Positive Free Speech: Rationales, Methods and Implications 9781509908295, 9781509908325, 9781509908318

Citation preview

POSITIVE FREE SPEECH Freedom of expression is generally analysed as a bare liberty against restraint by state action. However, rationales underpinning freedom of speech very often imply that the concept also has important positive aspects, and that to be truly ‘democratic’ the modern polity requires more than negative freedom. In contemporary conditions, this understanding of free speech raises matters such as media diversity or pluralism, the concept of voice and access to the public sphere, access to information, and the need to rethink the audience in relation to public speech. Whether securing positive free speech is a matter of politics or of law, a task for legislatures or for courts, is an open question. On one level, any programme of inculcating positive dimensions of free speech might be understood as inherently polycentric and hence political in character. Yet, a number of jurisdictions evince enhanced legal recognition for the principle. The aim of this collection is to interrogate the rationales of positive free speech, to consider the political and juridical methods by which it has been or may be more fully reflected in the modern state, and to consider the range of practical contexts in which its valorisation has or would have significant implications. The contributors are drawn from an array of European and international jurisdictions, and include academic lawyers and communications researchers.

ii

Positive Free Speech Rationales, Methods and Implications

Edited by

Andrew T Kenyon and

Andrew Scott

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK 1385 Broadway, New York, NY 10018, USA HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2020 Copyright © The editors and contributors severally 2020 The editors and contributors have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2020. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication data Names: Kenyon, Andrew T., editor. | Scott, Andrew, 1974- editor. Title: Positive free speech : rationales, methods and implications / edited by Andrew T Kenyon and Andrew Scott. Description: Oxford ; New York : Hart, 2020. |  Includes bibliographical references and index. Identifiers: LCCN 2020004871 (print) | LCCN 2020004872 (ebook) |  ISBN 9781509908295 (hardcover) | ISBN 9781509908301 (Epub) Subjects: LCSH: Freedom of speech—Europe. Classification: LCC KJC5160 .P67 2020 (print) | LCC KJC5160 (ebook) |  DDC 342.408/53—dc23 LC record available at https://lccn.loc.gov/2020004871 LC ebook record available at https://lccn.loc.gov/2020004872 ISBN: HB: 978-1-50990-829-5 ePDF: 978-1-50990-831-8 ePub: 978-1-50990-830-1 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.

TABLE OF CONTENTS List of Contributors�������������������������������������������������������������������������������������������������������������������� vii 1. Complicating Freedom: Investigating Positive Free Speech ������������������������������������������������ 1 Andrew T Kenyon 2. Providing a Platform for Speech: Possible Duties and Responsibilities����������������������������� 11 Thomas Gibbons 3. Positive Protection for Speech and Substantive Political Equality������������������������������������ 25 Jacob Rowbottom 4. The Access to Information Dimension of Positive Free Speech������������������������������������������ 43 Andrew Scott and Abbey Burke 5. Promoting Civic Discourse: A Form of Positive Free Speech under the Constitution of Ireland?������������������������������������������������������������������������������������������������ 65 Eoin Carolan 6. The State of Affairs of Freedom: Implications of German Broadcasting Freedom����������� 83 Andrew T Kenyon 7. The Collective Speech Rights of Minorities������������������������������������������������������������������������� 97 Sally Broughton Micova 8. The Positive Right to Freedom of Expression and Party Anonymity in Legal Proceedings���������������������������������������������������������������������������������������������������������� 117 Merris Amos 9. Positive Free Speech and Public Access to Courts������������������������������������������������������������ 135 Judith Townend 10. Hiding the Truth in the Shadow of the Law? Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts����������������������������������������������� 153 Andrew Scott 11. Speaking and Governing through Freedom of Access to Environmental Information����������������������������������������������������������������������������������������������������������������������� 173 Mélanie Dulong de Rosnay and Laura Maxim Bibliography������������������������������������������������������������������������������������������������������������������������������ 191 Index����������������������������������������������������������������������������������������������������������������������������������������� 199

vi

LIST OF CONTRIBUTORS Merris Amos is Professor of Human Rights Law in the Department of Law at Queen Mary University of London. She has previously held posts at the University of Westminster, the University of Essex and the Australian Human Rights Commission. Her research concerns the protection of human rights at the UK national level and the relationship between national institutions and the European Court of Human Rights. Her publications include Human Rights Law (Oxford, Hart Publishing, 2014). Sally Broughton Micova is Lecturer in Communications Policy and Politics at the University of East Anglia and a member of its Centre for Competition Policy. Her research focuses on media and communications policy and regulation. She is a Research Fellow at the Centre for Regulation in Europe (CERRE) in Brussels; a Visiting Fellow at the London School of Economics and Political Science; and a Visiting Lecturer at the Institute of Communication Studies in Skopje, Macedonia. Before entering academia, she spent over a decade working in international organisations, and continues to serve as an occasional expert for the Council of Europe and for the Organisation for Security and Cooperation in Europe. Abbey Burke is an Australian solicitor practising in human rights law. She holds a Master of Laws (Public Law) (Distinction) from the London School of Economics, and earlier degrees from the University of New South Wales. Her research interests lie in freedom of expression and other human rights, and refugee and asylum law. Eoin Carolan is Full Professor in Constitutional Law at University College Dublin, where he is founding Director of the Centre for Constitutional Studies, and has been a visiting researcher at Harvard and UCLA law schools. He is co-author (with Ailbhe O’Neill) of Media Law in Ireland (Dublin, Bloomsbury, 2019) and has published widely on Irish constitutional law, constitutional theory, privacy and freedom of expression. Mélanie Dulong de Rosnay is Associate Research Professor at the French National Centre for Scientific Research (CNRS) and director of its Centre for Internet and Society (UPR 2000). Her research focuses on digital commons, regulation by technology, information technology law and policy. It includes network infrastructure as a commons, algorithmic regulation, peer production platforms, citizen science, distributed architectures, open access and open licensing of public sector information, scientific data and publications, and digital native heritage. She is a board member of Internet Policy Review and the OpenEdition publishing platform. Thomas Gibbons is Emeritus Professor of Law at the University of Manchester. His main research is in media and communications law and regulation, with a particular interest in media pluralism. His publications include Regulating the Media (London, Sweet & Maxwell, 1998) and (with Peter Humphreys) Audiovisual Regulation under Pressure: Comparative Cases from North America and Europe (Abingdon, Routledge, 2012). He is a founding editor of the Journal of Media Law.

viii  List of Contributors Andrew Kenyon is a Professor in the Melbourne Law School, University of Melbourne and past Director of its Centre for Media and Communications Law. He researches in comparative media law, including defamation, privacy and free speech. His books include Comparative Defamation and Privacy Law (edited) (Cambridge, Cambridge University Press, 2016), Defamation: Comparative Law and Practice (London, UCL Press, 2006) and Democracy of Expression: Positive Free Speech and Law (Cambridge, Cambridge University Press, forthcoming). Laura Maxim is based in the Laboratoire Interdisciplinaire Sciences Innovations Sociétés (LISIS) (UMR 1326/9003) at the French National Centre for Scientific Research (CNRS) and has expertise in systems ecology and ecological economics. Her research focuses on the production, use and communication of scientific knowledge in chemical risk governance, in regulatory advisory activities and in practices of innovation in green chemistry. Her scientific advisory activities include participation in expert committees at the French Agency for Food, Environmental and Occupational Health & Safety (ANSES) and the European Food Safety Authority (EFSA). Jacob Rowbottom is an Associate Professor at the Faculty of Law, Oxford and a Fellow of University College, Oxford. He writes on issues relating to freedom of speech, elections, political participation and media law. He is the author of Democracy Distorted (Cambridge, Cambridge University Press, 2010) and Media Law (Oxford, Hart Publishing, 2018) and is a co-editor of the Journal of Media Law. Andrew Scott is an Associate Professor in the Department of Law at the London School of Economics and Political Science. His research focus is the constitution of the public sphere, with current projects concerning laws of defamation and privacy, the interplay between defamation and religious faith, freedom of expression and access to information, corporate power and the public sphere, and the regulation of journalistic newsgathering practices. His publications include (with Gavin Millar) Newsgathering: Law, Regulation and the Public Interest (Oxford, Oxford University Press, 2016). Judith Townend is Senior Lecturer in Media and Information Law at the University of Sussex, where her primary research interests lie in the protection and resourcing of public interest journalism, and freedom of expression more broadly. Prior to joining Sussex, she was the inaugural director of the Information Law and Policy Centre at the Institute of Advanced Legal Studies, University of London. She has published in a range of academic journals and co-edited (with Steven Barnett) Media Power and Plurality: From Hyperlocal to High-Level Policy (Basingstoke, Palgrave Macmillan, 2015).

1 Complicating Freedom: Investigating Positive Free Speech ANDREW T KENYON

I. Introduction Freedom of expression is commonly understood as limiting state restrictions on speech. Freedom of speech is a negative liberty; it is freedom from external restraint. And that is clearly an important part of free speech. But communicative freedom can also be understood to have positive dimensions; to involve actions that support speech as well as the absence of limitations.1 That is, free speech is not only a formal freedom, but also a freedom that is effective, existing or realised, to some degree at least. The rationales commonly said to underlie free speech – such as knowledge, autonomy and democracy – involve more than a negative right; the absence of restrictions on speech is insufficient for such rationales. While supports for speech could be offered by a host of actors, positive approaches to free speech, like negative ones, often focus on the state. Thus, legal analyses of positive free speech often entail the state having obligations to act in support of the freedom, to act in support of the goals or rationales underlying free speech. This reframes communicative freedom and, as I suggest below, complicates it. In a paper that prompted the workshop from which this collection emerged, I wrote:2 There seems little reason to support the idea that free speech exists primarily when the state is not directly limiting speech. If the analysis [of positive free speech] has plausibility, then arguments about free speech would be reframed. Free speech would be understood not merely as a negative legal right, but also in positive terms. New questions would emerge about what legal obligations should be applied in the name of free speech and through what methods.3 1 Free speech, free communication, freedom of expression and communicative freedom are used interchangeably here, and communicative freedom is not used in the specific sense adopted by Jürgen Habermas: see, eg J Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans W Rehg (Cambridge, Polity, 1996) 119–20. 2 The workshop ran over two days in 2015 at the London School of Economics (where my co-editor Andrew Scott is based, and where I was a visiting research professor at the time). While the article I quote above was a prompt for the event, participants were invited to examine the issues from their own varied perspectives, critique them, correct them and extend them. Contributions have been further developed and updated since the event, and were finalised in 2019. Thanks to all who participated in the workshop, which extended well beyond those whose work is presented in this book. A personal thanks also to Andrew Scott and his colleagues at the LSE for the invitation to visit. 3 AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379, 379–80.

2  Andrew T Kenyon From that beginning, the chapters that follow examine a range of ways in which such ideas about free speech obligations and methods might be extended. The chapters’ interests can be grouped across three broad areas, concerning justifications, developments and openness: how positive free speech might be justified; free speech obligations that might be developed; and ways in which positive communicative freedom could support openness of justice, government and business information.

II.  Positive and Negative Dimensions to Freedom An initial point to note is that positive and negative are labels of convenience. The idea is not to divide free speech into two forms of freedom, but to draw out particular dimensions of communicative freedom. That also means positive does not replace negative communicative freedom; rather, both dimensions are significant. In later chapters, authors use a variety of terms for positive free speech – including strong free speech,4 the effective enjoyment of free speech5 or facilitative freedom6 – a variety which reflects the diversity in existing commentary. Labels that take the place of ‘positive’ include active, affirmative, collective, effective, empowering, enabling, facilitative and functional.7 ‘Negative’ can also be labelled defensive, formal, freedom from, literal, passive and subtractive.8 In many ways, each of the terms can be collapsed into its opposite. The state always acts in ways that affect speech; recognising positive dimensions to the freedom simply increases the range of actions that register as matters of free speech. For example, media subsidies, competition law, public media (its creation, goals, governance and funding), media ownership controls and funding transparency, editorial and journalistic independence, public

4 T Gibbons, ‘Providing a Platform for Speech: Possible Duties and Responsibilities’, this volume, chapter 2. 5 J Rowbottom, ‘Positive Protection for Speech and Substantive Political Equality’, this volume, ch 3. 6 A Scott and A Burke, ‘The Access to Information Dimension of Positive Free Speech’, this volume, ch 4. 7 eg S Breyer, Active Liberty: Interpreting a Democratic Constitution (Oxford, Oxford University Press, 2008); WE Hocking, Freedom of the Press: A Framework of Principle, A Report from the Commission on Freedom of the Press (Chicago, University of Chicago Press, 1947) (‘affirmative’ and ‘effective’); RB Horwitz, ‘The First ­Amendment Meets Some New Technologies: Broadcasting, Common Carriers, and Free Speech in the 1990s’ (1991) 20 Theory and Society 21 (‘collective’); L Stein, Speech Rights in America: The First Amendment, Democracy, and the Media (Urbana, University of Illinois Press, 2006) (‘empowering’); P Keller, European and International Media Law (Oxford, Oxford University Press, 2011) 413 (‘empowered autonomy’); WC Durham, ‘General Assessment of the Basic Law: An American View’ in P Kirchhof and DP Kommers (eds), Germany and Its Basic Law: Past, Present and Future (Baden-Baden, Nomos, 1993) 37, 45–46 (‘facilitative’); MK Curtis, ‘Democratic Ideals and Media Realities: A Puzzling Free Press Paradox’ (2004) 21 Social Philosophy and Policy 385 (‘functional’). 8 eg Stein (n 7) (‘defensive’); Curtis (n 7) (‘formal’); Hocking (n 7) (‘freedom from’); Horwitz (n 7) (‘literal’); Hodge (n 7) 152 (‘passive’); R Levy and G Orr, The Law of Deliberative Democracy (Abingdon, Routledge, 2017) 81–83, 98 (‘subtractive’). Freedom from and freedom to are most often linked to Isaiah Berlin. See I Berlin, Two Concepts of Liberty: An Inaugural Lecture delivered before the University of Oxford on 31 October 1958 (Oxford, Clarendon Press, 1958), with later clarifications in I Berlin, ‘Introduction’ in Four Essays on Liberty (Oxford, Oxford University Press, 1969) ix. Elsewhere, I explain why Berlin’s approach has limited relevance to free speech. Among other things, it ‘is problematic for understanding the potential of structural, communicative supports for democratic forms of government. His analysis considers limits that must be applied to democracy to preserve individual freedom; his purpose is not to consider whether democracy has other requirements, including communicative ones’: AT Kenyon, Democracy of Expression: Positive Free Speech and Law (Cambridge, Cambridge University Press, forthcoming).

Complicating Freedom: Investigating Positive Free Speech  3 demonstrations, access to information and journalists’ safety all implicate positive dimensions of free speech.9 Despite the variety of terms, positive and negative are commonly used (and debated) across scholarly research on freedom, well beyond law.10 Here, positive free speech and similar terms are used simply as shorthand for the positive dimensions of communicative freedom. Positive does not necessarily mean a positive right – whether the freedom has positive dimensions is one question, whether they come within a positive legal right recognised by courts is another. While positive is a common term in the wider literature on freedom, it is worth noting different senses in which the term is often used. In simple terms, positive might mean something like self-mastery or self-perfection – a life of positive freedom being a reflective, self-directed one. Or it might mean enablement or capacity – the conditions that allow an opportunity of freedom to be exercised are part of freedom, not something separate from it.11 There is more subtlety in the wider analysis, but for present purposes that is p ­ robably sufficient. The chapters in this collection lean towards the second of these two senses. What might positive free speech mean in law? The chapters that follow are rich and detailed in their responses; little would be gained by summarising them here. They warrant careful reading. That said, I want to note the range of subjects with which the chapters engage, before drawing out five themes that recur across them and are relevant to positive free speech more generally. In terms of subject matter, the chapters explore: • media access (and criteria for denying access), media pluralism and public debate, including for networked platforms;12 • political equality and campaign finance, private censorship, public forums, media regulation (including public ownership and plurality) and the roles of court, parliament and executive, including the need for courts to recognise positive dimensions of free speech when reviewing government action on these matters;13 • access to state-controlled information, contrasting European Court of Human Rights developments and the UK common law principle of open government;14 • promoting civic discourse through a bifurcated approach to constitutional free speech (comprising deontological-individually focused and consequentialist-public debate

9 See, eg E Barendt, Freedom of Speech, 2nd edn (Oxford, Oxford University Press, 2005) 100–08; S Fredman, Comparative Human Rights Law (Oxford, Oxford University Press, 2018) 349–53; M Tushnet, Advanced Introduction to Freedom of Expression (Cheltenham, Edward Elgar, 2018) 83–103. 10 See, eg S Fredman, Human Rights Transformed: Positive Rights and Positive Duties (Oxford, Oxford University Press, 2008); CC Gould, Rethinking Democracy: Freedom and Social Cooperation in Politics, Economy, and Society (Cambridge, Cambridge University Press, 1988); V Held, Rights and Goods: Justifying Social Action (New York, Free Press, 1984); M Ananny, Networked Press Freedom: Creating Infrastructures for a Public Right to Hear (Cambridge MA, MIT Press, 2018). 11 Compare, for example, the approaches to positive freedom of Berlin (n 8) (positive freedom more as mastery) and GC MacCallum Jr, ‘Negative and Positive Freedom’ (1967) 76 Philosophical Review 312 (positive freedom more as enablement); see Kenyon (n 8). 12 Gibbons (n 4). 13 Rowbottom (n 5). 14 Scott and Burke (n 6).

4  Andrew T Kenyon

•

•

• •

•

•

focused dimensions) and the state’s role in ‘promoting conditions in which access to information or ideas is not dependent upon the exercise of another’s will’;15 the free speech basis of plural forms of media in terms of financing, mission, control and so forth, drawing on the example of Germany’s constitutional law on broadcasting freedom;16 collective minority language rights and the resulting need for state-supported media, with ‘scarce’ state resources being ‘devoted more towards enabling participation parity in … domestic political life’;17 party anonymity in court proceedings, and ways in which positive dimensions of free speech could help strengthen and make more consistent the making of orders;18 access to court information and court communication to the public, and the argument that recognising positive free speech could help courts facilitate public access to and understanding of law;19 confidentiality clauses in public contracts, in which effective regulation and even appropriate oversight mechanisms remain challenging for law, and ways in which positive free speech and government accountability might assist;20 and access to environmental data held by companies, including being able to reuse and mine the data, for the purposes of public regulation and wider involvement in that regulation.21

The examples of positive free speech are not comprehensive, just as a collection focused on negative dimensions of free speech could not be comprehensive. But the aim is to illustrate how the implications of free speech and their application can be rethought in comparison with analyses taking a substantially negative approach to the freedom. There is a l­ongstanding strand of legal writing and wider literature that addresses positive free speech,22 but this

15 E Carolan, ‘Promoting Civic Discourse: A Form of Positive Free Speech under the Constitution of Ireland?’, this volume, ch 5. 16 AT Kenyon, ‘The State of Affairs of Freedom: Implications of Broadcasting Freedom in Germany’, this volume, ch 6. 17 S Broughton Micova, ‘The Collective Speech Rights of Minorities’, this volume, ch 7. 18 M Amos, ‘The Positive Right to Freedom of Expression and Party Anonymity in Legal Proceedings’, this volume, ch 8. 19 J Townend, ‘Positive Free Speech and Public Access to Courts’, this volume, ch 9. 20 A Scott, ‘Hiding the Truth in the Shadow of the Law? Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts’, this volume, ch 10. 21 M Dulong de Rosnay and L Maxim, ‘Speaking and Governing through Freedom of Access to Environmental Information’, this volume, ch 11. 22 As well as the examples in nn 7–10 above; see, eg JM Balkin, ‘Some Realism about Pluralism: Legal Realist Approaches to the First Amendment’ [1990] Duke Law Journal 375; CE Baker, Media Concentration and Democracy: Why Ownership Matters (Cambridge, Cambridge University Press, 2007); JA Barron, ‘Access to the Press: A New First Amendment Right’ (1967) 80 Harvard Law Review 1641; R Delgado, ‘First Amendment Formalism Is Giving Way to First Amendment Legal Realism’ (1994) 29 Harvard Civil Rights-Civil Liberties Law Review 169; TI Emerson, The System of Freedom of Expression (New York, Random House, 1970); OM Fiss, The Irony of Free Speech (Cambridge, MA, Harvard University Press, 1996); MA Graber, Transforming Free Speech: The Ambiguous Legacy of Civil Libertarianism (Berkeley, University of California Press, 1991); AC Hutchinson, ‘Talking the Good Life: From Free Speech to Democratic Dialogue’ (1989) 1 Yale Journal of Law and Liberation 17; T McGonagle, ‘Positive Obligations Concerning Freedom of Expression: Mere Potential or Real Power?’ in O Andreotti (Council of Europe Task Force for Freedom of Expression and Media) (ed), Journalism at Risk: Threats,

Complicating Freedom: Investigating Positive Free Speech  5 collection has been prompted by the sense that further consideration is warranted. Indeed, as several chapters suggest, the contemporary communications environment might make positive communicative freedom more significant.23 Positive free speech remains relatively marginal in English language literature, and there is value in exploring why the freedom’s positive dimensions might be valuable, the sorts of things they encompass, how they might be pursued and some of the challenges in doing that. Given the ways in which negative free speech is often ‘entrenched’ and appears to be ‘the natural state of affairs’,24 in many common law jurisdictions at least, there is merit in merely suggesting that free speech has positive dimensions and considering questions ‘about the purpose, scope and permissible extent of government action’.25 All this complicates freedom. Free speech issues arise across more domains, there are more actors and actions relevant to it, and the issues can interact in a complex manner. In many ways, free speech law and analysis is already complicated, even when the focus is on negative free speech. The US First Amendment, for example, is stringently simple in its wording – ‘Congress shall make no law … abridging the freedom of speech’ – but the US law that has emerged and continues to evolve is anything but straightforward. Even so, one can well say negative communicative freedom is a ‘convenient simplicity’ arising from ‘an incomplete idea of liberty’.26 It is comparatively straightforward, but it is too narrow in its focus and is based on dubious assumptions.27 It can be the easier path for analysis, but its price is ‘ignoring the practicalities’ of speech.28 It is a fiction. Complicating matters also involves recognising that dealing with both positive and negative dimensions of the freedom would not make speech in some absolute sense ‘free’. It is always a contextual and relative freedom, subject to varied social infrastructures, practices and discursive contexts.29 Jacob Rowbottom offers one way to assess the legal salience of some of those wider influences – such as the silencing effect of social pressures that can limit how often people raise unpopular ideas in public speech (or limit how well the ideas are heard by others). He suggests that one might ask whether those wider influences result from the abuse of concentrated communicative power.30 Where that is the case, there is a free speech issue to consider.

Challenges and Perspectives (Strasbourg, Council of Europe Publishing, 2015) 9; F Schauer, ‘Positive Rights, Negative Rights, and the Right to Know’ in DE Pozen and M Schudson (eds), Troubling Transparency: The History and Future of Freedom of Information (New York, Columbia University Press, 2018) 34. 23 See also eg PM Napoli, Social Media and the Public Interest: Media Regulation in the Disinformation Age (New York, Columbia University Press, 2019) 191–93; T Wu, ‘Is the First Amendment Obsolete’ (Knight First Amendment Institute, 2017), https://knightcolumbia.org/content/tim-wu-first-amendment-obsolete; Ananny (n 10). 24 Gibbons (n 4). 25 Carolan (n 15). 26 Gibbons (n 4). 27 See, eg Kenyon (n 3). 28 Gibbons (n 4). 29 See, eg B Peters, Public Deliberation and Public Culture: The Writings of Bernhard Peters, 1993–2005, ed H Wessler, trans K Tribe (Basingstoke, Palgrave Macmillan, 2008) 81; NJ Hirschmann, The Subject of Liberty: Toward a Feminist Theory of Freedom (Princeton, Princeton University Press, 2002). See also Broughton Micova (n 17), who draws on Nancy Fraser’s public sphere analysis: N Fraser, ‘Rethinking the Public Sphere: A Contribution to the Critique of Actually Existing Democracy’ (1990) 25/26 Social Text 56. 30 Rowbottom (n 5).

6  Andrew T Kenyon

III.  Five Themes: Multiple Actors and Actions; Speakers and Audiences; Equality or Parity; Democracy; and Structural Conditions While a wide range of questions are examined in the chapters that follow, various points recur. It is not that authors always deal the same way with these issues, but they illustrate the matters that arise when free speech’s positive dimensions are taken into account. Here, I draw out five such themes. First, when free speech includes positive dimensions, multiple actors and actions are relevant.31 As noted above, freedom becomes complicated. Free speech does not only concern the legal review of government restrictions on speech. There are choices to make with regard to the freedom’s positive dimensions, and they necessarily involve discretion. What public funding is appropriate for public service media? How should competition law be applied to media entities? What limits and supports are appropriate with regard to public demonstrations? In short, who should do what in terms of free speech? Merely noting those examples suggests it is unlikely to be a matter for any one actor alone, and that so many of the actions that are taken will involve discretion. Legal analysis of positive free speech often considers the role of parliaments and courts,32 along with that of independent regulators.33 In this collection, chapters generally consider the importance of both parliaments and courts to free speech’s positive dimensions, with the legislature seen as well placed to decide which out of many possible actions supporting free speech are pursued. It is thought they can better consider the wide range of possibilities, their interactions, financial implications and so forth. At the same time, parliamentary power (which authors recognise may at times be subject to executive direction) is open to abuse. Governments can support speech in a manner that furthers their own political ends rather than freedom of communication as such. That would hardly be surprising of such political actors. Thus, courts have a vital role in reviewing actions that are taken and doing so with an understanding of both positive and negative dimensions of communicative freedom.34 If only negative dimensions of the freedom are recognised by courts, plausible efforts to support positive free speech can be overruled as infringing rather than supporting freedom of communication. Concerns include that courts will focus on formal or negative qualities of the right in question, with the risk of undermining effective freedom of speech,35 or will overreach their abilities or expertise if they are more assertive.36 However, quite what is entailed in the role of courts varies in the analyses. Even the German Federal Constitutional Court would say many choices about broadcasting freedom are for legislatures to make, but that court is undoubtedly significant in shaping what the freedom means.37 And the results in Germany would have been quite different if the 31 eg Townend (n 19). 32 eg Rowbottom (n 5), Scott and Burke (n 6), Carolan (n 15), Kenyon, ‘State of Affairs’ (n 16), focusing on one or both of parliaments and courts. 33 eg Gibbons (n 4), Dulong de Rosnay and Maxim (n 21), noting the position of regulators. 34 eg Rowbottom (n 5). 35 ibid. 36 eg Scott and Burke (n 6). 37 Kenyon (n 16).

Complicating Freedom: Investigating Positive Free Speech  7 court had not maintained so much of its positive approach to communicative freedom for nearly 60 years. While both actors are significant in that example, the court’s role can well be called active. It is a role the court justifies as being necessary to support the free formation of individual and collective opinion, which is required for a legitimate political process and is protected by the constitutional provision on freedom of speech and broadcasting. Similarly, European, UK and Irish courts have been active in developing positive dimensions of free speech in particular contexts.38 The examples add nuance to the frequent conception of positive dimensions of free speech as something to be supported politically rather than a right that is judicially enforceable. (The same point arises in relation to positive human rights in general, and the question is also debated in that literature.39) One of the courts that has recognised positive dimensions of free communication is the European Court of Human Rights. Its decisions are noted in some chapters, but are relatively tangential to the authors’ concerns.40 The decisions are more significant in Andrew Scott and Abbey Burke’s comparison of European rulings on access to information and the UK Supreme Court’s development of a parallel common law principle of open government. Second, analyses tend to note free speech interests of both speaker and audience and treat media freedom as derivative of them. Whether speaker or audience is emphasised, both are relevant to the freedom: free speech is not only an interest of speakers. This may be linked to with republican approaches to democracy and the role of ‘both speaker and listener interests in the process of civic engagement’,41 or social (rather than individual) conceptions of agency.42 It also means that the freedom in relation to media is derivative of free speech in general, and media might be regulated in particular ways to further free speech. The point extends to contemporary media, such as large data businesses operating search and social media.43 In that context, too, media’s concentrated communicative resources need to be exercised for the benefit of others and be publicly accountable.44 Media are ‘part of the structural conditions that are necessary for the effective operation’ of the right to free speech that is held primarily by others.45 Yes, the media can act as a participant in public speech, but it is also, significantly, a platform for civic discourse.46 Third, issues of communicative equality or parity, or at least freedom from domination, emerge in several chapters.47 Substantive political equality can be a basis for positive free speech rights, with an ideal of equal distribution of ‘access to communicative resources’ guiding evaluation of various forms of state intervention to support opportunities to communicate.48 The aim can be to ‘enable discursive spaces’ for minority languages that are

38 See Scott and Burke (n 6), Carolan (n 15). Many wider examples can be given, notably France, which I examine closely in Kenyon (n 8). 39 See, eg Fredman (n 10). 40 eg Rowbottom (n 5); Amos (n 18). 41 Carolan (n 15). 42 Gibbons (n 4). 43 eg Gibbons (n 4); Rowbottom (n 5); Carolan (n 15). 44 Rowbottom (n 5). 45 Carolan (n 15). 46 ibid. 47 See Rowbottom (n 5); Carolan (n 15); Broughton Micova (n 17); Dulong de Rosnay and Maxim (n 21). Gibbons (n 4) takes a more minimal approach, arguing for the freedom to extend an invitation to pursue a conversation more than some form of parity in public debate. 48 Rowbottom (n 5).

