlines’ of governance,43 which have been well charted by others.44 34 Compare that to what is now TEU, art 13 and was in the Treaty before hand—the statement that only particular bodies could exercise power. 35 J Weiler, The Constitution of Europe: Do the New Clothes Have an Emperor? (Cambridge, Cambridge University Press, 1999). 36 D Beetham and C Lord, Legitimacy and the European Union (London, Longmans, 1998); A Arnull and D Wincott (eds), Accountability and Legitimacy in the European Union (Oxford, Oxford University Press, 2002). 37 See in particular TFEU, arts 114(3), 192. 38 For example C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 2000); C Joerges, ‘Deliberative Supranationalism: Two Defences’ (2002) 8 European Law Journal 133; C Joerges, Y Mény and J Weiler (eds), Mountain Or MoleHill: A Critical Appraisal of the Commission White Paper on Governance (Florence, European University Institute, 2002). 39 E Fisher, ‘Framing Risk Regulation: A Critical Reflection’ (2013) 4 European Journal of Risk Regulation 125. 40 Most obviously the impressive analysis in C Anderson, ‘Contrasting Models of EU Administration in Judicial Review of Risk Regulation’ (2014) 51 Common Market Law Review 425. 41 V Heyvaert, ‘Governing Climate Change: Towards a New Paradigm for Risk Regulation’ (2011) 74 Modern Law Review 817 and M Lee, EU Environmental Law, Governance and Decision-Making 2nd edn (Oxford, Hart, 2014) ch 2. 42 For a longer discussion of that model see above n 8 at 28–30. 43 S Jasanoff, ‘The Songlines of Risk’ (1999) 8 Environmental Values 135. 44 Above n 41.

44

Elizabeth Fisher

It is also important to remember it is both a paradigm of risk evaluation and legitimate governance. For my purposes three aspects of that paradigm are important to emphasise. First, it presumes expert executive power is monolithic in nature. The classic example of the RI paradigm is an independent ‘expert’ agency that applies the facts to the legislative mandate through the application of specific analytical methodologies. Second, this understanding of executive power characterises such an agency as the robotic agent of the legislature. Third, this characterisation is appealing to those that fear the unchecked discretion of experts. The RI model is one response to the democratic accountability problems created by public administration in a polity committed to democratic constitutionalism. As that last line makes clear, the RI paradigm is a response to a particular governance dilemma in liberal democracies. Whether it is a ‘good’ response to those dilemmas in those democracies is one issue,45 but what I want to focus on is a different point here—that the acceptance of the RI paradigm in the EU is problematic because it is too quickly closing down debate on administrative constitutionalism. Saying this is not to say its promotion is not problematic for other reasons, but it is the lack of debates over administrative constitutionalism that I want to focus on. There are many ways I could do this. The approach here will feel tangential, particularly as the focus is upon expertise and not expert executive power, but I hope it will begin to show why debating administrative constitutionalism is so important in the EU.

III. Goshawks and Expertise This tangential approach is inspired by a statement of Helen MacDonald in H is for Hawk.46 That book is many things: a memoir about Macdonald grieving for her father; a story about her training a goshawk to deal with that grief; a literary biography of the author TH White who also trained a goshawk and wrote about the experience; and a social and cultural history of training goshawks. To put the matter another way, the book is about MacDonald’s own attempts to master a skill and exploring how that skill is embedded in a set of long held cultural practices. One statement in the book leapt out at me. MacDonald, in a passage discussing TH White’s anxieties over developing his own skills of training a goshawk, noted: Being a novice is safe. When you are learning how to do something, you do not have to worry about whether or not you are good at it. But when you have done something, have learned how to do it, you are not safe anymore. Being an expert opens you up to judgement.47 45 Compare T McGarity, ‘A Cost Benefit State’ (1998) 50 Administrative Law Review 7 and C Sunstein, The Cost Benefit State: The Future of Regulatory Protection (Washington DC, American Bar Association, 2002). 46 Above n 10. 47 ibid 146.

Administrative Constitutionalism

45

Now I admit goshawk training and the regulation of risk in the EU have little to do with each other, but this statement highlighted two things about expertise that I thought interesting and constructive—the temporal aspect of expertise and the way in which claims of expertise are ‘open to judgement’. The first of these aspects tends to be ignored and the second may appear paradoxical. Yet taken in tandem, they provide an avenue into understanding why having debates about administrative constitutionalism in the EU really matters and why the dominance of the RI paradigm is problematic.

A. The Temporal Aspects of Expertise Macdonald in her comment is highlighting that expertise is something that is gained by learning over time. It accrues in one person or institution where there are frames for learning and assessing expertise. I would describe this as the temporal aspect of expertise. Milestones that are reached through a process of learning mark the gaining of expertise. Those milestones do not appear out of thin air. They develop over time and become part of formal and informal learning processes that are embedded in a range of institutions. The expertise gained is a mixture of both the explicit and the tacit.48 Implicit in claims to expertise are thus understandings about how such expertise is gained (or normally gained). This is particularly the case in relation to expert executive power where, as noted above, the required expertise incorporates the expertise in governing. Thus behind claims to the expertise of the French public administration is an account of how they were educated. And the same is true historically, of the ‘generalist expertise’ of the British civil service.49 This link between education and understandings of expertise is one of the reasons why understandings of scientific expertise vary from culture to culture.50 It is also why a study of ‘regulatory science’ is so important. Expert executive power is not the same as research science, it has developed over time, and is shaped by legislative and institutional mandates.51 In work I recently carried out with Pasky Pascual and Wendy Wagner, we showed how understandings of the US Environmental Protection Agency’s expertise has been developing over 45 years due to developments created by ‘inside out’ and ‘outside in’ accountability.52 To put the matter another way, processes of learning are a dimension of administrative constitutionalism, and an important one. Those processes may come about in many different ways.

48

H Collins, Tacit and Explicit Knowledge (Chicago, University of Chicago Press, 2010). Porter, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton, Princeton University Press, 1995). 50 On that variation see Jasanoff, ‘Epistemic Subsidiarity’ above n 3. 51 Above n 19. 52 E Fisher, P Pascual and W Wagner, ‘Rethinking Judicial Review of Expert Agencies’ (2015) 93 Texas Law Review 1681. 49 T

46

Elizabeth Fisher

I am not talking only about formal education (and nor was Macdonald) but also about the learning created by the experience of governing. If we take administrative constitutionalism and co-production seriously, then processes of learning are important. Thus as the Expert Group on Science and Governance has noted: Learning from experience, including mistakes, is recognized as essential to a mature and robust knowledge society, not only for the instrumental reasons of avoiding h armful mistakes, but also for the democratically- accepted legitimacy of our institutions of policy and politics, and of the science which shapes these.53

There is a debate to be had about what type of learning is good, but that debate is a debate about administrative constitutionalism. The problem in the EU is that the promotion of the RI paradigm has neither been on the basis of such a debate or upon a process of learning. This can be seen most obviously by comparing the description of the RI paradigm above with a description of EU governance. The EU is a sui generis governance regime. It does not accord with the separation of powers. It is not a federal state. It has evolved from being grounded in international relations to becoming a ‘constitutional’ regime, but not a democratic one, and one in which its international relations’ DNA remains. The process of development has been incremental and messy. It has also defied easy explanation, particularly its ‘executive’ aspects.54 As Trondal has noted the European executive order ‘consists of a compound set of dynamics’ that ‘transforms an inherent Westphalian order to the extent that an intergovernmental dynamic is supplemented by different mixes of supranational, departmental and/or epistemic dynamics’.55 The end result is that in the EU ‘executive space’ is an interconnected web of national (and sub national) administrations, committees of different forms, Union institutions, and increasingly agencies. A study of decision-making processes also makes clear that such processes are a mixture of intergovernmental processes and classic administrative processes, very few of which take place within the boundaries of a single institution. More significantly it is very hard to identify any long-standing sites for the fostering of EU expert executive power in regard to risk regulation. All this stands in stark contrast with the ‘agent’ nature of public administration under the RI paradigm. Communications and White Papers cast expertise in RI terms but anyone who has ever had to do battle with the decision-making processes under a particular Directive finds it very hard to identify such instrumental

53

Expert Group on Science and Governance, above n 2 at 64. That is both descriptively and normatively. See P Craig, EU Administrative Law 2nd edn (Oxford, Oxford University Press, 2012); P Lindseth, Power and Legitmacy: Reconciling Europe and the Nation State (Oxford, Oxford University Press, 2010); C Harlow and R Rawlings, Process and Procedure in EU Administration (Oxford, Hart Publishing 2015). 55 J Trondal, An Emergent EU Executive Order (Oxford, Oxford University Press, 2010) 2. 54

Administrative Constitutionalism

47

understandings of power. Pfizer is a classic case in point.56 There was no legal duty under the Directive to follow the recommendations of the Scientific Committee on Animal Nutrition and the process was driven by Member State action. Likewise, the UK judicial review difference of opinion of the Divisional Court and Court of Appeal in Downs concerning the type of pesticide exposure model that the UK government utilised in the operation of Directive 91/414/EEC comes about because of the substantive exercise of power that Member States have under that Directive.57 Furthermore, the power of EU agencies does not fit easily into this matrix on a RI basis.58 Moreover, a limit to developing a robust approach to developing EU executive power is that the EU executive space is not an easily visible set of structures and as shown above it is one in constant flux. Vos’ ground-breaking book on Institutional Frameworks of Community Health and Safety Regulation59 was ground-breaking precisely because it charted a landscape that had been invisible to many. But since then that landscape has continued to evolve, most significantly with the Lisbon reforms to delegated power.60 And much of the institutional architecture is still in the process of being built and developed. As AG Jaaskinen noted in UK v Parliament and Council: ‘[t]the Lisbon Treaty clearly maps out the scheme for judicial review of laws and decisions made by agencies, the Treaty is more enigmatic when it comes to delimiting the powers of agencies’.61 The problems with a lack of a framework for fostering EU expert executive power have been hidden by policy and scholarly accounts of EU risk regulation that present a streamlined RI account of the expertise required in risk regulation. The risk assessment/risk management divide is one example of this. That divide has seemingly become an entrenched feature of risk regulation, but that is only a relatively recent phenomenon (as is the case in other jurisdictions). The 1997 Communication on Consumer Health and Food Safety set out the principle for a ‘new political departure’ that would result in a ‘proper food policy’.62 Risk assessment and risk management figured in a description of risk analysis alongside risk communication.63 The Communication emphasised that all three activities would be carried out by the Commission aided by a flexible structure of different committees. Risk analysis is the approach ‘to be followed’.64 It also noted that 56 E Fisher, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19. 57 Downs v Secretary of State for Environment Food and Rural Affairs [2008] EWHC 2666 and Secretary of State for Environment, Food and Rural Affairs v Downs [2009] EWCA Civ 664. 58 See the discussion about the role of EFSA in the TFEU, art 95(4) (now art 114(5)) and Case C-439/05 Land Oberosterreich v Commission of the European Communities [2007] ECR I-7141. 59 E Vos, Institutional Frameworks of Community Health and Safety Legislation: Committees, Agencies and Private Bodies (Oxford, Hart Publishing, 1999). 60 See TFEU art 290 in particular and for a general overview P Craig, The Lisbon Treaty: Law Politics and Treaty Reform (Oxford, Oxford University Press, 2010) ch 3. 61 Case C-270/12 UK v Parliament and Council (CJEU, 22 Jan 2014) para 75. 62 Commission, Communication on Consumer Health and Food Safety COM (1997) 183 final, 1. 63 ibid 19. 64 ibid 4.

48

Elizabeth Fisher

‘[s]cientific advice, risk analysis and control need to be used for the basis action’65 in specific areas but ‘[t]he depth of the risk analysis undertaken will reflect the nature of the hazard identified’.66 Three years later the risk assessment/risk management divide was taking on a more significant role. The White Paper on Food Safety in 2000 stated that: Risk analysis must form the foundation on which food safety policy is based. The EU must base its food policy on the application of the three components of risk analysis: risk assessment (scientific advice and information analysis) risk management (regulation and control) and risk communication.67

The White Paper went on to propose the European Food Safety Authority (now EFSA) and noted, ‘[t]he role of an Authority must be defined in the context of the process of risk analysis, which comprises risk assessment, risk management and risk communication’.68 Expert executive power was thus being framed in terms of a policy framework, a framework that originated in the US.69 It is not a framework that directly speaks to the fundamental issue in risk evaluation—drawing valid inferences from data.70 To put the matter another way it makes invisible the skills, experience and knowledge that need to be developed to make a claim of expertise. This can also be seen in regard to the utilisation of the risk assessment/risk management divide in the Communication on the Precautionary Principle, which characterises the application of the precautionary principle as part of risk management that occurs ideally after ‘a comprehensive assessment of risk’ or at least scientific evaluation.71 Yet as the principle is directly concerned with scientific uncertainty then the splitting off of the principle from scientific evaluation is odd.72 It is also interesting to note that in Annex III of the Communication, risk assessment is defined largely in terms of health risk assessment. But the Communication applies to all contexts in which the principle is relevant—many of which involve holistic questions that do not yield easily to sequential steps of hazard identification, hazard characterisation, exposure appraisal and risk characterisation. There is no discussion that the process of risk evaluation (and thus the assessment

65

ibid 8. ibid 19. 67 Commission, White Paper on Food Safety COM (1999) 719 final, 9. 68 ibid 14. 69 E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006). 70 P Pascual, W Wagner and E Fisher, ‘Making Methods Visible: Improving the Quality of Science Based Regulation’ (2013) 2 Michigan Journal of Environmental and Administrative Law 429. 71 Commission, Communication from the Commission on the Precautionary Principle COM (2000) 1 final, 14. 72 For a discussion of precaution within the context of thinking about scientific analysis see R von Schomberg, ‘The Precautionary Principle and its Normative Challenges’ in E Fisher, J Jones and R von Schomberg (eds), Applying the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006) and A Stirling, ‘Science, Precaution, and the Politics of Technological Risk: Converging Implications in Evolutionary and Social Scientific Perspectives’ (2008) 1128 Annals of the New York Academy of Sciences 95. 66

Administrative Constitutionalism

49

of risk) will need to be adapted to a context.73 These are just some of the reasons why I find the Communication, even on its own RI terms,74 deeply confusing. But the Communication is an exercise in expertise building and has as one of its purposes ‘build[ing] a common understanding of how to assess, appraise, manage and communicate risks that science is not yet able to evaluate fully’.75 Another example of the way in which expertise is taken as a given can be seen in discussions about expertise in EU policy documents in the early 2000s. For instance, in the White Paper on European Governance the emphasis was on boosting confidence in the way expertise influenced policy decisions through explaining how expertise was being deployed.76 But the assumption was that the role and nature of expertise were fixed and what was needed was greater explanation of how it was being used. The ensuing Communication ‘On the Collection and Use of Expertise by the Commission: Principles and Guidelines’ has a similar approach. Expertise is something to be utilised and ‘exploited’.77 The issue is the ‘access to the right expertise at the right time’.78 And while changing context is recognised and the fact that expertise can refer to different types of knowledge and experience,79 there is not an appreciation that expertise is a malleable source of authority. In the Communication on ‘The Operating Framework for the European Regulatory Agencies’ the role of agencies is described as follows: The independence of their [the agencies’] technical and/or scientific assessments is, in fact, their real raison d’être. The main advantage of using the agencies is that their decisions are based on purely technical evaluations of very high quality and are not influenced by political or contingent considerations.80

Agencies are also being cast in RI terms. In all cases the expert executive power in relation to risk regulation is a clearly identifiable thing. There is no need to discuss how such expertise came to be, and what fosters and sustains it. There is no temporal aspect to expertise—it just is. This depiction of the expertise in risk regulation as something perpetual is reinforced in the academic literature by accounts of EU regulation in ‘functionalist terms’. A prime example is Majone’s famous phrase in 1996 that the EU is a ‘regulatory state’, by which he meant it had a significant ‘regulatory function’ concerned with ‘attempts to increase the allocative efficiency of the market by correcting

73 Although note that the Communication also states that the precautionary principle need only be applied after scientific evaluation. 74 E Fisher and R Harding, ‘The Precautionary Principle and Administrative Constitutionalism: The Development of Frameworks for Applying the Precautionary Principle’ in Fisher, Jones and von Schomberg (eds), above n 72. 75 Above n 71 at 3. 76 ibid 3. 77 ibid 3, 4. 78 ibid 3. 79 ibid 6. 80 Commission, The Operating Framework for the European Regulatory Agencies COM (2002) 718 final, 5.

50

Elizabeth Fisher

various types of market failure’.81 What strikes me is how little of Majone’s argument is actually grounded in an empirical analysis of the legal frameworks of the EU. Much of his discussion in Regulating Europe is not about Europe at all, but about the growth of the US administrative state. It is however based on a presumption that ‘[s]tatutory regulatory by independent agencies—sometimes called, for historical reasons, ‘American style regulation’—is rapidly becoming the most important mode of regulation, indeed the leading edge of public policy making in Europe’.82 That was not descriptively the case in 1996 (the EFSA, the closest approximation to such a body, did not exist) and nor is that statement descriptively correct now. But what it, and other accounts, suggests, is that expertise is an ‘immutable mobile’ that can be easily transferred from one cultural context to another.83 What is clearer, particularly in later work, is that Majone’s functionalist vision is an RI one.84 It is not that Majone does not appreciate the complexity of the EU, and he is also critical of the risk assessment/risk management divide,85 but he views risk regulation through a particular lens, one heavily influenced by ‘principal agent’ theory. It is a normative account, not a descriptive one.86 But that of course is a feature of most accounts of expert executive power. Moreover, his focus is really upon promoting accountability and control.87

B. ‘Opening Up to Judgement’ The normative nature of Majone’s ‘regulatory state’ thesis and its influence on how we see the world is a reminder that discussions about expert executive power, whether in the EU or anywhere else, are always discussions about is and ought. They are ‘pictures of how our political world is organised’ and they shape not just ‘what we expect and what we see’ but also ‘what we demand and we affirm’.88 As seen in the last section the way in which expert executive power is currently characterised ‘disconnect[s] our knowledge, experience and imaginations from history’.89 Pictures which are not only normative but also performative.90 They act as blueprints for judging legitimate action and that returns me to the second and 81

G Majone, Regulating Europe (London, Routledge, 1996) 54. ibid 47. 83 B Latour, ‘Visualisation and Cognition: Drawing Things Together’ in H Kuklick (ed), Knowledge and Society Studies in the Sociology of Culture Past and Present, Volume 6 (Greenwich, JAI Press, 1986) and E Fisher, ‘Risk and Governance’ in D Levi-Faur (ed), Oxford Handbook of Governance (Oxford, Oxford University Press, 2012). 84 G Majone, Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth (Oxford, Oxford University Press, 2005). 85 G Majone, ‘Foundations of Risk Regulation: Science, Decision-Making, Policy Learning and Institutional Reform’ (2010) 1 European Journal of Risk Regulation 5, 18. 86 See in particular G Majone, ‘The Regulatory State and its Legitimacy Problems’ (1999) 22 West European Politics 1. 87 Above n 84 and n 86. 88 J Mashaw, Greed, Chaos and Governance: Using Public Choice to Improve Public Law (New Haven, Yale University Press, 1997) 1. 89 Expert Group on Science and Governance, above n 1 at 79. 90 ibid 75. 82

Administrative Constitutionalism

51

more striking aspect of Macdonald’s statement. That is: ‘Being an expert opens you up to judgement’. Such a statement is counterintuitive for many working in risk regulation scholarship. In this field, claims to expertise are often understood as closing down opportunities for judgement not opening them up. This is correct on one level. If expertise acts as a source of authority in a polity relative to other claims then claims based on expertise will indeed trump those other claims, but those expert claims will be difficult to verify.91 But MacDonald’s statement is making clear that claims to expertise are subject to some form of judgement.92 Expertise needs to accord with preordained yardsticks of what we think expertise is, and should be.93 Those yardsticks will connect with the temporal aspect of expertise. In learning how to be an expert we are learning about what makes a ‘good’ expert. They will also be the standards on which an expert will be held to account.94 Such yardsticks might operate as part of ‘inside out’ accountability operating on the basis of professional norms.95 Thus while our scholarly expertise is impenetrable and thus not easily assessed by the wider public, it is constantly ‘open to judgement’ by our peers and by our institutions. Multiple processes of judgement exist (peer review, management review etc). We may or may not agree with the judgement process and the yardsticks deployed, but we know full well we do not operate in an accountability vacuum. In a polity, ‘inside out’ accountability is not only supplemented by ‘outside in’ accountability but ‘outside in’ accountability also becomes a significant source of legitimacy.96 A major feature of administrative constitutionalism is debate over ‘outside in’ accountability and what yardsticks it should operate on the basis of. Those yardsticks will reflect a particular understanding of the role and nature of expert executive power. As those understandings shift, so do the yardsticks. Thus the promotion of the RI paradigm in regard to EU risk regulation is the promotion of a set of yardsticks to judge and thus to assess the legitimacy of expert executive power. The most obvious example of this relates to the EFSA: In order to achieve the general objective of a high level of protection of human health and life, food law shall be based on risk analysis except where this is not appropriate to the circumstances or the nature of the measure.97 91

Above n 20 at ch 1. Note here the non-adjudicative spelling. Such judgement may take place in many different forums. 93 An easy source of comedy is generated by expertise defying our expectations of what we think it is. Dr Leo Spaceman in the television show 30 Rock is a good example of this. 94 In saying this I am using ACL Davies 4 step schema of accountability—standard setting, giving an account, judging an account and consequences. See ACL Davies, Accountability: A Public Law Analysis of Government by Contract (Oxford, Oxford University Press, 2001) 81–82. 95 S Shapiro and R Wright, ‘The Future of the Administrative Presidency: Turning Administrative Law Inside-Out’ (2011) 65 University of Miami Law Review 577 and S Shapiro, ‘Why Administrative Law Misunderstands How Government Works: The Missing Institutional Analysis’ (2013) 53 Washburn Law Journal 1. 96 Which is why Majone focuses on it. Above n 86. 97 Regulation 178/2002/EC of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety [2002] OJ L31/1, art 6(1), emphasis added. 92

52

Elizabeth Fisher

Similarly, ‘[r]isk assessment shall be based on the available scientific evidence and undertaken in an independent, objective and transparent manner’.98 The Communication on the Precautionary Principle fulfils a similar role—it is not only fostering expertise, but yardsticks by which to judge that expertise by. Likewise, there have been shifts in how the CJEU and the General Court have reviewed the substantive exercise of discretion. Anderson has recently undertaken an important analysis of how different models of executive power underpin substantive review by the European courts.99 Thus for example, there are cases in which the approach is underpinned by an RI paradigm,100 but that is not the only approach he identifies. In other cases, the Court is basing its approach on expertise and not simply acting as an agent by applying facts to the rules.101 These can be thought of as closer to the DC paradigm of administrative constitutionalism. Reflecting on this second category of cases, what is striking is in how many of them the arguments put forward by litigants challenging a decision are based on RI premises often derived from policy documents and legislation.102 In some cases, the courts (often the Advocate Generals) are rebuffing these arguments with quite robust analyses firmly grounded in the EU legal order.103 In other cases the substantive review feels so broad ranging, almost like a merits review.104 This gives rise to questions about the legitimacy of courts in these cases.105 In other cases still, the judgment of the court appears to rest on a carving out of an exception from the RI paradigm. Take for example the case of Denka,106 that concerned the ‘non-inclusion’ (and thus withdrawal of authorisation) of the pesticide dichlorvos in Annex 1 of Directive 91/415 concerning the placing of plant protection products on the market.107 The reason given by the Commission, after a complicated process that involved the EFSA and Member States, was that: During the evaluation of this active substance, a number of concerns were identified. In particular, based on the available toxicological data and taking into account the uncertainties of the genotoxic and carcinogenic properties of the substance [and] also considering the overall poor quality of the dossier, it has not been demonstrated that the estimated operator, worker and bystander exposure, is acceptable.108

98

ibid, art 6(2). Above n 40. 100 The prime example he gives is Case T-229/04 Sweden v Commission [2007] ECR I-2437. 101 The example he gives is Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305. 102 eg, ibid and Case C-343/09 The Queen, on the application of Afton Chemical Limited v Secretary of State for Transport [2010] ECR I-7027. 103 See AG Kokott’s discussion of the precautionary principle as a general principle of EU in ibid. 104 Case T368/11 Polyelectrolyte Producers Group v Commission (GC, 1 Feburary 2013). This went on appeal but was dismissed. See C-199/13 Polyelectrolyte Producers Group and Others v Commission (CJEU, 27 March 2014). 105 A point also highlighted in the US context by Martin Shapiro. See M Shapiro, Who Guards the Guardians: Judicial Control of Administration (Athens, University Of Georgia Press, 1988). 106 Case T-334/07 Denka International BV v Commission [2009] ECR II-4205. 107 Council Directive 91/414/EEC of 15 July 1991 concerning the placing of plant protection products on the market [1991] OJ L230/1. 108 Above n 105 at para 34. 99

Administrative Constitutionalism

53

The applicant put forward many arguments all framed in RI terms, with the following point particularly interesting to note: The applicant submits that, according to the case-law, danger must be distinguished from risk. According to the applicant, the evaluation criteria set out in Article 5 of D irective 91/414 are based on risk assessment. The contested decision infringes that article in that it is not based on a risk assessment but only on the inherent hazard properties of dichlorvos.109

While Article 5 of Directive 91/414 sets out parameters for analysis, it does not mention risk assessment—not surprising given the era in which it was based. The Court did not accept this argument and noted the following (I quote it at length because it is striking that the Court does not directly rebut the assumption about risk assessment): In that context, it must be pointed out that the criteria set out in Article 5(1) of Directive 91/414 for a substance to be capable of being included in Annex I are framed in broad terms and based on an analysis of the risk of harmful effects on human or animal health or on groundwater or any unacceptable influence on the environment (Bayer CropScience and Others v Commission, paragraph 92 above, paragraph 184). Furthermore, that provision must be interpreted in conjunction with the precautionary principle. In accordance with that principle, where there is scientific uncertainty as to the existence or extent of risks to human health, the Community institutions may take protective measures without having to wait until the reality and seriousness of those risks become fully apparent (Case C-180/96 United Kingdom v Commission [1998] ECR I-2265, paragraph 99, and Case T-13/99 Pfizer Animal Health v Council [2002] ECR II-3305, paragraph 139). Moreover, in a situation in which the precautionary principle is applied, which by definition coincides with a situation in which there is scientific uncertainty, a risk assessment cannot be required to provide the Community institutions with conclusive scientific evidence of the reality of the risk and the seriousness of the potential adverse effects were that risk to become a reality (Pfizer Animal Health v Council, paragraph 142). In view of the foregoing and of the case-law cited in paragraphs 92 to 95 above, the Court notes that it is apparent from recital 5 of the contested decision that that decision is based not only on the available toxicological data and on the uncertainties of the genotoxic and carcinogenic properties of dichlorvos, but also, more generally, on the overall poor quality of the dossier. In that regard, it has already been stated, in paragraph 98 above, that there were gaps in the dossier submitted by the applicant and thus no reliable conclusion could be drawn as regards the genotoxic and carcinogenic properties of dichlorvos and, more generally, as regards the safety of dichlorvos. Accordingly, the EFSA report states that certain uncertainties can be dispelled only if the missing data and studies are provided. It must therefore be held, in view of the precautionary principle, that the Commission, in the light of the available toxicological data, the uncertainties of the genotoxic and

109

ibid para 111.

54

Elizabeth Fisher

c arcinogenic properties of dichlorvos and the gaps in the dossier which make it impossible to carry out a conclusive risk assessment, did not make a manifest error of assessment in adopting the contested decision or infringe Article 5 of Directive 91/414.110

Thus the Court, while rejecting the plea, did not find the RI paradigm argument invalid, but merely that the action taken here was an exception to it. No other positive vision of EU executive power was offered up. There was little in the way of reasoning. And this is the real problem with the dominance of the RI paradigm. It has stopped a more nuanced mainstream debate about the role and nature of EU expert executive power. Such a debate is particularly important when expert executive power has played an important and enduring role in the narrative of EU integration.111 Claims to expert authority are at the very core of the EU project and have been so since the days of its early neo-functionalist logic. Discussions about expertise are really discussions about our hopes and expectations for the EU as a polity. As this is the case, then to have no debate over what form it can and should take is problematic. It is thus perhaps not surprising that Habermas has commented about how the Euro-crisis highlights the need for a sense of a ‘legally constituted civil solidarity’.112 An integral aspect of that solidarity is that the nature and role of our institutions are up for debate. As the case law above, the work of others,113 and even my own historical reviews of the case law114 have shown, there are strands of other models of administrative constitutionalism embedded in legal frameworks in the EU. The problem is that currently they remain only fragments, rather than a promoted framework of risk governance. The end result is not just the promotion of the RI paradigm, but also little in the way of constructive debates about how administrative constitutionalism figures in European integration. Such a debate would have many dimensions and explore how questions of expertise relate to questions of conferral; that questions of expertise need to be considered against the multi-level institutional architecture of the EU; and how expertise relates to the development of through-put legitimacy.115

110

ibid paras 115–18. the collection of essays in the 2015 issue of www.cogitatiopress.com/ojs/index.php/ politicsandgovernance. 112 J Habermas, ‘Democracy in Europe: Why the Development of the EU into a Transnational Democracy is Necessary and How is it Possible’ (2015) 21 European Law Journal 54. See also H abermas, above n 11. 113 Above n 40 and Joerges, above n 38. 114 E Fisher, ‘Opening Pandora’s Box: Contextualising the Precautionary Principle in the European Union’ in E Vos, M Everson and J Scott (eds), Uncertain Risks Regulated: National, EU and International Regulatory Models Compared (London, Routledge-Cavendish, 2009). 115 V Schmidt, ‘Democracy and Legitimacy in the European Union Revisited: Input, Output and “Throughput”’ (2013) 61 Political Studies 2 and above n 8 ch 6. 111 See

Administrative Constitutionalism

55

IV. Conclusion As I was preparing this chapter I was out walking along the cliffs of Dover on a late April afternoon with my 10 year old son. It was a fortnight before the UK general election of 2015 and we had been visiting Dover Castle. We had driven through Dover, a town that has the transient feeling of so many border towns. Dover Castle had put on a St George’s Day pageant attended by history buffs, families, those wishing to celebrate St George Day as a statement of Englishness, and tourists who were in transit. As we walked we talked about the EU and the UK Independence Party. ‘The thing is,’ my son said gravely ‘is if they [UKIP] understood the law on the free movement of goods they would then understand why the EU is a good thing’. Such a view came from the fact that he has had to sit in the back of my EU Law lectures on occasion. Like many children he fears war, and Swedish Jet Ski116 and Omega117 with their possibilities of justified breach of the fundamental freedoms are cases that give him solace—they show how the EU Member States are tied together in complex and diverse ways that respect diversity. His comment does fall into the trap of many discussions about expert executive power and the EU—that the gaining of expertise must entail the gaining of a particular normative viewpoint. But he also has a point—the more we understand legal and governance structures the more we can debate their legitimacy in relation to what they really are. As I noted in the introduction, I am editing this piece in the aftermath of the EU referendum result. And in reading back through it, my son’s belief in the importance of knowing about the EU is poignant. Leading up to the referendum, the public discourse, while genuflecting to the ‘facts’ about the EU,118 encouraged little discussion and debate about the EU itself. Part and parcel of this, was that expertise was ‘kettled’ in a most remarkable way. Conservative politician and prominent leave campaigner Michael Gove argued that ‘I think people in this country have had enough of experts’119 in response to predictions by economists that the UK leaving the EU would have negative economic consequences. The experts Gove referred to were not just EU experts. But such a comment, pushing as it does institutional reservoirs of knowledge, experience and skills to one side, reflected the general disengagement from debates about governance. It also reflects the current dominance of the RI model in the EU, which discourages a richly textured debate

116

Case C-142/05 Aklagaren v Mickelsson & Roos [2009] ECR I-4273. Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609. 118 Anand Menon and Jonathan Portes, ‘You’re Wrong Michael Gove—Experts are Trusted Far More Than You’ Guardian (London, 9 June 2016) www.theguardian.com/commentisfree/2016/jun/09/ michael-gove-experts-academics-vote. 119 Michael Deacon, ‘EU Referendum: Who Needs Experts When We’ve Got Michael Gove?’ The Telegraph (London, 6 June 2016) http://www.telegraph.co.uk/news/2016/06/06/eureferendum-who-needs-experts-when-weve-got-michael-gove/. 117

56

Elizabeth Fisher

over administrative constitutionalism. Expertise was treated as timeless and easily black boxed. The point of this chapter has been that public debate about EU expert executive power is not just nice to have but is about fostering a demos. The EU referendum debate should have been about EU governance structures. It should have been a robust debate about constitutionalism, and administrative constitutionalism, but it was not. What this chapter has made clear is that administrative constitutionalism and co-production are integral to ‘legally constituted civil solidarity’. Generating some sense of an EU demos is not just debating issues, but it is also about debating how those issues are governed. In particular, it is about understanding that how we are governed does frame our understanding of the world. It is particularly about the role of expert executive power in governance. Such power is necessary. It has been shaped by history and culture. But its role and nature is not a given—it is up for debate and that debate is part of a healthy polity.

Part 2

Practices of Co-production

58

4 Recombinant Regulation: EU Executive Power and Expertise in Responding to Synthetic Biology ELEN STOKES

I. Introduction This chapter examines the implications in terms of European Union (EU) expert and executive power of ‘recombinant regulation’, whereby existing regulations recombine with newly emerging applications of technology. It will be rare for a new technology to emerge entirely unregulated, given the number and breadth of EU legislative measures regulating public health, environmental protection and workplace safety. A degree of interpretative flexibility may be required, however, to ensure that the regulations apply in a manner befitting their aims, especially if the technology brings new risks and societal challenges. This question of regulatory ‘fit’, arising when regulations are extrapolated from one social or technological context to another, raises interesting questions about whether and how—through what forms of expertise and decision making—regulatory regimes adapt to different circumstances.1 The focus here is on the EU regulation of synthetic biology, an emerging area of research that involves a range of techniques resulting in the insertion of synthetic genetic material into living cells. Synthetic biology is said to be ‘[i]n many ways … an extension of existing recombinant DNA technologies’.2 Like ‘conventional’ genetic modification, synthetic biology uses recombinant DNA techniques to alter the genetic make-up of organisms so that they develop new traits. There are varying schools of thought, however, on

1 There is extensive scholarship on the ability of law and regulation to keep pace with technological advances. See eg, MH Shapiro, ‘Is Bioethics Broke?: On the Idea of Ethics and Law “Catching Up” with Technology’ (1999) 33 Indiana Law Review 17; L Bennett Moses, ‘Recurring Dilemmas: The Law’s Race to Keep Up With Technological Change’ (2007) 2 Journal of Law, Technology and Policy 239; R Brownsword and M Goodwin, Law and the Technologies of the Twenty-First Century: Text and Materials (Cambridge, Cambridge University Press 2012) chs 15 and 16. 2 Science and Technology Committee, Bioengineering (HC 2009–10, 220) 66.

60

Elen Stokes

exactly where the similarities between synthetic biology and conventional genetic modification begin and end, in part because the terms encompass a variety of concepts and s pecific applications. This raises the question of novelty, which proves to be especially important for regulation. It means having to decide how to position synthetic biology in relation to conventional genetic modification, which already has its own regulatory regime. Does the regulatory regime on genetic modification cover synthetic biology, or is synthetic biology so new that it requires specially designed regulatory strategies and measures? How, in other words, does long-established legislation on genetic modification recombine with more recent applications of synthetic biology? And how does this play out in terms of expertise and executive power? The chapter begins with a brief explanation of synthetic biology, before introducing the EU regulatory framework within which synthetic biology may or may not fall. As the regulatory framework is geared towards conventional genetic modification (eg the marketing and cultivation of genetically modified crops), it is unclear whether the framework will adequately deal with newer products and processes of synthetic biology. The question of regulatory ‘fit’ or ‘connection’ is not unique to synthetic biology and arises whenever a new technology looks as though it might outrun existing regulatory arrangements.3 The case of synthetic biology is more unusual, however, in the sense that there is already technology-specific legislation in place. So, the question is not whether synthetic biology comes within the remit of generally-worded legislation governing health or the environment (which it invariably will do). Rather, it is whether synthetic biology is dealt with by legislation designed with a particular technological process in mind—the genetic modification of organisms and microorganisms. In answering this question, policymakers have set about comparing ‘conventional’ genetic modification with ‘more innovative’ synthetic biology, drawing on a range of legal norms and scientific practices along the way. The point is that the application of existing regulation is not a passive process, but involves the active translation, reinterpretation and recombination of existing regulation and new technological conditions. The chapter examines three such encounters between existing regulation and synthetic biology, in an attempt to understand how the conceptualisation of ‘synthetic biology’ as a regulatory target both shapes and is shaped by institutions and practices of executive power and expertise—they are co-produced.4 From this perspective, it immediately becomes apparent that issues of how synthetic biology comes to be defined, by whom (which experts?) and with what political power, cannot be easily separated from each other. First, the chapter focuses on the decision to deal with synthetic biology using ready-made regulation. The European Commission contends that, because synthetic biology can simply slot into the

3

See eg, Brownsword and Goodwin, above n 1 at 67. Jasanoff (ed), States of Knowledge: The Co-Production of Science and Social Order (London, outledge 2004). R 4 S

Recombinant Regulation

61

regulatory regime for genetically modified organisms (GMOs), there is no need for tailor-made legislation. One consequence is that issues of synthetic biology regulation are addressed not through the legislative process but rather at the level of implementation. A proposal for new or amended legislation would instigate an elaborate process of EU institutional negotiation, public and stakeholder consultation and integrated impact assessment. Measures of implementation, on the other hand, are non-legislative, meaning that they are not subject to the same processes of democratic accountability as legislative acts, but are instead adopted on narrower grounds of technical expertise and the exercise of executive power. Second, the chapter turns to examine in greater detail how and on what basis the Commission reached the conclusion that synthetic biology comes within the scope of GMO legislation. I argue that the Commission’s focus on legislative coverage (and on legislative definitions of ‘genetic modification’ and ‘organism’) reinforces the idea that the regulation of synthetic biology comes down to technical questions of the similarities and differences between synthetic biology and genetic modification. These questions have the appearance of being purely scientific; indeed, the greater the input of technical expert groups, the more they have become defined as questions of science and not politics. The chapter discusses the role of the newly established New Techniques Working Group and Synthetic Biology Working Group in defining both the problem and the terms of debate. But although such issues are conceived of as technical, they also have important social and political dimensions—in the sense that they have consequences for the regulation of synthetic biology, and for the production and distribution of its associated risks and benefits. The regulatory fate of synthetic biology was sealed when the Synthetic Biology Working Group, for example, determined that synthetic biology comes within the terms of existing legislation and is to be regulated in the same manner as ‘conventional’ genetic modification. Third, following on from this, the chapter shows how technical discourses (not just of scientific definition but also of legislative coverage) have given powerful impetus to the research and development of synthetic biology, but have effectively limited the extent to which broader considerations, such as the social effects, purposes and claimed benefits of the technology, are subject to public scrutiny. The result is a lack of systematic and real time public engagement with the prospects and implications of synthetic biology, and with the reasons for and consequences of using existing legislation. Treating synthetic biology as equivalent to genetic modification helps to maintain regulatory continuity and ‘normalise’ its novelty.5 There can be good reasons for this approach: it minimises market disruption, avoids the need to re-legislate (in an area already dense with legislation) and legitimates the current regulatory approach. But it also means that democratic

5 For discussion, see S Jasanoff, ‘Ordering Life: Law and the Normalization of Biotechnology’ (2001) 17(62) Politeia 34.

62

Elen Stokes

deliberation on the broader ethical, legal and social impacts of synthetic biology is sidelined—either because it is assumed to have already taken place (ie it is historicised) or because it is postponed to some unknown point in the future (ie it is deferred). Hence, regulatory stability in the present can come at a cost, in this case by limiting opportunities for incorporating more diverse knowledges and understandings in collective decisions about the future. Meanwhile forms of EU executive and expert power are maintained or produced—co-produced—as applications of synthetic biology emerge and ontological distinctions (genetic modification vs non-genetic modification) are drawn. Overall, then, the recombination of synthetic biology with legislation on ‘conventional’ GMOs provides an example of how new, emerging technological opportunities interact with existing legal and policy institutions and practices at EU level, and of how technical expertise and executive power are constituted and reaffirmed along the way.

II. Synthetic Biology Synthetic biology, which involves the application of engineering principles in the construction of novel biological systems, has received a great deal of attention in recent years based on its potential to deliver a wide array of technological benefits.6 New and revolutionary applications include the engineering of microbes to perform industrial tasks such as biofuel production,7 and the reprogramming of human cells for therapeutic purposes.8 Although its specific uses vary quite considerably, synthetic biology is generally said to involve the deliberate design and construction of new biological systems and the redesign of existing biological systems in order to perform particular functions.9 This can result in systems and organisms with properties not found in nature—such as bacteria that produce spider silk, and cells that change colour when they divide.10 The use of the term ‘engineering’ in biology is not new. Since the development of recombinant DNA technology in the 1970s, it has been used to describe the

6 E Andrianantoandro et al, ‘Synthetic Biology: New Engineering Rules for an Emerging Discipline’ (2006) 2(1) Molecular Systems Biology 1; M Heinemann and S Panke, ‘Synthetic Biology: Putting Engineering into Biology’ (2006) 22(22) Bioinformatics 2790; SA Benner and AM Sismour, ‘Synthetic Biology’ (2005) 6 Nature Reviews Genetics 533. 7 See eg, SK Lee et al, ‘Metabolic Engineering of Microorganisms for Biofuels Production: From Bugs to Synthetic Biology to Fuels’ (2008) 19(6) Current Opinion in Biotechnology 556. 8 See eg, WC Ruder, T Lu and JJ Collins, ‘Synthetic Biology Moving into the Clinic’ (2011) 333(6047) Science 1248. 9 See generally, Royal Academy of Engineering, ‘Synthetic Biology: Scope, Applications and Implications’ (2009). 10 Royal Society, ‘Synthetic Biology: Scientific Discussion Meeting Summary’ (2008) 2.

Recombinant Regulation

63

action of ‘cutting and pasting’ DNA sequences. The exciting thing about synthetic biology is said to be that ‘engineering stops being a metaphor to become a veritable methodology … for designing biological systems and objects with enhanced or altogether new-to-nature properties’.11 The promise of substantial benefits for health, the environment and the economy has not been lost on EU policymakers. For instance, synthetic biology is described as offering ‘novel opportunities for the creation of new industries with profound economic implications for the European Union’.12 Yet there are also potential risks involved, as there are with any s cientific and technological enterprise—although quite how risky (and whether it is unacceptably so) remains a matter of debate. Many issues are uncertain, including how synthetic biology relates to visions of ‘good’ and ‘attainable’ techno-scientific futures, but EU policy has so far been more narrowly focused on the question of regulatory ‘fit’—a question which, of course, has a bearing on the nature and direction of technological and scientific progress, but which tends to be debated only in terms of legislative coverage. In order to understand the regulatory implications of synthetic biology, it is useful to take a real-life example to see how and in what contexts the question of regulation arises. Take, for instance, the engineering of ‘bioluminescent’ (lightproducing) plants. The idea is that glow-in-the-dark plants might lessen, or one day even replace, the need for electric street lights. One project that has attracted widespread media attention uses synthetic biology techniques to endow a particular type of cress plant (Arabidopsis thaliana) with the genetic circuitry of fireflies, causing it to emit a weak, green-blue light. The plant, and others like it, can be purchased online from a distributor in the United States (US), although it is not without controversy that these glowing synthetic organisms are commercially available.13 Critics claim that the ‘widespread, random and uncontrolled release’ of bioengineered seeds poses ‘real world risks to the environment’, and cite ‘the controversial and experimental nature of synthetic biology’ as a reason for the need to desist from such activity.14 The purpose of this chapter is not to evaluate claims about either the benefits or risks of glowing plants, but to consider how their production and market circulation would be regulated in the EU, and, in particular, what the regulatory arrangements tell us about EU expert and executive power.

11 ERASynBio, Next Steps for European Synthetic Biology: A Strategic Vision from ERASynBio (ERASynBio, 2014) 2. 12 Scientific Committee on Health and Environmental Risks (SCHER), Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR), and Scientific Committee on Consumer Safety (SCCS), ‘Opinion on Synthetic Biology I—Definition’ (2014) 8. 13 Wilson Center, US Trends in Synthetic Biology Research Funding (Washington DC, Wilson Center, 2015) 9. 14 ETC Group, ‘ETC and Friends of the Earth request Kickstarter Inc to Cancel the “Glowing Plants” Project’ (ETC Group, 7 May 2013) www.etcgroup.org/content/kickstopper-letter-kickstarter.

64

Elen Stokes

III. Non-legislation and the Implementation of Existing Regulation To reiterate, this chapter is about how synthetic biology is regulated and whether synthetic biology fits within regulatory categories set out in existing GMO legislation. It does not deal with the specifics of GMO legislation, except insofar as they raise contentious issues of interpretation for the regulation of synthetic biology. GMO legislation has been dealt with in considerable detail elsewhere and it is not my intention to revisit that rich body of work here.15 My aim is simply to gain an understanding of how synthetic biology has been set on a particular regulatory path, who is involved in such decisions, and what are the effects on expertise and executive power in the EU.

A. Recalling Past Regulatory Arrangements As there is no EU legislation dealing specifically with synthetic biology, policymakers, regulators and regulatees are left to wonder whether existing regulatory regimes are up to the task. It is commonplace that emerging technologies, such as synthetic biology, raise questions about the scope and applicability of antecedent regulations, and the challenge is in striking a balance between a principled and a pragmatic regulatory response. On the one hand, legislation cannot be expected to address every techno-scientific eventuality; statutes can quickly become irrelevant and obsolete, and produce results that run counter to the spirit of the rules. On the other hand, it is implausible that new rules and regulations would be introduced every time a new, high-tech product, process or industry is developed. There are practical concerns not only about the financial and logistical inconvenience of having to re-legislate for each new technology, but also about the possibility that new, tailor-made legislation might stifle technological progress. These fears may not always be well-founded, but there is a tendency among policymakers and stakeholders—particularly in the field of biotechnology—to be wary of legal intervention that could impede techno-scientific development.16 Law and regulation are associated with control and the imposition of constraints on behaviour, whereas science and technology are imagined as endless sources of creativity and innovation. Although these are gross caricatures, they still animate many of the policy debates on the regulation of new technology—particularly debates that centre on the potential heavy-handedness of regulatory activity. 15 See eg, M Weimer, ‘Risk Regulation and Deliberation in EU Administrative Governance: GMO Regulation and its Reform’ (2015) 21(5) European Law Journal 622; M Lee, EU Regulation of GMOs: Law and Decision Making for a New Technology (Cheltenham, Edward Elgar, 2008); A Herwig, ‘Whither Science in WTO Dispute Settlement?’ (2008) 21(4) Leiden Journal of International Law 823. 16 For discussion, see Organisation for Economic Co-Operation and Development, ‘Regulatory Reform and Innovation’ (1996) 25.

Recombinant Regulation

65

This can be seen from early discussions of the regulation of synthetic biology in the US, which showed a strong preference for ‘softer’ and self-regulatory modes of governance over ‘hard’ regulation in the form of new legislation. In May 2006, over 300 researchers, government representatives, independent policy analysts, bioethicists, lawyers, economists, psychologists, anthropologists, philosophers, and members of the press and non-governmental organisations gathered at a conference—Synthetic Biology 2.0—in Berkeley, California to discuss the current state of research and expected future developments in synthetic biology. There, the delegates adopted a Declaration under which they resolved to, inter alia, ‘support ongoing and future discussions with all stakeholders for the purpose of developing and analyzing inclusive governance options, including self-governance’.17 Those familiar with the field will find the Declaration’s commitment to self- governance unsurprising, and some commentators have been quick to draw a comparison with the 1975 Asilomar Conference on Recombinant DNA.18 At Asilomar, a group of international experts agreed that research into recombinant DNA molecules should continue provided it was subject to certain guidelines, including, for example, that ‘experimenters must strictly observe the best microbiological practice in lab work’, and ‘it is appropriate to categorize experiments as low, moderate, and high risk activities and to develop distinct contaminant recommendations for each category of experiments’.19 Much detailed and reflective work has been done on the ramifications of the Asilomar conference, and on how it has come to be remembered: ‘memorialized simply as Asilomar, [it] became the twentieth century’s iconic example of scientific self-regulation’.20 In some quarters, it is even remembered as an alternative to regulation: Our aim as a community therefore should be to respond to this public concern, just as we did in 1975. And I think the best way we can do that is by agreeing as a community on a set of limited guidelines and restrictions, just as we did in 1975. If we ourselves act now to protect the public, we reassure them that we place our concern for their interests ahead of our self-interests. We also take away from those in government who would control every aspect of what we do both the moral high ground and the initiative.21

What is striking is how Asilomar (and particular interpretations of the event and its regulatory repercussions) is often the starting point for discussions about synthetic biology, and is seen as providing a working model for initiatives such as

17 ‘Declaration of the Second International Meeting on Synthetic Biology’ (MIT, 29 May 2006) http://syntheticbiology.org/SB2Declaration.html, emphasis added. 18 See eg, D Ferber, ‘Time for a Synthetic Biology Asilomar?’ (2004) 303(5655) Science 159; ‘Synthetic Biology: Life 2.0’ The Economist (London, 31 August 2006). 19 P Berg et al, ‘Summary Statement of the Asilomar Conference on Recombinant DNA Molecules’ (1975) 72(6) Proceedings of the National Academy of Sciences 1981. 20 S Jasanoff, ‘Introduction: Rewriting Life, Reframing Rights’ in S Jasanoff (ed), Reframing Rights: Bioconstitutionalism in the Genetic Age (Cambridge MA, MIT, 2011) ch 1, 7; S Wright, Molecular Politics: Developing American and British Regulatory Policy for Genetic Engineering, 1972–1982 (Chicago, University of Chicago Press, 1994) esp ch 3. 21 GA Petsko, ‘An Asilomar Moment’ (2002) 3(1) Genome Biology 1, 3.

66

Elen Stokes

the ‘Synthetic Biology 2.0 Declaration’. But memory is selective, and only certain recollections of Asilomar have gained ground—most notably its association with an anti- or self-regulatory stance. The recollected past involves constructive interpretation, as Mary Douglas observes: When we look closely at the construction of past time, we find the process has very little to do with past at all and everything with the present … History emerges in an unintended shape as a result of practices directed to immediate, practical ends.22

The practical end now sought by EU policymakers is the regulation of synthetic biology using legislation enacted for ‘conventional’ biotechnology. The European Commission is clear that any new regulation in this area should operate on a voluntary basis, at least in the first instance. It notes that such regulation should take the form of ‘a code of conduct drawn up by SynBio stakeholders’,23 much like those agreed upon in Asilomar in 1975 and Berkeley in 2006. Moreover, the Commission is reluctant to consider the adoption of synthetic biology-specific law because it considers that there is already comprehensive legislation in place, albeit legislation targeting ‘regular’ genetic engineering. That legislation was drafted and introduced before synthetic biology became more widely practised in scientific and industrial communities, but—crucially—because synthetic biology builds on ‘conventional’ techniques of genetic engineering, the legislation is said to apply by extension.

B. Exercise of Executive Rather than Legislative Power We can only speculate as to the many possible reasons for the policy of applying existing legislation, although it is by no means unusual for governments to resist legislative amendment in rapidly evolving fields of technology, especially in the early stages of technological development.24 More interesting and tangible are the

22 M Douglas, How Institutions Think (Syracuse NY, Syracuse University Press, 1986) 69–70. For detailed critique of assumptions about Asilomar, see C Marris and C Jefferson, ‘Synthetic Biology: Containment and Release of Engineered Micro-Organisms—Scoping Report’ (King’s College London, July 2013) www.kcl.ac.uk/sspp/departments/sshm/research/Research-Labs/CSynBI@KCL-PDFs/Publicationspage/SB-Containment-and-Release-Workshop-Scoping-Report-Final.pdf, 21: ‘It is often stated that existing containment measures established after the Asilomar conference on recombinant DNA technologies in 1975 have proved their effectiveness as biosafety measures, because there have been no serious health disasters, even though GMMOs [genetically modified micro-organisms] have been routinely used in numerous laboratories and factories around the world for several decades. But it is important to recognise that the absence of evidence of harm is not the same as the evidence of absence of harm. Rather than serving as evidence of safety, the lack of evidence of harm can be interpreted as evidence that failures in biosafety measures are not sufficiently tracked and investigated. Indeed studies indicate that data on the containment effectiveness of equipment and laboratories are scarce and fragmented’. 23 Commission, ‘Synthetic Biology: From Science to Governance—A Workshop Organised by the European Commission’s Directorate-General for Health & Consumers’ (18–19 March 2010) 15. 24 See eg, Commission, ‘Regulatory Aspects of Nanomaterials’ (Communication) COM (2008) 366 final; Commission, ‘Second Regulatory Review on Nanomaterials’ (Communication) COM (2012) 572 final.

Recombinant Regulation

67

consequences of this approach in terms of the balance between legislative and executive power. The fact that synthetic biology is construed as not requiring new legislation immediately narrows the sort of questions asked about regulation in this context. Instead of opening up synthetic biology regulation to a diverse range of views and interests,25 the issue is reduced to a series of decisions about how existing legislation can best be implemented. It is in implementing and making operational sense of the legislation, that the EU and its Member States exercise executive power. Let us return to the example of glow-in-the-dark plants. The two principal measures of legislation most relevant to their production and commercialisation are: Directive 2001/18/EC on the deliberate release into the environment of genetically modified organisms (‘Deliberate Release Directive’);26 and Directive 2009/41/EC on the contained use of genetically modified micro-organisms (‘Contained Use Directive’).27 The Deliberate Release Directive establishes a procedure for granting consent for the deliberate release or placing on the market of GMOs. Applications for consent are scrutinised by independent scientific experts to see whether they meet the safety standards set out in the Directive,28 and regulators make their decision based on expert advice and representations by the public. The Directive distinguishes between GMOs released ‘for any other purpose than for placing on the market’ (ie non-commercial releases) and GMOs placed on the market ‘as or in products’ (ie commercial releases). Decisions regarding non-commercial releases are taken at the Member State level, whereas responsibility for decisions involving commercial releases lies primarily at EU level, but there is Member State involvement there too (applications are made initially to the national Competent Authority, before a decision is taken by the European Commission with the assistance of Member States through comitology). The Contained Use Directive sets out safety requirements to prevent the contained use of genetically modified micro-organisms having negative consequences on human health and the environment. ‘Contained use’ means work in research laboratories and commercial bioscience facilities and does not refer, for example, to the deliberate release of genetically modified plants or animals into the environment (which is covered by the Deliberate Release Directive) or to the use of genetically modified products in food (which is covered by certain provisions of the Deliberate Release Directive, as well as legislative measures specifically on GMOs 25 For instance, new legislation is subject to minimum standards for stakeholder consultation, see Commission, ‘Better Regulation Guidelines’ (Staff Working Document) COM (2015) 215 final, 8: ‘It is important to consult as early and as widely as possible in order to maximize the usefulness of the consultation and to promote an inclusive approach where all interested parties have the opportunity to contribute to the timely development of effective policies’. 26 European Parliament and Council Directive 2001/18/EC of 12 March 2001 on the deliberate release into the environment of genetically modified organisms [2001] OJ L106/1 (‘Deliberate Release Directive’). 27 European Parliament and Council Directive 2009/41/EC of 6 May 2009 on the contained use of genetically modified micro-organisms [2009] OJ L125/75 (‘Contained Use Directive’). Note that the chapter focuses on applications of synthetic biology other than in medicinal and food or feed products. 28 Deliberate Release Directive, above n 26 at art 16.

68

Elen Stokes

in food and feed). The Contained Use Directive introduces, among other things, a classification system based on the risk and purpose of the contained use,29 and the requirement that all premises30 (and certain individual contained uses, depending on their risk profile)31 are notified to the national Competent Authority. The Directives define an ‘organism’ (or ‘micro-organism’) as ‘any biological [or microbiological] entity capable of replication or of transferring genetic material’.32 An organism (or micro-organism) is said to have been ‘genetically modified’ if it has undergone any of the techniques listed in specific annexes to the legislation.33 As neither Directive makes reference to ‘synthetic biology’, their applicability turns on the breadth of meaning attributed to ‘organism’ and ‘micro-organism’. In other words, are the definitions of ‘organism’ and ‘mirco-organism’ broad enough to encompass the organisms and micro-organisms produced using synthetic biology? Beginning with definitional coverage is not unusual in these circumstances, nor is it an unreasonable place to start in evaluating the capacity of regulation to remain connected with the technological state of the art.34 But it is important to highlight that existing regulations have inevitably conditioned the terms of the debate now taking place on synthetic biology. Synthetic biology is seen through the lens of the two Directives and their definitional categories. This, in turn, brings to prominence particular sites of knowledge-making and expertise. The following section illustrates how synthetic biology regulation has come to be perceived as a definitional, and ultimately a technical, problem.

IV. An Emerging ‘Problem’ and the Role of Technical Working Groups This section charts the emergence of synthetic biology as a ‘problem’ in EU policymaking spheres. The examples below shed light on the process by which the ‘problem’ of synthetic biology regulation emerges and becomes institutionally stabilised through the work of expert groups. Initial uncertainty about the appropriateness of existing GMO legislation has translated into an overriding concern for technical questions of definition. It is through this process of translation that contested meanings are settled and order is imposed—both material order, in the sense that synthetic biology is viewed as an extension of genetic modification, and regulatory order, by making it clear which legislative provisions apply. These states 29

Contained Use Directive, above n 27 at art 4. ibid art 6. 31 ibid arts 7, 8 and 9. 32 Deliberate Release Directive, above n 26 at art 2(1); also Contained Use Directive, above n 27 at art 2(a). 33 Deliberate Release Directive, above n 26 at art 2(2)(a) and annex 1A, pt 1; Contained Use Directive, above n 27 at art 2(b) and annex I, pt A. 34 For discussion, see Brownsword and Goodwin, above n 1. 30

Recombinant Regulation

69

of order are not cleanly separable, rather they have evolved in ways that influence each other. The aim here is to provide insights into how synthetic biology has been classified using mixed metrics of technical knowledge, regulatory categories and political aspirations. Such classification relies on, but also sustains, the implementation of existing legislation and scientific forms of expertise. Expert working groups have played a central role in the implementation of existing legislation as applied to synthetic biology, and to that extent are important ‘backstage actors’35 in the carrying out of EU executive functions.

A. The New Techniques Working Group In October 2007, at the request of the Competent Authorities under the D eliberate Release Directive,36 the European Commission established a specialist Working Group—called the ‘New Techniques Working Group’—to examine certain new techniques in plant breeding. Belgium, the Czech Republic, Hungary and the Netherlands had each reported problems in assessing whether the application of certain new techniques resulted in a ‘genetically modified organism’ (or ‘microorganism’) as defined by the Deliberate Release Directive and the Contained Use Directive. The fact that the legislation (as applied to new plant breeding techniques) was open to different interpretations among Member States prompted the Dutch delegation to propose a harmonised approach under which a new Working Group would develop common practices.37 As the setting up of the Working Group was not governed by specific rules of procedure,38 its terms of reference were decided by the Commission in consultation with the Competent Authorities. Competent Authorities were invited to nominate national experts, whose job would be to discuss the new techniques in light of the definition of genetically modified (micro-) organisms and the available scientific data. Although the Working Group did not operate under a clear legislative mandate, the Group noted that its establishment would contribute to the fulfilment of Article 31(1) of the Deliberate Release Directive—which requires that Member States and the Commission exchange information on the prevention of risks from the release and placing on the market of GMOs.39 35 See generally, D Curtin, ‘Challenging Executive Dominance in European Democracy’ (2014) 77(1) Modern Law Review 1. 36 Commission, ‘Report on the experience of Member States with Directive 2009/41/EC of the European Parliament and of the Council of 6 May 2009 on the contained use of genetically modified micro-organisms (recast) for the period 2006–2009’ (Commission Working Document) COM (2012) 398 final, 7. 37 Commission, ‘Working Group on the Establishment of a List of Techniques Falling under the Scope of Directive 2001/18/EC on the Deliberate Release of Genetically Modified Organisms into the Environment and Directive 90/219/EC on the Contained Use of Genetically Modified Micro- Organisms’ ENV B3/AA/ D(2008). 38 Note the general rules set out in Standard Rules of Procedures for Committees 2011/C206/06 [2011] OJ C206/11, art 6. 39 Deliberate Release Directive, above n 26 at art 31(1).

70

Elen Stokes

The Working Group agreed a non-exhaustive list of eight new techniques, from those proposed by the Competent Authorities,40 for which the interpretation of the legislation was unclear. The list included synthetic genomics, a field within synthetic biology that may include genetic modification depending on its usage. Synthetic genomics involves the synthesis of DNA molecules and their combination to form larger functional DNA molecules (ie genomes), which are then transferred into a recipient structure (eg a cell).41 Such practices of assembling the building blocks of DNA allow for the easy introduction of changes into the genetic material, as well as the redesign or new design of combinations of biological parts that do not necessarily exist in nature. The Working Group sought to determine whether the creation of synthetic genomes and their transplantation into recipient hosts would result in a ‘GMO’ according to the legislation. Recall that the definition of ‘organism’ (or ‘micro-organism’) covers only those entities capable of replication and/or transferring genetic material.42 The difficulty, as the Working Group saw it, lay in applying that definition in the context of synthetic genomics, since it was unclear whether ‘organism’ was intended to cover just the recipient host (the cell), or the resulting entity (the organism produced after having gone through all the steps of the particular technique). The Working Group noted that once the synthetic genome is introduced to a recipient cell then the resulting entity would meet the legislative definition, provided it is capable of replication or of transferring material.43 It also noted that the technique would be likely to constitute ‘genetic modification’, as defined in the annexes to the legislation.44 If, however, the emphasis is only on the recipient cell (rather than the resulting entity), then such activity would fail to trigger the legislation because a recipient cell is incapable of replication or of transferring genetic material within the meaning of the Directives—in other words, it would fall outside the definition of ‘organism’ (or ‘micro-organism’).45 So there are different possible interpretations depending on the point at which the legislation is deemed to apply, leading the Working Group to the following conclusion: Given the fact that synthetic genomics (and more generally synthetic biology) is a fastevolving field with a potential for very new developments as compared to what can be achieved with gene modification techniques currently listed in the Directives, the WG [Working Group] raised the question whether the GMO legislation would be the most appropriate place to deal with this technique, and with synthetic biology in general.46

40

ibid, art 20. New Techniques Working Group (NTWG), ‘Final Report’ (2012) 39. Release Directive, above n 26 at art 2(1); Contained Use Directive, above n 27 at art 2(a). 43 NTWG, above n 41 at 10. 44 Deliberate Release Directive, above n 26 at annex IA, pt 1; Contained Use Directive, above n 27 at annex 1, pt A, point 2. 45 NTWG, above n 41 at para 5.8.3. 46 ibid 41, emphasis added. 41

42 Deliberate

Recombinant Regulation

71

Moreover, the Working Group pointed out that it was impossible categorically to state whether or not synthetic genomics (and synthetic biology more broadly) constituted ‘genetic modification’, since the techniques were at such an early stage of development.47 As a consequence, it reported that the subject was ‘too broad to discuss in this context’,48 and called for further work to be done on the legal implications of the development of novel synthetic organisms.49 The activity of the New Techniques Working Group offers a useful entry point for discussion of the interplay between the material and normative dimensions of synthetic biology at a moment of ‘emergence’, when new technological objects (ie products of synthetic genomics) and institutional responses (ie the New Techniques Working Group) both come into being.50 It is in this phase of emergence, Sheila Jasanoff notes, that important normative choices are made, for example ‘in the resolution of conflicts; the classification of scientific and social objects; the standardization of technological practices’.51 Here, it is worthwhile looking at how authoritative technical knowledge about synthetic biology is produced in a context characterised by uncertainty. There is epistemic uncertainty, as reflected in the Working Group’s comments on the material lack of scientific data, making it diffi cult to draw clear ontological boundaries (are ‘new’ techniques of synthetic biology equivalent to or different from ‘conventional’ genetic modification?), which in turn creates a state of regulatory uncertainty. What is especially interesting about the activity of the Working Group is that it provides a snapshot of the inseparable connections between attempts to order nature (through synthetic genomics) and social practice (in seeking out definitional clarity). Moreover, the Working Group does not shy away from the fact that, in its development of the technical basis ‘for future guidelines or new categorisation of techniques of genetic modification’,52 it is engaging in politics. After all, the Working Group was created with the aim of achieving harmonisation between Member States, and to that end the ‘technical’ work of the Group is underpinned by an accepted need to avoid inter-state disagreement and market fragmentation.53 The Working Group’s report shows, not in so many words but practically, just some of the policy issues at stake in classifying synthetic genomics.54 In this regard, it is very much a site of ‘hybrid judgement’,55 which the Working Group seems to 47

ibid para 5.8.1.

48 ibid. 49

ibid 41. Jasanoff, above n 4 at 6. 51 S Jasanoff, ‘Afterword’ in Jasanoff, above n 4, ch 14, 278–79. 52 NTWG, above n 41 at 44. 53 See generally, S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2007). 54 For discussion on classification and its consequences in the context of biotechnology, see S Jasanoff, ‘Biotechnology and Empire: The Global Power of Seeds and Science’ (2006) 21(1) Osiris 273, 278–79. 55 See S Jasanoff, ‘Judgement Under Siege: The Three-Body Problem of Expert Legitimacy’ in S Maasen and P Weingart (eds), Democratization of Expertise? Exploring Novel Forms of Scientific Advice in Political Decision-Making (Dordrecht, Springer, 2005) ch 12, 213. 50

72

Elen Stokes

acknowledge but without actually saying so.56 The social, political and normative dimensions are also brought to the fore by the involvement of Member States, through their representatives, in the provision of technical advice. For example, the UK’s Advisory Committee on Releases to the Environment had this to say about the Working Group’s findings: ‘This flexibility in the EU’s definition of a GMO’—meaning the fact that the legislative terms are open to interpretation— ‘presents a problem for regulators and will affect innovation in plant breeding and crop science in the EU’.57 This also shows how knowledge-making in the domain of scientific expertise can advance certain ideas about the expected, or desired, trajectory of technological development. Approaching the issue from a co-productionist perspective is about uncovering ‘the normative that lurks within the technical’.58

B. The Synthetic Biology Working Group It is interesting to see how the findings of the New Techniques Working Group have filtered through the policy process. This next part of the story is about how other experts have picked up the baton, following the Working Group’s call for further investigation into the legal status of synthetic genomics. One gets a sense from the New Techniques Working Group report of some of the material and political contingencies of synthetic genomics; therefore, it is intriguing to find that these elements became less prominent in the deliberations of different expert committees. Here we see that, as the subject matter has been channelled through these committees, so it has crystallised into a technical problem requiring technical expertise. Shortly after the New Techniques Working Group published its report, the European Commission requested that further work be undertaken on synthetic biology by the three standing Scientific Committees: the Scientific Committee on Emerging and Newly Identified Health Risks (SCENHIR); the Scientific Committee on Health and Environmental Risks (SCHER); and the Scientific Committee on Consumer Safety (SCCS).59 Their mandate was to issue a joint scientific opinion on (1) the scope and definition of the phrase ‘synthetic biology’; (2) methodological and safety aspects; and (3) research priorities. To co-ordinate the activities of the Scientific Committees, a new Working Group on Synthetic Biology was

56 See eg, NTWG, above n 41 at 41: the regulatory status of synthetic genomics will depend on which interpretation ‘is considered to be more appropriate’. 57 Advisory Committee on Releases to the Environment (ACRE), ‘ACRE Advice: New Techniques Used in Plant Breeding’ (2013) 33. 58 J Corburn, ‘Community Knowledge in Environmental Health Science: Co-Producing Policy Expertise’ (2007) 10 Environmental Science and Policy 150, 159; S Jasanoff, ‘Technologies of Humility: Citizen Participation in Governing Science’ (2003) 41 Minerva 223, 240. 59 SCENIHR, SCCS and SCHER, ‘Mandate for a Joint Scientific Opinion on Synthetic Biology’ (26 February 2014).