8  Andrew T Kenyon ‘needed to achieve participation parity’ in a democracy,49 or ‘to create a closer-to-level playing field’ between companies, regulators and the public in relation to environmental risks.50 Fourth, as with free speech writing more generally,51 the analyses here often have a democratic focus.52 While not ignoring matters such as the cultural aspects of media ­pluralism,53 their attention tends to centre on democratic and political speech in a wide sense. What speech is included by that democratic focus varies, but it is not just electionrelated54 and can extend to broad conceptions of civic discourse or individual and public will formation.55 Chapters also consider changed democratic practices – such as the emergence of ‘monitory democracy’,56 and continual forms of democracy not centred on periodic elections – and changed ideas about citizenship, such as recognition of minority language groups within a polity.57 Older roots clearly exist for positive communicative freedom; republican democratic theories offer just one example.58 But ideas such as monitory democracy illustrate how the democratic value of positive free speech can become clearer in contemporary conditions. If ‘government is [to be] held accountable not just at the polling place on election day but continuously’, and by a wider variety of actors, including ‘news media, think tanks’ and ‘non-governmental organizations that take themselves to be watchdogs of government with respect to specific policy domains’,59 then matters of pluralism, public discourse and access to information are clearly significant. Fifth, positive dimensions move free speech beyond a formal freedom. They bring into focus the structural conditions underlying speech and they can change approaches to access to information and limits on speech. Attention moves from formal freedom alone to consider ‘the “architecture” or conditions which allow speech to occur’.60 There are questions about what speech actually happens and power over the means of communication.61 There is interest in ‘securing the structural conditions for civic discourse’62 or creating ‘an enabling environment for the exercise’ of minority speech rights ‘through positive state intervention’.63 And there is a need for access to media, information and ideas.64 The approach brings more ‘into’ free speech, such as the effects of people’s differing capacity and resources to speak, and (in many analyses) private controls and limitations on 49 Broughton Micova (n 17). 50 Dulong de Rosnay and Maxim (n 21). 51 See, eg Barendt (n 9) 20. 52 eg Gibbons (n 4); Rowbottom (n 5); Scott and Burke (n 6); Carolan (n 15); Broughton Micova (n 17). 53 Gibbons (n 4); Broughton Micova (n 17). 54 eg Rowbottom (n 5); Scott and Burke (n 6). 55 eg Carolan (n 15); Kenyon (n 16). 56 See, eg M Schudson, The Good Citizen: A History of American Civic Life (New York, Martin Kessler, 1998); M Schudson, The Rise of the Right to Know: Politics and the Culture of Transparency, 1945–1975 (Cambridge MA, The Belknap Press of Harvard University Press, 2015); J Keane, The Life and Death of Democracy (London, Simon & Schuster, 2009); J Keane, Democracy and Media Decadence (Cambridge, Cambridge University Press, 2013); Scott and Burke (n 6). 57 Broughton Micova (n 17). 58 As Carolan (n 15) notes, republican democratic theories provide an example. 59 Schudson, The Rise of the Right to Know (n 56) 25. 60 Dulong de Rosnay and Maxim (n 21). 61 Gibbons (n 4). 62 Carolan (n 15). 63 Dulong de Rosnay and Maxim (n 21). 64 eg Gibbons (n 4); Rowbottom (n 5); Scott and Burke (n 6); Carolan (n 15); Amos (n 18); Townend (n 19); Dulong de Rosnay and Maxim (n 21).

Complicating Freedom: Investigating Positive Free Speech  9 speech.65 Some of the analysis echoes the ‘realist’ approach to free speech in some 1990s US scholarship,66 but there is a wider recognition in the positive free speech literature of the importance of the framework, architecture or structural conditions for speech. This can also change what limits on speech are legal. It can suggest, for example, new approaches to the analysis of confidentiality clauses, especially when used by public bodies,67 and to the treatment of companies’ environmental information as not being merely their property, but having importance for public knowledge and the possibilities of public regulation.68 Or, if the ‘value of expression is the discourse it promotes’, there is less scope – but not no scope – for content-based restrictions on speech. Rather, limitations would be assessed in terms of the ‘impact of the [disputed] expression on discourse’.69 This point from Eoin Carolan about a new approach to assessing limitations in some ways mirrors Allan Hutchinson’s 1980s work on dialogue and a dialogic community in terms of free speech, which suggested ‘the state will have to intervene’ to support speech ‘and the intervention will be based on the value of dialogue itself ’.70 The point is not only suggested in that chapter; rather, the point illustrates the longer tradition which this book seeks to draw from and extend. At the least, the authors collectively suggest that positive dimensions of free speech should now be considered more often and more fully in scholarly analysis and in law, and they ask you to read all that follows to that end.



65 Gibbons

(n 4); Rowbottom (n 5). eg Balkin (n 22). 67 Scott (n 20). 68 Dulong de Rosnay and Maxim (n 21). 69 Carolan (n 15). 70 Hutchinson (n 22) 25. 66 See,

10

2 Providing a Platform for Speech: Possible Duties and Responsibilities THOMAS GIBBONS

Any attempt to advance a positive conception of speech will quickly encounter opposition based on traditional theory and media interests, bases so well entrenched that they appear to be the natural state of affairs. It is likely that the political feasibility of recognising the practical implications of adopting a positive view of speech is low, and change will take some time. But change will require the highlighting of different ways of thinking about speech rights, and that is the aim of this chapter. The reason for doing so is that the pattern of the traditional free speech argument is deficient. As Kenyon observes,1 it is often confined to the bare (negative) liberty to speak in the absence of government regulation, the presumption being that that is sufficient for an empirical state of free speech to exist. That presumption, in turn, depends on assumptions that equality exists in speech without government action, that rational debate will take place and allow truth to emerge and be identified, and that government intervention to protect speech would do more harm than non-intervention. All these assumptions have limited support. Further, where the deficiencies of a negative free speech right are recognised, the pattern of the traditional argument is to supplement it with bundles of various additional rights or policies in order to make it more socially palatable. This is because it is commonly conceded that an important reason why the negative liberty does not deliver an empirical state of free speech is that there are major social structural obstacles, in the shape of various kinds of power over the means of communication, to everybody’s speech being equally effective. In these circumstances, the foundation of the freedom as a negative liberty requires reconsideration. It will be suggested here that an examination of the nature of a liberty suggests that ‘positive’ dimensions of speech – what kinds of speech in fact take place – are actually an integral part of any principle of free speech. Such a ‘strong’ conception of free speech has the potential, however, to intrude into the activities of both individuals and commercial enterprises. The challenge will be to show that extended rights based on a strong conception of speech, together with associated policies for a more active form of media pluralism, can be made to work. In doing so, the underlying assumption is the desirability of a well-functioning democratic culture. Contemporary democracy entails, at least, that



1 AT

Kenyon, ‘Assuming Free Speech’ (2014) 77 Modern Law Review 379.

12  Thomas Gibbons citizens have the resources to take part equally in the political decision-making process,2 with the trend in recent democratic theory being to stress the deliberative or discursive aspects of such participation. That emphasis is found in discussion about the appropriate role of the media in a democracy,3 and that role will be considered here also.

I.  Deficiencies in the Traditional Conception of Freedom of Speech As indicated, a serious problem with the traditional conception of freedom of speech is that it tends to highlight only government interference with an individual’s wish to speak. While protection against government censorship is indeed an important condition for realising an empirical state of free speech, this approach thereby passes over the practical problem that private censorship, individual capacity and resource problems can also be significant impediments to speech. This is not to say that the problem is entirely ignored; rather, it is not regarded as within the provenance of a right to free speech. It is seen as a collateral issue, to be tackled in other ways. Those ways include recourse to bundles of rights or to media pluralism policy. Thus, Emerson wrote of the rights ‘to hear, inquire, associate’ as supplementary measures to complement the negative liberty of speech, and he favoured positive support and encouragement of free speech by means of providing facilities for speech, reducing distortions and making information widely available.4 Barron advocated a right of access to the media for similar reasons,5 as did Meiklejohn,6 in arguing for policy that would not amount to a right to be heard but would ensure that the substance of significant points could be aired. Not least, Schauer has acknowledged that the US judicial formulation of the First Amendment in negative terms is not an exclusive way of supporting the First Amendment values of obtaining knowledge and enabling democratic participation.7 Again, Lichtenberg has identified a ‘multiplicity of voices’ strand in free speech doctrine that justifies positive measures to increase media diversity in order to promote basic free speech values.8 It is interesting that none of these writers, nor more recent mainstream works on freedom of speech,9 have sought to reconsider the form of the liberty to speak. A much more significant failing is that the traditional notion of freedom of speech is based on an incomplete idea of liberty. By focusing only on interference with the intended activity, it diverts attention away from the nature of that activity, that is, the relevant meaning of ‘speech’.

2 See D Held, Models of Democracy, 3rd edn (Cambridge, Polity Press, 2006); JS Dryzek and P Dunleavy, Theories of the Democratic State (Basingstoke, Palgrave, 2009). 3 See J Keane, The Media and Democracy (Cambridge, Polity Press, 1991); CE Baker, Media Concentration and Democracy: Why Ownership Matters (Cambridge, Cambridge University Press, 2007); J Curran, Media and Democracy (London, Routledge, 2011). 4 TI Emerson, The System of Freedom of Expression (New York, Random House, 1970) 4. 5 JA Barron, Freedom of the Press for Whom? The Right of Access to the Mass Media (Bloomington, University of Indiana Press, 1973). 6 A Meiklejohn, Political Freedom: The Constitutional Powers of the People (New York, Harper, 1960). 7 F Schauer, Free Speech: A Philosophical Enquiry (Cambridge, Cambridge University Press, 1982). 8 J Lichtenberg, ‘Foundations and Limits of Freedom of the Press’ (1987) 16 Philosophy & Public Affairs 329. 9 eg E Barendt, Freedom of Speech, 2nd edn (Oxford, Oxford University Press, 2005).

Providing a Platform for Speech: Possible Duties and Responsibilities  13 While there is some discussion of the kinds of communicative acts that should be included in the definition of speech (that is, whether it be words, symbols or actions),10 it is not envisaged that the freedom should require consideration of what the speaker wishes to achieve by speaking. That narrow interpretation may seem rather odd, but its conceptual provenance can be traced to the distinction, most notably drawn by Berlin,11 between negative and positive liberty, between freedom from and freedom to do. By defining freedom in terms of rights against relatively obvious and tangible obstructions to an individual’s preferences, the scope of a right is not vulnerable to being dependent on contestable interpretations of the meaning of liberty. Further, that may be considered all the more important if such interpretations could be imposed through the power of government. It is therefore easy to appreciate the advantages of defining freedom of speech negatively, as a freedom from government interference, and to claim that that is the proper understanding. To follow that approach, however, is to accept convenient simplicity too readily, without considering the ways that liberties function in the real world.12 Half a century ago, Gerald MacCallum offered a cogent argument13 against the claim that there are two basic ‘kinds’ of freedom, negative and positive, and that one should be regarded as better than the other. He insisted that the use of the concept of freedom is only intelligible when it is realised that it always involves freedom ‘of something (an agent or agents), from something, to do, not to do, become or not become something; it is a triadic relation’.14 To take the typical route of concentrating on only two aspects of this, using the distinction ‘freedom from’ and ‘freedom to’, distracts attention from the important issues at stake. That is because it seems to imply that there are two different freedoms when there are not, but it also has the effect of emphasising one aspect of the relation against the other, without acknowledging that both (together with the third) are features of every case of agents’ freedom. MacCallum argued that it would be more illuminating to investigate what constitutes the components of the triadic relation instead of asking what the nature of ‘freedom’ is. Each of those components – agents, the character or actions which they seek to obtain, and the impediments or obstacles to achieving them – will span a range of variables, and it is only by assessing the arguments in favour of a particular conception of each of them that it will be possible to evaluate the relative merits of different social and political arrangements.15 Not only does any claim about liberty involve these three dimensions, there are a number of permutations for each.16 These insights are not, however, prominent in the freedom of speech literature. It is perplexing that the traditional approach does not really analyse the nature of the expressive activity that we might be free to engage in. It might be thought that even a general discussion of positive freedom of speech, if not expressed as the active ‘objectives’ component of the

10 ibid ch 3; Schauer (n 7) ch 7. 11 I Berlin, ‘Two Concepts of Liberty’ in I Berlin, Four Essays on Liberty (Oxford, Oxford University Press, 1969). 12 This section develops discussion in T Gibbons, ‘Free Speech, Communication and the State’ in M Amos, J Harrison and L Woods (eds), Freedom of Expression and the Media (Leiden, Martinus Nijhoff, 2012) 19. 13 GC MacCallum Jr, ‘Negative and Positive Freedom’ (1967) 76 The Philosophical Review 312. 14 ibid, 314 (original emphasis). 15 ibid, 320–27. 16 See also A Swift, Political Philosophy, 2nd edn (Cambridge, Polity Press, 2006) 52–54.

14  Thomas Gibbons triadic relation, would illuminate the scope of the right. But the distinction between negative and positive liberty is typically invoked to make the assertion that the negative element should dominate the formulation of the right,17 with its positive dimensions regarded as being within the province of political debate. As a consequence, the negative approach to freedom of expression is a very abstract way of thinking about the liberty. The major consideration is non-interference, yet, without knowing what it is that is being prevented from happening, the nature and significance of non-interference cannot be appreciated. The intelligibility of ‘freedom of speech’ depends on what would make sense to the agent who is claiming the freedom, and that depends on the way that the ‘active’ element is defined, in terms of what an interest in speech might be and why it might be considered valuable and worthy of protection. The beneficiary of the traditional approach to freedom of expression, however, appears to be a rather abstract representation of a human agent, who would seem to be content provided that there is no interference in a mere utterance, whether conveyed or not, and, indeed, whether intelligible or not. A more realistic depiction of human agents is as people who require more than this to satisfy their interest in speech. Such people will value speech or other forms of expression because it is through language and symbolic forms that thought can be articulated and their identity manifested. This is why being able to speak is so commonly understood to be a basic right. But it is usually taken to extend beyond the mere freedom to articulate, except perhaps for those who wish to ignore or reject human contact. Rather, human agents normally live in a social setting, and what typically gives meaning and purpose to their lives, and establishing and maintaining their identity, is the possibility of communicating their thoughts to others.18 At the same time, as already mentioned, most discussion of freedom of speech contains the implicit assumption that ‘speech’ involves more than the minimum requirement for the negative liberty of a mere expressive noise or gesture. What may be considered curious is that the general expectation, that speech will be communicated with a view to discussion or debate, is not actually incorporated into the free speech right. That is so notwithstanding that the rationales for giving special protection to speech (facilitating the discovery of truth, participating in democracy and enhancing autonomy) are invoked to support it because of the benefits that discussion provides through the exchange of ideas and opinion.

II.  The Basis for a Strong Conception of Freedom of Speech If the traditional conception of speech has such a narrow scope, what kind of activity should be the object of protection under a stronger conception of freedom of e­ xpression? 17 Schauer (n 7) 114. 18 This social nature of agency is most strongly emphasised in the work of Charles Taylor; see C Taylor, Sources of the Self: The Making of the Modern Identity (Cambridge MA, Harvard University Press, 1989). For criticism of a political culture inhabited by ‘atomistic’ individuals, see W Kymlicka, Liberalism, Community and Culture (Oxford, Oxford University Press, 1989); J Rawls, Political Liberalism (New York, Columbia University Press, 1993). For a similar line of thought, but drawing on Moon, see JA Chandler, ‘A Right to Reach an Audience: An Approach to Intermediary Bias on the Internet’ (2007) 35 Hofstra Law Review 1095, 1101 and 1130, highlighting the ability to communicate. See also D Mac Síthigh, ‘From Freedom of Speech to the Right to Communicate’ in ME Price, SG Verhulst and L Morgan (eds), Routledge Handbook of Media Law (London, Routledge, 2013) 175.

Providing a Platform for Speech: Possible Duties and Responsibilities  15 (Henceforth, this expression is preferred to the idea of ‘positive’ speech, in order to reflect the more unified, triadic approach discussed earlier.) The libertarian, individualistic perspective protects only the isolated actions of the speaker and imposes no obligations on others; any demand for reciprocation will be seen as an encroachment on the others’ interests. Thus, it is said that freedom to speak is balanced by freedom not to listen or to read. As a consequence, the protected activity is confined to a disengaged act of expression, whether or not it is paid attention, let alone acknowledged. From the perspective of a social view of human agency, this kind of freedom of speech is not only conceptually narrow, but is also practically useless. If the meaning and purpose of human activity is indeed identified by reference to interaction with others, the significant freedom to be protected in relation to speech is the freedom to be part of the social interchange that is the point of speech activity. A stronger and more socially useful conception of speech must go beyond the ability to make a lone utterance, but how far should its scope extend? At one end of the spectrum of possibilities, it could enable at least the basic steps towards starting a conversation. At the other end, it could require a full exchange of views and critical deliberation. One advantage of the traditional approach is that it gives priority to those who wish to be protected against unwanted speech, and that kind of wish cannot be ignored. Yet it may be that not all preferences against speech are legitimate in a normal social setting, where communicative interaction is fundamental to the individual’s proper functioning in society. Excluding somebody from even attempting such interaction requires greater justification than the mere wish to avoid any kind of engagement. Such justification would include the desire to avoid speech activity which does not compromise the speaker’s social identity, for example, unsolicited commercial speech, such as cold telephone calls, spam messages, leafleting in the street or junk mail. They may be regarded as sufficiently harmful or lacking in value to social identity to merit outright rejection. It is not so clear that it should be possible to avoid entirely attempts to initiate conversation about political viewpoints or religious beliefs, even in normal social interaction. Further, considering the matter at a higher plane, that would be especially so in a well-functioning democracy, because the ability to close down possible dialogue before it has even been attempted is inconsistent with the associated value of openminded willingness to consider different points of view. On the spectrum of possible formulations of a (strong) liberty to speak, coercing somebody to engage with a speaker’s wish to communicate is so intrusive that it compromises the listener’s own preferences about how to function in a social setting. To preserve the speaker’s interests in that same setting, I have suggested previously that the significant interest to be protected in a strong conception of freedom of speech is the freedom to seek out speech activity.19 However, that requires some further explanation. What it includes is the freedom to attract attention to the wish to speak to another about a stated topic, or, more simply, the freedom to extend an invitation to pursue a conversation. It does not demand that the request for attention, with a view to possible engagement, should be reciprocated. It has the effect of raising awareness that a speaker’s point of view is available, but without entailing that the person who is approached should react to that information. However, it may be noted that, once a conversation is joined, in certain organised settings, such as a

19 Gibbons

(n 12).

16  Thomas Gibbons discussion group or a political debate, there may be social or institutional expectations for further dialogue to take place.

III.  Media Implications: The Practicalities of Access So far, the discussion here, to lay a basis for a stronger conception of freedom of speech, has focused on the personal aspects of human interaction. It may be thought that any advance that has been made is only marginal, and probably unexceptionable if it essentially does no more than protect a minimal level of activity associated with normal social behaviour. Nevertheless, that protection cannot be taken for granted. While the stronger right might be unlikely to be regarded as intrusive in practice, its main function could be to act as a defence, for example, against possible claims of harassment by way of seeking conversation. However, a stronger conception of freedom of speech could also apply to writing, publishing and dissemination in other media. The ‘speech’ activity of human agents who function in a social environment would therefore include their writing and publishing with a view to attracting the attention of readers and audiences. Here, a strengthening of the free speech right would have more significant implications because it would serve to challenge some of the structural obstacles that communications media erect against speakers who want to attract wider audiences for their ideas. In doing so, it would raise difficult questions about the use of media and communications resources, especially those allocated by private commercial enterprises. These issues were explored in the 1970s in ‘access to the media’ debates,20 where they were ultimately considered as being non-compelling in terms of a negative conception of freedom of speech, or simply impracticable. The consequence has been an impasse because, in the absence of some stronger free speech right, the solution to structural obstacles to wider communication lies in policy initiatives through legislative intervention – for which there has been little political appetite. Acceptance of a strong conception of freedom of speech allows the question of access to media platforms to be reconsidered. Three issues will be examined: its application to different kinds of media; possible criteria for access; and relatedly, the place of partisan restrictions on access to particular viewpoints.

A.  Application to Different Kinds of Media In enabling access to media platforms, public service organisations appear well placed to take a lead, because their objects typically include support for values of social inclusion and democracy. Supporting a free speech right to attract attention for viewpoints is consistent with those objectives and, indeed, provides an additional rationale for preserving public service media. More controversial is the claim that private commercial enterprises should make provision for access to their platforms in support of such a stronger conception. There are good reasons, however, for imposing requirements to enable the minimal level of accessibility to an audience which is needed by a strong conception of freedom of speech.

20 A

right most notably advocated by Barron (n 5).

Providing a Platform for Speech: Possible Duties and Responsibilities  17 To the extent that they hold themselves out as providing a public forum, commercial media may be expected to provide facilities entailed by that role. But even if they do not make such an explicit representation, contemporary media have a de facto function as the preferred mechanism for reaching wide audiences. Furthermore, notwithstanding that they generally have that function, not for public reasons but to maximise revenue, they thereby exercise considerable power over communication.21 In a well-functioning democracy, major sources of power cannot be immune from scrutiny and intervention directed at controlling or shaping their behaviour towards public ends. Here, the object of supporting the strong conception of freedom of speech takes priority over media organisations’ commercial claim to be able to use their private resources as they think fit. Furthermore, it seems clear that the use of those private resources cannot plausibly be regarded as an exercise in institutional free speech that can be invoked as a negative liberty to resist public intervention.22 These points indicate that private media organisations can be expected to modify their behaviour to allow what will be a minimal level of accessibility to their users. In fact, much media use is already an integrated experience, with newspaper and broadcast material complemented with reader and audience comments through web-based forums and blogs. Such facilities reduce the financial burden on private operators of allowing access, not least because of the low additional marginal costs of widening access to a large number of speakers.

B.  Possible Criteria for Access Extending the strong conception of free speech to the media requires some differentiation between different modes of communication. To the extent that media content is selected by means of an editorial process, the focus will be on attracting the attention of the relevant decision maker. To the extent that material is distributed on a more non-selective basis, for example, across the internet and including social media, the focus will be on mechanisms to try to ensure that the speaker’s voice is not lost amidst the multiplicity of messages being carried. In both cases, the speaker’s wish to communicate with a mass audience is accepted as a legitimate goal. Although the strong conception of freedom of speech is rooted in a social interpretation of human agency, there seem no grounds for restricting the ambition of speakers to the relatively closed circles of face-to-face encounters. Where a wider audience is sought, a public platform is needed for the speaker to get noticed, what Fiss has described as the ‘street-corner’ tradition in free speech doctrine.23 In relation to the media, the analogous requirement is the facility for speakers to draw the audience’s attention to the fact that they have something to say. To do that, they have to confront the media’s power to select what is included in their content and to determine how it is conveyed. The need to control what is included in media output is obviously the consequence of the media not being an unlimited resource for communication, so that content has to be precisely mediated. In conventional discussion of access to media in the context



21 See

D Freedman, The Contradictions of Media Power (London, Bloomsbury, 2014). O’Neill, A Question of Trust, BBC Reith Lectures 2002 (Cambridge, Cambridge University Press, 2002) ch 5. 23 OM Fiss, ‘Free Speech and Social Structure’ (1986) 71 Iowa Law Review 1405, 1410. 22 O

18  Thomas Gibbons of a negative conception of free speech, it has been noted that the media’s choice to exclude content does not interfere with that right, because it is limited only to being able to utter an expression – and that can take place without recourse to the media. It follows that there are no liberty constraints on how the media use their resources to favour one source of content over another. It also follows that those who are concerned about the use of such media power have to resort to broader policy measures to effect control over it. However, those measures, such as ownership controls or competition regulation, are largely directed at the structure of the media industry and do not address the way that the media select their material. A strong conception of freedom of expression changes the position. In the face of a claim to be able to solicit audience interest, an entity which serves as a public platform for speech, whether it holds itself out as such or the situation arises de facto, needs to account for why it allows some and not other viewpoints. This does not mean that it cannot refuse certain kinds of access; the strong free speech right does not at all entail full access for every willing speaker. But there is a particular interest in the criteria for denying access. One kind of unacceptable reason would be discrimination against the views of particular groups of people. On the other hand, equal access for all viewpoints would not be practically possible. Another possibly unacceptable reason would be the popularity of a viewpoint; here, social media’s creation of trends, by aggregating users’ preferences, creates difficulties for the visibility of speakers with minority viewpoints. Because the available resources are limited, the important focus must be on the process of mediation. It is physically impossible to accommodate the nuances of every speaker’s opinion, yet if mediation is too drastic, it prevents speakers from indicating that they, personally, are associated with a point of view. For that reason, proper access to the media would not be satisfied if it did not apply to every individual but only ensured that, in Meiklejohn’s words, elaborating his right to be heard, ‘everything worth saying shall be said’.24 Essentially, a strong conception of freedom of speech requires that editors (in the widest sense) give serious attention to speakers’ wishes to connect with their audiences, readerships or users. This implies a relatively minimal access right, entailing practical measures for the media to be receptive to individual approaches, to enable individual speakers to associate themselves with sets of ideas, if not to reproduce their own precise formulation of their ideas, and to distribute that information to their audiences, readers or users. This form of access would impose a burden of moderation on the media, but that may not be as daunting as it first appears, given the widespread use of the internet to provide and manage blogs and discussion forums, and the ability to categorise information into themed threads. A corollary of this activity would be the need to signpost the availability of such an extended range of material, alongside the mainstream content selected by media editors.

C.  Partisan Restrictions on Content One feature of commercial media, which is particularly valued by the press, is its freedom to hold and espouse partisan viewpoints and to determine the scope of its coverage of topics.

24 Meiklejohn

(n 6) 28.

Providing a Platform for Speech: Possible Duties and Responsibilities  19 A media access right, even in the limited form outlined here, would constrain that freedom. However, the justification for doing so goes to the root of the politics of freedom of speech. The implication of a stronger conception of freedom of speech is that structural obstacles to drawing attention to one’s viewpoint should not be tolerated. Nevertheless, some real difficulties should be acknowledged. Should the media provide (even limited) access for every silly idea, or any idea that is inconsistent with their own editorial standpoint on the topic? Should specialist media be required to respond to approaches that are irrelevant to their subject theme? In the context of negative freedom of speech, a newspaper may say that it caters for a particular readership, perhaps representing a particular shade of political opinion, and that those who think differently should go elsewhere. But the problem is that those others do not have an opportunity to address the readership and to attempt to question the self-reinforcing formation of opinion that it represents. Resolution of these difficulties depends on the purposes and functions of the media in question. As suggested earlier, if they effectively communicate to the public at large and hold themselves out as being available to any user who might be interested in their content, the access right will apply to them because it supports the strong free speech right of the public at large to attract attention for its viewpoints. There is no requirement that media should endorse any point of view that they convey, still less one with which they disagree.

IV.  Strong Freedom of Speech and Media Pluralism As already mentioned, media pluralism policy is an important way of seeking to remedy the deficiencies in the negative approach to free speech. It helps to increase the pool of varied ideas in the hope of sparking a discussion. Rather than relying on the mere absence of state regulation to produce a range of viewpoints for speakers and audiences to consider, policy intervention can ensure that a plausible framework exists. Media pluralism policy has two main rationales: the provision of a wide range of sources that can be used for democratic discussion, and the ability of different groupings in a democratic society to express their cultural and value differences in media content. As the Council of Europe has put it, it involves: Political pluralism, which is about the need, in the interests of democracy, for a wide range of political opinions and viewpoints to be represented in the media. Democracy would be threatened if any single voice within the media, with the power to propagate a single political viewpoint, were to become too dominant. Cultural pluralism, which is about the need for a variety of cultures, as reflects the diversity within society, to find expression in the media.25

Media pluralism policy complements freedom of speech because it seeks to enhance the purposes of free speech that are considered valuable. Yet, it has a curiously passive quality, one which also reflects the negative conception of free speech that it supports. All it requires is that information should simply be accessible, which means that members of a

25 Council of Europe, ‘Recommendation No R(99)1 of the Committee of Ministers to Member States on Measures to Promote Media Pluralism: Explanatory Memorandum’ (Strasbourg, Council of Europe, 1999) (original emphasis).

20  Thomas Gibbons democratic society should be exposed to the range of different viewpoints that exist and should have their viewpoints adequately represented.26 The Council of Europe has stated, ‘it should be stressed that pluralism is about diversity in the media that is made available to the public, which does not always coincide with what is actually consumed’.27 In addition, the European Union’s High Level Group on Media Freedom and Pluralism has noted that it ‘encompasses all measures that ensure citizens’ access to a variety of information sources and voices, allowing them to form opinions without the undue influence of one dominant opinion forming power’.28 Yet this is a narrow formulation when compared to its intellectual roots in political pluralism. Indeed, the general pattern of media pluralism policy mirrors that adopted in traditional free speech theory: the reason for the policy assumes that certain benefits occur but without showing that the policy is actually responsible for them. The interesting point about different perspectives on political liberalism is that they anticipate that different parties in a stable political culture will engage with each other, whether to agree common ground, to reach a practical accommodation with each other or to achieve greater mutual understanding of different positions.29 Yet media pluralism policy is basically passive, providing little more than provision to increase the mere presence of differing points of view in the public domain. There is no requirement that the differing perspectives should connect with each other, and they may not even be aware of the existence of alternative views. It parallels the negative conception of free speech by envisaging somewhat abstract recipients of information who do not use it in their social settings. There is no constructive provision which moves beyond a right to do no more than merely speak, but to join debate. A more active approach to media pluralism would enable political pluralism to be taken seriously and would support and facilitate the kind of dialogue that is needed for a pluralistic democratic society to function effectively. That support would include policy measures to encourage conversation, dialogue and understanding engagement between different strands of opinion and identity. The link with a stronger conception of freedom of expression is clear; active media pluralism would be a policy extension of the stronger free speech right, both reflecting an understanding of human agents as social beings for whom interaction with others is the normal means of securing personal objectives and fulfilment. Those ideas suggest that, if pluralism is to be taken seriously, a wider and more proactive basis for policy is needed. A good starting point is acknowledgment that a key feature of the media’s role in a democracy is to provide an information resource for citizens so that they can weigh choices between alternative options. Another key aspect of its role is to provide a platform for standpoints to be expressed. Ideally, media can explain and ­mediate standpoints, and facilitate engagement between them. What is to be required is

26 See Council of Europe, ‘Media Diversity in Europe’, Report prepared by the AP-MD (Advisory Panel to the CDMM on Media Concentrations, Pluralism and Diversity Questions H/APMD (2003) 001 (Strasbourg, Council of Europe, 2003); L Hitchens, Broadcasting Pluralism and Diversity: A Comparative Study of Policy and Regulation (Oxford, Hart Publishing, 2007). 27 Council of Europe, ‘Recommendation’ (n 25) para 3. 28 Commission of the European Communities, ‘A Free and Pluralistic Media to Sustain European Democracy’, Report of the High Level Group on Media Freedom and Pluralism (Brussels, European Commission, 2013) 13. 29 For discussion, see T Gibbons, ‘Active Pluralism: Dialogue and Engagement as Basic Media Policy Principles’ (2015) 9 International Journal of Communication 1382.