Recombinant Regulation

73

formed in July 2013. To date, the Committees (via the new Working Group) have published opinions on the definition of synthetic biology and on the risk assessment and safety aspects of the technology. This section focuses on the first of these, the Opinion on Synthetic Biology—Definition,60 although it also comments on the Working Group’s other reports on the subject. It is useful to begin by outlining the rationale for a new definition of synthetic biology. As indicated by the European Commission’s request for a joint scientific opinion, the main concern was that the Scientific Committees produced a ‘science-based, operational’ definition.61 The Scientific Committees understood ‘operational’ to mean a definition that ‘enables risk assessment and is sufficiently broad to include new developments in the field’.62 In producing the definition, the Scientific Committees were instructed by the Commission to take into account: (1) the relationship between synthetic biology and the genetic modification of organisms; (2) current scientific, technical and commercial developments; and (3) any other, existing definitions. Based on these factors, the Opinion defined synthetic biology as: ‘the application of science, technology and engineering to facilitate and accelerate the design, manufacture and/or modification of genetic materials in living organisms’.63 This exercise raises questions of how the definition was arrived at, and with what effects. The first point to make is that the Scientific Committees (in the process of addressing the Commission’s questions and in the final text of their Opinion) have repeatedly claimed that their work is non-political and non-regulatory. This is unsurprising, given their constitutional background. The Scientific Committees are composed of independent scientific experts,64 and operate according to specific procedural rules governing matters such as transparency of process and stakeholder communication.65 The Rules of Procedure make it clear that the Scientific Committees’ opinions should be science-based and avoid considerations not related to issues of health, safety and environmental risk. According to the Rules, opinions on risk assessment ‘should not address or be influenced by economic, social, ethical aspects’ given that ‘[t]hose aspects are to be addressed, as appropriate in the Impact Assessment Procedure’.66 The Scientific Committees may require additional input from stakeholders in completing an opinion, in order to ‘gather specific comments and suggestions on the scientific basis of the opinion’, but the 60

SCHER, SCENIHR and SCCS, above n 12. ibid 46. 62 ibid 5. 63 ibid 30. 64 Commission, ‘Setting up an advisory structure of Scientific Committees and experts in the field of consumer safety, public health and the environment’ (Decision) [2008] OJ L241/21, see art 15 for the meaning of ‘independence’. 65 See eg, Commission, ‘Establishing Scientific Committees in the field of public health, consumer safety and the environment’ (Decision) C (2015) 5383 final, art 12. 66 Commission, ‘Rules of Procedure: the Scientific Committees on Consumer Safety (SCCA), Health and Environmental Risks (SCHER), and Emerging and Newly Identified Health Risks (SCENIHR)’ (2013) 50. 61

74

Elen Stokes

Rules stipulate that such a consultation process ‘shall not deal with policy or risk management needs and measures’.67 This demarcation between science and politics is evident in much of the Committees’ work on synthetic biology. Interestingly, the dividing line can be seen most clearly in the Committees’ responses to stakeholder views on the matter, collected during a process of public consultation. This is because the participating stakeholders were easily able to see the regulatory implications of defining synthetic biology, especially as the definition was based on a comparison with GMOs for which there were existing legislative regimes. Many of the stakeholder comments reflected ‘real world’ concerns about how such a definition might apply in practice. Yet the Scientific Committees insisted that ‘[t]he regulatory question was not part of the mandate’.68 When asked about the role of the precautionary principle and whether any lessons could be learned from the scientific risk management of genetic modification, the Committees responded: Policy decisions on how the precautionary principle should be applied go beyond the scope of work of the SCs [Scientific Committees] on SynBio … This is a critical point, but this decision is made by the politicians, not by the SCs as part of the risk assessment community.69

Similarly, the Committees emphasised the importance of remaining neutral about the possibilities of synthetic biology, noting that ‘our definition should not be unduly biased by the preoccupations of early-21st century researchers … We also need to avoid SynBio as a “vision” if we want to have a practically useful definition and avoid subjectivity and loopholes’.70 Yet, one does not have to look far to find the regulatory consequences of the Committees’ Opinion. Shortly before the Opinion was published, the Committees noted that ‘[t]he key message in our Opinion is that it remains unclear whether key principles underlying GMO regulatory framework would still apply for Synbio’.71 The Opinion itself concluded that synthetic biology ‘is currently encompassed within genetic modification as defined in the European Directives 2001/18/EC and 2009/41/EC and will likely remain so in the foreseeable future’.72 So, in terms of definition, synthetic biology is seen as equivalent to genetic modification, but in terms of underlying regulatory principles (such as risk assessment), it is unclear whether and how far the analogy can be stretched. Here we see how the classification of synthetic biology and the establishment of social order (through regulation) reinforce each other, in the mutual construction of sameness and difference, of conventionality and novelty.73 The fact that synthetic biology 67

ibid 44, emphasis added. See generally, annex IV on Stakeholder Dialogue Procedures. SCENIHR and SCEHR, ‘Results of Public Consultation on the Scientific Committees’ Preliminary Opinion on Synthetic Biology I—Definition’ (2014) 8. 69 ibid 17. 70 ibid 35. 71 SCCS, SCENIHR and SCHER, above n 68 at 17. 72 SCHER, SCENIHR and SCCS, above n 12 at 30. 73 For discussion of constructions of novelty, see Jasanoff, above n 5. 68 SCCS,

Recombinant Regulation

75

is defined in familiar terms (eg ‘the design, manufacture and/or modification of genetic materials in living organisms’) helps to sustain arguments that it is covered by existing legislation. Meanwhile, the simultaneous conclusion that it is ‘unclear whether key principles underlying GMO regulatory framework would still apply to Synbio’ suggests that synthetic biology is, in other respects, unfamiliar and in need of different policy. Such boundary-drawing between novelty and normalcy is evident in other EU fora of scientific expertise. Another example comes from the work of the New and Emerging Science and Technology (NEST) High-Level Expert Group (established before the New Techniques Working Group and the Synthetic Biology Working Group, discussed above); it typifies the EU’s general approach in so far as it demonstrates the ‘selective normalising’ of novelty in synthetic biology.74 In its report, the NEST High-Level Expert Group described synthetic biology in the following terms: It is often said that biology is the only existing nanotechnology that really works. But if we want to exploit this ‘natural nanotechnology’ for applied, engineering objectives, we will ultimately need to be able to intervene and to modify it at the level that synthetic biology is exploring.75

The coupling of synthetic biology to ‘natural nanotechnology’ is interesting as a means of normalising synthetic biology, as well as nanotechnology, which, as this edited collection shows, is another focus of EU attention.76 It also shows the tendency to separate out the negative effects of technological development from its benefits. Novelty exists in the possibility and promise of revolutionary change (ie by offering new ways of exploiting biology),77 whereas the process of synthetic biology is portrayed in less grand terms, as unexceptional and routine (eg involving the application of familiar principles of engineering). A similar practice of ‘separating out’ and ‘structuring’ the range of possibilities can be seen in relation to risks and other possible side-effects. The general categories of risk relating to synthetic biology are said to be broadly comparable to those associated with ‘conventional’ applications of biotechnology. For instance, it is noted that ‘[i]n terms of risks, abuses and safety measures, it is not obvious that there is any aspect of synthetic biology that is qualitatively different from the way such issues apply to biotechnology and genetic modification’.78 At the same time, however, it is recognised that several developments in the field (eg combining genetic parts, and interactions between certain synthesised and natural organisms) will require new 74 Detailing this phenomenon in a US regulatory context, L Levidow and S Carr, ‘Normalizing Novelty: Regulating Biotechnological Risk at the US EPA’ (2000) 11 Risk, Health, Safety and Environment 9. 75 Commission, ‘Synthetic Biology Bio—Applying Engineering to Biology’ (2005) 10. 76 See in this volume, T Ehnert, ch 5. 77 For example, ‘biology now seems poised to enter an era where significant advance in u nderstanding will derive from a fruitful dialogue between theory and experiment, from analytical and synthetic efforts, and from interdisciplinary interaction with the chemical, physical, engineering and computational sciences’: Commission, above n 75 at 10. 78 ibid 18.

76

Elen Stokes

methodological approaches, owing to the potential for new variants of risk and routes of exposure.79 This ‘slicing and dicing’ of newness—so that certain features of synthetic biology are understood as clearly novel (ie benefits) or potentially novel (ie risks) while others are expressed in more conventional terms (eg the process of genetic modification)—has repercussions not only for the application of regulation but also for the future role of expertise in this area.80 For example, the Scientific Committees point out that ‘[t]he present definition also enables the rapidly advancing nature of GM technologies [including synthetic biology] and adds an important nuance that supports the need for on-going updates of risk assessment methods’.81 Inevitably those ‘ongoing updates’ will take place in the scientific domain, and the new Working Group on Synthetic Biology will be just the body to steer such developments through. Through the updating of risk assessment methods, regulation is continually able to respond, all the while ‘normalising’ each incremental step and narrowing the policy focus on technical minutiae. The circularity of the argument appears complete. Existing legislation provides the vantage point from which synthetic biology is addressed; this informs the task of defining ‘synthetic biology’, which involves drawing a comparison with ‘conventional’ genetic modification as in the legislation; that task is presumed to fall within the remit of technical expertise given that the legislation is already predisposed to dealing with risk; this in turn leads to a concentration of effort on regulatory adaptation by means of ‘technical fixes’ (eg new methods of risk assessment); and, moreover, the understanding of synthetic biology as a variant of established practices of biotechnology serves to legitimise current legislative arrangements. As such, ‘none of the questions and hypothetical scenarios suggested a paradigm shift that would necessitate amending the … r egulatory framework’.82 This keeps synthetic biology out of the legislative process and subject instead to the various configurations of administrative executive power.

V. Regulatory Stability and the Deferral of Socio-ethical Deliberation So far, it has been argued that the production of techno-scientific ‘facts’ about synthetic biology (what it is and what it is not) is intimately connected with the 79 SCHER, SCENIHR and SCCS, ‘Opinion on Synthetic Biology II: Risk Assessment Methodologies and Safety Aspects’ (2015). 80 For discussion of law’s ‘knowledge moves’, see M Valverde, ‘Theoretical and Methodological Issues in the Study of Legal Knowledge Practices’ in A Sarat, L Douglas and MM Umphrey (eds), How Law Knows (Stanford CA, Stanford University Press, 2007) ch 3, esp 83–90. 81 SCHER, SCENIHR and SCCS, above n 12 at 31. 82 Biotechnology and Biological Sciences Research Council, as quoted by Marris and Jefferson, above n 22 at 6.

Recombinant Regulation

77

production of particular patterns of expert knowledge-making and regulatory interpretation. Such patterns reflect, as well as produce, a deference to existing technological categorisations and legislative rules, just as they reflect and produce regulatory stability. This final section briefly considers a further implication—that is, the development of synthetic biology becomes a matter of considerable urgency so that its societal and economic benefits might be unlocked in the here and now, but any wider debate about issues relating to the technology are either assumed to have been resolved or are postponed until some point in the future. The effect is that, while EU policymakers have sought to construe ‘evidence-based policy’ in broader terms than has been the case previously (so that it includes non-technical knowledge and non-traditional sites of expertise),83 doing ‘more science’ remains the priority. On the first point, the primacy accorded to technical matters (eg the difference or similarity between synthetic biology and genetic modification) is maintained in part by the assumption that the relevant socio-ethical questions have already been asked and answered. ‘Historicisation’ is a recognised strategy of political imagination which can involve placing an emphasis on certain continuities between then and now in order to produce a sense of shared experience.84 While such strategising is not immediately obvious in the context of synthetic biology, the policy seems implicitly to accept that the socio-ethical considerations are a stable and unproblematic area of inquiry. The assumption is that these considerations remain relatively unchanged and that fundamental socio-ethical conflicts (which arose in respect of early activity involving recombinant DNA but are presumed to be relevant for synthetic biology today) have been largely resolved. On the topic of biotechnology and bioethics, for example, the European Economic and Social Committee reiterated that: Since the 1975 Asilomar Conference, a large number of bioethics committees have addressed issues relating to all life science studies or medical applications that are likely to pose problems of respect for human dignity, the right of consent or the principle of non-discrimination.85

The focus on policymakers’ long track record of dealing with such issues serves to ‘naturalise’ what we know so that we take for granted the shape and measure of socio-ethical concerns.86 As has been noted, many of the ethical implications of

83 See eg, Commission, ‘Better Regulation Guidelines’ (Staff Working Document) SWD (2015) 111 final, 73, which emphasises the importance of stakeholder consultation and the incorporation of a ‘whole spectrum of views’. 84 Y Ezrahi, Imagined Democracies: Necessary Political Fictions (Cambridge, Cambridge University Press, 2012) esp ch 4. 85 Economic and Social Committee, ‘Communication from the Commission—Towards a s trategic vision of life sciences and biotechnology: consultation document’ (Opinion) [2002] OJ C94/23, para 3.11.1. 86 S Jasanoff, ‘Border Crossings: Social Sciences and Public Policy’ in AG Pereira and S Funtowicz (eds), Science For Policy: New Challenges, New Opportunities (Oxford, Oxford University Press, 2009) Foreword.

78

Elen Stokes

synthetic biology are ‘not new and have been addressed in many areas of life science research’.87 Another, much earlier, example of this comes from the European Commission’s Forecasting and Assessment in Science and Technology programme, which launched a multi-disciplinary working group (‘Social Dimensions of Biotechnology’) comprising industrialists, risk analysts and academics. The group produced a report offering five general principles to guide policy-making in the ‘management of our bio-society’: (1) there can be no policy of zero risk; (2) there are limits to the competence of experts; (3) even within these limits, their credibility/acceptability is questioned; (4) there is a need for adaptive strategic management, taking account of long-term, ‘total system’ effects so far as currently perceivable, and accepting that perceptions will be modified by experience; and (5) there is a need for continuing education at many levels, including decisionmakers, the general public and the scientists themselves. Again, when contemporary references are made to these findings, the implication is that there is no need to pursue similar lines of inquiry as the work has already been done. So, while socio-ethical considerations tend to take the back seat, techno- scientific development is the driving force for change. One gets a sense of this from the title of a European Commission workshop—Synthetic Biology: From Science to Governance—which reinforces the idea that the technical questions need to be resolved first, before being subject to wider public engagement.88 The European Commission is investing heavily in the science, as evidenced by its NEST programme, mentioned above, which pledges to fund only research that is outside or lies across traditional thematic priority areas and is highly innovative and involves correspondingly high (technical) risk.89 Some of that funding is specifically for projects involving synthetic biology, an area described by the Commission as ‘of enormous strategic and economic significance … in which open and public scientific knowledge will be progressively embedded in technological “solutions”’.90 The Commission notes that projects receiving funding ‘are expected to contribute to achieving the vision of synthetic biology in a coherent way’.91 Those projects are also expected to contribute to the EU’s own project of renewed technological growth, as reflected in various policies including the ‘Innovation Union’ initiative. For example, it is noted that: Much of the research [in synthetic biology] so far has been pioneered by individual groups in the US, and the European research community has been relatively slow to embrace the field … What is needed … is a framework for coordinating the current research, fostering a community of researchers.92 87 J Anderson et al, ‘Engineering and Ethical Perspectives in Synthetic Biology’ (2012) 13(7) uropean Molecular Biology Organization Reports 584, 588; cf AJ Newson, ‘Synthetic Biology: Ethics, E Exceptionalism and Expectations’ (2015) 15 Macquarie Law Journal 45. 88 A workshop title does not a policy make; however, there is, arguably, something to be read into the Commission’s phrasing. 89 Commission, ‘Synthetic Biology: A NEST Pathfinder Initiative’ (2007). 90 ibid 3. 91 ibid 6. 92 Commission, above n 75 at 5.

Recombinant Regulation

79

In other words, synthetic biology has as much to do with the EU’s political identity (as the ‘Innovation Union’) as it does with pathways of techno-scientific development. But it is as a result of their mutual production and reinforcement that ‘other’ conversations are, whether consciously or not, postponed until after the science and politics become settled.93 This is captured in the following extract: I think this is a moving feast and I think there hasn’t been any formal public consultation but I think that this is because some of us feel … [t]hat we are actually struggling a bit to understand whether we do understand the boundaries of what we are asking and how it relates to previous things. I think that is definitely ongoing business where we don’t know the answers.94

It certainly is ‘ongoing business’, not just in the sense of needing to fill gaps in the technical data, but also in the sense of the continual interplay between the finding of facts and the making of norms. The prioritisation of research and development (From Science to Governance) and technical input (the working groups, described above) underwrites the EU’s impressive narrative of technological progress and expert decision-making. These ‘technical’ activities, however, also have normative implications—they create meaning, determine the operation of the regulatory regime and define technological trajectories. If nothing else, the EU’s regulation of synthetic biology provides a useful example of the unavoidable entanglement of politics, power and technological change; states of uncertainty and stabilisation; and linkages between sites of knowledge and social/regulatory order. It tells of the complexities involved in new techno-scientific advances, recombined with more firmly established legislation. It reminds us of what can get lost in the process, such as opportunities for ongoing engagement with socio-ethical issues as the science unfolds. Finally, it shows that in situations involving broad regulatory coverage (eg GMO legislation) in which there is no pressing need for legislative reform, the emergence of new technological capabilities (eg synthetic biology) helps to reify and reinforce the co-production of EU expert and executive power. Here, recombinant regulation suggests a unity between scientific and regulatory orderings, and shows how the EU’s instrumental goals, and the knowledge and practices adopted to achieve them, are closely intertwined and dependent on each other.

93 For related discussion, see B Wynne, Rationality and Ritual: Participation and Exclusion in Nuclear Decision-Making 2nd edn (Abingdon, Earthscan, 2011) xxii, where Wynne describes ‘the chronic postponement of open confrontation’. 94 A Smith, Director General for Science and Research, Department for Business, Innovation and Skills, UK Government, speaking at the US National Academies conference on ‘Opportunities and Challenges in the Emerging Field of Synthetic Biology’ (Washington DC, 9 July 2009) https://www. oecd.org/sti/biotech/43459661.pdf, 119–20.

80

5 Defining the Invisible: Between Soft Norms and Hard Realities in the European Regulation of Nanotechnologies TANJA EHNERT

I. Introduction Slowly but surely nanomaterials are infiltrating the EU market. The tiny materials can already be found in various product groups, such as food, textiles, chemicals, pharmaceuticals and cosmetics.1 Frequently, nanotechnologies are associated with important economic and social advances. In 2009, the European Commission declared them a key enabling technology ‘at the forefront of managing the shift to a low carbon, knowledge-based economy’.2 Inseparably linked to their advances are, however, nanotechnologies’ potentially negative side effects. The impact and possible harm nanotechnologies might inflict upon the human body and the environment are largely unknown,3 which has made an objective calculation of the implied risks and benefits impossible.4 Value judgements inevitably enter the debate, rendering nanotechnologies’ regulation a controversial matter.5 ‘Sooner or 1 Milieu Ltd and RPA Ltd, Proposal for an EU Reporting System for Nanomaterials (study commissioned by European Commission DG Environment in 2008, 2010). 2 See www.ec.europa.eu/growth/industry/key-enabling-technologies/index_en.htm. 3 Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR), Risk Assessment of Products of Nanotechnologies (19 January 2009), in which SCENIHR advocates a case-by-case approach for risk assessment; European Food Safety Authority, ‘Scientific Opinion on the Potential Risks Arising from Nanoscience and Nanotechnologies on Food and Feed Safety’ (2009) 958 European Food Safety Authority Journal 1. 4 S Funtowicz, I Shephard, D Wilkinson and J Ravetz, ‘Science and Governance in the European Union: A Contribution to the Debate’ (2000) 27 Science and Public Policy 327, 330; S Jasanoff, The Fifth Branch: Science Advisors as Policy Makers (Cambridge MA, Harvard University Press, 1994) 232. 5 See for instance M Lee, ‘Risk and Beyond: EU Regulation of Nanotechnology’ (2010) 35 European Law Review 799; G Hodge, D Bowman and A Maynard, International Handbook on Regulating Nanotechnologies (Cheltenham, Edward Elgar Publishing, 2010); G van Calster, ‘Regulating Nanotechnology in the European Union’ (2006) 3 Nanotechnology Law & Business 359; G Marchant, D Sylvester

82

Tanja Ehnert

later’, Ulrich Beck infers, ‘the question of acceptance arises and with it anew the old question: how do we wish to live?’6 The controversy surrounding nanotechnologies evolves around some fundamental questions: when do we define a material a ‘nanomaterial’; where does the ‘nano’-scale begin; and is ‘size’ the only characteristic that should be taken into account? At the same time, rapid scientific and technological progress cast doubt as to whether norms enshrined in legal texts are still the appropriate way forward: isn’t ‘relentless change … best controlled by keeping options open’?7 Along this line, the European Commission has responded to the emergence of nanotechnologies not necessarily by exercising its right of legislative initiative but through regulatory measures in the executive sphere, including ‘softer’ options, such as recommendations or guidance documents aimed at the preparation of new and the enhancement of existing legislation.8 The purpose of this chapter is to provide a critical study of the reliance on rulemaking by experts in the EU administration in the field of nanotechnologies. To this end, the chapter will review in detail the adoption and implementation of one specific measure, namely the European Commission’s recommendation for a definition of ‘nanomaterial’, which will serve as an example of what this volume describes as the co-production of EU expert and executive power. First, the chapter will embed its empirical study of the Commission’s recommendation in an evaluative framework that adds a legitimacy dimension to the concept of co-production, as understood in this volume (Section II). Thereafter, the chapter turns to its empirical part, in which it will analyse the drafting history of the recommendation. For this purpose, the chapter will introduce the controversy accompanying the quest for a definition of the term ‘nanomaterial’, which is mirrored by the various (conflicting) reactions of stakeholders that followed the adoption of the recommendation. Notwithstanding the technical nature commonly associated with nanotechnologies, the chapter will show that decisions taken in the course of the drafting process for the recommendation extended well beyond the mere technical or scientific sphere (Section III). Subsequently, the chapter looks at the practical impact of the ‘nanomaterial’ definition. This definition has been incorporated, almost verbatim, into regulatory frameworks not only at the EU but also at the national level. This process

and K Abbott, ‘Risk Management Principles of Nanotechnology’ (2008) 2 Nanoethics 43; O Renn and M Roco, ‘Nanotechnology and the Need for Risk Governance’ (2006) 8 Journal of Nanoparticle Research 1; K-H Ladeur, ‘Kommunikation über Risiken im Rechtssystem. Das Beispiel Nanotechnologie’ in C Büscher and KP Japp (eds), Ökologische Aufklärung (Wiesbaden, VS Verlag für Sozialwissenschaften, 2010) 131. 6

U Beck, Risk Society: Towards a New Modernity (Delhi, SAGE Publications, 1992) 28. Korkea-aho, ‘Laws in Progress? Reconceptualizing Accountability Strategies in the Era of Framework Norms’ (2013) 2 Transnational Environmental Law 363, 367. 8 See on this shift towards executive rulemaking in the area of nanotechnologies, T Ehnert, ‘The Legitimacy of New Risk Governance: A Critical View in Light of the EU’s Approach to Nanotechnologies in Food’ (2015) 21 European Law Journal 44. 7 E

Defining the Invisible

83

deserves careful scrutiny, taking into account its two dimensions—the adoption of the European definition at the national level, as well as its evolution from a soft to a hard norm (Section IV). The impact of the ‘nanomaterial’ definition, in turn, raises further questions, most notably related to its adoption. The last part of this chapter therefore analyses the drafting process for the recommendation of a definition at the European level, in particular the involvement of stakeholders. It will become clear that, despite formal possibilities for affected actors to participate, the recommendation remains essentially an expert construct (Section V). Based on the evaluative framework introduced in the first section, the chapter will conclude with some critical remarks on the legitimacy of the Commission’s definition of a ‘nanomaterial’. (Section VI).

II. The Legitimacy Dimension of Co-production The co-production of expert and executive power or the close interaction between experts and the executive in EU risk regulation, as described by the editors of this volume, raises important normative questions. This chapter focuses on one of these questions by exploring the link between co-production and legitimacy: what does this intimate relationship between expert and executive power mean for the legitimacy of the EU administration’s regulatory output? In order to tackle this question, this chapter builds on a normative understanding of legitimacy that uses the democratic self-determination of citizens as its reference point.9 The concept of democratic self-determination will be understood to comprise Fritz Scharpf ’s ‘two faces’ of input and output legitimacy.10 According to this analytical distinction, political decisions have input legitimacy when they reflect the will of the citizens,11 which is determined via their participation in the decision-making process.12 Output legitimacy, on the other hand, serves to further the ‘common interest’ of the people, ie allowing for an effective solution of the problem.13 Furthermore, output legitimacy requires the prevention of abuse of political power by holding a regulator accountable for its conduct ex post.14 Fritz Scharpf argues that the development of input legitimacy is impeded at the EU level, due to the absence of a collective European identity.15 In his view, attention should instead be paid to the output aspect of legitimacy. This, however, conflicts with the observations made in the introduction to this chapter, according to which 9 J Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (Boston MA, MIT Press, 1996) 73. 10 F Scharpf, Governing in Europe: Effective and Democratic? (Oxford, Oxford University Press, 1999) 6–13. 11 ibid 6. 12 ibid 7. 13 ibid 6. 14 ibid 13. 15 ibid 12.

84

Tanja Ehnert

an objective determination of the most effective outcome in n anotechnology regulation is rendered impossible by the scientific uncertainty dominating the area. The ‘effective solution’ depends on the subjective perception of the inherent risks and benefits. Output legitimacy is therefore conditioned upon input legitimacy.16 Legitimacy, for the purpose of this chapter, is thus ultimately comprised of the two elements of participation and accountability.

III. Defining the Invisible: Walking a Tightrope A. ‘Nanomaterial’ and the Quest for a Definition Regulating nanotechnologies requires that the regulator qualifies what it intends to regulate: the inclusion of definitions in legislative frameworks forms the prerequisite for delineating an instrument’s scope and subject matter.17 Gaps in scientific knowledge in the area of nanotechnologies have so far hindered standardisation efforts. Different working definitions of ‘nanomaterial’ and ‘nanotechnology’ exist (and compete), which have been elaborated by international organisations such as the International Organisation for Standardisation,18 the European C ommission’s scientific committees as well as several countries such as Australia, Canada, Denmark, the UK and the US.19 To date, no international agreement on a definition has been reached.20 At the European level, the regulation of nanotechnologies has its origin in a Commission Communication of 2004 on a European Strategy for Nanotechnology.21 The Commission proposed in its Communication an ‘incremental approach’, which denotes the adaptation of relevant existing horizontal and sectorial legislative

16 cf A Føllesdal, ‘The Legitimacy Challenges for New Modes of Governance: Trustworthy Responsiveness’ (2011) 46 Government and Opposition 81, 84. 17 D Bowman, J D’Silva and G van Calster, ‘Defining Nanomaterials for the Purpose of Regulation within the European Union’ (2010) 2 European Journal of Risk Regulation 115, 115. 18 ISOTC229 has adopted standards on the core terms of nanotechnology in its series ISO/TS 80004-1:2010 ‘Nanotechnologies—Vocabulary—Part 1: Core terms’, see www.iso.org/iso/catalogue_ detail.htm?csnumber=51240. 19 E Bleeker et al, ‘Interpretation and Implications of the European Commission Recommendation on the Definition of Nanomaterial’ (National Institute for Public Health and the Environment, the Netherlands, RIVM Letter Report 601358001, 2012) 12 www.rivm.nl/dsresource?objectid=rivmp:181 801&type=org&disposition=inline. 20 ObservatoryNano, ‘Development in Nanotechnology Regulation and Standards’ (Report of the Observatory Nano, May 2009). An international workshop co-organised by the US National Institute of Standards and Technology (NIST) and the American National Standards Institute (ANSI) concluded in 2011 that there continues to be a need for the development of consensus-based standards regarding core aspects of nanotechnology; see NIST and ANSI, ‘ANSI: International Workshop on Challenges to Increased Use of Documentary Nanotechnology’ (Washington DC, 13–14 December 2011); for more information see also www.gsi.nist.gov/global/index.cfm/L1-8/L2-33/A-598. 21 European Commission, ‘Towards a European Strategy for Nanotechnology’ COM (2004) 0338 final, 17.

Defining the Invisible

85

frameworks to the specific characteristics of the nanoform.22 Yet, the Commission has so far been careful in introducing nano-specific provisions, including definitions, into legal texts.23 It hardly used its right of legislative initiative. Instead, it is the European Parliament, or even the Council, that drive the legislative review.24 In a Resolution of 2009, the Parliament stated that it: [D]oes not agree, before an appropriate evaluation of current Community legislation, and in the absence of any nano-specific provisions therein, with the Commission’s conclusions that a) current legislation covers in principle the relevant risks relating to nanomaterials, and b) that the protection of health, safety and the environment needs mostly be enhanced by improving implementation of current legislation, when due to the lack of appropriate data and methods to assess the risks relating to nanomaterials it is effectively unable to address its risks.25

The Parliament eventually asked the Commission to review all relevant legislation until June 2011.26 However, the Parliament did not wait two more years for the result, and started introducing nano-specific amendments into ongoing recasts and revisions of relevant legal texts within the ordinary legislative procedure. This is best illustrated by an example. The recast of the Cosmetics Directive, adopted in November 2009,27 constitutes the first EU legislative framework to explicitly mention ‘nanomaterials’. Whereas the Commission’s original legislative proposal did not mention ingredients in nanoform,28 the Parliament seized the opportunity to introduce particular nano-specific provisions at first reading stage.29 The new Cosmetics Regulation contains a definition of nanomaterial,30 and based on that definition, specific pre-market notification31 and labelling32 requirements for 22 European Commission, ‘Nanotechnologies: A Preliminary Risk Analysis on the Basis of a orkshop Organised in Brussels on 1–2 March 2004 by the Health and Consumer Protection W Directorate General of the European Commission’ (2014) 28, www.ec.europa.eu/health/ph_risk/documents/ev_20040301_en.pdf. In its 2008 Communication on ‘Regulatory Aspects of Nanomaterials’, the Commission concluded that ‘overall, … current legislation covers to a large extent risks in relation to nanomaterials and that risks can be dealt with under the current legislative framework. However, current legislation may have to be modified in the light of new information becoming available’, see COM (2008) 366, 3. The approach was again confirmed in European Commission, ‘Second Regulatory Review on Nanomaterials’ (Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee) COM (2012) 572 final. 23 See, for a detailed analysis of the Commission’s regulatory approach in the food sector, Ehnert, above n 8. 24 E Stokes, ‘Nanotechnology and the Products of Inherited Regulation’ (2012) 39 Journal of Law and Society 93, 98. 25 European Parliament Resolution of 24 April 2009 on regulatory aspects of nanomaterials, P6_TA(2009)03287, point 3. 26 ibid point 5. 27 Reg (EC) 1223/2009 of 30 November 2009 on cosmetic products [2009] OJ L342/59. 28 European Commission, ‘Proposal for a Regulation of the European Parliament and of the C ouncil on cosmetic products’ COM (2008) 49 final. 29 European Parliament, ‘Position of the European Parliament adopted at first reading on 24 March 2009 with a view to the adoption of Regulation (EC) No … /2009 of the European Parliament and of the Council on cosmetic products (recast)’ EP-PE_TC1-COD(2008)0035. 30 Reg (EC) 1223/2009, above n 27 at art 2(1)(k). 31 ibid arts 13(1)(f) and 16(3). 32 ibid art 19(1).

86

Tanja Ehnert

nanomaterials, as well as a public register for all cosmetic products containing nanomaterials.33 Following this pattern, the European Parliament subsequently proposed similar amendments to other legislative acts under revision and recast.34 However, the definitions of ‘nanomaterial’ that the Parliament applies have differed. For instance, the definition in the new Cosmetics Regulation is different from the definition in the final negotiated text of the first attempt to revise the Novel Food Regulation in 2011.35 The European Parliament recognised the potential (legal) complications arising from multiple definitions. A material may be characterised as ‘nano’ under one legislative framework but not under another. In its Resolution of 2009, the Parliament noted: ‘the current discussion about nanomaterials is characterised by a significant lack of knowledge and information, leading to disagreement starting at the level of definitions’.36 The Parliament therefore called on the Commission ‘to promote the adoption of a harmonised definition of nanomaterials at the international level and to adapt the relevant European legislative framework accordingly’.37

B. The Commission’s Recommended Definition and Stakeholders’ Reply The Commission responded to the Parliament’s call by adopting Recommendation 2011/696/EU in October 2011, which contains a definition of ‘nanomaterial’.38 The recommendation is addressed to the Member States, EU institutions and industry, who are encouraged to rely on the definition in the adoption and implementation of legislation and policy.39 The Commission found a recommendation preferable to a legally binding measure—for its flexibility and because it ‘does not contain any direct obligations for Member States and economic operators,’ and rather its ‘implementation will happen through various pieces of specific legislation’.40 Unlike the definition that was inserted into the new Cosmetics Regulation, the recommended definition is not limited to engineered nanomaterials; it equally

33

ibid art 16(10)(a). eg Reg (EU) 1169/2011 of 25 October 2011 on the provision of food information to consumers [2011] OJ L304; Reg (EU) 528/2012 of 22 May 2012 concerning the making available on the market and use of biocidal products [2012] OJ L167/1. 35 European Commission, ‘Statement by Commissioner Dalli on the lack of agreement in the conciliation procedure on the Novel Food Regulation’ (29 March 2011) MEMO/11/202. 36 Above n 25 at point F. 37 ibid point 8. 38 Commission Recommendation 2011/696/EU of 18 October 2011 on the definition of nanomaterial [2011] OJ L275/38, point 2. 39 ibid points 1 and 7. 40 European Commission, ‘Questions and answers on the Commission Recommendation on the definition of nanomaterial’ MEMO/11/704 (18 October 2011) www.europa.eu/rapid/pressrelease_MEMO-11-704_en.htm?locale=en. 34

Defining the Invisible

87

comprises natural or incidental materials.41 Moreover, it is exclusively based on the size of the material’s constituent particles. There are hence two core elements in the recommended definition: a material is considered a nanomaterial where at least 50 per cent of its constituent particles are in the size range of 1 to 100 nanometres. Given the scientific uncertainty and disagreements about the appropriate size range, and more fundamentally about whether size should be the only defining characteristic of a ‘nanomaterial’,42 the Commission’s recommended definition has been under attack from all sides. The European Environmental Bureau (EEB) promptly declared that it ‘is deeply disappointed by the European Commission’s decision released today to use a narrow definition for the term “nanomaterial”, indicating that industry lobbying has won over the Commission’s own scientific advisors’.43 The EEB criticised the threshold of 50 per cent as too restrictive, stating that it ‘had called upon the Commission to stick to a 1% threshold of the particle number size distribution as a way to put under scrutiny, certain materials, which may exhibit nano specific hazardous properties in the larger size range’.44 In a similar vein, the European Consumers’ Organisation (BEUC) wrote that: [W]e are critical of the fact that the threshold for the number of ‘nano’ particles needed in order for the product to qualify as a ‘nano-product’ has been raised from 0.15% to 50% and deviates in key points from the opinion of the scientific risk assessment committee and a previous Commission draft. Furthermore, the upper limit of 1–100 nm is too narrow and may lead to a situation in which products with bigger particles could escape the definition and thereby remain unregulated.45

Also the European Chemical Industry Council (CEFIC) expressed severe criticism: [T]he Commission’s recommendation released today is too broad in scope and therefore difficult to integrate into existing legislation in a meaningful way. … CEFIC fears implementing the proposed definition will add unnecessary burden for companies, leading to added costs and less efficient use of resources. In addition, its current form would define as nanomaterials some decades-old substances such as mineral pigments used in paints and other everyday products.46

The Commission’s recommended definition consequently strikes a delicate balance between the different interests at stake, most notably between the economic interest on the one hand, and the interest to protect environmental and health 41 Engineered nanomaterials refer to materials which are intentionally manufactured at nanoscale, while natural or incidental nanomaterials occur naturally or are unintentionally produced as a by-product of some other activity. 42 K Malkiewicz et al, ‘Nanomaterials in REACH’ (Project Report, funded by SKEP ERA-NET, 2011) 27. 43 EEB, ‘Nano Definition Too Narrow Says EEB’ (2011) www.eeb.org/index.cfm/news-events/news/ nano-definition-too-narrow-says-eeb/. 44 ibid. 45 The European Consumer Organisation/The European Consumer Voice in Standardisation (BEUC/ANEC), ‘Nano: Very Small and Everywhere’ (2012) 17, www.anec.eu/attachments/ANEC_ BEUC%20nano-silver%20brochure%20updated.pdf. 46 Cefic, ‘Practical nanomaterials definition needed to push forward next great innovation breakthroughs’ (2011) www.cefic.org/PressReleases/Press%20statement%20nano%20definition.FINAL.pdf.

88

Tanja Ehnert

safety on the other. The definition is hence all but void of political choices. As the Dutch National Institute for Public Health and the Environment (RIVM) highlights, ‘at present, the particle size range of 1nm–100nm has no scientific basis’ thus the decision ‘remain[s] a political one’.47 Likewise, the chosen particle number threshold of 50 per cent ‘has been a political decision because no scientific reasoning can be given for a threshold value’.48

IV. The Definition’s Impact: Analysing its Two Dimensions Notwithstanding the opposition of various stakeholders, the Commission’s definition managed to slip into several legislative texts as well as guidance documents at European and national level. Two dimensions of this process will be analysed: the definition’s migration from the European to the national level and its evolution from a soft norm to a hard legal reality for affected actors.

A. First Dimension: From European to National Level The Commission’s recommended definition has recently been used to formulate nano-specific laws in the EU’s Member States. Generally, nanotechnologies are or will be relevant to chemicals, food, cosmetics or medicines legislation—fields which are harmonised at the EU level.49 Most of the regulatory attention in the area of nanotechnologies has therefore been focused at the European rather than the national level. Yet, the Commission’s slow implementation of its ‘incremental approach’ has progressively caused discontent in several Member States.50 France,51 Belgium52 and Denmark53 introduced national mandatory reporting 47

Above n 19.

48 ibid.

49 The legislative frameworks identified by the Commission in its Communication ‘Regulatory Aspects of Nanomaterials’ are predominantly based on: Treaty on the Functioning of the European Union (TFEU) (internal market), art 114; TFEU (worker protection), art 153; TFEU (Common Agricultural Policy), art 43; TFEU (public health), art 168; and TFEU (environmental protection), art 192. 50 See for instance European Commission, ‘Summary Record 9th Meeting of the REACH Competent Authorities Sub-Group on Nanomaterials, 20 November 2012, CASG Nano/02/2013’ (2013) 2 (document on file with the author). 51 France adopted a decree for mandatory reporting of products containing nanomaterials in February 2012, which entered into force on 1 January 2013, Décret n° 2012-232 du 17 février 2012 relatif à la déclaration annuelle des substances à l’état nanoparticulaire pris en application de l’article L. 523-4 du code de l’environnement. 52 The Belgian Council of Ministers validated the national mandatory register on 7 F ebruary 2014. Registration of substances will begin on 1 January 2016. See www.presscenter.org/fr/pressrelease/ 20140208/la-belgique-met-en-place-un-registre-des-nanomateriaux. 53 See Statutory Order no 644 of 13 June 2014 on an inventory of mixtures and products that contain nanomaterials and the duty of manufacturers and importers to report to the inventory, www.eng. mst.dk/topics/chemicals/nanomaterials/.

Defining the Invisible

89

systems for nano-products, with similar ongoing discussions in several further Member States, including Germany.54 When France announced its intent to pass a respective law in 2009, critics immediately cried foul, raising doubts as to its compatibility with EU law, especially with the legislative framework for chemicals, ie the Regulation on the Registration, Evaluation and Authorisation of Chemicals (REACH). In a nutshell, the argument was that since ‘REACH should be considered as seeking full harmonisation for all chemicals produced and marketed in the EU’,55 ‘Member States cannot simply adopt measures on nanomaterials on the grounds that REACH is not specific enough to regulate nanomaterials’.56 France nevertheless notified its draft legislation to the Commission in 2011, which entered into force on 1 January 2013.57 The French legislation was, however, drafted before the Commission’s recommendation for a definition of ‘nanomaterial’ was adopted. Initially, the French draft legislation hence contained a different definition, resembling the one incorporated into the EU Cosmetics Regulation. However, following the recommendation’s adoption France adapted its definition to take into consideration and follow the Commission’s recommended definition.58 Belgium is currently developing a similar register, also with the Commission’s definition at its roots.59 The Danish statutory order establishing a register of nanoproducts explicitly states that it follows Commission Recommendation 2011/696/EU on the definition of nanomaterial.60 Consequently, the Commission’s recommended definition of ‘nanomaterial’ constitutes the basis of the only three nano-specific national laws currently existing in the EU. The recommended definition was therewith adopted from the European to the national legal context.

54 The SPD Fraction of the German Bundestag requested in December 2011 ‘ein öffentliches Produktregister und eine Meldepflicht erforderliche und angemessene Maßnahmen’, see Antrag der SPD Fraktion, ‘Chancen der Nanotechnologien nutzen und Risiken für Verbraucher reduzieren’ Drucksache 17/8158 (14 December 2011) www.dipbt.bundestag.de/dip21/btd/17/081/1708158.pdf. 55 J-P Monfort et al, ‘Nanomaterials under REACH: Legal Aspects’ (2010) 1 European Journal of Risk Regulation 51, 60. 56 ibid 61. cp E Vogelezang-Stoute, ‘Regulating Nanomaterials: Bottlenecks and Perspectives in EU Legislation on Chemicals and Products’ (2012) European Energy and Environmental Law Review 41, 44. 57 See J-P Monfort, ‘The French Nano Decree: A Model for Europe?’ (presentation given at Nanotechnologies Industry Association (NIA) Symposium, 19 December 2012) slide 15 www.mayerbrown. com/files/uploads/Documents/French_Nano_Decrees.pdf. 58 N Kaddour, ‘No Laws in Nanoland: How to Reverse the Trend? The French Example’ (2013) 30 Pace Environmental Law Review 486, 511. With the exception that the French definition is restricted to intentionally engineered nanoparticles. 59 Belgium notified the Commission in July 2013, under Dir 98/34, of its Draft Decree to create a register of nanomaterials, based on the Commission’s recommendation for a definition of nanomaterial, see notification to COM 2013/369/B. 60 Ministry of Environment and Food of Denmark, ‘Statutory order on a register of mixtures and articles that contain nanomaterials as well as the requirement for producers and importers to report to the register’ no 644 (13 June 2014), para 4 no 1, http://eng.mst.dk/topics/chemicals/nanomaterials/ the-danish-nanoproduct-register/.

90

Tanja Ehnert

B. Second Dimension: From Soft Law to Hard Law This adoption of the European definition at the national level also alters the Com mission definition’s legal status from a ‘soft law’ measure to a legally binding norm. Evidence for such a ‘legalisation’ or ‘hardening’61 of the recommended definition is also found at the European level. After the recommendation’s adoption in 2011, the definition was incorporated, almost verbatim, into the revision of the Biocidal Products Directive. The new EU Biocidal Products Regulation thus defines a ‘nanomaterial’ as comprising substances containing particles of which more than 50 per cent are in the size range of 1–100 nanometres.62 Moreover, in case of ambiguity, Member States may request the Commission to ‘decide, by means of implementing acts, whether a substance is a nanomaterial, having regard in particular to Commission Recommendation 2011/696/EU of 18 October 2011 on the definition of nanomaterial’.63 Similarly, the revision of the Medical Devices Directive that is currently under way in the ordinary legislative procedure contains a definition of ‘nanomaterial’, which follows the Commission’s recommendation.64 Yet, this ‘hardening’ process experienced a setback in February 2014 in the context of the Food Information Regulation.65 The Food Information Regulation was also adopted in October 2011, but originally contained a different definition of ‘nanomaterial’, which neither solely referred to size nor featured the size range of 1–100 nanometres.66 Following the Regulation’s entry into force, the Commission therefore proposed a delegated act, intending to bring the definition in line with its recommendation. The Parliament, however, used its right of objection to a delegated act—for the first time since the Lisbon Treaty entered into force in 2009.67 The Parliament contested, inter alia, that the 50 per cent threshold included in the Commission’s recommendation ‘[ran] contrary to the basic aim of the directive to pursue a high level of protection of consumers’ health and interests’.68 Instead, the Parliament considered a threshold of 10 per cent ‘more a ppropriate’.69 Stakeholders, too, seized the opportunity to comment. BEUC s ubmitted an open letter to the European Parliament before the plenary vote, urging its members 61 See for a recent article on the concept of ‘legalisation’ and the ‘hardening of soft law’, F Terpan, ‘Soft Law in the European Union—The Changing Nature of EU Law’ (2015) 21 European Law Journal 68. 62 Reg (EU) 528/2012, above n 34 at art 3(1)(z). See for a more detailed analysis, A Gergely and L Berzanskis,‘How the EU Nano Definition Will be Applied to Biocides’ (ChemicalWatch,April 2012) www. chemicalwatch.com/10657/nanowatch-how-the-eu-nano-definition-will-be-applied-to-biocides. 63 Reg (EU) 528/2012, ibid art 3(3). 64 European Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on medical devices’ COM (2012) 542 final, recital 13 and art 2(1)(15). 65 Reg (EU) 1169/2011, above n 34. 66 ibid art 2(2)(t). 67 European Parliament, ‘Motion for a Resolution on the Commission delegated regulation of 12 December 2013 amending Regulation (EU) No 1169/2011 of the European Parliament and of the Council on the provision of food information to consumers as regards the definition of “engineered nanomaterials”’ P7_TA(2014)0218 (18 February 2014). 68 ibid point T. 69 ibid.

Defining the Invisible

91

to support the opposition.70 The association of the EU food industry FoodDrinkEurope, to the contrary, considered the Parliament’s decision ‘regrettable’, ‘caus[ing] great confusion for consumers’.71 The Parliament thus exposed the controversy underlying the nano definition. The Parliament continued its opposition to the Commission’s recommendation in the second attempt to revise the Novel Food Regulation, which was eventually adopted in November 2015.72 While the Commission referred in its legislative proposal of 2013 to the definition of nanomaterial as laid down in the Food Information Regulation,73 the European Parliament inserted a new definition into the Novel Food Regulation, which no longer considers size as the only characterising property of a nanomaterial.74 The Parliament moreover demanded that this new definition of nanomaterial should replace the current definition in the Food Information Regulation.75 These amendments of the Parliament were accepted by the Commission and the Council, and are part of the final version of the new Novel Food Regulation.76 At the same time, the Parliament is not always able to scrutinise ex ante this ‘hardening’ process of the Commission’s recommendation, especially where it is not inserted into binding acts via a legislative or non-legislative procedure, but where it serves to prepare further soft law measures—such as EU agency guidance documents. This is exactly what happened in the context of the EU chemicals legislation REACH. Here, the European Commission announced in 2013 that it would not propose amendments to the REACH Regulation’s legal text. Instead, the Commission proposed to update the REACH Regulation’s annexes and mandated the European Chemicals Agency (ECHA) to develop guidance documents.77 The Commission justified its decision with reference to ‘legislative stability and predictability’.78 In the course of 2012, ECHA hence adopted updates to its REACH guidance documents in order to make them ‘fit for nano’.79 All of these updates are based on the Commission’s recommended definition of ‘nanomaterial’. 70 BEUC, ‘Plenary vote on a Resolution opposing the draft European Commission (EC) Delegated Regulation updating the definition of “engineered nanomaterials” in food’ (letter sent to the Members of the European Parliament, 10 March 2014). 71 FoodDrinkEurope, ‘MEPs’ vote on “engineered nanomaterials” regrettable for Europe’s food and drink industry’ (statement, 13 March 2014). 72 Reg (EU) 2015/2283 of the European Parliament and of the Council of 25 November 2015 on novel foods [2015] OJ L327/1, art 3(2)(f). 73 European Commission, ‘Proposal for a Regulation of the European Parliament and of the C ouncil on novel foods’ COM (2013) 894 final, art 2(a)(ii). 74 European Parliament legislative resolution of 28 October 2015 on the proposal for a regulation of the European Parliament and of the Council on novel foods COM (2013) 0894, P8_TAPROV(2015)0380, art 3(2)(f). 75 ibid art 33(1). 76 Above n 72 at arts 3(2)(f) and 33(1). 77 Above n 22 at 11. 78 European Commission, ‘General Report on REACH’, COM (2013) 49 final, 13. 79 Three of these updates hence formulate nano-specific recommendations for physicochemical properties (ECHA, ‘Guidance on Information Requirements and Chemical Safety Assessment: Appendix R7-1 Recommendations for Nanomaterials Applicable to Chapter R7a Endpoint Specific

92

Tanja Ehnert

Although formally non-binding, ECHA’s guidance documents play a central role in the day-to-day operation of the REACH regime.80 Furthermore, in their short lifetime, ECHA’s REACH guidance documents have already been invoked in EU courts.81 In her opinion to a preliminary reference, Advocate General Kokott relied upon the REACH guidance for monomers and polymers to support her conclusion that an obligation to register reacted monomer substances entailed in polymers is not disproportionate.82 Beyond REACH, the EU courts find it increasingly ‘useful’ for national courts to ‘take into account’ interpretations of secondary law stipulated in Commission guidance documents,83 but also to consider such documents themselves when interpreting EU legislation.84 Notwithstanding their ‘soft’ nature, ECHA’s guidance documents—including the Commission’s recommended definition of ‘nanomaterial’—thereby become a hard reality for the affected actors.

V. Putting the Recommendation’s Legitimacy to the Test The recommendation’s two-dimensional impact raises further questions, most notably as to its legitimacy. This last part of this chapter first focusses on the Guidance’ (April 2012) www.echa.europa.eu/documents/10162/13632/appendix_r7a_nanomaterials_ en.pdf), toxicological information requirements (ECHA, ‘Guidance on Information Requirements and Chemical Safety Assessment: Appendix R7-1 Recommendations for Nanomaterials Applicable to Chapter R7b Endpoint Specific Guidance’ (April 2012) www.echa.europa.eu/documents/10162/13632/appendix_r7b_nanomaterials_en.pdf) and ecotoxicological endpoints (ECHA, ‘Guidance on Information Requirements and Chemical Safety Assessment: Appendix R7-1 Recommendations for Nanomaterials Applicable to Chapter R7c Endpoint Specific Guidance’ (April 2012) www.echa.europa.eu/documents/10162/13632/appendix_r7c_nanomaterials_en.pdf). Two further updates lay down recommendations for nanomaterials in the characterisation of hazards, as regards human health (ECHA, ‘Guidance on Information Requirements and Chemical Safety Assessment: Appendix R8-15 Recommendations for Nanomaterials Applicable to Chapter R8 Characterisation of Dose [Concentration]—Response for Human Health’ (May 2012) www.echa.europa.eu/documents/10162/13643/appendix_r8_05-2012_en.pdf) and the environment (ECHA, ‘Guidance on Information Requirements and Chemical Safety Assessment: Appendix R10-2 Recommendations for Nanomaterials Applicable to Chapter R10 Characterisation of Dose [Concentration]—Response for Environment’ (May 2012) www.echa.europa.eu/documents/10162/13643/appendix_r10_05-2012_ en.pdf). The last update contains recommendations on occupational exposure estimation for nanomaterials (ECHA, ‘Guidance on Information Requirements and Chemical Safety Assessment: Appendix R14-4 Recommendations for Nanomaterials Applicable to Chapter R14 Occupational Exposure Estimation’ (May 2012) www.echa.europa.eu/documents/10162/13643/appendix_r14_05-2012_en.pdf). 80 See also Korkea-aho, above n 7; L Bergkamp and M Penman, ‘Conclusions’ in L Bergkamp (ed), The European Union REACH Regulation for Chemicals: Law and Practice (Oxford, Oxford University Press, 2013) 410, 412. 81 See also Korkea-aho, above n 7 at 380. 82 Opinion of AG Kokott in Case C-558/07 The Queen, on the application of SPCM SA CH Erbslöh KG Lake Chemicals and Minerals Ltd Hercules Inc v Secretary of State for the Environment, Food and Rural Affairs [2009] ECR I-5783, paras 127 and 131. 83 Case C-308/11 Chemische Fabrik Kreussler & Co GmbH v Sunstar Deutschland GmbH [2012] ECR 0000. 84 See, for instance, ibid para 29.

Defining the Invisible

93

rafting process of the recommendation at the European level, in particular on d the participation of stakeholders, and whether interested parties were involved in the adoption of the recommendation. Second, this part analyses to what extent the Commission could be held accountable for its procedural conduct during the drafting process.

A. The Spectacle of Participation Participation, meaning Scharpf ’s input ‘face’ of democratic self-determination as introduced in Section II of this chapter, may take two forms. Citizens may indirectly participate via their elected representatives, at the European level mostly via the European Parliament. Citizens may also directly participate in the decision-making process, via public consultations or dialogues.85 To this end, in 2002, the Commission issued a Communication entitled ‘Towards a reinforced culture of consultation and dialogue’, which lays down minimum standards for direct participation in the decision-making process.86 These standards were recently confirmed in May 2015 in the Commission’s ‘Better Regulation Package’.87 With these standards, the Commission commits itself, for example, to ensuring a balanced representation of affected parties in the consultation process, to adequate publicity of the consultation, to a sufficient consultation period which provides respondents time to reply and to display results of the consultations on its websites, including the provision of an ‘adequate feedback’.88 It is clear that the European Parliament was not involved in the r ecommendation’s drafting process. Yet, to what extent were interested parties able to participate directly in its adoption? In the drafting of the recommended definition, interested parties were able to submit comments on two occasions (see Figure 1). First, the Commission requested an opinion from one of its expert groups, the Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR),89 which, in the course of drafting its opinion, opened a preliminary version to public scrutiny. Second, the Commission itself launched a public consultation on the draft recommendation before its final adoption. At first glance, interested parties were able to participate.

85

cf TEU, art 11. Commission, ‘Towards a reinforced culture of consultation and dialogue—General principles and minimum standards for consultation of interested parties by the Commission’ COM (2002) 704 final. 87 European Commission, ‘Better regulation for better results—An EU agenda’ COM (2015) 215 final, 4. 88 Above n 86, at 21–22. 89 For the request see www.ec.europa.eu/health/scientific_committees/emerging/docs/scenihr_q_ 024.pdf. 86 European

94 Commission mandates SCENIHR to draft opinion on the scientific basis for a definition of ‘nanomaterial’

Commission drafts recommendation for a definition of ‘nanomaterial’

Tanja Ehnert

SCENIHR drafts scientific opinion

SCENIHR launches public consultation on its ‘preliminary’ scientific opinion

SCENIHR adopts scientific opinion

Commission launches public consultation on its draft recommendation

Commission adopts final recommendation for a definition of ‘nanomaterial’

Figure 1: Adoption of the Commission Recommendation on a Definition of ‘Nanomaterial’

Closer scrutiny, however, reveals a different picture. First, both consultations were performed in the absence of procedural standards. On the one hand, public consultations performed by the Commission’s scientific committees, such as SCENIHR, are generally not subject to the Commission’s minimum consultation standards of 2002.90 On the other hand, the minimum standards do also not apply to all measures adopted by the Commission but only to ‘major policy initiatives’.91 Whether a policy initiative qualifies as ‘major’ depends on the existence of an impact assessment requirement.92 Whilst impact assessments are compulsory for all legislative proposals, non-binding measures—such as recommendations—only fall under its scope where they are likely to ‘define future policies’.93 What exactly this criterion entails is not specified: does a measure ‘define future policies’ only where it contains concrete proposals for legislative or policy action, or is the threshold reached where a recommendation defines key concepts that will be inserted into future legislation or policies, such as the definition of ‘nanomaterial’? To be sure, in the drafting of its recommendation for a definition of ‘nanomaterial’, the Commission considered its impact assessment rules inapplicable. Neither, by implication, were the Commission’s minimum consultation standards hence applicable to the Commission’s public consultation on its draft recommendation. Second, an analysis of the two consultations’ practical operation yields sobering results. This is due, on the one hand, to imbalances in participation between the different societal interests involved, most notably between industry representatives 90 The scientific committees’ rules of procedure emphasise that ‘this particular consultation rocedure should not be confused with other consultations launched by the Commission regarding p policy or regulatory matters, for which a different scope, as well as rules and procedures apply’, see European Commission, ‘Rules of Procedure of the Scientific Committees on Consumer Safety, Health and Environmental Risks and Emerging and Newly Identified Health Risks’ (18 December 2009) annex IV point 5b, ec.europa.eu/health/scientific_committees/docs/rules_procedure_en.pdf. 91 Above n 86 at 15. 92 ibid. 93 European Commission, ‘Impact Assessment Guidelines’, SEC(2009) 92, 6.

Defining the Invisible

95

and environmental non-governmental organisations (NGOs) or consumer organisations. While in SCENIHR’s public consultation 49 per cent of the participants were industry representatives, a mere 15 per cent represented NGOs.94 Even more pronounced was this imbalance in the second public consultation organised by the Commission, where roughly 63 per cent of the respondents represented industry, and only four per cent NGOs.95 On the other hand, the influence of participants in SCENIHR’s consultation was substantially reduced, if not invalidated, by the fact that the Commission published its draft recommendation for a definition of ‘nanomaterial’ before SCENIHR adopted its final opinion: the Commission opened its draft recommendation for public consultation in October 2010,96 ie several weeks before SCENIHR had adopted its final opinion in December 2010. This means that the comments that SCENIHR received from participants during the consultation period were not reflected in the Commission’s draft recommendation (see Figure 2). March 2010

July 2010

Commission mandates SCENIHR to draft opinion on the scientific basis for a definition of ‘nanomaterial’

August 2010 September 2010 October 2010 November 2010

Commission drafts recommendation for a definition of ‘nanomaterial’ 21 October 2010

SCENIHR drafts scientific opinion 6 July 2010

SCENIHR launches public consultation on its ‘preliminary’ scientific opinion 12 July – 15 September 2010

Commission launches public consultation on its draft recommendation 21 October – 19 November 2010

SCENIHR adopts scientific opinion 8 December 2010

December 2010 October 2011

Commission adopts final recommendation for a definition of ‘nanomaterial’ 18 October 2011

Figure 2: Timeline for the Adoption of the Recommendation 94 See European Commission, ‘Explanatory Note how the Comments received during the Public Consultation were taken into Account for the final SCENIHR Opinion on the Scientific Basis for the Definition of the Term Nanomaterial’ SANCO C7, www.ec.europa.eu/health/scientific_committees/ emerging/docs/scenihr_o_032_note.pdf. 95 See www.ec.europa.eu/environment/consultations/nanomaterials.htm. 96 ibid.

96

Tanja Ehnert

The Commission’s subsequent consultation, however, ended after four weeks (contrary to the then eight—and currently 12—weeks stipulated in the Commission’s minimum consultation standards) on the grounds that: [T]he SCENIHR pre-consultation opinion on the ‘Scientific basis for the definition of the term “nanomaterial”’ was closed after 8 weeks of consultation on 15 September 2010. This draft proposal for the term ‘nanomaterial’ builds on the scientific opinion.97

Beyond doubt, this reasoning is inappropriate, most notably since the results of SCENIHR’s public consultation could not be considered by the Commission in its draft recommendation.98

B. A Weak Ex Post Scrutiny Accountability, the output aspect of Scharpf ’s conceptualisation of legitimacy, is narrowly construed in this chapter in order to distinguish it from the input that the citizens’ participation feeds into the decision-making process. Accountability is therefore understood as an ex post oversight mechanism. There are several kinds of fora to which an actor may render account, including political, legal, administrative and social fora.99 The locus of political accountability at the EU level is the European Parliament, which supervises, ex post, the EU’s regulatory activities. Besides the Parliament’s ‘grand supervisory powers’ over the Commission—its budgetary control, parliamentary questions and its right to approve and dismiss the college of Commissioners—the Parliament’s hold on the Commission’s day-to-day executive functions is limited.100 There is no specific parliamentary scrutiny over Commission recommendations (as it exists, for instance, with regard to delegated or certain implementing acts). The relevant annual activity report of the Commission merely informs the European Parliament (and the Council) that ‘the Commission took an important step towards greater protection for citizens by recommending a definition of “nanomaterials” which will help determine which materials need special treatment in specific legislation’.101 The European Parliament did not further examine the adoption process of the recommendation: Members of European

97 ibid.

98 European Commission, ‘Rules of Procedure of the Scientific Committees on Consumer Safety, Health and Environmental Risks and Emerging and Newly Identified Health Risks’ (18 December 2009) annex IV point 5b, www.ec.europa.eu/health/scientific_committees/docs/rules_procedure_en.pdf. 99 M Bovens, ‘Analysing and Assessing Accountability: A Conceptual Framework’ (2007) 13 European Law Journal 447, 441–44. I left out the ‘professional’ forum. 100 A Türk, H Hofmann and G Rowe, EU Administrative Law and Policy (Oxford, Oxford University Press, 2011) 777. 101 European Commission, ‘2011 Annual Activity Report DG Environment’ env_aar_2011_final, 16, www.ec.europa.eu/atwork/synthesis/aar/doc/env_aar_2011.pdf.

Defining the Invisible

97

Parliament neither posed questions regarding the Commission recommendation,102 nor was its content discussed in committees.103 The provision of accountability through the legal forum, ie through the EU courts, emerges as inherently problematic. This is because recommendations are explicitly of non-binding character.104 Yet, following the Court’s substantive approach in determining whether or not an act generates legal effects, even recommendations could in exceptional cases undergo an examination before the Court excludes ‘legal effects’.105 A recommendation must then de facto affect the legal position of third parties, for instance, by imposing a new obligation. This, however, is unlikely in the case of the recommendation on ‘nanomaterials’—for two reasons. First, it is possible to argue that the Commission’s recommendation is merely a ‘preparatory act’, which, only once inserted into the legislative framework (at the national or European level), generates legal effects vis-à-vis third parties.106 The situation, however, becomes complicated where the nano definition is used to prepare soft law, such as agency guidance documents, as is the case in the chemicals sector. In this case, the legal effect of the subsequent instrument, which is prepared by the nano definition, is also uncertain. Only binding decisions which are adopted based on the guidance document can then be challenged in court, rendering it increasingly difficult to trace the proper preparatory act underlying the final decision. What is more, treating the recommended definition as a ‘preparatory act’ negates the practical effects that it exerts on the behaviour of stakeholders— without resulting in a binding act. Second, in determining whether a recommendation could de facto affect the legal position of third parties, the Courts have looked in the past especially at a measure’s wording.107 In the recommendation at hand the Commission merely ‘invites’, as opposed to ‘requires’, its addressees to follow the proposed definition.108 Hence, the wording is far from authoritative.109 Again, this is in tension with the significant practical impacts of the recommendation. A potential remedy to the lack of legal accountability might be through bringing accountability in an administrative forum, in particular the European 102 See www.europarl.europa.eu/plenary/en/parliamentary-questions.html;jsessionid=2131FDAE7 88EA89305A5AB48AD658BB8.node2#sidesForm. 103 According to a search of documents of the parliamentary committees, see www.europarl.europa. eu/committees/en/search-in-documents.html. 104 Pursuant to TFEU, art 288. 105 This approach was pursued by the Court in Case C-322/88 Salvatore Grimaldi v Fonds des maladies professionnelles [1989] ECR 04407. 106 Case T-326/99 Nancy Fern Olivieri v Commission [2003] ECR II-6053, para 53; Case C-60/81 International Business Machines Cooperation v Commission [1981] ECR 2639, para 12. 107 See J Scott, ‘In Legal Limbo: Post-Legislative Guidance as a Challenge for European Administrative Law’ (2011) 48 Common Market Law Review 329, 340. 108 Commission Recommendation 2011/696/EU, above n 38 at points 1 and 7. 109 See for instance Case C-325/91 France v Commission [1993] ECR I-3283, in which the Court was confronted with a Commission Communication that ‘required’ Member States to report to the Commission, or Case C-303/90 France v Commission [1991] ECR I-5315. See for a more detailed discussion of these cases, Scott, above n 107 at 339–41.

98

Tanja Ehnert

Ombudsman.110 The Ombudsman can investigate complaints of ‘maladministration’, which ‘occurs if an institution fails to act in accordance with the law, fails to respect the principles of good administration, or violates human rights’.111 The Ombudsman could hence assume an important role in compensating for the lack of judicial review of non-binding measures, such as recommendations. However, it is questionable whether there are any procedural rules that the Commission could have failed to comply with, most notably due to the vague scope of application of the impact assessment guidelines and, by inference, the Commission’s minimum consultation standards. Finally, the social forum consists of civil society, which may not only participate in the decision-making process, but which may also scrutinise the regulator’s behaviour ex post. A prerequisite for social accountability is transparency,112 which is often understood as the accessibility and visibility of relevant information.113 Whilst the Commission is subject to Regulation 1049/2001 on access to documents, it is doubtful whether there are any documents regarding the adoption process of the recommendation that should have been directly accessible in electronic form via the Commission’s public register. This is because the requirement of ‘direct accessibility’ in Regulation 1049/2001 only applies to ‘documents drawn up or received in the course of procedures for the adoption of acts which are legally binding’114—ie not to those which are non-binding. The public is therefore required to explicitly request potential documents, which arguably requires a profound knowledge of the EU’s regulatory process.

VI. Conclusion Rulemaking by experts in the EU administration plays an important role in the controversial area of nanotechnologies’ regulation. The Commission’s recommendation 110 For a detailed study of administrative accountability through the European Ombudsman, see Tanja Ehnert, ‘Regulating the Invisible: A Critical Analysis of the EU’s Approach to Nanotechnologies’ (PhD thesis, Maastricht University 2015) 234–35. See also Lee, ch 11 in this volume. 111 See www.ombudsman.europa.eu/media/en/default.htm#hl1. To establish what the Ombudsman captures under the notion of ‘good administration’, it adopted the ‘European Code of Good Administrative Behaviour’ in 2001. According to this Code, the Ombudsman shall ‘help citizens to know what administrative standards they are entitled to expect from the EU institutions. It also serves as a useful guide for civil servants in their relations with the public. By making the principle of good administration more concrete, the Code helps to encourage the highest standards of administration’, European Ombudsman, ‘The European Code of Good Administrative Behaviour’ (2013) 4, www.ombudsman. europa.eu/resources/code.faces#/page/1. 112 This, however, does not imply that once the decision-making process is transparent, civil society scrutiny exists. Other factors, most obviously the existence of a European public sphere, need to be considered. 113 D Curtin, Executive Power of the European Union (Oxford, Oxford University Press, 2009) 205. Reflected is the transparency principle in TEU, art 11(2), complemented by the right to access to documents in TFEU, art 15(3). 114 Reg (EC) 1049/2001 regarding public access to European Parliament, Council and Commission documents [2001] OJ L145, art 12(1) and (2).

Defining the Invisible

99

for a definition of ‘nanomaterial’ constitutes a perfect example of what the editors of this volume have described as the co-production of expert and executive power in EU risk regulation. The aim of this chapter was to take a close look at this specific measure—its drafting history and eventually its practical impact—in order to examine the implications of co-production for the legitimacy of executive rulemaking. The analysis of the drafting history of the Commission’s recommendation showed that this seemingly scientific definition entails sensitive political choices and must therefore be understood as a balancing act between the conflicting interests of different stakeholders. By exercising policy choices, the EU executive hence determined a specific interpretation of how a ‘nanomaterial’ should be defined for regulatory purposes. At the same time, it could be demonstrated that, via a two-dimensional evolution from the European to the national level and from soft to hard law, the recommended definition has become a hard reality for affected actors. In several cases, the recommendation was slipped almost verbatim into legal texts, at European and also at national level in three Member States, and served to formulate agency guidance documents in the chemicals sector. This, in turn, drew the chapter’s attention to the recommendation’s drafting process and the potentiality of scrutinising its content ex post: in short, it raised the legitimacy question. Close examination of this question produced rather disappointing results. While stakeholders were invited to participate via two public consultations, both lacked procedural standards and were characterised by significant imbalances in stakeholders’ representation. Overall, the recommendation’s input legitimacy appears fragile. At the same time, it seems difficult to hold the Commission to account for its procedural conduct in the adoption of the definition, be it through political, legal or social fora, especially given the definition’s formally non-binding nature. This chapter hence identified, in the resolution of controversial nano questions, a shift of responsibility from the political fora, such as the European Parliament, to the realm of the EU executive. Ultimately, this has led to a depoliticisation of the nano controversy: decisive questions are not addressed in an open public debate. Instead, normative choices inherent in defining the term ‘nanomaterial’ are concealed behind (quasi-) scientific or technical terminology. Overall, this migration of political choices to their settlement in the EU administration is unmatched by adequate procedures that would ensure the recommendation’s legitimacy. This has become clear when analysing the spectacle of participation in the recommendation’s drafting, but is also manifested in the ineptness of existing ex post oversight mechanisms. This shows that a critical scrutiny of the EU’s reliance on (non-binding) rulemaking by experts in the EU administration is urgently needed, especially in light of these measures’ de facto influence on the behaviour of affected actors.

100

6 Mixing EU Security and Public Health Expertise in the Health Threats Decision ANNIEK DE RUIJTER It is very easy to adopt measures after Katrina for example or after the Tsunami or after something awful happens, everybody can be an expert then. But the real challenge, I think, is to look into the future and see what we think we need to do on the basis of our past experience.1

I. Introduction This chapter analyses the relationship between the changing nature of the expertise underlying and legitimising EU response to public health emergencies, and executive power. In general, in case of the spread of a highly contagious virus on a pandemic scale or when an emergency threatens the health and life of a population, the counter measures that are approved and coordinated by the executive in charge build on the advice of epidemiological experts, who will inform the policy makers on the speed of the spread, the likelihood that vaccines will work, or on what other public health measures may need to be taken, such as school closures or extra capacity in hospitals. Many of these experts are public health experts, epidemiologists, with a public health, or medical degree. The EU’s involvement in planning for public health emergencies, such as a communicable disease outbreak, has focused on the coordination and harmonisation of Member States’ emergency plans, ensuring the least impact on the internal market. The efforts of the EU address, for example, planning, coordination of an emergency response, assessment, prevention and communication.2 1 High level (public health) representative Commission Services, DG SANCO, interviewed in 2010 about the swine flu outbreak. 2 See, M Flear, Governing Public Health: EU Law, Regulation and Biopolitics (Haywards Heath, Bloomsbury Publishing, 2015) ch 5.