Providing a Platform for Speech: Possible Duties and Responsibilities  21 policy ­intervention to ensure that the viewpoints available in the media represent the range of ­different interests and communities in civic society; citizens’ perspectives should be dominant, so that they should be aware of the diversity of viewpoints and have access to them; and ideas promoted by any single media organisation should be open to challenge by an equivalent other.30 These requirements serve to enhance the limited media access right discussed earlier. Having gained access to a media platform, it is inconsistent with the rationale for enabling that access to allow different points of view simply to sit side by side.

A.  Ways of Implementing Active Pluralism A number of measures could be made available to improve the active dimension of media pluralism policy.31 The most basic approach to pluralism is to provide a wide range of content, which the user is expected to discover and explore. A minimal improvement would be for the media organisation to take the initiative in providing effective navigation and signalling that makes the user aware of what is available. More active media could take a stronger role in bringing together the representatives of different points of view, either to enable them to engage in dialogue with each other independently or to facilitate discussion on the media platform. That discussion might be face-to-face or mediated by journalists in a documentary format. This implies proactive management by editors, presenters or moderators if it is to be more than a parade of alternative perspectives that do not engage with each other. Promoting engagement and provoking debate would be the most active kind of media involvement, and could be seen as the culmination of policy to support a positive view of speech. For a democracy to function effectively as a legitimate basis for political decision making and to reflect the social agency of citizens, some form of public dialogue is needed to ensure that different positions are properly taken into account in a process of reasoning which leads to decisions that are acceptable to all. The interesting question is whether the media should try to stimulate engagement between speakers by drawing out the implications of different viewpoints and highlighting potential controversies. From an individual perspective, it may be sufficient that a speaker can express affiliation to a particular point of view without having to be tested for its cogency. But from an institutional, democratic perspective, to the extent that deliberation is the ideal way of validating agreement about differences and their practical resolution,32 standpoints do need to be challenged to their limits. Here the media can have a much stronger role. While public service media may appear to be the natural bearers of that role, commercial media could be asked to do much more to assist exchange and understanding between different points of view.

30 For discussion of UK policy and relevant practical options, see S Barnett and J Townend (eds), Media Power and Plurality: From Hyperlocal to High-Level Policy (Basingstoke, Palgrave Macmillan, 2015). For discussion of the policy implications of possible risk variables to be taken into account, drawing on work undertaken to develop a European Union Media Pluralism Monitor, see P Valcke, M Sükösd and RG Picard (eds), Media Pluralism and Diversity: Concepts, Risks and Global Trends (Basingstoke, Palgrave Macmillan, 2015). 31 Gibbons (n 29). 32 See JS Dryzek, Deliberative Democracy and Beyond: Liberals, Critics, Contestations (Oxford, Oxford University Press, 2000).

22  Thomas Gibbons

V.  Conclusion: Strong Speech and Media Pluralism as Positive Speech The idea of positive speech serves to correct a significant imbalance in analysis of freedom of expression. Rather than dwelling on state censorship of underspecified speech activity, it starts by examining the context and purpose of what happens when speech takes place. That enables a better appreciation of the obstacles to speech and the measures needed to remove them. This chapter has argued for an even sharper focus on the shape of the free speech right itself. Discussion of positive speech has tended to take the traditional negative liberty of speech as given and has offered various policy measures to ameliorate its negative effects. Here, the claim is that a realistic understanding of individuals as social agents leads to a stronger formulation of the freedom that should be protected. Furthermore, that enhanced right requires supplementary policy measures themselves to be strengthened, for, if it is accepted that the primary interest in freedom to speak is the capacity to engage in dialogue with others, a strong conception of freedom of speech therefore consists of the freedom to attract attention to the wish to speak to another about a stated topic. To supplement that, an active conception of media pluralism requires the media to act proactively to stimulate searching and critical dialogue. Contemporary concern about the implications for democracy of so-called ‘fake news’, on the internet and especially social media,33 serves to illustrate the themes of this chapter and the importance of thinking about freedom of expression in a different way.34 Formally, the deliberate dissemination of false information is an exercise in traditional free speech. The difficulty is that such speech seems especially resistant to the traditionally espoused method of challenging its veracity, namely, being exposed to opposing viewpoints. This is because, analogous to the control exercised by owners and editors in different types of publishing media, the very process of dissemination tends to exclude contributions from those who may disagree. It is typically rapid and extensive, in the context of an internet advertising model that rewards clicks on the most attention-seeking claims, and it may be targeted to particularly sympathetic and receptive users identified through their social media or internet profiles. As a result, the voices of dissenters are unlikely to be noticed. Yet their negative free speech rights, strictly, remain undiminished, even if they appear largely devoid of purpose. A strong conception of free speech also holds that the state is justified in intervening to secure protection, and that such intervention should extend to private entities’ control over speech activities. The latter is especially important, because it is in the selection of

33 For recent discussion, see LSE Truth, Trust and Technology Commission, ‘Tackling the Information Crisis: A Policy Framework for Media System Resilience’ (London, LSE, 2018); European Commission, Independent High Level Group on Fake News and Online Disinformation, A Multi-Dimensional Approach to Disinformation (Brussels, European Commission, 2018); Digital, Culture, Media and Sport Committee, ‘Disinformation and “Fake News”’, Final Report, Eighth Report of Session 2017–19, HC 1791 (London, House of Commons, 2019). 34 For a recent analysis drawing on traditional freedom of expression doctrine, see I Katsirea, ‘“Fake News”: Reconsidering the Value of Untruthful Expression in the Face of Regulatory Uncertainty’ (2018) 10 Journal of Media Law 159. Katsirea warns against the dangers of using the criminal law or regulatory interventions to curb fake news and thereby allowing government or corporations to be the arbiters of truth. The strong conception of freedom of expression advocated here is not inconsistent with that concern.

Providing a Platform for Speech: Possible Duties and Responsibilities  23 material to include in media content, whether in the press, via broadcasting or on the internet, that the power to exclude speakers from coming to attention is most evident. A negative conception of free speech does not have the conceptual apparatus to deal with that problem. Intervention to protect a stronger conception of free speech will entail the development of a regulatory scheme to provide remedies for failure to comply. While regulation is not novel, nor especially controversial, in relation to media pluralism, it may be more of an iterative process in relation to facilitating access to the media. But the underlying principle is sufficiently clear, requiring a measure of accountability for the exercise of power over the prominence that is given to speakers’ points of view, whilst acknowledging some practical difficulties for the media in allocating their resources in that way. At the same time, media pluralism policy will need to be expanded to achieve its immanent aspiration to enhance democratic activity, and there will be a continued and necessary role for public service media to take a lead in both areas of speech activity. Of course, it would be much easier to take no action and to follow the traditional approach of ignoring the practicalities of making speech work, and the most powerful speakers will want to say that that is the better route. But a democratic polity has to take responsibility for its public policy, rather than be passive because it involves difficult decisions. The awakening of new interest in positive speech is a welcome start to obtaining a place for free speech policy on the political agenda.

24

3 Positive Protection for Speech and Substantive Political Equality JACOB ROWBOTTOM

The effective enjoyment of free speech rights requires more than an absence of state interference. Even if government does not censor a speaker, a lack of communicative resources may nonetheless stop a speaker from being heard. This observation alone does not take us far in understanding any obligations on the state to provide resources to facilitate the exercise of expression rights. It is one thing for the state not to stop me from doing something, but another to demand that it assist me in that activity. To make a case for positive speech rights, we have to ask why freedom of expression is important in the first place and see if the answer provides convincing reasons for requiring the state to undertake positive measures to support expression rights, as well as to refrain from censorship. Freedom of expression has frequently been justified as a political right necessary for the functioning of a democracy. The democratic justification for expression rights has been dominant in UK and European jurisprudence.1 The argument runs that without free speech, people will lack the information necessary to make voting decisions, to hold government to account and to participate in the political process. However, in a democracy, what matters is not only that government does not curb our efforts to participate, but also that citizens have the chance to participate as equals. The fair workings of the democratic process can be jeopardised if people have a much greater chance to influence political decisions on account of their wealth or because they control the main channels for communication. State action can enable all citizens to have an opportunity to participate in collective decisions, while stopping others from having a disproportionate chance to influence the outcome. This chapter will argue that the need to secure political equality supplies an argument for positive speech rights. After setting out the basic argument, the chapter will examine the jurisprudence under Article 10 of the European Convention on Human Rights (ECHR), to show that the legal protection of positive speech rights is limited at best. Rather than calling on the courts to mandate further positive measures, this chapter will argue that there are a number of legislative steps and policies that can be pursued. These measures, however, pose a number of challenges and carry some risks. In particular, such measures allow government to make



1 See,

eg Lingens v Austria (9815/82) (1986) 8 EHRR 407.

26  Jacob Rowbottom decisions determining how political resources should be distributed, and there is a danger that government will misuse this power for its own advantage. To promote one person’s positive speech right may entail interfering with the speech of another. While being alert to the dangers of such measures and the risk of abuse, the courts should be careful not to block genuine efforts to promote opportunities for expression rights to be enjoyed. It will be argued that the courts should be sensitive to the demands of political equality and avoid the prioritisation of formal negative speech rights over the positive.

I.  Substantive Political Equality and Expression A case for the positive protection of freedom of expression can be based on arguments for substantive political equality.2 This approach requires a roughly equal distribution of certain political resources, so that each participant has an approximately equal chance to influence political decisions.3 This approach does not demand equal influence, as there is nothing undemocratic about a more persuasive argument influencing a greater number of people.4 Instead, an egalitarian political system requires that citizens have the equal opportunity to influence collective decisions. Speech provides an important means for political participation in which people seek to influence others and hear what others have to say. So, while not every argument is equally persuasive, the starting point for substantive political equality is that people have equal amounts of certain resources necessary to put forward their arguments. If freedom of expression is necessary for democratic participation, then it is important that people have access to the resources necessary to participate by exercising that right. Substantive political equality does not demand that people have access only to sufficient resources, but takes as its ideal the equal distribution of certain political resources.5 This means that the state might not only be required to provide positive support for speakers, but might also be expected to curb the unequal communicative power of others. This is important in relation to political influence, as the democratic process is used to make decisions that will bind others. Unequal communicative resources within the democratic process can therefore amount to a form of unequal power over other citizens.6 Furthermore, political equality may be especially important in a system where significant inequalities are to be found in other spheres. Joshua Cohen has written that we might find economic equalities less ‘troubling in a world of moral equals’ because ‘equality is already expressed through the equal standing of the individuals as citizens in the system of authoritative ­collective

2 See J Rowbottom, Democracy Distorted (Cambridge, Cambridge University Press, 2010), from which this chapter draws. The term ‘substantive political equality’ is taken from D Estlund, ‘Political Quality’ (2000) 17 Social Philosophy and Policy 127. 3 The resources under consideration in this paper are primarily wealth and property rights that affect communicative opportunities. 4 R Dworkin, Sovereign Virtue: The Theory and Practice of Equality (Cambridge, MA, Harvard University Press, 2000) 197. 5 See A Swift, Political Philosophy, 3rd edn (Cambridge, Polity, 2013) 198 on the difference between sufficient and equal. 6 H Brighouse ‘Democracy and Inequality’ in A Carter and G Stokes (eds), Democratic Theory Today: Challenges for the 21st Century (Cambridge, Polity, 2001) 52, 62; Rowbottom (n 2) 42.

Positive Protection for Speech and Substantive Political Equality  27 ­decision making’.7 Similarly, a political process that is dominated and influenced by economic inequalities is less likely to confer legitimacy to those government policies that reproduce and maintain those economic inequalities.8 Of course, a completely flat distribution of political resources would be impractical and undesirable. It is only certain political resources that should be distributed equally, and in this chapter I focus on access to communicative resources. Furthermore, in some cases, an unequal distribution of communicative resources may be consistent with substantive political equality – but on the condition that the larger concentrations of political resources are exercised for the benefit of others and the decisions on the use of those resources are in some way accountable to the public. To elaborate, imagine a scheme in which everyone is given an equal basic unit of communicative resources, which a person can use to advance their personal interests and preferences. Under this scheme, a person can exceed that basic unit of communicative resources, but on the condition that the use of the excess resources benefits the public. To put it in the terms of free speech theory, the basic equal unit of communicative resources is justified from the perspective of the speaker, while any communicative resources above that basic amount are justified from the perspective of the audience.9 The mass media provides an example. The concept of the mass media suggests a one-to-many model of communication that is seemingly incompatible with equal political resources. However, the difference in communicative power provides a reason why media freedom tends to be justified as a means of serving the audience, rather than the speaker.10 The audience-based justification for media communications means that we might subject the media to accountability mechanisms that would not be expected of an individual speaker. Such controls seek to constrain the exercise of concentrated political resources and ensure that the use of such resources is to some degree for the public benefit. The role of the media in serving the public can thereby provide some reconciliation with the demands of equality. If the case for the positive protection of expression rights is based on political equality, then that protection will apply only to political expression, whichever way that is defined. This focus on political speech and equality provides a limit in that the state is expected to secure the effective exercise of speech only insofar as it is necessary for a fair political process. This is far-reaching enough, but does not lead to a demand for the state to facilitate all speech at all times. The arguments for substantive political equality do not extend to, say, commercial advertising. Inequalities in the capacity to communicate are a generally accepted part of the economic sphere. Of course, there is nothing to stop the state providing resources to aid artistic or commercial expression if it wishes to do so. Support for such speech, however, does not rest on an argument for substantive political equality. Substantive political equality is not limited to equality in the electoral process, though that may be its most obvious application. The mechanisms of political influence that 7 J Cohen, ‘Money, Politics and Political Equality’ in A Byrne, R Stalnaker and R Wedgwood (eds), Fact and Value: Essays on Ethics and Metaphysics for Judith Jarvis Thomson (Cambridge MA, MIT Press, 2001) 47, 53. 8 The argument is referred to by Estlund (n 2) 129, although he goes on to refute this type of egalitarianism. 9 On the difference between speaker-based and audience-based justifications, see E Barendt, Freedom of Speech, 2nd edn (Oxford, Oxford University Press, 2005). 10 For example, in the first instance decision in R (on the application of Miranda) v Secretary of State for the Home Department [2014] EWHC 255 (Admin) [46], Laws LJ distinguished between media freedom and freedom of expression, insofar as the former is justified with reference to the rights of audience whereas the latter looks at the interest of the speaker.

28  Jacob Rowbottom connect citizens with government take place throughout the political cycle and not just in an ­election campaign. A pragmatic challenge to this approach is that substantive political equality, if applied to political communication in general, becomes an incredibly demanding standard. However, it is important to remember that equality is just one factor to be considered in the design of a political system, and the balance struck with other factors will vary according to the context. The way equality is promoted outside the context of an election will therefore be very different to the way that it is promoted during an election campaign. For example, election campaign spending in the UK is subject to caps to reduce the degree of inequality. A cap on political speech outside of elections would be undesirable, overly complex and impractical. Instead, the state can provide positive support in a number of ways, for example, in the provision of forums and structures that allow people to communicate and through the provision of certain subsidies. This conclusion might make the equality argument look much the same as the the argument for sufficient (as opposed to equal) resources in practice. However, by thinking in terms of equal resources as the ideal, the trade-off with other goals is made transparent and it allows us to recognise the problems of inequality even where there is no simple regulatory solution. The approach taken here stands or falls depending on whether one accepts the case for substantive political equality. Some defend only a formal and negative account of political equality, which requires only an absence of state censorship. If the justification for freedom of expression rests on suspicion of government, then one would naturally be sceptical of an account of positive speech rights that requires government support. The very measures taken to correct inequalities in the opportunities for expression would, on this view, be seen as interferences with expression. Such a formal approach, it is argued here, is undesirable in that it elevates the suspicion of the state to such a level that it precludes collective action to ensure the effective enjoyment of speech rights and to address threats to speech coming from non-state sources. In any event, the political system itself is never a matter of just formal protection of rights. The democratic process is the product of various state structures, for example in setting up polling stations and vote counting procedures during an election.11 The issue is not deciding whether the state should intervene or not, but to focus on the merits of the particular type of state intervention. It will be argued below that concerns about abuse of state power can be addressed without prioritising a formal approach to equality.

II.  Private Censorship If a positive right demands that people have the means and resources to speak and communicate in practice, then the state may be required to remove or limit restraints on speech imposed by private actors. Positive speech rights can thereby demand protection from private censorship. While private censorship is a powerful term, its meaning is often unclear. The term normally refers to the exercise of some privately held power to silence others. That, in turn, begs the question of what exercises of private power should be subject to free 11 Similarly, in relation to free speech, Cass Sunstein observed that ‘public authority creates legal structures that restrict speech’ including the property rules that give a media owner the power to exclude other speakers – see C Sunstein, ‘Free Speech Now’ (1992) 59 University of Chicago Law Review 255, 271–72.

Positive Protection for Speech and Substantive Political Equality  29 speech considerations. The traditional negative protection of speech avoids this difficult question by focusing only on the actions of the state. A positive approach to speech protection, by contrast, raises the difficult question of what the ideal system of communication and the proper distribution of communicative resources should look like.12 A broadcaster’s decision to blacklist certain political parties from panel discussion shows or a search engine’s decision to de-list the sites of a particular speaker might be thought of as censorship. Yet the question remains why such actions should be regarded as censorship rather than the exercise of the private organisation’s own speech rights. The broadcaster or search engine may claim the blacklist is an exercise of editorial autonomy. Private censorship thereby raises a dispute between two private actors both relying on speech rights. The question is which right to prioritise? Political equality provides a way to break this deadlock. If all communicative resources were distributed equally, then one private speaker would not have to share his or her political resources with other speakers. The situation changes when significant political inequalities arise. The owner of large-scale political resources, such as the mass media or a public forum, has greater opportunities to communicate than most other speakers. When the owner of the resource seeks to exclude a speaker, the owner’s speech claim is not supported by arguments of equality. Instead, the holder of the concentrated power seeks to use the speech right to maintain that unequal power. To be compatible with political equality under the approach outlined earlier, the speech rights of the owner need to be justified in terms of the benefits to the audience and the public more generally (as opposed to asserting a speakerbased interest). By contrast, the speaker seeking access to the communicative resource has his or her claim backed by equality as well as free speech arguments. Focusing on political equality shows that the issue is not a conflict between two equally weighted speech rights, and provides a framework for determining which rights are most deserving of protection. By thinking about the issue in this way, we can see certain cases of private censorship as the exploitation of inequalities in communicative power at the expense of other speakers. Measures to curb private censorship can thereby redistribute communicative power from the controller of the forum to the individual speaker. Under this approach, the rights of the owners of concentrated resources give way to demands of the public interest. To count as private censorship, the private body must usually exercise some form of special power that is not a legitimate basis to restrict or limit the political rights of others. Political inequality is one factor that can render private power non-legitimate in this context, but it is not the sole basis for understanding a private action as censorship. Private censorship can arise where an employer dismisses or disciplines an employee for engaging in political speech that is completely unrelated to the employer’s activities.13 More broadly, violence is not a legitimate source of power, so a chilling effect on speech generated through a threat of violence could also constitute private censorship. Such an effect might arise where a person’s home becomes a target for organised vandalism because of what that person said. Obviously,

12 Though Thomas Gibbons notes that the apparent simplicity of the negative approach rests on a number of assumptions rather than a clear view about ‘the nature of the actions which we might be free to engage in’ – see T Gibbons, ‘Free Speech, Communications and the State’ in M Amos, J Harrison and L Woods (eds), Freedom of Expression and the Media (Leiden, Martinus Nijhoff, 2012) 19–44. 13 For discussion of the impact of employer decisions on freedom of expression, see P Wragg, ‘Free Speech Rights at Work: Resolving the Differences between Practice and Liberal Principle’ (2015) 44 Industrial Law Journal 1.

30  Jacob Rowbottom the issue there is not one of equality, as the problem would not be remedied by equalising the power to vandalise.14 Such actions are never a legitimate response to expression. Political equality is just one way to identify private censorship, but it is not the only way. The account of private censorship given above raises some difficult questions in relation to the silencing of speech through social pressure. What if a person feels afraid to speak out for fear that his or her view will be widely criticised and ridiculed? Can the speaker consider himself or herself to be privately censored? It is surely legitimate for people to voice their opposition to a viewpoint. While a person may feel under pressure not to speak, that pressure comes not from concentrated communicative power, but from the aggregation of many voices. Where a response is not based on the exercise of some concentrated or non-legitimate power, but is merely the expression of opposition, then it is harder to see it as censorship. On this view, people are free to speak and others are free to react to that speech. Considering how others will react to what we say is a part of social life and is an important constraint. However, we should not be too dismissive of such threats that arise from non-concentrated private power. For example, certain forms of hate speech might invoke associations and symbols of societal power that generate a hostile environment, and could be seen as a form of censorship. Another example is online shaming through a mass of individual comments, which may be thought to unfairly inhibit speech because the social effects of the reaction are disproportionate to the initial statement. The problem is not one of concentrated power, but rather the power of a majority or active minority. This is not the place to resolve such difficult examples. The discussion shows that the concept of private censorship, while intuitively appealing, is also very complex. This complexity does not, however, detract from the central point that political equality nonetheless provides a useful way to understand those types of private censorship discussed earlier that are exercised through the control of communicative resources.

III.  Judicial Promotion of the Positive Aspects of Free Speech So far, the discussion has focused on the obligation on the state in terms of democratic and free speech theory. There are, however, circumstances where the state can be legally obliged to protect positive speech rights. The European Court of Human Rights (ECtHR) has stated that the ‘effective exercise’ of expression rights ‘does not depend merely on the State’s duty not to interfere, but may require positive measures of protection, even in the sphere of relations between individuals’.15 So the ECtHR has found the state to be under obligations to take positive steps to protect speakers from violent reactions to expression,16 to protect protestors from counter-demonstrations17 and to protect employees from being unfairly

14 Although one could attempt to work this into an equality framework, given that the power is being used to interfere with the individual’s exercise of their own basic unit of political influence, in the same way that voter intimidation undermines equality in voting power. 15 Gundem v Turkey (2001) 31 EHRR 49, [43]. 16 ibid. 17 Ollinger v Austria (76900/01) (2008) 46 EHRR 38. While the reasoning rests on Art 11, the court noted that the same issues arise under Art 10.

Positive Protection for Speech and Substantive Political Equality  31 punished by an employer on account of expressive activities.18 None of these positive duties are open-ended and should not be ‘interpreted in such a way as to impose an impossible or disproportionate burden on the authorities’.19 Even in these obvious cases of private censorship, the courts tend to require positive steps only in the more extreme cases. The positive obligations mandating equal access to communicative resources are limited. For example, Article 10 of the ECHR is unlikely to require the state to set aside large sums of money for expressive purposes or create any new forums for speakers, unless to do otherwise would have a dire effect on the opportunities for expression. To require such measures is more likely to be a disproportionate burden on the state, which will be left a margin of appreciation. Positive rights to communicative resources are more likely to be found in relation to an existing forum or facility that the speaker seeks to access. That, however, raises the question of when the decision to deny access or impose a condition on access is not a legitimate use of property or editorial judgment, but instead amounts to a form of censorship. In other words, which communicative resources are people entitled to access? When that claim is to access property, the ECtHR has indicated that Article 10 will be engaged only when to refuse access would destroy the essence of a right.20 Such destruction might arise where there are no alternative means to communicate with the targeted audience. That approach requires the state to prevent the dominance of a forum owner, rather than to secure a level of equality. The Article 10 jurisprudence requires only a minimal threshold of communicative resources to be provided. The position may be a little different if the speaker seeks access to a publicly owned forum that has a communicative purpose. In Gubi v Austria, the Austrian Ministry of Defence’s decision to exclude a magazine critical of military life from a list of titles to be distributed in army barracks under a publicly funded scheme was found to violate Article 10.21 While the government did not hold a monopoly on the sources of information for those within the barracks and alternative means were available, the exclusion interfered with expression rights as it ‘was bound to have an influence on the level of information imparted to the members of the armed forces’.22 While that finding seems to impose a significant obligation on the state, the approach does not guarantee equal access to government facilities, and the state retains considerable scope to exclude and restrict speakers as long as the measure is not arbitrary, unreasonable or unfairly discriminatory.23 The permissible grounds for excluding speakers will vary depending on the type of forum and the reason for the decision.24 The ECtHR has developed its jurisprudence on positive obligations with specific duties relating to the media, stressing the importance of media plurality, in particular in relation to broadcasting. The ECtHR has noted that a challenge to freedom of expression is posed where ‘a powerful economic or political group in a society is permitted to obtain a position of

18 In Sanchez v Spain (28955/06) (2012) 54 EHRR 24, the dismissal was found to be proportionate. 19 Appleby v United Kingdom (44306/98) (2003)37 EHRR 38, [40]. 20 ibid [47], stating that this analysis applies in relation to both publicly and privately owned property. 21 Vereinigung Demokratischer Soldaten Osterreichs and Gubi v Austria (A/302) (1995) 20 EHRR 56. 22 ibid. The decision to exclude a magazine was found to violate Art 10 as the government had shown that the decision was not necessary to secure discipline within the military. 23 R (on the application of ProLife Alliance) v BBC [2003] UKHL 23; [2004] 1 AC 185. Lord Hoffmann applied this standard in reference to whether Art 10 was engaged at all. 24 cf Core Issues Trust v Transport for London [2014] EWCA Civ 34, [2014] PTSR 785.

32  Jacob Rowbottom dominance over the audiovisual media and thereby exercise pressure on broadcasters’.25 This concern about undue dominance applies to both the state-owned and privately owned media. To address these concerns, the state is under a positive obligation to ensure that any public broadcasting system provides a ‘pluralistic service’.26 In Lentia v Austria, the ECtHR found that publicly owned broadcasting services can help promote the dissemination of diverse content, but found that this did not justify a state monopoly on broadcasting.27 The positive obligation also requires the state to ensure broadcasters provide the public with ‘impartial and accurate information and a range of opinion and comment, reflecting inter alia the diversity of political outlook within the country’ and that those working in the broadcast sector ‘are not prevented from imparting this information and comment’.28 The approach seeks to address the political power exercised by the media owner, either by ensuring that the state-run media provides such content to countervail the influence of the privately owned media or by regulating private broadcasters. The obligation is not, however, as far reaching as it may sound. For example, when looking at the arrangements for electoral communications in Communist Party of Russia, the ECtHR asked itself whether opposition parties were given ‘at least minimum visibility on TV’.29 Again, the standard required by the ECtHR seems to be for the minimum level of resources, rather than anything approaching equality. The Article 10 obligations in relation to communicative resources can be summarised as securing sufficient communicative resources to prevent the destruction of the essence of the right, ensuring that speakers are not excluded from state-provided forums for arbitrary reasons and securing at least a minimal level of coverage for major viewpoints in the broadcast media. The goal appears to be the provision of a minimal threshold of speech resources in some cases, and therefore goes a very small way to meeting the goals of political equality. While it is tempting to criticise courts for this limited approach, there are good reasons for courts to be wary of a particularly expansive approach to positive expression rights. The provision of positive speech rights will, in many cases, concern the distribution of economic resources, which may be seen as more appropriately decided by the political branches of government. The courts may also feel ill-equipped to decide exactly how a positive speech obligation should operate – such as who can claim it, when the positive speech rights of a person should outweigh the negative speech rights of another, and deciding what limitations may be appropriate. On an analogous issue, Lord Mance in Kennedy rejected an argument that Article 10 creates a right to access information, as such a right would be ‘established on an undefined basis without the normal checks and balances to be expected in the case of freedom of information legislation introduced by a State after public consultation and debate’.30 Positive speech rights raise issues of policy design, balance and detail, for which a range of choices are open to the authorities. Such managerial standards are less appropriate for the courts to determine on a case-by-case basis, and may be better developed by the legislature as part of a comprehensive scheme. 25 Manole v Moldova [2009] ECHR 13936/02, [98]. 26 Communist Party of Russia v Russia (29400/05) (2015) 61 EHRR 28. 27 Lentia v Austria (1994) 17 EHRR 93. 28 Manole (n 25) [100]. 29 Communist Party of Russia v Russia (2015) 61 EHRR 28, [126]. See also TV Vest AS and Rogaland Pensjonistparti v Norway (2009) 48 EHRR 51, [73], where the lack of free access to television and the fact that a political party was ‘hardly mentioned’ in the edited news coverage meant that the party had no means to get its message across through the medium. 30 Kennedy v Charity Commission [2014] UKSC 20; [2015] AC 455, [147].

Positive Protection for Speech and Substantive Political Equality  33 The biggest concern is not that the courts will take a minimalist approach to the positive aspects of freedom of expression, but rather that courts will be tempted to emphasise formal equality and inhibit attempts by the legislature to promote substantive equality. As stated earlier, methods to address private censorship will often interfere with the speech rights of the privately owned forum. For example, right of reply laws in relation to the mass media are often seen as a way to promote the positive aspect of expression by providing access to the means of communication. The right of reply imposes a limit on the prerogatives of the media controller and blunts the communicative power of the initial statement that the speaker wants to reply to. A court may thereby view the right to reply law as an interference with the media owner’s editorial autonomy.31 In the USA, the Supreme Court has used the free speech clause of the Constitution to strike down a statutory right of reply for candidates in an election, in relation to newspaper coverage.32 In such cases, a tension emerges between the positive aspect of freedom of expression and the negative protection from interference from the state. The danger is that courts will gravitate towards the negative protection from state interference at the expense of the positive. What is called for here is not for the court to embark upon an expansive protection for political equality and positive expression rights. Instead, the courts should at least recognise these values and not frustrate measures that properly protect positive expression rights simply by prioritising negative rights against the state. The different approaches can be seen in the majority and minority in Animal Defenders International, in which the ECtHR upheld a ban on political advertising on the broadcast media. One group of dissenters ­prioritised formal equality rights by arguing that positive duties should not be used to justify a restriction on the negative right of private companies to sell and purchase advertising time for political messages.33 By contrast, the majority recognised that a general measure to limit money in politics served the positive duties to promote pluralism and a fair electoral process.34 Under this latter approach, both courts and the legislature share responsibility for protecting expression rights, with the latter playing an important role in devising systems for promoting positive expression rights. The court should interpret Article 10 in a way that provides a safeguard to prevent abuse but without constructing prophylactic rules that block these efforts to secure positive speech rights.35

IV.  Promoting Equal Opportunities to Communicate In the previous section, I argued that the positive obligations under Article 10 are limited, but the state can still go beyond the requirements of the ECHR and devise policies that promote equal opportunities to engage in expressive activities that influence p ­ olitical ­decisions. The discussion that follows is non-exhaustive, but shows how there are a

31 See, eg the reasoning in Saliyev v Russia (35016/03), unreported, 21 October 2010, [54]. But cf Kaperzynski v Poland (43206/07), unreported, 3 April 2012, [66]. 32 Miami Herald v Tornillo, 418 US 241 (1974). 33 Animal Defenders International v United Kingdom (48876/08) (2013) 57 EHRR 21. See the dissenting opinion of Ziemele et al and the discussion in J Rowbottom, ‘Animal Defenders International: Speech, Spending, and a Change of Direction in Strasbourg’ (2013) 5 Journal of Media Law 1. 34 Animal Defenders International (n 33). 35 See Rowbottom, Democracy Distorted (n 2).

34  Jacob Rowbottom v­ ariety of different m ­ easures to promote ­positive expression rights. The discussion will show how areas such as campaign finance, media regulation, access to property and the digital media are all connected and have significant implications for political equality.

A. Elections Steps to promote political equality can be taken in relation to election campaigns, by ­attempting to reduce the influence of private wealth. The most obvious speech-enhancing step is through the provision of funds to political parties or to citizens to donate to parties. Other measures that restrict the use of private wealth in elections can also contribute to positive speech by opening up space for other speakers to contribute and be heard. The restrictions aiming to pursue this goal include a cap on overall campaign spending, as currently applied in the UK, and caps on the amounts of money that can be donated to politicians, as applied in the USA. In designing such a scheme, one question is whether the measures should seek to promote equality between parties or equality between citizens.36 An overall cap on election spending and the provision of state funds to political parties seeks to reduce differences in spending power between political parties. By contrast, caps on donations focus on citizen equality by limiting the scope for a single wealthy person to determine a party’s financial well-being. Capping donations does not mean that parties have equal spending power, but that differences in spending power between parties are to some degree reflective of levels of popular support. Similarly, matching grants and voucher schemes for public funding promote equality and the positive protection of speech by giving individual voters a choice to allocate a small sum of state resources to a particular party. In practice, a system of campaign finance regulation is likely to be a hybrid promoting equality both between citizens and between parties. There are a number of risks in this strategy. The government enacting the rules to regulate the conduct of elections will have a strong vested interest in designing a system that is most likely to entrench its power. There are also difficult questions in deciding what levels to limit election spending at and what subsidies are required. A heavy-handed approach could undermine free speech and reduce the overall level of political information. As will be argued later, the courts need to be alert to these risks. However, at the other extreme, the US experience has been to reject equality arguments and to rely on formal equality to strike down many campaign finance controls. What is important is to strike a balance that does not give government a free hand, but which does not inhibit genuine attempts to secure equality in speech opportunities.

B.  Public Land Positive expression rights can be protected through the provision of publicly run forums for communication. One example is publicly owned land. Streets, parks and town squares

36 See

LA Ringhand, ‘Concepts of Equality in British Election Financing Reform Proposals’ (2002) 22 OJLS 253.

Positive Protection for Speech and Substantive Political Equality  35 are not solely or primarily communicative forums, but are places where people can speak to sections of the public at a low cost. While the costs are low to the user, the provision of such space imposes costs. For example, there is the cost of providing the land, the cost of tidying up after users (which might include discarded leaflets) and the costs of policing. As these costs are typically met by the state, we might think of the provision of such space as a type of subsidy.37 The prerogatives of the property owner are, in the case of public bodies, constrained by public law controls.38 For example, as noted in the previous section, s­ peakers cannot be excluded from public land arbitrarily. Consequently, the state does not have a free hand to use its powers of ownership to regulate the subsidy. By providing access and meeting the costs of maintaining the space, the public body provides a resource for communicative activities. While access to public spaces might be thought to be the most obvious way the state can provide communicative resources, the use of that resource is coming under pressure from a number of directions. First, the pressure on public funding may cause public bodies to cut back on certain aspects of the subsidy. For example, the London Metropolitan Police guidelines from 2015 stated that protest organisers may have to arrange and pay for traffic management and stewarding in relation to their protests.39 When two protests were asked to pay for such costs in 2015, the organisers complained that they had not been asked to do so in relation to past events and that it imposed a financial burden on the demonstrators.40 Under such an approach, the costs of the expression right would be transferred from the public body to the speaker, rendering the use of streets less of a form of ‘cheap speech’. Such charges could arguably fall foul of Articles 10 and 11 of the ECHR, though the courts do not require police and other authorities to spend unlimited sums to support public speakers.41 The second limit to the use of public space is the legal controls that curtail the right to protest. This includes the use of police discretion to move people on, methods to contain speakers, such as kettling, and the filming of protestors. Controls on assembly rights can devalue the subsidy for speakers by increasing the potential risks and costs associated with protesting, and can diminish the effectiveness of the protest. Such measures may have a legitimate public order purpose underlying them, and any individual measures can be challenged in judicial review and human rights law. Again, legal challenges provide limited protection, given the potential for the courts to give considerable weight to the judgment of the police and other authorities. These problems are well known, but the impact on political equality is a factor that should be considered when striking a balance with public order. A further trend is the privatisation of public spaces. Many spaces that were previously publicly owned have been transferred to private bodies. In some cases, this might occur through selling off public spaces, or at least granting a long-term lease to a private owner.

37 See J Balkin, ‘Some Realism about Pluralism: Legal Realist Approaches to the First Amendment’ [1990] Duke Law Journal 375, 402–03. 38 See discussion in J Rowbottom, ‘Property and Participation: A Right of Access for Expressive Activities’ [2005] European Human Rights Law Review 186. 39 London Metropolitan Police, General Advice to Event Organiser: ‘it is likely that you shall be required to arrange your own traffic management/stewarding through accredited companies that are acceptable to the council’. 40 K Mathiesen, ‘Met Backs Down on Refusal to Police Climate and Women’s Marches’, The Guardian, 26 February 2015. While the police did not implement this requirement in relation to the reported protests, they did not rule out such a condition for future protests. 41 E v Chief Constable of the Royal Ulster Constabulary [2008] UKHL 66, [2009] 1 AC 536, [45].

36  Jacob Rowbottom In other cases, it may be that new privately owned developments attract new users and draw people out of traditional public spaces. These policies of privatisation may pursue valuable goals, such as the redevelopment of public spaces, but bring costs in other respects. The private landowner is not subject to public law obligations and is not required to respect the speech rights of others. As a result, the private landowner can normally exclude speakers on political grounds, raising concerns about political equality. As stated earlier, the courts will find that such exclusions engage Article 10 only in extreme circumstances. The discussion has shown that public land is an important form of positive support for expression rights. It is easy to forget this when thinking about the various competing uses of the land and challenges in managing public spaces. However, it is important to ensure that this more basic form of support is not eroded gradually, and new policies may be necessary to ensure the positive protection keeps pace with the changes. One solution to deal with the potential for private censorship would be to impose certain public law-like obligations on the private landowner through legislation, such as a right of access.42

C.  Mass Media The concerns about public space can be compared with another type of forum for expression, the mass media. A flat and equal distribution of mass media resources is a contradiction in terms.43 To function properly, the mass media is by definition a form of concentrated media power. The positive speech protection therefore cannot mean the equal distribution of communicative power or granting every person access to the mass media to carry their own speech. Instead, political equality requires that the media serves the interests of the audience equally and that each major group and viewpoint has access to some media outlets.44 The media should not become the political tool of its owner or controller. There are several measures that can be taken to promote equality, including the public ownership of the media, content regulations and media plurality laws. As with other forums, the media can be publicly owned, and its resources shared in a way that ensures that a range of leading viewpoints are carried. For example, in many European countries, parts of the broadcast media have been publicly owned. There are a number of ways to govern the publicly owned media. One model is to allocate channels to the leading political parties, while another is for the governing body of the public broadcaster to be composed of representatives from a range of political and civic organisations. However, the model in the UK has been for publicly owned broadcasters to be governed by a body that is independent from government direction and to be subject to content regulations, in particular requiring the political coverage to be impartial. The practical effect of impartiality obligations is that broadcasters will tend to include the major viewpoints in political coverage. Such inclusion can be seen as a type of subsidy to finance the communication of a range of political speakers. An individual speaker does not hold the subsidy as a matter

42 See Rowbottom (n 38). 43 CE Baker, Media Concentration and Democracy: Why Ownership Matters (Cambridge, Cambridge University Press, 2007). 44 ibid 10–12.

Positive Protection for Speech and Substantive Political Equality  37 of right and no one is entitled to demand access to the broadcaster. However, by serving the audience with diverse views, the system provides a state-funded structure that supports ‘free speech values’.45 Of course, the state cannot hold a monopoly of the media. In order for free political discussion to flourish and for uninhibited criticism of government, there is a need for privately owned media institutions to be protected. This, however, raises the problem of privately owned media outlets having an unequal chance to influence political decisions and engage in practices of private censorship. One solution to this issue is to extend some of the content regulations imposed on state broadcasters to privately owned broadcasters. Along these lines, the rules requiring impartiality in political coverage are applied to privately owned broadcasters in the UK. This regulatory regime can also be seen to generate a type of subsidy for speech, in which the power of the private broadcaster is redistributed by preventing the outlet simply reflecting the owner’s views and by requiring the inclusion of diverse voices. Given the number of broadcasting channels and the chance to disseminate audiovisual content on the digital media, the case for a special regulatory regime specifically targeting the broadcast media is coming under challenge. While there is a debate about whether the broadcast media warrants special regulation, political equality provides a separate principled reason for controls, such as impartiality, in relation to some parts of the media. However, impartiality and similar content controls cannot be desirable for all media entities. The broadcast regulations on impartiality require a level of internal pluralism, but there is also a need for external pluralism, with a multiplicity of media outlets representing specific viewpoints and perspectives. Under this approach, each media outlet is free to follow its own editorial line and is not required to provide any balance or diversity. The ideal is that all media sources taken collectively will then provide the audience with a range of views. The British press is seen to follow such a model insofar as it allows for each title to pursue a politically partisan editorial line, although it is questionable whether all the national titles really add up to a diverse range of views. The model of external pluralism may appear to be more closely connected with the negative protection of expression rights, in which media bodies are simply left to get on with things with less state interference.46 There are, however, positive steps that can be taken to ensure that this media sector does serve the values of free speech. For example, there are ways to fragment media power to ensure that a single owner or company cannot dominate the editorial stances taken within the media system. Journalists’ conscience clauses or guarantees of editorial freedom from proprietorial interference are ways to fragment the powers associated with media ownership.47 Such measures help to prevent the media being used instrumentally by owners or other powerful figures, and to ensure that editorial decisions follow the standards of professionalism. Media plurality laws provide another method to

45 J Lichtenberg, ‘Foundations and Limits of Freedom of the Press’ in J Lichtenberg (ed), Democracy and the Mass Media: A Collection of Essays (Cambridge, Cambridge University Press, 1990) 102. 46 Some content controls are consistent with the model of external pluralism, such as rights of reply for those defamed or attacked in such media. 47 This, of course, assumes that such clauses and guarantees can be effectively enforced. In the past, some ­guarantees of editorial independence made to assure regulators about the plurality impact of a media merger have had limited effect at best. See Rowbottom, Democracy Distorted (n 2) 195.

38  Jacob Rowbottom promote free speech values, by limiting the amount of media power that any person can accumulate. Media plurality laws redistribute communicative power to a limited extent in the hope that each relevant group and significant viewpoint has its own media outlet.48 If fragmentation of the market is not enough to pursue that goal, then an alternative response is some type of subsidy to media entities to ensure a more complete representation of major viewpoints, as has been implemented in relation to the press in Scandinavia. So far, we have seen that several strategies can be used to promote political equality in the media, each with costs and benefits. Which of these strategies are most appropriate will generally reflect the political culture in which the media operates.

D.  Digital Media It is often said that in the digital era, traditional media is in decline and that media power will be less of a concern. The internet, mobile technologies and the social media have given rise to a vast redistribution of communicative power. It might thereby be thought that equal political resources are now supplied without the need for any further government policies. The digital media has certainly led to a redistribution of sorts. Those who previously had no voice or potential to connect with a mass audience have more communicative opportunities than ever. New political movements can connect and mobilise. More people can make content available. Getting heard, however, is another matter. The power law distribution of attention means that those at the top get more attention than ever.49 For example, the New York Times, the BBC and The Guardian have become international publications with the power to reach a bigger audience than before. Utopian claims about digital communications alone being sufficient to generate substantive equality should be viewed with scepticism. Large and well-resourced institutions can continue to wield considerable communicative power. In addition, the new media has brought a number of powerful actors, such as Google, Facebook and Twitter – to mention the most obvious examples. At one level, these companies aim to facilitate the speech of others. In that respect, this is another private development that can foster some positive support for expression. The enjoyment of the freedom, however, exists only at the mercy of the owners of those services and is not a matter of right. Various scholars have focused on the potential impact and biases within the algorithms and architecture of the technology, and the way it can promote certain types of content. The recent controversies about the proliferation of extremist content and fake news via the digital media provide notable problems with the digital distribution of information. However, other reports have expressed concern at the potential for human editorial interventions in the promotion of content on the digital services, which brings its own risks of bias.50 Positive speech rights may in some cases require protection from the power of the large intermediaries. For example, some scholars have suggested that an online

48 Baker (n 43). 49 For discussion, see M Hindman, The Myth of Digital Democracy (Princeton, Princeton University Press, 2009); JG Webster, The Marketplace of Attention (Boston, MIT Press, 2014). 50 A Hern, ‘Facebook’s News Saga Reminds Us Humans Are Biased by Design’, The Guardian, 13 May 2016.

Positive Protection for Speech and Substantive Political Equality  39 e­ quivalent to a right of reply could be promoted by making certain links more prominent in the results displayed by a search engine.51 Another strategy is to provide some mechanisms of transparency and accountability, so that people can at least understand the basis on which algorithmic decisions are made and have an avenue to contest those decisions that have unfair effects on particular speakers or viewpoints. These issues pose significant challenges, but such mechanisms need to be explored to prevent the concentration of power in the ‘marketplace for attention’.52

V.  State Power and its Abuse The discussion has shown that there are numerous policies that government can pursue to promote political equality and that each comes with difficult issues of design. Many of the policies discussed may, however, face a broader objection about the use of state power. One line of objection focuses on policies in which the state finances or subsidises political discussion, namely that such policies may not be a good use of taxpayer resources particularly at a time when other government spending is under pressure. This has most commonly been asserted as an argument against state support for political parties. If a subsidy is seen as a state payment to a politician, then a hostile response is likely. One can, however, contrast such hostility with the public attitude to legal aid. Legal aid is not seen as a state handout to lawyers and is not condemned simply because it is provided to an unpopular claimant or defendant. Instead, it is seen as a component of a fair legal system (albeit one that has been subject to many cuts). So, positive support for speakers might also come to be accepted as an equivalent policy to ensure a fair democratic system. A more forceful objection to measures claiming to support positive expression rights is that the policy may really mask certain political purposes. Whether the measures entail the provision of state funds or the regulation of private resources, the policy maker takes a decision about the appropriate allocation of communicative resources. For example, state support for political speech may be seen as a means of channelling funds to political allies. Similarly, a subsidy to a newspaper may be thought to give the state leverage over that institution, and potentially inhibit criticism of the government. These arguments are particularly forceful if one takes the view that the media is primarily to act as a watchdog in relation to government action. Such concerns have been expressed about the use of the television licence fee negotiations, which provide the basis of a state subsidy, as a possible means to punish the BBC for critical comment. The powers exercised in the name of regulating access to or the fair use of communicative resources could also be abused to restrict opposing voices. Along similar lines, the Transparency of Lobbying, Non-party Campaigning and Trade Union Administration Act 2014, while claiming to regulate the use of money in elections, was criticised as being a vehicle to gag various campaigning groups and voices of criticism.

51 See F Pasquale, Black Box Society: The Secret Algorithms that Control Money and Information (Cambridge, MA, Harvard University Press, 2015) 75. 52 See Webster (n 49).

40  Jacob Rowbottom These concerns are very real and demonstrate the need for caution in the state support for expression. To minimise such concerns, wherever possible, resources should not be distributed by a government-directed block grant. The example of the right of access to land poses relatively minimal dangers of abuse, as the right will be generally available to speakers and there is less scope for discrimination. The distribution of election funds through voucher schemes provides an example of the funding decision being decentralised and removed from the hands of a government agency. Where there is no alternative but for government or an agency to make a decision about the allocation of resources, then that decision should be subject to a rigorous process of scrutiny and consultation by a variety of groups and independent institutions. In Animal Defenders International, the ECtHR found that a control on political advertising did not violate Article 10 because it had been considered by various independent bodies and been subject to external scrutiny.53 One of the difficulties with the Transparency of Lobbying, Non-party Campaigning and Trade Union Administration Act 2014, by contrast, was that it was rushed through Parliament with minimal external involvement. While the political branches bear part of the responsibility for promoting speech opportunities, it is also important to impose safeguards to minimise the risk of abuse. A further question is whether any restrictions should be imposed on extreme speakers and those challenging certain core democratic values. For example, should far-right political parties be entitled to state funds and claim access to the media under a scheme for positive rights? Under the current law, there is ample scope for such exclusion to be justified as non-arbitrary. Whether this should be the case is a matter for debate. Critics argue that such limits amount to censorship based on political viewpoints. This debate is a key point about the scope of political rights more generally and raises difficult issues. The positive protection of expression rights forces these issues to be confronted, but that does not supply a principled reason to reject the positive protection of expression rights in general.

VI. Conclusion Making a claim for the positive protection of speech rights is not straightforward. There are difficulties in deciding who is owed the duty and just how much support a person should be entitled to. Positive speech cannot simply depend on a general claim that people need the resources to exercise expression rights to the fullest extent. For example, there are many reasons why we should oppose any attempts by government to restrict artistic speech.54 However, rights to the negative liberty for artistic expression, while powerful, do not translate into a corresponding claim that the state is under a constitutional obligation to provide oil paints, a canvas and an easel for aspiring artists. There is a need for an additional argument to make the claim for a positive expression right. If freedom of expression is regarded as a right enabling democratic participation and the exchange of political ideas, then the free speech principle quickly becomes connected with substantive political equality. The differences in communicative resources become a

53 Animal 54 See

Defenders International (n 33). Judge Spielmann’s dissent in Muller v Switzerland (1991) 13 EHRR 212.

Positive Protection for Speech and Substantive Political Equality  41 problem of unequal citizenship, in which some individuals have a greater opportunity to influence political decisions. When viewed in such a way, the need for state action to promote opportunities for expression and curb the disproportionate influence of unequal economic resources becomes central to the fair operation of the democratic process. Substantive political equality supplies a reason why free speech, under the democracy-based justification, requires positive protection. While that positive protection is important in meeting the demands of democracy, that does not mean the protection lies primarily in the hands of the courts. The role of the courts is to intervene in those more extreme cases where a clear failure to meet the most minimal expectations is evident. Instead, the legislature and other public bodies share responsibility for securing equal communicative opportunities. This chapter has outlined a number of ways in which government can promote substantive political equality. These measures can span a wide range of areas, including the management of land, campaign finance law, media regulation and the regulation of internet intermediaries. The discussion here has not been exhaustive and could include other measures, such as the regulation of lobbying. Political equality provides the common thread that connects these diverse types of policy in various areas of activity. The discussion has also shown that political equality can be promoted through the provision of state resources, such as grants to private actors to facilitate speech-related activities, and the regulation of the use of private wealth and property in politics to redistribute speech opportunities. Which of these strategies, or combination thereof, is pursued is a matter of choice, and the most appropriate solution will depend on the context and political culture. Given this level of choice, the courts cannot be expected to flesh out the requirements of equality beyond a basic minimum. There are, however, reasons to proceed with caution when devising policies that promote equality. One risk is that controls seeking to equalise communicative resources may impose an excessive curb on the rights of some speakers. Another risk is that policies distributing speech opportunities may covertly pursue improper political purposes. While such concerns do not rule out the equality-promoting policies, there is a need for safeguards and checks to prevent positive speech rights being exploited and abused by government actors. The courts do have a role to play in guarding against the abuse of power. However, in performing that function, the courts should resist the temptation to prioritise the n ­ egative speech rights in a way that frustrates genuine measures to promote political equality. S­ triking the right balance is a delicate task, but is central in a system that values the equal status of citizens in the process for reaching collective decisions.

42

4 The Access to Information Dimension of Positive Free Speech ANDREW SCOTT AND ABBEY BURKE

The ‘right to know’, the expectation that citizens will be able to access state-held information subject to agreed limitations, is closely and intuitively associated with democracy and freedom of speech. It is often assumed that it was ever thus. The association is implicit, for example, in the frequent quotation of a letter written in 1822 by James Madison: a popular Government without popular information or the means of acquiring it is but a Prologue to a Farce or a Tragedy or perhaps both. Knowledge will forever govern ignorance, and a people who mean to be their own Governors must arm themselves with the power that knowledge gives.1

Yet, Madison wrote not of citizens’ access to information, but rather of the importance of schooling in Kentucky. As Michael Schudson explained in an illuminating book on the generalised cultural commitment to the right to know in the USA, ‘somehow we have collectively backdated ideas and innovations of the past half century’ when such commitment was in fact wrought only in the post-war period.2 Nonetheless, this cultural commitment to the right to know seems prevalent and strong. It is embodied most obviously in ‘freedom of information’ legislation passed by parliaments. Such legislation has become a commonplace feature of democratic constitutions.3 Still, in most jurisdictions, this is a relatively recent phenomenon.4 The UK passed its Freedom of 1 J Madison, The Writings of James Madison, ed G Hunt, vol 9 (New York, Putnam, 1910) 103. 2 M Schudson, The Rise of the Right to Know: Politics and the Culture of Transparency, 1945–1975 (Cambridge MA, The Belknap Press of Harvard University Press, 2015) 3. See also A Roberts, Blacked Out: Government Secrecy in the Information Age (Cambridge, Cambridge University Press, 2006) ch 1. 3 This includes both statutes of general application and those with more specific objects – consider, in respect of the UK, the right of access to environmental information under EU law, rights of access to public records and rights of access to information concerning planning, business or land under a disparate array of statutes – see P Coppel, Information Rights: Law and Practice, 4th edn (Oxford, Hart Publishing, 2014) chs 6–8. Moreover, the promise of greater openness has been assimilated into the statutory purposes and managerial language of some state bodies. This is typified in the statutory objectives of ‘accountability’ and ‘transparency’ to be furthered by the Charity Commission in conformity with regulatory best-practice that provided context to the judgment in Kennedy v Charity Commission [2014] UKSC 20. 4 The most obvious exception is Sweden, where the Freedom of the Press Act was introduced in 1766: see K Örtenhed and B Wennberg (eds), Press Freedom 250 Years: Freedom of the Press and Public Access to Official Documents in Sweden and Finland – A Living Heritage from 1766 (Stockholm, Sveriges Riksdag, 2016). The USA introduced the federal Freedom of Information Act (5 USC §552) in 1966, followed by Australia and New Zealand (both 1982) and Canada (1983). On the origins of the US federal legislation, see Schudson (n 2) ch 2;

44  Andrew Scott and Abbey Burke Information Act only in 2000.5 The primary focus of this chapter, however, is the even more recent phenomenon of judges beginning to fashion legal principles that install the default expectation of public access to information held by the executive branch of government. Such principles are distinct from, perhaps extend beyond, but certainly underpin legislative schemes. At least the judicial variant of this development has been taking place in the absence of a fully developed theoretical underpinning. In what it referred to as a ‘clarification of the classic principles’, the European Court of Human Rights (ECtHR, the Strasbourg court) has proved willing to found its emergent principle of access to information directly on the right to freedom of expression.6 British judges have been more circumspect. For instance, Lord Toulson has insisted that: if article 10 is to be understood as founding a right of access to information held by a public body, which the public body is neither required to provide under its domestic law nor is willing to provide, there is a clear need to determine the principle or principles by reference to which a court is to decide whether such a right exists in a particular case and what are its limits … what the new interpretation would require is a clear, high level exegesis of the salient principle and its essential components.7

His Lordship lamented further that ‘a consistent and clearly reasoned analysis of the “right to receive and impart information” within the meaning of article 10 … is so far lacking’. Yet, Lord Toulson showed himself content to base a principle of open government upon the atheoretic common law constitution. In both contexts, the pragmatic ramifications of this shift remain to be fully explored. The normative starting point of this chapter is that the right of access to information is properly conceived of as an aspect of what can be described as a ‘positive’ or ‘facilitative’ conception of free speech.8 Following a cursory reflection on the perceived dearth of theory linking free speech and access to information, the aim of this chapter is to chart and consider the development of judge-wrought principles of access to information held by the state. Although it has some antecedents, this project emerged most obviously in the jurisprudence of the ECtHR on the right to freedom of expression from around 2005. This shift was considered and rejected by a majority of the UK Supreme Court in the case of Kennedy v Charity Commission in 2014. Instead, that court located a previously unheralded principle of open government in the common law. This principle has been considered and somewhat honed in subsequent decisions of the British courts. The return of the ECtHR to the theme in 2016, in the case of Magyar Helsinki Bizottság v Hungary, saw a powerful reiteration of the principle founded on the speech right, and an implicit reproof to British judicial reluctance.9 on the early experience of the Antipodean laws, see R Snell, ‘The Kiwi Paradox: A Comparison of Freedom of Information in Australia and New Zealand’ (2000) 28 Federal Law Review 575. See generally JM Ackerman and IE Sandoval-Ballesteros, ‘The Global Explosion of Freedom of Information Laws’ (2006) 58 Administrative Law Review 85; see also the ‘Right to Information Rating’ maintained by the European and Canadian NGOs Access Info Europe and the Centre for Law and Democracy, available at https://www.rti-rating.org. 5 See generally B Worthy, The Politics of Freedom of Information: How and Why Governments Pass Laws That Threaten Their Power (Manchester, Manchester University Press, 2017); Coppel (n 3) ch 1. 6 Magyar Helsinki Bizottság v Hungary [2016] ECHR 975, [156]. 7 Kennedy (n 3), [145] and [148]. 8 See generally, AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379. 9 Helsinki (n 6).

The Access to Information Dimension of Positive Free Speech  45 Each of these stages in the genesis of legal principle(s) regarding the right to know is ­considered in the paragraphs that follow.