102

Anniek de Ruijter

In this chapter the Health Threats Decision of 2013 is analysed as signifying a shift in the EU public health response model, towards a ‘preparedness model’ mixing the underlying expertise that is used for EU health emergency response from the above-described public health experts towards security experts (economists, critical infrastructure and defence experts).3 The chapter also shows that this shift is the result of a transforming expert understanding of the nature of the health risks that we are facing. It describes a regulatory merging of the opinions and relevance of the scientific input of public health experts that believe that health risks that are manageable and perceived as statistically foreseeable, together with security experts and economists that view public health risks as unforeseeable disasters or catastrophes that should be addressed, not only in terms of health and life of humans, but also in respect to ‘security’ of the economy and preparing its ‘critical infrastructures’. The central question is whether the mixing of public health expertise at the EU level in the context of the Health Threats Decision is co-producing EU executive power in the field in ways unforeseen in the EU’s founding Treaty on the Functioning of the EU (‘TFEU’).4

A. The Health Threats Decision’s Legislative Context In recent years the EU has become increasingly involved in the coordination and response to a number of major health events such as Zika, Ebola, bird flu, SARS, swine flu and the E-Coli outbreak.5 In 2013, the EU adopted a new legal instrument, the Decision on Serious Cross-Border Threats to Health (‘Health Threats Decision’),6 to regulate its involvement in the response to public health threats, chemical attacks and bioterrorism. This Decision, adopted by the Council and the European Parliament, is an all-encompassing regulatory instrument, covering not only known, but also unknown health risks. It is a European effort to, in the words of the European Commission, ‘bridge the policy fields of health and security’.7 The main purpose for this Decision is to regulate the coordination of Member States action in responding to ‘major health threats’ by encompassing the already existing communicable disease control network and a number of health threats that before were mainly addressed, if at all, in the context of security policy.8 These other health threats include biochemical attacks and bio-terrorism, and hazards 3 Decision No 1082/2013/EU, ‘Decision No 1082/2013/EU of the European Parliament and of the Council of 22 October 2013 on Serious Cross-Border Threats to Health and Repealing Decision No 2119/98/EC’ (OJ L 293/1 5-11-2013). 4 Treaty on the Functioning of the EU (OJ 115/49). 5 ‘European Commission, Commission Staff Working Document, Health Security in the European Union and Internationally’ (SEC(2009)1622 Final) Brussels, 23/11/2009. 6 Decision No 1082/2013/EU, above n 3. 7 Commission Proposal, ‘Commission Proposal for a Decision of the European Parliament and of the Council on Serious Cross-Border Threats to Health Brussels’ (COM(2011)866 Final). 8 SJ Collier and A Lakoff, ‘Health Security, and New Biological Threats: Reconfigurations of Expertise’ in CE Bird and others (eds), Handbook of Medical Sociology, Sixth Edition (Nashville, Vanderbilt University Press, 2010).

Mixing EU Security and Public Health Expertise

103

created as a result of climate change.9 Moreover, the coordinating structures created by the Health Threats Decision: [S]hould, in exceptional circumstances, be available to the Member States and to the Commission when the threat is not covered by this Decision and where it is possible that public health measures taken to counter that threat are insufficient to ensure a high level of protection of human health.10

This means that potentially the Decision casts a wide net for its possible future application. The Decision is also encompassing in that it links ‘new’ threats in with older systems for information exchange, surveillance and preparedness. For example, the Decision extends the Early Warning and Response System (EWRS),11 which is operated by the European Centre for Disease Control (ECDC) that used to be available only for specific communicable diseases, to all health threats.12 In the following section, the chapter draws out the theoretical and analytical considerations that form the perspective through which the Health Threats Decision is analysed.

II. Reconfiguring Public Health: Globalisation and Security How is the changing understanding of public health risks that underlies the Health Threats Decision co-producing EU executive power? With regard to the underlying expert rationale for understanding public health risks, there are two highly interrelated frames of analysis applied at national and global levels that create a pathway for analysing the co-production of executive and expert power in the field of public health in the EU. The first frame relates to a reconfiguration of understanding public health risks. The second frame relates to globalisation and ‘securitisation’.

A. Reconfiguring Public Health Risks: From Risk Management to Crisis Preparedness At a global level there has been an observance of a ‘reconfiguration of biomedical and security expertise’.13 This reconfiguration leads to new ‘forms of expertise and

9

Decision No 1082/2013/EU, above n 3 at para 8. ibid, para 9. 11 Commission Decision (2000/57/EC), ‘Commission Decision (2000/57/EC) of 22 December 1999 on the Early Warning and Response System for the Prevention and Control of Communicable Diseases under Decision No 2119/98/EC of the European Parliament and of the Council’ (OJ L 21, 26-1-2000, 32). 12 Decision No 1082/2013/EU, above n 3, para 16. 13 Collier and Lakoff, above n 8. 10

104

Anniek de Ruijter

the practices of intervention through which disease threats are understood and managed’.14 One explanation for this reconfiguration is that the second part of the twentieth century has given rise, in the words of Sheila Jasanoff: ‘to risks that escape the framework of actuarial prediction, in part because there is little or no historical experience to fall back on in evaluating them’.15 In the famous account of Ulrich Beck, technical innovation and modernisation processes have created risks of which we can no longer foresee the possible future harm: ‘Along with the growing capacity of technical options grows the incalculability of their consequences’.16 Technological advances have produced new risks to human health. Expertise itself is one of the the main sources for debate on the nature of these new risks and on whether the values involved are adequately addressed in its regulation.17 As a result, the authority of the expert and scientific understanding of these risks itself becomes a political question.18 This new understanding and scholarly debate on risks and their nature both by experts and policymakers has had specific implications for different types of disaster management in the recent decades.19 This is particularly true for the field of public health as a science and policy field, when it concerns the response to a major health emergency.20 Public health as we now understand it in general terms, developed in the nineteenth century as a way of managing and understanding infectious diseases.21 Prior, communicable disease epidemics were largely seen as exogenous, unpredictable or as ‘Acts of God’. However, in the nineteenth century the social conditions of populations, for instance with respect to sanitation, living conditions and circulation, became predicaments that could be statistically analysed and managed through public policies and inoculation strategies. Within modernity’s public health rationale, cost-benefit analysis could be used to legitimise an eradication campaign of polio and small-pox.22 After the Second World War, experts, particularly also biodefence experts, started recognising and hypothesising the possible

14

ibid 364. S Jasanoff, ‘Beyond Calculation: A Democratic Response to Risk’ in A Lakoff (ed), Disaster and the Politics of Intervention (New York, Columbia University Press, 2010). 16 U Beck, Risk Society: Towards a New Modernity 1st edn (London, SAGE Publications, 1992) 22. 17 See further on the regulation of new technologies and the ‘value laden’ expertise underlying EU risk regulation a number of case studies: M Lee, ‘Beyond Safety? The Broadening Scope of Risk Regulation’ (2009) 62 Current Legal Problems 242. Also see M Weimer, ‘Risk Regulation, GMOs, and the Challenges to Deliberation in EU Governance: Politicisation and Scientification as Co-Producing Trends’ in C Joerges and C Glinski (eds), The European Crisis and the Transformation of Transnational Governance (Oxford, Hart Publishing, 2014). 18 Beck, above n 16, 30. 19 S Jasanoff, Learning from Disaster: Risk Management After Bhopal (Pennsylvania, University of Pennsylvania Press, 1994). Also see A Lakoff, Disaster and the Politics of Intervention (New York, Columbia University Press, 2010). 20 See ch 7 in this volume by Scott Greer on the nature of the concept of public health in EU policy. 21 G Rosen, A History of Public Health (Baltimore, John Hopkins University Press, 1958) 168. 22 F Fenner and others, Smallpox and Its Eradication (World Health Organization, 1988). 15

Mixing EU Security and Public Health Expertise

105

limitations of a public health approach to infectious diseases and to gauge the possibility of outbreaks as extraordinary events, without historical precedent.23 In recent decades, alongside a changing understanding of public health risks as a result of technological advances as described by Beck, the rationale and understanding of the risk of pathogens—that have been around in various forms for centuries—also changed and resulted in a new public policy consensus; from public health containment strategies to a ‘preparedness approach’ to infectious disease.24 The underlying idea is that increasing scientific knowledge on communicable diseases has created new threats to public health, for instance in the form of a possibly accidental or deliberate release of dangerous diseases. Prime examples here are the anthrax scare after 9/11 and the recent bird flu virus developed at the Erasmus University Rotterdam and the controversy surrounding publication on its modification in potentially human-contagious forms. This new context of ‘preparedness for the unknown and the deliberate rather than the accidental’ applied to consider disease, reconfigures experts’ understanding in the determination of public health risks.25 The difference between the ‘public health frame’ and the ‘preparedness frame’ in public policy terms is that a preparedness strategy does not use statistical forecasts based on past events. Rather, it uses scenario sketches and simulation models in order to prepare for uncertain future events.26

B. Securitisation and Globalisation: From ‘Public Health’ to ‘Health Security’ The second, interrelated development that has led to a changed expert framing and understanding of public health risks is a move from ‘public health policy’ to ‘health security policy’. The conceptualisation of ‘health security or biosecurity’, rather than ‘public health’, refers to a layering and integration of perceived security threats to human health and general public health concerns. ‘Biosecurity’ or ‘health security’ then is used as a term to recast the effort of public authority in relation to the protection of the health of the population. This move to security is relevant as it signifies a shift in governance discourse.27

23 Collier and Lakoff, above n 8. D Fidler and L Gostin, Biosecurity in the Global Age: Biological Weapons, Public Health, and the Rule of Law 1st edn (Stanford, Stanford Law and Politics, 2007). 24 WK Mariner, GJ Annas and WE Parmet, ‘Pandemic Preparedness: A Return to the Rule of Law’ (2009) 1 Drexel Law Review 341–382. 25 Fidler and Gostin, above n 23. A Kelle, ‘Securitisation of International Public Health: Implications for Global Health Governance and the Biological Weapons Prohibition Regime’ (2007) 13 Global Governance 217. Also see ‘WHO | The World Health Report 2007—A Safer Future: Global Public Health Security in the 21st Century’ (WHO), www.who.int/whr/2007/en/. 26 Collier and Lakoff, above n 8. MBA van Asselt, E Vos and I Wildhaber, ‘Symposium on Critical Infrastructures: Risk, Responsibility and Liability. Some Reflections on EU Governance of Critical Infrastructure Risks’ (2015) 6 European Journal of Risk Regulation 185. 27 A Boin and others, The Politics of Crisis Management: Public Leadership Under Pressure (Cambridge, Cambridge University Press, 2006).

106

Anniek de Ruijter

Biosecurity as Gostin and Fidler use it in the global context describes ‘society’s collective responsibility to safeguard the populations from dangers presented by pathogenic microbes’.28 However, the term also refers to efforts to protect populations against harm caused by biological weapons and biological terrorism and chemical attacks. The public policy rationale behind combining the public health and security strands of policy has been argued to result from the nature of the policy and the institutional actors that (should) be involved in harnessing these types of threats: The emergence of biosecurity reveals that the days when public health and security never intersected as governance tasks have ended. … The policy silos of security and public health are not integrated as they should be in order to respond to the threats that we face today.29

A crucial development in this integration has been the increased focus in the last decade on the ‘terrorist threat’ and an increased global interdependence.30 In Beck’s analysis the risks as they result from modernisation may be seen as unintentional side-effects of society’s decisions. However, terrorism is an intentional ‘bad’. The management of risks has the social function of engendering trust and legitimation in the public power and authority. Terrorism however undermines trust on a global scale. In the words of Beck: ‘Since the dissolution of trust multiplies risks, the terrorist threat triggers a self-multiplication of risks by the de-bounding of risk perceptions and fantasies’.31 This means that the focus on pathogens is not just on their natural occurrence, but shifts to their intentional release.32 At the same time this shift changes the experts involved and the legitimating force of the underlying scientific basis of this policy, since according to Beck, ‘the power of definition of experts has been replaced by that of states and intelligence agencies; and the pluralisation of experts has turned into the simplification of enemy images’.33 Globalisation, beyond terrorism, is also a factor in the recasting of public health as a security issue. The global circulation of people and goods has also increased the concern for pathogens and the responsibility of governments in this regard. In 2005 the International Health Regulations (IHR) were amended, which strengthened the obligations for the World Health Organization’s (WHO) members to notify public health threats.34 Under the IHR a ‘Public Health Emergency of International Concern’ is defined in Article 1 as: An extraordinary event, which is determined, as provided in these Regulations: —— to constitute a public health risk to other States through the international spread of disease and; 28

Fidler and Gostin, above n 23, 122. ibid 10. 30 U Beck, ‘The Terrorist Threat World Risk Society Revisited’ (2002) 19 Theory, Culture & Society 39. 31 ibid 44. 32 ibid 45. 33 ibid. 34 World Health Organisation (2008), International Health Regulations 2005 2nd Ed. 29

Mixing EU Security and Public Health Expertise

107

—— to potentially require a coordinated international response. This definition implies a situation that: is serious, unusual or unexpected; carries implications for public health beyond the affected State’s national border; and may require immediate international action.

Furthermore, in 2001, only a few months after the 9/11 terrorists attacks, a Global Health Security Initiative (GHSI) was launched by the G7, Mexico and the WHO. The initiative particularly focuses on chemical, biological, radio-nuclear threats (CBRN), bioterrorism, and influenza.35 In this respect in there is an international consensus,36 that, in the words of the UN Secretary General: ‘The threats presented by biological weapons and natural disease epidemics weave together to form an independent policy challenge the likes of which we have never seen before.’37 Globalisation however also entails a recasting of what ‘security’ means. Historically security was usually used to describe ‘state security’, ie the responsibility of the state to protect its citizens from harm caused by another state. However, recently there has been a move to the use of the term security in the broader sense of ‘human security’.38 In the words of Anne-Marie Slaughter: ‘From this perspective—of human security—it really doesn’t matter if you die from a bullet or you die from AIDS or you die from hunger. What matters is that you die’.39 Hence, state borders become blurred and health emergencies easily become the responsibility of international efforts that focus on short-term mitigation, rather than long-term structural public health reforms, which are deemed too complex and controversial.40 The recent Ebola crisis is a prime example of this dynamic.41 At the same time those people that are affected by a communicable disease in the ‘security frame’ become the ‘enemy’ and a threat to public health, rather than the disease itself.42

C. Reconfiguring Public Health, Co-producing Executive Power Executive power at the EU level (the European Commission), that is, the legal power to control public instruments for coercion, wealth and information,43 is not a clear-cut, unitary power with a single locus. Rather, it can be described as

35 ‘GHSI—Global

Health Security Initiative’ www.ghsi.ca/english/index.asp. C Calhoun, ‘A World of Emergencies: Fear, Intervention, and the Limits of Cosmopolitan Order’ (2004) 41 Canadian Review of Sociology/Revue canadienne de sociologie 373. 37 United Nations Secretary High Level Panel on Threats, Challenges and Change, ‘A More Secure World: Our Shared Responsibility’ (New York: United Nations)’. p. vii–x. 38 ibid. 39 A-M Slaughter, ‘A New UN for a New Century’ (2006) 74 Fordham Law Review 2961. 40 Collier and Lakoff, above n 8, 369 et seq. 41 Mohini Banerjee and Chelsea Jack, ‘Responding to Ebola: Retrofitting Governance Systems’ The Hastings Center www.thehastingscenter.org/Bioethicsforum/Post.aspx?id=7076&blogid=140. 42 Mariner, Annas and Parmet, above n 24. 43 DM Curtin, Executive Power of the European Union, Law, Practices and the Living Constitution (Oxford, Oxford University Press, 2009) 20. 36

108

Anniek de Ruijter

an accumulation of different layers of both internal (national constitutional law), and external executive power (diplomatic relations), with an administrative and political component.44 It is exercised by national civil servants acting within EU institutional actors, and by EU-level civil servants. Moreover EU executive power compounds political power (leadership, ability to propose policies and legislation), and administrative power (eg legal implementing powers and public revenue distribution).45 Emergencies and crises have long been recognised to ‘side-line’ representative and democratic actors within institutions and strengthen the executive.46 Disasters and crises can trigger an informal or formal response and open a ‘window of opportunity for change’, either changing a dominant policy discourse or regulatory approach.47 This dynamic also affects the role of experts who have a role in determining what is a disaster and the threat level. Security experts in this respect have a different frame of thinking than public health experts such as epidemiologists. Health emergencies and disasters can create ‘focusing events’ and create a change of discourse in a policy area that is dominated by scientists and technical experts, particularly also in policy areas where there is no intense public interest involvement. Birkland describes these ‘focusing events’ as an event that is sudden, relatively rare, can be reasonably defines as harmful or revealing the possibility of greater potential future harms, inflict harms, or suggest harms that are or could be concentrated on a definable geographical or community of interest and that is known to policymakers and the public virtually simultaneously.48

Focusing events can create opportunities for the executive given that disasters may focus the attention of experts and the public, and often there will be a demand for immediate change, which puts the executive at work, given that the emergency itself will give a clear indication that the previous regulatory approach was flawed.49 In the European Union there has been a long existing resistance against EU governance in the field of public health.50 However, emergencies—in the EU the BSE crisis is a notable example—can diminish existing policy-making constraints

44 ibid 29; also see D Curtin and M Egeberg, ‘Tradition and Innovation: Europe’s Accumulated Executive Order’ (2008) 31 West European Politics 639. 45 Curtin, above n 43 38. Also see S Hix, The Political System of the European Union (Basingstoke, Palgrave Macmillan, 2005). 46 D Curtin, ‘Challenging Executive Dominance in European Democracy’ (2013) 77 Modern Law Review 1. 47 SG Knowles, ‘Learning from Disaster?: The History of Technology and the Future of Disaster Research’ (2014) 55 Technology and Culture 773. 48 TA Birkland, Lessons of Disaster: Policy Change After Catastrophic Events (Washington DC, Georgetown University Press, 2006). 49 ibid. J Birkmann and others, ‘Extreme Events and Disasters: A Window of Opportunity for Change? Analysis of Organisational, Institutional and Political Changes, Formal and Informal Responses after Mega-Disasters’ (2008) 55 Natural Hazards 637. 50 TK Hervey and JV McHale, Health Law and the European Union (Cambridge, Cambridge University Press, 2004).

Mixing EU Security and Public Health Expertise

109

as they expose the weaknesses of existing policies and ‘create a sense of urgency’.51 An emergency may become viewed as proof that existing problems will become more serious if no serious steps are undertaken. Hence, during crises, existing constraints on institutional action are normally dropped or lowered to allow for swift action. The BSE crisis for instance broke down the resistance of Member States for more EU power regarding the regulation of food.52 In sum, emergencies often give the executive new solutions to consider, which were not previously possible as long as existing policy was strongly entrenched. But in an emergency, existing policies will suffer from declining support and at the same time there will be more pressure to make a decision and fewer constraints to deal with, particularly when the executive is to take swift action. The next section looks at the reconfiguration of public health and security experts and expertise in the context of the legislative history and the resulting adoption of the Health Threats Decision, and a growing role for the EU executive.

III. The Health Threats Decision: From ‘Public Health’ to ‘Health Security’ The Health Threats Decision defines a ‘serious cross-border threat to health’ in Article 3(g) as a: [L]ife threatening or otherwise serious hazard to health of biological, chemical environmental or unknown origin which spreads or entails a significant risk of spreading across the national borders of Member States, and which may necessitate coordination at Union level in order to ensure a high level of human health protection.

The legal base for the Health Threats Decision is found in Article 168 (5) of the TFEU. Paragraph 5 excludes the possibility for the EU to harmonise Member States’ laws with regard to public health. However, there is a limited role for the EU envisaged in Article 168 for responding to a public health emergency: Union action … shall cover the fight against the major health scourges, by promoting research into their causes, their transmission and their prevention, as well as health information and education, and monitoring, early warning of and combating serious crossborder threats to health.

This means that EU power for responding to a major public health threat is limited to the coordination of Member States responses.53 However, although the Health 51 JTS Keeler, ‘Opening the Window for Reform Mandates, Crises, and Extraordinary Policy- Making’ (1993) 25 Comparative Political Studies 433. 52 E Vos, ‘EU Food Safety Regulation in the Aftermath of the BSE Crisis’ (2000) 23 Journal of Consumer Policy 227. 53 Other treaty provisions that form the basis for related mechanisms on disaster prevention can be found in the context of civil protection (TFEU, art 196), the solidarity clause (TFEU, art 222), EU financial assistance to Member States (TFEU, art 122), and humanitarian aid to third countries (TFEU, art 214).

110

Anniek de Ruijter

Threats Decision stays within the legal parameters outlined in Article 168 of the TFEU, at the same time it allows the Commission to determine that there is a public health emergency (Article 12) and take subsequent ‘immediate implementing measures’, which allow the Commission to adopt temporary measures outside of the use of a comitology Committee.54 Keeping in mind that the structures established by the Decision may also be used in ‘exceptional circumstance, not covered by the Decision’ these legal provisions potentially create significant executive power for public health at the EU level.55 In order to draw an initial institutional map, Table 1 shows the principal institutional actors that come into play in a health emergency response at the EU level. Table 1: EU institutional actors involved in health emergency response Institutional Actor

Role

Legal Nature

World Health Organisation

Global coordination and policy on public health (events)

International Organisation, EU has observer Status in UN

Commission General Secretariat, Crisis management unit

Coordination between and among Commission Services

EU institutional actor

European Commission, DG Surveillance, Early Warning EU institutional actor SANCO, Health Threats and Response, manages Unit -3 information systems and exchange, including contact tracing. European Centre for Disease Control (ECDC)

Surveillance, Early Warning and Response, scientific input/output and information on public health threats. Limited operational capabilities through public health teams. Resorts under DG SANCO

EU Agency

European Medicines Agency (EMA)

Approval vaccines and antivirals, information dissemination, resides under DG SANCO

EU Agency

(continued) 54 Reg (EU) No 182/2011, ‘Regulation (EU) No 182/2011, of the European Parliament and of the Council of 16 February 2011 Laying down the Rules and General Principles Concerning Mechanisms for Control by Member States of the Commission’s Exercise of Implementing Powers’ (OJ L 55/13, 28-2-2011) art 8. 55 See para 9 and art 2 Health Threats Decision.

Mixing EU Security and Public Health Expertise

111

Table 1: (Continued) Institutional Actor

Role

Legal Nature

Early Warning and Response Committee (EWRS)

Information exchange and dissemination and coordination of surveillance and early warning and response on Communicable diseases based on Decision 2119/1998 on early warning and response to Communicable diseases

(Examination)Committee (Health Threats Decision Art. 8)

Health Security Committee (HSC)

Member States’ coordination of public health events, beyond communicable disease, including chemical, biological and radionuclear threats, resorts under DG SANCO

Intergovernmental cooperation (Health Threats Decision Art. 17)

Friends of the Presidency

Horizontal Council Group

Informal and ad hoc

A. Public Health Expertise in the EU: From Epidemiologists to ‘Health Security Experts’ Expertise and particularly also public health experts have been part and parcel of the existence of the modern democracies.56 Experts have a long-standing and significant role in contributing to the legitimacy and acceptability of the EU’s involvement in risk regulation.57 With the growth of the EU’s involvement in a variety of policy matters, the need for expertise in these areas grew as well.58 In the European Union, cooperation in the area of public health was from the very

56 S Jasanoff, The Fifth Branch, Science Advisors as Policymakers (Cambridge MA, Harvard University Press, 1998). A de Swaan, In Care of the State: Health Care, Education and Welfare in Europe and the USA in the Modern Era (Oxford, Oxford University Press, 1988). 57 See Weimer and de Ruijter, ch 1 in this volume, and see G Majone, ‘The Credibility Crisis of Community Regulation’ (2000) 38 Journal of Common Market Studies 273; G Majone, ‘The Regulatory State and Its Legitimacy Problems’ (1999) 22 West European Politics 1; E Vos and M Everson (eds), Uncertain Risks Regulated 1st edn (London, Routledge-Cavendish, 2012); E Vos, ‘EU Committees: The Evolution of Unforeseen Institutional Actors in European Product Regulation’ in C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999). 58 M Egeberg, Å Gornitzka and J Trondal, ‘A Not So Technocratic Executive? Everyday Interaction between the European Parliament and the Commission’ (2014) 37 West European Politics 1; Ã Gornitzka and C Holst, ‘The Expert-Executive Nexus in the EU: An Introduction’ (2015) 3 Politics and Governance 1.

112

Anniek de Ruijter

beginning a logical consequence of increased trade and integration.59 Although with respect to veterinary disease, such as bovine tuberculosis, the Community had already formulated some policy,60 with respect to human communicable disease control the Union’s involvement started in the 1970s. The Union first became involved in the WHO disease monitoring through an exchange of letters in 1972.61 Over the course of the 1970s European surveillance networks were set up for a number of communicable diseases, such as HIV/Aids, tuberculosis, legionella and influenza.62 This eventually led to a European public health research programme that was launched in 1984 on the serious spread and threat of HIV/Aids.63 In 1994 the surveillance networks became the subject of a public health network headed by the national communicable disease surveillance centres to create a unified framework for the surveillance of communicable diseases and training for epidemiological intervention (EPIET) under the auspices of the European Commission.64 The initial network was a bottom-up initiative on the basis of the advice of a Charter group of the heads of public health institutions with responsibility for communicable disease in the Member States.65 However, already since the start of the 1990s epidemiologists across the EU worked together in a network to tackle particular communicable diseases together, primarily for the sake of epidemiological data sharing and exchange.66 A Commission Decision in 1998 formalised these executive-expert networks into the Network Committee and established it as a regulatory comitology committee, composed 59 M McKee, TK Hervey and A Gilmore, ‘Public Health Policies’ in E Mossialos, G Permanand, R Baeten and T Hervey (eds), Health Systems Governance in Europe: The Role of European Union Law and Policy (Cambridge, Cambridge University Press, 2010) 233. Also see SL Greer, ‘The European Centre for Disease Prevention and Control’ (2012) 37 Journal of Health Politics, Policy and Law 101; SL Greer, Everything You Always Wanted to Know about European Union Health Policies but Were Afraid to Ask (World Health Organisation on behalf of the European Observatory on Health Systems and Policies, 2014). 60 EEC Commission, ‘EEC Commission (1964), Veterinary Matters: Harmonisation of Legislation, and Other Activities Newsletter on the Common Agricultural Policy Nr 24’. 61 Exchange of letters between the European Communities and the World Health Organisation laying down the procedure for cooperation between the two organisations; Memorandum defining the arrangements for cooperation between the World Health Organisation and the European Communities (72/725/ECSC, EEC, Euratom) (OJ L 300, 28-10-1982, 20–22). 62 M Liverani and R Coker, ‘Protecting Europe from Diseases: From the International Sanitary Conferences to the ECDC’ (2012) 37 Journal of Health Politics, Policy and Law 913; also L MacLehose, M McKee and J Weinberg, ‘Responding to the Challenge of Communicable Disease in Europe’ (2002) 295 Science 2047.; the surveillance of diseases in the EU currently is disseminated through a scientific, peer-reviewed publication, Eurosurveillance, available at www.eurosurveillance.org/. 63 Community Programme of Research into Aids (OJ C 46, 20-02-1984). 64 J Giesecke and J Weinberg, ‘A European Centre for Infectious Disease?ß£ (1998) The Lancet Vol. 352 (no. 9136). J. Weinberg et al, ‘On Behalf of the Charter Group: Establishing Priorities for European Collaboration in Communicable Disease Surveillance’ (1999) 9(3) European Journal of Public Health 236–40. 65 Weinberg et al, ibid. 66 L MacLehose et al, ‘Responding to the Challenge of Communicable Disease in Europe’ (2002) Science 295 2047–50; also see MR Roberts ‘The European Centre for Disease Prevention and Control: Science and Political Integration in Europe’ (2013) UCL STS Observatory blog at www.blogsuclacuk/ sts-observatory/2013/08/02/the-european-centre-for-disease-prevention-and-control-science-andpolitical-integration-in-europe/.

Mixing EU Security and Public Health Expertise

113

of two r epresentatives from each Member State, usually one epidemiologist and one representative of the Ministry of Health, chaired by DG SANCO (Directorate General for Health, European Commission).67 Under the auspices of the Network Committee, the surveillance of single diseases was consolidated and the Early Warning and Response System (EWRS) was set up.68 The Committee was in charge of determining the case definitions for communicable diseases, data and methods for surveillance and issue guidelines on the countermeasures to be taken in times of emergencies.69 In the EWRS, similarly to the surveillance network, the exchange of information for the operation of this system was legally limited to information on new or unknown communicable diseases or to the list of diseases covered under the annex of Decision 2119/98/EC.70 The Charter group as it was first conceived was a bottom-up network of experts in the field of communicable disease control as a public health matter. However, after the BSE crisis in the 1990s and with the increased perception of EU interdependence for responding to ‘health crises’, in 2004 the EU’s work on communicable diseases was further consolidated with the establishment of the ECDC.71 Important in this respect is that the locus of expertise on communicable disease control that was in the network setting a matter of national experts, was given a European counterpart. Particularly in the EU administrative space this was an important development given that the advice of the ECDC could now legitimise more far-reaching action of the executive actors involved in responding to a public health threat at the EU level, given that the input of science into the policy making at the EU level provided increased autonomy for the executive.72 Hence in 2003 in order to streamline the work in relation to public health threats, in DG SANCO a special administrative unit for Health Threats was established. With the establishment of the ECDC in 2004, the coordination of the surveillance systems and related public health networks from the communicable diseases network committee was handed over to this new agency.73 The ECDC ‘harmonised’ the public health expertise in issuing standard surveillance reports on the specific diseases and managing electronic systems, which allow Member States to upload health data in order for the ECDC to produce the statistics and carry out risk 67 Decision No 2119/98/EC, ‘Decision No 2119/98/EC of the European Parliament and of the Council of 24 September 1998 Setting up a Network for the Epidemiological Surveillance and Control of Communicable Diseases in the Community’ (OJ L 268/1, 03-10-1998). 68 See Decision No 2119/98/EC, art 1; also see MacLehose, McKee and Weinberg, above n 62. 69 Decision No 2119/1998/EC, art 3. 70 ibid art 4. 71 Reg (EC) No 851/2004 of the European Parliament and of the Council of 21 April 2004 establishing a European Centre for Disease Prevention and Control (OJ L 142/1, 30-04-2004). Liverani and Coker, above n 62. See further Ch 2 of A de Ruijter, The Expansion of EU Power in the Field of Human Health (Oxford, Oxford University Press, 2016). 72 Gornitzka and Holst, above n 58. 73 See Reg (EC) No 851/2004 of the European Parliament and of the Council of 21 April 2004 establishing a European Centre for Disease Prevention and Control (OJ L 142/1, 30-04-2004); Decision No 2119/98/EC (1998); Commission Decision 2009/312/EC of 2 April 2009 amending Decision 2000/96/ EC as regards dedicated surveillance networks for communable diseases (OJ L91/27 03-04-2009) at Resolution 7.

114

Anniek de Ruijter

assessments. The EWRS has been operated by the ECDC since 2007, and although the Commission remains responsible for a user’s manager’s authorisation, ie who has access to the information exchange,74 the ECDC otherwise manages the system and the information updates. The Centre has no regulatory powers,75 but it does coordinate and is in permanent contact with Member States, the ECDC, the WHO and Public Health Departments and Centres throughout the world through the above mentioned Global Health Initiative Channel.76 In terms of the reconfiguration of public health risks the 1998 Communicable Diseases Decision to ‘formalise’ the existing networks of national networks of epidemiologists and public health executives into a single European institutional actor, was an important stepping stone for decoupling the different national approaches and cultures to communicable diseases and creating one European ‘best practice’. At the same time the EU involvement remained limited to communicable diseases, particularly to a predetermined list of diseases. Thus, although the reconfiguration of public health expertise in the direction of surveillance and preparedness for communicable disease outbreaks as security threats had also reached the EU level, to a large extent the focus was still on the list of known diseases that formed the backbone of these public policy efforts. This changed rather swiftly after the 9/11 terrorist attacks.

B. Securitisation of EU Public Health Risks The 9/11 attacks in 2001 and the subsequent Anthrax attack played an important role in the securitisation of EU public health. Communicable disease control generally in the EU had little political traction.77 Scott Greer in this respect noted that: Politically, communicable disease control policy is caught in the logic of crisis and collective action: outside of crises, it is hard to find energy for collective action, whereas in crises, countries can sometimes overcome the barriers to collective measures and take actions.78

74 The national permanent representations formally assign the national contact points request the Commission and the EWR network committee for access. 75 See Commission Decision (2009/312/EC) of 2 April 2009 Amending Decision 2000/96/EC as Regards Dedicated Surveillance Networks for Communicable Diseases’ (OJ L91/27, 3-4-2009). Also see Commission Decision 2009/312/EC (2009) at 6. 76 The Global Health Security Initiative (GHSI) is an informal, international partnership between countries to strengthen health preparedness for pandemic influenza amongst other public health threats. This network was an initiative started in November 2001 by Canada, the European Union, France, Germany, Italy, Japan, Mexico, the United Kingdom and the United States. The World Health Organisation serves as an expert advisor to the GHSI. The network is made up of the Public Health Agency of Canada, the Health and Consumer Protection Directorate-General of the Commission, Ministère de la Santé et des Solidarités of France, the Federal Ministry of Health of Germany, the Ministry of Health of Italy, the Ministry of Health, Labour and Welfare of Japan, the Ministry of Health of Mexico, the Department of Health of the UK, the Department of Health and Human Services of the USA, and the WHO. 77 Greer, Everything You Always Wanted to Know about European Union Health Policies but Were Afraid to Ask, above n 59, 57. 78 ibid.

Mixing EU Security and Public Health Expertise

115

In response to the 9/11 attacks the Council immediately created a ‘Health Security Committee’ (HSC) which was an informal group of high level ministerial and health department representatives, principally conceived as being able to bind its Member States as a matter of intergovernmental, diplomatic executive action rather than EU action.79 As outlined above, not long after the creation of the HSC in 2003, Unit C3, a special health threat was established in DG SANCO. This ‘Health Threats Unit’, beyond managing the EWRS system and its management of the Network Committee and the HSC, was also made responsible for the surveillance and early warning for communicable diseases caused as a result of acts of bioterrorism and chemical threats. Furthermore, the unit was given the task to operate a Health Emergency Operations Facility in Luxembourg.80 With the setup of the HSC, beyond the Health Threats Unit in DG SANCO, a new set of experts entered the scene. The HSC, which has been since ‘formalised’ by the Health Threats Decision, was created to provide a setting in which emergency decisions could be taken on a European level. For the larger part of the last decade, the HSC has functioned as an informal, intergovernmental gathering of high level Member States’ ‘health security experts’. The intention was that the Member State representatives in the HSC are authorised by their health ministers to make coordinated decisions and commitments with respect to responding to major health threats.81 However, as practice evolved representatives of DG SANCO and other relevant Commission services and agencies, such as the ECDC and the EMA, also gave input for the HSC.82 The HSC invites experts and there are working groups that reside under the HSC. An important difference between the EWRS Network Committee and the HSC is that it is not mainly public health experts and epidemiologists that form the expert-base. Rather the Committee expects its members to be able to switch between engaging with public health and security issues. Beyond the scientific aspects of the pathogen or threat to health the Committee assesses the political,

79 ‘European Council, Declaration by the Heads of State or Government of the European Union and the President of the Commission. Follow up to the September 11 Attacks and the Fight Agianst Terrorism (19 October 2001, Brussels) (SN4296/2/01 REV 2)’. European Commission, ‘European Commission, Programme of Cooperation on Preparedness and Response to Biological and Chemical Agent Attacks (Health Security)’ Luxembourg, 17 December 2001 (G/FS D(2001) GG). 80 The Health Emergency Operations Facility is located in Luxembourg and is used for the management of alerts and emergencies notified by Member States. During an emergency situation the response of the Commission, Member States, and Agencies residing under the Commission including the liaison with international organisations such as the WHO are coordinated from this facility. European Commission, DG SANCO, The Commission Health Emergency Operations Facility: for a coordinated management of public health emergency at EU level, 2007, available at www.ec.europa.eu/ health/archive/ph_threats/com/preparedness/docs/heof_en.pdf. 81 See Communication from the Commission to the Council on transitional prolongation and extension of the mandate of the Health Security Committee in view of a future general revision of the structures dealing with health threats at the EU Level (COM (2006) 699 final); Council of the European Union, 2786th Council Meeting Employment, Social Policy, Health and Consumer Affairs, Brussels, 22 February 2007 (6226/07 Presse 23). 82 de Ruijter, above n 71.

116

Anniek de Ruijter

social and economic implications European health emergency decisions would have.83 It is meant to take decisions in an emergency, keeping in mind the interrelated systems of the economy, critical infrastructure and interlinkages of crucial sources and systems for sustaining human life.84 In the words of a Member State representative in the Council Working Group on Public Health, the HSC: ‘is about assessing the situation and acting’.85 In the HSC there is no divide between the ‘expert’ and the ‘executive’, or between ‘risk assessment’ and ‘risk-management’. The guise of security and the possibility of having to take immediate action mean that these divisions could be a ‘time-consuming threat’ in itself. Two Member State representatives described the HSC in the following manner: It is not a pure expert group, you also have experts group like: people meeting on cancer, … and then you have on the other side the formalised council working groups and in between you have such things as the Health Security Committee, where it is not only experts, it is about governmental questions, what is the government doing, but not in this EU legislative sense …86 The idea is that Member States send high-level representatives. If you want the Health Security Committee to really be able to decide on something … you put someone there that has a direct line with your minister. Since there are very important, politically sensitive issues if it is a crisis. I mean, how many ministers had to resign because of this kind of crises? It is very sensitive and once that people are dying for example, your minister, politically, is being asked questions almost daily. It’s fear. The national parliament is feared. The press is alert. Citizens or civil society groups put pressure on the minister. If you want the Health Security Committee to reach something, you need a representative who has the authority and the possibility to engage his minister. Otherwise it is useless, to have an agreement in the Health Security Committee when you have representatives who are not covered politically by the minister.87

After 9/11 a number of public health threats such as SARS, bird flu, swine flu, E-coli, and volcanic ash cloud strengthened the role of the HSC and its mandate became broader.88 The outbreak of swine flu was a turning point with respect to the way EU response was orchestrated. In 2005 pandemic preparedness plans had been implemented in many Member States on the basis of WHO guidelines and with the

83 ‘European Commission, Commission Staff Working Document, Health Security in the European Union and Internationally’ (SEC(2009)1622 Final) Brussels, 23/11/2009, above n 5. 84 Collier and Lakoff, above n 8. 85 Respondent 3 (MS Representative Working Party on Public Health in the Council, 2010). 86 Respondent 2 (Deputy Permanent Representative for Health in the Council, 2010). 87 Respondent 4 (MS Representative Working Party on Public Health in the Council, 2010) and further: ‘For us this is really the forum where Member States can talk and discuss and decide. So, for us it is very clearly a forum with an intergovernmental way of working. And it should be, because we are talking about Member States competences once again’. 88 Commission Communication, ‘Communication from the Commission to the Council on Transitional Prolongation and Extension of the Mandate of the Health Security Committee in View of a Future General Revision of the Structures Dealing with Health Threats at the EU Level’ (COM (2006) 699 Final).

Mixing EU Security and Public Health Expertise

117

g uidance of the ECDC.89 When the first news came that there was a swine flu outbreak in Mexico, the EU immediately became highly involved in response planning, particularly also given that a ‘fast track’ procedure for the authorisation of pandemic vaccines by the European Medicines Agency had to be activated, and without the declaration of an ‘Emergency of International Concern’ by the WHO, the fast track procedure could not be activated according to the EU regulations on medicines.90 Over the course of the unfolding of the swine flu pandemic, the HSC and the Network Committee, both acting under the auspice of the European Commission’s Health Threats Unit, were increasingly working together when decisions were made by the HSC that formally fell under the responsibility of the Network Committee, and where the HSC basically had no formal powers.91 Thus, as an institutional matter the public health experts and security experts were merged and the public health powers available at the EU level were used to formalise security measures that without the context of the Network Committee would be ‘mere’ diplomatic agreements. The EU’s response to the swine flu outbreak represents the first time a public health emergency of pandemic scale was coordinated at the EU level. Member States were highly vulnerable due to the availability of vaccines being dependent on the EU central authorisation procedures and with respect to the information available to the public, given the plethora of open access online sources.92 As a result, this health emergency opened a ‘window of opportunity’ for more far-reaching legislation at the EU level. After an elaborate ‘evaluation’ and ‘learning-lessons’ period,93 the Commission in 2011 proposed a new Decision on Health Threats, that would integrate the public health and security framework completely. 89 European Commission, ‘European Commission, Interim Document, Technical Guidance on Preparedness Planning for Public Health Emergencies, April 2005’. 90 European Commission Staff Document, ‘European Commission Staff Document, Regulatory Process for the Authorisation of Antiviral Medicines and Vaccines in the Protection against Pandemic Influenza H1N1 2009, Accompanying Document to the Communication to the Commission to the Council, the European Parliament and the European Economic and Social Committee and the Committee of Regions, Pandemic Influenza H1N12’ 2009 (COM(2009) 481-SEC(2009)119 Final). 91 See for instance HSC/EWRS Statement on priority groups for vaccination, ‘EU Health Security Committee (HSC) / Early Warning and Response System (EWRS)HSC EWRS Statement on Influenza A(H1N1) 2009: Target and Priority Groups for Vaccination 25 August 2009’, available at http:// ec.europa.eu/health/archive/ph_threats/com/influenza/docs/hsc_ewrs_statement_en.pdf. 92 EPSCO Council Conclusions, ‘Council Conclusions on Influenza A/H1N1 Infection 30 April 2009, 2965th Employment, Social Policy, Health and Consumer Affairs Council Meeting’ (Luxembourg, 12 October 2009). 93 European Medicines Agency, ‘European Medicines Agency, Pandemic Report and Lessons Learned Outcome of the European Medicines Agency’s Activities during the 2009 (H1N1) Flu Pandemic’ (29 April 2011) available at www.ema.europa.eu/docs/en_GB/document_library/Report/2011/04/ WC500105820.pdf; European Commission Staff Working Document, ‘Commission Staff Working Document, Vaccination Strategies against Pandemic (H1N1)2009 Accompanying the Communication from the Commission to the European Parliament, the Council, The European Economic and Social Committee and the Committee of the Regions (Pandemic H1N1 2009) (COM(2009)481-SEC (2009)1189 Final)’; Commission Communication, ‘Communication from the Commission to the Council, The European Parliament and the European Economic and Social Committee and the Committee of Regions, Pandemic (h1N1), Commission Staff Working Document on Joint Procurement of Vaccine against Influenza A H1N1 2009’ (COM (2009)481—SEC (2009)1188).

118

Anniek de Ruijter

IV. Conclusion This chapter analysed the reconfiguration of expertise in the area of public health and the mixing of public health policy with a security framework. The science and expertise behind these different policy strands is different. Yet, mixing security with public health in the Health Threats Decision co-produces increasing EU executive power with respect to the response to a health emergency. The lines of the executive and the expert, the ‘risk-assessor’ and the ‘risk manager’ become blurred in the Health Security Committee that is created by the Health Threats Decision. The scope of the field of public health is widened and enlarged to an ‘all hazard’ field of threats, whereby chemical attacks, bioterrorism and environmental threats are all on a par with communicable diseases. The public health experts are reconfigured as ‘health security experts’ in the HSC, in which and the mode of governance is intergovernmental, rather than part of a—however imperfect— system of procedural checks of executive action as the examination procedure that the Network Committee works under the earlier Communicable Diseases Decision of 1998 prescribed. The basic expert rationale of what major health threats require in the Health Threats Decision is that of preparedness for possible unknown risks. Given that it entails an open-ended article that allows the regulatory instruments, actors and surveillance and response mechanism to be legally used in case of emergencies unforeseen in the Decision itself, it creates a vast potential for the EU executive to engage in the response to anything that may be framed as a ‘health threat’.94 Communicable disease control and prevention have been long-standing champions of the welfare state. The public health experts that came together in the 1990s at the EU level to exchange ideas and strategies in order to cope with these public health challenges as a network of experts, have now been overtaken by a new rationality. And this rationality of public health preparedness, threats and security put the power of expertise and knowledge about risks in the hands of ‘health security’ specialists. This paradigm creates a new route for the EU to affect the welfare systems: vaccination schemes are no longer a way for national governments to protect its population from disease, but are reconfigured as a security instrument.95 In other words, the shift from public health expertise to security expertise also shifts the attention towards critical infrastructures, which entails a much broader ‘security view’ that also keeps in mind business continuity, competitiveness and other threats to security rather than the public health view that is mainly focused on the life and health of the population.96

94 European Commission, ‘European Commission, Interim Document, Technical Guidance on Preparedness Planning for Public Health Emergencies, April 2005’. 95 European Commission Staff Working Document, above n 93. 96 See Flear, above n 2.

Mixing EU Security and Public Health Expertise

119

Furthermore, the focus on security in the field of public health also shifts the attention and resources away from the public health problems that produce far greater morbidity and mortality, and can only be solved by socio-political interventions that focus on pervasive and on-going health problems rather than focusing on unlikely, extreme and rare events.97 The Health Threats Decision in this regard creates a stronger base for the EU executive to coordinate and respond to public health events in comparison to previously available powers. However, beyond the growing role of the EU executive in the field of public health as co-produced by a newly-blended landscape of public health-security expertise and experts, the public health policies of the Member States are also affected by the growing role of the EU level of the expert-executive in this respect.98 The role of the EU in this respect may affect the lives of individual Europeans significantly.99

97 ibid.

98 E Baekkeskov and P Öberg, ‘Freezing Deliberation through Public Expert Advice’ (2016) Journal of European Public Policy 1. 99 de Ruijter, above n 71.

120

7 Constituting Public Health Surveillance in Twenty-first Century Europe SCOTT L GREER

I. Introduction Public health surveillance, the collection of public health data for public purposes, is at the core of public health and the modern state. It is invasive, expensive, sometimes coercive, used to justify all sorts of public actions, and fraught with all manner of conflicts about personal data, privacy, state power and public action.1 It is therefore political. Its political nature, however, is easier to deduce than to see in the everyday work, publications and statements of people involved in an apparently low-profile, scientific and cumulative public enterprise. In the centre of a vortex of concerns about technology, privacy, disease, power and liberty, we find an oddly calm world of writing about epidemiological best practice. That world has shaped, among other institutions, the developing European Union (EU) engagement in public health. In other words, European public health surveillance is a case of science and formal politics co-producing each other, with the state co-opted into an image of a scientific public health practice and science co-opted into public sector risk and health politics, ‘solving problems of disorder’ by defining surveillance and what it means to do it well.2 The EU’s authority in the area depends in large part on this developing understanding of surveillance, shared among networks that it often helped to promote. Public health surveillance is an order-making concept that underpins expert claims to resources in a field whose most prominent characteristics are diversity, organisational complexity, and ambiguity about borders. The practice of what we 1 P Baldwin, Disease and Democracy: The Industrialized World Faces Aids (Berkeley and Los Angeles, University of California Press, 2005); AL Fairchild, ‘The Right to Know, the Right to Be Counted, the Right to Resist: Cancer, Aids, and the Politics of Privacy and Surveillance in Post-War America’ (2015) 3 Journal of Medical Law and Ethics 45. 2 S Jasanoff, ‘The Idiom of Co-Production’ in S Jasanoff (ed), States of Knowledge: The Co-Production of Science and Social Order (London, Routledge, 2006).

122

Scott L Greer

might call public health surveillance has existed for a long time. Reporting requirements for diseases and efforts by authorities to monitor health and incidence of communicable disease go back a long way, but the concept of public health surveillance (or ‘biosurveillance’), was defined in the midst of Cold War America. An order-making concept classifies activities, prioritising some and de-emphasising others, and enables the organisation of bureaucracies, research, training, careers and authoritative knowledge. It contains the promise that there is a best practice for these diverse activities, that there can be a scientific and technical underpinning for such a best practice and that certain people, organisations and activities are required to do it well. The act of naming creates a concept with some normative power that enables and energises some people while shaping the bounds of subsequent policy debate and science. The concept, by ordering a set of activities, enables the co-creation of scientific authority and warrants public policy action. This chapter discusses surveillance as a concept and a political campaign whose goal is to establish an ostensibly technical and scientific procedure. The co-production of power and expertise is visible in the intertwined efforts to define and justify both surveillance and the executive agencies that do it, at the domestic, European and international levels. The actual practice of surveillance is obscured by this work of co-production. We know very little about what is happening, what knowledge and activity is really being organised by the concept of surveillance, or what the effects on practice of the political campaign might be. The European level is one of the clearest and most dramatic cases of the co- production of surveillance available, since it involves explicit harmonisation of technical issues and the creation of named administrative capacity such as the European Centre for Disease Control and Prevention. The EU’s development of surveillance policy is in many ways a perfect case of co-creation, taking the internationally normative order-making concept of surveillance as a warrant for European action to promote convergence in details, practices and structures. Advocates of European action can draw on the presumed need for convergence on scientific surveillance, while advocates of surveillance can draw on European actions for normative reinforcement, technical help, and even a measure of formal Europeanisation.

II. Public Health Surveillance: A Powerful International Concept Surveillance is an order-producing concept. The concept renders apparently disordered activities coherent, and empowers those who appeal to it in their efforts to expand or reform surveillance practices. As such it has an irreducible component of imprecision, normativity and aspiration. The remarkably consensual definition of surveillance in public health justifies much activity, and might

Public Health Surveillance

123

produce international convergence when coupled with concrete action. It does not tell us much about the organisations, ideas and practices it would modify or replace, but it does seem to organise much activity and provide a base for policy recommendations.

A. Defining Public Health Surveillance Public health surveillance has a clear meaning in public health textbooks, ie: ‘(a) the systematic collection of pertinent data (b) the orderly consolidation and evaluation of these data and (c) the prompt dissemination of results to those who need to know’.3 This definition dates back to a crucial article by Alexander Langmuir of the United States (US) Centers for Disease Control and Prevention (CDC).4 Refinements of the definition have mostly either expanded the concept to include issues other than communicable diseases or, as the World Health Assembly agreed in 1968, a responsibility to act upon the data generated.5 A commitment to action was something Langmuir had resisted. For him and many public health practitioners since, surveillance stopped when information arrived at the right authorities.6 The public health definition nestles comfortably within the broader definition of surveillance found in the academic field of surveillance studies: ‘[t]he focused, systematic, and routine attention to personal details for purposes of influence, management, protection or direction’.7 Public health surveillance is part of a wide family of surveillance practices ranging from fiscal surveillance conducted by the International Monetary Fund (IMF) to intelligence agencies’ surveillance such as that famously conducted by the US National Security Agency (NSA) to the ubiquitous cameras found in cities such as London.8 It therefore is involved in all the human rights issues that come whenever governments start to collect data on people.9 The practical work of surveillance is also described in very similar terms in most public health textbooks and the publications of agencies that do it. It involves reporting by health providers (doctors, hospitals, laboratories, and possibly 3 SM Thacker, ‘Historical Development’ in LM Lee and others (eds), Principles and Practice of Public Health Surveillance (Oxford, Oxford University Press, 2010) 5. 4 AD Langmuir, ‘The Surveillance of Communicable Diseases of National Importance’ (1963) 268 The New England Journal of Medicine 182. 5 World Health Organization, ‘Report of the Technical Discussions at the Twenty-First World Health Assembly on ‘National and Global Surveillance of Communicable Diseases’ (World Health Organization, 1968). 6 BCK Choi, ‘The Past, Present, and Future of Public Health Surveillance’ (2012) 2012 Scientifica. 7 D Lyon, Surveillance Studies: An Overview (Cambridge, Polity, 2007) 14. 8 GT Marx, Windows into the Soul: Surveillance and Society in the Age of High Technology (Chicago, University of Chicago Press, 2015). 9 AL Fairchild and R Bayer, ‘In the Name of Population Well-Being: The Case for Public Health Surveillance’ (2015) Journal of Health Politics, Policy and Law; GJ Annas and WK Mariner, ‘(Public) Health and Human Rights in Practice’ (2015) 41 Journal of Health Politics, Policy and Law 129; Baldwin, above n 1.

124

Scott L Greer

thers such as schools or social welfare agencies). They collect information on the o appearance of a given disease in the population they serve, or sometimes broader information, and report it to a central agency which can then publish the aggregate information (eg, there were 10 cases of influenza in a given town last week) and take appropriate action. Every system must balance sensitivity (ability to pick up signs of a problem) and specificity (ability to provide useful information). Not only is there a trade-off in many cases between sensitivity and specificity, there is also such a thing as too much specificity or sensitivity. Communicable diseases of particular contagiousness or severity are ‘notifiable’, meaning that there is a formal obligation on providers or others to report a case to the state. Under the 2005 International Health Regulations (IHR), states then have to report evidence of a major public health threat to the World Health Organization (WHO). The 2005 IHR replaces a much older set of regulations that listed notifiable diseases in the treaty itself and therefore were unable to respond effectively to newly important diseases such as SARS or Ebola. Instead, the 2005 IHR invites states, the WHO, and to a limited extent others, to focus on threats to health in determining what should be reported.10 The 2005 IHR revisions mark a shift towards an ‘all-hazards’ approach in which the magnitude of the health threat, rather than its appearance on a preset list, justifies surveillance and action. From the perspective of communicable disease control, what is interesting is understood (including by the ECDC) as ‘epidemic intelligence’, in which information is categorised into signals, which can come through traditional, often state-run, surveillance or something else (eg media monitoring, academics, civil society).11 The signal is assessed by the central agency and, if it appears meaningful, is then investigated, with the investigation leading to the next action. This is an approach geared to identifying sudden outbreaks, with diverse instruments of different sensitivity and specificity from following Twitter to traditional reporting by doctors. Much weight is born by the investigation, since Twitter or the media are often overly sensitive and can spread rumour as quickly as fact. Surveillance of non-communicable diseases and population health, as well as some of the less frightening communicable diseases, start to shade into what looks like ordinary epidemiological research. The current mix of new and old surveillance data sources include explicit surveillance systems such as: vital registration; exposure registries; surveys, useful for issues such as broad health behaviours; d isease registries; medical, laboratory and pharmacy records and perhaps notification requirements on those providers; environmental monitoring of topics 10 S Rushton, ‘Gloal Governance Capacities in Health: WHO and Infectious Diseases’ in A Kay and OD Williams (eds), Global Health Governance: Crisis, Institutions and Political Economy (Basingstoke, Palgrave Macmillan, 2009); DP Fidler, ‘From International Sanitary Conventions to Global Health Security: The New International Health Regulations’ (2005) 4 Chinese Journal of International Law 325. 11 M French and E Mykhalovskiy, ‘Public Health Intelligence and the Detection of Potential Pandemics’ (2013) 35 Sociology of Health & Illness 174; C Paquet and others, ‘Epidemic Intelligence: A New Framework for Strengthening Disease Surveillance in Europe’ (2005) 11 Euro surveillance: bulletin européen sur les maladies transmissibles 212.

Public Health Surveillance

125

such as pollution and heat; criminal justice information; and media sources,12 not to mention the very old method of people who notice something and try to bring it to others’ attention. Calling all of these different things ‘public health surveillance’ looks like an attempt to bring order to what would otherwise be a diverse and opportunistic range of data collection efforts. Opportunism is recast as methodological creativity and diversity is recast in the framework of a single concept.

B. The Conventional History of the Concept of Public Health Surveillance Like all models, this model of surveillance has a history. Its birth in the very s pecific context of Cold War America was followed by quick worldwide diffusion through the WHO and use around the world. The speed of its diffusion as a concept is very striking, and contributes to both its power and its ability to obscure variations in practice. The popularisation of the word ‘surveillance’ in a public health context is most often attributed to Alexander Langmuir, founder of the Epidemic Intelligence Service of the US CDC and creator of the definition given above.13 Prior to the work of Langmuir and his colleagues, surveillance in health care had most often referred to monitoring individual patients who had been exposed to a pathogen. There was no blanket term for what we now call surveillance. Since Langmuir, surveillance has meant monitoring the health and threats to health of whole populations. The practice of notifying diseases might go back a long way, but organising it into the concept of surveillance, making it CDC’s core business, and trying to export the concept, started with Langmuir and his colleagues. A historian of CDC, indeed, identifies the CDC’s commitment to surveillance as its distinctive characteristic, part of its contribution to shaping global public health thinking, and a reason for its survival as a respectable US federal agency during the Cold War.14 For all the vital statistics traditions of Snow and Graunt, which are routinely invoked in public health textbooks, or for that matter the centuries of surveillance practices in European states, the concept of surveillance as we find it in textbooks has at its root the work of an agency in the US. Langmuir’s definition, the CDC’s actions and the context of Cold War America also bound surveillance to a broader set of issues somewhat removed from traditional public health—notably biological warfare research, both offensive and

12 LM Lee and others, ‘Evolving Challenges and Opportunities in Public Health Surveillance’ in Lee and others, above n 3; Choi, above n 6. 13 Langmuir, above n 4. 14 EW Etheridge, Sentinel for Health: A History of the Centers for Disease Control (Oakland, University of California Press, 1992) xvii.

126

Scott L Greer

defensive.15 Langmuir was deeply involved in the US national security establishment and was involved in its bioweapons and biodefence bureaucracies. CDC’s transformation from a World War Two era agency focused on malaria control and zoonoses into the much broader agency of today took place in part because of Langmuir’s successful push to focus its activities on surveillance and preparation for biological warfare.16 As with the origins of surveillance, the story of its turn towards securitisation reflected the comparative difficulty of securing funding for public health, especially the kind of invasive and expensive (and theoretically unlimited) programmes that public health surveillance might entail. In Cold War Washington, the Communist threat justified many things, including the growth of the CDC and its surveillance thinking. Security threats continue to play a big role in surveillance politics.

C. Debating the Concept of Surveillance Debate about public health surveillance is frequently focussed on the concept of surveillance as defined above and its intellectual history rather than about what happens in practice under the name ‘surveillance’. There is a very substantial degree of empirical overlap in the two literatures that frequently discuss public health surveillance—the public health surveillance literature epitomised by Thacker and his colleagues at CDC and reviewed by Choi,17 and the critical research tradition found in disciplines such as anthropology and countries such as Canada or the UK18 and often connected to the field of surveillance studies.19 The sources are similar and mostly secondary, as is the focus on a few people such as Alexander Langmuir and Donald Henderson. Both literatures tell a story that starts with brief mention of Europeans such as John Graunt, John Snow and Adolphe Quetelet but shifts, via Langmuir, to the US, and from there to definitions and practices of surveillance derived from Langmuir and diffused worldwide through, for example,

15 L Fearnley, ‘Epidemic Intelligence: Langmuir and the Birth of Disease Surveillance’ (2010) 3 Behemoth 36; E Fee and TM Brown, ‘Preemptive Biopreparedness: Can We Learn Anything from History?’ (2001) 91 American Journal of Public Health 721. 16 Etheridge, above n 14. 17 Lee and others, above n 3; SB Thacker, RL Berkelman and DF Stroup, ‘The Science of Public Health Surveillance’ (1989) Journal of Public Health Policy 187; Choi, above n 6. 18 MA French, ‘Woven of War-Time Fabrics: The Globalization of Public Health Surveillance’ (2009) 6 Surveillance & Society 101; A Lakoff and SJ Collier, Biosecurity Interventions: Global Health and Security in Question (New York, Columbia University Press, 2008); Fearnley, above n 15; L Weir and E Mykhalovskiy, Global Public Health Vigilance: Creating a World on Alert (Abingdon, Routledge 2009); A Bashford (ed), Medicine at the Border: Disease, Globalization and Security, 1850 to the Present (Basingstoke, Palgrave Macmillan, 2006). 19 KD Haggerty and RV Ericson, ‘The New Politics of Surveillance and Visibility’ in KD Haggerty and RV Ericson (eds), The New Politics of Surveillance and Visibility (Toronto, University of Toronto Press, 2006); GT Marx and GW Muschert, ‘Personal Information, Borders, and the New Surveillance Studies’ (2007) 3 Annual Review of Law and Social Science 375; M French and G Smith, ‘“Health”Surveillance: New Modes of Monitoring Bodies, Populations, and Polities’ (2013) 23 Critical Public Health 383.

Public Health Surveillance

127

WHO and the ECDC. In both literatures there is an element of the great-man theory of history, with a strong emphasis on the strategies and intentions of a few people. The first disagreement between the two literatures relates to theoretical emphasis—the extent to which authors wish to foreground the implication of surveillance in issues such as security, terrorism, privacy and state building. In the public health accounts, Langmuir and Henderson are impressive, even heroic, figures who saved lives as they shaped public health bureaucracy and thinking. In the critical accounts, they are effective politicians and exemplary figures of the relationship between state building, security and surveillance who are noteworthy for their deep links into some of the scarier areas of Cold War history such as bioweapons development. The second disjunction in the literatures is in their normative agendas. Broadly, the public health literature is in favour of surveillance as an essentially diagnostic tool for a population-focused field: ‘it is the essential activity that allows other public health actions to have effect’.20 A review of international health organisations’ frameworks for good public health, however they else they varied, found that they all started with good surveillance.21 How can public health be like a doctor to society if it knows nothing of society’s ills? The critical theorists are more prone to identify surveillance as a tool or structure of power and, for the Foucauldians among them, biopolitics. They are more likely to question the motivations and actions of the bureaucracies involved. In both literatures, however, the concept of surveillance and a few global intellectual-bureaucratic leaders such as Langmuir are centre stage. Previous activities that we now call surveillance, the concrete work of surveillance, many of the polemics around it, and variation within surveillance, are downplayed. But is that empirically correct? Debating the proper interpretation of a stylised intellectual-political history is not the same thing as studying the actual practices and people who use that stylised history.

III. The Empirical, Understudied, Diversity of Surveillance Diversity in actual surveillance, both divergence in theoretical ideal from the US CDC tradition associated with Langmuir, and in practice from any theoretical ideal, is much less studied. In secondary literature it must often just be inferred from either public health researchers’ discussions of data limitations or their policy 20 DF Stroup, R Brookmeyer and WD Kalsbeek, ‘Public Health Surveillance in America: A Framework’ in R Brookmeyer and DF Stroup (eds), Monitoring the Health of Populations: Statistical Principles and Methods for Public Health Surveillance (Oxford, Oxford University Press, 2004). 21 U Laaser and H Brand, ‘Global Health in the 21st Century’ (2014) 7 Global Health Action.

128

Scott L Greer

recommendations. In this sense the conventional public health writers’ accounts are often much more revealing than critical literature, which valorises resistance and diversity but focuses on the concept and its advocates. What we do not have is surveillance of surveillance systems. We thus do not know what the impact of this process of co-producing surveillance as expertise and surveillance bureaucracies actually might be. The spread of the concept and concrete practices of surveillance are inexplicable without CDC entrepreneurship (including internationally at the WHO), professionalism and not least money to finance state activities. But the CDC does not, of course, control what countries do internally- including within the United States. Researchers working in the US frequently note in p assing the patchiness of different systems within the US. For example, why does the state of Michigan participate in most occupational health surveillance networks but not in environmental health networks?22 The CDC can fund and cajole, but it cannot formally direct public health activities that are still lodged within US states and counties.23 The status and implications of the model in the US are therefore rather unclear. Attention to variation is particularly important because surveillance does not happen at the abstract level. Much of epidemiology is concerned with technical issues such as case definitions, measures and efforts to balance goals such as sensitivity and specificity. Even comparatively solid statistics such as death certificates can mislead. Many cholera deaths are first logged as ‘other’ before it is clear that there is an outbreak.24 Data systems are frequently not interoperable. Case definitions and measures can differ subtly. We have no idea why Latvia had five times more reportable diseases than France25 let alone whether one country’s list is ‘better’ in some way. Even extracting those lists is a surprisingly difficult research challenge. There is an agreed EU list of 52 notifiable diseases, with a procedure for designating them, but a commitment to notify them at the EU level does not come with a de facto commitment to operate surveillance systems that reliably identify their incidence and prevalence. This superficially explored and largely unexplained variation might be put down to historical legacies. But syndromic surveillance is developing in Europe in just as diverse a way26 despite the existence of a much larger ‘European’ public health infrastructure and intellectual community that co-produces ‘European’ public health expertise. 22 JD Shire and others, ‘Advances and Current Themes in Occupational Health and Environmental Public Health Surveillance’ (2011) 32 Annual Review of Public Health 109. 23 R Katz and S Rosenbaum, ‘Challenging Custom: Rethinking National Population Surveillance Policy in a Global Public Health Age’ (2010) 35 Journal of Health Politics Policy and Law 1027; SL Greer and PM Singer, ‘The United States Confronts Ebola: Suasion, Executive Action, and Fragmentation’ (2015) 12 Health Economics, Policy and Law 81–104. 24 SE Davies, ‘The Challenge to Know and Control: Disease Outbreak Surveillance and Alerts in China and India’ (2012) 7 Global Public Health 695. 25 R Reintjes, ‘Variation Matters: Epidemiological Surveillance in Europe’ (2012) 37 Journal of Health Politics, Policy, and Law 955. 26 S Medina and others, ‘Comparing Findings from Syndromic Surveillance Systems at a European Level’ (2014) 6 Online Journal of Public Health Informatics.

Public Health Surveillance

129

This variation fits into a larger point, better established in the literature, which is that there is nothing like a European or global model of public health. There is a great deal of variation in almost every possible axis. The few studies that have been conducted all find extraordinary diversity of public health systems—in their formal goals, structures and funding27 not to mention variation within decentralised countries.28 Evaluation is difficult because many surveillance systems do not have stated objectives.29 There is a great deal of historical and bureaucratic continuity over time—responses to cholera in the nineteenth century were much like responses to AIDS in the twentieth30—and very different relationships over space between healthcare systems, medicine, privacy, state bureaucracies and courts.31 In particular, healthcare systems have a longstanding and often difficult relationship with public health. Much public health, such as lifestyle advice and vaccinations, can be provided at the individual level by doctors. Strong healthcare systems coupled with effective local government (eg for environmental and food safety issues) can allow a country to operate with very limited formal public health bureaucracies, as is the case in much of Western Europe. In short: it is always wise to assume that any given proposition about ‘how public health works’ is confined to the speaker’s country of origin, and possibly only to textbooks in that country. Surveillance is no exception.

IV. Changing Justifications of Surveillance Public health surveillance, then, is an order-making concept that underpins expert claims to resources in a field whose most prominent characteristics are diversity, organisational complexity and ambiguity about borders. Atop a variety of systems with different histories sits an abstract set of criteria and objectives repeated in 27 L MacLehose and others, ‘Communicable Disease Outbreaks Involving More Than One Country: Systems Approach to Evaluating the Response’ (2001) 323 British Medical Journal 861; Reintjes, above n 25; R Reintjes and others, ‘Benchmarking National Surveillance Systems: A New Tool for the Comparison of Communicable Disease Surveillance and Control in Europe’ (2007) 17 European Journal of Public Health 375; MG Baker, S Easther and N Wilson, ‘A Surveillance Sector Review Applied to Infectious Diseases at a Country Level’ (2010) 10 BMC Public Health 332; H Elliott, DK Jones and SL Greer, ‘Mapping Infectious Disease Control in the European Union’ (2012) 37 Journal of Health Politics, Policy, and Law 935; D Porter, The History of Public Health and the Modern State (Amsterdam, Editions Rodopi, 1994); P Baldwin, Contagion and the State in Europe, 1830–1930 (Cambridge, Cambridge University Press, 2005); B Rechel, H Brand and M McKee, ‘Financing Public Health in Europe’ (2013) 75 Gesundheitswesen (Bundesverband der Arzte des Offentlichen Gesundheitsdienstes (Germany)) e28. 28 SL Greer, ‘Intergovernmental Governance for Health: Federalism, Decentralization and Communicable Diseases’ in S Greer, M Wismar and J Figueras (eds), Strengthening Health System Governance: Better Policies, Stronger Performance (European Observatory on Health Systems and Policies/ Open University Press, 2015). 29 JA Drewe and others, ‘Evaluation of Animal and Public Health Surveillance Systems: A Systematic Review’ (2012) 140 Epidemiology and Infection 575. 30 Baldwin, above n 1. 31 SL Greer and M Mätzke, ‘Bacteria without Borders: Communicable Disease Politics in Europe’ (2012) Journal of Health Politics, Policy and Law 887–915.

130

Scott L Greer

public health textbooks with an ostensible lineage tracing directly to the US in the early Cold War. The diversity of these systems, within and between countries, is striking, as is the limited attention given to even cataloguing let alone comparing and explaining them. Much of what attention they do get is evaluation rather than analysis. Public health surveillance is an expensive, labour-intensive, data-heavy, intrusive practice that can always be done better. It is also one that involves all sorts of different kinds of invasion of privacy, as we see especially with the collection of information about stigmatised diseases. Trusting public authorities with information about communicable diseases will always be contentious in democracies, and it is not always wholly clear that they need or will do something constructive with the data in many cases. This means that justifying surveillance—linking its development and evolution to broader public policy concerns—is a constant preoccupation for its practitioners and theorists. Proposals for new articulations between surveillance, health systems and the state are constant. This section presents three key proposals: technology, globalisation and security. They provide the case for policy and budgets for public health surveillance, since merely organising knowledge is not enough.

A. Technological Change: Circumventing Diagnosticians It is unsurprising that a data-centric activity like surveillance would be caught in the excitement and development of new data science practices around the world. The first element of the challenge, or possibility, is the rise of ‘big data’. Big data is the use of data sets that are so large that they are qualitatively different from previous data sources, which might not have been collected with the public health purpose in mind, and which are not structured like older public health data sets.32 It offers the possibility not just of better identification and analysis, but also of avoiding the expensive and apparently anachronistic practices associated with traditional surveillance. The problems with big data for public health are the problems with big data for most purposes: it is difficult to use data collected and structured for one purpose to achieve another. The second technological development is the rise of ‘syndromic surveillance’.33 This is an approach that disrupts the traditional surveillance focus on reporting diseases, with surveillance data following and reporting the judgements of health care professionals. Syndromic surveillance’s emergence came after a major 1993 cryptosporidium outbreak in the city of Milwaukee was especially noticed by pharmacies (which had sold out of stomach remedies) days before public health

32 SI Hay and others, ‘Big Data Opportunities for Global Infectious Disease Surveillance’ (2013) 10 PLoS Medicine e1001413; E Vayena and others, ‘Ethical Challenges of Big Data in Public Health’ (2015) 11 PLoS Computational Biology e1003904. 33 A Ziemann, ‘Syndromic Surveillance: Made in Europe’ (Maastricht, Maastricht University, 2015).