I.  A Brief Note on Theoretical Underdevelopment Access to state-held information lies at the heart of the relationship between the state and the citizen, such that in a democracy it can be assumed by default that the disclosure of information will normally follow on from its request. The possibility and quality of access to information regarding the development of public policy and the performance of executive functions is of vital concern to every citizen. Any undue constraint on the ability of citizens to obtain information regarding the state is likely to be doubly problematic. As Tony Blair once commented, on the one hand, ‘unnecessary secrecy in government leads to arrogance in governance and defective decision-making’, while on the other, ‘the perception of excessive secrecy [can become] … a corrosive influence in the decline of public confidence in government’.10 Executive government has become significantly more open over the recent decades.11 Perhaps it was the normative self-evidence of this easy assumption that has permitted British judges to revert to the underdetermined, amorphous norms of the UK common law constitution when insisting on the right of access. To rest anything on the more definite principle of freedom of speech would require an elaboration of the parameters of that principle. It would necessitate a definitive confirmation of what does, and what does not, fall within the compass of the right that a judge may reasonably shy from.12 It cannot be correct, however, to suggest that theory in this area has been left wholly unenunciated.13 Chronologically speaking, the emergence of the cultural commitment

10 Duchy of Lancaster, Your Right to Know: The Government’s Proposals for a Freedom of Information Act, Cm 3818 (London, Stationery Office, 1997) [1.1]. Famously, Blair has since reneged on that position: ‘freedom of information. Three harmless words. I look at those words as I write them, and feel like shaking my head till it drops off my shoulders. You idiot. You naïve, foolish, irresponsible nincompoop. There is really no description of stupidity, no matter how vivid, that is adequate. I quake at the imbecility of it … the truth is that the FOI Act isn’t used … by “the people”. It’s used by journalists. For political leaders, it’s like saying to someone who is hitting you over the head with a stick, “Hey try this instead”, and handing them a mallet’: T Blair, A Journey (London, Hutchinson, 2010) 516. 11 On the desirability of transparency in governance generally, see C Hood and D Heald (eds), Transparency: The Key to Better Governance? (Oxford, Oxford University Press, 2006). Historically at least, the executive arm of the British state has been among the most secretive in the democratic world – see generally I Cobain, The History of Thieves: Secrets, Lies and the Shaping of a Modern Nation (London, Granta, 2016); C Moran, Classified: Secrecy and the State in Modern Britain (Cambridge, Cambridge University Press, 2013); D Vincent, The Culture of Secrecy: Britain 1932–1998 (Oxford, Oxford University Press, 1997); R Crossman, ‘The Real English Disease’, New Statesman, 24 September 1971. Indeed, for some, secrecy in Britain is endemic: ‘as much a part of the English landscapes as the Cotswolds … it goes with the grain of our society … its curtailment, not its continuity would be aberrational’: see P Hennessy, Whitehall (London, Secker & Warburg, 1989) 347. The first formal move towards instituting a ‘right to know’ in the UK – the Home Office, Code of Practice on Access to Government Information (London, HMSO, 1994) – was introduced by John Major in 1994 as an adjunct to the Citizens’ Charter initiative. 12 The same lack of confidence might be read in the insistence of the Strasbourg court in Helsinki that it did not rest its rationale on a ‘positive’ conception of freedom of expression, preferring instead to infer a foundation from the sometime instrumentality of access to the right to receive and impart information: Helsinki (n 6) [156]. Alternatively, this may have been merely an accommodation of precedent. 13 See generally CA Bishop, Access to Information as a Human Right (El Paso, LFB Scholarly, 2012), especially chs 2, 5 and 6 (encompassing, inter alia, ‘freedom of expression’ and ‘right to truth’ conceptualisations of ­freedom

46  Andrew Scott and Abbey Burke to openness in government is aligned with a deepening intellectual conception of free speech. There is plainly a close association between the ‘argument from democracy’ or ‘selfgovernment’ justification for freedom of speech and the basis for the right to know. This explanation of the importance of freedom of speech holds that constitutional guarantees are intended to protect and promote the free discussion of governmental affairs and wider deliberative debate that are deemed necessary to the process of self-government.14 More particularly, a linkage of freedom of speech and access to information is evident in Vincent Blasi’s more focused variant of the argument from democracy which emphasises the value of checks on the exercise of state power that freedom of speech can afford to citizens.15 Indeed, one of the core premises of Blasi’s approach has been that ‘the general populace must be the ultimate judge of the behavior of public officials’,16 with one manifestation being that ‘the First Amendment may require that journalists have access as a general matter to some records’.17 British judges’ appreciation of this position has previously been visible in their discussion of the open justice principle and the general insistence that the operation of the courts be open to public scrutiny. It can be discerned more fully in the occasional expansion of that logic to encompass the transparency of executive action achieved through open justice: there is … a distinct aspect of the principle which goes beyond proper scrutiny of the processes of the courts and the judiciary. The principle has a wider resonance, which reflects the distinctive contribution made by the open administration of justice to what President Roosevelt described in 1941 as the ‘first freedom, freedom of speech and expression’. In litigation, particularly litigation between the executive and any of its manifestations and the citizen, the principle of open justice represents an element of democratic accountability, and the vigorous manifestation of the principle of freedom of expression. Ultimately it supports the rule of law itself. Where the court is satisfied that the executive has misconducted itself, or acted so as to facilitate misconduct by others, all these strands, democratic accountability, freedom of expression, and the rule of law are closely engaged.18

There are two essential points here that expand the ambit of state transparency. The first is that the principle of open justice extends beyond the parameters that are often claimed for it – to permit scrutiny of the exercise of power by officers of the court – to encompass also the need for transparency of the exercise of (state) power more broadly. The second essential of information, and concluding that the free speech conceptualisation is the most persuasive among all those available); P Birkinshaw, ‘Freedom of Information and Openness: Fundamental Human Rights?’ (2007) 58 Administrative Law Review 177; A Mason, ‘The Relationship between Freedom of Expression and Freedom of Information’ in J Beatson and Y Cripps (eds), Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams (Oxford, Oxford University Press, 2000) ch 13; S Sedley, ‘Information as a Human Right’ in Beatson and Cripps (ibid) ch 14; R Stubbs, ‘Freedom of Information and Democracy in Australia and Beyond’ (2008) 43 Australian Journal of Political Science 667. 14 This strand of thought on freedom of speech is particularly associated with the writings of Alexander ­Meiklejohn – see, eg Political Freedom: The Constitutional Powers of the People (New York, Harper, 1960); ‘The First Amendment is an Absolute’ [1961] Supreme Court Review 245. See generally E Barendt, Freedom of Speech, 2nd edn (Oxford, Oxford University Press, 2005) 17–21. 15 V Blasi, ‘The Checking Value in First Amendment Theory’ (1977) 2 American Bar Foundation Research Journal 521. 16 ibid 542. 17 ibid 609–10. 18 R (on the application of Binyam Mohamed) v Secretary of State for Foreign and Commonwealth Affairs [2010] EWCA Civ 65, [39] (Lord Judge CJ).

The Access to Information Dimension of Positive Free Speech  47 point is that the open justice principle, serving as it does to secure the transparency of court proceedings, is only a specific manifestation in a particular context of a broader norm that compels transparency in the exercise of power. The bottom line is that, as the Strasbourg court has appreciated, access to information facilitates the process of informed democratic will-formation upon which the Western polity is founded. Schudson describes this as a new form of ‘monitory democracy’: in which government is held accountable not just at the polling place on election day but continuously … through the watchfulness of the news media, think tanks, public opinion polling, a variety of non-governmental organizations that take themselves to be watchdogs of government with respect to specific policy domains, and even watchdog agencies within government itself.19

The virtue of openness in government in such a polity can be summarised as its ability to act as a prophylactic, and, when required, a scalpel, to the weakness, ineptitude, face-saving, self-dealing or plain corruption to which public administrators, as much as any person, are liable.20

II.  Emergence of an Article 10 Right of Access to Information in Europe The proposition that Article 10 of the European Convention on Human Rights (ECHR) includes a right to receive information from a public authority enjoyed inauspicious origins. It was considered, but rejected, in four judgments of the Grand Chamber of the ECtHR between 1987 and 2005.21 This was done on the basis of a literal interpretation of the text of Article 10(1) ECHR, and involved the issue of ‘apparently clear-cut statements of principle’.22 The first and most prominent of these cases was Leander v Sweden. In that case, the ECtHR noted in passing that: Article 10 does not, in circumstances such as those of the present case, confer on the individual a right of access to a register containing information on his personal position, nor does it embody an obligation on the Government to impart such information to the individual.23

Notably, three of the four decisions of the Grand Chamber were primarily based on Article 8 ECHR. The 1998 decision in Guerra v Italy, however, arguably stands as a direct rejection of an Article 10 right to official information.24 In that case, it was stated explicitly that while 19 Schudson (n 2) 25; see also J Keane, The Life and Death of Democracy (London, Simon & Schuster, 2009); J Keane, Power and Humility: The Future of Monitory Democracy (Cambridge, Cambridge University Press, 2018). 20 Peter Hennessy’s magisterial tome, Whitehall, includes a chapter on government secrecy that opens by quoting a former civil servant in the Thatcher government: ‘with confidence and competence so much lower than they should be, it is not surprising that Whitehall fiercely defends its tradition of secrecy … [the rules that secure it] protect ministers and officials from embarrassment … [but simultaneously] ensure that there is no learning curve’: see Hennessy (n 11) 344 (quoting Sir John Hoskyns). 21 Leander v Sweden (1987) 9 EHRR 433; Gaskin v United Kingdom (1990) 12 EHRR 36; Guerra v Italy (1998) 26 EHRR 357; Roche v United Kingdom (2006) 42 EHRR 30. Such a right has been read into other Convention rights. 22 Kennedy (n 3) [59] (Lord Mance). 23 Leander (n 21) [74]. 24 On this point, contrast Kennedy (n 3) [59] (Lord Mance) and [180] (Lord Wilson).

48  Andrew Scott and Abbey Burke one role of Article 10 is to prevent a public authority from interfering in the communication of information between two parties, it could not also impose a positive obligation upon a public authority itself to impart information.25 Notwithstanding the seemingly clear statements in Leander and related cases, a subsequent series of judgments handed down by a number of sections of the ECtHR since 2006 have reached the opposite conclusion.26 In Társaság v Hungary, the Second Section explained that, despite the earlier jurisprudence, ‘the Court has recently advanced towards a broader interpretation of the notion of “freedom to receive information” and thereby towards the recognition of a right of access to information’.27 The court held that the state faced a positive obligation to facilitate access to information where a person or body exercised ‘the functions of a social watchdog, like the press’, and where ‘barriers to the exercise of press functions … [o]n issues of public interest … exist solely because of an information monopoly held by the authorities’.28 The broadened version of Article 10, the right of access to state-held information, was assumed to exist by the Grand Chamber in Gillberg v Sweden.29 It was also recognised explicitly in other cases (at least insofar as a right of access had already been granted under state law).30 In Österreichische Vereinigung zur Erhaltung, Stärkung und Schaffung v Austria, the First Section of the court recognised the broadened right, notwithstanding that there had been no breach of any domestic right of access.31 Again, the court emphasised that the ‘most careful scrutiny’ – a high standard of review – is necessary when measures taken by national authorities may potentially discourage the participation of the press, or similar ‘watchdogs’ in society, in public debate on matters of legitimate public concern.32 Domestic courts in England and Wales were asked to affirm or reject the Article 10 right of access to information in a number of cases. Reception of the developing principle was initially somewhat mixed, and the strength of the precedent deriving from the disparate judicial statements was uncertain. The Supreme Court has considered the issue on two occasions. In BBC v Sugar (dec’d) (No 2), Lord Brown offered a clear rejection of

25 Guerra (n 21) [53]. The European Commission of Human Rights had previously agreed with the applicants that such positive obligations did flow from Article 10. 26 For early comment on this shift, see W Hins and D Voorhoof, ‘Access to State-Held Information as a Fundamental Right under the European Convention on Human Rights’ (2007) European Constitutional Law Review 114; P Tiilikka, ‘Access to Information as a Human Right in the Case Law of the European Court of Human Rights’ (2013) 5 Journal of Media Law 79; K Steyn and H Slarks, ‘Positive Obligations to Provide Access to Information under the European Convention on Human Rights’ (2012) 17 Judicial Review 308. In Kennedy, Lord Wilson noted that, in fact, ‘the first straws in the wind’ could be discerned in Guerra v Italy. He explained that the majority in that case had only rejected an overstatement of the broad view – one that would extend the obligation on states also actively to collect information – while seven judges (of 20) explicitly left the door open to a general principle obliging public authorities to impart information: see Kennedy (n 3) [180]. 27 Társaság a Szabadsagjogokert v Hungary (2011) 53 EHRR 3, [35]–[36]. The court cited Matky v Czech R ­ epublic, app no 19101/03, unreported, 10 July 2006 as the origin of this shift. In both this case and Kenedi v Hungary, app no 31475/05, unreported, 26 May 2009 (also decided by the Second Section), the respondent government had conceded that Art 10(1) had been infringed. 28 Társaság (n 27) [36]. 29 App no 41723/06, unreported, 3 April 2012, [93]. 30 Shapovalov v Ukraine, app no 45835/05, unreported, 31 July 2012 (Fifth Section); Youth Initiative for Human Rights v Serbia, app no 48135/06, unreported, 25 June 2013 (Second Section). 31 App no 39534/07, unreported, 28 November 2013. 32 ibid [33]–[34].

The Access to Information Dimension of Positive Free Speech  49 the broadened view.33 He considered that the first three of the new wave of Strasbourg decisions – Matky, Társaság and Kenedi – fell ‘far short’ of grounding the new principle.34 As Lord Mance acknowledged in Kennedy, however, Lord Brown’s view was not strictly part of the ratio decidendi of the case.35 Moreover, it preceded the more recent and more clearly stated Strasbourg decisions. Nevertheless, in Kennedy, a majority of a seven-member Supreme Court affirmed the view expressed by Lord Brown in Sugar (No 2). Again, the views expressed by this majority on point were explicitly obiter, but they were forcefully put.36 Lord Mance reviewed what were then the four most recent decisions of the Strasbourg court,37 and concluded that – had it been decisive for the case in hand – he would have affirmed the general principle to be found in the Leander line of cases.38 Along with Lord Toulson, he expressed concern that the opposite conclusion would entail recognition of a general European freedom of information law derived from Article 10 ECHR.39 Other domestic jurisprudence has been more accommodating of the new wave of Article 10 jurisprudence. In A v Independent News & Media Ltd, the Court of Appeal applied the principle in a case involving media access to the Court of Protection.40 In passing, Lord Judge CJ affirmed the description of the wellspring of the development, the Társaság decision, as ‘a landmark decision on the relation between freedom to information and the … Convention’.41 In the Court of Appeal’s consideration of the Sugar (No 2) case, Moses LJ had similarly referred to Társaság as ‘a landmark decision on freedom to information’ that demonstrated that Article 10 could be invoked ‘not only by those who seek to give information but also by those who seek to receive it’.42 He considered it to be ‘plainly consistent with Article 10, as it is now interpreted, that there should be freedom of access to [official documents]’.43 The Court of Appeal in Kennedy commended the conclusion of the First Tier Tribunal that Article 10 ECHR had advanced towards a broader interpretation so as to include some form of right to receive information from a public authority as based on an ‘excellent, clear and comprehensive analysis’.44 The court concluded, however, that it was bound by the decision of the Supreme Court in Sugar (No 2).45

33 BBC v Sugar (dec’d) (No 2) [2012] UKSC 4, [90]–[97]. Lord Mance agreed: [113]. 34 ibid [94]. 35 Kennedy (n 3) [62]. See also [188] (Lord Wilson). 36 ibid [57]. 37 ibid [76]–[89]. 38 ibid [94]. This perspective was acknowledged by Green J in R (on application of Privacy International) v HMRC Commissioner [2014] EWHC 1475 (Admin), [169]. Counsel accepted in that case that a contrary contention was no longer open. 39 Kennedy (n 3) [94] and [96], and [147] respectively. 40 [2010] EWCA Civ 343, [41]–[45]. 41 ibid [42]. 42 [2010] EWCA Civ 715, [76]. 43 ibid [77]. The judge acknowledged that limitations on this right might be properly imposed. In the instant case, his difficulty arose because there were arguments based on freedom of expression going either way, in which context he considered that the ‘invocation of the freedoms enshrined in Article 10 provide no assistance’. 44 Kennedy v Charity Commission [2012] EWCA Civ 317, [26]. 45 ibid [36]–[58]. This conclusion can be juxtaposed with Lord Mance’s view in Kennedy that Lord Brown’s view – and his – on the Art 10 point could not be treated as part of the ratio decidendi of the case. Notably, Etherton LJ also noted that were the point not part of the ratio in Sugar (No 2), he would nonetheless have followed Lord Brown’s view as persuasive authority.

50  Andrew Scott and Abbey Burke In his minority judgment in Kennedy, Lord Wilson offered both a normative and a ­jurisprudential critique of the position adopted by the majority. First, he noted that: a brief reflection on the nature of freedom of expression suggests difficulties with [the approach in Leander] … without freedom to receive certain information, there is no freedom to proceed to express it; and a person’s freedom to express the information is likely to carry much greater value for the public if the person holding the information is unwilling to impart it to him.46

He then conducted his own review of the European case law.47 He insisted that the Társaság decision could not be considered ‘irregular’, nor ‘some arguably rogue decision which, unless and until squarely validated by the Grand Chamber, should be put to one side’.48 Rather, he depicted it as a judgment whose ‘importance was quickly and generally recognised’; both it and the development in the jurisprudence more broadly were ‘not in conflict with the basic Leander approach’, but were rather ‘a dynamic extension of it’.49 Lord Carnwath agreed that ‘at first sight [Társaság] represents an unexpected departure … [that could be] seen as a “landmark decision”, or as an aberration by a single section of the court’.50 In his view also, however, it was possible to discern ‘a general “direction of travel” away from a strict application of article 10, at least in cases involving journalists or other “watchdogs” seeking information of genuine public interest’.51 In that light, Lord Wilson asserted that the Supreme Court should ‘confidently conclude that a right to require an unwilling public authority to disclose information can arise under article 10’.52 He emphasised that ‘in no sense [would] this betoken some indiscriminate exposure of sensitive information held by public authorities to general scrutiny’.53 Rather, any review would be conducted in accordance with Article 10(2) ECHR, with the state being permitted a wide margin of appreciation in its determination of the contours of an appropriate scheme for the release of information. It seems likely that the position adopted by the majority of the Supreme Court in Kennedy was attributable less to outright distaste for a tightly drawn association between Article 10 and the right of access to information and more to a desire definitely to inculcate rights values into the common law. Without a readily available theoretic justification for the linkage, the majority sought a basis in positive law. However, for them to have located this basis in Strasbourg jurisprudence could have generated its own problems. As the following paragraphs describe, it may be that with the principle now installed in the common law, subsequent developments in Strasbourg will allow an assimilation with the expanded right to freedom of expression. The theme will almost certainly return to the Supreme Court at some point, and on that occasion their Lordships will be free to determine the issue afresh.

46 Kennedy (n 3) [179]. 47 ibid [181]–[187]. 48 ibid [188]. 49 ibid [188]. 50 ibid [215]–[216]. 51 ibid [217]. 52 ibid [189]. 53 ibid [189]. This was a response to the view that a general principle of access would permit the harmful release of information that would not occur under a more tailored statutory scheme of access.

The Access to Information Dimension of Positive Free Speech  51

III.  United Kingdom: A Common Law Principle of Open Government At the same time as they rejected the development of an equivalent right as a component part of the Article 10 right to freedom of expression, in the case of Kennedy v Charity Commission, the Supreme Court developed a previously unheralded common law principle of open government.54 This can be seen as an uncontroversial development. It has long been expected that, in line with the constitutional principle of responsible government, the UK executive will provide an account of itself in legal and political forums. Yet, the precise contours of this new principle remain somewhat underdetermined, and its field of application uncertain. Perhaps unsurprisingly, it has been cited and deployed in a range of cases before the British courts as lawyers seek to discern its limits and exploit the opportunities it offers to the benefit of their clients.55 In Kennedy, their Lordships were faced with a question over the operation of the absolute exemption from disclosure found in section 32(2) of the Freedom of Information Act 2000 (the 2000 Act). The provision applies to documents placed in the custody of, or created by, a person conducting an inquiry, for the purposes of that inquiry. The appellant was a journalist from The Times newspaper. He wished to gain access to documents deriving from an inquiry conducted into the ‘Mariam Appeal’ – a charity established by the former MP George Galloway ostensibly to provide medical treatment for sick children. His intention was better to understand a conclusion drawn by the Commission regarding the integrity of the charity. Notwithstanding the legal right of access to information under section 1 of the 2000 Act, the Commission refused to release the information sought, citing the section 32(2) exemption. The case had been argued solely by reference to the putative Article 10 ECHR right to receive information from public authorities. A core contention was that section 32 should be ‘read down’ so as to apply only during the period in which an inquiry was ongoing. As noted above, notwithstanding the emergent Strasburg jurisprudence, a majority of the court found that there was no such strand to the Article 10 right. They concluded therefore that section 32 could not be read down in the manner requested.56 This was not the end of the matter. Without it being argued by the parties, the majority of the court also concluded – explicitly obiter – that a result equivalent to that sought by the journalist could be achieved under section 78 of the 2000 Act. That section retained in place any pre-existing scheme for the disclosure of information based on statute or the common law. By reference to the new-found common law principle, the court found itself able to interpret the general principles of the Charities Act 1993 – a statute which at that time prescribed the regulatory regime for charities – so as to provide a right of access to public documents that might be relied on to secure the release of the information sought.57 54 ibid. 55 One of the first, and still most notable, such cases was that of Privacy International (n 38); for further, see text accompanying nn 83–94 below. 56 For the contrary view, see Kennedy (n 3) [200] (Lord Wilson) and [221]–[233] (Lord Carnwath). 57 ibid [101]. The general functions of and duties on the Commission were set out in the 1993 Act, and included the task of disseminating information in connection with the performance of its functions (s 1C(2)) and the requirement that, in performing its functions, the Commission should have regard to the principles of best regulatory practice, including those of accountability and transparency: s 1D(2).

52  Andrew Scott and Abbey Burke The court suggested that the journalist should apply for such disclosure, and then pursue proceedings by way of judicial review should it not be made. Their Lordships described their approach as the development of a common law discretion. Lord Mance noted that this approach has ‘become a fruitful feature of United Kingdom jurisprudence’.58 He drew an analogy between court proceedings and inquiries such as those undertaken by the Charity Commission under the 1993 Act, and identified ‘a common law presumption in favour of openness’ that was common to both.59 Important to this presumption was the strong public interest in both the functioning and regulation of the charity in question and its reflection in the statutory objectives prescribed for the Commission under the Act.60 Lord Mance suggested that, in the absence of strong countervailing interests, the Charity Commission should accede to any future request for information of this type.61 The majority of the court considered that proceedings intended to give effect to this common law principle should be brought by way of an application for judicial review. Lord Mance noted that ‘the common law no longer insists on the uniform application of the rigid test of irrationality once thought applicable under the so-called Wednesbury principle’.62 While accepting that appropriate weight must be given to the decision-maker’s findings of fact and exercise of discretion, he emphasised that the more grave the issue that the challenged decision has determined, the more rigorous can be the scrutiny to which it is subjected by the courts.63 Among the categories of more serious circumstances are ‘those where a common law right or constitutional principle is in issue’.64 In the immediate context, Lord Mance invoked ‘the principles of accountability and transparency … contained in the Charities Act and reinforced by common law considerations’.65 Hence, ‘it would be necessary for the court to place itself so far as possible in the same position as the Charity Commission, including perhaps by inspecting the material sought’.66 Their Lordships emphasised repeatedly that this would entail an approach to the review function equivalent to that which might proceed under a rights-based analysis, and that the outcome would almost certainly be the same.67 Their Lordships’ judgments were not entirely clear on the question of how far this constitutional principle runs. As noted, Lord Mance drew an explicitly imperfect analogy between courts and bodies that conduct inquiries.68 Lord Toulson explained that ‘although

58 Kennedy (n 3) [38] (Lord Mance). 59 ibid [47]. Lord Mance recognised, however, that it would be wrong automatically to assimilate the ­requirements of openness as between the two contexts: see [48]. 60 ibid [49]. 61 ibid [49]–[50], [56] and [59]. 62 ibid [51]. 63 ibid [51], citing R v Secretary of State for the Home Department, ex parte Bugdaycay [1987] AC 514, 531 (Lord Bridge). 64 Kennedy (n 3) [55]. 65 ibid [55]. 66 ibid [56]. See also [132] (Lord Toulson). 67 ibid [40], [49]–[50] and [56] (Lord Mance); [133] (Lord Toulson); and [219] (Lord Carnwath). A similar point was made by Supperstone J in R (on the application of Good Law Project Ltd) v Secretary of State for Exiting the European Union [2018] EWHC 719, [5]: ‘the analysis to be conducted when considering whether to disclose documents under the common law and Art 10 routes … is not … materially different from the balancing exercise to be conducted under FOIA’. 68 Kennedy (n 3) [48]–[49].

The Access to Information Dimension of Positive Free Speech  53 such inquiries and hearings may vary considerably in nature and scope, it is fair to describe the conduct of them as a quasi-judicial function’.69 This might suggest that the obligation is a close relation to the principle of open justice, and that – without understanding the term too narrowly – it applies only to bodies that might be considered ‘judicial’ or ‘quasi-judicial’ in character. The principle introduced by the Supreme Court could also be understood to be of wider application. While the vehicle for discussion of the emergent principle was its application to inquiries, the overall tenor of the majority judgments suggested a broader field of application. Lord Mance indicated that the requirements imposed by the principle would depend on the circumstances surrounding each exercise of power. On one side, the public interest in disclosure would rely, in part at least, on the degree of public interest in the subject matter of the decision.70 On the other side, countervailing arguments running against disclosure of particular information would ‘differ in nature and weight, according to whether one is considering court or inquiry documents, and in the latter case according to the nature of the inquiry’.71 Of course, every public body is engaged, almost by definition, in activity in which there is a public interest. As Lord Toulson recognised: every public body exists for the service of the public, notwithstanding that it may owe particular duties to individual members of the public which may limit what it can properly make public … there may … be other reasons, apart from duties of confidentiality, why it would not be in the public interest or would be unduly burdensome for a public body to disclose matters to the public, but the idea that, as a general proposition, a public body needs particular authority to provide information about its activities to the public is misconceived.72

This would suggest a default presumption that every public authority should disclose information on request, subject to the possibility that this presumption might be outweighed by countervailing interests. Of course, for many public bodies, a statutory scheme is set out in the Freedom of Information Act 2000. Against this interpretation sits the strident criticism offered of the equivalent general principle based on Article 10 that counsel had asked the court to recognise.73 Not all the judges in Kennedy recognised the proposed new common law principle. Indeed, those in the minority were bemused by the development. Lord Wilson considered the introduction of the principle ‘surprising’ and ‘profoundly unsatisfactory’, noting that ‘there is certainly no example of its prior operation or other recognition of its existence’.74 Lord Carnwath was concerned that the common law approach had been ‘unsupported by any of the parties before us, in my view for good reasons … [and hence that the court had] not had the advantage of full argument’.75

69 ibid [122]. See also [108]–[111]. It would appear that Lord Toulson’s motivation to draw this analogy arose from a desire to explain the seemingly parallel treatment meted to courts and inquiries in s 32 of the 2000 Act. 70 ibid [49]. 71 ibid [50]. 72 ibid [107]. 73 ibid [147]–[148] (Lord Toulson). 74 ibid [191] and [198]. He highlighted potential legal and practical pitfalls in the operation of the regime proposed by the majority: see [199]. 75 ibid [202] and [234].

54  Andrew Scott and Abbey Burke Lord Carnwath also queried the suggestion that inquiries exercise ‘judicial power’ or some form of ‘quasi-judicial function’. He insisted that in the instant case the role of the Commission ‘[was] administrative, rather than judicial’.76 Indeed, he baulked at the use of the term ‘quasi-judicial’, which he considered to be a ‘somewhat imprecise and outmoded expression’ that was unsupported by authority or explanation.77 For Lord Carnwath, the analogy drawn between courts and inquiries was premised on a ‘basic fallacy’ of equivalence. In his view, there was no existing authority to support the suggestion that common law principles relating to disclosure of documents in the courts could be transferred directly to inquiries.78 Lord Carnwath did allow that ‘the courts might … in due course develop a more general principle of openness, applicable also to different forms of statutory inquiry’.79 He argued, however, that this ‘would involve a significant extension to the existing law – arguably a bolder leap into the unknown than the modest step we are being asked to take (after full argument) in relation to article 10’.80 Like Lord Wilson, Lord Carnwath preferred that the Supreme Court should endorse the emergent Strasbourg principle.81 It is tempting to understand the position of the majority in Kennedy v Charity Commission as being motivated by a general aversion to a perceived overemphasis on the rights-based jurisprudence and the associated diminution of the importance of the common law. Lord Mance complained that: since the passing of the Human Rights Act 1998, there has too often been a tendency to see the law in areas touched on by the Convention solely in terms of the Convention rights … [when] in view of the contribution which common lawyers made to the Convention’s inception, they may be expected, at least generally even if not always, to reflect and to find their homologue in the common or domestic statute law … the natural starting point in any dispute is to start with domestic law, and it is certainly not to focus exclusively on the Convention rights, without surveying the wider common law scene.82

He added that a ‘greater focus in domestic litigation on the domestic legal position might also have the incidental benefit that less time was taken in domestic courts seeking to interpret and reconcile different judgments’.83 Lord Toulson concurred, emphasising that the common law method of ‘gradual adaption and development … to meet current needs’ had not been eviscerated on the enactment of the Human Rights Act 1998. In that context, he lamented that ‘since then there has sometimes been a baleful and unnecessary tendency to overlook the common law … [when] it was not the purpose of the Human Rights Act that the common law should become an ossuary’.84 76 ibid [236]–[237]. 77 ibid [236]. 78 ibid [240]–[241]. Lord Carnwath emphasised that there was no underpinning constitutional rule, comparable to the open justice principle applicable to courts, that required the Commission to sit in public. Hence, the extension of the open justice principle to the provision of court documents could have no parallel in this other context: ‘the essential foundation that is needed for application of the Guardian News approach is wholly absent’. 79 ibid [240]. 80 ibid [240]. 81 See the foregoing section above. 82 Kennedy (n 3) [46]. He also cited Toulson LJ in R (on the application of Guardian News and Media Ltd) v City of Westminster Magistrates’ Court [2012] EWCA Civ 420, [88] to the effect that comparative common law ­scholarship is often preferable. 83 Kennedy (n 3) [46]. See also [59]. 84 ibid [133].

The Access to Information Dimension of Positive Free Speech  55 Certainly, frustration at the need to decipher principle from often contradictory Strasbourg authority is one explanation for the approach adopted by the majority. The complaint that ‘the Strasbourg jurisprudence is neither clear nor easy to reconcile’ and that ‘Strasbourg has spoken on a number of occasions to apparently different effects’, however, was perhaps something of a smokescreen.85 We would suggest that the real purpose of the court in asserting the primacy of the common law in this context was to imbue domestic law with the values of rights in order to hedge against the risk of any future repeal of the Human Rights Act.86 As such, the Kennedy case stands as a significant moment in the resurgence of common law constitutionalism.87

IV.  United Kingdom: Subsequent Development of the Common Law Principle Since Kennedy, journalists and NGOs have sought to rely on the principle announced in that case on a number of occasions. They have enjoyed varying degrees of success. The principle certainly offers a promising rhetorical tool for those keen to access information held by reluctant bureaucracies. This potential has been embraced by some judges. Conversely, other courts have been swayed by competing impulses, with concerns over well-meaning but injudicious overreach coming to the fore. The following paragraphs note three significant decisions that postdate, and in some measure flow from, the emergence of the common law principle in Kennedy: Privacy International, Ingenious Media and the Good Law Project cases.88 Insight can also be drawn from the further interplay between journalist and public authority in the aftermath of the Kennedy request itself.

A.  Privacy International A first application of the common law principle was seen very shortly after the judgment of the Supreme Court in the approach adopted by the Administrative Court in R (on the application of Privacy International) v Commissioner for HM Revenue & Customs.89 That case concerned the disclosure of information on the export control functions of HMRC. HMRC had refused to disclose whether it was investigating a UK-based company, Gamma International, for an alleged breach of export control laws. Privacy International – an NGO focused on issues at the intersection between technologies and rights – applied for judicial review of the decision. It had earlier provided HMRC with a dossier of evidence suggesting that the company had, contrary to UK law, supplied surveillance software to repressive

85 ibid [59] (Lord Mance). 86 C Gearty, On Fantasy Island (Oxford, Oxford University Press, 2016) 196. 87 R Clayton, ‘The Empire Strikes Back: Common Law Rights and the Human Rights Act’ [2015] PL 3. Other important cases in this regard include A v British Broadcasting Corporation [2014] UKSC 25; R (on application of HS2 Action Alliance Ltd) v Secretary of State for Transport [2014] UKSC 3; Osborn v Parole Board [2013] UKSC 6. 88 See Privacy International (n 38); R (on the application of Ingenious Media Plc) v Revenue and Customs Commissioners [2016] UKSC 54; Good Law Project (n 67). 89 Privacy International (n 38).