Public Health Surveillance

131

authorities became aware of the problem. The idea this event spawned was that the pharmacies and employers had real-time, valuable public health information that was already contained in digital form in the pharmacies’ order books and employers’ records of absenteeism, and that if the city had been able to tap into this information it would have been able to take action much earlier. Syndromic surveillance was subsequently adopted and formalised as a system by New York City and the concept has spread, particularly in the wake of the 9/11 attacks in 2001. It inverts the traditional logic of surveillance by tracking syndromes (eg upset stomachs) rather than diagnoses, and appealing to data sources such as pharmacies that can identify widespread syndromes rather than provide diagnoses.34 Syndromic surveillance promises a technical solution to the blockage that can arrive when patients do not go to reporting providers, providers fail to diagnose or providers fail to report. Circumventing the providers’ monopoly has all sorts of advantages. Just as older forms of surveillance rely on reporting by hospitals and laboratories which, as bureaucracies, were more reliable about transmitting data than individual doctors, syndromic surveillance promises to cut out the need for a potentially faulty reporting process altogether. The problem is that syndromic surveillance cannot replace medical diagnosis or microbiology. It does not identify the pathogen and cannot identify diseases which are contracted in small numbers, such as Ebola or anthrax. It is unlikely that syndromic surveillance could even pinpoint influenza.35 Every fluctuation in behaviour could possibly trigger an investigation while a much more frightening pathogen could start to spread unnoticed. In epidemiological terms, its sensitivity can be too high, alerting too often, or too low, missing rare but dangerous issues, and its specificity is almost always too low. That means it is probably most used as part of a toolkit, particularly for epidemic intelligence alerting.36 The question is whether its expense and priority, born of its novelty and technology, is appropriate to its place in the toolkit. These two distinctive movements, of big data and syndromic surveillance, come together in, for example, Google’s effort to track the incidence and prevalence of influenza by monitoring the rate and geographical distribution of influenzarelated searches (‘Flu Trends’). There has been some controversy over Flu Trends,37 notably when after an accurate and useful initial year Google then substantially overestimated influenza relative to the CDC’s more traditional surveillance data.

34 L Fearnley, ‘Redesigning Syndromic Surveillance for Biosecurity’ in Lakoff and Collier (eds), above n 18; L Fearnley, ‘Signals Come and Go: Syndromic Surveillance and Styles of Biosecurity’ (2008) 40 Environment and Planning A 1615. 35 A Ziemann and others, ‘Meeting the International Health Regulations (2005) Surveillance Core Capacity Requirements at the Subnational Level in Europe: The Added Value of Syndromic Surveillance’ (2015) 15 BMC Public Health 107. 36 JW Buehler and others, ‘Situational Uses of Syndromic Surveillance’ (2009) 7 Biosecur Bioterror 165. 37 S Cook and others, ‘Assessing Google Flu Trends Performance in the United States During the 2009 Influenza Virus a (H1n1) Pandemic’ (2011) 6 PLoS One e23610; D Butler, ‘When Google Got Flu Wrong’ (2013) 494 Nature 155,

132

Scott L Greer

That said, the possibilities continue to fascinate public health and data specialists, including those in Europe.38 Google abandoned Flu Trends in 2015, but at the same time began supplying information about searches for serious sexually transmitted diseases to researchers so that they could identify potential outbreaks.39 The third technological development is the effort to turn Electronic Health Records (EHR) into tools of surveillance.40 In principle, most people will have an EHR and it will contain most health-relevant information about them, including their medications, previous treatments, chronic conditions and interactions with providers. In principle, use of the EHR for surveillance circumvents the problem of spotty reporting by providers and avoids the logistical problems of communicating each case. Public health information derived from EHRs would be useful for a wide range of public health tasks, such as tracking vaccinations, identifying clusters of disease and even evaluating the largely unknown postmarket performance of drugs and devices (currently dependent on very imperfect voluntary reporting by doctors and patients). In some European countries, EHRs are already used in this basic way. It would be possible to find out, in theory, that years of taking a given combination of medicines creates a previously unknown risk. Given the remarkable, if unsurprising, weakness of reporting on side effects of drugs or medicines,41 the idea of being able to track the effects of devices and medicines over time by querying medical records is very appealing from a safety and quality perspective. In general, an additional attraction of EHRs is that, unlike big data or syndromic surveillance, they promise to incorporate clinical, diagnostic information to make accurate conclusions, without requiring additional information or calculations.42 The problems with EHRs are less obvious, but comprehensive research efforts have identified several challenges. Notably: it is difficult to establish any functioning EHR; data entry is often spotty and imperfect; data exchange is difficult; privacy

38 N Rosenkötter and others, ‘Non-Infectious Events under the International Health Regulations (2005) in Europe–a Case for Syndromic Surveillance’ (2014) 35 Journal of Public Health Policy 311; Ziemann and others, above n 35. 39 MC Jaklevic, ‘Google Helps Researchers Track Down Worst STD Cases’ Kaiser Health News (21 December) http://edition.cnn.com/2015/12/11/health/google-stds-data-tracking/; D Lazer and R Kennedy, ‘What We Can Learn from the Epic Failure of Google Flu Trends’ Wired (2015). 40 GS Birkhead, M Klompas and NR Shah, ‘Uses of Electronic Health Records for Public Health Surveillance to Advance Public Health’ (2015) Annual Review of Public Health 345. 41 AA Daemmrich, Pharmacopolitics: Drug Regulation in the United States and Germany (Chapel Hill NC, University of North Carolina Press, 2004); E Lopez-Gonzalez, MT Herdeiro and A Figueiras, ‘Determinants of under-Reporting of Adverse Drug Reactions’ (2009) 32 Drug Safety 19; L Hazell and SAW Shakir, ‘Under-Reporting of Adverse Drug Reactions’ (2006) 29 Drug Safety 385; RG Hauser, ‘Here We Go Again—Another Failure of Postmarketing Device Surveillance’ (2012) The New England Journal of Medicine 10.1056/NEJMp1114695; PV Rajan, DB Kramer and AS Kesselheim, ‘Medical Device Postapproval Safety Monitoring Where Does the United States Stand?’ (2015) 8 Circulation: Cardiovascular Quality and Outcomes 124; Daniel M Fox and Diana M Zuckerman, ‘Regulatory Reticence and Medical Devices’ (2014) 92 Milbank Quarterly 151. 42 D Heisey-Grove and others, ‘Using Electronic Clinical Quality Measure Reporting for Public Health Surveillance’ (2015) 64 MMWR: Morbidity and Mortality Weekly Report 439.

Public Health Surveillance

133

concerns are serious; and there are many contradictory patient, commercial, vendor and professional interests standing in the way of data sharing. Furthermore, EHRs are already polyvalent texts, and their content and structure are shaped by day-to-day power struggles.43 Adding surveillance needs to the contests over their structure and contents might not work for surveillance agencies or anybody else.

B. Globalisation Appeal to new transportation technologies is a longstanding argument for b etter surveillance—how are we to know if a traveller, or a shipborne rat, has introduced a pathogen without surveillance? This sentiment supports the argument that transportation technologies have ‘annihilated distance’, as phrased by Eric H obsbawm and invoked by Barnett and Sorensen.44 As a result, the world is newly vulnerable to ‘emerging infectious diseases’ (EIDs) such as SARS or Ebola and new influenza strains that can move quickly from Liberia to Texas, Mexico to Scotland, or south China to Toronto. There is an element of anachronism in these arguments, undoubtedly. Just as surveillance happened before Langmuir’s order-making concept was published in 1963, governments have policed their borders against the threat of disease from abroad. The tendency of diseases to cross borders has long been known, and much of the history of public health is a story of quarantines, letters from Venetian ambassadors certifying disease-free ships, and such. The threat of new circulating risks would justify surveillance, but also reorient it from domestic towards global communicable diseases.45 The idea is that surveillance of EIDs which present a risk to us all is a ‘global public good for health’ (GPGH). The term GPGH46 is another interesting example of public health rhetoric. It includes an argument for international public expenditure. A public good that we all want but are tempted to let somebody else supply while we reap the benefits, and is therefore a textbook economic argument for public action. Calling something a GPGH is therefore a case for global action to provide it, and communicable disease surveillance seems to fit the definition of a global public good.47

43 M Berg and G Bowker, ‘The Multiple Bodies of the Medical Record’ (1997) 38 The Sociological Quarterly 513. 44 T Barnett and C Sorenson, ‘Infectious Disease Surveillance in the United States and the United Kingdom: From Public Goods to the Challenges of New Technologies’ (2011) 36 Journal of Health Politics Policy and Law 165. Also SE Davies, A Kamradt-Scott and S Rushton, Disease Diplomacy: International Norms and Global Health Security (Baltimore, John Hopkins University Press, 2015). 45 P Calain, ‘Exploring the International Arena of Global Public Health Surveillance’ (2006) 22 Health Policy and Planning 2. 46 D Woodward and RD Smith, ‘Global Public Goods and Health: Concepts and Issues’ in R Smith and others (eds), Global Public Goods for Health: Health Economic and Public Health Perspectives (Oxford, Oxford University Press, 2003). 47 Barnett and Sorenson, above n 44.

134

Scott L Greer

Whether communicable disease surveillance is actually a global public good, rather than a club good,48 and whether policymakers think about GPGHs are both empirical and unanswered questions. The concept of the EID, well analysed by Weir and Mykhalovskiy,49 is easy to criticise on the same grounds of blindness to international inequalities. EIDs might be ‘emerging’ in rich countries and endemic in poor ones. The concept directs attention to ‘circulating risks’ rather than the ones confined to poor countries.50 It was nonetheless an effective political tool for promoting communicable disease surveillance and response in the face of the persistent tendency to focus on biomedicine or non-communicable diseases, and once again finds its ideational entrepreneurs in US academia and the CDC. The concept of the EID points to the reinforcement and generalisation of epidemic intelligence as a focus, as well as the globalisation of surveillance. It places emphasis on the ever-changing nature of diseases, whether they are jumping across national or species borders. A system to provide early warning and response (epidemic intelligence) seems like a natural response. The essentially political category of EIDs—diseases emerging from the point of view of human medicine in the richer countries—justifies surveillance. Response is of course a thornier issue than surveillance. The recent history of global outbreaks, notably SARS and Ebola and to a lesser extent influenza, might be read as showing that a global web of surveillance is sufficient for rich countries’ purposes. Their various domestic surveillance capacities are sufficient to track people who might be infected and their pharmaceutical industries and health care systems can, with the warning provided by global surveillance, cope with the spread of disease. However, the Indonesian Government (longstanding host of a major CDC station) has pointed out forcefully that it is asked to supply viral samples for free that are later turned by for-profit firms into vaccines that Indonesia cannot afford.51 The payoff for participation by poorer countries is often unclear, and the ‘global public good’ looks more like a club good for countries that can afford to respond. Even when rich countries respond to outbreaks like Ebola in poor countries, they might only focus on issues that suit their interests such as EIDs and their circulation rather than helping to build local health systems. Much of the published literature on surveillance practice and politics, and much of the change, focuses on this reconfigured globalisation of public health

48 A public good is non-excludable and non-rival: it can be consumed by multiple people at once and it is not possible to stop people enjoying it. A club good is excludable and non-rival: one that can be consumed by multiple people at once, and it is possible to stop people enjoying it. To say that public health is a public good is to say that we all enjoy it or none of us do. To say that it is a club good is to say that jurisdictions or people can buy it without sharing it with others. 49 Weir and Mykhalovskiy, above n 18. 50 G Blouin Genest, ‘World Health Organization and Disease Surveillance: Jeopardizing Global Public Health?’ (2014) 19 Health 595. 51 R Gatter, ‘The New Global Framework for Pandemic Influenza Virus- and Vaccine-Sharing’ in I Glenn Cohen (ed), The Globalization of Health Care: Legal and Ethical Issues (Oxford, Oxford University Press, 2014).

Public Health Surveillance

135

s urveillance (most of which is epidemic intelligence: a system for collecting alerts rather than constant effective monitoring). In particular, focus is placed on the extent to which the law (2005 IHR) and practice permit surveillance and signals from a wider variety of actors than the traditional states. Governments are often unaware of or lie about the presence of a notifiable disease within their borders, and the world learns about outbreaks through specialist networks such as ProMedMail (which brought SARS to light) or the media. In Davies’ words, there is much writing on the technical capacity of governments to comply with the IHR requirements, but there is less writing about, and just as much variation in, their political capacity to report.52 In an effort to circumvent this problem the WHO and others can rely on reporting from a broader variety of sources—a shift away from the state-centrism of the older reporting system that David Fidler calls a ‘post-Westphalian’ global public health.53 Implementation of the revised International Health Regulations demands political and technical capacities few countries have. IHR implementation should have created worldwide understanding of surveillance priorities, while richer countries have invested in expanded global surveillance and CDC models have diffused around the world.54 Given the diversity of surveillance, much attention has also gone into attempts to start harmonising the content and communication of the Regulations,55 with predictably complex politics of all sorts.56 The construction of a much better global alerting system, which is the core of the IHR, does not always involve changes in the organisation, resources or priority of the surveillance systems that underpin reporting. In fact, expanding the range of those who receive alerts in global epidemic intelligence to include journalists or NGOs is partly a response to the political and technical problems of public health surveillance systems. The Ebola outbreak of 2014, of course, showed that the WHO is not perfectly adept at collecting and making timely decisions in response to such information.57 Perhaps the WHO would have been better off sticking with Henderson’s definition of surveillance instead of adding ‘response’ to the definitional characteristics of surveillance. At least in the initial stages, the challenge in addressing the Ebola outbreak was not related to surveillance, but to response.

52

Davies, above n 24. Fidler, ‘Sars: Political Pathology of the First Post-Westphalian Pathogen’ (2003) 31 Journal of Law, Medicine and Ethics 485. 54 S Binder and AM Laing, ‘The National Public Health Institutes of the World- Update’ (2009) 30 Journal of Public Health Policy 467. 55 C Castillo-Salgado, ‘Trends and Directions of Global Public Health Surveillance’ (2010) Epidemiologic Reviews 10.1093/epirev/mxq008. 56 SE Davies, J Youde and R Parker, ‘The Shared Responsibility of Disease Surveillance’ (2012) 7 Global Public Health 667. 57 Ebola Interim Assessment Panel, Report of the Ebola Interim Assessment Panel (Geneva, World Health Organization, 2015); S Moon and others, ‘Will Ebola Change the Game? Ten Essential Reforms before the Next Pandemic. The Report of the Harvard-Lshtm Independent Panel on the Global Response to Ebola’ (2015) The Lancet 10.1016/S0140-6736(15)00946-0. 53 D

136

Scott L Greer

C. The Challenge of Securitisation It is easy to focus on the civilian aspects of surveillance. History is full of cases of local governments that engaged in surveillance in order to protect the health of their populations, with doctors and civilian officials in the leading positions.58 But surveillance achieved its contemporary meaning in a context entangled with security bureaucracies.59 ‘Securitisation’—the construction of an issue as one of security60—is an important political act with both benefits (money and attention) and costs (diversion into worries about immediate hostile threats, entanglement with security bureaucracies that might be fierce competitors for resources). Securitisation is particularly helpful in establishing surveillance of any sort, because the promise of trading privacy for security is apparently more attractive than other potential trade-offs (such as trading privacy for research, or trading privacy for reduced non-communicable disease incidence). After the Cold War, the case for surveillance started to change, with an increasing focus on EIDs, and then with a return to securitisation after 2001. The 9/11 terrorist attacks on the US were followed by still unsolved anthrax attacks. In a country newly attentive to threats of terrorism and making broad changes to its domestic and international security apparatus, the result was a major push for increased counter-bioterrorism funding and the expenditure of a great deal of money on the issue. US policy emphasised the importance of ‘uninterrupted global surveillance’ and the alignment of global health surveillance efforts with US national security interests.61 EU Member States with a tradition of military engagement, particularly the UK, also began to reconsider and strengthen their existing biosecurity infrastructure (eg microbiology and surveillance of diseases such as anthrax), including the creation of a new agency in the UK that rationalised and centralised existing resources.62 The risk of securitisation, repeatedly stated, perhaps most eloquently by historians,63 is that the promise of security from biosurveillance will be traded 58 DM Fox, ‘Social Policy and City Politics: Tuberculosis Reporting in New York, 1889–1900’ (1975) 49 Bulletin of the History of Medicine 169; AA Rusnock, Vital Accounts: Quantifying Health and Population in Eighteenth-Century England and France (Cambridge, Cambridge University Press, 2002). 59 D Fidler and L Gostin, Biosecurity in the Global Age Biological Weapons, Public Health, and the Rule of Law (Palo Alto, Stanford University Press, 2008). 60 B Buzan, O Wæver and J de Wilde (eds), Security: A New Framework for Analysis (Boulder, Lynn Rienner, 1998). 61 The key legislation is the Pandemic and All-Hazards Preparedness Act, PAPHA, passed in 2006 and reauthorised in 2011. M Mair, B Maldin and B Smith, ‘Passage of S 3678: The Pandemic and All-Hazards Preparedness Act’ (2007) 5 Biosecur Bioterror 72; R Morhard and C Franco, ‘The Pandemic and All-Hazards Preparedness Act: Its Contributions and New Potential to Increase Public Health Preparedness’ (2013) 11 Biosecur Bioterror 145. Unrelated is the Global Health Security Agenda, the Obama administration’s tool to strengthen global health surveillance and response through a US-led alliance that circumvented the WHO and put the CDC at the centre of the networks. 62 Department of Health, Getting Ahead of the Curve: A Strategy for Combating Infectious Diseases Including Other Aspects of Health Protection (London, Department of Health, 2002). 63 Fee and Brown, above n 15; NB King, ‘The Influence of Anxiety: September 11, Bioterrorism, and American Public Health’ (2003) 58 Journal of the History of Medicine and Allied Sciences 433.

Public Health Surveillance

137

off against not just privacy but also effective public health. US experiments with smallpox vaccination under the Bush administration did not inspire confidence in the prioritisation produced by a security-focused agenda.64 Expenditure on security-led programmes such as the US ‘BioWatch’ system, led by the Department of Homeland Security and composed mostly of expensive devices that ‘sniff ’ for airborne toxins, also suggests that a great deal of money can be spent on technology in a way that does little for health or security.65 Europeans have largely been spared such exercises in technological futility. It would be interesting to explore why.

V. Europeanisation: Health Security and State Security Against this background of shifting justifications for resources, legitimacy and authority, Member States have increasingly authorised the EU to operate in public health surveillance. The European public health practice that is being Europeanised, to whatever extent, is extremely diverse and is neither well mapped nor justified.66 The combination of Europeanisation, with its implicit promise of convergence, and an order-making, global and putatively scientific concept like surveillance, means that policymakers and advocates can focus on the concept rather than the disorder and diversity. New data and technology, new globalised health threats, and new security threats are invoked differently in different political systems. In the EU, they have been invoked by a variety of actors to promote the basic Langmuirian concept of public health surveillance, framing its promotion as a response to all three, a practical form of European action, and an opportunity to engage in ‘capacity building’ that converts the diverse European scene into something more standardised and compatible with the concept. In other words, a project for Europeanisation is yoked to the concept of surveillance in order to advance the creation of isomorphic bureaucracies and functions across Member States. From the perspective of EU studies, the whole sector of communicable disease control broadly confirms general theories about how EU integration develops in technical market-related regulatory areas, with ECDC’s activities, history and 64 Committee on Smallpox Vaccination Program Implementation and others, The Smallpox Vaccination Program: Public Health in an Age of Terrorism (National Academies Press, 2005). 65 Committee on the Effectiveness of National Biosurveillance Systems BioWatch and Public Health Surveillance:: Evaluating Systems for the Early Detection of Biological Threats: Abbreviated Version (Washington, National Academies Press, 2011). One problem afflicting DHS in health is that while it is the principal coordinating department for emergencies, it has relatively little to contribute to health emergencies except coordination, and its culture is a far cry from that of public health. 66 MacLehose and others, above n 27; Reintjes, above n 25; Reintjes and others, above n 27; Baker, Easther and Wilson, above n 27; Elliott, Jones and Greer, above n 27; Porter, above n 27; Baldwin, above n 27); Rechel, Brand and McKee,above n 27.

138

Scott L Greer

olitics resembling other EU agencies such as the European Food Safety Authority, p European Medicines Agency, the European Aviation Safety Agency and also other areas of the EU work such as Health Technology Assessment that have not been formalised in an agency.67 From a public health perspective, it fits within a small international movement to advocate for a model of public health agency (called a public health ‘institute’ by proponents) that looks much like CDC with optional in-house laboratory capacity68 and creates the enticing possibility of a prestigious scientific and actionoriented public health actor where there had been ministries, bureaucrats, doctors and local government. As ever, the influence of the CDC hangs over European thinking on the topic (not just in the ECDC, which joins a Chinese CDC and even now a prospective African CDC in using the name). Core features of the ECDC, however, such as its small staff and reliance on networking, make it quite different from the CDC and quite typical of EU agencies.69 The story of EU public health policy has been widely discussed.70 In short, the EU has been slowly developing a structure for joint responses to health issues, starting with cancer in the 1980s. Over time, it has funded more and more EUwide research and benchmarking activities, creating networks dedicated to data collection, research, analysis and comparison. With the creation of ECDC, it created a dedicated agency to improve surveillance, coordination and response across the continent. ECDC is a networking agency without in-house laboratory capacity that operates through networks of Member State ‘competent authorities’. Its activities focus on information sharing, coordination of the work of others’ laboratories and field epidemiology teams, and capacity-building. Creating competent bodies that are actually competent in a practical sense often means helping them to adhere to shared European standards, promoting a European model of surveillance. The fate of competent practices that do not fit within the European approach is largely unresearched—perhaps variation continues, unreported, and perhaps convergence is taking place. The affinity between the EU agency model, the model of CDC, the practice of surveillance as defined in textbooks and the framing of divergence as gaps, all point 67 SL Greer and O Löblová, ‘European Integration in the Era of Permissive Dissensus: Neofunctionalism and Agenda-Setting in European Health Technology Assessment and Communicable Disease Control’ (2016) Comparative European Politics 10.1057/cep.2016.6. 68 S Binder and others, ‘National Public Health Institutes: Contributing to the Public Good’ (2008) 29 Journal of Public Health Policy 3. 69 C Waterton and B Wynne, ‘Knowledge and Political Order in the European Environment Agency’ in Jasanoff (ed), above n 2; B Rittberger and A Wonka, ‘Introduction: Agency Governance in the European Union’ (2011) 18 Journal of European Public Policy 780. 70 SL Greer and P Kurzer (eds), European Union Public Health Policies: Regional and Global Perspectives (Abingdon, Routledge, 2013); A de Ruijter, Uncovering European Health Law (Amsterdam, University of Amsterdam, 2013); SL Greer, ‘The European Centre for Disease Prevention and C ontrol: Hub or Hollow Core?’ (2012) 37 Journal of Health Politics, Policy, and Law 1001; M Steffen (ed), Health Governance in Europe: Issues, Challenges and Theories (Abingdon, Routledge, 2005); SL Greer and others, Everything You Always Wanted to Know About European Union Health Policy but Were Afraid to Ask (Brussels, European Observatory on Health Systems and Policies, 2014).

Public Health Surveillance

139

to the EU as a significant new part of the co-creation of public health k nowledge and power. Publishing standardised data, building capacity, standardising case definitions and advising on preparedness strategies, insofar as they work, standardise a given relationship across Europe between public health agencies, states and societies. To whatever intellectual power public health texts, WHO and CDC had in Europe, the EU adds cognitive Europeanisation of surveillance71 and an agency dedicated to promoting networks that can harmonise technical as well as politico-scientific practice. In this framing, the ECDC’s work promotes a certain model of Europe and public health surveillance, giving new force to the long-stable definition of surveillance. The ECDC’s work combines epidemic intelligence with aggregation of data from public health surveillance. The European Surveillance System (TESSy) and the Annual Epidemiological Report (AER) carry out ongoing surveillance; Member States that identify a case of a disease, with or without cross-border significance, provide the information to ECDC and it is reported via TESSy and the AER. Potential cross-border health threats are treated via the Early Warning and Response System (EWRS). The core of ECDC’s work is epidemic intelligence, which resembles the outbreaks monitoring of an agency such as EFSA. Beyond that, it is also engaged in a wide variety of projects aimed at collecting and diffusing broader population health data about topics such as obesity. Behind the ECDC stands the Commission and the committees it services whose related functions are now structured, following an all-hazards approach, as ‘health threats’, incorporating human and animal diseases as well as other threats to health and funding more communication mechanisms such as Medisys.72 If ECDC is a specialist public health agency on the evolving global template, the Commission is an executive agency, with a broader mission and tendency to use the language of preparedness, threats and hazards rather than public health. Thus, for example, the Commission is the intervening agency in the process of selecting diseases as notifiable at the EU level. This Commission-ECDC relationship fits a basic template in which the concept of a specialist agency is co-produced along with a realm of specialist knowledge while the Commission occupies a more clearly political role. Insofar as there is political acceptance of the existence of a technical realm that can be relatively depoliticised, there is a case for an agency to do it. Insofar as there is an agency doing it, there is a case that the area can be depoliticised. Of course, the agency’s actions help to shape the border between what is and what is not seen as political and therefore fit for closer executive and legislative involvement, and whatever the Commission does is likely to be seen as more political simply because the

71 S Guigner, ‘L’européanisation Cognitive De La Santé: Entre Imposition Et Persuasion’ in O Baisnee and R Pasquier (eds), L’europe Telle Qu’elle Se Fait (Paris, CNRS Editions, 2007). 72 The treaty base for cooperation on nuclear weapons and related radiological issues is lacking and there is little interest in revising the EURATOM treaty, but Member States have agreed to cooperate on radio-nuclear incidents as well.

140

Scott L Greer

Commission does it. The civilian nature of the EU then means that an i nteresting mix of formal and informal mechanisms are required to ensure coordination between health and security administration during health emergencies.73 It seems possible that differences have arisen between Europe and the US in the interplay of security and surveillance.74 Both have adopted an all-hazards approach. And in both cases the argument for surveillance together with a strong classic public health approach, as supported by CDC and ECDC, means that there is a new category of risk incorporating bio- (and other kinds of) terrorism, systemic breakdowns (such as a transport system collapse), and major disease outbreaks. The US, however, had a ready-made military and domestic security infrastructure, one long linked to surveillance that was substantially reinforced after 2001, not least through the creation of the Department of Homeland Security. By contrast, in 2001 there was relatively little European capacity in biosecurity. The security approach to surveillance, which does not just link surveillance to the security apparatus but distances it from the preferred logics of public health professionals, is weaker. Where biosecurity approaches existed they tended to be found in the countries with some military ambitions (especially Britain and France). Furthermore, the EU, unlike NATO or a domestic government, is an essentially civilian organisation. The reconstruction and Europeanisation of surveillance systems, therefore, involved an essentially civilian EU working with countries that mostly had little or no pre-existing military capacity in surveillance. EU surveillance interrelates to another very common form of EU action, namely benchmarking and Member State comparisons. Much EU activity focused on data collection and benchmarking tries to achieve policy change and convergence.75 The European Semester is probably the most consequential result of such effort, but there has also been a great deal of technical activity in health that harnesses specialist networks and Member State data to specific ends such as addressing particular causes of ill health.76 The ECDC’s Annual Epidemiological Report and the European Core Health Indicators are examples of cognitive Europeanisation that uses surveillance data to press for both better public health and better surveillance and even redefines diseases.77

73

A de Ruijter, Uncovering European Health Law (Amsterdam, University of Amsterdam, 2013). See Anniek de Ruijter’s contribution on this topic, ch 6. CF Sabel and J Zeitlin (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2010); SL Greer, ‘The Weakness of Strong Policies and the Strength of Weak Policies: Law, Experimentalist Governance, and Supporting Coalitions in European Union Health Care Policy’ (2011) 5 Regulation & Governance 187. 76 HAK Elliott, ‘European Union Information Infrastructure and Policy’ in Greer and Kurzer (eds), above n 70; F Briatte, The Europeanization of Health System Performance: The Eurocare Study and C ancer Control in England (Lille, CERAPS, 2010); F Briatte, ‘The Politics of European Public Health Data’ in Greer and Kurzer (eds), above n 70; M Steffen, ‘The Europeanization of Public Health: How Does It Work? The Seminal Role of the AIDS Case’ (2012) 37 Journal of Health Politics, Policy, and Law 1057. 77 RCR Taylor, ‘The Politics of Securing Borders and the Identities of Disease.’ (2013) 35 Sociology of Health and Illness 241. 74

75

Public Health Surveillance

141

VI. Conclusions: Surveying Surveillance Public health surveillance is an order-making concept that co-produces surveillance bureaucracies, authorising them as trustworthy and scientific even as they produce the concept of surveillance. Making such order is apparently not enough to justify expenditure, and so its advocates have harnessed it to a variety of security-based arguments from health emergencies to bioterrorism. But its order-making properties, which draw the attention of critical as much as public health scholars, actually obscure its descriptive accuracy. The closest we come to understanding surveillance is a set of public health articles doing ‘gap analysis’ and ‘benchmarking’ and public health reports. They generally produce the concept of a scientific and apolitical surveillance and advocate for more convergence and resources, but their data shows the diversity of practice while rendering it as a sort of policy failure that should be addressed. Order-making concepts can be more useful in co-production of power and expertise than in illuminating the concrete practice they might influence. Part of their effectiveness is precisely in obscuring inconvenient disorder. Thus we have the emblematic order-making of the AER. At first glance it is extremely impressive, with its European concern and ever-increasing European standardisation of concepts and definitions. But the attentive reader will then spot the lengthy explanatory notes on data in each section showing the very serious and complex limitations in the data for most countries. There might be many unexploited stories behind those omissions, variations and partial coverage, but the co-production of European public health surveillance frames them only as obstacles. The model of public health surveillance first formulated by Langmuir has spread remarkably quickly, bringing order to activities of great diversity that in some cases went on for hundreds of years without being joined in a single concept. We need a far better understanding of the histories, operation and politics of surveillance systems in order to understand the impact of changing models of surveillance, transplanted international notions of good practice and political justifications for surveillance. Endlessly rehashing the story of the model from critical and supportive perspectives is not enough; first we need to know how realistic the model is before we can understand the impact of its changes.

142

8 Behavioural Expertise and Regulatory Power in Europe HOLGER STRASSHEIM

I. Introduction1 On 14 June 2012, ‘European Voice’ published a statement by Paola Testori Coggi, at that time Director-General of the European Commission’s health and consumers unit DG SANCO (now DG SANTE), asserting that the ‘European Commission supports the study and application of behavioural insights for policymaking. Indeed, it has been using this approach since 2008’.2 Stressing the importance of behavioural insights for EU policymaking, Coggi was replying to an op-ed by Alberto Alemanno, professor of European Union law and regulation at HEC Paris, who had criticised the EU for not having shown ‘a commitment to integrating behavioural research into its policymaking’.3 In her response, Coggi pointed to several examples of behaviourally informed policies at the EU, emphasising that ‘behavioural economics can play an important role in achieving the Commission’s aim of smarter legislation’.4 The fact that EU Director-General Coggi felt obliged to publicly respond to professor Alemanno’s comment is an interesting illustration of the role modern media plays in science-policy interaction. More importantly, however, it is a symptom of the significance behavioural insights and interventions have gained all over Europe and beyond.5 In the United Kingdom, the rise of behavioural economics 1 The author wishes to thank the editors of this book and the participants of the workshop on ‘Regulating Risks in the European Union: the co-production of expert and executive power’ (Amsterdam, 21–22 May 2015) for various helpful comments and suggestions. 2 PT Coggi, ‘The European Commission supports the study and application of behavioural insights for policymaking’ (POLITICO, 13 June 2012, updated 2014) www.politico.eu/article/ behavioural-insights-in-the-commission/. 3 A Alemanno, ‘Nudging Europe’ European Voice (London, 16 May 2012) 9. 4 Coggi, above n 2. 5 A Alemanno and A-L Sibony (eds), Nudge and the Law: a European Perspective (Oxford, Hart Publishing, 2015); JS Lourenço, E Ciriolo, SR Almeida and X Troussard, Behavioural Insights Applied to Policy (European Commission, 2016); P Lunn, Regulatory Policy and Behavioural Economics (OECD, 2014).

144

Holger Strassheim

is associated with the creation of the Behavioural Insights Team (BIT) in 2010, an organisation that has gained both national and international influence by experimentally developing and disseminating behavioural interventions. It has become a paradigmatic example, inspiring the creation of similar project units in other countries, such as the Social and Behavioural Sciences Team at the United States White House Office of Science and Technology Policy (OSTP), the BVA nudge unit in France or the project group ‘Wirksam Regieren’ (‘governing e ffectively’) at the German chancellery. In Denmark and Norway, bottom-up networks such as ‘iNudgeyou’ or ‘greenudge’ engage in research projects focusing mainly, but not exclusively, on environmental policy and public health, eg reducing food waste, changing littering behaviour or decreasing smoking. The Organisation for Economic Co-operation and Development (OECD) has published a report on ‘Regulatory Policy and Behavioural Economics’6 and the World Bank7 discusses in its most recent development report the advantages and complexities of b ehavioural and psychological interventions in public policy. On the European level, one of the most prominent examples of using behavioural approaches is the EU Directive on Consumer Rights, which limits the use of pre-ticked boxes for online purchases. In several Directorate Generals of the Commission, such as Health and Food Safety (DG SANTE), Justice (DG JUST), Communications Networks, Content and Technology (DG CONNECT), Climate Action (DG CLIMATE), Environment (DG ENV) and the Taxations and Customs Unit (TAXUD), behavioural and psychological studies have been issued and integrated in the development of regulations. The Commission’s in-house science service, the DG Joint Research Centre (JRC), has recently tasked a Foresight and Behavioural Insights Unit (FBIU) with coordinating behavioural research across the Commission’s units and integrating it in policy-making, strategic foresight, forecasting and assessment of future technologies. In many ways, the rise of behavioural insights and interventions as a novel mode of public policy in the past 10 years is a puzzling case. The basic principles of behavioural economics such as bounded rationality and biased judgement were already formulated in the 1950s.8 More than 40 years ago, Tversky and Kahneman9 identified the central heuristic principles—representativeness, availability and anchoring—that form the core of behavioural economics until today. From the beginning, behavioural economics has been systematically related to policy issues: behavioural economists have extensively suggested corrective procedures, modes of removing biases and instruments of behavioural intervention to improve collective judgement heuristics and risk perceptions as a core challenge for public 6

Lunn, ibid. World Bank, Mind, Society and Behavior (Washington, World Bank, 2015). 8 HA Simon, Models of Man: Social and Rational: Mathematical Essays on Rational Human Behavior in a Social Setting (New York, Wiley, 1957). 9 D Kahneman and A Tversky, ‘Judgement under Uncertainty: Heuristics and Biases’ in D Kahneman, P Slovic and A Tversky (eds), Judgement under Uncertainty: Heuristics and Biases (Cambridge/ New York/Sidney, Cambridge University Press, 1982). 7

Behavioural Expertise and Regulatory Power

145

management.10 When Kahneman received the Nobel Prize in Economics in 2002, behavioural economics and related fields of research were already well established in academia and known to a wider audience.11 In administrative law, calls for integrating behavioural science date back to the 1960s.12 This raises some challenging questions: how can we explain the inter- and transnational appeal of behavioural approaches developed in the past years? What are the factors leading to the ‘recent renaissance of behavioural studies’13 in the European Union? And, more concretely, what makes the idea that designing better regulation ‘can greatly benefit from a better understanding of people’s behaviour’14 so attractive, even publicly irrefutable, for the European Commission? While there is a growing literature on approaches to behavioural change, the sources of their contemporary success and the practices by which they are translated into public policy instruments are still unclear. The case of the European Union’s hesitant, but over the last five years much firmer, embrace of behavioural approaches, has been especially puzzling for observers. Tentative explanations range from the influence of the digital economy, the appeal of cheap alternatives to regulation, the low invasiveness of behavioural interventions and their alleged effectiveness and efficiency to the global rise of a ‘psychological state’.15 So far, there has been no systematic study on the increasing influence of behavioural approaches, the conditions for and the mechanisms of their rise in the context of EU regulation and policymaking. Such a study, however, would be a necessary precondition for a better understanding of the immediate effects and long-term consequences of behavioural change policies in Europe. Drawing on some first findings from recent research16 I am proposing three main assumptions: first, the ‘ecosystem of expertise’ in the EU17 is characterised 10 APSC, Changing Behaviour. A Public Policy Perspective (Canberra, Australian Public Service Commission, 2007); Kahneman, Slovic and Tversky (eds), ibid. 11 D Kahneman, Thinking, Fast and Slow (London, Penguin Books, 2011); D Kahneman and A Tversky, Choices, Values, and Frames (Cambridge, Cambridge University Press, 2000). 12 A Alemanno and A Spina, ‘Nudging Legally: On the Checks and Balances of Behavioural Regulation’ (2014) 12 International Journal of Constitutional Law 429, 434–36. 13 ibid 436. 14 Rv Bavel and others (European Commission), Applying Behavioural Sciences to EU Policy-making (Joint Research Centre, Scientific and Policy Reports, 2013) 3. 15 Alemanno and Sibony, above n 5; Alemanno and Spina, above n 12; R Jones, J Pykett and M Whitehead, Changing Behaviours. On the Rise of the Psychological State (Cheltenham UK/Northampton MA, Edward Elgar, 2013); P Lunn, ‘Behavioural Economics and Policymaking: Learning from the Early Adopters’ (2012) 43 The Economics and Social Review 423; Lunn, above n 5; E Shafir (ed), The Behavioral Foundations of Public Policy (Princeton, Princeton University Press, 2013). 16 H Straßheim, A Jung and R-L Korinek, ‘Reframing Expertise: The Rise of Behavioural Insights and Interventions in Public Policy’ in A Berthoin Antal, M Hutter and D Stark (eds), Moments of Valuation. Exploring Sites of Dissonance (Oxford, Oxford University Press, 2015); H Straßheim and R-L Korinek, ‘Behavioural Governance in Europe’ in R Doubleday and J Wilsdon (eds), Future Directions for Scientific Advice in Europe (London, University of Cambridge/University of Sussex and others, 2015); H Straßheim and R-L Korinek, ‘Cultivating “Nudge”: Knowing Behavioural Governance’ in R Freeman and J-P Voss (eds), Knowing Governance (Palgrave Studies in Science, Knowledge and Policy) (Houndmills and others, Palgrave, 2016). 17 Doubleday and Wilsdon (eds), ibid.

146

Holger Strassheim

by a highly fragmented and complex constellation of actors, contested principles, unclear guidelines and overlapping, potentially competing spheres of sciencepolicy-interactions. After more than five years of what Marten Hajer has termed an ‘institutional void’,18 problems cumulated in 2014 with the abolishment of the Chief Scientific Advisor (CSA). Under such conditions, unquestioned certainties about the validation and justification of knowledge become less taken for granted, opening up an opportunity window for the authoritative re-allocation of meaning. Second, behavioural expertise in the EU has profited from this opportunity window by using easily demonstrable evidence from experimental trials (object dimension), by positioning behavioural experts as intermediaries between science, politics and citizens (social dimension) and by formalising experimental policy procedures through guidelines and emphasising their potential for forecasting (temporal dimension), thus gaining both political and epistemic authority. Third, behavioural approaches promise to reframe the principles of EU regulation during a period of increasing public scepticism, providing legitimatory orientation by emphasising simplicity and a more ‘human’ approach to regulation.19 On the basis of such a justificatory narrative, diverging political positions on individual autonomy are reconciled, while policy-makers are given the capacity to ‘nudge’ citizens to make better choices to their own benefit. By highlighting the multiple ways expertise is (re)configured in political opportunity structures, attributions of authority and narratives of legitimacy, this chapter is following more recent approaches of ‘co-production’.20 In this debate, two competing strands of argument can be distinguished: that of ‘objectivist epistemologies’ and that of ‘political epistemologies’. From the perspective of ‘objectivist epistemologies’, the policy process itself is to be conceptualised in analogy to a scientific process of falsification and progressive problem-solving.21 Since the 1960s, inspired by very similar ideas, the ‘policy analysis movement’ has searched for ways to both minimise policy failures and maximise ‘policy analytical capacity’ by a more systematic application of evaluation methods, scientific expertise and evidence.22 Approaches such as ‘evidence-based’ or ‘evidence-informed’ 18 MA Hajer, Authoritative Governance. Policy-making in the Age of Mediatization (Oxford, Oxford University Press, 2009). 19 CR Sunstein, ‘Empirically Informed Regulation’ (2011) 4 University of Chicago Law Review 1349; CR Sunstein, Simpler. The Future of Government (New York and others, Simon & Schuster, 2013); CR Sunstein, Valuing Life. Humanizing the Regulatory State (Chicago/London, University of Chicago Press, 2014). 20 R Adler-Nissen and K Kropp, ‘A Sociology of Knowledge Approach to European Integration’ (2015) 37 Revue d’Integration Européenne 155; S Jasanoff, ‘A World of Experts. Science and Global Environmental Constitutionalism’ (2013) 40 Boston College Environmental Affairs Law Review 439; H Straßheim, ‘Politics and Policy Expertise: Towards a Political Epistemology’ in F Fischer, D Torgerson, A Durnová and M Orsini (eds), Handbook of Critical Policy Studies (Cheltenham UK/ Northampton MA, Edward Elgar, 2015). 21 G Majone, ‘Research Programmes and Action Programmes, or Can Policy Research Learn from the Philosophy of Science?’ in P Wagner and others (eds), Social Sciences and Modern States. National Experiences and Theoretical Crossroads (Cambridge, Cambridge University Press, 1991). 22 M Howlett, ‘Policy Analytical Capacity and Evidence-based Policy-making: Lessons from Canada’ (2009) 52 Canadian Public Administration 153.

Behavioural Expertise and Regulatory Power

147

olicy-making are the latest expression of this movement, relying on the belief p that the apparent certainties provided by sound science will lead to a more rational problem-solving process or, at least, to increased policy learning.23 In contrast, interpretive and related approaches propose analysing the relationships among science, policy and the public in terms of what we may call ‘political epistemologies’.24 Political epistemologies are understood as research perspectives that focus on how various practices used to validate and justify beliefs about the world, and to collectively order and evaluate the world, are interconnected. While objectivist epistemologies insist upon separating knowledge and values, political epistemologies are explicitly aimed at reconstructing their interaction and mutual constitution. While objectivist epistemologies take boundaries between science and policy for granted, political epistemologies squarely challenge on these presumptive beliefs on societal boundaries, asking for their constitution in social processes of sense-making. Finally, while objectivist epistemologies tend to treat the rationalisation of politics by science as their analytical vantage point, political epistemologies are more sceptical. Proponents assume that, depending on the institutional and cultural context, there are always multiple and potentially contested perceptions of how science and politics should interact and what rationalisation could actually mean. They advocate recognising the social complexity of expertise resulting from often controversial and culturally framed ways of attributing authority and reputation across different levels of social interaction. Understanding the emergence, embeddedness and erosion of expert authority under the conditions of the post-national constellation has become a core question for research on transnational governance and regulation.25 This chapter is structured as follows: the second section provides a brief overview of the use of behavioural expertise in EU policy-making and regulation; in the third section I focus on the ways behavioural expertise generates politico- epistemic authority at the science-policy-public nexus; the fourth section shows how principles of regulation are influenced and reframed in behavioural narratives of justification; and the chapter concludes by summarising the findings and developing some preliminary assumptions about the future relationship of behavioural expertise and regulatory power in the EU.

23 H Straßheim and P Kettunen, ‘When does Evidence-based Policy turn into Policy-based E vidence? Configurations, Contexts and Mechanisms’ (2014) 2 Evidence & Policy 259. 24 Understanding and use of the term differ: F Fischer and H Gottweis (eds), The Argumentative Turn Revisited: Public Policy as Communicative Practice (Durham/London, Duke University Press, 2012); J Friedman, ‘Political Epistemology’ (2014) 26 Critical Review i; B Latour, Politics of Nature. How to Bring the Sciences Into Democracy (Cambridge/London, Harvard University Press, 2004); Straßheim, above n 20. In 2014, the American Political Science Association (APSA) approved an organised section on ‘Political Epistemology’ in order to study how people ‘interpret either their own interests or the public interest; from what sources are these interpretations drawn; and how do these interpretations motivate political action’ (www.apsanet.org/section46). 25 S Quack, ‘Regime Complexity and Expertise in Transnational Governance: Strategizing in the Face of Regulatory Uncertainty’ (2013) 3 Oñati Socio-Legal Series 647; M Zürn, M Binder and M Ecker-Ehrhardt, ‘International Authority and Its Politicization’ (2012) 1 International Theory— A Journal of International Politics, Law and Philosophy 69.

148

Holger Strassheim

II. Behavioural Expertise in EU Policy-making and Regulation It has been argued repeatedly that ‘the EU’s engagement with behavioural sciences is far from systematic and remains largely circumstantial’.26 This observation, however, is largely based on anecdotal evidence. There has never been a comprehensive and systematic study of the use of behavioural insights and expertise in EU policy-making and regulation. Behavioural approaches can be defined ‘as every mode of governing informed, designed or implemented by focusing on psychological as well as cognitive mechanisms of behaviour in both individuals and collectives’.27 This includes behavioural change instruments in non-smoking policies or food safety but also the design of certain policies with respect to the ways people’s behaviour affects their effectiveness. Behavioural approaches are based on two core insights. First, when situations are complex or ambiguous the behaviour of individuals is influenced by heuristics and biases. In contrast to standard models of rational choice, people often use mental shortcuts and simple solutions even if this means acting against their own interests. According to the research by Tversky and Kahnemann,28 people trying to predict the future intuitively rely on similarities and stereotypes (representativeness); the probability of risks is assessed by the ease with which instances or occurrences can be brought to mind (availability); and in many situations, estimates are biased by an initial value or a starting point that works as ‘default’ (anchoring). People may ignore that they do not know enough or fail to learn from new information. When consumer decisions in credit markets are influenced by such mental shortcuts, they may resort to borrowing at extremely high interest rates. More information rarely solves the problem since it adds up to the already existing amount of information while increasing uncertainty. Second, the social environment against which people make decisions may trigger some of these heuristics and inhibit others. Everyday action is embedded into informational infrastructures, simplifying the presentation of options, evoking certain associations or making certain options more visible than others. In their popular book titled Nudge Thaler and Sunstein speak of ‘choice architectures’.29 Following the authors’ choice architectures is ubiquitous: the order of traffic lights, the display of food in a cafeteria and the design of a web page structure, are

26 Alemanno and Sibony, above n 5; Alemanno and Spina, above n 12 at 441; E Ciriolo, ‘Behavioural Economics in the European Commission: past, present, and future’ (Oxera Consulting LLP, January 2011); Lunn, above n 5. 27 Straßheim and Korinek, ‘Behavioural Governance in Europe’, above n 16 at 154. 28 Kahneman and others, above n 10. 29 RH Thaler and CR Sunstein, Nudge. Improving Decisions About Health, Wealth and Happiness (London, Penguin Books, 2008).

Behavioural Expertise and Regulatory Power

149

e xamples of both the salience of options and the likelihood of certain choices. Since choice architectures are inevitable, policy-makers are advised to actively engage in designing arrangements that support desirable policy goals and reduce behaviour seen as suboptimal. This is the core argument of behavioural approaches: avoiding pre-structured contexts is not an option. Not engaging in behaviourally informed policies and regulations still has behavioural consequences. Following this line of thought, both policy-makers and experts should become choice architects. The earliest case of the use of behavioural insights in EU legislation is the directive on the protection of consumers in respect of distance contracts from 1997.30 It allows for a ‘cooling-off period’ of at least seven working days to withdraw from the contract without penalty or giving reasons, thus compensating for typical ‘availability’ biases under the circumstances of distant communication, leading to spontaneous and mostly ill-advised purchases (eg in the aftermath of an earthquake online purchases of insurances rise). The regulation on nutrition and health claims on food from 200631 harmonises the rules for the use of nutritional or dietary claims. It is targeting the ‘anchoring effects’ related to a misleading starting value on a products description, for example a cheese is declared as being 80 per cent fat-free instead of containing 20 per cent fat.32 Another well-known example of designing choice architecture on the European level is the EU ban on pre-ticked boxes for online purchases. Orthodox economics predict that decisions should not be influenced by a pre-selection with respect to certain products or add-on components, such as in travel insurances. In contrast, behavioural economics has shown that defaults significantly increase the probability of specific purchasing behaviour. The EU Directive on Consumer Rights33 prohibits pre-ticketing, a targeting problem connected to anchoring heuristics. Under this directive, using behavioural insights means removing defaults.34 In the context of the Commission’s case against Microsoft for abuse of its market position,35 the solution chosen by the Commission was based on similar behavioural insights. It ensures that consumers have to actively choose a browser instead of being confronted with a default option.36 However, these are only the more well-known cases. A closer view reveals that the influence of behavioural economics should not be underestimated. On the 30 European Parliament and Council Dir 97/7/EC of 20 May 1997 on the protection of consumers in respect of distance contracts [1997] OJ L144/19. 31 European Parliament and Council Reg (EC) 1924/2006 of 20 December 2006 on nutrition and health claims made on foods [2006] OJ L404/9. 32 Ciriolo, above n 26. 33 European Parliament and Council Dir 2011/83/EU of 25 October 2011 on consumer rights, amending Council Dir 93/13/EEC and Dir 1999/44/EC of the European Parliament and of the C ouncil and repealing Council Dir 85/577/EEC and Dir 97/7/EC of the European Parliament and of the Council [2011] OJ L304/64. 34 Alemanno and Spina, above n 12 at 441. 35 Case T-201/04 Microsoft Corp v Commission [2007] ECR 2007 II-03601. 36 Lunn, above n 5 at 33.

150

Holger Strassheim

basis of a very first and selective analysis of documents related to behaviourally informed studies, regulations and communications in the Commission, the following table 1 provides a preliminary account of behavioural insights in EU policy-making and regulation.37 Table 1: Using Behavioural expertise in EU policies & regulations (selection) 1997

EU Parl/Council

Directive: Cooling off period for distance contracts

2006

EU Parl/Council

Health and Nutritional Claims Regulation

2007

COM

EU Consumer Policy Strategy 2007–2013

COM

White Paper on a Strategy for Europe on nutrition, overweight and obesity-related health issue

DG Research

Seventh Framework Program

2008

DG SANCO

Conference: How can behavioural economics improve policies affecting consumers?

2009

EU Parl/Council

Directive: users’ rights relating to electronic communications networks and services

2010

DG JRC DG SANCO

Behavioural Studies for European Policies (BESTEP) and Framework Contract for the Provision of Behavioural Studies (FCPBS)

DG SANCO

Publication on Consumer Behaviour: The Road to Effective Policy Making

DG SANCO

Conference: Behavioural Economics, So What: Should Policy-Makers Care?

DG SANCO / DG GROW

Study on Consumer Decisions in Retail Investment Services

DG CNECT

Study: Consumer 2020—From Digital Agenda to Digital Action

DG JRC-IPTS

Study: Nudging lifestyles for better health outcomes

EU Parl/DG Internal Affairs

Consumer behaviour in a digital environment

EU Parl/Council

Regulation: Provision of Food Information to Consumers

EU Parl/Council

Directive on Consumers Rights

DG JRC-IHCP

Behavioural Economics Team

2011

(continued)

37 The analysis included only documents from the EU Commission and its agencies that explicitly mention behavioural economics, behavioural research, behavioural psychology, ‘nudge’ and related fields or political practices. To get a full picture, a more comprehensive analysis would systematically include also those documents that only indirectly relate to or are inspired by behavioural expertise.

Behavioural Expertise and Regulatory Power

151

Table 1: (Continued) 2012

2013

2014

DG CLIMA

Behavioural Climate Change Mitigation Options

COM

European Consumer Agenda—Boosting confidence and growth

DG SANCO

Bank Fees Behaviour Study

DG SANCO

Conference: Applying Behavioural Insights to Policy-Making—Results, Promises and Limitations

DG JRC-IPTS

Study: Applying Behavioural Insights to Policy-Making

DG CLIMA

Testing CO2 labelling options and consumer behaviour

DG ENERGY

Study on the impact of the energy label—and potential changes to it—on consumer understanding and on purchase decisions

DG JUST

Testing of a Standardized Information Notice for Consumers on the Common European Sales Law

DG JRC-IPTS

Conference: Good behavioural research for EU policymaking

JRC-IHCP

PEER-ACTIVE: a school based study on physical activity

DG JRC

Foresight & Behavioural Insights Unit (FBIU)

DG JRC-FBIU

Tomorrow’s Healthy Society. Research Priorities for Foods and Diets

DG JRC-IPTS/ DG SANCO

Good Behavioural Research for EU Policy Making

DG SANCO

Report on Consumer Policy 2012–2013

DG JRC-IPTS

Conference: Nudging Internet Citizens: Lessons from behavioural studies on online privacy

DG JRC-IPTS

Behavioural Responses to Privacy Visceral Notices (BREVE)

SG-IA

Consultation: Revision of the European Commission Impact Assessment Guidelines

DG CNECT

Online Behavioural Advertising (OBA) Roundtables with Industry

EU Parl/Council

Directive concerning the manufacture, presentation and sale of tobacco and related products

DG ENV

Influences on Consumer Behaviour—Policy Implications beyond Nudging

DG TAXUD

Behavioural Economics and Taxation (continued)

152

Holger Strassheim

Table 1: (Continued)

2015

DG SANCO

Study on the Impact of Food Information on Consumers’ Decision Making

DG SANCO-CHAFEA

Study on the effects on consumer behaviour of online sustainability information displays

DG SANCO-CHAFEA

Impact of information on patients’ choice within the context of the Directive 2011/24/EU of the European Parliament and of the Council on the application of patients’ rights in cross-border healthcare

DG SANCO-CHAFEA

Study on online gambling

DG SANCO/JRC-IHCP Conference: New interventions to reduce overconsumption of alcohol DG SANCO-CHAFEA

Milan BExpo 2015: A behavioural study on food choices and eating habits

DG TAXUD/JRC-IHCP Conference: How to increase tax compliance—new tools from behavioural economics Source: own research

Based on this preliminary overview, it becomes clear that the Commission is engaging in an increasingly broad array of behavioural research, ranging from obesity-related issues and retail investment services to climate change mitigation, bank fees, CO2 labelling, online privacy and gambling addiction. We can distinguish between three overlapping policy waves of using behavioural expertise that are still in progress, namely (1) experimentation, (2) institutionalisation; and (3) diffusion as follows: (1) Experimentation: In 2008, DG SANCO organised the first of its high- profile conferences on the subject, asking tentatively, ‘How can behavioural economics improve policies affecting consumers?’38 Such insights of behavioural studies are taken up selectively and mostly in the field of consumer protection. DG SANCO acts as a policy entrepreneur, referring to behavioural insights in communications on the ‘EU Consumer Policy Strategy 2007–2013’ and the ‘White Paper on a Strategy for Europe on nutrition, overweight and obesity-related health issue’.39 The seventh framework programme already provides more systematic incentives for studies in the disciplines of behavioural economics and behavioural psychology. In 2010, the Commission established the Framework Contract for the Provision of Behavioural Studies (FCPBS) with the purpose of facilitating research and 38 DG SANCO (European Commission, Health and Consumers), ‘Consumer Behaviour: The Road to Effective Policy Making’ (2010) http://ec.europa.eu/consumers/consumer_behaviour_en.htm. 39 Commission, ‘EU Consumer Policy strategy 2007–2013. Empowering consumers, enhancing their welfare, effectively protecting them’ COM (2007) 99 final.

Behavioural Expertise and Regulatory Power

153

experimental trials in support of EU regulation and policy-making. In a formal collaboration with DG SANCO, the Institute for Prospective Technology Studies (JRC-IPTS) started the Behavioural Studies for European Policies (BESTEP) programme, providing technical advice in the conception, design and execution of experimental trials and other behavioural studies. DG GROW and DG CONNECT also started to collaborate with DG SANCO, resulting in two influential pilot studies on ‘Consumer Decisions in Retail Investment Services’ and on a digital agenda for consumer policy. At its second conference, DG SANCO asked more provocatively: ‘Behavioural Economics, So What: Should Policy-Makers Care?’ (2) Institutionalisation: The structural and organisational manifestation of behavioural economics starts with the Behavioural Economics Team at the Institute for Health and Consumer Protection at DG JRC. Set up under the leadership of Benedikt Herrmann, an economist also trained in natural sciences, the Behavioural Economics Team began to systematically carry out and coordinate studies not only in the field of consumer protection but also in health, nutrition and food safety. In 2013, members of the JRC-IPTS published a report on ‘Applying Behavioural Insights to Policy-Making’, based on the third conference organised by DG SANCO. This report differed from other publications in that it provided general guidelines and methodological advice on how to incorporate behavioural insights into the design, implementation and evaluation of EU policies.40 In 2014, DG-JRC established the Foresight and Behavioural Insights Unit (FBIU) aimed at coordinating behavioural research including its application to policies across various Commission units. In its revision of the Impact Assessment (IA) Guidelines, the Secretariat-General of the Commission referred to behavioural biases and behavioural change as components and key principles of IA.41 (3) Diffusion: At that point in time in 2014, the sheer number of behaviourally informed studies and conferences had already exploded, covering a wide range of policy areas such as obesity, climate change, bank fees, CO2 labelling, sales law, physical activity, online privacy, online behavioural advertising (OBA) and tax compliance. Institutionally, behaviourally informed policies were somewhat limited to certain DGs including DG SANTE and DG JRC as forerunners, DG CLIMATE, DG TAXUD, DG ENV, DG CNCT, DG GROW and DG JUST. Moreover, it seems that horizontally more DGs were cooperating in carrying out studies and applying behavioural insights while vertically agencies such as the Consumer, Health, Agriculture and Food Executive Agency (CHAFEA) were becoming more active.

40 B Piniewski and others (European Commission), Nudging Lifestyles for Better Health Outcomes: Crowdsourced Data and Persuasive Technologies for Behavioural Change (DG JRC, 2011). 41 Commission, ‘Regulatory Fitness and Performance Programme (REFIT): State of Play and Outlook’, COM (2014) 368 final.

154

Holger Strassheim

While the overall spread of behavioural expertise in EU policy-making and regulation might be far from revolutionary, during the past five years we can observe a steadily increasing number of adoptions and applications of behavioural insights across policy areas and organisational units. It is a puzzling case, shedding light on the ways novel modes of expertise gain both authority and legitimacy on the European level. As the next section shows, the remarkable career of behavioural expertise in the European advisory system can only be understood by taking into account its institutional and organisational fragmentation.

III. The European ‘Ecosystem of Expertise’: Institutional Voids Debates on the organisation and structure of science-policy advice rarely receive public attention. For this reason it is even more telling what happened in 2014 in the dispute concerning the European Commission’s Chief Scientific Adviser (CSA) and the molecular biologist Anne Glover.42 Previously in 2009, six months before the post of the CSA was created, European Commission President Barroso had already emphasised that ‘we also need a fundamental review of the way European institutions access and use scientific advice’.43 After an unexplained gap of two years, Barroso appointed Glover in 2012 for a three-year term. Her mandate included, inter alia, providing independent expert advice and authoritative guidance on interpretations of scientific evidence, building relationships with high-level advisory groups, the scientific committees of the Commission, the EU agencies and the European Group on Ethics in Science and New Technologies, and communicating the scientific values on which specific Commission proposals are based ‘in order to enhance public confidence in science and technology’.44 From the beginning, the CSA’s office was underfunded and lacked informational channels to the DG JRC and other science units.45 Based within the Bureau of European Policy Advisers (BEPA) and its successor body, the European Political Strategy Centre (EPSC), its competences overlapped with the mandate of other units in the Commission and were especially challenging to the DG JRC and the

42 For the following see A Glover, ‘A Moment of Magical Realism in the European Commission’ in Doubleday and Wilsdon (eds), above n 16; J Wilsdon, R Doubleday, J Hynard, ‘Future Directions for Scientific Advice in Europe’ in Doubleday and Wilsdon (eds),ibid. 43 Glover, ibid 58. 44 ibid 59f. 45 For different interpretations of the role of the CSA see ibid; D Parr, ‘Why it Made Sense to Scrap the Post of Chief Scientific Adviser’ in Doubleday and Wilsdon (eds), above n 16.

Behavioural Expertise and Regulatory Power

155

Commissioner for Research, Innovation and Science. In 2013, complaints by scientists and non-governmental organisations (NGOs) reached the Commission, which expressed suspicion that the CSA could override the political process by offering a short-track for lobbyists to bypass Commission procedures and undermine the work by expert committees. Soon afterwards, a public dispute started that would later be described as a ‘war of letters’.46 Arguing that the post of the CSA was ‘fundamentally problematic’, a coalition of NGOs, including Greenpeace, wrote to Juncker, in-coming president of the Commission, only to be countered by a letter in support of the CSA role, signed by 40 scientific organisations and 773 individuals, stressing the ‘integrity and independence of scientific advice received at the highest level of the European Commission’. Further letters of support and criticism were exchanged. After Barroso left office, Juncker decided to not renew the post of the CSA, asking his Commissioner for research, innovation and science in January 2015 to ‘reflect and present options to me before the summer on how to better institutionalise future scientific advice to the Commission, based on the experience made in all Member states’.47 In summer 2015, the Commission had finally decided to establish a collective body for scientific advice, the Scientific Advice High Level Group (SAM HLG), appointed on the basis of a search process carried out by an Identification Committee.48 This matter sheds some light on the institutional and organisational complexi ties of expert advice in the EU. In the EU ecosystem of expertise, both the provision of expertise and its translation in science-policy interactions ranges across a broad spectrum of public bodies, including not only DG JRC with its institutes and the other DG’s executive and independent agencies, but also the Science & Technology Advisory Council (STAC), the European Political Strategy Centre (EPSC), the European Research Area Board (ERAB), the EU Agencys’ Network of Science Advisers (EU-ANSA), the Research, Innovation and Science Policy Experts (RISE) High Level Group (HLG), the European Group on Ethics in Science and New Technologies (EGE) and a large number of standing scientific committees and ad hoc expert groups. As the former CSA observed, ‘there are no harmonised guidelines for gathering, using and communicating scientific evidence which leads to very diverse approaches by different Commission services, and opens the possibility of evidence being procured and used selectively’.49 Indeed, the first and only guidelines on the ‘collection and use of expertise by the Commission’ were issued in 2002 and since then have never been changed.50

46

Wilsdon and others, above n 42 at 9. Cited after ibid. 48 J Wilsdon, ‘Watchdogs of the European System’ (2015) 6238 Science 947. 49 Glover, above n 42 at 75. 50 Commission, ‘On the Collection and Use of Expertise by the Commission: Principles and Guidelines’ COM (2002) 713 final. 47

156

Holger Strassheim

Marten Hajer has termed such constellations of high fragmentation, complex interaction, unclear hierarchies and ad hoc agreements ‘institutional void[s]’, in which: [T]here are no clear and generally accepted rules and norms according to which politics is to be conducted and policy measures are to be agreed upon. … In the institutional void, politics is about the authoritative allocation of meaning in its broadest possible sense.51

With President Barroso’s call for a ‘fundamental review’ of scientific advice in 2009 at the latest, it becomes clear that the ecosystem of expertise in the EU can also be characterised by such an institutional void. The efforts by DG JRC and DG SANCO from 2009 onwards to support behavioural studies and to establish a technical and methodological infrastructure have to be understood in the context of an increasing pressure to re-establish new modes of authoritative expertise. A brochure on ‘Consumer Behaviour: The Road to Effective Policy Making’ published by DG SANCO in 2010 described as major goals for future action the improvement of academic-policy-links, the better integration of scientific knowledge into policy-making and a better collaboration with other Commission services, eg DG Research.52 Programmes like BESTEP or FCPBS and the establishment of the Behavioural Economics Team in 2011 were direct expressions of these efforts. In one of its recent reports on ‘Tomorrow’s healthy society—research priorities for food and diets’ the Foresight and Behavioural Insights Unit (FBIU) promoted as its first research priority an ‘integrated policy-making’ to ‘improve the evidencebase’, ‘develop a scientific framework for a system approach’ and ‘provide a framework to design, monitor and evaluate policies’.53 As the director of policy support coordination at JRC emphasised, it is one of the essential tasks of FBIU: [T]o support systemic thinking and the consideration of complexity for reflections on the future …, to proactively shape policies, taking into account the dynamics of change rather than linear extrapolation and wishful thinking … to further improve its scientific and technical policy support [by] pooling relevant expertise.54

This is not to say that by using behavioural insights and interventions the Commission is actually succeeding in improving science-policy interaction and integrating scientific knowledge into regulation. Behavioural expertise might even add just another layer to the complexity of policy-advice in the EU. The increasingly salient lack of coordination in the advisory system notably creates a critical moment in which political actors and experts alike are highly receptive to promising ideas whose time seems to have come.55 By framing behavioural approaches as

51

Hajer, above n 18 at 34. DG SANCO, above n 38. 53 A-K Bock and others (European Commission), Tomorrow’s Healthy Society—Research Priorities for Foods and Diets (final report of the Joint Research Centre Foresight Study, 2014). 54 ibid. 55 JW Kingdon, Agendas, Alternatives, and Public Policies (Boston, Little Brown and Company, 1984). 52

Behavioural Expertise and Regulatory Power

157

solutions to the lack of science-policy coordination, behavioural experts combine the capability for an authoritative re-allocation of meaning with the right timing. The next section shows how the behavioural movement manages to be perceived as both scientifically reputable and politically attractive, capable of filling the institutional void in European policy making (expert-executive policy-making?).

IV. Combining Epistemic and Political Authority Experts are essentially already boundary workers. Drawing on insights from pragmatism, social phenomenology and institutionalist approaches, expertise is conceptualised as a nexus of political and epistemic authority attributions:56 to recognise the expertise of an individual or collective actor means to attribute to that actor both the competence to validate and justify knowledge claims (‘epistemic authority’) and the capability to make these knowledge claims relevant for collectively ordering and evaluating society (‘political authority’). To understand how behavioural expertise has emerged and gained both epistemic authority and political relevance across different units of the Commission, I suggest a three-dimensional theory, focusing on the multiple ways facts and forms of evidence are designed to function as ‘boundary objects’ at the sciencepolicy interface in an object dimension, on the attribution of competence to individuals or organisations in a social dimension, and on the temporal sorting of policy advice and decision-making in the dimension of time.57 The persuasive power of behavioural expertise is arguably based on: (1) the demonstration of relevance for politically salient issues using easily understandable experimental evidence, that appeals to common sense reason, while at the same time being linked to scientific norms and standards (object dimension); (2) the positioning of behavioural experts as intermediaries between science, politics and citizens on the one hand and the empowerment of European policy-makers to act as ‘choice

56 A Jung, R-L Korinek, H Straßheim, ‘Embedded Expertise: a Conceptual Framework for Reconstructing Knowledge Orders, their Transformation and Local Specificities’ (2014) Innovation: The European Journal of Social Sciences 398–419. 57 Straßheim, above n 20; Straßheim and others, above n 16. These three dimensions of sensemaking have been used in remarkably similar ways by different strands of social theory, see P Berger, T Luckmann, The Social Construction of Reality: A Treatise in the Sociology of Knowledge (New York, Anchor Books, 1967); N Luhmann, Social Systems (Stanford CA, Stanford University Press, 1995); WR Scott, Institutions and Organizations (London, Thousand Oaks and New Delhi, Sage, 1995). See also Jasanoff ’s typology of expertise, identifying ‘three bodies of expertise’ that are connected to the problems of expert legitimacy: S Jasanoff, ‘Judgement Under Siege: The Three-Body Problem of Expert Legitimacy’ in S Maasen, P Weingart (eds), Democratization of Expertise? Exploring Novel Forms of Scientific Advice in Political Decision-Making (Berlin, Heidelberg and New York, Springer Verlag, 2005). KC Elliott, Is a Little Pollution Good for You? Incorporating Societal values in Environmental Research (Oxford, Oxford University Press, 2011) has presented a fruitful application of Jasanoff ’s three body problem on environmental policy and research.

158

Holger Strassheim

architects’ on the other (social dimension); and (3) the temporal integration of experimental studies into the policy-making process and the systematic implementation of behaviourally informed foresight activities (temporal dimension). These three dimensions can be described as follows: (1) In the object dimension the epistemic authority of behavioural economics can be attributed to the demonstration of immediate practical relevance to politically salient issues such as efficiency and applicability. For instance, text message reminders to save money increased savings balances by six per cent; emails to homeowners comparing their electricity bill with that of neighbours and rating them (‘great’, ‘good’, ‘below average’) led to reductions in power consumption equivalent to what would have happened if energy prices had been raised 11–20 per cent; automatically enrolling people in pension plans dramatically increased participation and retention.58 The promise to deliver effective and efficient solutions to contemporary political problems is, moreover, based on evidence forms that appeal to common sense reasoning and can be understood without specialised knowledge. Cognitive rules such as ‘loss aversion’ or ‘overconfidence’ seem to be taken directly from everyday experience.59 While appealing to common sense, the epistemic authority of behavioural experts is, however, also safely nested in the academic recognition of peers. From the perspective of behavioural economists, orthodox economics has been a deductive endeavour, while behavioural economics rest on an inductive and experimental approach to produce easily demonstrable evidence and facts that are understood by every lay person. The heuristics and biases clouding the assessment of probabilities in everyday behaviour are identified and demonstrated in numerous psychological experiments. A constitutive aspect of the politico-epistemic authority that is attributed to behavioural expertise is the dual approach of referring to the scientific ‘gold standard’ of experiments while at the same time presenting evidence that is easily graspable.60 In short, the ‘persuasive power’61 of behaviour insights on the object dimension can be attributed the way scientific norms and standards are linked to politically acceptable ‘rules of thumb’.62 (2) Essential to understanding the persuasive power of behavioural economics in the social dimension is that both judgemental and decision-making competence is attributed to behavioural experts and economists. In Thaler and Sunstein’s perspective, policy-makers in government agencies need to be aware of the ‘misconception … that it is possible to avoid influencing people’s choices’.63 Since they are changing the ‘choice architecture’ with 58 N Rodriguez-Priego and Rv Bavel (European Commission), Good Behavioural Research for EU Policy-Making (Report of the Joint Research Centre Workshop, 2014); World Bank, above n 7 at 197. 59 Kahneman, above n 11. 60 DG JRC, above n 14; DG JRC, above n 40. 61 Lunn, above n 15 at 432. 62 Thaler and Sunstein, above n 29 at 125. 63 ibid 10.

Behavioural Expertise and Regulatory Power

159

every decision, policy-makers are supposed to rely on behavioural experts— both for the development of effective instruments of behavioural intervention and the use of randomised controlled trials to test and improve these instruments: ‘Nudgers will be able to make good guesses when they have much more expertise at their disposal’.64 Policy-makers and behavioural experts cooperate to become what Thaler and Sunstein call ‘choice a rchitects’: ‘Choice architects can preserve freedom of choice while also nudging people in directions that will improve their lives’.65 The collaboration with behavioural economists is portrayed as enabling policy-makers to become behavioural experts themselves, combining both the political legitimacy to decide upon new instruments and the interpretive competence to judge whether these instruments are effective. This idea has resonated well with parts of the Commission (see section V). (3) In the temporal dimension, behavioural experts have developed schemes for structuring the practices and sequences of decision-making. These include both guidance on how to ensure that ‘empirically informed regulation’ gains real-time access to the rulemaking docket and checklists on how to use techniques to ‘quantify anticipated present and future benefits as accurately as possible’.66 By defining the procedures and the temporal sorting of ‘behavioural policies’,67 the incorporation of behavioural expertise becomes routine and a necessity in public policy-making. Furthermore, the whole process of policy-making is modelled on experimental designs. Agencies are advised to analyse in advance the effects of alternative policy designs and to adapt policy instruments in accordance with the results of experimental or quasi-experimental studies. In several reports prepared for DG JRC, the authors argue that the incorporation of behavioural sciences into the policymaking process changes the very stages and time horizons of agenda setting, decision-making, implementing and monitoring. ‘Good policy-making’, it is argued, is based on the right timing of identifying and studying behavioural elements to develop preventive measures and prospective policy-making.68 The institutionalisation of the Foresight and Behavioural Insights Unit (FBIU) at the DG JRC in 2014 highlights the importance of the temporal dimension in developing behavioural expertise. It is thus the amalgamation of experimental evidence, professional expertise and foresight competence at the science-policy-nexus that has helped behavioural economics to gain authority for the re-allocation of meaning across a rising number of organisations and policy areas in the Commission. As will be argued in the following section, however, it is not only the politico-epistemic authorisation of

64

ibid 247. ibid 253. 66 Sunstein, ‘Empirically Informed Regulation’, above n 19 at 1388. 67 Shafir, above n 15. 68 DG JRC, above n 14. 65

160

Holger Strassheim

behavioural economics, but also its capability to re-define regulatory power and to promote new, legitimate ways of ordering the relationship between science, policy and civil society.

V. Reframing Regulation Cass Sunstein, professor of law at Harvard and former ‘regulatory czar’ heading the White House Office of Information and Regulatory Affairs (OIRA), has repeatedly argued that the future of government depends on simple regulation.69 Of course, the quest for simplified messages and reduced complexity as guiding principles of regulatory action resonates well with behavioural insights: behavioural studies have shown that simplifying messages and reducing complexity may have a significant impact on people’s behaviour. Making registration forms or information letters more clear and options more salient can move citizen’s choices into desired directions, eg, changing energy providers. More information does not solve the problem of biased and non-rational behaviour since it adds up to the already existing information, increases uncertainty and reinforces biased decision-making. Sunstein, however, also argues from a normative perspective. In an extensive discussion of John Stuart Mill’s ‘On Liberty’, he challenges Mill’s harm principle.70 While Mill defends the sovereignty of each individual over his and her body and mind, Sunstein rejects this notion ‘on the ground that in certain contexts, people are prone to error, and paternalistic interventions would make their lives go better’.71 In their philosophy of ‘libertarian paternalism’, Sunstein and Thaler justify the need ‘for self-conscious efforts, by private and public institutions, to steer people’s choices in directions that will improve the chooser’s own welfare [and to] impose … trivial costs on those who seek to depart from the planner’s preferred option.’72 The apparent oxymoron of ‘liberal paternalism’ provides policy-makers with the capacities of ‘choice architects’ to create situations where citizens make better choices to enhance their own welfare, thereby reconciling competing ideological positions between liberalism and paternalist intervention. Moreover, Sunstein is also a strong proponent of cost-benefit-analysis as a b ehaviourally informed ‘form of political algebra’ with the purpose of ‘humanizing the r egulatory state’.73 In its initiatives on Smart Regulation and in its Regulatory Fitness and Performance programme the Commission draws heavily on the idea of simple

69

Sunstein, ‘Simpler. The Future of Government’, above n 19. Sunstein, Why Nudge? The Politics of Libertarian Paternalism (New Haven/London, Yale University Press, 2014). 71 ibid 4. 72 CR Sunstein and RH Thaler, ‘Libertarian Paternalism is not an Oxymoron’ (2003) 70 The University of Chicago Law Review 1162. 73 Sunstein‚ Valuing Life. Humanizing the Regulatory State, above n 19. 70 CR

Behavioural Expertise and Regulatory Power

161

government.74 Sunstein has been one of the key experts advising the Commission in developing this position. Since 2008, there has been a dialogue between the European Commission services and the US Office of Management and Budget (OIRA) on the improvement of Impact Assessment guidelines.75 The most recent draft for a revision of European Commission Impact Assessment Guidelines is invoking behavioural insights.76 As the authors of an early study on ‘Designing policy to influence consumers’ carried out on behalf of the DG Environment put it: Critiques of policy-making based on insight from behaviour economics sometimes accuses such policies of being overly paternalistic, leading to accusations of the ‘nannystate’. Policy-makers should not be put off by such accusations. Policy instruments that are uninformed by research from behavioural science are not necessarily less paternalistic, they are simply less likely to be effective.77

Following this line of thought, regulation appears as a ‘persuasive technology’ based on the normative stance that ‘citizens can be unobtrusively and intelligently helped to make optimal choices supporting both their individual well-being as well as group well-being policy goals’.78 Some argue that some current EU legislation, such as on labelling and disclosure requirements for tobacco or alcohol, demonstrates the awareness that information is a tool to change consumer’s behaviour in a politically preferred direction, thus mirroring the debate on ‘libertarian paternalism’ on an instrumental level.79 Others are sceptical, pointing to differences in terms of cultures of regulation and expertise between the EU and the US. Whereas in EU regulation paternalism has never been as problematic and loaded an issue as in the US, the concept of libertarianism could raise criticism in the context of the EU.80 Indeed, it is not very likely that the debate on libertarian paternalism will feature strongly in EU debates. Behavioural approaches, however, resonate very well with known tendencies in the EU administration to technocratic practices and regulation by stealth. With its emphasis on evidence-based policy

74 Commission, ‘Smart Regulation in the European Union’ COM (2010) 543 final; Commission, ‘Regulatory Fitness and Performance Program (REFIT): State of Play and Outlook’, COM (2014) 368 final. 75 Commission, ‘Review of the Application of EU and US Regulatory Impact Assessment Guidelines on the Analysis of Impacts on International Trade and Investment’ (Final Report and Conclusions of EC and OIRA, 2008). 76 Commission, ‘2014 Revision of the European Commission Impact Assessment Guidelines’ (10 May 2015) ec.europa.eu/smart-regulation/impact/consultation_2014/index_en.htm. 77 Policy Studies Institute, Designing Policy to Influence Consumers: Consumer Behaviour Relating to the Purchasing of Environmentally Preferable Goods 2nd edn (A project under the Framework contract for economic analysis, 2009) 80, ENV.G.1/FRA/2006/0073. 78 JRC, above n 40 at 37. 79 G Howells and J Watson, ‘The Role of Information in Pushing and Shoving Consumers of Tobacco and Alcohol to make Healthy Lifestyle Choices’ in A Alemanno and A Garde (eds), Regulating Lifestyle Risks: The EU, Alcohol, Tobacco and Unhealthy Diets (Cambridge, Cambridge University Press, 2015). 80 A Alemanno and A-L Sibony, ‘Epilogue: The Legitimacy and Practicability of EU Behavioural Policy-Making’ in Alemanno and Sibony (eds), above n 5, 325.

162

Holger Strassheim

and its reliance on expertise, behavioural approaches might be the foundation for a re-strengthened rationale of technocratic intelligence in European regulation: Regulatory science needs to be understood as being shaped by normative commitments, institutional cultures, and existing mechanisms of translating scientific understanding into policy action. … Yet, the policy discourse (at EU level and in individual member states) tends to adopt a rather less critical attitude towards behavioural approaches to regulation. Our argument is that the turn to more behavioural science for greater effectiveness and legitimation brings its own set of ambiguities, which need to be properly addressed, not glossed over.81

Clearly, a closer analysis of the interactions and influences between behavioural ideas on legitimacy and regulatory frameworks in the EU is needed. Concluding from these first observations, however, it seems that behavioural approaches are at least partly successful in reframing some of the legitimatory principles of EU regulation, while re-enforcing technocratic dynamics based on a new alliance between economics, behavioural sciences, psychology and neurosciences. It remains to be seen how these dynamics affect basic political positions such as the precautionary principle or the leitbild of the sovereign consumer.82

VI. Conclusions Summing up, I have suggested that the rise of behavioural expertise in the European Union is the result of a three-part movement. First, behavioural experts have cultivated politico-epistemic authority by claiming the role of ‘choice architects’, mobilising easily demonstrable forms of evidence and modelling the policy process after experimental designs. Second, the institutional void in the ecosystem of expertise in Europe has created a critical moment, allowing behavioural experts to combine the capability for an authoritative re-allocation of meaning with the right timing. Third, it seems that behavioural approaches are at least partly successful in reframing some of the legitimatory principles of EU regulation, while re-enforcing dynamics of technocracy and regulation by stealth based on behavioural sciences. In a highly fragmented advisory system, behavioural expertise gains its persuasive power not by the diffusion of ideas or the provision of new insights about human rationality. Rather, it is the result of a complex interlinkage between expertise, political authority and regulatory ideas on the future statecitizen relationships. There is a further aspect to the behavioural movement that might prove problematic in the context of the European Union and that has yet been neglected 81 M Quigley and E Stokes, ‘Nudging and Evidence-Based Policy in Europe: Problems of Normative Legitimacy and Effectiveness’ in Alemanno and Sibony (eds), ibid. 82 On the precautionary principle as ‘collective loss aversion’ see Sunstein, Valuing Life. Humanizing the Regulatory State, above n 19 at 151.

Behavioural Expertise and Regulatory Power

163

in most of the literature on the topic.83 It is related to the well-known tension between unity and diversity that is so characteristic of the European Union. The translation of behavioural insights into policy-making differs greatly between regional and national contexts. In Denmark, behavioural interventions are associated with networks between civil society actors, policy consultants and academics aiming at improving sustainability and energy efficiency by ‘greenudges’. In contrast, the behavioural units in the UK or Germany consist of more or less secluded expert teams working on nudges in diverse policy areas such as taxation, pensions or crediting. In each country, science-policy-interactions are structured by different cultures of expertise, ie more or less unquestioned ways public knowledge claims are validated and perceived as politically relevant.84 These cultural preferences may affect the resilience and receptiveness of knowledge production, influencing the basic understanding of what nudges actually could be, how they could be embedded into the existing regulatory system and what could be the legitimate goals of using them. When implementing behavioural governance in Europe, both experts and policy-makers need to critically re-examine unquestioned certainties and normative implications of behaviourally informed policies. Otherwise they might find themselves confronted with challenges posed not only by political, but more importantly by ‘epistemic subsidiarity’.85 If the European Union is aiming at transnationalising behavioural change it will have to find new modes to both support and coordinate the coexistence of different ways of public knowledge production. First and foremost, this means that the ‘one-size-fits-all’ model of policy advice86 is no longer applicable in the ecosystem of expertise in Europe.

83

For an exception see Quigley and Stokes, above n 81. Straßheim, above n 20. 85 S Jasanoff, ‘Epistemic Subsidiarity—Coexistence, Cosmopolitanism, Constitutionalism’ (2013) 4 European Journal of Risk Regulation 133. 86 S Beck, ‘Between Tribalism and Trust: The IPCC Under the “Public Microscope”’ (2012) 7 Nature and Culture 151. 84

164

Part 3

Rethinking Constitutionalism: Legitimacy and Accountability

166

9 Expertise as Justification: The Contested Legitimation of the EU ‘Risk Administration’ MARIA WEIMER AND GAIA PISANI1

I. Introduction Risk regulation in the European Union (EU) over the last two decades can be seen as a process of relentless soul-searching with regard to the normatively desirable and functionally adequate model of taking collective decisions in the face of technological progress, scientific uncertainty and societal contestation. Since the wake of the BSE crisis at the turn of the century, the EU has undergone major legal and constitutional developments in this field,2 including the promotion of the precautionary principle,3 to a general principle of EU law applicable to all policy fields touching upon environmental, health, animal or plant health protection.4 These developments seem to reflect the recognition that while scientific expertise plays a crucial role in democratic decision-making on risk, its potential to legitimate such decision-making is limited.5 In the well-known Pfizer ruling, the EU General Court expressed this idea by pronouncing that the exercise of public authority by the EU executive is rendered legitimate by way of its political responsibility as well as accountability towards democratically established institutions, such as the European Parliament. Scientific experts on the other hand, ‘although they have

1 The authors wish to thank the participants of the Workshop, ‘Regulating Risks in the EU—The Co-Production of Expert and Executive Power’ organised by the University of Amsterdam from 21–22 May 2015, for their feedback and valuable comments. 2 E Vos, ‘EU Food Safety Regulation in the Aftermath of the BSE Crisis’ 23 Journal of Consumer Policy 227. 3 Now TFEU, art 191(2). 4 Case T-74/00 Artegodan v Commission ECLI:EU:T:2002:283; T-13/99 Pfizer v Commission ECLI:EU:T:2002:209; see also European Commission, Communication from the Commission on the Precautionary Principle (COM 2000/1 final). 5 See M Weimer, ‘Risk Regulation and Deliberation in EU Administrative Governance—GMO Regulation and Its Reform’ (2015) 21 European Law Journal 622.

168

Maria Weimer and Gaia Pisani

scientific legitimacy, have neither democratic legitimacy nor political responsibilities. Scientific legitimacy is not a sufficient basis for the exercise of public authority.’6 In this way the General Court was careful to preserve a space for the exercise of political discretion bound by law in the face of scientific uncertainty, finding that the Commission was not bound by the scientific opinion of its expert committee when banning the use of a certain antibiotic in animal feed.7 So far so good. The considerable policy, legal and scholarly efforts seem to indicate a change of approach in EU risk regulation from technocratic to inclusive and precautionary governance, and from a naïve positivist understanding of science as infallible authority to a more nuanced view of its contextualised, value-laden and uncertain nature as shown by years of social studies of science, risk and technology.8 The politics of EU risk regulation are now well understood. Or are they? Those following the topic know, of course, that the EU’s search for the ‘right’ model of risk regulation has been accompanied by an inner struggle, and many contradictions both in legal norms and regulatory practices.9 Perhaps the biggest paradox is that despite the widespread acknowledgement of the limitations of science as the basis for public decision-making, EU decision-makers continue to justify regulatory decisions on risk and technology almost exclusively in scientific terms.10 This chapter helps to unpack this paradox by placing the discussion about the role of science in EU risk regulation within the institutional context of the EU public administration and EU administrative law. It is within this context that a vast number of risk decisions in the EU are being made, which in turn shapes the way in which scientific expertise is being used in these decisions. By doing so, this chapter supports the insight that controversies surrounding risk regulation are at their core controversies about the legitimacy of an unelected public administration within modern liberal democracies,11 including ideas on how law could legitimate the exercise of wide discretionary powers. This concern

6

Case T-13/99, above n 4, para 201. On the continuing relevance of Pfizer see C Anderson, ‘Contrasting Models of EU Administration in Judicial Review of Risk Regulation’ (2014) 51 Common Market Law Review 425. 8 See S Jasanoff, The Fifth Branch: Science Advisers as Policymakers (Cambridge MA, Harvard University Press, 1990); S Funtowicz and J Ravetz, ‘Science for the Post Normal Age’ in L Westra and J Lemons (eds), Perspectives on Ecological Integrity (Dordrecht, Springer Netherlands, 1995); B Wynne, ‘Uncertainty and Environmental Learning: Reconceiving Science and Policy in the Preventive Paradigm’ (1992) 2 Global Environmental Change 111. See also European Commission, ‘Taking European Knowledge Society Seriously—Report of the Expert Group on Science and Governance’ (2007). 9 See MBA van Van Asselt and E Vos, ‘Wrestling with Uncertain Risks: EU Regulation of GMOs and the Uncertainty Paradox’ (2008) 11 Journal of Risk Research 281; M Lee, ‘Beyond Safety? The Broadening Scope of Risk Regulation’ (2009) 62 Current Legal Problems 242; M Kritikos, ‘Traditional Risk Analysis and Releases of GMOs into the European Union: Space for Non-Scientific Factors?’ [2009] European Law Review 405. 10 It is undisputed here that scientific expertise is a crucial element of risk regulation. However, the uncertain nature of many late-modern technological risks necessarily changes the role of science in regulation; see ibid and above n 8. 11 E Fisher, Risk Regulation and Administrative Constitutionalism (Oxford, Hart Publishing, 2007); Anderson, above n 7. 7

Expertise as Justification

169

is particularly pertinent to the EU public administration, the institutional complexity of which sets it apart from political-administrative orders both at national and international levels.12 Moreover, the expansion of the EU administrative state intensifies the concern for administrative legitimation13 in an institutional context strongly reliant on expertise yet relatively remote from national democratic processes. This chapter, therefore, explores the legitimacy challenge posed by the emergence of the EU ‘risk administration’—a public administration whose main task is to assess, manage and communicate risks on the basis of available, and often limited, scientific evidence. EU law plays a vital role in addressing that challenge. On the one hand, it is a mechanism of constituting, limiting and holding the EU public administration to account, thereby ensuring its formal legitimation. On the other hand, EU law itself ‘provides both arenas and discourses for disputing the role and nature of public administration’.14 This chapter explores the co-production15 of law, administrative legitimation and scientific expertise in a particular field of the EU risk administration, namely the EU market authorisations of genetically modified organisms (GMOs). Based on an in-depth empirical study of two particularly controversial authorisation processes, it shows that the practice of administrative implementation of EU rules on GMOs as well as of judicial review thereof is characterised by competing views on how the EU public administration ought to be legitimated in the face of scientific uncertainty and contestation, which in turn influences ideas about the role of scientific expertise in this process.

II. Shifting Sands—Administrative Legitimation Between Control and Deliberation EU regulation of GMOs illustrates well the struggle over the appropriate balance between the impact of science and politics on EU administrative decision-making, as well as the role of EU law in that struggle. For over a decade the administrative implementation of EU rules on the marketing of GMOs has been the arena in which the political battle over this controversial technology has taken place. The resulting problems of regulatory deadlock, failure of deliberation and controversial legal reform have been substantively discussed.16 In this chapter we aim 12 See special issue A Gornitzka and C Holst (eds), ‘The Role of Expert Knowledge in EU Executive Institutions’ (2016) 1 Politics and Governance. 13 We use the term legitimation instead of legitimacy to distance ourselves from a sociological understanding of legitimacy as actual social acceptance of the exercise of public authority. Legitimation instead refers to mechanisms of legitimation, which justify the exercise of public authority leading to the assumption that it is acceptable. 14 See Fisher, above n 11, 23. 15 See Introduction to this volume. 16 M Lee, EU Regulation of GMOs (Cheltenham, Edward Elgar, 2008); Weimer, above n 5.

170

Maria Weimer and Gaia Pisani

to show that the struggle over GMO authorisations has also been a struggle over ideas of what constitutes legitimate EU risk administration. Models of public administration are reflections of constitutional theories.17 Our understandings of administrative legitimation are thus infused with constitutional theories regarding the legitimacy of public administration within constitutional democracies. These understandings are therefore shaped within, to use Elizabeth Fisher’s term, discourses of administrative constitutionalism, which manifest themselves, among others, in legal and administrative decision-making.18 Naturally, concepts of administrative legitimation will vary from one legal system to another.19 Yet also within one legal system, the contours of such concepts are essentially contested, in flux and subject to an ongoing constitutional re-settlement.20 In other words, normative ideals about public administration and how it should be constituted, limited and held to account so as to ensure its legitimacy are constantly competing with one another within administrative legal practice. In turn, those ideals and discourses shape the process of technological risk evaluation in different ways. Contestations with regard to administrative legitimation are also (and perhaps even more so)21 present in the EU legal system, and are particularly visible in the field of EU risk regulation. For the purposes of this chapter, we identify two ideal models of EU administrative legitimation, namely the control model and the deliberative model. A key difference between these models is the way they define the role of public administration, including the nature of administrative discretion. Moreover, each model conceptualises the nature of safety, risk, scientific evidence and uncertainty differently. The control model follows principalagent or ‘transmission belt’ theories of the administrative state.22 Applied to the EU, according to the control model the EU public administration acts as the agent of the EU legislature as its democratically legitimated principal. Wide discretionary powers are seen as problematic, and need to be limited and controlled in order to maintain the democratic transmission between EU legislative commands and their administrative implementation. EU administrative law then is seen as the main instrument to control administrative power, and to protect individual liberties from arbitrary actions via stringent judicial control.23 17 See C Harlow and R Rawlings, Law and Administration 3rd edn (Cambridge, Cambridge University Press, 2009) 1–48. 18 Fisher, above n 11. 19 ibid. 20 See Anderson, above n 7. 21 This is given that EU administrative law is a relatively young field, and its doctrinal developments are lagging behind their national counterparts. See J Schwarze, European Administrative Law 1st edn revised (London, Sweet & Maxwell 2006); E Schmidt-Assmann and B Schöndorf-Haubold, Der Europäische Verwaltungsverbund: Formen und Verfahren der Verwaltungszusammenarbeit in der EU (Tubingen, Mohr Siebeck, 2005). 22 Harlow and Rawlings refer to it as the red-light model; see Harlow and Rawlings, above n 17, 22. 23 Harlow and Rawlings associate the red-light model with liberal theories in the spirit of Friedrich Hayek; see ibid. In the US context, Anderson refers to it as the private law model; see Anderson, above n 7.

Expertise as Justification

171

In the field of risk regulation, scientific expertise contributes to that control. By rationalising administrative decision-making, it becomes an ex ante control mechanism that facilitates ex post judicial review. Therefore, the scientific advice of the European Food Safety Authority (EFSA) regarding the Commission authorisation process on GMOs is seen as instrumental to delivering the facts, which the Commission can apply to the legislative commands in this field. Because of its control function, scientific advice constitutes the main input into the authorisation process leaving the Commission with little political discretion. Moreover, in the control model, technological risks in general and GMO risks in particular are understood as quantifiable, scientific uncertainty as manageable and knowledge as an objective body of truth claims based on established facts.24 In contrast, according to the deliberative model, the EU public administration acts as a relatively autonomous institution serving the public good. Instead of emphasising judicial control, it emphasises the value of deliberation in administrative decision-making.25 According to this model the Commission, comitology and the EFSA are all part of a composite EU administration,26 which legitimises its action on the basis of the cooperative and deliberative nature of its joint decisionmaking.27 The Commission is granted substantive discretion to achieve broadly formulated legislative goals and objectives in cooperation with other actors, which includes the right to deviate from EFSA opinions as well as to base decisions on socalled other legitimate factors, such as socio-economic and ethical concerns. The exercise of that discretion is legitimated on the basis of the deliberative nature of Commission decision-making, which is supported by reason-giving requirements. The role of EFSA is to inform this process. However, both the limits and valueladen nature of scientific risk assessment are recognised, and the role of EFSA is framed as one of communicating openly about these limits including knowledge gaps and scientific uncertainties. Since complexity and uncertainty are recognised as being pervasive, the Commission is granted a wide range of discretion to make policy judgements on the basis of available scientific evidence. In this way EFSA’s advice is seen as only one input into the process. In this model, knowledge is seen as an unstable body of contested truth claims.28 Finally, the role of EU courts is,

24 See also S Shapiro, E Fisher and W Wagner, ‘The Enlightenment of Administrative Law: Looking Inside the Agency for Legitimacy’ (2012) 47 Wake Forest Law Review 469. 25 Fisher, above n 11. Harlow and Rawlings associate what they call the green-light model with progressive functionalist theories of the administrative state. In the US context, Anderson refers to it as the public law model; see Anderson, above n 7. 26 See HCH Hofmann and AH Turk, Legal Challenges in EU Administrative Law: Towards an Integrated Administration (Cheltenham, Edward Elgar Publishing Ltd, 2009). 27 C Joerges and J Neyer, ‘From Intergovernmental Bargaining to Deliberative Political Processes: The Constitutionalisation of Comitology’ (1997) 3 European Law Journal 273; CF Sabel and J Zeitlin, Experimentalist Governance in the European Union: Towards A New Architecture (Oxford, Oxford University Press, 2012). 28 See European Commission, ‘The Future of Europe Is Science’, A Report of the President’s Science and Technology Advisory Council, October 2014 available at https://ec.europa.eu/digital-agenda/en/ news/report-science-and-technology-advisory-council-stac-future-europe-science.

172

Maria Weimer and Gaia Pisani

while acknowledging substantive discretion on the part of the Commission, to act as catalysts of deliberation. In other words, judicial review should help ensure the functioning of ‘process values’ in the administrative process such as reason-giving, transparency, consistency, participation, and the quality of the scientific basis of GMO authorisations.29 It should be noted that neither of the above models of administrative legitimation provide for a fully satisfactory answer to the challenges of contemporary risk administration. The control model relies on legal formalism, and tries to preserve the traditional idea of rule of law in the liberal constitutional state, including the idea of democratic control and accountability of public administration. Yet it does so at the risk of becoming a fiction, and of not being able to normatively capture the actual complexity of the exercise of administrative power, especially in a multilevel supranational system. As early as 1975, US scholar Richard Stewart observed that: [D]isintegration of the traditional model, which has proven unsuccessful in its effort to reconcile the discretionary power enjoyed by agencies with the basic premise of the liberal state that the only legitimate intrusions into private liberty and property interests are those consented to through legislative processes.30

It can be assumed that this problem intensified over the last few decades due to increased global interdependence, stronger regional integration and the subsequent transfer of executive powers beyond the state. In contrast to the control model, the deliberative model offers a more realistic and pragmatic approach to administrative problem solving which is guided by general public interest. However, it can be criticised as problematic for sacrificing legal certainty, overestimating the potential of deliberation in democratic problem-solving, and giving raise to concerns over the arbitrary use of discretionary power.31 Moreover, rather than being a fully-fledged model, at present it rather serves as a container notion for a series of alternative visions of public administration which are still lacking conceptual sharpness.32 The two models therefore should be understood as representing contrasting points, the tension between which fosters a productive search for an adequate concept of administrative legitimation, in this case in the EU. Looking through this lens at concrete examples of the EU administrative legal practice helps to ascertain the suitability of one or the other model in a particular context, as well as confirms

29 J Scott and S Sturm, ‘Courts as Catalysts: Rethinking the Judicial Role in New Governance’ (2007) 13 Columbia Journal of European Law 565. 30 R Stewart, ‘The Reformation of American Administrative Law’ (1975) 88 Harvard Law Review 1667. For a different type of critique of the US administrative state see M Dorf and C Sabel, ‘A Constitution of Democratic Experimentalism’ (1998) 98(2) Columbia Law Review 267. 31 For a discussion of advantages and disadvantages of both models see Fisher, above n 11, 251. 32 See F Rodriguez, ‘The Scientification of Participation’ in M Van Asselt, E Versluis, and E Vos (eds), Balancing Between Trade and Risk. Integrating Legal and Social Science Perspectives (Abingdon, Routledge, 2013).

Expertise as Justification

173

instances of gradual legal and administrative change that is already taking place. What is of interest here is to observe the process whereby the challenges of the techno-scientific progress are transforming ideas and concepts of administrative legitimation, including concepts of administrative law. It therefore should be seen as part of the EU’s ongoing process of introspection and contestation over the ‘right’ model of risk regulation. In the remainder of this chapter we apply these two models in our analysis of two controversial cases involving the EU’s authorisation of GMOs. The goal is to trace through past discourses about EU administrative legitimation, and to analyse the way in which they shape the practice of GMO decision-making, including the role of scientific expertise therein.

III. Administrative Legitimation and GMO Authorisations: The Long Tale of Herculex and Amflora In recent years two processes of EU authorisation of GMOs have been particularly controversial and also emblematic of the regulatory crisis in this field more generally. The first case concerns the GM Maize 1507 branded as Herculex, and the second the GM potato EH92-527-1 branded as Amflora.33 Both concern applications for cultivation, ie for the agricultural use of GMOs, which has been particularly controversial because of its potential environmental and socio-economic implications.34 Moreover, both cases involved a long and complicated authorisation process marked by: persistent contestation of both the EFSA’s risk assessment and the Commission’s risk management; several EFSA opinions as a result of recurrent requests by the Commission; and administrative delays. Moreover, in both cases the EU General Court has (for the first time in the field of GMO regulation) reviewed the legality of the Commission’s administrative actions. Previous research suggested that EFSA opinions on GMOs ‘generally indicate no uncertainty’ that might trigger extra risk management measures, and that EFSA has ‘framed scientific uncertainties in such a way that it can be resolved by extra information, or can be readily managed, or deemed irrelevant to any risk’.35 Therefore, EFSA’s risk assessment has been described as ‘uncertainty intolerant’ which means that it is reluctant to both acknowledge and communicate scientific

33

Our analysis is based on document review and individual semi-structured interviews. Lee, above n 16; Lee, above n 9; Weimer, above n 5. 35 L Levidow and S Carr, ‘Europeanising Advisory Expertise: The Role of “Independent, Objective, and Transparent” Scientific Advice in Agri-Biotech Regulation’ (2007) 25 Environment and Planning C: Government and Policy 880; L Levidow, S Carr and D Wield, ‘European Union Regulation of AgriBiotechnology: Precautionary Links Between Science, Expertise and Policy’ (2005) 32 Science and Public Policy 261, 270, 273. 34

174

Maria Weimer and Gaia Pisani

uncertainty, instead of genuinely and systematically investigating it.36 The Commission, on the other hand, was found to merely rubberstamp EFSA’s opinions acting ‘as a blind driver taking directions from the passenger in the back seat,’37 thereby inheriting EFSA’s uncertainty intolerance. Our analysis of decision-making in both cases (see table 1 below) offers new findings in this respect. We register a slight yet meaningful gradual change in the way both EFSA and the Commission have acted within the authorisation process, notably with regard to how they as EU administrative bodies have dealt with scientific advice, risk and uncertainty. The following two sections present our findings in more detail. Table 1: Chronology of authorisation processes of Herculex and Amflora Herculex (maize 1507)38

Amflora (Potato EH92-527-1)39

January 2005

EFSA issues first opinion

April 2004

EFSA issues opinion on the use of antibiotic resistance marker genes in GM plants

November 2006

EFSA issues update with Annex on non-target organisms

February 2005

The World Health Organization (WHO) report classifies kanamycin and neomycin (antibiotics to which Amflora is resistant) as critically important for human medicine

2007 (date unknown)

Commission drafts a decision rejecting authorisation (but never submits it to the comitology committee)

December 2005

EFSA issues first opinion on Amflora

October 2008

EFSA issues updated opinion reviewing 11 independent studies

December 2006

Commission submits draft authorisation to the comitology committee (which later fails to reach a decision)

February 2009

Commission submits proposal to authorise to the comitology committee (which later fails to reach a decision)

February 2007

The European Medicine Agency (EMEA) issues opinion concluding that neomycin and kanamycin cannot be classified as of no or minor importance (continued)

36

Van Asselt and Vos, above n 9. Vos and F Wendler, Food Safety Regulation in Europe: A Comparative Institutional Analysis (Mortsel, Intersentia, 2007). 38 Application submitted in 2001. 39 Application submitted in 2003. 37 E

Expertise as Justification

175

Table 1: (Continued) Herculex (maize 1507)

Amflora (Potato EH92-527-1)

April 2010

A new (Testbiotech) study criticizes EFSA’s assessment

March 2007

EFSA states that the use of nptII gene in GM plants will not compromise the therapeutic effects of kanamycin and neomycin

November 2011

EFSA issues updated opinion acknowledging differences between the Bt toxins Cry1F (contained in Herculex) and Cry1Ab (contained in other GMOs) and acknowledging some potential risks

June 2007

Commission submits draft authorisation to the Council (which later fails to reach a decision)

October 2012

EFSA issues a supplementing opinion acknowledging the need for further studies on the impact of Cry1F on nontarget species

June 2009

EFSA issues consolidated opinion acknowledging some uncertainty; a minority opinion of the BIOHAZ panel concludes that a risk assessment is not possible due to scientific uncertainty

November 2012

EFSA issues another supplementing opinion considering new scientific studies

March 2010

Commission authorises Amflora without a re-submission to the comitology committee

September 2013

The General Court condemns the Commission for failure to act during the authorisation process

December 2013

The General Court annuls the authorisation on procedural grounds

November 2013

Commission submits draft authorisation decision to the Council

February 2014

Council votes with 19 Member States against authorisation (at the time of writing, the Commission had not yet issued its final decision)

176

Maria Weimer and Gaia Pisani

A. The EFSA as Expert Agency: From Control to Deliberation Our empirical analysis reveals a gradual change in approach in recent years on the part of the EFSA GMO panel,40 as expressed in its willingness to both acknowledge and communicate knowledge gaps, flaws in methodology, potential risks and remaining scientific uncertainty. We find that this shift has been the result of external pressure from several Member States, other scientific authorities as well as recurrent requests of the Commission.

i. Risk assessment of Herculex41 In 2001, Pioneer-DuPont requested authorisation to place on the market and to cultivate maize 1507 under Directive 2001/1842 and Regulation 1929/2003.43 Maize 1507 is an insect-resistant genetically modified maize characterised by the insertion of two genetic constructs: one which produces Cry1F, a so-called Bt-protein44 that makes the maize resistant to certain Lepidoptera;45 and another which makes the maize tolerant to glufosinate ammonium, a herbicide the use of which will be phased out in the EU by 2017.46 Herculex has raised concerns with regard to the risk of resistance development in target organisms, as well as the uncertain impact of the Bt-toxin on non-target organisms. In its initial risk assessments the EFSA did not seriously engage with these concerns. Likely being aware of the political pressure to provide authoritative scientific advice as the new EU expert agency, at the time it framed its advice in definite terms following the ‘uncertainty intolerant’ pattern previously identified by Van Asselt and Vos.47 For example, in its first opinion in 2005,48 EFSA employed a so-called ‘plausibility proof ’49 by stating that the cultivation of maize 1507 ‘will not have an 40

The GMO panel is the scientific panel responsible for risk assessment of GMOs at EFSA. See for more detailed empirical analysis Maria Weimer and Gaia Pisani, ‘Biotechnology: The EU Adventures of Herculex’ (2014) 5 European Journal of Risk Regulation 208. 42 Dir 2001/18/EC on the deliberate release into the environment of genetically modified organisms and repealing Council Dir 90/220/EEC, [2001] OJ L/206. 43 Reg (EC) 1929/2003 on genetically modified food and feed, [2003] OJ L 268/1. On the legal framework, see M Weimer, ‘Applying Precaution in EU Authorisation of Genetically Modified Products—Challenges and Suggestions for Reform’ (2010) 16 European Law Journal 624. 44 Bt (Bacillus thuringiensis) are bacteria which naturally occur in soil. Some produce a protein crystal which is toxic to insects. Bt Cry1F is the result of a modification of the natural occurring Cry1Fa2 gene. 45 Larvae of the European corn borer and of the Mediterranean corn borer. 46 Commission Implementing R (EU) No 365/2013 of 22 April 2013 amending Implementing R (EU) No 540/2011, OJ (2013) L111/27. 47 Van Asselt and Vos, above n 9. 48 EFSA, Opinion of the Scientific Panel on Genetically Modified Organism on a request from the Commission related to the notification (Reference C/ES/01/01) for the placing on the market of insecttolerant genetically modified maize 1507 for import, feed and industrial processing and cultivation, under Part C of Directive 2001/18/EC from Pioneer Hi-Bred International/Mycogen Seeds, Question No EFSA-Q-2004-072, The EFSA Journal (2005) 181, 1–33. 49 Van Asselt and Vos define those as ‘statements by experts in which their risk judgments are cast as conclusive evidence on the existence or non-existence of an uncertain risk. There is no room for uncertainty in plausibility proofs: risks exist or they do not’; see Van Asselt and Vos, above n 9, 282. 41

Expertise as Justification

177

adverse effect’50 on human health and the environment. It thus framed its advice as conclusive evidence, in which there is no room for uncertainty. Moreover, EFSA employed so-called ‘boundary work’, a process by which the agency draws a boundary between risk assessment and risk management, which in turn allows it to dismiss certain risk claims.51 Several Member States pressured EFSA to include monitoring and mitigation issues in its risk assessment, which was followed by a Commission request to cooperate with the national authorities and to update its risk assessment in this respect.52 EFSA, however, rejected these requests by stating that monitoring arrangements in relation to potential adverse effects on nontarget organisms are outside of the scope of risk assessment.53 Finally, EFSA based its initial risk assessment on controversial analogies between different types of the Bt-toxin. As mentioned above, Herculex contained the Bt-toxin Cry1F, which differed from Bt-toxins contained in other transgenic plants (eg the Monsanto’s Mon810 maize containing Cry1Ab). A controversial issue was to what extent different Bt-toxins could be expected to have either similar or different effects on the environment. While no peer-reviewed literature on the effects of the Herculex toxin Cry1F existed at that time, some studies highlighted the significantly higher toxicity of Cry1F vis-à-vis Cry1Ab on certain insects.54 In 2008, the Commission asked EFSA to review its previous opinion in the light of 11 independent scientific studies,55 which explored the effects and potential toxicity of Cry1Ab on some target and non-target organisms. However, the GMO panel only vaguely engaged with these studies, dismissing all of them as not providing new scientific evidence.56 Most notably, EFSA continued to assert that the two proteins (Cry1F and Cry1AB) had high affinities to one another and that their concentration in pollen was relatively low, which was based on EFSA’s reliance on existing studies of Cry1Ab for the assessment of Herculex.57

50

See above n 48, 2, emphasis added. According to Gyrien, boundary work is defined as a strategic and purposeful act in which boundaries are drawn between realms, for example, between science and non-science and between science and politics; see TF Gieryn, Cultural Boundaries of Science: Credibility on the Line (Chicago, University of Chicago Press, 1999); Van Asselt and Vos, above n 9, 288. 52 EFSA, Clarifications of the Scientific Panel on Genetically Modified Organisms following a request from the Commission related to the opinions on insect resistant genetically modified Bt11 (Reference C/F/96/05.10) and 1507 (Reference C/ES/01/01) maize, 19 November 2006, 2. 53 ibid 5–7. 54 AV Hanley, ZY Huang, WL Pett, ‘Effects of Dietary Transgenic Bt corn Pollen on Larvae of Apismelliferaand Galleria mellonella’ (2003) 42(4) Journal of Apicultural Research 77–81. 55 For a comprehensive list, see EFSA, Scientific Opinion of the Panel on Genetically Modified Organisms on a request from the European Commission to review scientific studies related to the impact on the environment of the cultivation of maize Bt11 and 1507, Question No EFSA-Q-2008-679, The EFSA Journal (2008), 851, 1–27 at 6–7. 56 ibid. 57 In contrast, in 2007 EFSA launched a call for proposal to gather scientific data on Bt proteins, in order to better analyse their behaviour in GM plants. See EFSA Call CFP/EFSA/GMO/2007/01, Cry proteins and their expression in microorganisms and genetically modified plants. This call was also mentioned in the proposal for the rejection of the authorisation of Herculex as drafted by Commissioner Dimas in 2007 (see n 77 below). 51

178

Maria Weimer and Gaia Pisani

However, in 2010 the Commission yet again asked EFSA for an update following the release of a new scientific study conducted by a non-profit research institute.58 The research institute raised a number of objections to EFSA’s conclusions, namely the inadequacy of basing an environmental impact assessment on analogies with other Bt toxins.59 It argued that, instead of relying on partial information and overlooking the existing uncertainties, EFSA should have requested further data from the applicant. This time EFSA delivered a more nuanced response partially acknowledging methodological flaws and the need for further studies.60 A new opinion also acknowledged some potential risks.61 First, EFSA acknowledged the higher concentration of Cry1F in Herculex as compared to other Bt-toxins in other GMOs.62 Second, it identified potential risks in relation to the evolution of resistance in target pests and the toxicity of the Bt-protein to sensitive non-target lepidoptera, such as European butterflies. The Panel acknowledged the need for further studies to fill the knowledge gap,63 but did not consider this as sufficient to alter the outcome of its risk assessment. Instead, it proposed ‘appropriate management measures’, notably insect resistance management strategies to delay the evolution of resistance in target pests and field studies to monitor the effects on non-target Lepidoptera. It concluded that Herculex is ‘unlikely to raise safety concerns for the environment’.

ii. Risk Assessment of Amflora BASF Plant Science GmbH requested the authorisation for the placing on the market of the potato Amflora under Directive 2001/18/EC and Regulation (EC)

58 The study at issue is A Bauer-Panskus and C Then, ‘Testbiotech opinion on the application for market approval of genetically modified maize 1507 (DAS-Ø15Ø7-1)’, April 2010 available at www. testbiotech.org/sites/default/files/Opinion%20Testbiotech%201507%20Maize%20engl_0.pdf. EFSA makes explicit reference to it in: EFSA, Minutes of the 61st plenary meeting of the scientific panel on genetically modified organisms held on 20–21 October 2010 in Parma, Italy, adopted on 1 December 2010, 7. 59 Notably, they claimed that maize 1507 produced a high concentration of Bt toxin in pollen, and that this evidence should have significantly altered the risk assessment outcome as to the development of insect resistance and the impact on soil, air and water. 60 See EFSA (2010), above n 58, 7. See also two supplementing opinions: EFSA, Scientific Opinion supplementing the conclusions of the environmental risk assessment and risk management recommendations on the genetically modified insect resistant maize 1507 for cultivation, EFSA Journal 2012; 10(11):2934; and EFSA, Scientific Opinion updating the risk assessment conclusions and risk management recommendations on the genetically modified insect resistant maize 1507, EFSA Journal 2012; 10(10):2933. 61 EFSA, Scientific Opinion updating the evaluation of the environmental risk assessment and risk management recommendations on insect resistant genetically modified maize 1507 for cultivation, EFSA Journal 2011; 9(11):2429. 62 The concentration of Cry1F in pollen of Herculex was about 350 times the concentration of Cry1Ab in MON810. 63 See EFSA (2011), above n 61, which at 3 states ‘The EFSA GMO Panel recommends caution when predicting future responses of the European and Mediterranean corn borer in the EU based on experiences elsewhere, as resistance evolution in target insect pests is dependent upon many factors’.

Expertise as Justification

179

1829/2003 in 2003. Amflora is a potato intended for industrial use only, and its modification allows greater production of starch. It is also characterised by the insertion of the marker gene nptII, which is antibiotic resistant. The use of nptII as a marker gene was the main focus of the objections raised by Member States and NGOs to the authorisation of Amflora. The major concern regarded the risk that antibiotic resistance could be transferred horizontally (namely from plants to bacteria), with potential implications in the treatment of infections and transmissible diseases such as tuberculosis in humans and animals. The first EFSA opinion64 on Amflora issued in 2005 followed the familiar pattern of framing advice in certain terms, and providing plausibility proofs.65 EFSA concluded that the GM potato was ‘no more likely to cause adverse effect on human and animal health or the environment than conventional potato’.66 Most notably, EFSA classified kanamycin and neomycin as antibiotics with ‘no or only minor therapeutic relevance in human medicine and only restricted use in defined areas of veterinary medicine’,67 which is why the use of the Amflora marker gene nptII was not found to pose any risks. However, EFSA’s classification contradicted a WHO report published in February 2005 in which kanamycin and neomycin were classified as ‘critically important’ antibacterial agents for human medicine, used in the treatment of E-coli and tuberculosis.68 This led the Commission to request the advice of another EU agency, the European Medicines Agency (EMEA), on the validity of the EFSA assessment of the nptII. In line with the WHO report the EMEA challenged the EFSA opinion affirming that kanamycin and neomycin belong to a class of antibiotics (aminoglycosides) that has become more and more important in the treatment of some serious infectious diseases such as tuberculosis.69 This led the Commission to request EFSA to reconsider its previous assessment in the light of both the WHO and the EMEA opinions. In a statement adopted in March 2007, the GMO Panel reiterated its findings claiming that the EMEA and

64 EFSA, Opinion of the Scientific Panel on Genetically Modified Organisms on an application (Reference EFSA-GMO-UK-2005-14) for the placing on the market of genetically modified potato EH92527-1 with altered starch composition, for production of starch and food/feed uses, under Regulation (EC) No 1829/2003 from BASF Plant Science; Question No EFSA-Q-2005-070; EFSA Journal 2006; 324, 1–20. 65 On this term see above n 49. 66 See above at n 64, 17, emphasis added. 67 The classification used by EFSA was developed in a previous opinion evaluating the use of antibiotic resistance genes as marker genes in GM plants more generally, see EFSA, Opinion of the Scientific Panel on Genetically Modified Organisms on the use of antibiotic resistance genes as marker genes in genetically modified plants; Question No EFSA-Q-2003-109; EFSA Journal 2004; 48, 1–18, at 11. 68 World Health Organization, Critically Important Antibacterial Agents for Human Medicine for Risk Management Strategies of Non-human Use, report of a WHO working group consultation, 15–18 February 2005, Canberra, Australia, available at http://apps.who.int/iris/bitstream/10665/43330/1/9241593601_ eng.pdf. 69 European Medicine Agency, Presence of the antibiotic resistance marker gene nptII in GM plants for food and feed uses, EMEA/CVMP/56937/2007—Final, 22 February 2007.

180

Maria Weimer and Gaia Pisani

WHO reports would not change its previous assessment on the safety of using nptII in GMOs. It concluded: The GMO Panel agrees with the EMEA that the preservation of the therapeutic potential of the aminoglycoside group of antibiotics is important. The Panel is also of the opinion that the therapeutic effect of these antibiotics will not be compromised by the presence of the nptII gene in GM plants, given the extremely low probability of gene transfer from plants to bacteria and its subsequent expression. Furthermore, the GMO Panel considers it very unlikely that the presence of the nptII gene in GM plants will change the existing widespread prevalence of this antibiotic resistance gene in bacterial sources in the environment. The GMO Panel also points to evidence which indicates that integration of the nptII gene would only be one of many mechanisms by which bacteria could become resistant to aminoglycosides such as kanamycin. Therefore, the GMO Panel reiterates its earlier conclusions (EFSA, 2004).70

Hence, EFSA barely engaged with the contesting scientific views expressed by the WHO and EMEA. It defined its response in a decisive and clear-cut way (‘will not be compromised’), and evaded the uncertainty surrounding the use of nptII by employing terms which create the appearance of certainty and low risk (‘extremely low probability’ and ‘very unlikely’). However, given the continuing criticism of EFSA’s work including pressure from the Member States and some NGOs the Commission asked the GMO Panel to work together with another EFSA scientific panel, the Biological Hazard (BIOHAZ) Panel, to deliver a joint scientific opinion on the use of antibiotic resistance marker genes in GM plants. In June 2009, EFSA issued a consolidated opinion (‘EFSA consolidated opinion’) acknowledging some uncertainties, knowledge gaps, and admitting a number of limitations.71 Similar to the Herculex case, EFSA gradually shifted towards tolerating more uncertainty while at the same time concluding that the identified risks were not significant enough to justify a modification of its assessment of the safety of the nptII gene. The Panel concluded that: Notwithstanding these uncertainties, the current state of knowledge indicates that adverse effects on human health and the environment resulting from the transfer of these two antibiotic resistance genes from GM plants to bacteria, associated with the use of GM plants, are unlikely.72

Importantly, for the first time since the establishment of EFSA, two scientists (of the BIOHAZ Panel) disagreed with the majority opinion and raised objections 70 EFSA, Statement of the Scientific Panel on Genetically Modified Organisms on the safe use of the nptII antibiotic resistance marker gene in genetically modified plant, adopted on 22–23 March 2007, emphasis added. 71 EFSA, Consolidated presentation of the joint Scientific Opinion of the GMO and BIOHAZ Panels on the ‘Use of antibiotic resistance genes as marker genes in genetically modified plants’ and the Scientific Opinion of the GMO Panel on ‘Consequences of the opinion on the use of antibiotic resistance genes as marker genes in genetically modified plants on previous EFSA assessments of individual GM plants’; EFSA-Q-2009-00589 and EFSA-Q-2009-00593; EFSA Journal 2009; 1108, 1–82 at 3, ‘There are limitations related among others to sampling, detection, challenges in estimating exposure levels and the inability to assign transferable resistance genes to a defined source’; see also at 67–82. 72 ibid at 3.

Expertise as Justification

181

to the assessment of the likelihood of horizontal gene transfer and its potential implications for the medical use of kanamycin and neomycin. In their minority opinion they found that ‘it would be imprudent to regard resistance to any antibiotic as being of little or no relevance to human health’;73 and that the level of risk for human health and the environment could not be assessed, as the probabilities of horizontal gene transfer were below the detection limits. Hence, the minority opinion openly acknowledged that a risk assessment was not possible under the given circumstances. The EFSA consolidated opinion, however, dismissed these arguments without further explanation, by simply stating that it had taken into due account the minority opinion and that ‘from a scientific point of view, the joint opinion of 2009 did not require further clarification or additional work of a scientific nature’.74 We observe, therefore, that the pressure exerted on EFSA by national authorities, independent scientific studies and other agencies, as well as the Commission’s recurrent requests, had forced the agency to engage more seriously with competing views and scientific uncertainty. This resulted in a more nuanced risk assessment combined with more explicit communication of uncertainty information. However, this opening up in EFSA’s opinions ultimately did not result in a modification of the overall conclusions regarding the risks of both GMOs. This reveals that EFSA understood uncertainty as something negligible for the overall risk assessment, because it was assumed that uncertainty is not pervasive, but can be reduced through risk management. Moreover, EFSA’s tendency to conclude the risk assessment with a definitive and certain response indicates an understanding of its role as risk assessor in line with the control model. EFSA sees itself as instrumental for the provision of facts, to which the Commission can apply legislative commands. The role of EFSA therefore is not merely to facilitate the exercise of administrative discretion by the Commission, but to provide for a decisive basis for it thereby controlling discretion.

B. The Commission as Risk Manager: From Deliberation to Control The Commission’s conduct in both the Herculex and Amflora cases follows a similar pattern. A cross-analysis of these cases demonstrates the ambiguous role of the Commission in its attempt to overcome the exacerbation of scientific, 73

See above n 71 at 81, emphasis added. As explained in Case T-240/10, Hungary v European Commission, see below n 93 at para 39: ‘On 28 April 2009, the Director of EFSA asked the chairmen of the GMO and BIOHAZ panels and the chairman of the joint working group whether the two dissenting opinions required further work of a scientific nature. On 25 May 2009, those chairmen replied that, during the preparation of the joint opinion of 2009, the content of the two dissenting opinions had largely been taken into account, so that, from a scientific point of view, the joint opinion of 2009 did not require further clarification or additional work of a scientific nature’. 74

182

Maria Weimer and Gaia Pisani

regulatory and political conflicts in this area, and in maintaining its own legitimacy as the EU’s public administrator. While EFSA gradually shifted from supporting an uncertainty intolerant to a more uncertainty tolerant approach in both authorisations, the opposite holds true for the Commission. Initially, in both cases the Commission had operated in a rather cautious and politically sensible way, arguably at the expense of an efficient and timely administrative process. It engaged with different scientific authorities encouraging their collaboration with EFSA. It repeatedly put pressure on EFSA to respond to competing scientific views expressed either by national authorities or independent studies. Therefore, the Commission arguably acted as an administrative body with substantive discretion whose task it is to foster a deliberative process, which should inform the final authorisation decision on a scientifically complex and politically sensitive application. Likely aware of the resistance on the part of the Member States that it might encounter in comitology, the Commission tried to accommodate different concerns in a pre-emptive manner. It thereby acted in a manner indicative of the deliberative model. In the Herculex case, its self-understanding as a body with wide disretion is evidenced by the fact that for the first time in the field of GMOs it appears that the Commission attempted to depart from EFSA’s advice, and to refuse authorisation for Herculex. In 2007, former Commissioner Stavros Dimas presented a draft decision proposing not to approve the maize for cultivation on the basis of the high degree of scientific uncertainty and the lack of scientific knowledge over significant traits of this GMO. The draft proposal stressed the need for an approach based on the precautionary principle, and indicated that the degree of uncertainty was so high that it was ‘not possible to establish appropriate management measures which would effectively mitigate the potential damage on the environment’.75 However, instead of pursuing this proposal in comitology, the Commission again returned to EFSA. After the agency reaffirmed its previous findings about the safety of Herculex, the Commission reversed its position. It drafted a new proposal, this time in favour of authorisation by relying on EFSA’s statement that there is no evidence of the adverse effects of Herculex on human health and the environment. It is striking that the previous concerns about existent knowledge gaps no longer appeared in the new draft.

75 The document reported a number of objections raised by MSs concerning the impact on soil organisms and anthropods, the long-term effects of the Bt toxin on the environment, and it also pointed out that, ex lege, environmental risk assessments need to consider potential effects on nontarget organisms and cumulative long-term effects which, at that time, were not adequately evaluated. Importantly, the Commission seemed to question the reliability of a meta-review of the applicant’s assessment, marking the lack of crucial data such as the concentration of Bt toxin, and contesting the methodology used by EFSA in its assessment. See EU Commission (2007), Draft Commission Decision of … concerning the placing on the market, in accordance with Directive 2001/18/EC of the European Parliament and of the Council, of a maize product (Zea mays L, line 1507) genetically modified for resistance to certain lepidopteran pests and for tolerance to the herbicide glufosinateammonium. ENV/07/.

Expertise as Justification

183

This clearly shows that different ideas about the nature of the Commission’s discretion, including its ability to deviate from EFSA’s advice, have been in contention internally within the Commission. Interestingly, both decision proposals on Herculex have been drafted under Commissioner Dimas, so the change of approach is clearly not due to a change in the Commission’s leadership. Instead, there are indications that external rather than internal factors were decisive in this change in position. Following the WTO EC-Biotech panel report in 2005,76 which was unfavourable to the EU, the Commission found itself under strong external pressure to adopt an interpretation of the Commission’s discretion along the lines of the control model. The latter implied that the Commission’s discretion to reject authorisation was considerably reduced in light of EFSA’s findings. In the words of a Commission official, ‘[t]he Commission felt that it should apply the legislation’.77 Both processes of authorisation were marked by delay, inaction and ultimately rulings of the EU General Court.78 As usual for GMO authorisations,79 the comitology process failed to produce the necessary majority, which meant that, under the rules applicable at that time, the Commission had to forward the draft authorisation proposals to the Council. However, in the Herculex case, it was not until November 2013, 12 years since the filing of the Pioneer application, that the Commission submitted its authorisation proposal to the Council. The Commission justified the proposal on the basis that the EFSA had not identified new scientific publications that would have invalidated its previous opinions on the safety of Herculex.80 What is most striking about this proposal is the absence of any references to the potential risks and uncertainties identified by the EFSA in its latest more nuanced opinions.81 In February 2014 the Council voted on the authorisation of Herculex. 19 Member States opposed the proposal, five were in favour, and four abstained.82 Despite the high number of opposing Member States, no qualified majority was

76 WTO Panel Report, European Communities—Measures Affecting the Approval and Marketing of Biotech Products, WT/DS291 from 29 September 2006. 77 Interview with a member of the Commission Legal Service on 21 February 2014. 78 See below at s IV. Overall, in the Herculex case it took two court actions by Pioneer to move the procedure along. The first one was brought in 2007 (Case T-139/07) forcing the Commission to submit a draft proposal to the comitology committee, after which the EU Court closed the case because there was no need to further adjudicate the issues. 79 See Weimer, above n 5. 80 European Commission, Proposal for a Council decision concerning the placing on the market for cultivation, in accordance with Directive 2001/18/EC of the European Parliament and of the Council, of a maize product (Zea mays L, line 1507) genetically modified for resistance to certain lepidopteran pests, COM(2013) 758 Final, 6 November 2013. 81 With the exception of the potential Bt toxin resistance in corn borers, for which the Commission refers to an EFSA opinion from 2005. See ibid, European Commission Proposal (2013), para 17. 82 Spain, the UK, Finland, Estonia and Sweden voted in favour, while Portugal, Czech Republic, Belgium and, quite surprisingly, Germany abstained. France, Italy, Hungary, the Netherlands, Slovenia, Slovakia, Austria, Poland, Greece, Romania, Bulgaria, Croatia, Cyprus, Latvia, Ireland, Denmark, Malta, Luxembourg and Lithuania voted against the Commission’s proposal for approval.

184

Maria Weimer and Gaia Pisani

reached against the Commission proposal, paving the way for the Commission to approve Herculex. Following the vote, 12 ministers83 addressed a letter to former Commissioner Tonio Borg asking the Commission to withdraw its proposal in the light of the broad opposition of the majority of stakeholders. The Commission has until present not proceeded with the authorisation, nor taken a formal position in this case.84 In the Amflora case, the Commission presented a proposal to the Council in June 2007. After the Council failed to gather the necessary majority to vote on the proposal, the Commission went on to authorise Amflora in March 2010.85 In the preamble of the authorisation decision, the Commission included a general reference to the debate between the EFSA, WHO and EMEA over the risks of using nptII as marker gene. However, it did not engage with the minority opinion included in the 2009 EFSA consolidated opinion. Moreoever, it did not consider it necessary to again consult the comitology committee in the light of the new consolidated opinion. Later on, this resulted in the General Court annulling the authorisation for breach of procedural rules (see below at section IV). We conclude that the EFSA’s shift towards a more uncertainty-tolerant deliberative approach was ultimately not followed by the Commission, but was, instead, met with a strikingly uncertainty intolerant response along the lines of the control model. We argue that an important factor influencing the Commission has been that it interpreted its mandate to implement GMO legislation along the lines of the control model: as being restrained by the EFSA majority opinion. For example, in its press release on the authorisation of Herculex, the Commission has explained its proposal with the obligation to comply with the ruling of the EU General Court and thereby with the rule of law.86 This is astonishing at least in the sense that the Court found the Commission guilty of failure to act, but did not impose on the Commission how to act in this case. Therefore, it seems that the Court’s ruling did not prevent the Commission from adopting a different proposal, for example, rejecting the authorisation. It shows, however, that when it comes to taking the final decision the Commission does not see itself as a body with substantive discretion able to make a policy judgement on the basis of scientific information, if it means deviating from its expert agency.

83 Letter dated 12 February 2014, signed by the Ministers of Austria, Bulgaria, Cyprus, France, Hungary, Italy, Latvia, Lithuania, Luxembourg, Poland, Slovenia and Malta. The text of the letter is available at http://m.greenpeace.org/greece/Global/greece/image/2014/gmos/gmo_general_petition/2379_ 001.pdf. 84 At the time of writing, no decision on Maize 1507 was yet adopted. 85 Commission Decision of 2 March 2010 concerning the placing on the market, in accordance with Directive 2001/18/EC of the European Parliament and of the Council, of a potato product (Solanumtuberosum L line EH92-527-1) genetically modified for enhanced content of the amylopectin component of starch, C (2010) 1193, OJ (2010) L 53/11. 86 European Commission MEMO/13/960, Statement by EU Health Commissioner Tonio Borg, on Commission’s decision on GM Pioneer 1507, 6 November 2013, available at http://europa.eu/rapid/ press-release_MEMO-13-960_en.htm.

Expertise as Justification

185

IV. Judicial Responses In both the Herculex and the Amflora cases, the EU General Court had for the first time the opportunity to scrutinise the administrative process of GMO authorisations, thereby acting in its function as an administrative court. In doing so, it revealed an understanding of administrative legitimation of the Commission along the lines of the deliberative model. While granting the Commission a wide discretion to make factual assessments on the basis of the EFSA opinions, the Court stressed the importance of the requirement to act consistently and to provide reasons on the part of the Commission. However, the legal nature and significance of these judgments are rather different. In the Herculex case, the Court ruled on an action brought by Pioneer regarding the Commission’s alleged failure to act.87 While the applicable comitology rules at the time required the Commission to submit its draft decision on Herculex to the Council ‘without delay’,88 this did not occur. It was not disputed among the parties that the Commission exceeded the time period for forwarding its draft decision to the Council as laid down in the relevant rules. The question was whether this delay could be justified. What is interesting about this case is the Commission’s defence strategy before the Court. It framed its role in the authorisation procedure heavily within the deliberative model, asserting that when deciding on when, and if at all, to proceed with the Council stage of the comitology procedure it must be granted a wide discretion to consider a variety of factors. Making a reference to the Pharos judgment of the Court of Justice,89 the Commission argued that ‘it must have sufficient time to consider the possible courses of action in the light of the complexity of the problem and the need to ensure that its proposal was not rejected by the Council’.90 Moreover, the Commission invoked the need at the time to consider the political circumstances of the ongoing legislative reform on GMO cultivation. The Court, however, rejected these arguments. It ruled against the Commission rebuking both the inconsistency91 and insufficiency of reason-giving by the Commission as follows: [T]he Commission had not raised any issue of scientific uncertainty that has been reported to it concerning the placing on the market of maize 1507. It merely stated that the period

87

Case T-164/10 Pioneer Hi-Bred International v European Commission ECLI:EU:T:2013:503. Art 18(1) of Dir 2001/18 together with art 5(2) of Council Decision laying down the procedures for the exercise of implementing powers conferred on the Commission (1999/468/EC). 89 C-151/98 P Pharos v Commission ECLI:EU:C:1999:563, in which the Court ruled that when confronted with highly sensitive and complex matters, the Commission had the right to request a new scientific opinion. 90 Case T-164/10, above n 87 at para 50. 91 The Court pointed out that on 2 March 2010 the Commission authorised another controversial GMO, namely the Amflora potato. 88

186

Maria Weimer and Gaia Pisani

… was necessary in order to examine the various possible courses of action, but without explaining what the various available options were.92

The Court’s language reflects an understanding of administrative legitimation along the lines of the deliberative model in two ways. First, the phrase ‘any issue of scientific uncertainty that has been reported to it’ appears to recognise the scientific complexity of risk assessment entailing that not only EFSA opinions, but also uncertainty information coming from other sources, could have been accepted as justification for delaying the procedure. Second, the Court seems to also reject the Commission’s arguments on the ground that they were not sufficiently explained. The Commission ‘merely’ asserted that it needed to exercise its discretion ‘without explaining’ properly to the Court what the relevant considerations were in this regard.

A. Case T-240/10 Hungary v Commission (Amflora) The Court’s Amflora judgment, in contrast, contains a more far-reaching engagement with the Commission’s role as risk manager acting under the conditions of scientific uncertainty. Following the Commission authorisation of Amflora, Hungary brought an action before the General Court seeking the annulment of this decision.93 Hungary claimed that the Commission erred manifestly in its assessment, and infringed the precautionary principle; and that the authorisation was based on a risk assessment that was ‘deficient, inconsistent, and incomplete’, which amounted to a breach of the Commission’s obligations under Directive 2001/18,94 namely to follow the principles of environmental risk assessment laid down therein, and to take into particular consideration GMOs, such as Amflora, which contain antibiotic resistant marker genes.95 The plaintiff therefore asked the Court for a substantive review of the Commission’s decision including of the quality of the factual basis it applied.96

92

Case T-164/10 above n 87 at para 53, emphasis added. Case T-240/10 Hungary vs Commission ECLI:EU:T:2013:645. 94 Art 4(2) of Dir 2001/18 states that: ‘Any person shall, before submitting a notification under part B or part C, carry out an environmental risk assessment. The information which may be necessary to carry out the environmental risk assessment is laid down in Annex III. Member States and the Commission shall ensure that GMOs which contain genes expressing resistance to antibiotics in use for medical or veterinary treatment are taken into particular consideration when carrying out an environmental risk assessment, with a view to identifying and phasing out antibiotic resistance markers in GMOs which may have adverse effects on human health and the environment. This phasing out shall take place by the 31 December 2004 in the case of GMOs placed on the market according to part C and by 31 December 2008 in the case of GMOs authorised under part B’. 95 Case T-240/10 above n 93 at paras 67–68. 96 Thereby the Court followed a recent trend in EU case law to intensify the standard of review of science-based decisions; see E Vos, ‘The European Court of Justice in the Face of Scientific Uncertainty and Complexity’ in M Dawson, B De Witte and E Muir (eds), Judicial Activism at the European Court of Justice (Cheltenham, Edward Elgar, 2013). 93

Expertise as Justification

187

The Court, however, did not rule on these claims, but chose a procedural detour around them instead. It raised, on its own motion, a new potential ground for annulment, namely the infringement of essential procedural requirements.97 The Court ruled that the Commission breached a procedural rule governing the adoption of the authorisation decision, namely the obligation to re-submit its draft decision on Amflora to the competent comitology committee following the 2009 consolidated EFSA opinion. As explained above, the Commission, before adopting the final decision, did not re-submit to the comitology committee the amended draft of this decision, together with the consolidated EFSA opinion and the dissenting opinion, which would have effectively restarted the comitology process. Instead, the Commission, following the Council’s failure to adopt a decision on the initial draft, went on to authorise the product. In its defence, the Commission argued that it was not necessary to resubmit the proposal, because its enacting terms were identical to the initial draft which had already been discussed by the committee. Moreover, the new proposal contained no changes in substance.98 The Court disagreed. It found that the new proposal was not identical to the version initially submitted to the committee, since the scientific basis which the Commission relied upon when adopting the former had changed. The Court considered the updated scientific basis of the proposal as part of the reasoning on which it was based, and therefore concluded: It must therefore be noted that the Commission, in deciding to seek a consolidated opinion from EFSA as result of the observations of an NGO and of the Danish Government, and in using that opinion in particular as a basis for the contested decisions without allowing the competent committees to take a position either on the opinion or on the statement of reasons on which the amended draft decisions were based, disregarded the regulatory procedure laid down in Article 5 of Decision 1999/468.99

From the point of view of administrative legitimation, two aspects of this judgment deserve further attention. First, the Court avoided reviewing the EFSA opinion as the basis of the Commission authorisation decision.100 It did not examine the quality of the scientific risk assessment nor did it scrutinise Hungary’s claims that the Commission had breached the principles of environmental risk assessment and the precautionary principle. In this judgment the Court was arguably too pragmatic and also too narrow in its assessment of the role of judicial review in risk regulation. It has been argued that a more proactive and less deferential role of the Court is desirable when reviewing the informational basis of science-based measures. The principle of the public administrator’s duty of care (ie the duty to

97

Case T-240/10 above n 93 at, para 70. ibid, para 74. ibid, para 83. 100 Which the Court could have done, as in other similar cases, by reviewing the Commission’s discretion in ascertaining the factual basis of the decision. 98 99

188

Maria Weimer and Gaia Pisani

examine carefully and impartially all the relevant aspects of the case) is particularly important in this regard. EU courts have applied this principle in a line of case law aiming to scrutinise the Commission’s interpretation of information of a complex economic or technical nature.101 More importantly, the duty of care is inherently linked to the Commission’s duty to pursue public interests enshrined in EU legislation when exercising its administrative discretion. In the words of Mendes, ‘It is this assessment of the circumstances in relation to the interpretation—and construction—of the public interests envisaged by law that ought to guide the substantive choices of the administrative decision-maker’.102 Judicial review of the Commission’s compliance with its duty of care, then, becomes an important mechanism in ensuring that public interest considerations are prioritised in administrative decisions. In the field of risk regulation, the Court could contribute to unveiling the value-laden choices behind scientific risk assessment including the uncertainties surrounding it by scrutinising more closely both the process and informational basis of scientific opinions (including eg consistency, independence and transparency of methodological choices).103 It seems that the Court in this case has missed an opportunity to fulfil its function as an ‘informational catalyst’.104 Second, the reasoning of the Court in this case reveals an understanding of administrative legitimation along the lines of the deliberative model. This is evident in its understanding of the nature of the EU comitology procedure as fulfilling an important function in securing the composite or cooperative nature of the EU public administration.105 Comitology has been described as providing a network structure, which transforms top-down supranationalism into a process of deliberative cooperation between national representatives and supranational institutions, and which marries techno-scientific and political rationalities in this process.106 Therefore, the Court’s procedural approach can be read as aiming to preserve the shared political responsibility of the participating institutions for administering GMOs, and to prevent the Commission from acting unilaterally.107 It is obvious from this judgment that given the complexity of the risk assessment of Amflora, expressed among various conflicting scientific opinions, that the Court refused to accept the Commission’s unilateral move to proceed to authorisation without seeking the opinion of national representatives in the comitology

101 For an overview of the case law see J Mendes, Discretion, Duty of Care and Public Interests in the EU Administration: the Limits of Law (unpublished manuscript at file with the authors, 2016). 102 ibid. 103 See Shapiro, Fisher and Wagner, above n 24; Scott and Sturm, above n 29. 104 Scott and Sturm, ibid. Note however that there is a fine line between a more intense procedural review of the scientific basis of administrative decisions and a more intrusive substantive review, in which the Court could substitute the scientific judgement by its own judgment; see for the latter, Case T-229/04 Sweden v Commission ECLI:EU:T:2007:217. 105 Schmidt-Assmann and Schöndorf-Haubold, above n 21; Hofmann and Turk, above n 26. 106 Joerges and Neyer, above n 27; Sabel and Zeitlin, above n 27. 107 See Weimer, above n 42; Weimer, above n 5.