56  Andrew Scott and Abbey Burke regimes in Ethiopia and Bahrain, among other places.90 The NGO believed that the technology had been used covertly to monitor persons who were subsequently arrested, tortured or killed.91 HMRC was prohibited from disclosing information it held in connection with its export control functions in response to a request under the Freedom of Information Act 2000.92 Through section 18(2) of the Commissioners for Revenue and Customs Act 2005, however, it enjoyed a power to make disclosures ‘for the purposes of a function of the Revenue and Customs’.93 HMRC had written to the NGO insisting that it lacked any power to provide the requested information, could not ‘comment on particular cases’ and was ‘unable to keep you or any other third parties informed of the progress of the investigations’.94 It accepted at trial that this was incorrect. In his judgment, Green J noted that while the decision in Kennedy involved a very different factual and statutory context, like the Supreme Court, he insisted that a decisionmaker must balance the factors for and against disclosure, paying close attention to the particular circumstances of the case. These factors should include the status of the persons seeking information and the nature of the information sought.95 In particular, given its status and prior role in the case, a body such as Privacy International would have a special claim to information. He considered that such organisations act as ‘guardians of the public conscience’, and that ‘their very existence and the pressure they bring to bear on particular issues and upon those who are responsible for governance of those issues, is one of the significant checks and balances in a democratic society’.96 At trial, HMRC suggested a number of factors that it considered weighed against disclosing the information sought – the possibility of forewarning a suspect, causing reputational harm or disclosing confidential material – but the judge considered that it had not in fact taken such factors into account when reaching its decision. Indeed, he delivered an eight-point excoriation of the body’s handling of the request, noting that it jumped straight to refusal, ‘[relying] on abstract reasoning divorced from the actual evidence and hence operat[ing] on the premise that the actual facts were an irrelevance’.97 The approach of the Administrative Court in the Privacy International case supports a broad understanding of the reach of the common law principle. The HMRC functions at issue were described by Green J as ‘involv[ing] legal process but at a much earlier stage’, with ‘investigations conducted by HMRC [being] equivalent to police investigation which might or might not lead to a prosecution … operations which are necessary precursors to court proceedings’.98 Clearly, such investigations are not ‘judicial’ or ‘quasi-judicial’ in 90 ibid [1]. 91 ibid [13]. 92 Commissioners for Revenue and Customs Act 2005, s 23(1). The restriction applies insofar as the information sought specifies the identity of the person to whom it relates or would enable the identity of such a person to be deduced. 93 The Act also provides in s 20 for ‘public interest disclosures’, but these must be made to one of a number of specified public authorities, and not to third parties or the world at large. 94 Privacy International (n 38) [17]. 95 ibid [61]. 96 ibid [77]. He added that ‘such bodies, like the press, hold Government to account, campaign on issues of public importance, act as focal points for complaints and pursue injustices’: [79]. 97 ibid [68] and [63]–[72] generally. 98 ibid [62].

The Access to Information Dimension of Positive Free Speech  57 nature. Green J expressly rejected the invitation by counsel to read the principle narrowly in this way.99 The availability of the common law right of access to government information is not restricted to information held by inquiries, statutory watchdogs or similar bodies, but would appear to pertain to public authorities generally. As Green J noted, the Supreme Court had been ‘at pains to point out that the common law treated openness as very important and, with all the necessary provisos and caveats, that message can in some measure carry through’.100 Privacy International illustrates a shift in judicial attitude towards a position where a public body faced with a request for information about its practices will be expected to apply a presumption of transparency, rebuttable only for concrete and compelling reasons. This presumption is given life in the law independent of parliamentary dispensation.

B.  Decision of the Supreme Court in Ingenious Media When hearing the case of R (on the application of Ingenious Media Holdings plc) v HM Revenue & Customs, the Supreme Court was directed to neither Kennedy nor Privacy International.101 Yet, Ingenious Media offers the converse factual scenario to Privacy International: it saw a public authority release information without due attention to and careful weighing of the countervailing interests that militate in favour of non-disclosure. The judgment in Ingenious Media highlights how a de facto limitation on access to information based on the common law principle might easily develop. This is especially likely if courts are presented with circumstances in which a public authority fails adequately to consider the full breadth of interests at stake when deciding whether to disclose information that has been requested. Like Privacy International, Ingenious Media concerned the power of HMRC to disclose information ‘for the purposes of a function of the Revenue and Customs’, albeit that it focused on tax rather than export control functions.102 The applicant, who marketed film investment schemes, had applied for judicial review of the decision of a senior HMRC official to give an interview to journalists from The Times newspaper. In a subsequent article, the applicant was identified as a prominent promoter of tax avoidance schemes and a ‘threat’ to the Revenue. In remarks that were not reported, the film schemes were also described by the official to the journalists as ‘scams for scumbags’.103 HMRC defended the legality of its actions by arguing that it was a part of its revenue-collecting functions to cultivate good

99 ibid [62]. 100 ibid [62]. In doing so, the judge echoed the view expressed by Lord Toulson in Kennedy (n 3) [107] that public authorities do not need any explicit statutory basis to disclose information to the public – see text accompanying n 72 above. 101 Ingenious Media (n 82). 102 Commissioners for Revenue and Customs Act 2005, s 18(2). A further connection between the cases is that in Privacy International (n 38) Green J cited the first instance decision handed down by Sales J in Ingenious Media [2013] EWHC 3258 approvingly, considering it ‘informative in a general sense’ regarding the scope of the HMRC discretion to disclose information: [55]. The Supreme Court in Ingenious Media, however, reversed the first-instance decision in the case. 103 Ingenious Media (n 82) [8]–[11].

58  Andrew Scott and Abbey Burke relationships with the press in order to publicise disapproval of elaborate tax avoidance schemes.104 Giving the judgment of the court, Lord Toulson adopted a narrow approach to the type of disclosures permitted for the purpose of the HMRC revenue collecting function, rejecting its contention that the disclosure was permissible. In so doing, he emphasised the ‘fundamental’ common law right of taxpayers to confidentiality when they provide information to HMRC about their affairs, and expressed abhorrence at officials ‘discussing the affairs of individual taxpayers with journalists’.105 The decision of the Supreme Court serves as a reminder that the common law principle of openness exists alongside other long-standing common law and statutory principles – in this case, confidentiality – which the courts will also guard fiercely. The reasoning it offered reflects a broader concern with the power of public authorities to damage the reputation of individuals. This ambivalence over the installation of a broad common law right to know was presaged in the judgments handed down in Kennedy. In his dissent in that case, Lord Wilson complained that the incipient common law principle of open government expounded by the majority was ‘so vague and generalised’ that he regarded the determination of any future request for disclosure as being ‘impossible to predict’.106 Lords Mance and Toulson were specifically concerned about overgenerosity in disclosure by public authorities. They worried that factors in the balance against disclosure may not be adequately considered by public authorities, and that unwitting harm could be caused to individuals. For instance, Lord Toulson noted that information held by the state regarding individuals ‘may include material detrimental to a person’s reputation’,107 and that ‘unlimited access to every item … could cause serious harm to innocent people’.108 Underpinning these concerns is the spectre of government hoovering up swathes of sensitive information about individuals and then incontinently releasing it, whether through inattention, inadvertence or incompetence. There is a risk, however, that successful claims against authorities – such as Ingenious Media – may induce a risk-averse approach. If authorities withhold information without justification, as was the case in Privacy International, this can be corrected by judicial review. Equally, however, if authorities disclose information without justification, as occurred in Ingenious Media, they will leave themselves open to legal action. For this reason, public authorities faced with common law requests for access to information may become unnecessarily and highly cautious, lest they incur liability for disclosures that are later shown to be illegal. Any default practice against disclosure would serve as a de facto constraint on the use of common law claims as a timely and effective means of accessing information.

C.  Good Law Project A second decision that indicates potential limitations upon the common law principle of openness is that of the Administrative Court in Good Law Project.109 This case concerned 104 ibid [13]. HMRC also maintained that it had not intended that the applicant should be identified publicly, but understood that the interview was to be an ‘off the record’ briefing about HMRC tax enforcement activities. 105 ibid [19] and [34]. 106 Kennedy (n 3) [199]. 107 ibid [138]. 108 Lord Toulson was quoting a US court decision: United States v Amodeo, 71 F 3d 1044 (1995). 109 Good Law Project (n 67).

The Access to Information Dimension of Positive Free Speech  59 a request made by a newly created, crowd-funded NGO for two categories of expert report held by the Brexit Secretary.110 The documents sought comprised detailed studies on the likely impact of Brexit across 58 sectors of the UK economy and an HM Treasury report comparing the likely economic costs of Brexit with likely economic gains from entering new free trade agreements.111 The sectoral analyses, in particular, were the subject of much political contention. The then Brexit Secretary, David Davis MP, had first told MPs that these analyses existed in ‘excruciating detail’, but later clarified that no detailed information could be published because it still needed to be ‘collated’ from various sources.112 The applicants had made their request to the Secretary of State on the basis of the emergent common law principle of open government (and Article 10 ECHR), although it might equally have been made under the Freedom of Information Act.113 They argued that they required the information to contribute, in their capacity as public watchdogs, to a debate of vital public interest: whether, and on what terms, the UK should leave the European Union. The Secretary of State refused to disclose the economic studies or confirm the existence of the Treasury report, purporting to have decided this under the freedom of information regime.114 Before the Administrative Court, the claimants contended that the freedom of information regime should not have been used, as a rejected request appealed under the 2000 Act would not have been determined in time to inform public debate on Brexit.115 In support of this position, the claimants relied on a sample of 31 recent First-tier Tribunal decisions concerning government departments. This showed that the average time from initial request to Tribunal decision was 467 working days.116 Supperstone J rejected the claim, holding that the government was correct to consider the requests under the FOI regime: Parliament, by [the Freedom of Information Act 2000], has created a specialist statutory mechanism for addressing requests for information held by public authorities … [and] the claimants cannot by framing their requests [on the basis of the common law and Article 10 ECHR] avoid the legal regime established by Parliament to deal with disputes arising from information requests.117

In particular, this required that the appeal against the decision to refuse disclosure should proceed through the process prescribed in the statute: internal review, followed by consideration by the Information Commissioner and then the First-tier Tribunal, from which there could be an appeal on a point of law. The judge concluded that there was ‘nothing fundamentally unfair about the timeframe imposed by the specialist statutory mechanism created

110 The NGO had been established in 2017 by Jolyon Maugham QC and Green Party MEP Molly Scott Cato. 111 Good Law Project (n 67) [1]. 112 See generally L Maer and G Ryan-White, ‘Exiting the EU: Sectoral Assessments’, House of Commons Library Briefing Paper no 8128 (9 March 2018). 113 The applicants cited both Kennedy (n 3) and Helsinki (n 6): for discussion, see the sections above and below, respectively. 114 Regarding the sectoral studies, the Secretary of State relied on the qualified exemptions in ss 27 (‘international relations’) and 29 (‘the economy’) of the 2000 Act, on the basis that releasing the studies would harm the UK’s Brexit negotiating position; and on s 35 (‘formulation of government policy’) on the basis that the ministers making policy on Brexit needed a ‘safe space’ for discussion. 115 At the time, both sides of the negotiation expected to conclude by October 2018. 116 T Pitt-Payne QC, ‘Statement of Facts and Grounds in Support of Claim for Judicial Review’ (7 December 2017) [36], https://mollymep.org.uk/wp-content/uploads/Statement-of-Facts-And-Grounds-Dec17.pdf. 117 Good Law Project (n 67) [3].

60  Andrew Scott and Abbey Burke by Parliament’.118 Hence, there was a suitable alternative remedy to an application for judicial review, and permission to proceed with the judicial review application was refused.119 The decision of the Administrative Court in Good Law Project suggests that judges may be resistant to recognising any freestanding common law or Article 10-based claims for disclosure unless the freedom of information regime is somehow expressly excluded. The rhetorical platform that Article 10 and the Kennedy common law principle afforded the litigants in the case was not altogether wasted, however. After the NGO announced its intention to appeal, the Brexit Select Committee released a document containing a summary of the economic studies it had sought. The NGO noted that the battle to release this information was ‘won by political pressure’, and suggested that their litigation helped keep the government under such pressure.120

D.  The Aftermath of Kennedy As noted above, Kennedy opened the question of how public authorities are likely to respond to a request for information based on the common law principle of openness. The continuation of the Kennedy story in the aftermath of the Supreme Court decision itself provided an early illustration. In the wake of the decision, the journalist submitted a further request to the Charity Commission, relying on his common law rights ‘as identified by the Supreme Court’.121 Two months later, the Commission disclosed some documents, but set out prolix reasons for denying access to the remainder of those requested. These justifications were based on legal professional privilege, confidentiality, the need to maintain good working relationships with other public authorities and not inhibit ‘the free and frank exchange of views’, ‘preventing disorder and crime’, protecting ‘the rights of others’ and other factors.122 Such reasons comprise a melange of those available under Article 10(2) ECHR and the 2000 Act. Hence, the Commission’s response satisfied Lord Wilson’s prediction that in dealing with common law access requests, authorities would be likely to avoid reinventing the wheel and instead ‘reach for the helpful prescriptions in the FOIA [Freedom of Information Act] and, in effect, work in its shadow’.123 For the sake of completeness, it can be noted that Kennedy did not then seek judicial review of the refusal to disclose the information. He considered that this would add both to the unreasonable delay that had occurred and to the enormous legal costs already incurred. Instead, he complained to the ECtHR, arguing that the absolute exemption in section 32 violated his Article 10 rights, and that to require him to pursue judicial review would constitute an excessive and disproportionate burden on his enjoyment of the right to freedom

118 ibid [17]. In reaching this conclusion, the judge accepted the Information Commissioner’s evidence that these timescales could be curtailed by her practice of marking high-profile cases as a priority, while the First-tier Tribunal could also expedite important cases. 119 The judge distinguished Kennedy, noting that as there was an absolute exemption in play in that case there would be no grounds for appeal available under the FoI regime – see Good Law Project (n 67) [4]. 120 Good Law Project, ‘We Got What We Came For’, Good Law Project Blog, 13 March 2018, https://goodlawproject.org/we-got-what-we-came-for/. 121 Kennedy (n 3) [14]. 122 ibid [14]–[15]. 123 ibid [199].

The Access to Information Dimension of Positive Free Speech  61 of expression.124 His application was stayed for a period, however, when the claim in Helsinki was referred to the Grand Chamber. Moreover, his application was ultimately rejected as inadmissible by the First Section of the ECtHR on the basis that domestic ­remedies – the freedom to apply for judicial review – had not been exhausted.125

V.  Dénouements à Strasbourg: Confirmation of the Article 10 Right of Access In Magyar Helsinki Bizottság v Hungary, the Grand Chamber of the ECtHR offered a rejoinder to the conclusions of the majority of the Supreme Court in Kennedy.126 Having conducted its own review of Strasbourg jurisprudence, the ECtHR expressly recognised that a government’s refusal to disclose information to a public watchdog may violate Article 10 ECHR. Like Privacy International, Helsinki involved a request for information by a well-regarded human rights organisation, the Hungarian Helsinki Committee.127 As part of a long-standing project concerning access to justice, the committee was undertaking research on the mode of appointment of defence counsel. Such counsel were appointed by the police departments that had investigated the crimes to be tried, with many of the lawyers concerned undertaking only work of this type. The lawyers were therefore heavily dependent on the exercise of the discretionary powers of the police to secure their livelihoods. The NGO wished to test the theory that the quality of legal defence received by suspects was adversely affected by this conflict of interest, and so sought information on the pattern of appointments. It requested the names of defence counsel and the number of times each lawyer had been appointed to cases from 28 police departments. The NGO maintained that this was public interest data. Most of the police departments complied with the request; the NGO brought legal proceedings against two that refused to disclose the information sought.128 The Hungarian Supreme Court did not compel the release of the information by the police departments. Rather, it found that the defence counsel were not ‘persons performing public duties’, and that the information requested was therefore exempt ‘personal data’ under the Hungarian Data Act. The Grand Chamber decided, by a majority of fifteen-to-two, that the NGO enjoyed an Article 10(1) right of access to the information held by the public authorities concerned.129 In doing so, it reiterated its general teleological approach, explaining that ECHR rights must be rendered ‘practical and effective, not theoretical and illusory’. It insisted that the absence of a right of access to information under Article 10 could ‘lead to situations where the freedom to “receive and impart” information is impaired in such a manner and to such a degree

124 Communicated Case app no 64367/14 Times Newspapers Ltd and Kennedy v UK [2015] ECHR 359. 125 Times Newspapers Ltd and Kennedy v UK, app no 64367/14, unreported, 13 November 2018. 126 [2016] ECHR 975. 127 The group monitors the implementation of international human rights standards in Hungary, provides legal representation to victims of alleged human rights abuses, and promotes legal education and training – see Helsinki (n 6) [10]. 128 See generally, ibid [11]–[30]. 129 ibid [171]–[180]. This was in the face of an intervention by the UK Government.

62  Andrew Scott and Abbey Burke that it would strike at the very substance of freedom of expression’.130 Moreover, it explained that ‘the time has come to clarify the classic principles’.131 This clarification involved the restatement of accepted principles,132 but then also – demurring from the general principle that Article 10 afforded no right of access nor imposed any positive obligation on a state to impart information to an individual – the confirmation that a right of access may arise where such access to the information ‘is instrumental for the individual’s exercise of his or her right to freedom of expression, in particular “the freedom to receive and impart information” and where its denial constitutes an interference with that right’.133 The ECtHR identified four criteria that would assist in the determination as to when such a right is engaged: 1. whether the purpose of the person in requesting access to the information held by a public authority is to enable his or her exercise of the freedom to ‘receive and impart information and ideas’ to others;134 2. whether the information sought is of legitimate public interest;135 3. whether the applicant is acting in the capacity of a ‘public watchdog’ (whether as a journalist, NGO, academic, blogger or social-media poster);136 and 4. whether the authority has the information ‘ready and available’.137 Having determined that these criteria were satisfied on the facts of the instant case, the court concluded further that the interference with the NGO’s right of access was not proportionate. It noted that the very broad definition of ‘personal data’ applied by the domestic courts, which caught even statistical data about public defenders’ professional activities, effectively ‘excluded any meaningful assessment of the applicant’s freedom of expression rights’.138 In light of the Helsinki decision of the Grand Chamber, there would seem to be no clear reason why the Kennedy common law principle of open government and the emergent ECHR-based right of access to government information should not now be assimilated. Both the principle and the right emphasise the special status of actors who seek information in order to disseminate it in the public interest. Both also recognise that there are numerous countervailing reasons for non-disclosure, which decision-makers must weigh carefully in each case. Litigants before the UK courts have already noted these similarities, pleading their case on the basis of both the principle and the right.139 It seems likely that, in future,

130 ibid [155]. 131 ibid [156]. 132 These principles were that: (i) the right to freedom to receive information prohibits a government from restricting a person from receiving information that others wish or may be willing to impart to him; (ii) the right to receive information cannot be construed as imposing on a state positive obligations to collect and disseminate information of its own motion; and (3) Art 10 does not confer on the individual a right of access to information held by a public authority nor oblige the government to impart such information to the individual – see Helsinki (n 6) [156]. 133 ibid [156]. 134 ibid [158]–[159]. 135 ibid [160]–[163]. 136 ibid [164]–[168]. 137 ibid [169]–[170]. 138 ibid [199]. 139 This was done, for example, in Good Law Project (n 67).

The Access to Information Dimension of Positive Free Speech  63 the UK courts will explicitly recognise the shared ground covered by these two paths to disclosure. As Lord Mance anticipated in Kennedy, ‘in time … a synthesis may emerge’.140

VI. Conclusion Almost irrespective of its legal basis, the emergence of the legal principle of openness as a component of a positive, or facilitative, free speech right is potentially a development of profound significance. The reach of such a principle could become momentous indeed. The domestic and European iterations of this principle represent promising legal and rhetorical tools for those seeking information to wield against reluctant bureaucrats. Insofar as the principle is a component of the common law, this opens questions as to the potential for its use in an array of contexts where statutory actualisation has been inadequate or altogether lacking. Moreover, given that the traditionally conservative English judges have been able to discern the principle in English common law, this raises the question as to whether the judiciaries of other common law jurisdictions might similarly unearth such principles. The most immediate context in which the principle of open government may have some play is in the review of the boundaries of freedom of information legislation.141 Overbroad exemption clauses may be read down. Access obligations might be extended to private entities to whom governments outsource many of their traditional functions.142 Heavy-handed use of the exemption applicable to vexatious or repeated requests, such as that found in section 14 of the 2000 Act, might be more easily challenged.143 The same logic applies to the framing of other, more specific, legislation that allows individuals to seek access to information held by the state, such as the Local Government Act 1972 or the Environmental Information Regulations 2004. The principle may also have significant purchase outside of the freedom of information domain. For instance, it might bear on questions of party anonymity in legal proceedings,144 or public access to courts.145 It might found an argument that a public authority acts unlawfully should it include a confidentiality clause in a dispute settlement contract concerning a matter in which there was a legitimate and overriding public interest.146 Similarly, it might 140 Kennedy (n 3) [46]. 141 C Knight, ‘Article 10 and the Right of Access to Information’ [2013] PL 468, 476. 142 Some public authorities already require contractors to opt-in to fulsome access obligations at the point of contract: see House of Commons Justice Committee, ‘Post-Legislative Scrutiny of the Freedom of Information Act 2000’, First Report of Session 2012–13, HC 96 (London, Stationery Office, 2012) [237]; G Millar and A Scott, Newsgathering: Law, Regulation and the Public Interest (Oxford, Oxford University Press, 2016) [9.13]. An expansion of FoI of this nature has recently been recommended by the UK Information Commissioner’s Office – see, Outsourcing Oversight? The Case for Reforming Access to Information Law, Report of the Information Commissioner to Parliament (London, Stationery Office, 2019). 143 In 2015, for example, the Metropolitan Police was criticised for issuing generic refusals to Press Gazette requests for information on the eminently important subject of police surveillance of journalists on the basis the requests were ‘intended to be annoying or disruptive’ – see Millar and Scott (n 142) [9.66]. See also Cabinet Office v Ashton [2018] UKUT 208 (AAC). 144 See further M Amos, ‘The Positive Right to Freedom of Expression and Party Anonymity in Legal Proceedings’, this volume, ch 8. 145 See further J Townend, ‘Positive Free Speech and Public Access to Courts’, this volume, ch 9. 146 See further A Scott, ‘Hiding the Truth in the Shadow of the Law? Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts’, this volume, ch 10.

64  Andrew Scott and Abbey Burke conceivably see a court reject any attempt to rely on technical legal arguments, for instance based on Crown copyright, to prevent the publication of information contained in leaked public documents. As noted above, it has been used already to compel a public authority to release information outside of the scheme of the Freedom of Information Act 2000 in the absence of a countervailing reason not to disclose. One reservation about the principle of access concerns the question of how far it runs. Statutory renditions of the principle, reflected in specific and general legislation, specify more or less clearly the circumstances in which information will be made available under the schemes they provide, and conversely when it will not. Concerns about the disadvantages of openness, codified in such statutory exemptions, are also evident in the competing directions in which Kennedy and its progeny have pulled. Broad principles developed by judges in the abstract are almost bound to be under-specified in this respect. Statements of principle are welcome, but ultimately the hard questions are asked at the limits of transparency: the points at which the logics underpinning claims to secrecy, confidentiality or privacy become irresistible. It is in understanding that tension that difficult choices must be made. As Lord Toulson once noted in a related context, ‘fine words butter no parsnips’.147



147 Guardian

News and Media (n 82) [1].

5 Promoting Civic Discourse: A Form of Positive Free Speech under the Constitution of Ireland? EOIN CAROLAN

Given the enduring and international influence of the ‘almost entirely negative’ conception of free speech found in much US case law and commentary,1 the simple recognition that positive free speech is possible can, of itself, have merit. Defending the idea of government intervention to promote or protect expression is, however, only a first step in fleshing out an alternative approach. Questions naturally arise about the purpose, scope and permissible extent of government action. The answers to these questions will, in turn, have implications for citizens, for public bodies and for media actors, as well as for issues around plurality, diversity of content or of voices, access to information and media regulation generally. Many of these difficult conceptual questions are considered elsewhere in this volume. This chapter is, by comparison, much more limited in its aims and ambitions. In charting recent developments in Irish freedom of expression jurisprudence, its primary purpose is, in all probability, to introduce the material to readers to whom it may be unfamiliar. The chapter also, however, aims to provide a critical overview of both the potential and the pitfalls of an approach under which the courts have begun to develop a view of freedom of expression capable of being understood as a positive constitutional commitment to civic discourse. That is not to suggest that the Irish experience provides the only or even an optimal approach to positive free speech. In fact, as the chapter suggests, this emerging approach may reflect quite specific aspects of Ireland’s republican constitutional traditions. Even allowing for this particularity, however, these decisions may have some utility as an example of a constitutional system that has begun to grapple with the practical challenges of a free speech guarantee with potentially positive dimensions. In this regard, the chapter focuses in particular on the implications this may have for the regulation of media organisations which, partly because of an express constitutional acknowledgement of their pedagogical role, have been held to be vested not only with rights, but also with certain constitutional responsibilities.



1 F

Schaeur, ‘Hohfeld’s First Amendment’ (2008) 76 George Washington Law Review 914, 917–18.

66  Eoin Carolan

I.  The Emergence of a Binary Model of Expression in Ireland A.  Freedom of Expression in the Irish Constitution Any discussion of Irish law on freedom of expression naturally begins with the text of the 1937 Constitution of Ireland (the Constitution), and, in particular, with the specific, if heavily qualified, guarantee set out in Article 40.6.1: The State guarantees liberty for the exercise of the following rights, subject to public order and morality: – i The right of the citizens to express freely their convictions and opinions. The education of public opinion being, however, a matter of such grave import to the common good, the State shall endeavour to ensure that organs of public opinion, such as the radio, the press, the cinema, while preserving their rightful liberty of expression, including criticism of Government policy, shall not be used to undermine public order or morality or the authority of the State. The publication or utterance of blasphemous, seditious, or indecent matter is an offence which shall be punishable in accordance with law.

The article, which censoriously seems to bear the influence of Catholic social teaching of the era, provides for substantial restrictions on media expression. The general evolution of freedom of expression jurisprudence in Ireland has been well covered elsewhere.2 This chapter is concerned with a specific subplot to that story: namely, how the articulation of a binary model of freedom of expression facilitated the development of a more positive conception of expression under Article 40.6.1. The first stage in this process was the identification of an alternative rubric for expressive claims under Article 40.3.1’s protection of the unspecified ‘personal rights of the citizen’. In AG v Paperlink,3 Costello J identified a right to communicate as a separate constitutional entitlement for the reason that ‘the act of communication is the exercise of such a basic human faculty that a right to communicate must inhere in the citizen by virtue of his human personality and must be guaranteed by the Constitution’.4

B.  The Relationship between Expression and Communication For some time, the precise scope and significance of the right to communicate was obscured by the circumstances in which it had first been identified. Paperlink involved a challenge to the monopoly of the Post Office. This meant that the court was obliged, as a preliminary issue, to determine whether the physical communication of information by means of a postal service had a sufficient constitutional dimension to sustain the initial claim of an interference with protected rights. This, together with the reference to ‘convictions and

2 See, eg GW Hogan and GF Whyte, JM Kelly: The Irish Constitution, 4th edn (Haywards Heath, Tottel, 2003) 1726–79. 3 [1984] ILRM 373. 4 ibid 383 (Costello J).

Promoting Civic Discourse  67 opinions’ in Article 40.6.1, led to a concentration in subsequent cases on distinguishing expressive claims on the basis of whether they involved information or opinions.5 This distinction was problematic for a number of reasons. First of all, it led to considerable and arguably pointless complications in the not uncommon instances where the communication at issue involved a mixture of fact and opinion.6 The practical difficulties of treating parts of a single communication under different constitutional and conceptual regimes are obvious. Secondly, and for similar reasons, the existence of a range of expression that might reasonably be defined as either fact or opinion underlined the operational artificiality of the distinction. Thirdly, while Article 40.6.1 does refer in its opening paragraph to the expression of convictions or opinions, the remainder of the text does not necessarily support the scope of the guarantee being so confined. That point had particular force with regard to the references to media organisations. An interpretation of Article 40.6.1 that protected the media when expressing opinions but not when conveying facts seems both impractical and implausible, especially in light of the expressly educative character of its constitutional role.

C.  A New Model? Dividing Expression into Personal and Public Spheres A recasting of the relationship between Article 40.3 and Article 40.6.1(i) began with the decision of the Supreme Court in Irish Times v Ireland.7 The case concerned a successful challenge by media organisations to a court order prohibiting contemporaneous reporting of a criminal trial. Notably, the Supreme Court’s decision strongly emphasised the public dimension to the Article 40.6.1 entitlement. The court rejected the argument that, because the court remained open to members of the public to attend and witness the trial, the order made did not interfere with the public administration of justice. Rather, the court recognised a specific constitutional position for the media as ‘the eyes and ears of the public’ on whom ‘the ordinary citizen is almost entirely dependent … for his knowledge of what goes on in court’.8 Although this was sufficient to dispose of the case on other grounds, one judge (Barrington J) went on to address the distinction drawn in previous cases between the communication of facts under Article 40.3.1 and the expression of opinion under Article 40.6.1(i). In rejecting that approach, he outlined an alternative account of the relationship between the two articles: It appears to me also that it is important to look at the context in which the right of the citizens to express freely their convictions and opinions is placed in the Constitution. Article 40.6.1 deals with three rights, the right of the citizens to express freely their convictions and opinions, the

5 See, eg Kearney v Minister for Justice [1986] IR 116; Oblique Financial Services Ltd v The Promise Production Co Ltd [1994] 1 ILRM 74; Carrigaline Community Television Broadcasting Co Ltd v Minister for Transport, Energy and Communications (no 2) [1997] 1 ILRM 241; SPUC v Grogan (no 5) [1998] 4 IR 343. 6 See, eg the arguments about whether memoirs should be regarded as containing fact or opinion in AG for England and Wales v Brandon Books Publishers Ltd [1986] IR 597. 7 [1998] 1 IR 359. 8 ibid 409 (Keane J).

68  Eoin Carolan right of the citizens to assemble peaceably and without arms and the right of the citizens to form associations and unions. All of these relate to the public activities of the citizens and to the practical workings of a democratic society. They are part of the dynamics of political change.9

The issue was considered again by the court in Murphy v IRTC.10 This case (which was later the subject of an application to the European Court of Human Rights in Murphy v Ireland11) involved a challenge to section 10(3) of the Radio and Television Act, 1988, which prohibited advertising directed towards ‘any religious or political end or which has any relation to an industrial dispute’. Delivering on this occasion the unanimous decision of the court, Barrington J returned to the theme of his earlier judgment in Irish Times v Ireland: It appears to the Court that the right to communicate [under Article 40.3] must be one of the most basic rights of man. Next to the right to nurture it is hard to imagine any right more important to man’s survival. But in this context one is speaking of a right to convey one’s needs and emotions by words or gestures as well as by rational discourse. Article 40.6.1 deals with a different though related matter. It is concerned with the public activities of the citizen in a democratic society. That is why, the Court suggests, the framers of the Constitution grouped the right to freedom of expression, the right to free assembly and the right to form associations and unions in the one sub-section. All three rights relate to the practical running of a democratic society … If this means that there is a certain overlapping between the right to communicate impliedly protected by Article 40.3 and the right of the citizens freely to express their convictions and opinions guaranteed by Article 40.6.1, so be it. The overlap may result from the different philosophical systems from which the two rights derive.12

What is noteworthy about this rationalisation of the binary model is the extent to which it, seemingly deliberately, tracks the distinction between deontological and consequentialist theories of freedom of expression. The right to communicate under Article 40.3 is described in strongly individualistic terms which emphasise the relationship between expression and the intrinsic personality of individuals. Furthermore, the existence of the right is specifically disconnected from its content. It is valued for what it is rather than what it involves. This is an approach which lends itself quite naturally to the non-censorship principle at the core of many negative philosophies of free speech. By contrast, expression under Article 40.6.1 is concerned with the public activities of the citizen. This owes more to consequentialist accounts of expression, particularly those associated with the promotion of democracy rather than the pursuit of truth. The characterisation of expression in the public sphere as a consequentialist value carries with it important implications. If expression is valuable because it promotes certain consequences, then it may be susceptible to constraints where it does not do so. Similarly, if the structural conditions for expression are deficient, it may – in principle – be justifiable for the state to intervene to improve them. The decisions in Irish Times and Murphy thus provided a theoretical template for a positive approach to regulating expression in the public sphere. Locating deontological notions of expression as self-actualisation in a separate article created the possibility that these claims could be dealt with under separate regimes, each of which took appropriate account of their normatively distinct character.

9 ibid

404. 1 IR 12. 11 (2003) 38 EHRR 212. 12 Murphy (n 10) 24–25. 10 [1999]

Promoting Civic Discourse  69 The identification of public expression as a democratic value accordingly provided a starting point for a clearer and more rigorously conceived approach to expressive claims. However, substantial ambiguities remained. How did expression promote democracy? Whose interests were primarily engaged by attempts to restrain public expression? What was meant by democracy itself? And what, if any, implications did this have for media expression?

II.  The Implications of a Binary Model of Constitutional Expression A.  Consequentialism in Ireland: Expression as a Democratic Value The claim that expression serves democracy is a common one which, in the modern era, was arguably first associated with the work of Alexander Meiklejohn in the USA.13 There are, however, various versions of this theory, which differ in the emphasis they attach to distinct aspects of the democratic process: Some suggest that free speech facilitates the informed decision-making that self-rule requires. Others argue that free speech furthers democracy by allowing individuals to recognize themselves as self-governing. Still others simply conclude, without elaboration, that democracy would be ‘meaningless’ without the freedom to discuss government and its policies.14

Clarifying the specific democratic function which expression is understood to perform is important as this naturally shapes the system’s conception of the type of expression which it ought to protect. Some focus on the value of voters exercising an informed choice at elections. This variant logically prioritises the education of voters by the provision of accurate information about issues relevant to elections. This can cover factual information about government conduct or more opinionated commentary on political policies or ideas. However, it tends to derive the constitutional protection of expression from its political or governmental content. This election-oriented model of expression also tends to focus on the interests of the audience rather than those of the speaker. This can generate complications when there is a divergence between the interests of the speaker in expression and the interest of the audience in having access to that expression. A model of expression which envisages the interests of the audience as predominant is at risk of lapsing into a repressive popular majoritarianism, where the opinions of the majority on particular content become elided with an assessment of that expression’s social or constitutional value. It may also, again, be more consistent with

13 A Meiklejohn, Free Speech and its Relation to Self-Government (New York, Harper, 1948). 14 MH Redish and AM Mollen, ‘Understanding Post’s and Meiklejohn’s Mistakes: The Central Role of ­Adversary Democracy in the Theory of Free Expression’ (2009) 103 Northwestern University Law Review 1303, 1304, citing, respectively: Meiklejohn (n 13); RC Post, Constitutional Domains: Democracy, Community, M ­ anagement (Cambridge, MA, Harvard University Press, 1995); and D Held, Models of Democracy (Stanford, Stanford University Press, 1987).

70  Eoin Carolan a negative conception of expression under which the role of the constitution is primarily one of prohibiting any regulation of political speech. This demonstrates how the courts’ explanation of the democratic value of expression has implications for the nature, scope and level of the protection provided in practice by the legal or constitutional guarantee. From that point of view, it is significant that the courts in Ireland appear to have developed an understanding of expression as a democratic value which is premised not on the content of the expression, but on its value as expression of a particular type. Furthermore, the value of that expression seems to be associated not with a narrow focus on the democratic process,15 but with a broader understanding of public expression as a civic value which relates to the individual’s involvement with his or her community. As an approach which is primarily concerned with the conditions in which expression occurs, it is also one which is more clearly connected with positive ideas of free speech. This civic conception of expression finds support in aspects of Barrington J’s judgment in Murphy v IRTC. The judge’s explanation of the basis for both rights is expressed in terms of the interests and objectives of the speaker rather than his or her audience. The audience gives purpose and focus to the speaker’s expression, but it remains an entitlement which emanates from the speaker’s role. This Article 40.6.1 variant is connected with democracy, but it is an understanding of democracy which is – in the court’s articulation – oriented in the first instance towards the activities of the individual qua citizen rather than the more general secondary social or institutional interests which are also associated with democracy.

B.  A Republican Model of Expression? It is notable that this emphasis on the citizen’s active participation in public life, broadly conceived, is consistent with the republican strand of Irish constitutional thought.16 This envisages the citizen as an engaged member of the country’s civic life, as an observer of and contributor to the ‘contestatory citizenry’ that is central to the republican polity.17 What this republican concern for civic action implies in this context is that the claim to freedom of expression derives from, and is defined by, the interests of the citizen rather than those of the collective. This was confirmed by the court’s rejection in Murphy of the argument that Article 40.6.1 could only be invoked in relation to public expression in its stricter sense. The Article 40.6.1 entitlement to expression as part of a citizen’s participation in public life was not constrained by a notion of ‘democracy-as-elections’, which could, in principle, limit the right to either expression aimed at political policy or expression directed to the public as a whole. The distinction drawn did not mean that ‘to invoke the

15 This seems broader than the understanding of civic republicanism as direct participation in the electoral process that appears to underpin the account in WJ Rinner, ‘Maximizing Participation through Campaign Finance Regulation: A Cap and Trade Mechanism for Political Money’ (2010) 119 Yale Law Journal 1060. 16 See I Honohan (ed), Republicanism in Ireland: Confronting Theories and Traditions (Manchester, Manchester University Press, 2008). 17 Philip Pettit regards this as one of the three core ideas of traditional republican thought: see P Pettit, On the People’s Terms: A Republican Theory and Model of Democracy (Cambridge, Cambridge University Press, 2012) 5–8.

Promoting Civic Discourse  71 protection of Article 40.6.1 a person must be attempting to influence the citizens at large’.18 This suggests a much broader conception of what constitutes public-facing expression under Article 40.6.1 than might initially have been understood. It is, however, worth also briefly considering what the apparent influence of r­ epublican theory might mean in terms of the state’s role in promoting or protecting expression. One of the key distinctions between republican and liberal conceptions of the state is that classical republicanism views positive action in defence of freedom to be a legitimate, if not the primary, purpose of the state. This is related to the fact that the republican notion of freedom as ‘freedom from domination’ requires more than egalitarian neutrality or liberal non-intervention. The fact that a person is subject to the potential dominion of another, whether exercised or not, is sufficient to justify action being taken to address that imbalance of power. Crucially, that is so whether the imbalance occurs between a public body and a citizen or between private actors. The basic notion of a positive obligation to promote a constitutional value is, accordingly, entirely consistent with this republican tradition. In the context of freedom of expression, that means that the Constitution is concerned with matters beyond the individual’s moment of intended expression. Freedom of expression is also engaged by the individual’s freedom to choose from a range of possible ideas.19 One consequence of this is that the state’s obligations are not limited to ensuring a citizen can speak, but may also include the creation of conditions in which the citizen’s freedom to engage with information and ideas – with civic discourse – is not susceptible to external control. While this bears some similarities to Post’s reference to an ‘uncoerced domain of public discourse’, it would share Kenyon’s scepticism about the related assumption that ‘public speech achieves something like an ideal form, or comes closer to doing so, when the state is “absent”’.20 Nor would it necessarily be satisfied by a ‘diversity of voices’ approach if the choice and number of voices remained subject to third party control. Rather, there is a role for the state in securing or promoting conditions in which access to information or ideas is not dependent upon the exercise of another’s will. This is not, of course, a guarantee of access to all possible ideas in all possible worlds. It is instead an approach that is concerned with ensuring the integrity and freedom of the processes by which civic discourse occurs: a concern in which the potential dominion of media organisations naturally looms large.

C.  Expression as Discourse: The Role of Citizens? The adoption of a discourse-oriented understanding of Article 40.6.1 has had i­mportant implications for the constitutional position of media organisations. Most obviously, if the instrumental goal is the promotion of the individual’s ability to participate in public discourse, media expression rights seem a somewhat awkward fit, especially when their actions inevitably privilege some speakers over others.

18 Murphy (n 10) 25. 19 This idea of freedom is also linked with certain conceptions of autonomy (and privacy) as an entitlement to a level of psychological or ethical freedom in which the individual is equipped to make the choices they would make in the exercise of their true will. For this argument in more detail, see H Delany and E Carolan, The Right to Privacy: A Doctrinal and Comparative Analysis (Dublin, Round Hall, 2007) ch 1. 20 AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379, 385.

72  Eoin Carolan This is particularly relevant to contemporary debates about the boundaries between traditional journalism and newer, less institutionally structured forms of public expression. The fact that the constitutional distinction is based on the civic or public character of the expression rather than on the identity of the speaker allows for the extension of privileges associated with the press to non-traditional participants in public discourse. This was confirmed by the willingness of the High Court in Cornec v Morrice to allow a blogger to claim confidentiality over the identity of his sources.21 In Cornec, an application was made to require persons to give evidence about an alleged religious cult which was the subject of legal proceedings in another jurisdiction. One of these persons was a blogger who was, it was suggested in evidence, a regular and expert commentator on religious cults. Hogan J adopted a strongly consequentialist analysis of Article 40.6.1 as intended to facilitate public discourse, citing an earlier judgment of his in a case about the accuracy of the information provided to the public during a referendum campaign.22 In that earlier case, Hogan J had underlined the civic nature of Ireland’s democracy, specifically highlighting the value of freedom of expression as a means of providing the active and informed citizenry anticipated by the Constitution: The Constitution envisaged a plebiscitary as well as a parliamentary democracy … it envisaged a society in which all citizens would be called upon from time to time to make critical ­decisions regarding their future, the future of their neighbourhood and, ultimately, the future of their country. It is [necessarily] implicit in this [that the] Constitution thus places a premium on honest and fearless debate. The drafters of the Constitution must have understood that an inert, supine and indifferent public posed the greatest threat to the public welfare, since a plebiscitary democracy will simply not function under such circumstances. The Constitution, therefore, calls, especially at a time of referendum, for robust political debate from an informed public. This is, in many respects, also reflected in Article 9.3 of the Constitution, which speaks of the duty which all of us as citizens owe, namely, fidelity to the Nation and loyalty to the State. It calls upon each of us to perform that duty for ourselves, for our neighbours and for our country, to inform ourselves and to make our own decision, informed as best as we can be, on difficult issues to which there is often no easy answer. It is not easy, but we must all try. Article 9.3 of the Constitution accordingly places considerable emphasis on individual civic responsibility of each citizen.23

In applying those principles in Cornec, the High Court accepted that the blogger, by making a distinct contribution to civic discourse, acted ‘squarely within the “education of public opinion” envisaged by Article 40.6.1’, and was therefore entitled to invoke its protection.24 This emphasises once again that the distinction drawn in Murphy is based on the value of the expression itself rather than on the identity of the speaker, his or her intention in engaging in such expression, or the specific content of the expression. It also highlights that the democratic value of expression under this civic conception is not linked to elections, referenda or even government. It is sufficient for the speaker to be seeking to engage with others on an issue of broader public value or interest.



21 [2012]

1 IR 804. v Referendum Commission [2012] 2 IR 596. 23 ibid 604. 24 Cornec (n 21) 825–26. 22 Doherty

Promoting Civic Discourse  73

D.  Expression as Discourse: Content-Based Constraints? The corollary of this discourse-oriented consequentialism is that there is less scope for content-based restrictions than might be the case under other consequentialist models. Where expression is justified as a means of pursuing truth, for example, or of informing voters on matters of political concern, it may logically be permissible to proscribe the making of false or inaccurate statements, or to impose greater limitations on non-political expression. Where the value of expression is the discourse it promotes, however, this comes closer to a situation in which the act of expression itself supplies both the means of achievement and the end to be achieved. Some support for this interpretation can be found in the decision of the Irish Supreme Court in Mahon v Post Publications.25 The case involved an application by a judge-led Tribunal of inquiry for an injunction restraining a newspaper from publishing information which had been circulated to certain individuals on a confidential basis. What is relevant for the purposes of this discussion was the argument advanced that the newspaper could not claim protection for material which it could not – in the Tribunal’s view – prove to be in the public interest. This argument clearly sought to rely on the broadly consequentialist character of the post-Murphy case law to confine the scope of the Constitution’s protection to public expression which could be positively shown to serve a publicly valuable purpose. The Tribunal argued that, in this instance, the disclosure of confidential information was not in the public interest but was aimed solely at increasing circulation figures. The suggestion that the absence of such an established consequentialist value deprived the newspaper of the entitlement to publish the material was firmly rejected by the Supreme Court. Fennelly J was clear that, while freedom of expression could be subject to certain constraints, there was no a priori obligation on a publisher to prove the value of an individual piece as a prerequisite to publication. He wrote: The courts do not pass judgment on whether any particular exercise of the right of freedom of expression is in the public interest. The media are not required to justify publication by reference to any public interest other than that of freedom of expression itself. They are free to publish material which is not in the public interest … The right of freedom of expression extends the same protection to worthless, prurient and meretricious publication as it does to worthy, serious and socially valuable works. The undoubted fact that news media frequently and implausibly invoke the public interest to cloak worthless and even offensive material does not affect the principle.26

While some of Fennelly J’s language came close to a deontological conception of expression, the context of the discussion and the remainder of the decision indicated that the judgment instead involved a particular form of consequentialist analysis. The court went on to make clear that categories of expression could be restricted where that was legitimately done by common law or statute. Mahon accordingly involved the application of the Murphy analysis that certain types of expression have social value and an associated constitutional status in a categorical rather than situation-specific manner. This is consistent with the emphasis in



25 [2007] 26 ibid

3 IR 338. 374–75.

74  Eoin Carolan Article 40.6.1 and in Murphy on promoting civic participation and public discourse. This category-consequentialism allows for the possibility of constraints on civic expression – some of which could, in principle, be content-based – but avoids the imposition of a priori burdens on individual participants to consider the objective or demonstrable value in their intended expression. This reflects the fact that the model adopted in Murphy is focused on the value of civic discourse itself, rather than on the content of the information or ideas discussed. The primary value of discourse under the Murphy model is that it encourages active engagement in civic issues. Through discourse, the individual acquires not only the information, but also the attitudes, qualities, relationships and skills that facilitate active citizenship. Thus, even where the expression involves information of limited or questionable value, the fact that engagement in civic discourse has occurred can have value both for the individual and for society as a whole. Expression, even if for less than noble purposes, can encourage the active and mutual engagement of citizens on issues of common interest or concern that, in turn, fosters a participatory civic culture.

E.  Expression as Discourse: Can it be Constrained at All? This does not mean that the model would not, in principle, permit the limitation of expression rights in some situations. The difference with the discourse model is that limitations may be more likely to be justified by an assessment of the impact of the expression on discourse rather than on the content of the expression itself. As Clarke J acknowledged in Cogley v RTE: The form of parliamentary democracy enshrined in the Constitution requires that there be a vigorous and informed public debate on issues of importance. Any measures which would impose an excessive or unreasonable interference with the conditions necessary for such debate would require very substantial justification.27

The highlighted section seems to envisage a system focused not on the content of expression, but on its potential impact on the ability of others to participate in public discourse. This would support, for example, restrictions on expression which might adversely affect the ability or willingness of others to engage in expression or debate. This approach could, in principle, cover several familiar forms of regulation of expression. Rules prohibiting hate speech, for example, could be justified on this analysis where the expression in question could be adjudged likely to deprecate civic discourse by discouraging or devaluing the expression of others. Rules prohibiting harassment or other forms of threatening conduct would also fit within a model predicated on promoting civic discourse. Likewise, the connection between autonomy, identity and expression means that this model could, in principle, accommodate laws on defamation or privacy. The threat of damage to a person’s good name may cause them to withdraw from discourse, just as the causing of such damage can adversely – and, if false, unfairly – affect how others engage with them in civic discourse. In a similar vein, privacy laws might be justified as supportive of the conditions for uninhibited individual expression.28



27 [2005]

IR 79, 94 (emphasis added). v Vopper, 532 US 514 (2001); LSN v Persons Unknown [2010] EWHC 119 (QB).

28 Bartnicki

Promoting Civic Discourse  75 Nor would any permissible regulation necessarily be confined to conduct which impacts upon the willingness of individuals to speak. It could also apply to situations where one form of speech may affect the capacity of another to form or express views by, for example, occupying a dominant position in the public discourse or otherwise obstructing access to information or ideas. This could potentially provide a basis for regulating political advertising or donations.29 Thus, the discourse model does provide for the possibility of prohibiting expression in certain circumstances. Unlike other consequentialist models, it does not value or evaluate expression by reference to the quality of its content. Rather, it logically permits expression to be constrained on the basis of its impact on civic discourse and, in particular, on the expression of others. Promoting civic engagement and expression is the core value and objective of protecting freedom of expression in the public sphere. How, if at all, this tallies with the actions, attitudes and existence of the traditional press is the question considered in the next sub-section.

F.  Expression as Discourse: The Limitations of Civic Value? In its emphasis on the value of civic participation, it might be thought that this approach approximates to the deliberative, republican or communitarian theories of expression that have been advanced in the American literature on the First Amendment. Fiss,30 Post31 and Sunstein32 have all put forward theories which – while differing in the details of how they would distinguish forms of expression – locate its value in the way in which it connects the individual with the collective self-rule of the democratic polity. These models have been criticised, however, as imprecise and under-inclusive,33 subjectively biased34 and stipulating an unduly stylised form of rational discourse.35 Critics have contended that these theories err by substituting the author’s own, often ambiguous, preferences for particular types of speech for the clear, decisive and comparatively neutral approach mandated by the free market of ideas model. By contrast, the Irish courts’ approach to ‘expression-as-discourse’ has avoided at least some of these concerns. In particular, the Irish jurisprudence has thus far eschewed an approach which would define (or exclude) expression as publicly valuable by reference to the identity of the speaker or to their intention in engaging in such expression (as would, for example, Post36 or Sunstein37). An approach which treats discourse as a valuable act in itself 29 M Youn, ‘First Amendment Fault Lines and the Citizens United Decision’ (2011) 5 Harvard Law and Policy Review 135. 30 OM Fiss, ‘Free Speech and Social Structure’ (1986) 71 Iowa Law Review 1405. 31 RC Post, ‘The Constitutional Concept of Public Discourse: Outrageous Opinion, Democratic Deliberation, and Hustler Magazine v Falwell’ (1990) 103 Harvard Law Review 601; ‘Meiklejohn’s Mistake: Individual Autonomy and the Reform of Public Discourse’ (1993) 64 Colorado Law Review 1109. 32 CR Sunstein, Democracy and the Problem of Free Speech (New York, Free Press, 1993). 33 Redish and Mollen (n 14). 34 BK Lemley, ‘Effectuating Censorship: Civic Republicanism and the Secondary Effects Doctrine’ (2002) 35 John Marshall Law Review 189. 35 RH Fallon Jr, ‘Post on Public Discourse under the First Amendment’ (1990) 103 Harvard Law Review 1738. 36 Post’s model would consider expression as public if it involved ‘a protected effort to shape public opinion’: see R Post, ‘The Constitutional Status of Commercial Speech’ (2000) 48 UCLA Law Review 1, 17. 37 Sunstein defines expression as political on the basis that it ‘is both intended and received as a contribution to public deliberation about some issue’: (n 32) 103.

76  Eoin Carolan allows for a broader conception of public value. In particular, it allows for the possibility of taking into account both speaker and listener interests in the process of civic engagement. It is instructive that Hogan J’s discussion of civic duty in Doherty, as cited in Cornec, is not limited to the speaker who engages in ‘honest and fearless debate’ – it includes also the listener’s duty to become informed and to make his or her own decision. This approach does not therefore necessarily discount the potential value of particular expression to its audience. Nor, however, does it subjugate the interests of the speaker to a majoritarian notion of collective action. Rather, the focus on the activity of discourse as of value to both speaker and listener can be understood as acknowledging the autonomy interests of both parties. This ties in with the republican requirement that the free individual has choices available to him or her, and has the information and opportunity necessary to make that choice in an authentic and autonomous manner. Discourse may be valuable as a means of allowing the speaker to engage with other speakers in a robust form of direct civic engagement – but it also has value for the silent spectator who takes seriously their obligation to act as an autonomous participant in collective self-rule. This is a much broader and more robust conception of publicly valuable expression than one which is focused on content, context or speaker intention alone. It potentially includes categories or forms of expression which may not conform to ideal types of rational debate but which nonetheless have value as forms of civic engagement or education. Satire, vulgar comment or popular entertainment that deals with social issues could all potentially qualify as valuable discourse. Discourse which may not directly relate to the electoral process may nonetheless attract some degree of protection,38 thus reflecting the fact that forms of intercitizen engagement or learning may occur (and arguably more often occur) outside strictly electoral discussions. Furthermore, the fact that the model continues to treat the interests of the individual as critical, even under the more public-oriented rubric of Article 40.6.1, reduces the scope for the sort of communitarian-inspired sanctions of individual conduct feared by Redish and Lippmann.39

G.  Expression as Discourse: What is the Role of the Media? What does this mean for the role and status of the media in Irish society? Are they entitled to protection as one voice among many? Might they be privileged for their specific contribution to civic discourse? Or does a focus on protecting the interests of the citizen as speaker in fact indicate scope for additional restrictions on media speech? The outcome in Mahon highlighted some of the positive implications of the ‘expressionas-discourse’ approach for media organisations. If public-facing discourse is itself valuable, this effectively establishes a presumption in favour of its protection, regardless of the content or subject matter of the expression itself. More directly, the Supreme Court’s repeated references in Irish Times v Ireland to the media’s value as the ‘eyes and ears’ of the

38 See J Weinstein, ‘Participatory Democracy as the Central Value of American Free Speech Doctrine’ (2011) 97 Virginia Law Review 491. 39 MH Redish and G Lippman, ‘Freedom of Expression and the Civic Republican Revival in Constitutional Theory: The Ominous Implications’ (1991) 79 California Law Review 267.

Promoting Civic Discourse  77 everyday citizen implies that, insofar as Irish law is concerned, the media retains a particular constitutional position as part of the structural conditions which support and promote civic discourse. This was reiterated by the High Court in the more recent decision in Kivlehan v RTE.40 In that case, Baker J – in undertaking a judicial review of the public broadcaster’s policy on who to invite to a leaders’ debate during an election campaign – stated that: I regard the provisions of Article 40.6.1(i) as importing a positive obligation on a public service broadcaster to produce programmes and to inform and educate. I consider that the framers of the Constitution, in recognising the unique and central role of the media, recognised both its duty and any rights that flow therefrom.41

This again acknowledges a specific role – and duty – under the constitution for media expression as a catalyst for discourse by drawing attention to a range of information or arguments that may be capable of prompting further debate. It can also offer an outlet for others to engage in or encourage public debate. Article 40.6.1’s reference to the media as educators and organs of public opinion can be read to support both these pedagogical and participatory roles. Thus, while the post-Murphy focus on the speaker’s interest may not justify the protection of the press as primary rights holders, it does allow for the recognition of their position as part of the structural conditions that are necessary for the effective operation of that right. However, theirs is a secondary status which is parasitic upon the promotion and support of the speaking rights of others. Irish constitutional law provides specific protection for the media insofar as it fulfils its particular role. What the law also appears to allow for, however, is the drawing of a further distinction between pedagogical and participatory dimensions of media expression. This has not, as yet, been considered in the case law. Nonetheless, this is a dichotomy which may have something to say about the role and future regulation of the modern media environment. Where the Constitution’s core concern is the promotion of civic discourse, the media’s potential contribution is – it would seem – either as an educator or as a conduit. In other words, it serves as a participant or to provide a platform. Some may, of course, be both. This distinction does appear to track relevant differences in how media actors may provide structural support for civic discourse. It may, accordingly, provide a useful template for the treatment – and regulation – of both forms of support. This could have value for traditional media actors not only by acknowledging and accommodating their specific strengths, but also because of the regulatory implications it could have for their ‘new’ media competitors. The image of media as participant is familiar from traditional accounts of freedom of expression. The media’s value as a participant in public discourse can extend beyond educating and informing the public to include challenging orthodoxies, highlighting unpopular or unknown issues, or otherwise encouraging more wide-ranging or multifaceted public debates. The media’s role here is to add value to civic discourse, whether as a source of reliable information, independent views, expert analysis or challenging ideas.42 Support for this view can be found in Kivlehan, where Baker J, in considering why the Constitution

40 [2016] IEHC 88. 41 ibid [59]. 42 AN Folami, ‘Using the Press Clause to Amplify Civic Discourse Beyond Mere Opinion Sharing’ (2013) 85 Temple Law Review 269.

78  Eoin Carolan singled out the media for particular protection, felt that ‘what is singular and unique is found in the editorial process’.43 This is also, arguably, the role which offers traditional media organisations the best prospect of success (or survival) in the face of fragmentation. The proliferation of alternative – and often faster – sources of information means that there is diminishing value in the traditional media’s function as the provider of information. Where they may continue to serve a purpose is in the added value they can bring to that information, whether by verifying, contextualising or further investigating it. An analysis of the media as a participant in structuring civic discourse neatly captures this type of expression, thereby supplying a workable template for its constitutional protection. Protection under this model would be available to organisations with a demonstrable expertise in, or commitment to, this kind of added-value activity. This finds echoes in the approach adopted in various contexts or countries: from the efforts in Reynolds to protect ‘responsible journalism’44 to the incentivist approach to press regulation adopted in Ireland45 and more recently advocated by the New Zealand Law Reform Commission.46 The idea that the assumption of additional responsibilities, whether in terms of neutrality, fact-checking, expertise or robust procedures, can be rewarded by the conferral of additional protections is one which holds out the possibility of a special (albeit likely a niche) position for certain media organisations. Critically, the nature of the incentives seems linked to the development or maintenance of qualities which are most likely to set traditional media apart from other forms of user-generated or digital discourse. That is not to suggest that the statement that a media organisation has committed to a particular code is likely to have any impact on sales. The point is rather that, insofar as the media may seek to establish a point of distinction between itself and other forms of expression, it seems most likely to succeed by building its reputation and brand on its less replicable resources and qualities: experience, expertise, ­independence, scale, consistency of procedures and so on. The market for media ­organisations which emphasise those qualities might be smaller than previously, but it is – at least – a market in which they retain a competitive advantage. The attraction of a model which recognises a specific role for organisations which promote discourse-enhancing values is thus not that it will, by virtue of that recognition, transform those organisations’ commercial fortunes; rather, it is that it is capable of providing an answer to the questions of whether there is any future for the press by supporting the sort of conduct which could provide a future for at least some traditional media actors. It may also be consistent with the distinct argument that this is an activity which warrants public funding.47 Another notable consequence of the discourse model may be in its implications for media organisations that fulfil the alternative function as platforms for civic discourse. This is the type of activity which might today be more closely associated with online publishers who provide a mechanism for the dissemination of individual views, or with sites which collect and aggregate material from other places online. Typically, these organisations

43 Kivlehan (n 40) [59]. 44 [2001] 2 AC 127, [1999] 3 WLR1010. 45 E Nagle, ‘Keeping Its Own Counsel? The Irish Press Council, Self-Regulation and Media Freedom’ (2009) 20 Entertainment Law Review 93. 46 New Zealand Law Commission, The News Media Meets ‘New Media’: Rights, Responsibilities and Regulation in the Digital Age, Report 128 (Wellington, NZLC, 2013). 47 See the discussion in Kenyon (n 20) 405.