Expertise as Justification

189

committee. This is especially evident in the Court’s finding that: the ‘great political sensitivity’ and the ‘complexity of the subject-matter’ of GMO authorisations support the argument ‘exactly in favour of the Commission’s obligation to submit the amended drafts of the authorisation decisions in relation to the Amflora potato to the competent regulatory committees, and where applicable, to the Council’.108 Furthermore, despite of its restraint in reviewing the Commission decision including its informational basis, the Court nevertheless managed to act as an accountability forum for the Commission’s actions in this case. It did so by weaving process values such as reason-giving and consistency on the part of the Commission into its review of procedural compliance. The Commission claimed that the scientific basis of its last amended draft (namely the 2009 EFSA consolidated opinion) did not change in comparison to the previous drafts, and that therefore there was no ‘substantive amendment’ that would have required going back to the committee. In response the Court dedicated a significant part of the judgment to examining the consistency of the Commission’s actions throughout the authorisation process, and effectively held the Commission to account for the exercise of its administrative discretion.109 In doing so it revealed an understanding of the role of the Commission in this process according to the deliberative model. The Court found three major differences between EFSA’s consolidated opinion of 2009 and EFSA’s earlier opinions, which also confirm our previous findings of the EFSA’s gradual shift towards more uncertainty tolerant risk assessments. Regrettably, the Court walked a tight rope between comparing the different EFSA opinions without actually evaluating their quality, explicitly insisting that it was not necessary to rule on whether the risk assessments were well-founded. The first difference, according to the Court, was that the 2009 opinion was authored not only by the GMO panel, but also by the BIOHAZ panel, and was drafted in collaboration with the EMA. The other two differences were identified as follows: Secondly, the findings of EFSA’s consolidated opinion of 2009, on which the amended proposals are based, place greater emphasis on scientific uncertainty (‘not fully understood’, ‘limitations’, ‘uncertainties’, ‘unlikely’) and dangers (‘cause for global concern’) than the findings of EFSA’s 2005 opinion (‘no reason to assume’, ‘would not pose any additional risk’, ‘no significant risk’, ‘no adverse environmental effects were observed or would be likely’) and the EFSA declaration of 2007 (‘will not be compromised’, ‘extremely low probability’, ‘very unlikely’, ‘does not pose a risk’), on which the earlier drafts are based. Thirdly, EFSA’s consolidated opinion of 2009 contains dissenting opinions from two members of the BIOHAZ panel emphasising scientific uncertainty, whereas EFSA’s 2005 opinion and the EFSA declaration of 2007 did not include any dissenting opinion.110

Moreover, the Court referred to the Commission’s written submissions made in the course of another judicial action against the Commission, namely the action

108

See above n 93, para 110. Case T-240/10 above n 93, paras 93–104. 110 ibid, para 98. 109

190

Maria Weimer and Gaia Pisani

for failure to act, which was filed by the applicant BASF in 2008 before the Commission adopted the Amflora authorisation.111 In that procedure the Commission defended the delay in the authorisation procedure as being caused by unclear, ambiguous and contradictory EFSA opinions, which justified the need to ask EFSA for a new consolidated opinion. It argued that given the scientific uncertainty underlying Amflora’s risk assessment it had exercised its discretion appropriately, and according to the precautionary principle decided to ask EFSA for another opinion. The Court summarised the Commission’s approach as follows: It is apparent from those assertions that the Commission, at the very least after receiving the letters from an NGO and the Danish Government, considered that the EFSA declaration of 2007, which was inconsistent with EFSA’s 2004 opinion read in conjunction with the EMA declaration of 2007, constituted a scientific basis too uncertain for the adoption of the proposed decisions already submitted to the regulatory committees and to the Council and that, having regard to the pervading scientific uncertainty, it was incumbent upon the Commission, pursuant to the precautionary principle, to consult EFSA once more in order to obtain clarification on the scientific assessment of the risks linked to the Amflora potato, in particular to the nptII gene.112

The Court was rigorous in its finding that it was contradictory of the Commission to argue that the EFSA consolidated opinion was merely a confirmation of previous EFSA opinions, and not significant enough to alter substantively the scientific basis on which the final Amflora authorisation was based. And, en passant, the Court also proved to be quite sensitive to the pervasive scientific complexity of the authorisation process. Read along the lines of the deliberative model of administrative legitimation, the Court’s reasoning indicates that the EU public administration, understood as a composite and cooperative institution, has a relatively wide substantive discretion when determining GMO authorisations. In exercising this discretion it must however apply the correct procedure, cooperate, avoid contradictions and provide for a sound and consistent reasoning of its actions. In contrast, nothing in this judgment points to an understanding of the role of the Commission, along the lines of the control model, as being obliged to follow the EFSA’s final recommendation, to disregard the uncertainties communicated by EFSA as irrelevant, and as not having the discretion to take into account dissenting opinions.

V. Conclusions This chapter has argued that key notions of EU risk regulation, such as science, risk, uncertainty, risk assessment and risk management are significantly shaped in 111 Case T-293/08 BASF vs Commission closed by order of the Court of 9 June 2010 due to lack of the need to adjudicate. 112 See above n 93 at para 103, emphasis added.

Expertise as Justification

191

the legal administrative discourse through which EU risk regulation takes place. It has analysed EU administrative and judicial decision making in the case of two highly controversial GMO authorisations tracing how notions of administrative legitimation (ie ideas of how the EU public administration is constituted, controlled and held to account) have co-produced notions of GMO risks as well as understandings of the role of scientific experts in administrative decision-making on risk. The findings indicate that different models of administrative legitimation, namely the control model and the deliberative model, are competing with each other in EU legal administrative practice. Moreover, different actors (EU agency, Commission and courts) have embraced different notions depending on the stage of the administrative process (risk assessment or risk management) and the level of contestation and pressure from other actors (eg national scientific authorities, NGOs or courts). In the GMO cases analysed, the EU expert agency EFSA initially understood its role mainly in terms of the control model, namely as being instrumental in enabling decision-making by the Commission by providing the latter with ascertained facts about quantifiable risks. It therefore framed its risk assessments in an uncertainty intolerant way, and tried to establish scientific authority vis-à-vis other scientific sources. However, over time EFSA has responded to pressure and contestation coming from national authorities and NGOs by gradually moving towards the deliberative model. This resulted in more nuanced risk assessment communicating scientific uncertainty more openly, and in one case including a dissenting opinion. The Commission, in contrast, has shifted in the opposite direction. During the phase of risk assessment (which took many years in both cases investigated) the Commission assumed the role of facilitating deliberation among different actors and using its discretion to delay the process in order to enable all actors to voice their concerns. Interestingly, it interpreted the precautionary principle as justifying delay. In contrast, the closer the Commission approached the final decision, the more it followed the control model. The Commission attributed a major role to EFSA in determining the outcome of the authorisation procedure. EFSA’s majority opinion was seen as a constraint on the Commission’s exercise of discretion not leaving space to consider instances of scientific uncertainty or dissenting opinion. Moreover, when adopting the final decision the Commission did not see itself as entitled to invoke the precautionary principle. Thus, the final decision was justified in mainly scientific terms (of the majority opinion), which barely left any space for the exercise of policy judgement on the part of the Commission. And for that purpose, scientific expertise was construed as being able to provide sufficient facts and certainty about GMO risks. This shows that the Commission regarded scientific expertise as a surrogate for legislative control in the control model. Given the wide framework nature of EU legislative provisions, the fiction of an objective and prescriptive law in the control model has been replaced with the fiction of an objective expert that is able to tame administrative discretion.

192

Maria Weimer and Gaia Pisani

Interestingly, the Commission’s preference for the control model cannot be explained by judicial approaches to administrative legitimation. The rulings of the EU General Court analysed in this chapter reveal the continuing relevance of Pfizer and the deliberative model expressed therein. In other words, the Court grants the Commission a wide discretion in ascertaining the factual basis for risk decisions while imposing procedural requirements of reason giving, consistency and compliance with all legally prescribed procedures. While shying away from directly reviewing the quality of EFSA’s risk assessments, in the Amflora case the Court was nonetheless able, along the lines of the deliberative model, to hold the Commission to account by emphasising the cooperative nature of the EU public administration as well as process values such as reason-giving and consistency. This chapter, therefore, reveals an interesting discrepancy between the Commission’s and the Court’s approach to administrative legitimation, which should be subject to further research. From the perspective of the EU courts, the Commission’s exercise of discretion is legitimated in terms of the deliberative model: as risk manager the latter has the responsibility, in a scientifically informed way, to exercise a political judgement. The Commission, however, according to the control model, sees its discretionary space radically reduced to following the lead of its expert agency. In a highly politicised and contested field of EU risk regulation the Commission sees applying scientific expertise as the main, if not the only way to justify its decisions. It does, in other words, not exercise political judgement when it comes to GMO authorisations other than to delay the process, and not to decide on authorisations. It follows that in the Pfizer case, the Court’s emphasis of political legitimacy together with the recognition of the limits of scientific legitimacy is not meaningful in practice. To put it provocatively, the Court constructs political legitimacy of the EU administration without questioning the reality of that legitimacy.113 This raises questions with regard to the viability of the deliberative model, and indicates that the search for an adequate model of administrative legitimation of the EU public administration is far from over.

113

See also Anderson, above n 7.

10 Evolving Conceptions of Science and Legitimacy: Insights from American Administrative Law CHRISTOPHER ANDERSON

I. Introduction In the classical account, American administrative law is premised on a Weberian model of administration in which ‘power is exercised on the basis of knowledge’.1 In this telling, administration furthers the public good by bringing expertise to bear on the resolution of public policy problems, and it justifies the exercise of executive power by providing a politically neutral basis for the formulation of regulatory policy.2 In its strongest forms, expertise is portrayed as providing objective answers to at least some regulatory problems, so that the need for political resolution is effectively eliminated.3 Despite frequent acknowledgment of its limitations, the classical account has dominated American administrative law scholarship and has become the standard model for explaining the role of expertise in bureaucratic regulation generally and risk regulation in particular. In this chapter, I provide a different account of the relationship between expertise and executive power in American administrative law, which I will argue has animated the case law on risk regulation for some time. For ease of reference, I will call my version the contemporary account. My analysis begins by recognising that the classical account has long been held in disrepute. Since the 1970s, courts and 1 That phrasing is Mashaw’s. JL Mashaw, ‘Small Things Like Reasons Are Put in a Jar: Reason and Legitimacy in the Administrative State’ (2001) 70 Fordham Law Review 17, 23. Others who have described what I am calling the classical account include RB Stewart, ‘The Reformation of American Administrative Law’ (1975) 88 Harvard Law Review 1667, 1671–81 and E Kagan, ‘Presidential Administration’ (2001) 114 Harvard Law Review 2245, 2260–64. 2 Mashaw, ibid at 26: ‘The promise of the administrative state was to bring competence to politics. It is the institutional embodiment of the enlightenment project to substitute reason for the dark forces of culture, tradition, and myth’. 3 GE Frug, ‘The Ideology of Bureaucracy in American Law’ (1984) 97 Harvard Law Review 1276, 1318–21.

194

Christopher Anderson

scholars have recognised that science, and in particular risk science, is virtually always tentative, uncertain and ambiguous, and that as a consequence it cannot provide a politically neutral basis for the formulation of regulatory policy.4 Because of this recognition, courts have increasingly come to understand the essential role of judgement in drawing scientific conclusions.5 The ambiguous concept of judgement is crucial to the contemporary account, in which it denotes a complex process of opinion formation that may incorporate a range of considerations drawn from scientific, public policy and legal discourses. Though individual judgements may take on more of an ‘expert’ or ‘policy’ character, they can never be fully analysed into their constituent components.6 By acknowledging the essential role of judgement in regulatory decision making, the contemporary account in effect recognises that administrative agencies’ scientific conclusions and policy decisions are co-produced.7 The challenge for the contemporary account has been to construct a theory of administrative legitimacy that does not depend on the neutrality of bureaucratic expertise.8 The prevailing judicial solution to this problem has been to locate the foundations of administrative legitimacy jointly in agencies’ technical expertise and in a presumed democratic mandate derived from oversight by the elected president. In the contemporary account, administrative risk regulation is legitimate not because it is the product of neutral expertise, but because it reflects the expert judgement of a democratically legitimate policymaker. Just as the contemporary account incorporates a more complex picture of the relationship between science and politics, it also posits a complex relationship between expert and executive authority. In the classical account, scientific expertise and executive authority are related in a linear fashion. The contemporary account, by contrast, presents the relationship between expertise and authority in dialectical terms: expertise supports the legitimacy of administrative regulation by providing a normative framework for the exercise of bureaucratic power in which the tools of science are used to pursue good public policy. At the same

4 Some early cases that explicitly discuss the need for judgement in interpreting risk science include: Industrial Union Department, AFL-CIO v Hodgson, 499 F2d 467, 476 (DC Cir 1974), Synthetic Organic Chemical Manufacturers Association v Brennan, 503 F2d 1155, 1158–59 (3d Cir 1974), and Society of the Plastics Industry, Inc v OSHA, 509 F2d 1301, 1308 (2d Cir 1975). Similar observation were made by scholars, eg, TO McGarity, ‘Substantive and Procedural Discretion in Administrative Resolution of Science Policy Questions: Regulating Carcinogens in EPA and OSHA’ (1979) 67 Georgetown Law Journal 729; Stewart, above n 1 at 1681–88. 5 Perhaps the most detailed statement of this position is the DC Circuit’s monumental en banc opinion in Ethyl Corp v EPA, 541 F2d 1, 20–29 (DC Cir 1976) (en banc): ‘how can the Administrator determine that a risk is a significant risk if he cannot assess risks? And how can he assess risks if he cannot make policy judgments? Surely reliance on “facts” as contemplated by petitioners will provide little guidance’. 6 Consider in this regard the courts’ references to the concept of ‘expert policy judgment’, eg, National Cable and Telecommunications Association v Brand X Internet Services, 545 US 967, 1003 (2005); American Coke and Coal Chemicals Institute v EPA, 452 F3d 930, 945 (DC Cir 2006); Public Citizen Health Research Group v Food and Drug Administration, 740 F2d 21, 28 (DC Cir 1984). 7 M Weimer and A De Ruijter, ch 1 in this volume. 8 cf above n 3 at 1331–34.

Insights from American Administrative Law

195

time, however, the authoritativeness of agencies’ scientific conclusions is grounded not only in their expertise, but also in the administration’s democratic legitimacy. Recognising that scientific conclusions rely on judgement and that many conclusions will be contested, the courts have held that agencies’ scientific conclusions are authoritative not because they are the work of experts, but because administrative agencies are democratically legitimate institutions for making such judgements. In this way, agencies’ democratic credentials underwrite the authoritativeness of their scientific conclusions.9

II. Early Case Law on Risk Regulation and the Breakdown of the Classical Account During the 1970s, the courts of appeals began to grapple with the problems of scientific expertise in administrative decision making in a sustained way. These cases concerned the legality of regulations promulgated by administrative agencies to implement several new statutes designed to address health and environmental risk. The circumstances of these cases, particularly the need to address regulatory problems at the outer bounds of scientific understanding, laid bare the deficiencies of the classical account. Courts were faced with the problem of how to interpret statutes that on their face required science-based decision making in circumstances in which scientific answers seemed to be woefully lacking.10 They also had to confront the problem of their own role in the regulatory process, including their role as guardians of the constitutional order. Broadly speaking, the lower courts took two divergent approaches. The first approach, championed by the Fifth Circuit Court of Appeals, tracked the classical account. It treated the scientific basis of risk regulation as an objective matter of fact for expert resolution. As a consequence, agencies bore a burden of proof to adduce sufficient scientific evidence to demonstrate that a product or process had a significant propensity to cause harm.11 If agencies could not meet

9

cf S Jasanoff, ‘Science, Politics, and the Renegotiation of Expertise at EPA’ (1992) 7 Osiris 194, 197. above n 4 at 749–53; M Shapiro, ‘The Frontiers of Science Doctrine: American Experiences with the Judicial Control of Science-Based Decision-Making’ in C Joerges, K-H Ladeur and E Vos (eds), Integrating Scientific Expertise into Regulatory Decision-Making: National Traditions and European Innovations (Baden-Baden, Nomos, 1997) 328–29. 11 Aqua Slide ‘N’ Dive Corp v Consumer Products Safety Commission, 569 F2d 831, 838 (5th Cir 1978). What this burden required is not clear. The Fifth Circuit apparently did not believe that the administration had to demonstrate that it was more likely than not that specified harm would occur, but it did have to adduce enough evidence to indicate that harm was more than a mere possibility. Because the Fifth Circuit’s approach fell out of favour, the exact nature of the administration’s burden was never resolved. See also Dry Color Manufacturers’ Association v Department of Labor, 486 F2d 98, 105–06 (3d Cir 1973), discussing OSHA’s burden of proof. 10 McGarity,

196

Christopher Anderson

that burden, they could not regulate.12 Because the assessment of risk was viewed as an objective matter, the Fifth Circuit’s approach denied that there was much room for agencies to exercise policy judgement when interpreting the significance of scientific evidence. This highly instrumental model of administrative risk regulation was rooted as much in theories of bureaucratic legitimacy as it was in understandings of science. Administrative regulation was legitimate because all important policy questions were decided by Congress in the applicable statute.13 The agency’s job was merely to determine the facts of particular situations and to implement Congress’s instructions. The courts’ duty was to ensure that the agency did not stray beyond its statutory power by verifying that the agency’s evidence adequately supported its regulatory action.14 Competing with the Fifth Circuit’s approach was a second model of administrative risk regulation developed principally by the DC Circuit. That model was much more realistic about the nature and limits of risk science. Foremost, it recognised that scientific evidence is virtually always uncertain and open to interpretation.15 It also explicitly recognised that waiting for a high degree of scientific confidence before taking protective measures could put humans and the environment at risk, thereby undermining the goals of the new risk regulation statutes.16 Given these realities, the DC Circuit understood risk regulation to call for ‘an essentially legislative policy judgment’ and adapted its approach to review accordingly.17 Although the court continued to stress the importance of science for administrative decision making, it treated the interpretation of scientific e vidence as a matter for the agency’s policy judgement.18 As such, administrative risk regulation could not be reviewed as a question of fact and subject to a burden of proof analysis. Instead, the court had to assess the consistency of the agency’s decision with relevant legislation and its reasonableness in light of the available evidence.19 The DC Circuit’s approach thus allowed an important constitutive role for the 12 American Petroleum Institute v OSHA, 581 F2d 493, 502–03 (5th Cir 1979). The Fifth Circuit’s approach had strong resonances with tort law approaches to dangerous products, eg, Aqua Slide ‘N’ Dive, above n 11 at 839–40. 13 Aqua Slide ‘N’ Dive, above n 11 at 839–40; American Petroleum Institute, above n 12 at 504–05. 14 Aqua Slide ‘N’ Dive, above n 11 at 838; American Petroleum Institute, above n 12 at 505. 15 Hodgson, above n 4 at 475. As the Second Circuit, which also tended to follow the DC Circuit’s approach, famously put it, ‘though the factual finger points, it does not conclude’. Society of the Plastics Industry, above n 4 at 1308. 16 Above n 5 at 25. 17 Hodgson, above n 4 at 475. 18 The canonical statement is Ethyl, above n 5 at 27: ‘Decision making must in that circumstance depend to a greater extent upon policy judgments and less upon purely factual analysis’. 19 ibid 29. Because it continued to recognise the importance of scientific expertise for high-quality decision making, the DC Circuit also considered the adequacy of the agency’s scientific investigation. In some cases, that court set aside administrative decisions on the grounds of inadequate evidentiary support because the court believed that the agency had failed to obtain readily available evidence, eg, Leather Industries v EPA, 40 F3d 392, 408 (DC Cir 1994); Natural Resources Defense Council, Inc v Herrington, 768 F2d 1355, 1391 (DC Cir 1985). These decisions differ from the Fifth Circuit approach in that they do not focus on a burden of proof analysis but on consideration of whether the agency had discharged its responsibility to examine reasonably available evidence.

Insights from American Administrative Law

197

agency in formulating social responses to uncertain technological risks. While the general policy direction was set by Congress in legislation, it was up to the agency to make the numerous judgement calls necessary to elaborate that policy in regulatory practice. One result of this approach was a more deferential form of review, as the American courts have historically been reluctant to override administrative judgements on matters of policy.20 The differences between the Fifth Circuit and DC Circuit approaches neatly illustrate the classical account and the challenges posed to it by modern risk regulation programmes. The Fifth Circuit’s approach relied heavily on the expertise of the agency to justify the exercise of regulatory authority. So much so, in fact, that the agency’s regulatory power could be seen as coterminous with its expertise: the agency could only regulate to the extent it could establish the nature and extent of risk as a factual matter.21 The DC Circuit’s approach recognised the inadequacy of the classical account for dealing with the problems of risk regulation. It appreciated that administrative decision making could not be captured in the adjudicative metaphor of burdens of proof, but instead required the weighing of competing societal priorities in light of the agency’s regulatory mandate. To allow the administration to deal with those issues effectively, courts following the DC Circuit’s approach were prepared to accord the administration a substantial margin of discretion in setting risk policy. The problem, however, was that the thendominant theories of administrative legitimacy were insufficient to legitimate that policy making authority. If scientific expertise was necessarily an insufficient basis for risk regulation, then the classical account was also insufficient to justify vesting regulatory power in administrative agencies. Some courts attempted to fill the gap by crafting a ‘surrogate political process’, but that effort was short-lived (and largely unsatisfactory).22 Others courts simply skirted the issue.23 The lack of a cogent theory to legitimate vesting substantial policy discretion in administrative agencies left the DC Circuit’s approach constitutionally vulnerable. Ultimately, it would be up to the Supreme Court to develop a coherent theory of administrative risk regulation.

III. Early Supreme Court Cases on Risk Regulation Risk regulation first reached the Supreme Court in the Benzene case,24 which was an appeal from a decision of the Fifth Circuit. Although the Benzene decision was 20

Bowman Transportation, Inc v Arkansas-Best Freight System, Inc, 419 US 281, 297–98 (1974). American Petroleum Institute, above n 12 at 504: ‘Until OSHA can provide substantial evidence that the benefits to be achieved … bear a reasonable relationship to the costs imposed … it cannot show that the standard is reasonably necessary to provide safe or healthful workplaces’. 22 Stewart, above n 1 at 1760–89. 23 eg, above n 5 at 31. 24 Industrial Union Dep’t, AFL-CIO v American Petroleum Institute, 448 US 607 (1980) (Benzene). 21

198

Christopher Anderson

much-anticipated, the court failed to reach a majority in support of any opinion, depriving the judgment of precedential authority.25 As Justice Scalia, then a professor at the University of Chicago, put it, the Benzene decision ‘literally provides no conclusive answer to any legal question more general than whether the benzene exposure regulation promulgated by the Occupational Safety and Health Administration (OSHA) on February 10, 1978, is valid’.26 Despite that important limitation, the opinions in the Benzene case provide useful insights regarding how the justices understood the role of science in risk regulation, as well as their concerns about the constitutional limits of administrative authority. Benzene considered the lawfulness of OSHA’s new permissible exposure limit (PEL) for benzene in the workplace. The agency had lowered the main limit for airborne exposure from 10 ppm to 1 ppm on the basis of studies indicating that benzene could cause leukaemia via the inhalation pathway.27 Although the scientific evidence was strong that benzene caused leukaemia at high levels of exposure, there was little direct evidence of effects at lower levels. Given that uncertainty, the agency did not attempt a quantitative assessment of leukaemia risk.28 Instead, OSHA relied on its Cancer Policy, under which the agency asserted that the Occupational Safety and Health Act (OSH Act) required it to set PELs for carcinogens at the most stringent level technically and economically feasible because no level of exposure to a carcinogenic substance could be considered safe.29 After surveying conditions in the relevant industries and the available control measures, OSHA determined that 1 ppm was the lowest feasible level.30 Although the Benzene case was nominally about the benzene PEL, the main concern of Justice Stevens’s plurality opinion was the Cancer Policy.31 The plurality was transparent in its uneasiness with the scope of authority that the C ancer Policy seemed to claim for the agency and with the regulatory costs that it potentially entailed.32 Its solution was to cabin the agency’s power. Relying on a convoluted construction of statutory language, Justice Stevens held that the OSH Act only authorised the agency to regulate risks that rendered a workplace ‘unsafe’, which he equated with ‘significant risk’.33 The agency was therefore required to make a finding that a substance posed a significant risk before regulating; such

25

Marks v United States, 430 US 188, 193 (1977). A Scalia, ‘A Note on the Benzene Case’ (1980) 4 Regulation 25, 25. 27 Above n 24 at 619–21. 28 ibid 632–33. 29 ibid 624–25. 30 ibid 626. 31 Justice Stevens’s opinion was joined in full by Chief Justice Burger and Justice Stewart and by Justice Powell, except as to part III-D. Justice Powell would have held that the OSH Act required OSHA to show that the benefits of regulation justify the regulation’s costs. ibid 670–71 (Powell J concurring in part). 32 ibid 639–40, 651 (plurality opinion). This uneasiness was even more pronounced in Justice Powell’s and Justice Rehnquist’s concurrences. ibid 669 (Powell J concurring in part); ibid 673 (Rehnquist J concurring in the judgment). 33 Above n 24 at 642. 26

Insights from American Administrative Law

199

risk could not be presumed from the presence of a hazardous property as under the Cancer Policy. Because OSHA had not made the necessary finding, its rule was invalid.34 The plurality never resolved the question whether there was adequate evidence in the record to support a finding of significant risk. Indeed, it rather pointedly noted that it was declining to do so.35 Nor, despite certain illadvised dictum, did the court hold that OSHA could only regulate on the basis of quantitative risk assessment.36 To the contrary, both the plurality and the d issent explicitly recognised that the ‘significance’ of a risk could not be determined mathematically.37 Although it did not definitively determine what reasons OSHA would need to give to support a finding of substantial risk, the plurality did offer some clues. Unfortunately, those clues pointed in different directions. On one hand, the plurality held that the agency bore a burden of proof to show that ‘it is at least more likely than not’ that a substance poses a significant risk.38 It was quick to add, however, that the agency was ‘not required to support its finding that a significant risk exists with anything approaching scientific certainty’.39 Nonetheless, by characterising the agency’s task as meeting a burden of proof, the plurality seemed to conceive of the question of significance as a question of fact, echoing the Fifth Circuit’s approach. On the other hand, the plurality clearly did not understand the question of significance in purely scientific terms. Drawing heavily on precedents from the DC and Second Circuits,40 the plurality held that ‘while the Agency must support its finding that a certain level of risk exists by substantial evidence, we recognize that its determination that a particular level of risk is “significant” will be based largely on policy considerations’.41 Importantly, the plurality noted its agreement on this point with Justice Marshall’s dissent for four justices.42 Thus, at least eight justices in the Benzene case agreed that a finding of 34

ibid 653. ibid 659. 36 ibid 655, discussing in dictum what evidence might show a risk to be significant; see also, ibid 666–67 (Powell J concurring in part), interpreting the plurality opinion not to require quantification. Puzzlingly, several commentators have nonetheless claimed that Benzene does require quantitative risk assessment, at least as a practical matter, eg, L Bergkamp, ‘The Quiet Revolution in Administrative Procedure: Judicial Vetting of Precautionary Risk Assessment’ [2014] European Journal of Risk Regulation 102; G Charnley and ED Elliott, ‘Risk Versus Precaution: Environmental Law and Public Health Protection’ (2002) 32 Environmental Law Reporter 10363; H Latin, ‘Good Science, Bad Regulation, and Toxic Risk Assessment’ (1988) 5 Yale Journal on Regulation 89; P Van Zwanenberg and E Millstone, BSE: Risk, Science and Governance (Oxford, Oxford University Press, 2005) 25. That interpretation has been consistently rejected by the courts of appeals, however, eg, National Maritime Safety Association v OSHA, 649 F3d 743, 751 (DC Cir 2011); Miami-Dade County v EPA, 529 F3d 1049, 1069 (11th Cir 2008); Cactus Corner, LLC v Department of Agriculture, 450 F3d 428, 433 (9th Cir 2008); American Dental Association v Martin, 984 F2d 823, 827 (7th Cir 1993) (Posner J). 37 Above n 24 at 655. 38 ibid 653. 39 ibid 656. 40 ibid 656, citing Hodgson, above n 4, and Society of the Plastics Industry, above n 15. Notably, these were the only precedents the plurality relied on in this section of the opinion. 41 ibid 655 n 62, quoting ibid 706 (Marshall J dissenting). 42 ibid. 35

200

Christopher Anderson

significant risk would have to be based on a combination of empirical and policy considerations. The messy set of opinions43 in Benzene is best interpreted in light of the breakdown of the classical account. On one hand, every opinion in the case recognised in some way that science and expertise were insufficient bases for making difficult decisions about risk regulation. The plurality and the dissent went further and recognised that drawing conclusions from the scientific data depended on the exercise of expert and political judgement and that, accordingly, the science was inherently contestable.44 The issue that divided the justices was what the fluidity of risk science meant for the constitutional legitimacy of administrative risk regulation. Justice Rehnquist, writing alone, would have held that delegating decisions of such policy consequence to administrative agencies was flatly unconstitutional.45 The plurality was not prepared to go so far, but instead attempted to ameliorate the legitimacy problem by construing the statutory language in a way that would guide the agency in exercising its discretion and, presumably, narrow the scope for administrative policy making. What divided the justices, in other words, was not so much a dispute about the nature of risk science but about the constitutionally legitimate scope for administrative policy making on risk. Because of its opacity, Benzene did little to clarify how agencies should review administrative risk regulation. The Supreme Court next returned to the problem in Baltimore Gas and Electric Co v Natural Resources Defense Council, Inc,46 which concerned a Nuclear Regulatory Commission (NRC) rule providing that, for purposes of considering licensing applications, long term storage of nuclear waste would be deemed to have no environmental impact. Baltimore Gas was the second time that rule had been reviewed by the court. The first was Vermont Yankee,47 in which the court famously declared that courts could not impose procedural requirements on agencies beyond those specified in legislation. Following remand, the DC Circuit had once again vacated the rule, but this time on the basis that it was substantively irrational.48 As in Vermont Yankee, the Supreme Court unanimously reversed. Writing for the court, Justice O’Connor held that a reviewing court must remember that the Commission is making predictions, within its area of special expertise, at the frontiers of science. When examining this kind of scientific determination, as opposed to simple findings of fact, a reviewing court must generally be at its most deferential.49

43

A total of five opinions were filed in the case. Chief Justice Burger also underscored this point. ibid 663 (Burger CJ concurring). 45 ibid 686–87 (Rehnquist J concurring in the judgment). 46 Baltimore Gas and Electric Co v Natural Resources Defense Council, Inc, 462 US 87 (1983). 47 Vermont Yankee Nuclear Power Corp v Natural Resources Defense Council, Inc, 435 US 519 (1978). 48 Natural Resources Defense Council, Inc v Nuclear Regulatory Commission, 685 F2d 459 (DC Cir 1982). 49 Above n 46 at 103. 44

Insights from American Administrative Law

201

This holding came to be known as the ‘frontiers of science doctrine’.50 How to interpret that doctrine, however, continues to be a matter of dispute. Some take it to mean that when agencies are regulating on the basis of highly technical scientific conclusions, courts must be especially hesitant to second-guess those conclusions. Put differently, it has been argued that Baltimore Gas stands for the proposition that agency rules are entitled to greater deference, even ‘super-deference’, when they are based on sophisticated science.51 This interpretation of the ‘frontiers of science’ doctrine follows the classical account in that it identifies scientific conclusions with ostensibly neutral expertise and accords deference to agencies on the basis of their technical competence. As a consequence, proponents of the superdeference interpretation sometimes argue that Baltimore Gas creates incentives for agencies to couch the justification for their rules in complex scientific terms, even when decisions are actually being made for other reasons.52 Careful analysis of Justice O’Connor’s opinion, however, suggests that the super-deference interpretation is incorrect or at least incomplete. For one thing, technical issues were not at the heart of the dispute in the case: all parties agreed that, in theory, nuclear waste could be stored safely.53 The parties also agreed that there were deep and irreducible uncertainties as to whether safe storage would be accomplished in practice.54 What the parties disagreed about were the implications of those uncertainties for regulatory policy. The frontiers of science doctrine should thus be understood as addressing the situation, rife in risk regulation, in which regulators must respond to ambiguous and uncertain evidence.55 Although it required the NRC to bring its technical expertise to bear on the issue, the Supreme Court openly acknowledged that the final decision depended on policy considerations. Indeed, what set issues ‘on the frontiers of scientific knowledge’ apart from ‘simple finding[s] of fact’ was that the issue could not be resolved by scientific methods alone, but instead required the agency to exercise its ‘policy judgment’.56 Deference was owed at least as much because of the policy-laden nature of the agency’s conclusions as because of their technical complexity.57 By focusing on the importance of policy considerations in the Commission’s rule,

50 Shapiro, above n 10 at 331. The ‘frontiers of science’ idiom originated in the DC Circuit’s opinion in Hodgson, above n 4 at 474. 51 EH Meazell, ‘Super Deference, the Science Obsession, and Judicial Review as Translation of Agency Science’ (2011) 109 Michigan Law Review 733, 736–37 (collecting sources). 52 ibid 752; WE Wagner, ‘The Science Charade in Toxic Risk Regulation’ (1995) 95 Columbia Law Review 1613. 53 Above n 46 at 98, noting that ‘no party seriously challenges’ the Commission’s determination ‘that the probabilities favor the zero-release assumption’. 54 ibid. 55 ibid 105; Shapiro, above n 10 at 332. 56 Above n 46 at 105, citing Bowman, above n 20. 57 Above n 46 at 105–06; see also Motor Vehicle Manufacturers Association v State Farm Mutual Automobile Insurance Co, 463 US 29, 52 (1983).

202

Christopher Anderson

the court in effect held that the agency’s exercise of its expert judgement and its policy judgement were inseparable.58 The unanimous opinion in Baltimore Gas, along with the plurality and dissenting opinions in Benzene,59 shows a strong majority of the Supreme Court endorsing a view of risk science similar to that advanced by the DC Circuit.60 On this view, regulatory conclusions about risk cannot be viewed as matters of either science or policy, but must instead be understood as complex judgements grounded in multiple sorts of reasons. As the result in Benzene made clear, however, the court was not entirely comfortable with the implications of that understanding for the scope of the administration’s policy making power. What continued to be lacking was a theory of how such policy making authority could be reconciled with democratic values and the constitutional separation of powers. The debate over those issues would be reshaped by two landmark decisions that laid the groundwork for a new theory of the constitutional legitimacy of administrative regulation.

IV. State Farm and Chevron The Supreme Court’s struggles with risk science were part of a much larger upheaval in American administrative law in the late 1970s and early 1980s brought about by the break-down of the classical account. Since the early 1970s, the lower courts had been struggling to develop a new theory of administrative legitimacy that would account for the increasingly apparent role of politics in administrative regulation. Those struggles were not limited to risk regulation, but extended to administrative law generally, including fields such as telecommunications, utilities regulation and immigration. Of particular importance, the newly elected Reagan administration had embarked on an explicit programme of deregulation that laid bare the political origins of regulatory decision making and called into question many administrative law orthodoxies.61 In a pair of cases, the Supreme Court responded to these challenges by reformulating standards for judicial review of agency action. In doing so, the court developed a revised model of administrative legitimacy that departed in important ways from the classical account. State Farm, which was decided just 18 days after Baltimore Gas, addressed the role of courts in reviewing the substance of administrative rules. The case concerned

58 Shapiro, above n 10 at 334; see also Natural Resources Defense Council, Inc v EPA, 902 F2d 962, 973–74 (DC Cir 1990). 59 Only a year after the Benzene decision Justice Stevens, the author of the plurality opinion, joined the Benzene dissenters to form a majority in American Textile Workers v Donovan, 452 US 490 (1981). Although that case focused on issues of economic feasibility rather than risks to health, the majority largely followed the analytical approach of Justice Marshall’s Benzene dissent. 60 In particular, as elaborated in the DC Circuit’s opinion in Hodgson, above n 4. 61 MB Garland, ‘Deregulation and Judicial Review’ (1985) 98 Harvard Law Review 505, 508–12.

Insights from American Administrative Law

203

a decision by the National Highway Traffic Safety Administration (NHTSA) to rescind a safety standard requiring passive restraint systems in new automobiles.62 Affirming the Courts of Appeals, the Supreme Court vacated the rescission on the ground that it was arbitrary and capricious in light of the evidence before the agency. The court observed that in an earlier rulemaking NHTSA had found both that passive restraints would save lives and that two passive restraint technologies, automatic seatbelts and airbags, were feasible and effective. In rescinding the standard, the agency had revised its earlier opinion and found that automatic seat belts would not in fact be effective.63 The court unanimously held that the revised finding was insufficient to support rescission of the standard because the agency had not disturbed its determination regarding airbags.64 On the question whether NHTSA’s revised finding regarding automatic seatbelts was itself supportable, the court divided five-four against the agency.65 State Farm reaffirmed the centrality of technocratic rationality to lawful administrative decision making, and the majority opinion emphasised that it was imperative for the agency to exercise its ‘expert’ judgement.66 To that extent, the case reinforced the notion that it is the administration’s job to gather and analyse information and to make decisions on the basis of that analysis rather than on the basis of pure political preference.67 The opinion should not, however, be read as denying a legitimate role for administrative policy judgement. Although it focused on the importance of expertise, the majority opinion also acknowledged that often ‘the available data does not settle a regulatory issue and the agency must then exercise its judgment in moving from the facts and probabilities on the record to a policy conclusion’.68 It also held that an agency was entitled to change its position, provided that its revised view was supportable in light of the record. Justice Rehnquist’s partial dissent went further and recognised that an agency’s view of the evidence will be influenced by the political preferences of the incumbent president.69 In this way, Justice Rehnquist, apparently for the first time in the Supreme Court’s case law, identified the president’s electoral accountability as a basis for the legitimacy of administrative policy making.70

62 State Farm, above n 57 at 36–38. Passive restraint systems are ‘devices that do not depend for their effectiveness upon any action taken by the occupant except that necessary to operate the vehicle’. See also 34–35. 63 ibid 38–39. 64 ibid 47–48. 65 Compare ibid 51–55 (majority opinion) with ibid 58–59 (Rehnquist J concurring in part and dissenting in part). 66 ibid 48. This aspect of State Farm’s holding tends to be emphasised in the academic literature, eg, KA Watts, ‘Proposing a Place for Politics in Arbitrary and Capricious Review’ (2009) 119 Yale Law Journal 2, 5–6 and n 4 (collecting sources). 67 CR Sunstein, ‘Deregulation and the Hard-Look Doctrine’ [1983] Supreme Court Review 177, 210–12. 68 State Farm, above n 57 at 52. 69 ibid 59 (Rehnquist J concurring in part and dissenting in part). 70 ibid.

204

Christopher Anderson

The second, and arguably more important, case was the court’s unanimous decision in Chevron,71 which concerned the Environmental Protection Agency (EPA)’s interpretation of the term ‘stationary source’ in the Clean Air Act. EPA had interpreted the term to apply to an industrial installation as a whole, rather than to each emissions point within the installation. The courts of appeals below, in an opinion that relied on a traditional statutory construction analysis, vacated the rule on the ground that an interpretation that applied the term to each emission point would better effectuate the purpose of the statute.72 In reversing, the court announced its now-famous two-part test: when reviewing an agency’s interpretation of a statute, courts must first determine ‘whether Congress has directly spoken to the precise question at issue’.73 If so, the will of Congress prevails. If, however, the statute is ambiguous, the court must defer to the agency’s interpretation, so long as it is reasonable.74 Chevron dealt at some length with the basis for judicial deference to administrative decision making. At the outset, the court emphasised the agency’s expertise as the source of its superior competence for setting regulatory policy, particularly in technical areas.75 That justification for deference was consistent with the classical account’s understanding of administration as an alternative to politics for regulatory decision making. Later in the opinion, however, the court articulated a second justification for judicial deference grounded in the agency’s democratic mandate derived from its oversight by the president: [A]n agency to which Congress has delegated policy-making responsibilities may, within the limits of that delegation, properly rely upon the incumbent administration’s views of wise policy to inform its judgments. While agencies are not directly accountable to the people, the Chief Executive is, and it is entirely appropriate for this political branch of the Government to make such policy choices.76

That rationale picked up on Justice Rehnquist’s opinion in State Farm, but in this case it was embraced by a unanimous court. It was by far the Supreme Court’s strongest endorsement to that point of the legitimate role for political preferences in administrative regulation. Chevron was a powerful statement on the nature of the administration in two ways. First, it openly acknowledged the indeterminacy inherent in most regulatory programmes and statutes. In doing so, it undermined the notion, central to the classical account, that administrative rationality could be purely instrumental in nature. Second, the court made a self-conscious decision to place primary

71

Chevron USA Inc v Natural Resources Defense Council, Inc, 467 US 837 (1984). Natural Resources Defense Council, Inc v Gorsuch, 685 F2d 718 (DC Cir 1982) (RB Ginsburg J). Note that the court felt itself bound by circuit precedent on this point. ibid 720. 73 Above n 71 at 842. 74 ibid. 75 ibid 865. 76 ibid. 72

Insights from American Administrative Law

205

responsibility for the elaboration of regulatory policy in the administration itself, thereby affirming an important constitutive role for the administration.77 To be sure, the administration’s policy making role was a subordinate one; plenary power to issue regulatory commands remained the monopoly of Congress.78 But the size and complexity of the modern regulatory state ensures that the administration will have significant power to shape regulatory policy. Crucially, Chevron reconciled the exercise of this constitutive power with the tripartite constitutional structure by tying administrative authority to the president’s electoral mandate.79 Taken together, the frontiers of science doctrine, as elaborated in Baltimore Gas, and the dual rationales of State Farm and Chevron present a distinct picture of the role of the administration in regulating risk. State Farm requires that the agency undertake a thorough scientific investigation of the potential risk in q uestion.80 This requirement is consistent with the longstanding American view that administrative regulation is legitimate, in part, because it is capable of improving regulatory policy by applying expertise.81 Baltimore Gas recognises, however, that scientific evidence will often be inconclusive and that when science runs out, the agency must exercise its judgement to determine the significance of the evidence for regulatory policy. Courts owe such policy judgements no less—and perhaps more82—deference than the agency’s technical determinations. Finally, Chevron supplies the theoretical foundation for the legitimacy of administrative policy making by stressing the administration’s democratic legitimacy.83 In the model of administrative rationality described by these cases, scientific evaluation and policy judgement are essentially inseparable. What is more, both are vested in a single decision maker, often literally embodied in the agency head.

77 Although Chevron sanctioned conferral of constitutive interpretive power on the administration, it left open many questions regarding how that power should be exercised and what its limits might be. These questions are beyond the scope of the present chapter. Some recent cases that have addressed the limits of agencies’ policy making power include: Gonzalez v Oregon, 546 US 243, 262–64 (2006); Whitman v American Trucking Associations, 531 US 457, 474–76 (2001); FDA v Brown & Williamson Tobacco Corp, 529 US 120, 159–61 (2000). Justice Thomas has written most explicitly on this topic, arguing that Chevron’s approach to administrative policy making is unconstitutional in some circumstances, eg, Michigan v EPA, 135 S Ct 2699, 2712–14 (2015) (Thomas J concurring); Whitman, 531 US at 486–87 (Thomas J concurring). 78 Bowen v Georgetown University Hospital, 488 US 204, 208 (1988). 79 CR Farina, ‘The “Chief Executive” and the Quiet Constitutional Revolution’ (1997) 49 Administrative Law Review 179, 182–83; PL Strauss, ‘From Expertise to Politics: The Transformation of American Rulemaking’ (1996) 31 Wake Forest Law Review 745, 764–66; see also JL Mashaw, ‘Prodelegation: Why Administrators Should Make Political Decisions’ (1985) 1 Journal of Law, Economics & Organization 81, 95–96. 80 Accord Gas Appliance Manufacturers Association v Department of Energy, 998 F2d 1041, 1046 (DC Cir 1993). 81 Mashaw, above n 1 at 23; RL Rabin, ‘Federal Regulation in Historical Perspective’ (1986) 38 Stanford Law Review 1189, 1219–20, 1263. 82 See Competitive Enterprise Institute v National Highway Traffic Safety Administration, 956 F2d 321, 323–24 (DC Cir 1992). 83 Above n 71 at 866.

206

Christopher Anderson

Rather than making a choice between politics and expertise, the Supreme Court’s case law suggests that administrators should be understood as politically accountable experts and that it is this unique mix of qualities that makes administrative regulation legitimate. The role of the court is not to scrutinise either the technical or the policy substance of administrative decisions, except at the margins. Instead, the courts’ duty is to ensure that administrative actions fall within the scope of agencies’ lawful authority and that the agency has exercised is discretion on the basis of its consideration of the relevant evidence.

V. Risk Regulation in the Courts after Chevron The Supreme Court’s landmark decisions of the early 1980s did not change the lower courts’ approach to risk regulation overnight. Initially, there was considerable debate in the courts of appeals regarding the application of these cases, and it would be several years before they fully worked out the implications of the Supreme Court’s decisions. Eventually, however, two trends emerged in the case law. First, following the logic of Baltimore Gas, courts increasingly came to see questions of risk science and questions of regulatory policy as inextricably intertwined. As a result, courts increasingly expressed the view that the interpretation of scientific evidence was inevitably a matter of judgement that invoked the agency’s policy making authority, as well as its scientific expertise. The second trend was the forging of a link between agencies’ judgement on disputed scientific questions and the rationale of Chevron, which held that statutory ambiguity should normally be interpreted as a delegation of policy making authority to the administration. Following Chevron’s logic, some courts have treated statutory silence (hence, ambiguity) regarding scientific issues as an implicit delegation of authority to agencies to rely on policy considerations in their interpretation of scientific evidence. The significance of the courts’ reliance on Chevron is that it grounds deference to administrative science-policy judgements not only in the agencies’ technical expertise, but also in their putative democratic legitimacy. A prominent early case that touches on both these themes is the DC Circuit’s unanimous en banc judgment in Natural Resources Defense Council, Inc v EPA.84 The case is interesting because while the court recognised that regulatory decisions about risk ‘depend to a greater extent upon policy judgments’, it nonetheless interpreted the Clean Air Act85 to require the agency to exercise its expertise and ‘determine an acceptable risk to health’.86 The case underlines the tension at the

84

Natural Resources Defense Council, Inc v EPA, 84 F2d 1146 (DC Cir 1987) (en banc). Specifically, s 112 regarding controls on hazardous air pollutants. 42 USC § 7412. 86 Above n 84 at 1163. 85

Insights from American Administrative Law

207

heart of the Supreme Court’s case law: recognition that science does not provide a politically neutral basis for decision making coupled with a continued commitment to a Weberian conception of administration in which the application of scientific expertise is central to the legitimacy of administrative policy making. What the court of appeals essentially required EPA to do was to make a policy judgement, but to make that judgement on the basis of its best assessment of the available scientific evidence.87 The case is also important because it is one of the earliest examples of the court explicitly linking administrative evaluation of risk to the Chevron framework.88 The DC Circuit further developed the application of Chevron to judicial review of risk regulation in Edison Electric Institute v EPA.89 The issue in that case was whether EPA had permissibly adopted a ‘generic mismanagement scenario’ for determining whether wastes are hazardous, instead of assessing the management of each waste separately.90 The court held that statutory silence regarding risk assessment methodology should be interpreted as a delegation of policy making authority and that, accordingly, EPA’s choice of methodology was entitled to deference so long as its choice was reasonable.91 Because EPA’s preferred approach was consistent with the underlying goals of the statute, the court upheld the agency’s decision. The relationship between Chevron and agencies’ scientific judgement was raised again in Chemical Manufacturers Association v EPA.92 In that case, the court held that because the Clean Air Act was silent on the matter, Chevron deference extended to EPA’s development of an air dispersion model for estimating pollutant concentrations.93 In doing so, the court stressed that the choice of modelling parameters was not merely a question of scientific validity, but also a question of policy, and that such policy judgements could not be overcome with technical arguments.94 Both Edison Electric and Chemical Manufacturers are examples of the court reviewing issues that on their surface appear purely scientific—how to estimate health risk from hazardous waste disposal, how to model dispersal of an air pollutant—using the Chevron framework. In both cases, the court acknowledged that, although the agency’s analysis required the use of scientific expertise, its conclusions on these technical issues could not be separated from its policy judgements about how best to implement its regulatory programmes. The court then framed those judgements as matters of policy and relied on Chevron to hold that it must defer to such judgements.

87 EPA had attempted to avoid the scientific issue by substituting a feasibility analysis for risk assessment. The court ruled that approach unlawful. ibid 1164–65. 88 ibid. 89 Edison Electric Institute v EPA, 2 F3d 438 (DC Cir 1993). 90 ibid 443. 91 ibid 445. 92 Chemical Manufacturers Association v EPA, 28 F3d 1259 (DC Cir 1994). 93 ibid 1264, reaffirming that Chevron requires courts to defer to agencies’ risk assessment methodologies if the statute ‘can reasonably be read to authorize the agency’s choice’. 94 ibid 1264–65.

208

Christopher Anderson

Since the 1980s, cases acknowledging the importance of policy judgement in reaching scientific conclusions have become common, and examples can be found in virtually every circuit.95 In addition, courts have been increasingly ready to re-characterise challenges to agency science as challenges to administrative policy choices.96 The sheer ubiquity of these statements suggests that the courts have rejected the technocratic premises of the classical account, and that they have accepted that (in many circumstances) an agency’s scientific conclusions will also embody policy judgements about how scientific evidence should be interpreted. The result is that courts have been increasingly unwilling to second-guess agencies’ conclusions about risk on evidentiary grounds alone. To prevail, challengers must show either that the agency’s conclusion is scientifically indefensible—an extremely high bar97—or that the policy judgements implicit in the agency’s scientific conclusions are somehow inconsistent with the governing statute—a question that Chevron largely places within the agency’s discretion. The lower courts have been less consistent in linking deference to agency science with the Chevron framework, although many examples can be found.98 But even when they do not rely upon it expressly, Chevron’s influence can be felt throughout the case law. As several commentators have observed, Chevron is part of a major theoretical shift in American administrative law away from a technocratic theory of administrative legitimacy and toward a president-centred democratic

95 eg, Upper Blackstone Water Pollution Abatement District v EPA, 609 F3d 9, 22–24 (1st Cir 2012), holding that determining how to proceed when scientific evidence is uncertain lies within the agency’s policy discretion; Cellular Phone Task Force v Federal Communications Commission, 205 F3d 82 (2d Cir 2000): ‘As a policy matter, an agency confronted with scientific uncertainty has some leeway to resolve that uncertainty by means of more regulation or less’; Public Citizen Health Research Group v Department of Labor, 557 F3d 165, 183–84 (3d Cir 2009); Ohio Valley Environmental Coalition v Arcoma Coal Co, 556 F3d 177, 192 (4th Cir 2009); BCCA Appeal Group v EPA, 355 F3d 817, 833–34 (5th Cir 2003); National Truck Equipment Association v National Highway Traffic Safety Administration, 711 F3d 662, 672 (6th Cir 2013), noting that ‘predictive’ judgements necessarily rely on ‘policy considerations’ as well as expertise; Natural Resources Defense Council v Jackson, 650 F3d 662, 665–66 (7th Cir 2011); Friends of the Boundary Waters Wilderness v Dombeck, 164 F3d 1115, 1127–28 (8th Cir 1999); Lands Council v McNair, 537 F3d 981, 993 (9th Cir 2008) (en banc); Morris v Nuclear Regulatory Commission, 598 F3d 677, 684 (10th Cir 2010), holding that resolution of technical issues ‘necessarily require[s] significant expertise and entail[s] the exercise of judgment grounded in policy concerns’; Miami-Dade County v EPA, 529 F3d 1049 (11th Cir 2008); American Coke and Coal Chemicals Institute v EPA, above n 6 at 945. 96 eg, Mississippi v EPA, 744 F3d 1334, 1343 (DC Cir 2013). 97 Troy Corp v Browner, 120 F3d 277 (DC Cir 1997), holding that agencies’ scientific conclusions can only be overcome with ‘substantial and weighty evidence to the contrary’. 98 eg, Catawba County v EPA, 571 F3d 20, 35 (DC Cir 2009); Miami-Dade County, above n 95 at 1063, linking the holdings of Chevron and Ethyl; American Coke and Coal Chemicals Institute, above n 6 at 942; Pennsylvania Department of Environmental Protection v EPA, 429 F3d 1125, 1130 (DC Cir 2005); BCCA Appeal Group, above n 95 at 834–35; National Wildlife Federation v EPA, 286 F3d 554, 560 (DC Cir 2002), linking the holdings of Chevron and Baltimore Gas; Allied Local and Regional Manufacturers Caucus v EPA, 215 F3d 61, 71 (DC Cir 2000); Animal Legal Defense Fund, Inc v Glickman, 204 F3d 229, 234–35 (DC Cir 2000); above n 97 at 283–84; Maier v EPA, 114 F3d 1032, 1043 (10th Cir 1997); Schering Corp v Food and Drug Administration, 51 F3d 390, 399 (3d Cir 1995); Alliance for Cannabis Therapeutics v Drug Enforcement Administration, 930 F2d 936, 939 (DC Cir 1991).

Insights from American Administrative Law

209

theory.99 When courts invoke the greater legitimacy of administrative agencies as policymakers, they are drawing on that trend. By recognising the policy-laden nature of scientific determinations, the courts are frequently able to justify deference to those determinations on the grounds that administrators are better placed than courts to make such judgements without having to delve deeply into the technical merits of the agency’s position. As a result, judicial review of administrative risk regulation has become considerably more deferential, and the Fifth Circuit’s approach, which treated risk as a straightforward empirical matter, has all but died out.100

VI. A Retreat to the Classical Approach? The prevailing view of the US courts, that issues of science and policy are inseparable and that administrators are entitled to exercise their judgement on both, came under strain during the administration of George W Bush.101 It has been widely reported that certain officials in the Bush administration—relying broadly on the contingent and subjective nature of science—used various forms of influence in an effort, sometimes successful, to cause administrative agencies to produce scientific findings that would support pre-established political positions. The most prominent issue, of course, was climate change, for which the Bush administration confidently asserted there was little scientific evidence, but other examples have been documented.102 The Bush administration in some ways upturned traditional positions on s cience and policy. Whereas regulated interests have often tried to impose ‘good science’ requirements in an effort to limit regulatory intervention,103 environmental and 99

Farina, above n 79; Kagan, above n 1; Strauss, above n 79. said, it is important to acknowledge that the case law is not perfectly uniform, and it is possible to find relatively recent examples of courts seeming to treat science-policy judgements as pure questions of fact, eg, Sierra Club v EPA, 671 F3d 955, 963–68 (9th Cir 2012); Bluewater Network v EPA, 370 F3d 1, 21–22 (DC Cir 2004); Dithiocarbamate Task Force v EPA, 98 F3d 1394, 1404 (DC Cir 1996). There are simply too many courts and too many cases for uniformity to be a realistic goal. Much also depends on the particular case; sometimes fully supportable decisions are undermined by a sloppy administrative record. The existence of counterexamples does not, however, invalidate the broader point that cases treating issues of risk science as simple questions of fact have become quite rare in the case law. 101 Further references to the ‘Bush administration’ are to the administration of George W Bush from 2001 to 2009, not to that of George HW Bush from 1989 to 1993. 102 A number of examples are discussed in TO McGarity and W Wagner, Bending Science (Cambridge MA, Harvard University Press, 2010). Other sources include H Doremus, ‘Science Plays Defense: Natural Resource Management in the Bush Administration’ (2005) 32 Ecology Law Quarterly 249 and SA Shapiro, ‘“Political” Science: Regulatory Science After the Bush Administration’ (2009) 4 Duke Journal of Constitutional Law and Public Policy 31. For a preliminary view on the Obama administration’s record, see H Kitrosser, ‘Scientific Integrity: The Perils and Promise of White House Administration’ (2011) 79 Fordham Law Review 2395. 103 WE Wagner, ‘The “Bad Science” Fiction: Reclaiming the Debate over the Role of Science in Public Health and Environmental Regulation’ (Autumn 2003) 66 Law and Contemporary Problems 63. 100 That

210

Christopher Anderson

consumer groups now found themselves advocating for an emphasis on science over politics in decision making.104 The Bush administration’s practices challenged the traditional view of science and policy issues at a much deeper level than political positions, however. Although US courts had come to recognise that policy and science could not be separated, they had continued to treat science as a discourse independent from politics that could discipline administrative decision making. The Bush administration’s assertion that it could act without regard to science or, worse, that it could use political means to dictate scientific outcomes threatened that understanding and with it the theory on which the constitutional legitimacy of administrative regulation rests. The most high-profile case challenging the Bush administration’s approach to science was brought by the Commonwealth of Massachusetts against EPA’s denial of a rulemaking petition seeking limits on greenhouse gas emissions from mobile sources.105 In denying the petition, EPA had refused to decide whether greenhouse gases ‘endanger public health’.106 Instead, the agency declined to regulate on the grounds that it preferred other strategies for addressing possible climate change. At first instance, Judge Randolph, writing the lead opinion for the DC Circuit, held that by offering numerous policy reasons weighing against greenhouse gas regulation EPA had provided a supportable rationale for denying the petition. Judge Randolph’s opinion can be seen as something of a perverse apotheosis of the view that drawing scientific conclusions requires an exercise of policy judgement. He recognised that the scientific evidence regarding climate change is to some degree uncertain and open to interpretation, and then relied on that uncertainty to hold in effect that the agency could proceed solely on the basis of its policy predilections without first making a judgement about the implications of the scientific evidence. Notably, Judge Randolph’s opinion relied heavily on the DC Circuit’s landmark decision in Ethyl Corp v EPA,107 which was one of that Court’s earliest and most forceful statements on the policy-laden nature of risk science and had come to be seen as something of an exemplar of sophisticated judicial review of risk regulation.108 The Supreme Court granted certiorari and reversed. Writing for a five justice majority, Justice Stevens held that EPA’s proffered reasons for declining to regulate were arbitrary and capricious. First, Justice Stevens held that it was incumbent upon the agency to make an endangerment finding. If the agency thought the evidence too inconclusive to warrant action, it was required to make a finding

104

See, eg, McGarity and Wagner, above n 102. Massachusetts v EPA, 415 F3d 50 (DC Cir 2005). 106 This is the statutorily required finding. 42 USC § 7521(a)(1). 107 Above n 105 at 57–58. 108 eg, WE Wagner, ‘Ethyl: Bridging the Science-Law Divide’ (1995) 74 Texas Law Review 1291. Notably, the petitioners challenging EPA’s refusal to regulate had themselves relied heavily on Ethyl. Above n 105 at 289–90. 105

Insights from American Administrative Law

211

of no endangerment and to justify that finding by reference to the uncertainty of the evidence.109 Second, the statute limited the considerations that EPA could take into account in deciding whether to regulate. In the court’s view, EPA had to make its decision solely by reference to the public health implications of greenhouse gas emissions. The other considerations EPA had offered for declining to regulate, whatever their merit, were irrelevant and reliance on them was therefore arbitrary.110 In Massachusetts v EPA, the Supreme Court engaged in a more assertive form of judicial review than had been typical in risk regulation cases since the mid-1990s. Part of the explanation for that scrutiny was the court’s apparent, if implicit, scepticism that EPA was treating the scientific issues openly and honestly. It is also possible, however, that the court was reasserting the primacy of scientific considerations over policy in setting risk standards. Indeed, at least two scholars have argued that the court’s message was that politics needed to take a back seat to expertise in administrative decision making.111 The court’s opinion unquestionably sets limits on the role of policy preferences in administrative regulation. Most importantly, it reaffirms that the administration may only take into account policy considerations that are germane to the goals of the underlying statute. It is a reminder that Chevron’s recognition of a constitutive administrative power to elaborate statutory policy is not a licence to rewrite legislation, no matter how reasonable the administration’s motives may be.112 It reaffirmed, in other words, that the administration is only a subordinate policy maker. It also suggested that the integrity of the agency’s scientific findings matters and that courts need not always assume an agency’s conclusions were reached in good faith. The court’s opinion should not, however, be read as a rejection of the understanding that reaching scientific conclusions requires the exercise of judgement, or as holding that policy considerations can or should be purged from this process. An important factor in the Massachusetts opinion was that EPA had refused to make any judgement about the science, policy-laden or otherwise. Instead, the agency had offered policy reasons for not addressing the scientific issues, and it was that attempt to avoid confronting the scientific evidence that the court found to be arbitrary. The issue of the role of policy considerations in reaching scientific judgements simply did not arise. Indeed, its discussion of EPA’s refusal to make an endangerment finding suggested that the court understood that when science is uncertain an agency will have to look to policy considerations to reach the necessary conclusions.113

109

Massachusetts v EPA, 549 US 497, 533 (2007). ibid 533–34. 111 J Freeman and A Vermeule, ‘Massachusetts v. EPA: From Politics to Expertise’ [2007] Supreme Court Review 51. 112 Indeed, Justice Stevens’s majority opinion relied heavily on his earlier opinion in Chevron, eg, above n 109 at 534. 113 ibid 533–34. 110

212

Christopher Anderson

All in all, it seems doubtful that in Massachusetts the Supreme Court was advocating a significant retreat from the prevailing judicial understanding that considerations of science and considerations of policy are essentially inseparable in administrative risk regulation. Instead, the court’s judgment seems to be a reminder that for the exercise of administrative policy discretion to be legitimate, it must be based on both the agency’s expert evaluation of the situation and its politically informed judgement regarding the appropriate regulatory response. Thus far, the lower courts appear to have interpreted the decision in this way, and recent cases do not show a trend away from an understanding that risk regulation demands the exercise of both scientific and policy judgement.114

VII. Conclusion A review of the US case law on judicial review of administrative risk regulation shows the gradual, but decisive, abandonment of the classical account. It also shows that this development was motivated in large part by an increasing recognition that the formation of scientific judgements regarding risk is inevitably bound-up with normative analysis of those risks. Because US courts recognise the inherent ambiguity and contestability of risk science, they have increasingly come to accept that when agencies draw scientific conclusions, they are simultaneously making policy judgements. As a result, the courts have moved away from the classical model’s approach to review, which treats scientific conclusions as neutral matters of fact.115 The evolution in the US courts’ case law is to be welcomed for adopting a more sophisticated and realistic understanding of risk science and its limits. A theory of administrative law that incorporates naïve understandings of risk science serves no one’s interests. The cost of this salutary development, however, was the collapse of the classical account of administrative legitimacy. Without some claim to neutrality, science could no longer provide an apolitical foundation for the exercise of bureaucratic power.116 If anything, the recognition that scientific analysis itself is a policy-laden undertaking seems to require deeper forms of political legitimacy than the classical account could provide. This breakdown in theories of administrative legitimacy accounts for much of the confusion and inconsistency that has been observed in the early case law dealing with administrative risk regulation.

114 eg, Center for Biological Diversity v EPA, 749 F3d 1079 (DC Cir 2014); Coalition for Responsible Regulation, Inc v EPA, 684 F3d 102 (DC Cir 2012) affd in part and revd in part sub nom Utility Air Regulatory Group v EPA, 134 S Ct 2427 (2014); Upper Blackstone Water Pollution Abatement District, above n 95; Miami-Dade County, above n 95. 115 cf CF Edley, Administrative Law: Rethinking Judicial Control of Bureaucracy (New Haven, Yale University Press, 1990) 74–77. 116 Above n 3 at 1331–32.

Insights from American Administrative Law

213

The solution that eventually developed posited a much more complex relationship between scientific expertise and administrative power. Although it no longer aspires to be an alternative to politics, the promise that the application of scientific expertise to social problems will produce superior public policy continues to be the primary justification for delegating substantial law making authority to administrative agencies.117 The application of expert analysis thus remains essential to the legitimacy of administrative risk regulation. The acknowledgment of the policy-laden nature of scientific conclusions, however, means that science is no longer sufficient to justify specific regulatory choices. To the contrary, agencies’ scientific conclusions—to the extent they are based on an exercise of policy judgement—are themselves made authoritative by the agencies’ presumed democratic legitimacy. The courts have in effect held that when the science is uncertain or disputable, the agency’s position is entitled to deference not because it is right (or even better) as a scientific matter, but because the agency is the democratically appropriate institution to make the choice among competing interpretations of the evidence. In the contemporary account, scientific expertise and administrative authority are two sides of the same coin; they are in effect co-produced. Although the contemporary account improves on the classical account in its understanding of risk science and its limitations, it is not without significant problems of its own. Most obviously, its core premise that administrative agencies enjoy limited democratic legitimacy is highly controversial. The constitutional and empirical arguments for and against a ‘presidential’ vision of administration are complex and cannot be reviewed here,118 but the centrality of the problem is obvious. If policy choices made by administrative agencies are not democratically legitimate to some degree, then the contemporary account falls apart. Whether one accepts the (limited) democratic legitimacy of administrative policy making ultimately depends on one’s theory of democracy,119 and resolution of the question is thus necessarily a matter for the political process. Given the highly contestable nature of the question, it is perhaps not surprising that the courts have retreated to a rather formalistic theory that focuses on the president’s constitutional position as chief executive instead of a more complex analysis of the political and administrative realities. Even if one accepts the democratic legitimacy of administrative policy making, however, there remain important questions about the role of law in the contemporary account. One of the effects of the shift to the contemporary account has been a substantial increase in the deference courts afford agencies. This change

117

Above n 111 at 93–96; Mashaw, above n 1 at 19–23. and normative arguments against a presidential theory of administration are welldeveloped in CR Farina, ‘The Consent of the Governed: Against Simple Rules for a Complex World’ (1996) 72 Chicago-Kent Law Review 987; perhaps the most robust defence of this theory is Kagan, above n 1. 119 PP Craig, Public Law and Democracy in the United Kingdom and the United States of America (Oxford, Clarendon Press, 1990) 3–5. 118 Empirical

214

Christopher Anderson

is attributable, in part, to greater judicial reluctance to review technical issues closely. To a much greater degree, however, increased judicial deference is a result of the understanding that scientific judgements, particularly in the area of risk regulation, are inextricably bound-up with judgements about regulatory policy. Once a court determines that an issue turns on policy judgements, it will typically limit its review to confirming that the agency’s decision falls within its statutory authority; only in extreme cases will the court second-guess the substance of those judgements.120 In effect, by treating mixed science-policy questions as matters of policy, the contemporary account places those questions outside the ambit of legal analysis. The question is whether limiting the scope of judicial review in this way is desirable. On one hand, there is much to be said for limiting the judicial role in regulation. It is far from clear that an unelected, generalist judiciary is wellplaced to intervene in issues that mostly turn on technical detail and democratic preference.121 Indeed, there are plenty of examples of over-eager courts committing technical errors, making highly contentious policy judgements, and failing to appreciate the systemic effects of their decisions. On the other hand, restricting the intensity of judicial review may result in an intolerable increase in administrative power, with attendant opportunities for sloppy, biased, or even corrupt regulation. In addition to these practical worries, the very practice of judicial review performs an important legitimating function.122 If courts can review neither the technical analysis nor the policy content of administrative action, there is a risk not only that they will hand too much power to the executive, but also that the perception of weak judicial review will itself undermine confidence in the administrative state. For now at least, these fears seem unfounded. Although judicial review of administrative risk regulation is undoubtedly less intrusive than it once was, it has hardly become a perfunctory exercise. In recent decades, US courts have developed a number of techniques of substantive review directed at rooting out unprincipled decision making123 and, as Massachusetts v EPA and similar cases show, they have had some success in doing so. That is not to say that no bad decisions have avoided judicial disapproval; most likely some have. It does, however indicate that judicial review remains sufficiently rigorous that agencies run a real risk of having their actions set aside if they fail to engage in reasoned decision making. These successes demonstrate that the contemporary account’s recognition of the co-production of

120 That hesitancy is traceable to a longstanding American suspicion of judicial law making. LS Bressman, ‘Beyond Accountability: Arbitrariness and Legitimacy in the Administrative State’ (2003) 78 New York University Law Review 461. 121 Mashaw, above n 79 at 93–95; SF Williams, ‘The Roots of Deference’ (1991) 100 Yale Law Journal 1103, 1105–06. 122 LL Jaffe, Judicial Control of Administrative Action (Boston, Little, Brown & Co, 1965) 320. 123 E Fisher, P Pascual and W Wagner, ‘Rethinking Judicial Control of Expert Agencies’ (2015) 93 Texas Law Review 1681.

Insights from American Administrative Law

215

scientific conclusions and policy positions need not lead to a break down in the ability of courts to enforce the rule of law in administrative risk regulation. Although they do not use the term, the US courts seem to have grasped the nub of co-production. In particular, they have internalised the view that an agency’s interpretation of scientific evidence cannot be fully separated from its views on good public policy. The recognition of co-production has not lessened the challenges for administrative law, however. If anything, its acknowledgment of the judgement-laden nature of many scientific conclusions has complicated the courts’ task by making the questions about the legitimacy of administrative regulation more urgent. What the case law also demonstrates is that questions about the role of expertise in administrative regulation cannot be separated from normative visions of administration and its role in the constitutional order. It was only by redefining the theoretical foundation of administrative legitimacy that the Supreme Court was able to create the legal preconditions for development of the contemporary account. However one evaluates the court’s solution, the contingency and contestability of the administration’s role cannot be ignored. Recognising co-production is thus only the first step in analysing the relationship between expert and executive power. The next and more fundamental step is to build a normative vision of the role of the administration within a particular democratic polity. Only once we have an articulable theory of the administration’s role is it possible to assess normatively the role of expertise in the administrative process.

216

11 Accountability and Co-production beyond Courts: The Role of the European Ombudsman MARIA LEE1

I. Introduction Accountability is never straightforward: what we might mean by accountability is likely to depend on our (possibly unexamined) assumptions about what administrative (or executive) decision-making is and should be, so that when we disagree about whether a decision maker is properly accountable, we also disagree about deeper commitments.2 A rich literature discusses variously the meaning, goals, components and frameworks of accountability, in all sorts of contexts. A comprehensive definition is elusive, but accountability can be simply described as a relationship between two parties in which one has an ‘obligation to explain and justify conduct’.3 The actor being held to account is subject to some form of external scrutiny, as well as to the possibility of ‘facing consequences’.4 The ‘ombudsman institution’ is frequently understood as an institution of accountability.5 This chapter explores some of the ways in which the E uropean

1 I am grateful to participants at ‘Regulating Risks in the European Union: The Co-production of Expert and Executive Power’ 21–22 May 2015, particularly to Maria Weimer for her detailed feedback, and to Steven Vaughan for comments on a draft of this chapter, as well as to Amarvir Sidhu for his research assistance. 2 eg S Shapiro and E Fisher, ‘Chevron and the Legitimacy of “Expert” Public Administration’ (2013) 22 William & Mary Bill of Rights Journal 465; R Mulgan, ‘“Accountability”: An Ever-Expanding Concept?’ (2000) 78 Public Administration 555; on judicial review, C Anderson, ‘Contrasting Models of EU Administration in Judicial Review of Risk Regulation’ (2014) 51 Common Market Law Review 425. 3 M Bovens, ‘Analysing and Assessing Accountability: A Conceptual Framework’ (2007) 13 European Law Journal 447, 450; J Black, ‘Calling Regulators to Account: Challenges, Capacities and Prospects’ LSE: Law, Society and Economy Working Papers 15/2012, 4, www.lse.ac.uk/collections/law/ wps/WPS2012-15_Black.pdf. 4 Bovens, ibid. 5 eg T Buck, R Kirkham and B Thompson, The Ombudsman Enterprise and Administrative Justice (Farnham, Ashgate, 2011) for whom ombudsmen promote integrity and accountability; P Magnette,

218

Maria Lee

Ombudsman (EO) might contribute to accountability.6 The EO describes its ‘mission’ as ‘to serve democracy by working with the institutions of the European Union to create a more effective, accountable, transparent and ethical administration’.7 It is ‘empowered’ by Article 228 of the Treaty on the F unctioning of the European Union (TFEU) ‘to receive complaints from any citizen of the Union or any natural or legal person residing or having its registered office in a Member State concerning instances of maladministration … He or she shall examine such complaints and report on them’.8 The particular context for this examination of the EO is the co-production of expert and executive authority. I reflect upon the insight that not only are ‘facts’ socially constructed, but that ‘society’ does not exist independently of and prior to the facts; social and natural worlds are mutually constitutive. In respect of expertise and executives, neither form of authority is independent of the other, and each shapes and rests upon the other; neither straightforwardly controls the other or has an autonomous definition of the purpose of their interactions.9 This perspective raises particular (if not unique) challenges for accountability: at its simplest, when power is co-produced, it is not easy to see where or by whom it is exercised, and accordingly how or with whom accountability relationships might most appropriately be constructed. The next section introduces both the complexity of EU governance and the space occupied by co-produced authority in the EU. It also reflects upon the perennial scholarly concern around the accountability of the EU’s dense processes of governance, especially the ways in which the most familiar and powerful legal and political/democratic routes to accountability are left wanting.10 There is relatively little detailed discussion of the EO in the literature on EU governance and accountability.11 Nor is co-production an explicit feature of this literature.

‘Between Parliamentary Control and the Rule of Law: the Political Role of the Ombudsman in the European Union’ (2003) Journal of European Public Policy 677 for whom the EO has ‘hybrid’ role, between political and legal accountability. 6 On the EO, see eg C Harlow and R Rawlings, Process and Procedure in EU Administration (Oxford, Hart Publishing, 2014) ch 3; Magnette, ibid; A Tsadiras, ‘The Ombudsman’ in P Craig (ed), EU Administrative Law (Oxford, Oxford University Press, 2012). 7 EO, Strategy of the European Ombudsman: Towards 2019 (2014), 3. All EO documents discussed here can be found at www.ombudsman.europa.eu/en/home.faces. 8 Ombudsmen are often personalised so that ‘he’ or ‘she’ reports, which reflects the importance of the personal qualities of the ombudsman. However, to reflect the institutional status of the EO (and for simplicity), I de-personalise, other than when discussing speeches or publications in the EO’s own name. 9 On co-production, see S Jasanoff (ed), States of Knowledge: The Co-Production of Science and Social Order (London, Routledge, 2004); Weimer and de Ruijter, ch 1 in this volume. 10 Although not powerless, eg C Harlow, ‘Composite Decision-Making and Accountability Networks: Some Deductions from a Saga’ (2013) 32 Yearbook of European Law 3. 11 Important exceptions (from different perspectives) include M Dawson, ‘Transforming into What? New Governance in the EU and the “Managerial Sensibility” in Modern Law’ [2010] Wisconsin Law Review 389 and C Harlow and R Rawlings, ‘Promoting Accountability in Multilevel Governance: A Network Approach’ (2007) 13 European Law Journal 542.

Accountability and the European Ombudsman

219

Co-production is however deeply entwined in the existing focus on the complexity of EU governance, given shared interests in both the relationship between knowledge and authority, and the blurring of taken for granted lines between different stages and types of governance. After this discussion, I explore more specifically the EO’s general promise as an institution of accountability. The following section then turns to a reading of EO decisions,12 hoping to provide some modest insights into the ways in which the EO grapples with complexity and engages with co-production. It is perhaps not surprising that what we learn directly about co-production from the EO is rather thin. But the EO does take a keen interest in the generation both of knowledge, scrutinising the various groups providing scientific or technical advice and expertise in the EU, and (perhaps less keenly) of authority. The intention in this chapter is not to propose the EO as a solution to our accountability challenges, or even to encourage the EO to engage differently with co-production. In fact, a ‘solution’ to the EU’s accountability challenges is probably impossible. Nonetheless, the EO is an important institution and its contribution to governance deserves critical attention.

II. Co-production and Accountability in EU Governance The complexity of EU governance is the subject of large and powerful literatures in a range of disciplines and sub-disciplines.13 At least four matters might be borne in mind. First, decision-making processes are obscure, and highly varied; the diversity of institutional structures for the delivery of expertise can be difficult to keep track of, and invisible without careful scrutiny of individual regimes. Second, the multiple (especially central and Member State) levels of governance involved in decisions are often impossible to disentangle, so that processes and decisions can no longer (if they ever could) be easily categorised as intergovernmental or supranational, ‘national’ or ‘European’. Third, public and private actors

12 I scanned all OIIs, the opening sections of all Draft Recommendations between January 2010 and May 2015, and of all Decisions between July 2013 and May 2015, selecting the cases that seemed most relevant for closer examination. In particular, I dismissed the employment, contract and tender complaints, state aids and competition complaints, and cases about Commission enforcement decisions. One absence from the EO’s ‘docket’ is striking: I have not encountered complaints about the ways in which the Commission (especially) uses, or does not use, scientific advice; by contrast, these are central to the most interesting litigation. 13 Selecting from the literature is daunting, but as well as work cited herein, see eg K Armstrong, ‘The Character of EU Law and Governance: From “Community Method” to New Modes of G overnance’ (2011) 64 Current Legal Problems 179; CF Sabel and J Zeitlin (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2010); G de Búrca and J Scott (eds), Law and New Governance in the EU and the US (Oxford, Hart Publishing, 2006).

220

Maria Lee

work together in ways that elide any public/private divide. And finally, the lines between institutionally separated areas of responsibility (including EU/national and public/private, but for current purposes, most importantly ‘science’ and ‘politics’) are difficult to maintain. Trying to isolate co-production from these broader features of EU governance, or to map features of co-production precisely upon them, would probably be unhelpful. The mutual shaping, influence and dependence of expert and executive resources and authority pervades EU governance. Exploring these four features provides useful insight into EU governance. A brief review of the arrangements for the delivery of expertise and scientific or technical advice in the EU brings out these four layers of complexity. On the obscurity and diversity of institutional structures, even at its simplest there are at least three broad approaches,14 within which lies considerable diversity, to knowledge generation. First, the Commission calls frequently on ‘expert groups’, defined as bodies ‘set up by the Commission or its departments to provide it with advice and expertise, comprising at least 6 public and/or private-sector members and meeting more than once’.15 Basic rules on appointment, operation and transparency apply,16 including mandatory registration on the Expert Group Register, although that is incomplete and sometimes inaccurate.17 Second, the EU agencies vary in their precise architecture and responsibilities, but in most cases they have an expert advisory or information generation role. They generally contain specialist scientific or technical committees, ‘the beating hearts of agencies’,18 and Management Boards composed of representatives of EU institutions, M ember States and sometimes stakeholders. In some cases there is space for explicit political orientation, for example in the European Chemicals Agency (ECHA)’s Member State Committee, which has a consensus seeking role and, as its name suggests, represents national interests.19 In some cases there are standing stakeholder consultative groups.20 Third, ad hoc approaches to expertise in particular pieces of legislation can be harder to pin down. For example, the Industrial E missions

14 I am not suggesting that this exhausts the possibilities, consider also eg networks or epistemic communities, and even these three categories might be differently arranged. Still at an embryonic stage is the new ‘Scientific Advice Mechanism’, which will ‘draw on the wide range of scientific expertise in Europe through a close relationship with national academies and other bodies, as well as the expertise of a High-Level Group of independent scientific advisors’, see www.ec.europa.eu/research/sam/index. cfm. 15 See ec.europa.eu/transparency/regexpert/index.cfm?do=faq.faq&aide=2. 16 European Commission, Framework for Commission Expert Groups: Horizontal Rules and Public Register C(2010) 7649 final. The rules also apply to ‘other similar entities’, which were not set up by the Commission, but play a similar role and are administered by the Commission. 17 Decision in complaint 1682/2010 against the European Commission (expert groups). 18 E Vos, ‘EU Agencies: Features, Framework and Future’ Maastricht Faculty of Law Working Paper 2013/3, 15, www.maastrichtuniversity.nl/web/Faculties/FL/2013_maastricht_working_papers.htm. 19 Reg 1907/2006/EC of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency [2006] OJ L396/1, art 76. 20 eg ECHA’s Directors Contact Group, see www.echa.europa.eu/web/guest/about-us/partnersand-networks/directors-contact-group.

Accountability and the European Ombudsman

221

Directive21 provides a bare framework for the ‘Seville Process’ (the European Integrated Pollution Prevention and Control Bureau22 is based in Seville) in which Technical Working Groups, composed of a range of national and European, public and private, actors produce draft Best Available Techniques (BAT) reference documents (BREFs) to describe ‘best available techniques’ for different sectors.23 Turning to the multi-level, public/private nature of these groups, again variety in detail is the rule. ‘National’ experts are routinely, but with varying intimacy, involved in these ‘EU’ agencies, committees and groups, often alongside EU officials.24 Although participants often have obligations of ‘independence’, concern with even-handedness of representation suggests that the potential for national affiliations to provide different perspectives on a problem is tacitly understood. Private actors, including public interest groups such as environmental non-governmental organisations (NGOs), but predominantly economic actors who are the subject of the regulation, are also ubiquitous as experts, stakeholders and sometimes representatives of the Member State.25 And finally, the purported divide between science or expertise and politics is firmly institutionalised, but difficult to maintain.26 The groups and committees discussed so far are generally institutionally and rhetorically presented as providers of expertise or of scientific or technical opinions and advice. The more complex reality of their task flares into view during the occasional controversy, but their normative role is also routine in less high profile cases. The formally ‘political’ decision is generally taken by the Commission.27 Even at this final stage, we have multi-level arrangements, since the Commission is supported or supervised by comitology, a set of processes for providing Member State oversight of Commission decision-making. The Member State representatives are generally

21 Dir 2010/75/EU of 24 December 2010 on industrial emissions (integrated pollution prevention and control) [2010] OJ L334/17. 22 Part of the Sustainable Production and Consumption Unit of the Institute for Prospective Technological Studies, and in turn one of the Commission’s Joint Research Centre institutes. 23 M Lee, EU Environmental Law, Governance and Decision-Making (Oxford, Hart Publishing, 2014) ch 5. 24 eg Member States are represented on Technical Working Groups in Seville. But whilst Member States nominate members to the ECHA committees, and the ECHA Management Board appoints at least one, and no more than two, members from each nominating Member State to each committee (REACH, above n 19 at art 85), members of EFSA’s scientific panels are recruited by open calls for expression of interest. 25 C Abbot and M Lee, ‘Economic Actors in EU Environmental Law’ (2015) 34 Yearbook of European Law 1. 26 By the Court from Case 9/56 Meroni v High Authority [1957/8] ECR 133 through to cases following Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305, emphasising the need for politically legitimate decisions; risk-regulating legislation (such as IED and REACH) tends to reinforce the distinction; in policy, eg European Commission, Communication on the Precautionary Principle COM (2000) 1 final. The distinction could plausibly be softened by any increased delegation of decision-making authority to agencies following Case C-270/12 UK v European Parliament and Council (Short Selling) (CJEU, 22 January 2014); however, this case does not revolutionise existing practice, and agencies already have their own internal institutional divides. 27 In some cases final decisions, eg the grant of permits to operate, are taken by national authorities.

222

Maria Lee

civil servants, but sometimes industry representatives, and occasionally high level political representatives, including ministers of state. Since the Lisbon Treaty, comitology no longer applies to Commission ‘delegated’ acts,28 but the intention seems to be to continue to consult Member State experts in committee. The focus of this volume is co-production rather than the complexity of governance. The explicit language of co-production is rarely used by scholars of EU governance, but some of its features are evidently well understood. The blurred lines between European and national, public and private, for example, are often noted, as is the elusive line between politics and science, and the role of expertise as a legitimating mechanism. However, completing the circle of co-production, the Commission not only relies on ‘the facts’ as a source of its own political authority, but also participates in shaping the establishment of facts.29 For example, the Commission is one of the political actors influencing the ways in which regulation is applied and interpreted by the technical bodies. Its demands for, approval of, or simple articulation of standards for technical/scientific decision-making, does some work ‘enhancing scientific credibility in public contexts’.30 The accumulation of detail on how the risk assessment of GMOs should be conducted, for example, may be seen as a conscious effort to enhance the status of the European Food Safety Authority (EFSA)’s facts, in the face of challenges to EU authority.31 This idea that ‘what one knows in science significantly depends on prior or concurrent choices about how one chooses to know it’32 has been implicitly raised before the EO. To extrapolate a little, a complainant alleged that the integrity of EFSA’s whole approach to risk assessment (said to be the approach preferred by industry) was impugned by the private interests of the Chair of one of its science panels.33 The EO however did not address the significance of standard setting for the production of facts, focusing instead on EFSA’s conflict of interest procedures. It might even be fair to say that there is a self-conscious effort to co-produce authority in the EU, again fitting reasonably neatly into a discussion framed around the density of EU governance. So both the European ‘scientific’ bodies 28 Although the legislators continue to be involved. The new distinction between ‘implementing’ and ‘delegated’ acts (TFEU, arts 290–91) has not been much clarified by the Court, Case C‑427/12 Commission v Parliament and Council (CJEU 25 April 2014). 29 The focus here is largely on the facts as they emerge from risk assessment; a similar phenomenon is arguably emerging with respect to cost benefit analysis (CBA), which is increasingly called on in legislation and policy, since there is arguably also an increasing reliance on the ‘facts’ of costs and benefits to legitimise a decision, eg Lee, above n 23 at ch 2. Whilst political reliance on the facts of costs and benefits shape and enhance the authority of those producing CBAs, the production of CBAs shape decisions and their legitimacy. 30 CA Miller, ‘Climate Science and the Making of a Global Political Order’ in Jasanoff, above n 9 at 56. 31 See also the examples discussed by Fisher in this volume, and the more problematic cases discussed by A Stirling, ‘Power, Truth and Progress: Towards Knowledge Democracies in Europe’ in J Wilsdon and R Doubleday (eds), Future Directions for Scientific Advice in Europe (Cambridge, Centre for Science and Policy, 2015). 32 S Jasanoff, ‘Making Order: Law and Science in Action’ in E Hackett et al (eds), The Handbook of Science and Technology Studies (Cambridge MA, MIT Press, 2007) 772. 33 Decision in complaint 622/2012 against EFSA (Test Biotech).

Accountability and the European Ombudsman

223

and the European ‘political’ bodies assert and enhance their own legitimacy and authority by reference to the institutional appropriateness of their own activity, and the ‘scientific’ or ‘political’ legitimacy of their interlocutor. Each contributes to the shaping of, and is shaped by, the other. Equally importantly, the insights of co-production may apply beyond natural and social life, scientific and political authority. ‘Europe’ calls on European science, knowledge and authority to present issues (sometimes contentiously) as of European importance.34 We ‘choose to know’ in Europe, and the ‘making’ of ‘knowledge’ contributes to ‘the making and constant re-ordering of Europe as an institutional and political entity’.35 The inclusion and authority of national experts shore up the authority of collaboration at the EU level. Similarly, ‘private’ knowledge is legitimised by its presentation and promulgation by an ‘official’ forum composed of a more mixed group; outputs may in turn be legitimised by the inclusion of private knowledge. In each case, calls on expertise and objectivity, or on the other hand on deliberation, consultation and representativeness, constitute an effort to render authority less problematic. The ways in which these legitimating strategies might fit together (or not) in the EU is unclear, and they may simply be uncoordinated efforts to appeal to different legitimacy communities.36 Whilst some worry about ‘too much science’, others worry equally about ‘too much politics’, when actually the two categories mutually reinforce each other.37 Accountability and legitimacy have been perennial concerns for those interested in the fragmentation and density of EU governance mechanisms. I hope that co-production does not add to ‘contestation over naming rights’,38 but thinking explicitly about co-production could enhance our sense and understanding of authority building, the ways in which authority is divided up and patched back together, and sharpen our thinking on the elusiveness of hierarchy.39 The institutionalisation of the (fictional)40 separation between scientific knowledge and politics could be interpreted as a particular way of thinking about accountability. 34 eg S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005). 35 C Waterton and B Wynne, ‘Knowledge and Political Power in the European Environment Agency’ in Jasnoff, above n 9 at 88, specifically on environmental knowledge. 36 On which, see eg J Black, ‘Constructing and Contesting Legitimacy and Accountability in Polycentric Regulatory Regimes’ (2008) 2 Regulation & Governance 137, 144. They are also part of the task of constructing objectivity, discussed below from n 106. 37 M Weimer, ‘Risk Regulation, GMOs, and the Challenges to Deliberation in EU Governance— Politicization and Scientification as Co-Producing Trends’ in C Joerges and C Glinski (eds), The European Crisis and the Transformation of Transnational Governance—Authoritarian Managerialism versus Democratic Governance (Oxford, Hart Publishing, 2014); M Lee, ‘Beyond Safety? The Broadening Scope of Risk Regulation’ (2009) 62 Current Legal Problems 242. 38 B Karkkainen, ‘“New Governance” in Legal Thought and in the World: Some Splitting as Antidote to Overzealous Lumping’ (2004–05) 89 Minnesota Law Review 471, 478. 39 M Lee, ‘The Ambiguity of Multi-Level Governance and (De)-harmonisation in EU Environmental Law’ (2014) 15 Cambridge Yearbook of European Legal Studies 357. 40 S Jasanoff, ‘A Century of Reason: Experts and Citizens in the Administrative State’ in S Skowronek, S Engel and B Ackerman (eds), The Progressives’ Century: Democratic Reform and Constitutional Government in the United States (New Haven, Yale University Press, 2014).

224

Maria Lee

It may reinforce the plausibility of peer review of technical decisions, rather than any broader approach, as a satisfactory form of accountability. The governance arrangements for the generation of knowledge discussed above tend to focus on peer accountability, meaning accountability towards others within the process, in an effort to fill the gaps left by the fragmentation of political and legal accountability. This is potentially rather powerful, and may provide the committed, resourced and informed account holder that we need.41 It is however obviously limited, leaving unquestioned the identification of those peers, and the assumptions and approaches shared by ‘peers’.42 The rhetorical and institutional separation of science from politics may also attempt to provide one answer to ‘the enduring question of how experts, with their specialist knowledge, can be held accountable to public values’.43 The insistence that final decisions are the responsibility of politically legitimate institutions44 reinforces a formal delegation model, in which the accountability of the superior satisfies the accountability demands on the committee or working group. This model of accountability is found seriously wanting in the EU’s complex governance framework. The inadequacy of the delegation model is brought out by the fragile political legitimacy and weak political accountability of EU political authority, not least Commission plus comitology,45 and in some cases, control of the ‘lower’ level actor will be resisted precisely because its independence is valued.46 As suggested above, trying to attribute particular accountability challenges to co-production, rather than other ways of thinking about governance, is problematic. But to go back to the fundamental perspective of c o-production: nature and society, facts and governance, mutually shape each other, and neither exists in the world independently of the other. Expertise and executive authority each depends on and is constituted by the other. So as with other complexityembracing perspectives on EU governance, traditional (legal, political/democratic) mechanisms of accountability are not hopeless, but are lacking. No individual actor can be found fully responsible; but nor can any individual actor be entirely without responsibility.47 41 Black, above n 3, on the challenges faced by the party doing the holding to account; CF Sabel and J Zeitlin, ‘Learning from Difference: The New Architecture of Experimentalist Governance in the EU’ in Sabel and Zeitlin, above n 13 on the advantages of peer accountability. 42 eg Harlow and Rawlings, above n 11. 43 Jasanoff, above n 40; although not to the other half of her question. 44 Above n 26. The robustness of that claimed legitimacy is not explored in those sources. 45 D Curtin, ‘Holding (Quasi-) Autonomous EU Administrative Actors to Public Account’ (2007) 13 European Law Journal 523 on the weakness of links to political accountability; D Sarewitz, ‘How Science Makes Environmental Controversy Worse’ (2004) 7 Environmental Science and Policy 385 on the importance of political legitimacy. 46 At issue in Draft Recommendations in complaint 1151/2008 against the European Commission (biofuels), para 83. 47 See also Sabel and Zeitlin, above n 41, on the difficult of identifying ‘principal’ and ‘agent’ for the purposes of ‘principal-agent’ accountability, at 11; Harlow, above n 10 on the difficulty of holding the ‘administrative’ answerable to the ‘executive’, with the blurring of any line between administrative and executive roles, D Curtin, Executive Power of the European Union (Oxford, Oxford University Press, 2009).

Accountability and the European Ombudsman

225

III. The EO as a Forum for Accountability The EO has a number of institutional advantages in terms of accountability.48 First and perhaps most importantly, the EO is able to take a strategic approach to governance: ‘firewatching’ as well as ‘firefighting’.49 It has made a number of strategic contributions to the shaping of EU governance, perhaps most famously in respect of Commission infringement proceedings against Member States (particularly the treatment of complainants), and in respect of obligations of transparency.50 The EO’s power to undertake an own initiative inquiry (OII) allows for a more holistic view of administration than would be possible simply by responding to complaints.51 The ongoing OII into Commission Expert Groups is discussed further below.52 The EO visits agencies, proactively scrutinising such policies as language use, conflicts of interest, whistleblowing and transparency.53 And the EO can articulate standards of good administration, ‘a vital element in the process of securing accountability’.54 Along with its guidelines and codes of practice, annual reviews of decisions may in part be an effort to rationalise its approach, and to shape a body of something like precedent.55 In addition to this strategic overview, the EO has a potentially important ‘dialogic and political role’,56 emphasising the ‘relationship’ part of the ‘accountability relationship’. It has strong investigative powers, and ‘can conduct all the enquiries which [it] considers justified’.57 It has access to files, can commission reports and 48

For more detailed discussion of the EO’s origins and role, see above n 6. and Rawlings, above n 5 at 65. The distinction is of course not clear cut, eg Magnette, above n 5. 50 eg R Rawlings, ‘Engaged Elites, Citizen Action and Institutional Attitudes in Commission Enforcement’ (2000) 6 European Law Journal 28; P Dyrberg, ‘Accountability and Legitimacy: What Is the Contribution of Transparency?’ in A Arnull and D Wincott (eds), Accountability and Legitimacy in the European Union (Oxford, Oxford University Press, 2002). 51 TFEU, art 228; Decision of the European Ombudsman adopting implementing provisions (2002, as amended 2008), art 9. OIIs are reserved for ‘matters of significant public importance or principle’, EO Strategy, above n 7 at 3. 52 Own-initiative inquiry OI/6/2014 concerning the composition of Commission expert groups. 53 Through a series of OIIs, eg OI/9/2011 (EMA), OI/11/2011 (European Environment Agency), OI/12/2012 (ECHA). 54 C Harlow, Accountability in the European Union (Oxford, Oxford University Press, 2002); E Fisher, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495. There is some debate on the appropriateness of EO standard setting, see eg ME de Leeuw, ‘The European Ombudsman’s Role as a Developer of Norms of Good Administration’ (2011) 17 European Public Law 349. The line between setting and applying standards is not however easy to draw, and the EO is in part addressing the plethora of other standard setters in the EU. 55 Certainly this is suggested by EO, Good Administration in Practice: The European Ombudsman’s Decisions in 2013 (2014); earlier overviews are less substantive. 56 Dawson, above n 11 at 432. 57 Decision of the European Parliament on the Regulations and General Conditions Governing the Performance of the Ombudsman’s Duties OJ L 113/25 (1994) as amended in 2002 and 2008 (the ‘Statute’), art 3; see also Implementing Regulations, above n 50 at art 5. See A Tsadiras, ‘Unravelling Ariadne’s Thread: The European Ombudsman’s Investigative Powers’ (2008) 45 Common Market Law Review 757. 49 Harlow

226

Maria Lee

consult the public, and hear from officials or ask questions of the challenged body. Submissions from respondents seem generally to provide substantial detail on the complaint, literally providing a public account, but this capacity to engage in ongoing dialogue with respondents,58 and to require responses, enhances the EO’s capacity to demand an account.59 The EO’s possible responses to maladministration also tend to emphasise ongoing reflection and dialogue, both in resolving the individual issue, and improving standards.60 On a finding of maladministration, the EO initially tries to ensure a ‘friendly solution’ between the institution and the complainant.61 The case is closed with a reasoned decision if the friendly solution is effective. If a friendly solution is not possible, and the maladministration has no general implications, the EO closes the case with a critical remark. Otherwise, a report with ‘draft recommendations’ is issued.62 If the EO is not satisfied with the institution’s response to its draft recommendations, it can provide a Special Report to the European Parliament. These are rare: there were none in 2014, and only one in both 2013 and 2012.63 The EO now reports systematically on responses to its inquiries, and concluded that in 2013 the institution responded ‘constructively’ to its intervention in four out of five cases.64 It promises to use its ‘further remarks’, which are intended to raise the quality of administration for the future and do not necessarily imply maladministration, to make ‘concrete suggestions’ for systemic improvements, or to invite the institution to make its own suggestions and report back. The EO says it will also systematically check that the institution does what it says it will do,65 and it can revisit difficult or stubborn areas through OIIs.66 Whilst many approaches to accountability demand the possibility of sanctions,67

58 And with others, eg consultees, complainants, the European Parliament. Harlow and Rawlings, above n 11, argue that the EO network (with national ombudsmen) is capable of seeing into multi-level governance, organised around networks of actors at all levels of governance. 59 Black, above n 3. 60 I am interested in a sub-set of cases, but for data on the processing of all complaints, see the EO’s Annual Reports. 61 EO Statute, above n 57 at art 3(5)–(7). The EO ‘finds it more constructive to avoid stating, even provisionally, that there could be maladministration’, instead identifying a ‘problem or shortcoming … that could be put right’, EO, Putting it Right? How the EU Institutions Responded to the Ombudsman in 2013 (2014) at 5. The EO resists being described as a mediation service, Decision in OI/12/2011 concerning the European Monitoring Centre for Drugs and Drug Addiction, para 7. 62 Followed by a reasoned decision. 63 Special Report in OI/5/2012 concerning Frontex (being considered by joint committee, 2014/2215(INI); Special Report in 2591/2010 against the European Commission (Vienna airport) (European Parliament resolution of 12 March 2013 on the Special Report of the European Ombudsman concerning his inquiry into complaint 2591/2010/GG against the European Commission). 64 EO, above n 61. M Hertogh, ‘Coercion, Cooperation, and Control: Understanding the Policy Impact of Administrative Courts and the Ombudsman in the Netherlands’ (2001) 23 Law & Policy 47 suggests that ombudsmen have more policy impact than administrative courts, but the empirical evidence on both courts and ombudsmen is sparse. 65 EO, ibid 3. 66 OI/6/2014 (Expert Groups), above n 52 follows up on the commitment to ‘keep a watchful eye on the situation’ in 1682/2010, above n 17 at para 193. 67 A Benz, C Harlow and Y Papadopoulos, ‘Introduction’ (2007) 13 European Law Journal 441.

Accountability and the European Ombudsman

227

the EO, like most ombudsmen, has no formal coercive p owers. The party being held to account does however ‘face consequences’:68 informal and indirect sanctions abound, for example through publicity69 and reports to the European Parliament,70 which in turn has formal sanctions at its disposal, including its role in the EU budget process.71 The EP’s representative democratic credentials are not straightforward, but nevertheless this speaks to a relatively familiar form of democratic accountability. Third, lawyers are often especially impressed by the EO’s procedural openness, relative to the Court. Rules on standing are notoriously restrictive at the EU level, so that judicial review of EU acts by environmental or other public interest groups is rare, and judicial review by economic interests more likely.72 The EO, by stark contrast, can receive complaints from ‘any citizen of the Union or any natural or legal person residing or having its registered office in a Member State’.73 This provides an additional route for the construction of an accountability relationship, with the complainant as well as the EO. Even if access is formally equal, we might expect powerful, well-resourced interests to make disproportionate use of any way of challenging unwelcome developments.74 The relative speed and informality of the EO, as well as its independent investigative capacity, may contribute to evening out practical access.75 As well as a particular approach to standing, the Court reviews only ‘acts’ that are ‘intended to produce legal effects’.76 Whilst the Court in principle looks to form rather than substance,77 this leaves an enormous amount of ‘non-binding’ guidance, opinions and advice in an uncertain position. This non-binding material addresses a range of questions, from the safety of a product, to the meaning of 68

Bovens, above n 3. Harlow and Rawlings, above n 11, are concerned that including informal responses like publicity in notions of ‘sanction’ can collapse the criterion of ‘sanctioning’. 70 The EO describes its reports to Parliament in individual inquiries as ‘the most powerful tool’ at its disposal, EO, Annual Report 2012 at 35. 71 The Parliament has twice (2012, 2014) voted to withhold part of the Commission’s budget in a disagreement over expert groups; in 2012 it withheld the budget of agencies, including EFSA, referring to an EO decisions, in part because of concerns over conflict of interests (A7-0106/2012), and citing an EO decision. A search of the European Parliament’s Register of Documents suggests that EO reports frequently feature in written questions. 72 M Lee, ‘Access to Justice at EU Level in Environmental Law’ (2012) papers.ssrn.com/sol3/papers. cfm?abstract_id=2062252. 73 TFEU, art 228. 74 The EO’s data on the source of complaints in 2013 simply tells us that 77.1% came from individual citizens, 22.9% from ‘companies, associations and other legal entities’, EO, Annual Report 2013; the Petitions Commission of the European Parliament has asked for better data on complaints, A8-0058/2014. 75 S Gilad, ‘Why the Haves do not Necessarily Come out Ahead in Informal Dispute Resolution’ (2010) 32 Law & Policy 283 (although Gilad is especially concerned by the absence of precedent in cases where individuals challenge decisions made about them, whilst I am concerned with broader accountability of governance). 76 TFEU, art 263. 77 See J Scott, ‘In Legal Limbo: Post-Legislative Guidance as a Challenge for European Administrative Law’ (2011) 48 Common Market Law Review 329. 69

228

Maria Lee

legislation, to acceptable methodologies for testing. Even if expressly not legally binding, choices on these matters by powerful actors shape future findings of fact, including facts about the law, and the final, legally binding decision.78 The EO is limited neither to reviewing ‘acts’ (it often addresses the general constitution and behaviour of groups within the governance system), nor to a consideration of legal effects. The limitations placed on the Court and the EO do overlap in one respect: Article 228 (like 263) applies to ‘the activities of the Union institutions, bodies, offices or agencies’.79 Whilst this captures much of the EU governance landscape, there may be challenges if, for example, an ‘act’ is not formally authored at EU level.80 Again, the fact that the EO is not constrained to particular ‘acts’ enlarges its scope of action.81 The limits on the ability of the EO to hear a case are relatively light (although not non-existent).82 Its procedural advantages can be seen in its handling of a complaint about the Air Passenger Rights Regulation.83 The Regulation provides a right to compensation if a flight is cancelled, unless the cancellation is caused by ‘extraordinary circumstances’. The Commission put a list of ‘extraordinary circumstances’ (agreed by most National Enforcement Bodies) on its website. The publication of this list is suggestive of the ways in which authority and facts are casually co-produced between national bodies and the Commission, and between technical (legal) expertise and political judgment; and the EO in turn plays its own role in reinforcing the authority of these collaborations. The complainant alleged that the list was incompatible with the Regulation. In judicial review, the General Court would have first considered the standing of the complainant, a firm of solicitors, and rejected the challenge at that stage. In the unlikely event that another suitable litigant could be found, the ambiguous legal effects and authorship of the document would have been problematic. By contrast, the EO is not concerned with the formal status of the document. It acknowledges that publication on the Commission website ‘added credibility and authority’ to the list.84 However, substantively, it seems to be satisfied that the document’s status would

78 E Korkea-aho, ‘Laws in Progress? Reconceptualizing Accountability Strategies in the Era of Framework Norms’ (2013) 2 Transnational Environmental Law 363; T Hervey, ‘“Adjudicating in the Shadow of the Informal Settlement?”: The Court of Justice of the European Union, “New Governance” and Social Welfare’ (2010) 63 Current Legal Problems 92. 79 The EO cannot hear a complaint against Court acting in its judicial role, art 228. The EO is subject to judicial review, C-234/02P European Ombudsman v Lamberts [2004] ECR I-2803. 80 Scott, above n 77; Decision in complaint 1581/2013 against the Commission (passenger rights), discussed below. 81 And note the European network of ombudsmen, above n 58. 82 EO Statute, above n 57. Eg Decision in complaint 1892/2012 against the European Commission (renewables) is indicative: some of the complaints missed the limitation period of two years; some were not first raised with the challenged body; and the EO cannot hear complaints against the ‘merits of EU legislation’. 83 Reg 261/2004 establishing common rules on compensation and assistance to passengers in the event of denied boarding and of cancellation or long delay of flights [2004] OJ L 46/1. 84 1581/2013 (passenger rights), above n 80 at para 8.

Accountability and the European Ombudsman

229

be properly reflected by its description as a ‘draft’, and a disclaimer in terms of it being for information and guidance only, and not adopted by the Commission.85 Such disclaimers are routine in the EU, and do not fully grapple with a document’s authority.86 The EO also however insists that ‘good administrative practice requires the Commission to ensure that the … list is compatible’ with the Regulation, a question of legal interpretation to which I return below.87 The concept of maladministration allows the EO to look beyond strict legality. The current Ombudsman said in a report released early in her appointment that she had been ‘struck by the extent to which EU institutions respond to complaints primarily in terms of the legality of their actions’. Whilst this is often reasonable, ‘in some other cases, it is almost as if the law is being used to limit the options’, resonating with the possibility that legitimacy is sought in apparently ‘technical’ (legal) reasons for decisions.88 The EO by contrast ‘will expect an institution to do whatever is possible within the law in order to achieve outcomes which are fair and reasonable in all the circumstances’.89 The broad approach to ‘maladministration’, defined by the EO as something that ‘occurs when a public body fails to act in accordance with a rule or principle which is binding upon it’,90 is developed in part through the cases, but also as suggested above, through the EO’s Code of Good Administrative Behaviour (said to make ‘the principle of good administration more concrete’),91 and its summaries of decisions. Sometimes, the EO uses a language of self-denial that is familiar to those of us more used to reading judicial decisions.92 Its approach to access to documents often looks a little legalistic (and it is bound by the access to documents regulation),93 but supplemented with broader advice on enhancing access, such

85 ibid paras 9, 27. Compare this with the EO’s approach to a different sort of ‘disclaimer’, that a report was co-authored by the President of the European Central Bank (ECB) in ‘his personal capacity’: ‘the ECB could not reasonably expect citizens and other stakeholders to regard such a statement as credible if the subject-matter of a report related to the areas of responsibility of the ECB … all statements by members of the decision-making bodies of the ECB relating to the ECB’s areas of responsibility, and all actions by members of the decision-making bodies of the ECB relating to the areas of responsibility of the ECB, will have an impact, in the eyes of EU citizens, and of other stakeholders, on how the ECB is perceived’, Decision in complaint 1339/2012 against the ECB, para 77. 86 Korkea-aho, above n 78; Scott, above n 77. 87 1581/2013 (passenger rights), above n 80 at para 12. 88 de Leeuw, above n 54. 89 EO, above n 55 at 2. 90 EO, Annual Report 2012 at 13. 91 www.ombudsman.europa.eu/en/resources/code.faces#/page/1; see also the Five Public Service Principles. The Code is rarely directly cited in the cases discussed here, although seems to be referred to more often in individual (eg employment/contract) complaints; for an argument that it has had limited impact, see P Leino, ‘Efficiency, Citizens and Administrative Culture: The Politics of Good Administration in the EU’ (2014) 20 European Public Law 681. 92 Decision in complaint 364/2013 against the European Medical Agency (EMA) (migraine medication) looked to ‘whether a procedural error has occurred or whether there is a manifest error in the reasoning of the contested decision’, para 38. 93 Reg 1049/2001/EC of 30 May 2001 regarding public access to European Parliament, Council and Commission documents [2001] OJ L 145/43; see EO Statute, above n 57 at art 4.

230

Maria Lee

as encouraging the development of improved archives or of mechanisms to help applicants identify the documents they need.94 But in particular, the EO focuses primarily on procedure rather than substance, as will be seen in more detail in the discussion of the decisions below.95 The EO does not ‘reassess technical or scientific evaluations’,96 ‘assess the outcome of the work of expert groups’,97 or ‘[settle] scientific disputes between EU agencies and complainants’.98 Investigations turn easily into an examination of the way the Commission dealt with the initial complaint.99 The procedural focus extends to the interpretation of legal requirements. The EO does not review the legality of legislation: a ‘measure of general application … must be presumed to be valid unless and until annulled by the Court of Justice’.100 Similarly, only the Court can provide a binding interpretation of legislation, but in the absence of such a binding interpretation, the EU institution is ‘entitled to develop its own rules of interpretation’, and the EO would consider ‘whether the interpretation was correct and reasonable’.101 In another case, the EO required the complainant to demonstrate that the Commission’s interpretation of legislation was ‘manifestly unreasonable or incorrect’.102 But the EO focused on the reasonableness of the interpretation, and on the process (including consultation) that was followed in reaching it.103 The elusiveness of any single ‘correct’ answer to legal interpretation is clearly recognised, if not explored; the EO’s approval arguably contributes to the authority (perhaps the construction of objectivity) of the ‘reasonable’ interpretation, in practice and probably judicially. Avoiding contested substantive judgments in favour of process is a common tactic, and may contribute to the EO’s own legitimacy in the eyes of those with whom it interacts.104 94 Decision in complaint 1877/2010 against the EMA; Decision in complaint 693/2011 against the EMA, para 34. Note the ‘serious concerns’ about the EMA’s inability to retrieve documents: ‘The keeping of adequate records constitutes a principle of good administration which helps to ensure both effectiveness and accountability’, at para 39. 95 This observation may not apply in eg employment or contract cases. 96 Draft Recommendations in complaint 1171/2013 against the European Aviation Safety Agency (EASA) (flight times), para 25. 97 EO, above n 55 at 15. See also eg Decision in complaint 51/2011 against the EASA, para 40. Nor does the EO substitute its own choice of members of committees, Decision in complaints 1874/2011 and 1877/2011 against the European Insurance and Occupational Pensions Authority (EIOPA), para 19. 98 Decision in complaint 1301/2013 against the ECHA, para 30. 99 eg Decision in complaint 2202/2012 against the European Commission (Ryanair). 100 Decision in complaint 1047/2013 against the European Commission, para 12. T-294/03 Gibault v Commission [2005] ECR-SC II-635: ‘the Ombudsman is empowered only to investigate and give his views in cases of maladministration, which cannot include infringement of a legal provision or of a general principle amenable to review by the Community judicature’, at para 45. 101 Decision in complaint 1826/2010 against the ECHA, para 44, emphasis added. 102 1892/2012 (renewables), above n 82 at para 34. 103 1581/2013 (passenger rights), above n 80. The EO does seem to take a more substantive approach to interpretation of Reg 1901/2006 of 12 December 2006 on medicinal products for paediatric use [2006] OJ L378/1 in Decision in complaint 2575/2009 again the EMA, see the discussion in the first half of the decision and the EMA response beginning at para 160. 104 Although accountability to parliaments may turn on the merits, Black above n 2. In 1892/2012 (renewables), above n 81, the EO suggested that the ‘merits of EU legislation’ are more suitable for the European Parliament’s Petitions Committee.

Accountability and the European Ombudsman

231

IV. Knowledge and Co-production in EO Decisions We might hope that the EO would be able to look through the EU’s complex governance arrangements to require an accounting from otherwise obscure parts of the process, and beyond the ‘science’ of risk assessment and the ‘politics’ of the final outcome to take a more sophisticated and realistic picture of authority. Not surprisingly, the EO does not explicitly discuss ‘co-production’. It does however scrutinise the various groups providing scientific or technical advice and expertise in the EU, and this is a good place to start. It is difficult to draw clear or consistent conclusions on the EO’s approach to and understanding of coproduced authority, but this section tentatively organises the material along two dimensions, of objectivity and boundary drawing. In thinking about objectivity, I rely on Jasanoff ’s description of objectivity as a ‘highly sought-after and hard-won epistemic achievement’.105 This does not require an unrealistic belief that neutral, universally applicable technical opinions are achievable, but engages with the idea that claims to objectivity are crucial in establishing the stability and authority of facts. In turn, this means that objectivity is crucial to the ways in which facts are used to claim a legitimate basis for the outputs of both knowledge generators and policy makers. The holding to account, and construction, of objectivity by the EU in the ways discussed below, speaks to co-production in a number of ways. Most importantly, it has the potential to recognise the constructed nature of facts, the role of political actors in constructing those facts, and then the dependence of political legitimacy on objectivity of the facts. Turning to institutional boundary drawing, I am concerned with the ways in which EU governance arrangements attempt to maintain clear lines between both actors and concepts: science and politics, EU and national, private and public. These lines are not natural, but are effortfully drawn and maintained.106 If it is blind to some of these boundaries, the EO will be able to forge accountability relationships with any part, or many parts, of a decision-making process, and in so doing to explore and challenge attempts to create space between them, in turn creating the potential to see into their mutual dependence. The observations in the following sub-sections reinforce both concerns about the difficulties of holding the EU’s dense thicket of administration to account, and the potential of the EO in this respect.

105 S Jasanoff, ‘The Practices of Objectivity in Regulatory Science’ in C Camic, N Gross and M Lamont (eds), Social Knowledge in the Making (Chicago, University of Chicago Press, 2011). 106 A Irwin, ‘STS Perspectives on Scientific Governance’ in EJ Hackett et al (eds), The Handbook of Science and Technology Studies (Cambridge MA, MIT Press, 2008).

232

Maria Lee

A. The Construction of Objectivity in the EU The ‘objectivity’ of expertise is directly or indirectly challenged in a number of the cases. The EO often requires an account to be provided of objectivity, and in some cases implicitly or explicitly rejects institutional claims of objectivity by a finding of maladministration. But importantly, the EO does not simply hold claims of objectivity to account. Objectivity is constructed and requires ongoing work, both by the generator of the facts, and the user of those facts.107 Even in the cases where it rejects the institution’s insistence that all is well, the EO reinforces the work being done to achieve objectivity, opening a route to authoritative fact-finding, in particular through better record-keeping or reason-giving, more inclusive or better rationalised procedures. For the avoidance of doubt, my argument is not that these processes always work in terms of either achieving neutrality, or achieving acceptance; contestation continues, as indeed it should. But we do see the EO both holding objectivity claims to account, and contributing to the achievement of objectivity. In this latter respect, the EO’s work on ‘balance’ is suggestive of the ways in which the EU may be constructing a peculiarly ‘European’ approach to the work of achieving objectivity and legitimacy,108 particularly in its response to perennial concern that economic interests are disproportionately present or influential in EU governance, and to questions of geographical balance. The challenge posed by the ubiquity of economic actors may not seem distinctively ‘European’, but it has been a dramatically visible ‘EU’ issue, including in the EO’s ‘docket’, for example in numerous complaints about conflicts of interest.109 Although the detailed legislative definition of the ‘balance’ provides the crucial context,110 the EO’s decisions in a number of cases about the stakeholder groups111 providing input into financial services regulation clearly indicate the many ways in which economic actors might come to dominate in EU governance.112 First,

107 ibid.

108 Of course, the EO, which is operating within a particular legislative and policy context, and may not be the most important place to look for this work on constructing EU objectivity. But some interesting issues are raised. Note that cases on ‘balance’ (or cases about the effectiveness of consultation more generally) do not often find their way before the Court. 109 The EO emphasises the potential for influence, rather than requiring evidence that influence has actually been exercised, often relying on the OECD definition of ‘conflict of interest’, eg Decision in complaint 297/2013 against the European Commission (ad hoc ethical committee), para 51; EO, above n 55 at 20. The EO’s responses in this area emphasise the importance of having proper procedures in place. 110 Reg 1094/2010/EU of 24 November 2010 establishing a European Supervisory Authority (European Insurance and Occupational Pensions Authority) [2010] OJ L 331/48. 111 So not expert groups. 112 1874/2011 (EIOPA), above n 97, in which the EO refers to some similar cases about almost identical legislative provisions (the UNI I and UNI II decisions, para 20); Decision in complaint 1966/2011 against the European Banking Authority (EBA); Decision in complaint 1875/2011 against the EBA, and cases 1876/2011 and 1321/2011.

Accountability and the European Ombudsman

233

the EO held that the mere fact that an individual is a consumer of financial services is not ‘an adequate and valid justification’ for appointment to the stakeholder group as a consumer (!); the question is whether the individual ‘is able to act as an objective and dedicated consumer representative’.113 Second, the EO had to clarify that it was ‘not acceptable for profit-making suppliers of remunerated services’ (lawyers, accountants, auditors, actuaries and analysts) to be included in the category of ‘users’ of financial services.114 And third, an ‘employers’ representative’ had been appointed; the EO pointed out that employers were already included in the ‘industry’ representation, and also found that there was nothing in the legislation to suggest that its list of interests was indicative only. Similarly prompted (in part) by the place of economic interests in EU governance, the EO has begun an OII Concerning the Composition of Commission Expert Groups. In its letter to the Commission containing initial proposals, the EO proposes a ‘legally binding’ framework in which the ‘balanced representation of all relevant interests’ would be a ‘mandatory requirement’, rather than, as currently, required ‘as far as possible’.115 Recognising the complexity of the notion of balance, the EO does not define or require a single approach.116 It proposes instead that ‘an individual definition of “balance”’, including the appropriate ratio of economic and non-economic interests, be published for each group. General criteria should be developed for working that out case by case. This may sound rather bland, but an obligation to articulate the meaning of ‘balance’ in any particular case could be a major step, providing criteria for assessing and challenging balance.117 In another case, it was alleged that the ‘European Biofuels Technology Platform’ was dominated by commercial interests, resulting in ‘one-sided advice’ to the Commission on biofuels.118 The EO confirmed the legitimacy of what the Commission called the ‘deliberate industrial focus of technology platforms’, and the possibility that an overall balance could be achieved by referring to other sources of advice. However, whilst the Commission enjoys ‘wide discretion in deciding how to achieve the necessary overall balance’,119 it cannot simply assert that ‘the sheer quantity of procedures, and the mass of input’ makes all well.120 Given the concern that the Platform was a ‘privileged interlocutor’, the Commission should ensure that others had a ‘genuine opportunity’ to express their views, with an ‘expectation

113 1874/2011 (EIOPA), ibid para 39. In this case, the ‘consumer’ was an academic, but as the EO points out, industry representatives also consume products, ibid. 114 1874/2011 (EIOPA), ibid para 50. 115 Letter to the European Commission (27 January 2015), OI/6/2014. Commission, above n 16. 116 The EO has agreed with the Commission that ‘general criteria’ are not helpful, 1682/2010, above n 17 at para 113. 117 The Commission has rejected this proposal, Opinion of the European Commission in OI/6/2014. The EO will respond to this Opinion. 118 1151/2008 (biofuels), above n 46 at para 6. 119 ibid para 33. See also 1682/2010 above n 17. 120 ibid para 36.

234

Maria Lee

that these views will be taken into account’.121 The Commission should have been able to explain to the complainants the weight given to various inputs.122 So the EO seems to contribute a route to the construction of objectivity by making particular procedural demands in respect of the embedding of economic actors in the generation of EU knowledge. Equally pertinently, the issue of g eographical balance points to a particular vision of European objectivity in the mutual shaping of science and Europe. The financial services legislation discussed above requires that ‘to the extent possible’ there should be ‘an appropriate geographical … balance’.123 In response to a complaint about an alleged failure to include any industry representatives from a ‘new’ Member State in one group, the EO concluded that the Agency had not provided good reasons for limited geographical representation. In particular, there was nothing in the legislation to suggest that applicants should be discounted on the basis of ‘limited English language skills’ or ‘exclusive national professional focus’.124 Of different possible ‘practices of knowledge making’, when policy or legislation is concerned by geographical balance,125 we see ‘common knowledge’ being created through ‘group reasoning explicitly based on principles of political representation’, with an emphasis on the representation of different types (new and old) of Member State.126 The approach to geographical balance might be contrasted with the EO’s discomfort at being asked to assess the balance between Christian and secular groups on the European Group on Ethics in Science (EGE).127 The EO sidestepped the substance of the complaint about over representation of ‘the Christian world’, satisfied that members were required not to accept external instructions, for example from their church. The ‘pluralism’ of the EGE is judged in terms of ‘geographical origin, gender, age, as well as knowledge and expertise’; not in terms of religion. The EO agreed with the Commission that it would be legally ‘highly questionable’ to exclude anyone from the EGE simply because of allegiance to a religious group,128 but seemed unconcerned that ensuring geographical balance would lead to some being excluded simply for their (over-represented) nationality.129 Whilst we generally see procedural routes to objectivity, the EO does not speak with a single voice. In the biofuels case, the EO took what looks like a more substantive approach to ‘objectivity’: Objectivity … raises more concrete issues regarding the technical content and quality of the output, and the basis on which it is formulated. It raises specific questions as to 121

ibid paras 81, 91. ibid para 36. 123 1874/2011 (EIOPA), above n 97 at para 6. 124 1874/2011 (EIOPA), ibid para 23. 125 Again, it should be noted that the EO is always working in a particular legislative and policy context. We are more accustomed to seeing explicit references to national/geographical representation in eg Agency and other committees. 126 Jasanoff, above n 105. 127 Decision in complaint 203/2013 against the European Commission (EGE), para 19. 128 ibid paras 32–33. 129 1874/2011 (EIOPA), above n 97. 122

Accountability and the European Ombudsman

235

whether the output is, or can reasonably be expected to be, factually well-founded and in line with informed and expert opinion.130

The Commission needs to provide information on the ‘mechanisms … to ensure the factual objectivity of the Platform’s input, namely, the objectivity of its advice/recommendations’.131 Even here, however, with this ostensibly substantive approach, other than requiring arrangements that ensure that the Platform ‘[takes] into account all relevant considerations’ (a procedural requirement), the EO did not elaborate on what this means in practice.132

B. Institutional Boundary Drawing In many cases, including for example the passenger rights case discussed above,133 the EO simply ignores the EU’s tenacious institutional boundaries (such as politics/science, EU/national, public/private), in a way that enhances its capacity to hold muddled authority to account. Perhaps centrally for the purposes of co-production, the EO is generally emphatic that the ‘technical’ nature of an opinion or process in no way isolates that opinion or process from politics. For example, ‘a choice in the field of research policy—however technical in nature or narrow in scope—cannot be dissociated from numerous other environmental, social and economic considerations’.134 Similarly, whilst the EO does not engage directly with a Commission argument that a discussion ‘limited to legal issues’ could not amount to ‘lobbying’, its conclusion that ‘private interests’ were being represented in the relevant meeting seems to reject the proposition that legal argument is ‘neutral’.135 Nor is it acceptable simply to assume that particular actors, such as academics136 or those in an Agency’s secretariat,137 raise no conflict of interest issues: ‘an examination of the specific relevant facts’138 or a ‘thorough assessment’ of tasks139 is necessary in every case. More generally, although the required degree of ‘representation’ versus ‘independence’ varies according to the particular legislative or policy context, the EO’s concern with the identification and ‘balance’ of participants implies a fundamental acceptance that science is not wholly autonomous of interests, or inevitable in the sense that anyone with access to the data would reach the same conclusion.

130

1151/2008 (biofuels), above n 46 at para 29, emphasis added. ibid para 35. 132 ibid para 77. Consultation by the Platform, and the inclusion of one NGO in the Platform was not a sufficient response at para 82. 133 Above n 80. 134 EO, above n 55, at 15–16. See also 1151/2008 (biofuels), above n 46 at para 77. 135 297/2013 (ad hoc ethical committee), above n 109 at para 14. 136 Decision in complaint 346/2013 against EFSA (conflict of interests). 137 Draft Recommendations in complaint 775/2010 against EFSA (revolving doors). 138 346/2013 (conflict of interests), above n 136 at para 12. 139 775/2010 (revolving doors), above n 137 at paras 65, 63. 131

236

Maria Lee

The EO’s focus on economic interests, discussed above, could be bolder, but it at least implicitly recognises the role of interests, if not of power, in the production of knowledge and expertise, and the impossibility of separating these categories.140 The current Ombudsman, Emily O’Reilly, has indirectly acknowledged that the focus on ‘lobbying’ as an activity that can be isolated and so dealt with, may be unhelpful.141 Private influence is a pervasive fact of EU governance, and she recognises that officials may not always be aware that they are being lobbied. She does not use the term epistemic communities, but she talks about elites shaping the debate in terms of knowledge, and about the role of industry in that process.142 Along similar lines, a decision on the Commission’s ad hoc ethical committee did not engage with whether someone is employed as a ‘lobbyist’, but confirmed that the person involved ‘represents private interests’ when contacting EU institutions,143 looking to substance rather than form, enhancing the ability of the EO to scrutinise activities. The ways in which the EO splits up and reassembles decision-making processes in its forging of accountability relationships, recognising expressly the mutual dependence of different parts of the process, is also relevant to the boundary between technical and political responsibilities. For example, the mere fact that advice is not binding does not allow the Commission to avoid accountability for that advice: ‘the complainant’s point was not that the Platform purports somehow to take over the Commission’s decision-making … but that its guidance influenced that decision-making’.144 Similarly, the Commission ‘is not exonerated from the obligation to ensure objectivity by the fact that it is not bound to follow the recommendations made by a particular committee or group’.145 Being part of ‘a wider rulemaking process’ will not necessarily ‘mitigate any potential conflict of interest in the rulemaking group’, if a particular source of advice is central to the decision.146 This wider process may however be important. A case about ‘containing’ conflicts of interest in a scheme to develop parts of the River Danube with EU funding, emphasises the totality of the process.147 A study was steered by an engineering company with a financial interest in the outcome, leading to a potential conflict of interest. But the Commission had taken appropriate action: a Monitoring Group had not said anything to raise concerns; a ‘regional conference’

140

Stirling, above n 31. Is Brussels the new Washington, D.C.? Lobbying transparency in the EU, European Ombudsman Opening Address, 11 May 2015. 142 297/2013 (ad hoc ethical committee), above n 109 at para 56. 143 ibid para 54. Elsewhere, the EO does define lobbying, EO, above n 55 at 21, citing 1339/2012, above n 85. 144 1151/2008 (biofuels), above n 46 at para 31. 145 EO, above n 55, 16; 1151/2008 (biofuels), ibid paras 78, 30. Also 775/2010 (revolving doors), above n 137. 146 1171/2013 (flight times), above n 96 at paras 19, 22; similar language is used in Draft Recommendation of the EO in her Inquiry into Complaint 726/2012/FOR against the EASA, para 33. 147 Decision in complaint 2265/2011 against the European Commission. Also 1151/2008 (biofuels), above n 46 and text at n 119, although the Commission had not explained adequately. 141

Accountability and the European Ombudsman

237

provided for the inclusion of ‘a wider circle of interested parties … reducing the likelihood of a single interest dominating the project’; and it was ‘relevant’ that ‘the Commission closely followed the entire process’.148 Notwithstanding its powerful potential to disrupt institutional boundary drawing, the EO is nevertheless often happy to reinforce lines between science and politics.149 In a revealing response to a complaint, the ECHA argued that disclosing the positions of the Member States on animal testing proposals would mean a ‘shift from decision-making based on the efficient provision of objective scientific and technical advice to decision-making based on policy considerations’. As well as the familiar and plausible concern that unanimous decision-making would be more difficult and the quality (perhaps read sincerity) of debate would be reduced, the ECHA was concerned that disclosure would ‘lead to subjective decisions’.150 The EO avoided commenting explicitly on the ECHA’s line drawing between science and subjectivity,151 but may have been reflecting the ECHA dichotomy when noting that ‘pressure’ from third parties and registrants ‘on science alone’ is ‘entirely legitimate and useful pressure’.152 Perhaps the most frustrating example of the EO uncritically adopting institutional boundaries is found in its OII into expert groups. The very framing of the inquiry is dependent on the institutional category of ‘expert group’, rather than on (for example) ‘knowledge generating activities’ or something similarly inclusive. Whilst we might expect some ‘good practice’ spillover into other areas, the EO’s approach is obviously limited by prior and problematic categories drawn up by the very institutions being scrutinised. Explaining the inconsistencies in the EO’s approach is difficult, and we might more positively frame this as an observation that the EO does not insist on any single model of accountability. But the phenomena of objectivity and boundary drawing are not the conscious focus of the EO’s attention, and although the EO is building a framework for good administration, the absence of any formal system of precedent, and an associated ability to respond to the fairness of any facts before it, is part of the power of ombudsmen institutions. Further, whilst we speak of ‘the’ ombudsman, and tend to personalise the decisions (although I have not here),153 the EO is an institution, with a multi-lingual staff, and four ‘Complaints and inquiries units’.154 This glancing interest in co-production, fact-specific decision-making, and institutional diversity means that differences in 148

2265/2011 (Danube), ibid paras 61–62. eg in its resistance to substantive judgments, above nn 96–98. 150 Draft recommendation in complaint 2186/2012 against the ECHA (animal testing) para 24. 151 Concluding, in words familiar from the Court, that the ECHA had not demonstrated that ‘undue pressure on decision-makers is reasonably foreseeable, and not purely hypothetical’, nor that this pressure ‘would be of such a nature and intensity as to undermine seriously the decision-making process’, ibid para 47. 152 ibid paras 56, 58. Although note the wording of REACH, above n 19, which provides that only ‘scientifically valid information and studies’, rather than ethical concerns, must be taken into account, at art 40(2). 153 See above n 8. 154 www.ombudsman.europa.eu/en/atyourservice/team.faces. 149

238

Maria Lee

approach are to be expected. And importantly for current purposes, like the rest of us, the EO has to work with the institutions it finds. Sometimes it can be difficult to engage with, even criticise, risk regulation as it operates without reinforcing its conceptual limitations.

V. Conclusions The EO can hardly be called an unsophisticated observer of EU administration, or indeed of the co-production of science and EU society, and there are many cases in which it simply walks straight through the boundaries constructed for the convenience of governing, and engages quite directly (if implicitly) with the reality of co-produced authority, exposing the mutual dependence of different strands of the EU governance framework. This has the potential to provide a far reaching and distinctive sort of accountability. In other cases, however, the EO follows and reinforces, rather than challenges, the institutionalisation of boundaries. At best, this might strengthen more familiar (limited in this context, although not powerless) legal and political forms of accountability. Whilst this chapter provides a preliminary examination of the EO’s role in co-production, and confident conclusions on what it offers to accountability would be premature, we can see that the EO does not insist on any single model of accountability. The EO puts considerable faith in procedural mechanisms, taking a particularly striking approach to reason-giving, requiring not just reasons for an outcome, but also for the process that has been followed.155 For example, whilst the institutions ‘necessarily have a margin of discretion’ in respect of ‘the precise manner by which participatory democracy is made effective in any given circumstance’, they should be able to ‘justify objectively how they exercise that margin of discretion’.156 The Commission cannot merely assert that subsequent broad consultation mitigates concerns about industry dominance in one particular part of the process, but must articulate how it does so.157 Similarly, in its initial proposals in the OII on expert groups the EO addresses an obligation of reason-giving to the way in which balance has been understood in any particular case.158 These are potentially crucial contributions to accountability, since the obligation to explain159 provides

155 This seems to be the sort of thing that the authors have in mind in J Scott and S Sturm, ‘Courts as Catalysts: Rethinking the Judicial Role in New Governance’ (2007) 13 Columbia Journal of European Law 565, see also Vos in this volume. Also of course, the EO requires more routine processes of eg consultation and transparency. 156 EO, above n 55 at 13. 157 eg discussion of 1151/2008 (biofuels) in text at n 119. 158 EO, ibid 16. The Commission has rejected this proposal, above n 117. Also Draft Recommendation in Complaint 2558/2009 against the European Commission (non-human primates) at paras 31–32. 159 See more generally Sabel and Zeitlin, above n 41.

Accountability and the European Ombudsman

239

not only an accountability standard against which to assess individual cases, but may also allow for broader deliberation on the articulation of that very standard. The EO is able to address the diversity of fora and processes for the construction of knowledge and executive authority in the EU, and to hold different combinations of actors to account. The information generated or revealed by the EO investigation can provide important resources for others seeking to hold public bodies to account, perhaps in other fora, including the European Parliament. Further, the EO’s decisions and reports are public, providing ‘avenues for political contestation and scrutiny of governance procedures’160 even beyond parliament. The EO’s structural role may also contribute to the development of governance values, such as transparency and reason-giving, which shape the space for accountability more generally. What a ‘successful’ accountability framework for the EU administration would look like is far from clear, not least because of the diversity of processes, but also because this question is fundamentally associated with the sort of governance and democracy to which we aspire.161 The EO’s openness has the capacity to recognise the challenges of EU democracy. But accountability is at least as complicated, and at least as contestable, as the thing(s) being held to account. If the EO provides only a partial and flawed response to the accountability of co-produced authority, it shares that characteristic with other accountability fora (political, judicial and peer). Exploring this particular forum, in respect of this particular accountability conundrum, is part of the continual revisiting and striving for accountability in the EU. Moreover, it is important that the EO receives the scrutiny of any institution that purports to speak with authority on administration in the EU.

160 161

Dawson, above n 11 at 393. Above n 2.

240

BIBLIOGRAPHY Abbot, C and Lee, M, ‘Economic Actors in EU Environmental Law’ (2015) 34 Yearbook of European Law 26 Adam, B, Beck, U, Van Loon, J (eds), The Risk Society and Beyond: Critical Issues for Social Theory (London, Sage Publications, 2000) Advisory Committee on Releases to the Environment (ACRE), ACRE Advice: New Techniques Used in Plant Breeding (London, Advisory Committee on Releases to the Environment, 2013) Anderson, B, Imagined Communities (New York, Verso, 1983) Anderson, C, ‘Contrasting Models of EU Administration in Judicial Review of Risk Regulation’ (2014) 51 Common Market Law Review 425 Anderson, J et al, ‘Engineering and Ethical Perspectives in Synthetic Biology’ (2012) 13(7) European Molecular Biology Organization Reports 584 Andrianantoandro, E et al, ‘Synthetic Biology: New Engineering Rules for an Emerging Discipline’ (2006) 2(1) Molecular Systems Biology 1 Annas, GJ and Mariner, WK, ‘(Public) Health and Human Rights in Practice’ (2015) 41 Journal of Health Politics, Policy and Law 129 Armstrong, K, ‘The Character of EU Law and Governance: From “Community Method” to New Modes of Governance’ (2011) 64 Current Legal Problems 179 ——, Regulation, Deregulation, Reregulation (London, Kogan Page, 2000) Arnull, A and Wincott, D (eds), Accountability and Legitimacy in the European Union (Oxford, Oxford University Press, 2002) Baekkeskov, E and Öberg, P, ‘Freezing Deliberation through Public Expert Advice’ (2016) 23 Journal of European Public Policy 1 Baert, P and Rubio, FD (eds), The Politics of Knowledge (Abingdon, Routledge, 2011) Baisnee, O and Pasquier, R (eds), L’europe Telle Qu’elle Se Fait (Paris, CNRS Editions, 2007) Baker, MG, Easther, S and Wilson, N, ‘A Surveillance Sector Review Applied to Infectious Diseases at a Country Level’ (2010) 10 BMC Public Health 332 Baldwin, P, Contagion and the State in Europe, 1830–1930 (Cambridge, Cambridge University Press, 2005) ——, Disease and Democracy: The Industrialized World Faces Aids (Berkeley and Los Angeles, University of California Press, 2005) Barnett, T and Sorenson, C, ‘Infectious Disease Surveillance in the United States and the United Kingdom: From Public Goods to the Challenges of New Technologies’ (2011) 36 Journal of Health Politics, Policy and Law 165 Bashford, A (ed), Medicine at the Border: Disease, Globalization and Security, 1850 to the Present (Basingstoke, Palgrave Macmillan, 2006) Bauer-Panskus, A and Then, C, Testbiotech opinion on the application for market approval of genetically modified maize 1507 (DAS-Ø15Ø7-1) (Munich, Testbiotech, 2010) Beck, U, Risk Society: Towards a New Modernity (London, Sage Publications, 1992)

242

Bibliography

——, ‘The Terrorist Threat World Risk Society Revisited’ (2002) 19 Theory, Culture & Society 39 ——, World at Risk (Cambridge, Polity, 2007) ——, Giddens, A and Lash, S, Reflexive Modernization (Stanford, Stanford University Press, 1994) Beetham, D and Lord, C, Legitimacy and the European Union (London, Longmans, 1998) Benner, SA and Sismour, AM, ‘Synthetic Biology’ (2005) 6 Nature Reviews Genetics 533 Bennett Moses, L, ‘Recurring Dilemmas: The Law’s Race to Keep Up With Technological Change’ (2007) 2 Journal of Law, Technology and Policy 239 Benz, A, Harlow, C and Papadopoulos, Y, ‘Introduction’ (2007) 13 European Law Journal 441 Berg, M and Bowker, G, ‘The Multiple Bodies of the Medical Record’ (1997) 38 The Sociological Quarterly 513 Berg, P et al, ‘Summary Statement of the Asilomar Conference on Recombinant DNA Molecules’ (1975) 72(6) Proceedings of the National Academy of Sciences 1981 Bergkamp, L (ed), The European Union REACH Regulation for Chemicals: Law and Practice (Oxford, Oxford University Press, 2013) ——, ‘The Quiet Revolution in Administrative Procedure: Judicial Vetting of Precautionary Risk Assessment’ (2014) European Journal of Risk Regulation 102 Binder, S and Laing, AM, ‘The National Public Health Institutes of the World- Update’ (2009) 30 Journal of Public Health Policy 467 Binder, S and others, ‘National Public Health Institutes: Contributing to the Public Good’ (2008) 29 Journal of Public Health Policy 3 Bird, CE and others (eds), Handbook of Medical Sociology 6th edn (Nashville, Vanderbilt University Press, 2010) Birkhead, GS, Klompas, M and Shah, NR, ‘Uses of Electronic Health Records for Public Health Surveillance to Advance Public Health’ (2015) Annual Review of Public Health 345 Birkland, TA, Lessons of Disaster: Policy Change After Catastrophic Events (Washington DC, Georgetown University Press, 2006) Birkmann, J and others, ‘Extreme Events and Disasters: A Window of Opportunity for Change? Analysis of Organisational, Institutional and Political Changes, Formal and Informal Responses after Mega-Disasters’ (2008) 55 Natural Hazards 637 Black, J, ‘Calling Regulators to Account: Challenges, Capacities and Prospects’ LSE: Law, Society and Economy Working Papers 15/2012 ——, ‘Constructing and Contesting Legitimacy and Accountability in Polycentric Regulatory Regimes’ (2008) 2 Regulation & Governance 137 Bleeker, E et al, Interpretation and Implications of the European Commission Recommendation on the Definition of Nanomaterial (National Institute for Public Health and the Environment, the Netherlands, RIVM Letter Report 601358001, 2012) Blouin Genest, G, ‘World Health Organization and Disease Surveillance: Jeopardizing Global Public Health?’ (2014) 19 Health 595 Boin, A and others, The Politics of Crisis Management: Public Leadership Under Pressure (Cambridge, Cambridge University Press, 2006) Boullier, H and Laurent, B, ‘La précaution réglementaire: Un mode européen de gouvernement des objets techniques’ (2015) 3 Politique européenne 30 Bovens, M, ‘Analysing and Assessing Accountability: A Conceptual Framework’ (2007) 13 European Law Journal 447

Bibliography

243

Bowman, D, D’Silva, J and van Calster, G, ‘Defining Nanomaterials for the Purpose of Regulation within the European Union’ (2010) 2 European Journal of Risk Regulation 115 Bressman, LS, ‘Beyond Accountability: Arbitrariness and Legitimacy in the Administrative State’ (2003) 78 New York University Law Review 461 Briatte, F, The Europeanization of Health System Performance: The Eurocare Study and Cancer Control in England (Lille, CERAPS, 2010) Brookmeyer, R and Stroup, DF (eds), Monitoring the Health of Populations: Statistical Principles and Methods for Public Health Surveillance (Oxford, Oxford University Press, 2004) Brownsword, R and Goodwin, M, Law and the Technologies of the Twenty-First Century: Text and Materials (Cambridge, Cambridge University Press, 2012) Buck, T, Kirkham, R and Thompson, B, The Ombudsman Enterprise and Administrative Justice (Farnham, Ashgate, 2011) Buehler, JW and others, ‘Situational Uses of Syndromic Surveillance’ (2009) 7 Biosecur Bioterror 165 Busch, L, Standards: Recipes for Reality (Cambridge MA, MIT Press, 2011) Büscher, C and Japp, KP (eds), Ökologische Aufklärung (Wiesbaden, VS Verlag für Sozialwissenschaften, 2010) Bush, G, Public Papers of the Presidents of the United States: George Bush, 1990 (Washington DC, Government Printing Office, 1991) Butler, D, ‘When Google Got Flu Wrong’ (2013) 494 Nature 155 Calain, P, ‘Exploring the International Arena of Global Public Health Surveillance’ (2006) 22 Health Policy and Planning 2 Calhoun, C, ‘A World of Emergencies: Fear, Intervention, and the Limits of Cosmopolitan Order’ (2004) 41 Canadian Review of Sociology/Revue canadienne de sociologie 373 Camic, C, Gross, N, and Lamont, M (eds), Social Knowledge in the Making (Chicago, University of Chicago Press, 2011) Castillo-Salgado, C, ‘Trends and Directions of Global Public Health Surveillance’ (2010) 32(1) Epidemiologic Reviews 93 Champeua, S, Closa, C, Innerarity, D and Maduro, MP (eds), The Future of Europe: Democracy, Legitimacy and Justice After the Euro Crisis (London, Rowan & Littlefield International, 2015) Charnley, G and Elliott, ED, ‘Risk Versus Precaution: Environmental Law and Public Health Protection’ (2002) 32 Environmental Law Reporter 10363 Chilvers, J and Kearnes, M (eds), Remaking Participation: Science, Environment and Emergent Publics (Abingdon, Routledge, 2016) Choi, BCK, ‘The Past, Present, and Future of Public Health Surveillance’ (2012) 2012 Scientifica 875253 Cole, SA, ‘Forensic Culture as Epistemic Culture: The Sociology of Forensic Science’ (2013) 44 (1) Studies in History and Philosophy of Biological and Biomedical Sciences 36 —— and others, Truth Machine: The Contentious History of DNA Fingerprinting (Chicago, University of Chicago Press, 2008) Collins, H, Tacit and Explicit Knowledge (Chicago, University of Chicago Press, 2010) Committee on Smallpox Vaccination Program Implementation and others, The Smallpox Vaccination Program: Public Health in an Age of Terrorism (Washington DC, The National Academies Press, 2005)

244

Bibliography

Cook, S and others, ‘Assessing Google Flu Trends Performance in the United States During the 2009 Influenza Virus a (H1n1) Pandemic’ (2011) 6 PLoS One e23610 Corburn, J, ‘Community Knowledge in Environmental Health Science: Co-Producing Policy Expertise’ (2007) 10 Environmental Science and Policy 150, 159 Craig, P, EU Administrative Law 2nd edn (Oxford, Oxford University Press, 2012) ——, Public Law and Democracy in the United Kingdom and the United States of America (Oxford, Clarendon Press, 1990) ——, The Lisbon Treaty: Law Politics and Treaty Reform (Oxford, Oxford University Press, 2010) Curtin, D, ‘Challenging Executive Dominance in European Democracy’ (2014) 77 (1) Modern Law Review 1 ——, Executive Power of the European Union, Law, Practices and the Living Constitution (Oxford, Oxford University Press, 2009) ——, ‘Holding (Quasi-) Autonomous EU Administrative Actors to Public Account’ (2007) 13 European Law Journal 523 —— and Egeberg, M, ‘Tradition and Innovation: Europe’s Accumulated Executive Order’ (2008) 31 West European Politics 639 Daase, C, Geis, A and Nullmeier, F (eds), Der Aufstieg der Legitimitätspolitik. Rechtfertigung und Kritik politisch- ökonomischer Ordnungen (Baden-Baden, Nomos, 2012) Daemmrich, AA, Pharmacopolitics: Drug Regulation in the United States and Germany (Chapel Hill NC, University of North Carolina Press, 2004) Davies, ACL, Accountability: A Public Law Analysis of Government by Contract (Oxford, Oxford University Press, 2001) Davies, SE, ‘The Challenge to Know and Control: Disease Outbreak Surveillance and Alerts in China and India’ (2012) 7 Global Public Health 695 ——, Kamradt-Scott, A and Rushton, S, Disease Diplomacy: International Norms and Global Health Security (Baltimore, Johns Hopkins University Press, 2015) ——, Youde, J and Parker, R, ‘The Shared Responsibility of Disease Surveillance’ (2012) 7 Global Public Health 667 Dawson, M, ‘Transforming into What? New Governance in the EU and the “Managerial Sensibility” in Modern Law’ (2010) Wisconsin Law Review 389 ——, De Witte, B and Muir, E (eds), Judicial Activism at the European Court of Justice (Cheltenham, Edward Elgar, 2013) de Búrca, G and Scott, J (eds), Law and New Governance in the EU and the US (Oxford, Hart Publishing, 2006) de Leeuw, ME, ‘The European Ombudsman’s Role as a Developer of Norms of Good Administration’ (2011) 17 European Public Law 349 de Ruijter, A, The Expansion of EU Power in the Field of Human Health (Oxford, Oxford University Press, 2016) ——, Uncovering European Health Law (Amsterdam, University of Amsterdam, 2013) de Swaan, A, In Care of the State: Health Care, Education and Welfare in Europe and the USA in the Modern Era (Oxford, Oxford University Press, 1988) Department of Health, Getting Ahead of the Curve: A Strategy for Combating Infectious Diseases Including Other Aspects of Health Protection (London, Department of Health, 2002) Doremus, H, ‘Science Plays Defense: Natural Resource Management in the Bush Administration’ (2005) 32 Ecology Law Quarterly 249

Bibliography

245

Dorf, MC and Sabel, CF, ‘A Constitution of Democratic Experimentalism’ (1998) 98 (2) Columbia Law Review 267 Douglas, M, How Institutions Think (Syracuse NY, Syracuse University Press, 1986) Drewe, JA and others, ‘Evaluation of Animal and Public Health Surveillance Systems: A Systematic Review’ (2012) 140 Epidemiology and Infection 575 Edley, CF, Administrative Law: Rethinking Judicial Control of Bureaucracy (New Haven, Yale University Press, 1990) Egeberg, M, Gornitzka, Å and Trondal, J, ‘A Not So Technocratic Executive? Everyday Interaction between the European Parliament and the Commission’ (2014) 37 West European Politics 1 Ehnert, T, Regulating the Invisible: A Critical Analysis of the EU’s Approach to Nanotechnologies (Maastricht, Maastricht University, 2015) ——, ‘The Legitimacy of New Risk Governance: A Critical View in Light of the EU’s Approach to Nanotechnologies in Food’ (2015) 21 European Law Journal 44 Elliott, H, Jones, DK and Greer, SL, ‘Mapping Infectious Disease Control in the European Union’ (2012) 37 Journal of Health Politics, Policy, and Law 935 Estlund, D, Democratic Authority: A Philosophical Framework (Princeton, Princeton University Press, 2008) Etheridge, EW, Sentinel for Health: A History of the Centers for Disease Control (Berkeley and Los Angeles, University of California Press, 1992) European Commission, Taking European Knowledge Society Seriously—Report of the Expert Group on Science and Governance (Luxembourg, Office for Official Publications of the European Communities, 2007) European Medicines Agency, Pandemic Report and Lessons Learned: Outcome of the European Medicines Agency’s Activities during the 2009 (H1N1) Flu Pandemic (London, European Medicines Agency, 2011) European Ombudsman, Annual Report 2012 (Luxembourg, European Union, 2012) ——, Annual Report 2013 (Luxembourg, European Union, 2013) ——, Good Administration in Practice: The European Ombudsman’s Decisions in 2013 (Luxembourg, European Union, 2014) ——, Putting it Right? How the EU Institutions Responded to the Ombudsman in 2013 (Luxembourg, European Union, 2014) ——, Strategy of the European Ombudsman: Towards 2019 (Luxembourg, European Union, 2014) ——, The European Code of Good Administrative Behaviour (Luxembourg, European Union, 2013) Ezrahi, Y, Imagined Democracies: Necessary Political Fictions (Cambridge, Cambridge University Press, 2012) ——, The Descent of Icarus: Science and the Transformation of Contemporary Democracy (Cambridge MA, Harvard University Press, 1990) Fairchild, AL, ‘The Right to Know, the Right to Be Counted, the Right to Resist: Cancer, Aids, and the Politics of Privacy and Surveillance in Post-War America’ (2015) 3 Journal of Medical Law and Ethics 45 —— and Bayer, R, ‘In the Name of Population Well-Being: The Case for Public Health Surveillance’ (2015) Journal of Health Politics, Policy and Law 3445650 Farina, CR, ‘The “Chief Executive” and the Quiet Constitutional Revolution’ (1997) 49 Administrative Law Review 179

246

Bibliography

——, ‘The Consent of the Governed: Against Simple Rules for a Complex World’ (1996) 72 Chicago-Kent Law Review 987 Fearnley, L, ‘Epidemic Intelligence: Langmuir and the Birth of Disease Surveillance’ (2010) 3 Behemoth 36 ——, ‘Signals Come and Go: Syndromic Surveillance and Styles of Biosecurity’ (2008) 40 Environment and Planning A 1615 Fee, E and Brown, TM, ‘Preemptive Biopreparedness: Can We Learn Anything from History?’ (2001) 91 American Journal of Public Health 721 Fenner, F, and others, Smallpox and Its Eradication (Geneva, World Health Organization, 1988) Ferber, D, ‘Time for a Synthetic Biology Asilomar?’ (2004) 303(5655) Science 159 Fidler, D, ‘Sars: Political Pathology of the First Post-Westphalian Pathogen’ (2003) 31 Journal of Law, Medicine and Ethics 485 ——, ‘From International Sanitary Conventions to Global Health Security: The New International Health Regulations’ (2005) 4 Chinese Journal of International Law 325 —— and Gostin, L, Biosecurity in the Global Age: Biological Weapons, Public Health, and the Rule of Law (Stanford, Stanford Law and Politics, 2007) Fisher E, ‘Chemicals as Regulatory Objects’ (2014) 23 Review of European, Comparative and International Environmental Law 163 ——, ‘Climate Change Litigation, Obsession and Expertise: Reflecting on the Scholarly Response to Massachusetts v EPA’ (2013) 39 Law And Policy 236 ——, ‘Framing Risk Regulation: A Critical Reflection’ (2013) 4 European Journal of Risk Regulation 125 ——, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19 ——, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006) ——, Risk Regulation and Administrative Constitutionalism (Oxford, Hart Publishing, 2007) ——, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495 ——, Jones, J and von Schomberg, R (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006) ——, Pascual, P and Wagner, W, ‘Rethinking Judicial Review of Expert Agencies’ (2015) 93 Texas Law Review 1681 Flear, M, Governing Public Health: EU Law, Regulation and Biopolitics (London, Bloomsbury Publishing, 2015) Føllesdal, A, ‘The Legitimacy Challenges for New Modes of Governance: Trustworthy Responsiveness’ (2011) 46 Government and Opposition 81 Foucault, M, Discipline and Punish (New York, Vintage Books, 1979) ——, Security, Territory, Population: Lectures at the Collège de France 1977–1978 (trans, Paris, Gallimard, 2004) Fox, DM and Zuckerman, DM, ‘Regulatory Reticence and Medical Devices’ (2014) 92 Milbank Quarterly 151 Freeman, J and Vermeule, A, ‘Massachusetts v EPA: From Politics to Expertise’ (2007) Supreme Court Review 51 French, M, ‘Woven of War-Time Fabrics: The Globalization of Public Health Surveillance’ (2009) 6 Surveillance & Society 101

Bibliography

247

—— and Mykhalovskiy, E, ‘Public Health Intelligence and the Detection of Potential Pandemics’ (2013) 35 Sociology of health & illness 174 —— and Smith, G, ‘“Health” Surveillance: New Modes of Monitoring Bodies, Populations, and Polities’ (2013) 23 Critical Public Health 383 Frug, GE, ‘The Ideology of Bureaucracy in American Law’ (1984) 97 Harvard Law Review 1276 Funtowicz, S, Shephard, I, Wilkinson, D and Ravetz, J, ‘Science and Governance in the European Union: A Contribution to the Debate’ (2000) 27 Science and Public Policy 327 Garland, MB, ‘Deregulation and Judicial Review’ (1985) 98 Harvard Law Review 505 Gibbons, M and others, The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies (London, Sage Publications, 1994) Gieryn, TF, Cultural Boundaries of Science: Credibility on the Line (Chicago, University of Chicago Press, 1999) Giesecke, J and Weinberg, J, ‘A European Centre for Infectious Disease?’ (1998) 352 (9136) The Lancet 1308 Gilad, S, ‘Why the Haves do not Necessarily Come out Ahead in Informal Dispute Resolution’ (2010) 32 Law & Policy 283 Glenn Cohen, I (ed), The Globalization of Health Care: Legal and Ethical Issues (Oxford, Oxford University Press, 2014) Gornitzka, Å and Holst, C, ‘The Expert-Executive Nexus in the EU: An Introduction’ (2015) 3 Politics and Governance 1 Greer, SL, Everything You Always Wanted to Know about European Union Health Policies but Were Afraid to Ask (Copenhagen, World Health Organization on behalf of the European Observatory on Health Systems and Policies, 2014) ——, ‘The European Centre for Disease Prevention and Control: Hub or Hollow Core?’ (2012) 37 Journal of Health Politics, Policy, and Law 1001 ——, ‘The Weakness of Strong Policies and the Strength of Weak Policies: Law, Experimentalist Governance, and Supporting Coalitions in European Union Health Care Policy’ (2011) 5 Regulation & Governance 187 —— and Kurzer, P (eds), European Union Public Health Policies: Regional and Global Perspectives (Abingdon, Routledge, 2013) —— and Löblová, O, ‘European Integration in the Era of Permissive Dissensus: Neofunctionalism and Agenda-Setting in European Health Technology Assessment and Communicable Disease Control’ (2016) 10.1057/cep.2016.6 Comparative European Politics —— and Mätzke, M, ‘Bacteria without Borders: Communicable Disease Politics in Europe’ (2012) Journal of Health Politics, Policy and Law 1813763 —— and others, Everything You Always Wanted to Know About European Union Health Policy but Were Afraid to Ask (Brussels, European Observatory on Health Systems and Policies, 2014) —— and Singer, PM, ‘The United States Confronts Ebola: Suasion, Executive Action, and Fragmentation’ (2015) Health Economics, Policy and Law http://dx.doi.org/10.1017/ S1744133116000244 ——, Wismar, M and Figueras, J (eds), Strengthening Health System Governance: Better Policies, Stronger Performance (Berkshire, Open University Press, 2015) Guston, D, Between Politics and Science: Assuring the Integrity and Productivity of Research (Cambridge, Cambridge University Press, 2000)

248

Bibliography

Guyer, P (ed), The Cambridge Companion to Kant (Cambridge, Cambridge University Press, 1992) Habermas, J, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (Boston MA, MIT Press, 1996) ——, ‘Democracy in Europe: Why the Development of the EU into a Transnational Democracy is Necessary and How is it Possible’ (2015) 21 European Law Journal 54 ——, The Lure of Technocracy (Cambridge, Polity Press, 2015) Hackett, EJ et al (eds), The Handbook of Science and Technology Studies (Cambridge MA, MIT Press, 2008) Hacking, I, ‘A Tradition of Natural Kinds’ (1991) 61(1/2) Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition 109 Haggerty, KD and Ericson, RV (eds), The New Politics of Surveillance and Visibility (Toronto, University of Toronto Press, 2006) Hamlin, C, Public Health and Social Justice in the Age of Chadwick, Britain 1800–1854, (Cambridge, Cambridge University Press, 1998) Hanley, AV, Huang, ZY, Pett, WL, ‘Effects of Dietary Transgenic Bt Corn Pollen on Larvae of Apismelliferaand Galleria mellonella’ (2003) 42 (4) Journal of Apicultural Research 77 Harlow, C, Accountability in the European Union (Oxford, Oxford University Press, 2002) ——, ‘Composite Decision-Making and Accountability Networks: Some Deductions from a Saga’ (2013) 32 Yearbook of European Law 3 —— and Rawlings, R, Law and Administration 3rd edn (Cambridge, Cambridge University Press, 2009) —— and ——, Process and Procedure in EU Administration (Oxford, Hart Publishing, 2014) —— and ——, ‘Promoting Accountability in Multilevel Governance: A Network Approach’ (2007) 13 European Law Journal 542 Harr, J, A Civil Action (New York, Random House, 1995) Hauser, RG, ‘Here We Go Again—Another Failure of Postmarketing Device Surveillance’ (2012) New England Journal of Medicine http://dx.doi.org/10.1056/NEJMp1114695 Hay, SI and others, ‘Big Data Opportunities for Global Infectious Disease Surveillance’ (2013) 10 (4) PLoS medicine e1001413 Hazell, L and Shakir, SAW, ‘Under-Reporting of Adverse Drug Reactions’ (2006) 29 Drug Safety 385 Heinemann, M and Panke, S, ‘Synthetic Biology: Putting Engineering into Biology’ (2006) 22 (22) Bioinformatics 2790 Heisey-Grove, D and others, ‘Using Electronic Clinical Quality Measure Reporting for Public Health Surveillance’ (2015) 64 MMWR: Morbidity and Mortality Weekly Report 439 Hertogh, M, ‘Coercion, Cooperation, and Control: Understanding the Policy Impact of Administrative Courts and the Ombudsman in the Netherlands’ (2001) 23 Law & Policy 47 Hervey, T, ‘“Adjudicating in the Shadow of the Informal Settlement?”: The Court of Justice of the European Union, “New Governance” and Social Welfare’ (2010) 63 Current Legal Problems 92 —— and McHale, JV, Health Law and the European Union (Cambridge, Cambridge University Press, 2004) Herwig, A, ‘Whither Science in WTO Dispute Settlement?’ (2008) 21 (4) Leiden Journal of International Law 823

Bibliography

249

Heyvaert, V, ‘Governing Climate Change: Towards a New Paradigm for Risk Regulation’ (2011) 74 Modern Law Review 817 Hilgartner, S, Science on Stage: Expert Advice as Public Drama (Stanford, Stanford University Press, 2000) Hix, S, The Political System of the European Union (Basingstoke, Palgrave Macmillan, 2005) Hodge, G, Bowman, D and Maynard, A, International Handbook on Regulating Nanotechnologies (Cheltenham, Edward Elgar, 2010) Hofmann, HCH and Turk, AH, Legal Challenges in EU Administrative Law: Towards an Integrated Administration (Cheltenham, Edward Elgar, 2009) Jaffe, LL, Judicial Control of Administrative Action (Boston, Little, Brown & Co, 1965) Jasanoff, S, ‘A New Climate For Society’ (2010) 27 Theory, Culture and Society 233 ——, ‘Biotechnology and Empire: The Global Power of Seeds and Science’ (2006) 21(1) Osiris 273 ——, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005) ——, ‘Epistemic Subsidiarity—Coexistence, Cosmopolitianism, Constitutionalism’ (2013) 4 European Journal of Risk Regulation 133 ——, Learning from Disaster: Risk Management After Bhopal (Philadelphia, University of Pennsylvania Press, 1994) ——, ‘Ordering Life: Law and the Normalization of Biotechnology’ (2001) 17(62) Politeia 34 —— (ed), Reframing Rights: Bioconstitutionalism in the Genetic Age (Cambridge MA, MIT Press, 2011) ——, ‘Science, Politics, and the Renegotiation of Expertise at EPA’ (1992) 7 Osiris 194 —— (ed), States of Knowledge: The Co- Production of Science and Social Order (Abingdon, Routledge, 2004) ——, ‘Technologies of Humility: Citizen Participation in Governing Science’ (2003) 41 Minerva 223 ——, The Fifth Branch: Science Advisers as Policy Makers (Cambridge MA, Harvard University Press, 1990) ——, The Fifth Branch: Science Advisors as Policy Makers2nd edn (Cambridge MA, Harvard University Press, 1994) ——, ‘The Songlines of Risk’ (1999) 8 Environmental Values 135 ——, ‘What Judges Should Know about the Sociology of Science’ (1992) 32 (3) Jurimetrics 345 Joerges, C, ‘Deliberative Supranationalism: Two Defences’ (2002) 8 European Law Journal 133 —— and Glinski, C (eds), The European Crisis and the Transformation of Transnational Governance (Oxford, Hart Publishing, 2014) ——, Ladeur, K-H, and Vos, E (eds), Integrating Scientific Expertise into Regulatory DecisionMaking: National Traditions and European Innovations (Baden-Baden, Nomos, 1997) ——, Mény, Y and Weiler, J (eds), Mountain Or MoleHill: A Critical Appraisal of the Commission White Paper on Governance (Florence, European University Institute, 2002) —— and Neyer, J, ‘From Intergovernmental Bargaining to Deliberative Political Processes: The Constitutionalisation of Comitology’ (1997) 3 European Law Journal 273 —— and Vos, E (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999)

250

Bibliography

Kaddour, N, ‘No Laws in Nanoland: How to Reverse the Trend? The French Example’ (2013) 30 Pace Environmental Law Review 486 Kagan, E, ‘Presidential Administration’ (2001) 114 Harvard Law Review 2245 Karkkainen, B, ‘“New Governance” in Legal Thought and in the World: Some Splitting as Antidote to Overzealous Lumping’ (2004–05) 89 Minnesota Law Review 471 Katz, R and Rosenbaum, S, ‘Challenging Custom: Rethinking National Population Surveillance Policy in a Global Public Health Age’ (2010) 35 Journal of Health Politics, Policy and Law 1027 Kay, A and Williams, OD (eds), Global Health Governance: Crisis, Institutions and Political Economy (Basingstoke, Palgrave Macmillan, 2009) Keeler, JTS, ‘Opening the Window for Reform Mandates, Crises, and Extraordinary PolicyMaking’ (1993) 25 Comparative Political Studies 433 Keene, E, ‘Social Status, Social Closure and the Idea of Europe as a “Normative Power”’ (2013) 19 European Journal of International Relations 939 Kelemen, D, ‘Globalizing European Union Environmental Policy’ (2010) 17 Journal of European Public Policy 335 Kelle, A, ‘Securitisation of International Public Health: Implications for Global Health Governance and the Biological Weapons Prohibition Regime’ (2007) 13 Global Governance 217 Kennedy, D, A World of Struggle: How Power, Law and Expertise Shape Global Political Economy (Princeton, Princeton University Press, 2016) King, NB, ‘The Influence of Anxiety: September 11, Bioterrorism, and American Public Health’ (2003) 58 Journal of the History of Medicine and Allied Sciences 433 Kitrosser, H, ‘Scientific Integrity: The Perils and Promise of White House Administration’ (2011) 79 Fordham Law Review 2395 Knowles, SG, ‘Learning from Disaster?: The History of Technology and the Future of Disaster Research’ (2014) 55 Technology and Culture 773 Korkea-aho, E, ‘Laws in Progress? Reconceptualizing Accountability Strategies in the Era of Framework Norms’ (2013) 2 Transnational Environmental Law 363 Kritikos, M, ‘Traditional Risk Analysis and Releases of GMOs into the European Union: Space for Non-Scientific Factors?’ (2009) 34 (3) European Law Review 405 Kuklick, H (ed) Knowledge and Society Studies in the Sociology of Culture Past and Present Vol 6 (Greenwich, JAI Press, 1986) Laaser, U and Brand, H, ‘Global Health in the 21st Century’ (2014) 7 Global Health Action 23694 Lakoff, A (ed), Disaster and the Politics of Intervention (New York City, Columbia University Press, 2010) —— and Collier, SJ, Biosecurity Interventions: Global Health and Security in Question (New York City, Columbia University Press, 2008) Lange, B, Haines, F and Thomas, D (eds), Regulatory Transformations: Rethinking EconomySociety Interactions (Oxford, Hart Publishing, 2015) Langmuir, AD, ‘The Surveillance of Communicable Diseases of National Importance’ (1963) 268 The New England Journal of Medicine 182 Lash, S, Szerszynski, B, and Wynne, B (eds), Risk, Environment and Modernity: Towards a New Ecology (London, Sage Publications, 2000) Latin, H, ‘Good Science, Bad Regulation, and Toxic Risk Assessment’ (1988) 5 Yale Journal on Regulation 89

Bibliography

251

Latour, B, We Have Never Been Modern (Cambridge MA, Harvard University Press, 1993) Lee, LM and others (eds), Principles and Practice of Public Health Surveillance (Oxford, Oxford University Press, 2010) Lee, M, ‘Beyond Safety? The Broadening Scope of Risk Regulation’ (2009) 62 Current Legal Problems 242 ——, EU Environmental Law, Governance and Decision-Making (Oxford, Hart Publishing, 2014) ——, EU Regulation of GMOs: Law and Decision Making for a New Technology (Cheltenham, Edward Elgar, 2008) ——, ‘Risk and Beyond: EU Regulation of Nanotechnology’ (2010) 35 European Law Review 799 ——, ‘The Ambiguity of Multi-Level Governance and (De)-harmonisation in EU Environmental Law’ (2014) 15 Cambridge Yearbook of European Legal Studies 357 Lee, SK et al, ‘Metabolic Engineering of Microorganisms for Biofuels Production: From Bugs to Synthetic Biology to Fuels’ (2008) 19(6) Current Opinion in Biotechnology 556 Leino, P, ‘Efficiency, Citizens and Administrative Culture: The Politics of Good Administration in the EU’ (2014) 20 European Public Law 681 Levi-Faur, D (ed), Oxford Handbook of Governance (Oxford, Oxford University Press, 2012) Levidow, L and Carr, S, ‘Europeanising Advisory Expertise: The Role of “Independent, Objective, and Transparent” Scientific Advice in Agri-Biotech Regulation’ (2007) 25 Environment and Planning C: Government and Policy 880 —— and ——, ‘Normalizing Novelty: Regulating Biotechnological Risk at the U.S. EPA’ (2000) 11 Risk, Health, Safety and Environment 9 ——, —— and Wield, D, ‘European Union Regulation of Agri-Biotechnology: Precautionary Links Between Science, Expertise and Policy’ (2005) 32 Science and Public Policy 261 Lezaun, J, ‘Bees, Beekeepers, and Bureaucrats: Parasitism and the Politics of Transgenic Life’ (2011) 29 Environment and Planning D: Society and Space 738 Lindseth, P, Power and Legitmacy: Reconciling Europe and the Nation State (Oxford, Oxford University Press, 2010) Liverani, M and Coker, R, ‘Protecting Europe from Diseases: From the International Sanitary Conferences to the ECDC’ (2012) 37 Journal of Health Politics, Policy and Law 913 Lopez-Gonzalez, E, Herdeiro, MT and Figueiras, A, ‘Determinants of under-Reporting of Adverse Drug Reactions’ (2009) 32 Drug safety 19 Lyon, D, Surveillance Studies: An Overview (Cambridge, Polity, 2007) Maasen, S and Weingart, P (eds) Democratization of Expertise? Exploring Novel Forms of Scientific Advice in Political Decision-Making (Dordrecht, Springer, 2005) Macdonald, H, H is for Hawk (London, Jonathan Cape, 2014) MacLehose, L and others, ‘Communicable Disease Outbreaks Involving More Than One Country: Systems Approach to Evaluating the Response’ (2001) 323 BMJ 861 ——, McKee, M and Weinberg, J, ‘Responding to the Challenge of Communicable Disease in Europe’ (2002) 295 Science 2047 Macleod, R (ed), Government and Expertise: Specialists, Administrators, and Professionals 1860–1919 (Cambridge, Cambridge University Press, 1988) Magnette, P, ‘Between Parliamentary Control and the Rule of Law: the Political Role of the Ombudsman in the European Union’ (2003) Journal of European Public Policy 677 Mair, M, Maldin, B and Smith, B, ‘Passage of S. 3678: The Pandemic and All-Hazards Preparedness Act’ (2007) 5 Biosecur Bioterror 72

252

Bibliography

Majone, G, Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth (Oxford, Oxford University Press, 2005) ——, ‘Foundations of Risk Regulation: Science, Decision-Making, Policy Learning and Institutional Reform’ (2010) 1 European Journal of Risk Regulation 5 ——, Regulating Europe (Abingdon, Routledge, 1996) ——, Rethinking the Union of Europe Post-Crisis (Cambridge, Cambridge University Press, 2014) ——, ‘The Regulatory State and its Legitimacy Problems’ (1999) 22 West European Politics 1 ——, ‘The Credibility Crisis of Community Regulation’ (2000) 38 Journal of Common Market Studies 273 Marchant, G, Sylvester, D and Abbott, K, ‘Risk Management Principles of Nanotechnology’ (2008) 2 Nanoethics 43 Mariner, WK, Annas, GJ and Parmet, WE, ‘Pandemic Preparedness: A Return to the Rule of Law’ (2009) 1 Drexel Law Review 341 Marty, MS, Carney, EW and Craig Rowlands, J, ‘Endocrine Disruption: Historical Perspectives and Its Impact on the Future of Toxicology Testing’ (2011) 120 Toxicological Sciences S93 Marx, GT, Windows into the Soul: Surveillance and Society in the Age of High Technology (Chicago, University of Chicago Press, 2015) —— and Muschert, GW, ‘Personal Information, Borders, and the New Surveillance Studies’ (2007) 3 Annual Review of Law and Social Science 375 Mashaw, J, Greed, Chaos and Governance: Using Public Choice to Improve Public Law (New Haven, Yale University Press, 1997) ——, ‘Prodelegation: Why Administrators Should Make Political Decisions’ (1985) 1 Journal of Law, Economics & Organization 81 ——, ‘Small Things Like Reasons Are Put in a Jar: Reason and Legitimacy in the Administrative State’ (2001) 70 Fordham Law Review 17 McGarity, T, ‘A Cost Benefit State’ (1998) 50 Administrative Law Review 7 ——, ‘Substantive and Procedural Discretion in Administrative Resolution of Science Policy Questions: Regulating Carcinogens in EPA and OSHA’ (1979) 67 Georgetown Law Journal 729 —— and Wagner, W, Bending Science (Cambridge MA, Harvard University Press, 2010) Meazell, EH, ‘Super Deference, the Science Obsession, and Judicial Review as Translation of Agency Science’ (2011) 109 Michigan Law Review 733 Medina, S and others, ‘Comparing Findings from Syndromic Surveillance Systems at a European Level’ (2014) 6 Online Journal of Public Health Informatics e124 Monfort, J-P et al, ‘Nanomaterials under REACH: Legal Aspects’ (2010) 1 European Journal of Risk Regulation 51 Moon, S and others, ‘Will Ebola Change the Game? Ten Essential Reforms before the Next Pandemic. The Report of the Harvard-Lshtm Independent Panel on the Global Response to Ebola’ (2015) 386 The Lancet 2204 Mooney, C, The Republican War on Science (New York, Basic Books, 2005) Morhard, R and Franco, C, ‘The Pandemic and All-Hazards Preparedness Act: Its Contributions and New Potential to Increase Public Health Preparedness’ (2013) 11 Biosecur Bioterror 145 Mossialos, E, Permanand, G, Baeten, R and Hervey, TK (eds), Health Systems governance in Europe: The role of European Union law and policy (Cambridge, Cambridge University Press, 2010)

Bibliography

253

Mulgan, R, ‘“Accountability”: An Ever-Expanding Concept?’ (2000) 78 Public Administration 555 National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academies Press, 1983) Newson, AJ, ‘Synthetic Biology: Ethics, Exceptionalism and Expectations’ (2015) 15 Macquarie Law Journal 45 Organisation for Economic Co-Operation and Development, Regulatory Reform and Innovation (Paris, Organisation for Economic Co-Operation and Development, 1996) Paquet, C and others, ‘Epidemic Intelligence: A New Framework for Strengthening Disease Surveillance in Europe’ (2005) 11 Euro surveillance: bulletin européen sur les maladies transmissibles = European communicable disease bulletin 212 Pascual, P, Wagner, W and Fisher, E, ‘Making Methods Visible: Improving the Quality of Science Based Regulation’ (2013) 2 Michigan Journal of Environmental and Administrative Law 429 Pereira, AG and Funtowicz, S (eds), Science For Policy: New Challenges, New Opportunities (Oxford, Oxford University Press, 2009) Peterson Myers, J and others, ‘Concerns over Use of Glyphosate-Based Herbicides and Risks Associated with Exposures: A Consensus Statement’ (2016) 15 Environmental Health 19 Petsko, GA, ‘An Asilomar Moment’ (2002) 3(1) Genome Biology 1 Porter, D, The History of Public Health and the Modern State (Amsterdam, Editions Rodopi, 1994) Porter, T, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton, Princeton University Press, 1995) Rabin, RL, ‘Federal Regulation in Historical Perspective’ (1986) 38 Stanford Law Review 1189 Rajan, PV, Kramer, DB and Kesselheim, AS, ‘Medical Device Postapproval Safety Monitoring Where Does the United States Stand?’ (2015) 8 Circulation: Cardiovascular Quality and Outcomes 124 Rawlings, R, ‘Engaged Elites, Citizen Action and Institutional Attitudes in Commission Enforcement’ (2000) 6 European Law Journal 28 Rechel, B, Brand, H and McKee, M, ‘Financing Public Health in Europe’ (2013) 75 Gesundheitswesen (Bundesverband der Arzte des Offentlichen Gesundheitsdienstes (Germany)) e28 Reintjes, R, ‘Variation Matters: Epidemiological Surveillance in Europe’ (2012) 37 Journal of Health Politics, Policy, and Law 955 ——, and others, ‘Benchmarking National Surveillance Systems: A New Tool for the Comparison of Communicable Disease Surveillance and Control in Europe’ (2007) 17 European Journal of Public Health 375 Renn, O and Roco, M, ‘Nanotechnology and the Need for Risk Governance’ (2006) 8 Journal of Nanoparticle Research 1 Rittberger, B and Wonka, A, ‘Introduction: Agency Governance in the European Union’ (2011) 18 Journal of European Public Policy 780 Rosen, G, A History of Public Health (Baltimore, John Hopkins University Press, 1958) Rosenkötter, N and others, ‘Non-Infectious Events under the International Health Regulations (2005) in Europe–a Case for Syndromic Surveillance’ (2014) 35 Journal of Public Health Policy 311 Royal Academy of Engineering, Synthetic Biology: Scope, Applications and Implications (London, Royal Academy of Engineering, 2009)

254

Bibliography

Ruder, WC, Lu, T and Collins, JJ, ‘Synthetic Biology Moving into the Clinic’ (2011) 333(6047) Science 1248 Sabel, CF and Zeitlin, J (eds), Experimentalist Governance in the European Union: Towards a New Architecture (Oxford, Oxford University Press, 2010) Sarat, A, Douglas, L and Umphrey, MM (eds) How Law Knows (Stanford, Stanford University Press, 2007) Sarewitz, D, ‘How Science Makes Environmental Controversy Worse’ (2004) 7 Environmental Science and Policy 385 Scalia, A, ‘A Note on the Benzene Case’ (1980) 4 Regulation 25 Scharpf, F, ‘After the Crash: A Perspective on Multilevel European Democracy’ (2015) 21 European Law Journal 384 ——, Governing in Europe: Effective and Democratic? (Oxford, Oxford University Press, 1999) Schmidt-Assmann, E and Schöndorf-Haubold, B, Der Europäische Verwaltungsverbund: Formen und Verfahren der Verwaltungszusammenarbeit in der EU (Tübingen, Mohr Siebeck, 2005) Schmidt, V, ‘Democracy and Legitimacy in the European Union Revisited: Input, Output and “Throughput”’ (2013) 61 Political Studies 2 Schwarze, J, European Administrative Law (London, Sweet & Maxwell, 2006) Science and Technology Advisory Committee, The Future of Europe is Science (Luxembourg, Publications Office of the European Union, 2014) Science and Technology Committee, Bioengineering (London, House of Commons, 2010) Scott, J, ‘In Legal Limbo: Post-Legislative Guidance as a Challenge for European Administrative Law’ (2011) 48 Common Market Law Review 329 —— and Sturm, S, ‘Courts as Catalysts: Rethinking the Judicial Role in New Governance’ (2007) 13 Columbia Journal of European Law 565 Shapiro, M, ‘Is Bioethics Broke?: On the Idea of Ethics and Law “Catching Up” with Technology’ (1999) 33 Indiana Law Review 17 ——, Who Guards the Guardians: Judicial Control of Administration (Athens, University Of Georgia Press, 1988) Shapiro, S, ‘Why Administrative Law Misunderstands How Government Works: The Missing Institutional Analysis’ (2013) 53 Washburn Law Journal 1 ——, ‘“Political” Science: Regulatory Science After the Bush Administration’ (2009) 4 Duke Journal of Constitutional Law and Public Policy 31 —— and Fisher, E, ‘Chevron and the Legitimacy of “Expert” Public Administration’ (2013) 22 Willam & Mary Bill of Rights Journal 465 ——, Fisher, E and Wagner, W, ‘The Enlightenment of Administrative Law: Looking Inside the Agency for Legitmacy’ (2012) 47 Wake Forest Law Review 463 —— and Wright, R, ‘The Future of the Administrative Presidency: Turning Administrative Law Inside-Out’ (2011) 65 University of Miami Law Review 577 Shire, JD and others, ‘Advances and Current Themes in Occupational Health and Environmental Public Health Surveillance’ (2011) 32 Annual Review of Public Health 109 Skowronek, S, Engel, S and Ackerman, B (eds), The Progressives’ Century: Democratic Reform and Constitutional Government in the United States (New Haven, Yale University Press, 2014) Slaughter, AM, ‘A New UN for a New Century’ (2006) 74 Fordham Law Review 2961 Snyder, R, ‘Leukemia and Benzene’ (2012) 9 (8) International Journal of Environmental Research and Public Health 2875

Bibliography

255

Steffen, M (ed), Health Governance in Europe: Issues, Challenges and Theories (Abingdon, Routledge, 2005) ——, ‘The Europeanization of Public Health: How Does It Work? The Seminal Role of the Aids Case’ (2012) 37 Journal of Health Politics, Policy, and Law 1057 Stewart, RB, ‘The Reformation of American Administrative Law’ (1975) 88 Harvard Law Review 1667 Stirling, A, ‘Science, Precaution, and the Politics of Technological Risk: Converging Implications in Evolutionary and Social Scientific Perspectives’ (2008) 1128 Annals of the New York Academy of Sciences 95 Stokes, E, ‘Nanotechnology and the Products of Inherited Regulation’ (2012) 39 Journal of Law and Society 93 Strauss, PL, ‘From Expertise to Politics: The Transformation of American Rulemaking’ (1996) 31 Wake Forest Law Review 745 Sunstein, CR, ‘Deregulation and the Hard-Look Doctrine’ [1983] Supreme Court Review 177 ——, The Cost Benefit State: The Future of Regulatory Protection (Washington DC, American Bar Association, 2002) Taylor, RCR, ‘The Politics of Securing Borders and the Identities of Disease’ (2013) 35 Sociology of Health and Illness 241 Terpan, F, ‘Soft Law in the European Union—The Changing Nature of EU Law’ (2015) 21 European Law Journal 68 Thacker, SB, Berkelman, RL and Stroup, DF, ‘The Science of Public Health Surveillance’ (1989) Journal of Public Health Policy 187 Thompson, EP, ‘The Moral Economy of the English Crowd in the Eighteenth Century’ (1971) 50 Past & Present 76 Trondal, J, An Emergent EU Executive Order (Oxford, Oxford University Press, 2010) Tsadiras, A, ‘The Ombudsman’ in P Craig, EU Administrative Law (Oxford, Oxford University Press, 2012) ——, ‘Unravelling Ariadne’s Thread: The European Ombudsman’s Investigative Powers’ (2008) 45 Common Market Law Review 757 Türk, A, Hofmann, H and Rowe, G, EU Administrative Law and Policy (Oxford, Oxford University Press, 2011) 777 United Nations Secretary General, A More Secure World: Our Shared Responsibility: Report of the High-level Panel on Threats, Challenges, and Change (New York, United Nations, 2004) Van Asselt, MBA, Versluis, E and Vos, E (eds), Balancing Between Trade and Risk. Integrating Legal and Social Science Perspectives (Abingdon, Routledge, 2013) —— and Vos, E, ‘Wrestling with Uncertain Risks: EU Regulation of GMOs and the Uncertainty Paradox’ (2008) 11 Journal of Risk Research 281 ——, —— and Wildhaber, I, ‘Symposium on Critical Infrastructures: Risk, Responsibility and Liability ∙ Some Reflections on EU Governance of Critical Infrastructure Risks’ (2015) 6 European Journal of Risk Regulation 185 van Calster, G, ‘Regulating Nanotechnology in the European Union’ (2006) 3 Nanotechnology Law & Business 359 Van Zwanenberg, P and Millstone, E, BSE: Risk, Science and Governance (Oxford, Oxford University Press, 2005) Vayena, E and others, ‘Ethical Challenges of Big Data in Public Health’ (2015) 11 PLoS Comput Biol e1003904

256

Bibliography

Vogelezang-Stoute, E, ‘Regulating Nanomaterials: Bottlenecks and Perspectives in EU Legislation on Chemicals and Products’ (2012) European Energy and Environmental Law Review 41 Vos, E, ‘EU Agencies: Features, Framework and Future’ Maastricht Faculty of Law Working Paper 2013/3 ——, ‘EU Food Safety Regulation in the Aftermath of the BSE Crisis’ (2000) 23 Journal of Consumer Policy 227 ——, Institutional Frameworks of Community Health and Safety Legislation: Committees, Agencies and Private Bodies (Oxford, Hart Publishing, 1999) —— and Everson, M (eds), Uncertain Risks Regulated (London, Routledge-Cavendish, 2012) —— and —— (eds), Uncertain Risks Regulated: National, EU and International Regulatory Models Compared (London, Routledge-Cavendish, 2009) —— and Wendler, F, Food Safety Regulation in Europe: A Comparative Institutional Analysis (Antwerpen, Intersentia, 2007) Wagner, WE, ‘Ethyl: Bridging the Science-Law Divide’ (1995) 74 Texas Law Review 1291 ——, ‘The “Bad Science” Fiction: Reclaiming the Debate over the Role of Science in Public Health and Environmental Regulation’ (Autumn 2003) 66 Law and Contemporary Problems 63 ——, ‘The Science Charade in Toxic Risk Regulation’ (1995) 95 Columbia Law Review 1613 Waterton, C and Wynne, B, ‘Knowledge and Political Order in the European Environment Agency’ in S Jasanoff (ed), States of Knowledge: The Co-Production of Science and Social Order (Abingdon, Routledge, 2004) Watts, KA, ‘Proposing a Place for Politics in Arbitrary and Capricious Review’ (2009) 119 Yale Law Journal 2 Weiler, J, The Constitution of Europe: Do the New Clothes Have an Emperor? (Cambridge, Cambridge University Press, 1999) Weimer, M, ‘Applying Precaution in EU Authorisation of Genetically Modified Products— Challenges and Suggestions for Reform’ (2010) 16 European Law Journal 624 ——, ‘Risk Regulation and Deliberation in EU Administrative Governance: GMO Regulation and its Reform’ (2015) 21(5) European Law Journal 622 —— and Pisani, G, ‘Biotechnology: The EU Adventures of Herculex’ (2014) 5 European Journal of Risk Regulation 208 Weinberg, J and others, ‘On Behalf of the Charter Group: Establishing Priorities for European in Communicable Disease Surveillance’ (1999) 9(3) European Journal of Public Health 236 Weingart, P and Maasen, S (eds), Democratization of Expertise? Exploring Novel Forms of Scientific Advice in Political Decision-Making (Dordrecht, Springer, 2005) Weir, L and Mykhalovskiy, E, Global Public Health Vigilance: Creating a World on Alert Vol 10 (Abingdon, Routledge, 2009) Westra, L and Lemons, J (eds), Perspectives on Ecological Integrity (Dordrecht, Springer, 1995) Williams, SF, ‘The Roots of Deference’ (1991) 100 Yale Law Journal 1103 Wilsdon, J and Doubleday, R (eds), Future Directions for Scientific Advice in Europe (Cambridge, Centre for Science and Policy, 2015) Winickoff, D and others, ‘Adjudicating the GM Food Wars: Science, Risk, and Democracy in World Trade Law’ (2005) 30 (1) Yale Journal of International Law 81

Bibliography

257

Woodward, D and Smith, RD, ‘Global Public Goods and Health: Concepts and Issues’ in R Smith and others (eds), Global Public Goods for Health: Health Economic and Public Health Perspectives (Oxford, Oxford University Press, 2003) World Health Organization, Critically Important Antibacterial Agents for Human Medicine for Risk Management Strategies of Non-human Use: Report of a WHO Working Group Consultation (Geneva, World Health Organization, 2005) ——, Report of the Ebola Interim Assessment Panel (Geneva, World Health Organisation, 2015) ——, Report of the Technical Discussions at the Twenty-First World Health Assembly on ‘National and Global Surveillance of Communicable Diseases’ (Geneva, World Health Organization, 1968) ——, The World Health Report 2007—A Safer Future: Global Public Health Security in the 21st Century (Geneva, World Health Organization, 2007) Wright, S, Molecular Politics: Developing American and British Regulatory Policy for Genetic Engineering, 1972–1982 (Chicago, University of Chicago Press, 1994) Wynne, B, Rationality and Ritual: Participation and Exclusion in Nuclear Decision-Making 2nd edn (Abingdon, Earthscan, 2011) ——, ‘Uncertainty and Environmental Learning: Reconceiving Science and Policy in the Preventive Paradigm’ (1992) 2 Global Environmental Change 111 Ziemann, A, Syndromic Surveillance: Made in Europe (Maastricht, Maastricht University, 2015) —— and others, ‘Meeting the International Health Regulations (2005) Surveillance Core Capacity Requirements at the Subnational Level in Europe: The Added Value of Syndromic Surveillance’ (2015) 15 BMC Public Health 107 Zimmern, A, ‘Democracy and the Expert’ (1930) 1 Political Quarterly 7

258

INDEX

accountability administrative constitutionalism 51–2 nanotechnologies regulation 96–8 risk governance constraints 34–5 see also under European Ombudsman (EO) administrative constitutionalism accountability 51–2 co-production background 37–8 debates over 41–3 deliberative-constitutive (DC) paradigm 38 EU, debates over 39, 42–3, 55–6 EU as interconnected web (national/subnational administrations) 46 EU referendum in UK 39, 55–6 expert executive power 38, 40–4 learning process 45–6 expertise accountability yardsticks 51–2 in EU policy documents 49 explicit/tacit 45 opening up to judgment 50–4 tangential approach 44–5 temporal aspects 45–50 key issues/conclusion 10, 13, 39, 55–6 precautionary principle 48–9, 52–3 rational instrumental (RI) paradigm 38, 43–4 discretion 52–4 learning process 46–9 risk assessment/risk management divide 47–9 risk regulation (health and safety) 39 expertise in 40–1, 49–50 nature of institution 41 policy narrowing 43–4 see also co-production of expert and executive power; risk governance administrative law see US administrative law insights administrative legitimation see under risk administration (EU), contested legitimation Administrative Procedure Act 1946 (US) 42 Agriculture, Fisheries and Food, Ministry of 35 Alemanno, Alberto 143 Amflora (potato) case see under risk administration (EU), contested legitimation

anthrax scare see under public health expertise, and EU security Arabidopsis thaliana (cress plant) 63 Asilomar Conference on Recombinant DNA (1975) 65–6 Baltimore Gas case see under US administrative law insights Barnett, T 133 Barroso, José Manuel 32, 154, 156 Beck, Ulrich 21–2, 82, 104–5 behavioural expertise background 143 Behavioral Insights Team (BIT) 144 Behavioural Economics Team (Institute for Health and Consumer Protection (DG JRC)) 153, 156 Chief Scientific Adviser (EC) 146, 154–5 choice architectures 148–9, 162 complex/ambiguous situations 148 core insights 148–9 definition 148 DG SANCO 152–3, 156 diffusion 153–4 ecosystem of expertise (EU) 145, 155–7, 162–3 epistemic/political authority, combination 157–60 European interventions 144 experimentation 152–3 Foresight and Behavioural Insights Unit (FBIU) 156 fragmentation/complexity of expertise ecosystem of 145–7 Framework Contract for the Provision of Behavioural Studies (FCPBS) 152–3 harm principle 160 history of 144–5 institutional voids 154–7, 162 definition 156 institutionalisation 153 international interventions 143–4 key issues/conclusions 11–12, 145–7, 162–3 liberal paternalism 160 nudge techniques 144, 146, 148, 159, 163 object dimension 157–8

260

Index

Office of Information and Regulatory Affairs (OIRA) (US) 160, 161 policy-making and regulation 149–54 account of use 150–2 Table political epistemologies 147 regional/national contexts (EU) 163 regulation reframing 160–2 normative perspective 160–1 persuasive technology 161–2 technocratic intelligence, re-strengthening 162 Regulatory Fitness and Performance programme (EC) 160–1 Smart Regulation programme (EC) 160–1 social dimension 157–9 temporal dimension 158–9 benchmarking see under public health surveillance Benzene case see under US administrative law insights big data, use see under public health surveillance Biocidal Products Regulation see under nanotechnologies regulation biology see synthetic biology, and recombinant regulation bioluminescent (light-producing/glow-in-thedark) plants see under synthetic biology, and recombinant regulation biosecurity infrastructure see under public health surveillance biosecurity/health security, reconfiguration see under public health expertise, and EU security bird flu virus controversy see under public health expertise, and EU security Blair, Tony 27 Brexit debate 1, 55–6 BSE (mad cow) crisis 34–5 Busch, Lawrence 29 Bush, President George HW 25–6, 35 Bush, President George W 26, 31, 209–10 Centers for Disease Control and Prevention (CDC) (US) see under public health surveillance Chemical Manufacturers Association v EPA see under US administrative law insights Chevron case see under US administrative law insights Choi, BCK 126 climate change challenge, in US 31 co-production of expert and executive power as analytical frame 8–9 Brexit debate 1 constitutionalism rethink 13–15 definition of co-production 23 delegation of powers 3 EU executive empowerment 4–5, 9–13

European Ombudsman (EO) see under European Ombudsman (EO) key issues 2–3 knowledge/expertise infrastructures 6–7 nanotechnologies regulation see under nanotechnologies regulation, legitimacy objective epistemologies 7–8 policy-relevant expertise, EU trends 5–6 post-truth background 1 public health surveillance see under public health surveillance rational-instrumental model of evidencebased regulation 7, 8 risk regulation 3–4 theme 2 US administrative law see under US administrative law insights see also administrative constitutionalism; risk governance Coggi, Paoli Testori 143 communicable diseases see under public health expertise, and EU security; public health surveillance Cosmetics Regulation see under nanotechnologies regulation Daubert v Merrell Dow 34 Declaration of the Second International Meeting on Synthetic Biology (2006) 65 delegation see under co-production of expert and executive power; risk governance Deliberate Release/Contained Use Directives see under synthetic biology, and recombinant regulation democratic participation see under nanotechnologies regulation Downs case 47 E-Coli see under public health expertise, and EU security Ebola see under public health surveillance Edison Electric Institute v EPA see under US administrative law insights Electronic Health Records (EHR) see under public health surveillance emergency responses see under public health expertise, and EU security emerging infectious diseases (EIDs) see under public health surveillance Environmental Protection Agency (EPA) (US) 31 epistemic communities see under European Ombudsman (EO) Ethyl Corp v EPA see under US administrative law insights European Biofuels Technology Platform see under European Ombudsman (EO)

Index European Centre for Disease Control (ECDC) see under public health surveillance European Chemical Industry Council (CEFIC) see under nanotechnologies regulation European Consumers’ Organisation (BEUC) see under nanotechnologies regulation European Environmental Bureau (EEB) see under nanotechnologies regulation European examples of epistemic delegation 227–8 European Food Safety Authority (EFSA) see under European Ombudsman (EO); risk administration (EU), contested legitimation European Food Standards Agency (EFSA) 27–8 European Ombudsman (EO) accountability institution 217–18, 225–30 accountability and legitimacy 223–4 co-production 218, 222–4 and knowledge 231–8 dialogic/political role 225 differences of approach to institutions 237–8 direct/indirect challenges to objectivity 232 disciplines/sub-disciplines, governance range 219–20 diversity of fora/processes 239 economic interests 232–4 epistemic communities 236 European Biofuels Technology Platform 233 European Food Safety Authority (EFSA) 222 expertise/scientific or technical advice, delivery arrangements 220–1 geographical balance 234 governance complexity 219–24 institutional boundary drawing 231, 235–8 interests balance/conflicts of 232–3 role of 236 investigative powers 225–7 key issues/conclusions 14, 218–19, 238–9 knowledge 231–8 maladministration concept 229 margins of discretion 238 mutual dependence of parts of process 236–7 non-binding guidance 227–8 objectivity construction 231, 232–5 OII Concerning the Composition of Commission Expert Groups 233, 238 openness of proceedings 227, 229–30 procedural advantages 228–9 reason-giving obligation 238–9 reasonableness 230 science/expertise/politics, divide/ connection 221–2, 235, 237 stakeholder groups 232–3

261

strategic approach 225 substantive approach to objectivity 234–5 see also under nanotechnologies regulation executive power see under public health expertise, and EU security exogenous risk see under risk governance expert executive power see under administrative constitutionalism expertise public health see public health expertise, and EU security see also under administrative constitutionalism Fidler, D 106 Flu Trends controversy see under public health surveillance Foresight and Behavioural Insights Unit (FBIU) see under behavioural expertise France expert executive power, learning process 45 nanotechnologies regulation 89 genetically modified organisms (GMOs) authorisations see under risk administration (EU), contested legitimation genetically modified organisms (GMOs)/ micro-organisms bioluminescent plants see under synthetic biology, and recombinant regulation EU regulation harmonisation 32–3 labelling provision, US controversy 31 synthetic biology 69 Glover, Anne 154 GM/non-GM crop contamination (in Germany) 30 Gostin, L 106 Gove, Michael 1, 55 governable subjects see risk governance Greer, Scott 114–15 H is for Hawk (Macdonald) 39, 44–5 Hacking, Ian 19 Hajer, Marten 146 harmonisation of regulatory standards see under risk governance Health Threats Decision (2013) see under public health expertise, and mixing EU security Herculex (maise) case see under risk administration (EU), contested legitimation Herrmann, Benedikt 153 HIV/Aids see under public health expertise, and EU security institutional boundary drawing see under European Ombudsman (EO) Institutional Frameworks of Community Health and Safety Regulation (Vos) 47

262

Index

institutional voids see under behavioural expertise International Health Regulations, revised see under public health surveillance Jaaskinen, AG 47 Jasanoff, Sheila 231 Kahneman, D 144–5, 148 knowledge/expertise infrastructures see under co-production of expert and executive power Langmuir, Alexander 125–6, 137, 141 Latour, Bruno 23 liberal paternalism see under behavioural expertise Maastricht Treaty 42–3 Macdonald, Helen 39, 44–6, 51 Majone, G 49–50 maladministration concept see under European Ombudsman (EO) margins of discretion see under European Ombudsman (EO) Massachusetts v EPA 31 see also under US administrative law insights Medisys see under public health surveillance Mill, John Stuart 160 Ministry of Agriculture, Fisheries and Food 35 nanotechnologies regulation accountability 96–8 administrative 97–8 legal 97 political 96–7 social 98 Biocidal Products Regulation 90 as controversial matter 81–2 Cosmetics Regulation, definition 86 definition quest 84–6 at European level 84–5 European Parliament legislation review 85–6 international agreement, unattained 84 see also recommended definition (EC) below definition recommended (EC) 86–8 engineered/natural/incidental materials 86–7 impact dimensions 88–92 legitimacy test 92–8 size range 87 stakeholders’ reactions 87–8 see also definition quest above democratic participation 93–6 direct/indirect 93 drafting process 93–4 imbalances in 94–6

depoliticisation 99 European Chemical Industry Council (CEFIC) 87 European Consumers’ Organisation (BEUC) 87 European Environmental Bureau (EEB) 87 European Ombudsman 97–8 European Strategy for Nanotechnology (EC) 84–5 European to national level impacts 88–9, 99 Food Information Regulation 90–1 key issues/conclusion 11, 82–3, 98–9 legitimacy and co-production 83–4, 99 test 92–8 National Institute for Public Health and the Environment (RIVM) (Netherlands) 88 Novel Food Regulation 91 REACH Regulations 91–2 SCENIHR (Scientific Committee on Emerging and Newly Identified Health Risks) 93–6 soft to hard law impacts 90–2, 99 synthetic biology link see under synthetic biology, and recombinant regulation National Academy of Sciences (NAS) 25 Natural Resources Defense Council, Inc v EPA see under US administrative law insights Netherlands, National Institute for Public Health and the Environment (RIVM) 88 New Techniques Working Group see under synthetic biology, and recombinant regulation non-communicable diseases see under public health surveillance Novel Food Regulation see under nanotechnologies regulation nudge techniques see under behavioural expertise Obama, President Barack 26 objective epistemologies see under co-production of expert and executive power Office of Information and Regulatory Affairs (OIRA) (US) see under behavioural expertise Ombudsman see European Ombudsman (EO) pandemic preparedness plans see under public health expertise, and EU security Pascual, Pasky 45 pathogenic risk see under public health expertise, and EU security Pfizer case 47 see also under risk administration (EU), contested legitimation Platonic ideal of an interest-free enterprise 26 policy-making and regulation see under behavioural expertise

Index policy-relevant expertise, EU trends see under co-production of expert and executive power political epistemologies see under behavioural expertise Popper, Karl 34 precautionary principle administrative constitutionalism 48–9, 52–3 risk administration (EU), contested legitimation 167–8, 182, 186–7, 190–1 public health expertise, and EU security anthrax scare 105 biosecurity/health security, reconfiguration 105–7 globalisation 106–7 human security context 107 public health to health security, movement 105 terrorism 106 bird flu virus controversy 105, 116 E-Coli 116 emergency responses 101 EU actors (health emergency response) 110–11 executive power description 107–8 emergencies’ effect on constraints 108–9 focusing events 108 expertise 111–14 communicable diseases (ECDC) 113 coordination of surveillance systems (ECDC) 113–14 DG SANCO (Directorate General for Health, EC) 113, 115 disease monitoring (WHO) involvement 112–13 formalisation of existing networks 114 long-standing/significant role 111–12 single diseases surveillance (EWRS) 113 surveillance and training (EPIET) 112 Global Health Security Initiative 107 Health Threats Decision (2013) adoption 102 basic purpose 102–3 EU power 109–10 health emergency response, EU actors 110–11 Health Security Committee (HSC) 115–16 key issues/conclusion 102, 118–19 public health preparedness 118 serious cross-border threat to health 109 HIV/Aids 112 International Health Regulations (IHR) amendments 106–7 key issues/conclusion 102, 118–19 legal basis 109 pandemic preparedness plans 116–17 pathogenic risk 105, 106, 115

263

risk management to crisis preparedness, reconfiguration 103–5 biomedical/security expertise, reconfiguration 103–4 innovation/modernisation, and new risks 104 preparedness approach 105 public health and infectious diseases 104–5 SARS 116 securitisation of public health risks 114–17 Health Security Committee (HSC) 115–16 H9/11 attacks’ role 114–15 swine flu 116–17 volcanic ash cloud 116 public health institute see under public health surveillance public health surveillance airborn toxins 137 Annual Epidemiological Report (AER) 139, 141 benchmarking 140–1 big data, use 130–2 biosecurity infrastructure 136, 140 Centers for Disease Control and Prevention (CDC) (US) 125–6, 128 and co-production 121, 141 communicable diseases 124 Europeanisation 137–8 concept 121–2 debate on 126–7 history of 125–6 normative agendas 127 theoretical emphasis 127 definition/meaning 123–5 diversity of 127–8 Ebola 134 Electronic Health Records (EHR) 132–3 emerging infectious diseases (EIDs) 133–4, 136 European Centre for Disease Control (ECDC) 137–40 Europeanisation 137–40 cognitive Europeanisation 139 Commission-ECDC relationship 139–40 communicable diseases 137–8 data collection/benchmarking 140 public health policy 138 Flu Trends controversy 131–2 globalisation 133–5 International Health Regulations, revised 135 justification changes 129–30 key issues/conclusions 12–13, 122, 141 Medisys 139 non-communicable diseases 124–5 power and expertise 122, 141

264

Index

public health institute 138 SARS 134–5 securitisation challenge/risk 136–7 surveillance definition 123 practical work 123–4 syndromic surveillance 130–2 technological change 130–3 variation 128–9 rational instrumental (RI) paradigm see under administrative constitutionalism rational-instrumental model of evidence-based regulation see under co-production of expert and executive power REACH Regulations see under nanotechnologies regulation recombinant regulation see synthetic biology, and recombinant regulation Regulating Europe (Majone) 49–50 regulation reframing see under behavioural expertise Regulatory Fitness and Performance programme (EC) see under behavioural expertise risk administration (EU), contested legitimation administrative legitimation 169, 170 control/deliberative models 170, 172–3, 187–9 Amflora (potato) case 173 chronology of authorisation process 174–5 Table control model role 181 deliberation to control (Commission’s role) 181–4 General Court, administrative legitimation 187–8 General Court scrutiny (Hungary v Commission) 185, 186–90 horizontal transfer 179, 181 plausibility proofs 179 request for authorisation 178–9 responses to criticism 179–81 risk assessment 178–81 background developments 167–8 Commission as risk manager, deliberation to control 181–4, 185–92 control model 170–1 deliberative model 171–2 European Food Safety Authority (EFSA) 171–2 as expert agency, control to deliberation 176–81, 189–90, 191–2 genetically modified organisms (GMOs) authorisations 169–70, 172, 173–84, 186, 191–2 Herculex (maise) case 173 boundary work process 177

chronology of authorisation process 174–5 Table controversial analogies 177 deliberation to control (Commission’s role) 181–4 General Court scrutiny (Pioneer action) 185–6 plausibility proof 176–7 potential risks 178 request for authorisation 176 risk assessment 176–8 uncertainty intolerant pattern 176 key issues/conclusions 168–9, 190–2 Pfizer case 167, 192 precautionary principle 167–8, 182, 186–7, 190–1 science’s role 168 risk assessment/risk management divide, administrative constitutionalism 47–9 risk governance co-productionist view 22–4, 36 constitutional role of science 24 definition of co-production 23 definition of risk 19 delegation 24–8 civic epistemology 27 democratic governance 24–5 epistemic delegation 25 European examples 27–8 United States’ example 25–7 EU as political formation 19–20 exogenous risk 22 faces of risk governance 19–22 framing of risk 20–1 harmonisation of regulatory standards 28–30 divergent scientific perspectives 29–30 GM/non-GM crop contamination (German example) 30 inclusion/exclusion 29 market importance 28–9 key issues 9–10, 13–14, 20–2 natural kind concept 19 political dimension 20, 22–3 science and technology studies (STS) 23–4 science’s place in EU governance 33–5 accountability constraints 34–5 impartial objectivity 33–4 legitimation role 33 presumed autonomy 36 tacit delegation principles 34, 35 unexamined role 35–6 subject-matter governance 20 subsidiarity 30–3 European/member state regulation 31–3 multilevel governance functions 30–1 US federal government experience 31

Index summary/conclusion 35–6 technical assessment of risk 21–2 see also administrative constitutionalism; co-production of expert and executive power risk management to crisis preparedness, reconfiguration see under public health expertise, and EU security risk regulation (health and safety) see under administrative constitutionalism Royal Society 27 SARS see under public health expertise, and EU security; public health surveillance SCENIHR (Scientific Committee on Emerging and Newly Identified Health Risks) see under nanotechnologies regulation Scharpf, Fritz 83, 93 science/expertise/politics, divide/connection see under European Ombudsman (EO) science’s place in EU governance see under risk governance securitisation of public health risks see under public health expertise, and mixing EU security Seville Process (European Integrated Pollution Prevention and Control Bureau) 221 Slaughter, Anne-Marie 107 Smart Regulation programme (EC) see under behavioural expertise socio-ethical deliberations, deferral see under synthetic biology, and recombinant regulation Sorenson, C 133 stakeholder groups see under European Ombudsman (EO) State Farm case see under US administrative law insights Stewart, Richard 172 subsidiarity see under risk governance Sunstein, CR 148, 160 surveillance see under public health surveillance swine flu see under public health expertise, and EU security syndromic surveillance see under public health surveillance Synthetic Biology 2.0 conference 65 synthetic biology, and recombinant regulation bioluminescent (light-producing/glow-inthe-dark) plants 63, 67–8 conferences/declarations 65–6 conventional biotechnology legislation, proposed use 66–7 Deliberate Release/Contained Use Directives 67–8 emerging problem, synthetic biology as 68–9

265 engineering in biology 62–3 European Economic and Social Committee, Opinion (2002) 77–8 executive power/legislative power, balance 66–7 existing regulation 64–6 anti/self-regulatory stance 65–6 techno-scientific development, concerns 64–5 Forecasting and Assessment in Science and Technology programme (EC) 78 genetically modified organisms (GMO)/ micro-organisms 69, 70–1, 79 key issues/summary 10–12, 60–2, 64 meaning/application of synthetic biology 62–3 nanotechnology link 75–6 New Techniques Working Group 69–72 genetically modified organisms (GMO)/ micro-organisms 69, 70–1 new techniques, agreed list 70 social/political/policy consideration 71–2 recombinant DNA technologies 59, 62 regulatory fit 59 socio-ethical deliberations, deferral 76–9 EU as ‘innovation union’ 78–9 historicisation of socio-ethical issues 77–8 techno-scientific development as driving force 78–9 Synthetic Biology: From Science to Governance (EC) 78 Synthetic Biology Working Group 72–6 arriving at/rationale for new definition 73–4 conventionality/novelty difference 74–6 rationale for new definition 73 science/politics demarcation 74 and standing scientific committees 72–3 updating of risk assessment methods 76

terrorism see under public health expertise, and EU security, biosecurity/health security, reconfiguration Thaler, RH 148 Thompson, EP 29 Trondal, J 46 Trump, President Donald J 26–7 Tversky, A 144, 148 UK v Parliament and Council 47 United Kingdom administrative state, debates over 42 expert executive power, learning process 45 United States administrative state, debates over 41–2

266

Index

Centers for Disease Control and Prevention (CDC) see under public health surveillance as epistemic delegation example 25–7 expert executive power, learning process 45 GMO labelling provision controversy 31 New Deal as catalyst 42 US administrative law insights Baltimore Gas case 200–2, 205 dissenting/plurality opinions 202 frontiers of science doctrine 200–1, 205 Benzene case 197–200, 202 Cancer Policy and plurality opinion 198–200 inconclusiveness 197–8, 200 OSHA’s new permissible exposure limit (PEL) 198 Bush administration policy 209–10 Chemical Manufacturers Association v EPA 207 Chevron case 204–6 administrative rationality 204–6 judicial deference grounds 204, 208 post-Chevron risk regulation 206–9 stationary source in industrial emissions, EPA interpretation 204 classical approach/account 193 abandonment 193–4, 196–7, 212 return to 209–12 co-production 215 contemporary approach/account 193–4 co-production 215 judicial review limitation 194, 213–15 scientific expertise/administrative power, complex relationship 213 early case law (courts of appeal) 195–7

classical approach (Fifth Circuit) 195–6 contemporary approach (DC Circuit) 196–7 differences between approaches 197 Edison Electric Institute v EPA 207 Ethyl Corp v EPA 210 expertise/executive power, relationship 193, 194–5 key issues/conclusion 193–5, 212–15 Massachusetts v EPA 210–12 Natural Resources Defense Council, Inc v EPA 206–7 State Farm case 202–3, 205 passive restraint systems in new automobiles 202–3 technical rationality reaffirmed 203 Supreme Court cases 197–206 Vermont Yankee case 200 Weberian model see classical approach above US Environmental Protection Agency 45 utilitarian (Mode 2) technoscience 26 Van Asselt, MBA 176 Vermont Yankee case see under US administrative law insights volcanic ash cloud see under public health expertise, and EU security Vos, E 47, 176 Wagner, Wendy 45 Waterton, Clair 27 Weberian model see US administrative law insights, classical account White, TH 44–5 Wynne, Brian 27