Promoting Civic Discourse  79 have adopted (and enjoyed) a hands-off approach to regulation on the basis that they exist primarily as a platform for the expression of third parties. The discourse model raises issues, however, for this laissez-faire approach. After all, if traditional media have the potential to constrain or inhibit discourse, so, too, do the more dominant or influential online publishers, aggregators or curators. The tendency of Google to direct online traffic either to or away from particular sites will, for example, often be the single most influential determinant of whether a site attracts wider public attention or engagement. The ease with which online platforms like Twitter can be used aggressively to hound those holding unpopular positions can have a detrimental effect on such individuals’ willingness to engage in or contribute to further discourse. Meanwhile, the more recent success of Facebook Instant Articles means that that site will increasingly control the whole environment in which news stories are identified, displayed and accessed. As a platform with a commercial interest in promoting certain responses to news stories (and avoiding others, most notably users leaving the site), a user might be thought more likely to find themselves fed a diet of the familiar rather than information they might regard as challenging or objectionable. Even leaving aside the distinct and more difficult regulatory issues around the capacity of online publishers to use cognitive techniques to promote certain behaviour,48 the existence of such a powerful level of control has serious ramifications for a republican conception of civic discourse. What this all demonstrates is that a focus on discourse may be useful because of the way in which it concentrates attention on what is one of the main challenges and defining features of digital communication: namely, that the relative ease and vast scope for dissemination renders critical – and arguably suspect – the role of those with the power to shape the direction of discourse online. This reflects the reality that platforms are not passive, so they should, on a republican conception at least, be subject to some form of regulation. The difference with the regulation of media-as-participants outlined above is that a consequentialist focus on discourse would, in the context of media-as-platforms, favour regulation which strives to ensure diversity of voice and content, plurality, accessibility49 and – critically – neutrality not only in terms of content, but also in terms of access and distribution. A focus on securing the structural conditions for civic discourse would imply that there may be scope to require greater regulatory standards for those who provide these platforms, and greater regulatory interest in the algorithms and cognitive techniques that increasingly govern them.

III. Conclusion On the whole, Ireland’s emerging approach to freedom of expression shows the potential of positive free speech. From a conceptual point of view, the model’s most attractive feature is probably the way in which it strives to take seriously the logical and practical 48 On which, see E Carolan and A Spina, ‘Behavioural Sciences and EU Data Protection Law: Challenges and Opportunities’ in A Alemanno and A-L Sibony (eds), Nudge and the Law: What Can EU Law Learn from Behavioural Sciences? (Oxford, Hart, 2015) 161. 49 A right of non-discriminatory access to the airwaves has been advocated on broadly similar grounds – see D Matheson, ‘No Moderator Needed: A Liberty Tradition Right to Broadcast Advertorials’ (2006) 33 Hastings Constitutional Law Quarterly 255, 259.

80  Eoin Carolan differences between different forms of expression. Unlike the more monolithic notions of negative free speech associated with the First Amendment, this provides a rationally defensible and workable basis for distinguishing between expressive claims. This provides the clarity and consistency necessary to calibrate the law’s response to freedom of expression litigation. While negative free speech absolutists may criticise the very notion of permitting government to differentiate between different instances of expression, the American experience suggests that some form of classification is, in practice, inevitable.50 If this is correct, it seems preferable for any calibration to be performed on an assessment of the value of the impugned expression rather than by defining expression to exclude whole categories of conduct. The latter approach risks preserving the integrity of the freedom of expression principle at the cost of denying protection to less popular expressive activities. A second advantage of the Irish model is that its design allows a distinction to be drawn between negative and positive approaches to expression. This could address at least some of the concerns some freedom of expression advocates may have with permitting a ­consequentialist balancing of the social benefits and costs of expression by providing a heading under which deontological claims to ‘expression-as-non-censorship’ can be made. In an era of global communications and major media corporations, it is naive to insist on an all-or-nothing approach to expression. There are valid and important differences, both in theory and practice, between the expression of the individual amongst intimates or on a soapbox at Speakers’ Corner and that dictated by a media mogul or wealthy purchaser of political advertising. The binary distinction between personal and public-facing expression makes it possible for the law to develop and apply alternative forms and levels of protection for what are conceptually distinct constitutional values. The adoption of a consequentialist model, however, inevitably introduces points of contention or criticism. First of all, it has the potential – as the litigation in Mahon v Post Publications showed – to generate claims that expression which does not clearly support the identified consequentialist value is presumptively suspect. Secondly, it gives rise to a further question about the nature of values the model aims to promote. There is scope for substantial disagreement about whether expression might be variously justified as supportive of truth, democracy, participation, education and so on. Thirdly, and importantly from the perspective of media organisations, the distinction drawn by Barrington J is one which tends to be made in conceptual contemplation of the expression of the individual citizen. Tying at least some of the Constitution’s protection of expression to deontological ideas of self-actualisation or autonomy emphasises principles and values which it is difficult for the media plausibly to invoke. While there are other legitimate candidates for a consequentialist concept of expression, the emergence of a discourse model has had broadly positive effects. Most fundamentally, the model’s commitment to expression as generally valuable in itself avoids the common consequentialist dangers of unduly narrowing the category of expression protected, or of encouraging content-based limitations on expression which might objectively be regarded as less socially valuable. Acknowledging that the act of expression in itself has social value encourages a presumption in favour of all forms of expression, as the Supreme Court’s decision in Mahon showed.

50 US

v Alvarez, 567 US 709 (2012).

Promoting Civic Discourse  81 The model’s emphasis on civic discourse does, however, reiterate the primacy of the autonomous individual – whether as speaker or listener – under this approach. This has the potential to create uncertainty for the press and other traditional media organisations. The recognition by the Irish courts that non-media actors may invoke privileges that have been traditionally reserved for journalists highlighted how peripheral the media may be to a theory premised on civic discourse. That Cornec conferred the privilege on the very kind of free, enthusiastic and expert online commentator with whom traditional media sometimes struggles to compete underscores the secondary or parasitic importance of the media’s position. These are issues that the Irish courts have not had to determine in recent years. The Supreme Court’s last endorsement of the position of the press as the eyes and ears of the public pre-dated the exponential increase in user-generated information or analysis. Nonetheless, the court’s discussion in Irish Times v Ireland of the media as a bulwark for broader public values (in that case, the public administration of justice) could – if updated in the kind of manner broadly sketched out in Kivlehan – deliver a workable constitutional framework for media expression. The notion of the media as structural support for individual discourse might assist in the articulation and defence of a role for traditional media organisations in the digital era. This is largely because the model’s focus on civic discourse echoes the way in which the social and digital environment has evolved over the last decade. Public discourse – in the sense of engagement with other members of the community beyond an immediate core of personal relationships – has become easier, more accessible and more central to social activity. Conceptualising claims of expression in terms of discourse encourages claimants – including the media – to reflect upon the nature and asserted value of their contributions. This may be less comforting than an approach which simply accepts as self-evident the value and importance of an uninhibited press. What it may allow, however, is the possibility for the press and other traditional media outlets to identify a particular place for themselves within this fragmented and disaggregated system, and to demand the legal and constitutional supports necessary to fulfil that role. It may not return the press to the perceived heyday of a pre-digital age. However, the fact that the model conceives of the media’s contribution to civic discourse in a manner which may, ultimately, coincide with the commercial necessity for a contemporary ‘unique selling point’ suggests that it offers a useful exemplar for a positive approach to free speech.

82

6 The State of Affairs of Freedom: Implications of German Broadcasting Freedom ANDREW T KENYON

I. Introduction1 Free speech is often understood as an individual liberty, as a limitation on government action that would restrict speech. But free speech is not only understood in those terms; it can also be said to impose obligations on states to foster a ‘state of affairs’ that contains plural public speech – or diverse public discourse – as well as the absence of censorship.2 That state of affairs recognises positive dimensions, or structural implications, of free speech. Then free speech involves multiple diverse voices, as well as the absence of prior censorship and (often judicial) scrutiny of legal limitations or penalties on published speech.3 This approach suggests that there are implications for the ‘architecture’ underlying public speech that flow from a free speech principle. Research in human rights law and political theory, as well as legal scholarship on freedom of speech itself, can be used to suggest just that, and here I will largely take the point as given. I address the ideas at length elsewhere.4 The concern here is that, given it is possible to conceptualise free speech as having positive dimensions, what might that mean in practice? How might a state foster such a communicative environment? As this book suggests, there are many possible responses to those questions. This chapter considers Germany’s constitutional protection for speech and its influence on broadcasting law as an illustration of how different arms of a state can have distinct roles from those commonly imagined in English language writing on free speech. The example is considered in part because of the stature and influence of the German Federal Constitutional Court. 1 This chapter draws on AT Kenyon, Democracy of Expression: Positive Free Speech and Law (Cambridge, Cambridge University Press, forthcoming), especially ch 6, examining the German broadcasting decisions, although the analysis there is more detailed. For a briefer overview of some of the same issues considered here, see AT Kenyon, ‘Positive Free Speech: A Democratic Freedom’ in A Stone and F Schauer (eds), Oxford Handbook on Freedom of Speech (Oxford, Oxford University Press, forthcoming). 2 The idea of a ‘state of affairs’ is taken from R Alexy, A Theory of Constitutional Rights, trans J Rivers (Oxford, Oxford University Press, 2002). 3 See, eg J Lichtenberg, ‘Foundations and Limits of Freedom of the Press’ in J Lichtenberg (ed), Democracy and the Mass Media: A Collection of Essays (Cambridge, Cambridge University Press, 1990) 102. 4 Kenyon, Democracy of Expression (n 1).

84  Andrew T Kenyon But the German example also holds interest because of the reasoning used in the decisions. Once the stated reasons for the approach are understood, free speech is seen to hold new possibilities and challenges. At the outset, I want to note one point of debate about the role of courts in relation to free speech’s positive dimensions: should courts play any substantial role about positive free speech or not? On one approach, positive free speech primarily concerns a quality of freedom, whatever constitutional actors seek to foster it. Under that approach, positive dimensions of free speech may be matters primarily for parliaments, executives, regulators or media professionals. If their efforts to promote diverse speech were legally challenged, courts would understand that free speech is not only a negative liberty. For example, courts would allow suitably framed legislation that sought to support plural public speech. This approach to positive free speech is commonly considered or argued for in Commonwealth and US legal literature.5 However, it is also possible to understand positive dimensions of free speech being supported by a legal right. The ‘positive’ here concerns the quality of the legal right, not only the quality of free communication: free speech involves a positive right that can be asserted in a court. While there are reasons to doubt the likelihood of many courts being more active, or the value of them doing so, it is possible for courts to have such a role. The German example illustrates this and, significantly, the court’s reasons for acting suggest that real weaknesses may well arise where courts do not take an active role. Just as importantly, the example suggests aspects of free speech that states should seek to foster whether courts take an active role or not. Thus, the German example suggests points of value for both approaches to ‘­positive’ described above, points about what a positive freedom of communication would be. The analysis also suggests a paradoxical quality to free speech. State actors under democratic constitutions – executives, legislature, courts, independent regulators – are all compromised in their ability to support free speech. Yet structures that support diverse public speech are required by those states’ claims to be democratic, and there are reasons to doubt that the structures can create themselves. That foundational challenge ‘needs to be recognised if free speech is to have substance and be more than a merely formal freedom’.6 The responses that any legal system offers to that challenge are likely to be partial and qualified. Free speech is a state of affairs with no single, simple composition. Thus, the German example is not a blueprint that could simply be adopted elsewhere, and other courts that support positive free speech do so in different ways (the French Conseil constitutionnel offers an example that is broadly similar, but less precise and less demanding).7 The German example, however, suggests aspects of positive free speech that should be addressed elsewhere, whatever actors are most relevant there. English language literature has given some attention to the German broadcasting decisions, but the example is worth considering further. The court’s approach has been

5 Kenyon, Democracy of Expression (n 1) examines many examples; in this volume see, eg J Rowbottom, ‘Positive Protection for Speech and Substantive Political Equality’, ch 3. 6 AT Kenyon, ‘Who, What, Why and How: Questions for Positive Free Speech and Media Systems’ in M Edström, AT Kenyon and E-M Svensson (eds), Blurring the Lines: Market-Driven and Democracy-Driven Freedom of Expression (Gothenburg, Nordicom, 2016) 29. 7 See, eg J Morange, La liberté d’expression (Brussels, Bruylant, 2009) 127. I consider the French example at length in Kenyon, Democracy of Expression (n 1).

The State of Affairs of Freedom  85 sustained, in broadly similar terms, across nearly six decades, and the court has become a particularly powerful actor in German media policy. German public service media ‘has been ruled, at least partially, by decisions of the Federal Constitutional Court throughout its history’ and, despite concerns about ongoing politicisation, the system remains ‘one of the better structured and well managed’ public service media systems in Europe.8 The key point for present purposes is the degree to which the requirements of free speech, as they are understood in these cases, underlie the approach. Free speech is not an individual right alone; it has structural implications that call for certain forms of state action, which can include action by courts. While arising in a particular historical, political and legal context, the court’s reasoning is not overtly limited to that context: the approach can certainly be linked to German constitutional culture, but I do not think it should be restricted to that. Aspects of the approach suggest something more general about free speech in democratic legal systems.

II.  Points of Caution or Contrast As I began by noting, free speech is commonly understood as a restriction on government action that limits or penalises speech. That approach has well-recognised theoretical strengths: an independent judiciary can review and control restrictions on speech favoured by a parliamentary majority (or by the executive, where that is subject to review). While that common approach remains important, it also has evident weaknesses. As has long been recognised, litigation over free speech has little effect on the capacity of ordinary people to communicate publicly.9 Removing explicit government restrictions, while valuable, does not in itself support the ‘voice’ through which people can seek to influence their individual and collective lives.10 The traditional focus on free speech as a negative liberty can flow from a suspicion of state power, with social conflict understood in ‘dyadic’, not collective, terms.11 And courts are commonly seen to face ‘structural constraints on [their] capacity to engage in a significant or coherent redistribution of communicative power’.12 Negative dimensions of the freedom constrain explicit restrictions on speech imposed by government, but that does little to support diversity in public speech. Commentary also raises concerns about the likely results if judges take on more substantial roles, due to the relative conservativism of the legal process. Some of these concerns do not appear to hold in quite the same way for the German example – free speech is not merely a matter of an absence of restrictions by the state; rather, it requires

8 K Arriaza Ibarra, ‘Management and Organization of Public Service Media Companies: Basic Concepts Related to Efficiency and Failure in European Public Service Media, with Two Case Examples (Germany and Spain)’ in K Arriaza Ibarra, E Nowak and R Kuhn (eds), Public Service Media in Europe: A Comparative Approach (Abingdon, Routledge, 2015) 147, 155–56 (emphasis added). 9 See, eg J Bakan, Just Words: Constitutional Rights and Social Wrongs (Toronto, University of Toronto Press, 1997); R Moon, The Constitutional Protection of Freedom of Expression (Toronto, University of Toronto Press, 2000). 10 N Couldry, Why Voice Matters: Culture and Politics After Neoliberalism (London, Sage, 2010). 11 Bakan (n 9) 47. 12 Moon (n 9) 7.

86  Andrew T Kenyon state action of particular forms, and so forth. But the law’s conservativism – and questions about how well law could promote diverse public speech – are likely to remain. Further, the German example is not intended to suggest that its public media is exempt from commonplace criticisms, such as public media’s relatively high degree of politically mainstream content and its connections to establishment views.13 Another point of caution concerns geography. The role of the German judiciary in sustaining public media as part of protecting free speech (and the somewhat similar aim pursued by parliaments rather than courts in Nordic ‘media welfare states’14) could suggest this approach to free speech is a peculiarity of what Daniel Hallin and Paolo Mancini famously called ‘democratic corporatist’ systems of media and government.15 In terms of history, that may hold some truth, although the geographical reach of similar approaches is wider. In any event, understanding more about the German example still suggests weaknesses that might exist elsewhere. And it offers a striking counterpoint to regular debates about public media in many places which are so often framed outside the terms of free speech.16

III.  Free Speech Ideas and Techniques How might law address positive dimensions of free speech? The German Federal Constitutional Court has set out detailed requirements about television broadcasting, drawing on the free speech protection in Article 5 of Germany’s Basic Law. For readers unfamiliar with the court, there is a reasonably extensive English language literature.17 Two points worth noting about the court here are that its decisions have the weight of precedent in a manner that would be familiar in common law systems, and that one of its central roles is the protection of rights. Article 5 of the Basic Law protects freedom of expression; it includes negative dimensions of free speech, but it is not limited to them. Its first paragraph is traditionally translated: (1) Every person shall have the right freely to express and disseminate their opinions in speech, writing and pictures, and to inform himself without hindrance from generally accessible sources.

13 An example in the UK context is T Mills, The BBC: Myth of a Public Service (London, Verso, 2016). 14 See, eg T Syvertsen, G Enli, OJ Mjøs, and H Moe, The Media Welfare State: Nordic Media in the Digital Era (Ann Arbor, University of Michigan Press, 2014). 15 See DC Hallin and P Mancini, Comparing Media Systems: Three Models of Media and Politics (Cambridge, Cambridge University Press, 2004). 16 See, for example, in relation to the UK’s BBC, the report on the inquiry chaired by Lord Puttnam that examined the future of public service television in the UK in the 21st century: D Freedman, J Curran and D Puttnam, ‘A Future for Public Service Television: Content and Platforms in a Digital World’ (London, Goldsmiths, 2016), http://futureoftv.org.uk/report/; D Puttnam and D Freedman, ‘Charter Plans Stifle BBC’s Independence’, Broadcast, 19 September 2016; UK, Department for Culture, Media and Sport, A BBC for the Future: A Broadcaster of Distinction, Cm 9242 (London, HMSO, 2016). 17 The most useful entry point in English probably remains DP Kommers and RA Miller, The Constitutional Jurisprudence of the Federal Republic of Germany, 3rd edn (Durham, Duke University Press, 2012), especially ch 8: ‘Freedom of Speech, Press, and Art’. English translations of many older free speech decisions are available via the Institute for Transnational Law at the University of Texas: https://law.utexas.edu/transnational/ foreign-law-translations/german/toc.php. Some decisions are available in English directly from the court: www. bundesverfassungsgericht.de.

The State of Affairs of Freedom  87 Freedom of the press and freedom of reporting by means of broadcasts and films shall be guaranteed. There shall be no censorship.18

A more contemporary translation offers gender neutral language and renders the second sentence as: ‘Freedom of the press and freedom of reporting through audiovisual media shall be guaranteed.’19 The second sentence makes positive dimensions of broadcasting freedom necessary ‘to guarantee that the diversity of existing opinion is articulated as broadly and thoroughly as possible and that comprehensive information is made available’.20 Market forces are ‘unable to prevent the accumulation and abuse of power in the broadcasting sector’,21 so having only commercial broadcasting would breach the constitutional requirements. The German Länder, or states, which have constitutional responsibility for almost all of broadcasting law, are obliged ‘to establish the conditions necessary for the effective exercise of speech rights’.22 The merits of the broadcasting case law are subject to ongoing domestic debate,23 but my primary concern here is what the decisions suggest more generally about free speech. I leave aside debates about the overall effectiveness of the court’s approach.24 Elsewhere, I engage with some criticisms of the court’s effectiveness regarding broadcasting freedom.25 Here, my argument is not that the broadcasting system resulting from the decisions is without fault, or that the decisions’ requirements are all substantially achieved in practice; rather, my interest is in what the decisions state on their own terms. Central to the broadcasting cases is the rationale underlying Article 5 – the formation of individual and collective opinion – which is a general democratic free speech rationale. Ideas about democratic free speech that challenge a purely negative approach emerge from the decisions, as well as legal techniques to support structural dimensions of the freedom. The decisions suggest reasons why courts might act, and how they could act. These ideas have parallels elsewhere; the German approach is not an entirely ‘separate path’ from those taken elsewhere.26 The German example, however, offers a particularly developed judicial analysis.

18 See, eg www.gesetze-im-internet.de/englisch_gg/englisch_gg.html#p0030 (emphasis added). 19 C Potschka, Towards a Market in Broadcasting: Communications Policy in the UK and Germany (Basingstoke, Palgrave Macmillan, 2012) 161. 20 W Hoffmann-Riem, Regulating Media: The Licensing and Supervision of Broadcasting in Six Countries (New York, Guilford Press, 1996) 119. 21 ibid. 22 Kommers and Miller (n 17) 449; see also eg Hoffmann-Riem (n 20) 119. 23 See, eg B Grzeszick, ‘The “Serving” Freedom to Broadcast: Subjective versus Objective Dimensions of a Fundamental Right’ in H Pünder and C Waldhoff (eds), Debates in German Public Law (Oxford, Hart Publishing, 2014) 75, 85. 24 See, eg C Witteman, ‘Constitutionalizing Communications: The German Constitutional Court’s Jurisprudence of Communications Freedom’ (2010) 33 Hastings International and Comparative Law Review 95; T Gibbons and P Humphreys, Audiovisual Regulation under Pressure: Comparative Cases from North America and Europe (Abingdon, Routledge, 2012); Potschka (n 19); W Schulz, ‘How to Control Informal Practices by Formal Law? The German Constitutional Court’s Ruling on the Independence of the Public Service Broadcaster ZDF from the State’ (2015) 7 Journal of Media Law 145; G Taylor, ‘Diversity within Government Broadcasters’ in G Gounalakis and G Taylor (eds), Media Diversity Law: Australia and Germany Compared (Frankfurt, Peter Lang, 2016) 221; D Dörr, ‘The Federal Constitutional Court’s ZDF Ruling and its Consequences’ (1 November 2016) (a revised and translated version of speech at 14th Mainz Media Forum, 2014), https://friends.ca/explore/article/the-federalconstitutional-courts-zdf-ruling-and-its-consequences-by-dieter-doerr. 25 Kenyon, Democracy of Expression (n 1); see, eg Taylor (n 24). 26 eg B Grzeszick (n 23) 90–93 notes Spain, Italy, Portugal, the Netherlands, Japan and France, including French decisions more recent than those dealt with in E Barendt, Freedom of Speech, 2nd edn (Oxford, Oxford University

88  Andrew T Kenyon My point is not to suggest that other legal systems should necessarily follow the German example in detail; there are many ways to support the state of affairs involved – courts may play a greater or lesser role, other actors have quite varied options to promote plural public speech and it is not just a question of broadcasting after all (not even in the mass media age, in which the earlier German decisions arose). Rather, the chapter seeks to make two points: first, thinking about free speech as including positive freedom suggests that state actors should seek to support plural public speech; and second, the German example shows why courts might do that and how they could then go about it. Here, five issues are drawn from the cases.27 They concern: positive dimensions of free speech; a dual broadcasting system; funding public broadcasting; internal pluralism; and the court’s ‘precautionary’ approach. These issues illustrate the power of the idea that free speech requires that efforts be made to provide comprehensive, universally available and diverse public speech, efforts supported and perhaps prompted by constitutional rulings.

A.  Free Speech Has Positive Dimensions Free speech is not merely a negative liberty for the court. Article 5 is an instrumental freedom and the interests it serves in individual and public opinion formation require state action. This makes free speech both negative and positive: ‘Its negativity protects the individual against official restraints on speech; its positivity obliges the state and its agents to establish the conditions necessary for the effective exercise of speech rights.’28 Free opinion formation cannot be met by ‘a mere negative duty’. Instead, ‘a positive order is necessary, which ensures … the variety of existing opinion is expressed … as widely and completely as possible’.29 Broadcasting freedom is a state of affairs with diversity in content and freedom from both state and market control. It is a freedom of reception as well as a freedom for speakers. In the wider German legal context, the state clearly has a ‘protective function’30 and constitutional rights establish objective principles as well as subjective protections for individuals.31 While more could be explored about those aspects of the German situation, here I merely want to suggest that a similar result may be reached for free speech rights in other constitutional systems due to the nature of speech within a democratic constitutional order. That could follow, for example, from the ways in which diverse speech can be seen Press, 2005). C Calliess, ‘Dimensions of Fundamental Rights: Duty to Respect versus Duty to Protect’ in Pünder and Waldhoff (n 23) 27, 40 notes Austria, France, Ireland and the European Court of Human Rights. One could also consider Argentina; see, eg CSJN, Grupo Clarín SA y otros c/ Poder Ejecutivo Nacional y otros/ acción meramente declarativa, Fallos, G 439 XLIX (REX), G 445 XLIX, G 451 XLIX; A Del Campo, ‘Current Trends in Argentina’s Freedom of Expression Jurisprudence’ (2 May 2014), https://globalfreedomofexpression.columbia.edu. There are also proposals elsewhere to support media diversity, sometimes drawing on parallel ideas, eg in France, P Rimbert, ‘Projet pour une presse libre’, Le Monde Diplomatique, December 2014, 20, www.monde-diplomatique.fr/2014/12/ RIMBERT/51030, who draws directly on Art 19 of the Universal Declaration of Human Rights. 27 I consider more in Kenyon, Democracy of Expression (n 1). 28 Kommers and Miller (n 17) 449. 29 BVerfGE 57, 295 (1981) (Third broadcasting case). 30 Eg D Grimm, ‘The Protective Function of the State’ in G Nolte (ed), European and US Constitutionalism (Cambridge, Cambridge University Press, 2005) 137. 31 See, eg W Heun, The Constitution of Germany: A Contextual Analysis (Oxford, Hart Publishing, 2011) 198–200.

The State of Affairs of Freedom  89 to have a special or primary role in the claims of democratic government: for people to express an opinion, they must be able to form an opinion first and form it through plural public speech.32

B.  A Dual Broadcasting System A dual system of public and commercial broadcasting can meet the constitutional requirements. In 1961, the court reasoned that spectrum scarcity and broadcasting’s high production costs limited how many broadcasters could exist. In that context, a public monopoly structured for internal pluralism met the constitutional imperative of freedom.33 In the 1980s, a dual public and commercial broadcasting system was held to meet the requirements.34 Within the dual system, however, public broadcasting’s role is primary and vital. Commercial broadcasting, subject to lesser obligations, is constitutionally permissible under a dual system only where public broadcasting is able to meet the Basic Law’s mandate for individual and public opinion formation.35 While, theoretically, that could change if commercial broadcasting by itself met the constitutional requirements, there is no sign of that in the court’s analysis to date, including its decisions in the most recent decade. In earlier decisions, the court said the Constitution requires the ‘basic provision’ to come from public broadcasting.36 While the term ‘basic’ was used, the requirement is not ‘a minimum provision to which public broadcasting is confined or could be reduced’.37 As the court has more recently emphasised (and using its current terminology), the requirement is for a comprehensive service, available to the entire population, containing substantial diversity of opinion.38 That is the functional mandate or constitutional role of public broadcasting.

C.  Public Broadcasting Requires Adequate, Non-politically Determined Funding Article 5 requires public broadcasting to receive adequate funding for a comprehensive service through a relatively depoliticised funding process. Free speech requires public broadcasting to be outside the control of state and market forces: ‘Broadcasting may not be abandoned either to the state or to any societal group.’39 One aspect of the required 32 See further Kenyon, Democracy of Expression (n 1); S Boyron, The Constitution of France: A Contextual A ­ nalysis (Oxford, Hart Publishing, 2013) 194. 33 BVerfGE 12, 205 (1961) (first broadcasting case). 34 BVerfGE 73, 118 (1986) (fourth/Lower Saxony broadcasting case). 35 BVerfGE 90, 60 (1994) (eighth/Cable Penny broadcasting case). 36 Eg BVerfGE 83, 238 (1991) (sixth/North Rhine–Westphalia broadcasting case). 37 BVerfGE 74, 297 (1987) (fifth/Baden–Württemberg broadcasting case). 38 BVerfGE 83, 238 (1991) (sixth/North Rhine–Westphalia broadcasting case). The same requirements were presented in a tripartite form in the fifth broadcasting decision (BVerfGE 74, 297): that is, transmission in a form that can reach the entire population; a full scope of programming to allow for the constitutional guarantee of free opinion formation to be met; and diversity in the presentation of opinion, pursued through organisation and procedural measures. 39 BVerfGE 90, 60 (1994) (eighth/Cable Penny case). See also BVerfGE 83, 238 (1991) (sixth/North Rhine– Westphalia broadcasting case).

90  Andrew T Kenyon independence concerns economic control. While allowing private broadcasters to exist, the court has emphasised that private programming ‘cannot manage fully to do justice to the task of comprehensive information’.40 An important reason for this is economic, with private broadcasting relying primarily on advertising.41 ‘Transmissions of interest to only a small part of the audience and often – as especially the case with ambitious cultural transmissions – requiring high expenditure will as a rule take second place, if not be entirely absent.’42 However, the court sees such programming as vital for free speech. It is part of the comprehensive, universally available and diverse information required by Article 5. The inherent weakness of private broadcasting in this regard underlies a guarantee of public broadcasting’s existence, development and funding.43 Notable in the court’s analysis is the way in which economic liberties are quite distinct from free speech. This is one of the most striking aspects of the German law when compared with the US Supreme Court’s tendency to equate speech and markets.44 ‘Market opportunities may be a question of economic freedom, but not of freedom of opinion.’45 As Christopher Witteman summarises the court’s position, the market ‘by itself, and by definition, cannot provide for a fully democratic media’.46 In its approach, the court has drawn on research in media economics in a manner that is broadly parallel to analyses like that of Edwin Baker.47 The court holds that financing by advertising will lead to ‘a host of negative results’, including ‘standardisation’ of programming, ‘one-sided reporting’ and an ‘erosion’ of the identity of public broadcasting.48 As well as independence from markets, Article 5 requires separation from institutional politics. Otherwise, funding decisions for public broadcasting could amount to indirect political influence. (Given the dominance of political parties in the German state, this is an aspect that has been quite strongly doubted in some scholarly analyses that suggest the separation is not so great in practice.) The funding implications were addressed at length in the court’s 1994 ‘Cable Penny’ decision.49 It required the legislature to structure an appropriate procedure for determining funding, and the licence fee that underlies it, ‘a multi-stage, cooperative’ process ‘that limits the opportunities for political influence’.50 The procedure must begin from public broadcasters’ own assessment of their funding needs based on their programming decisions. That assessment is then judged on two issues: first, whether the 40 BVerfGE 73, 118 (1986) (fourth/Lower Saxony case). 41 As well as pressures towards the largest audience in any given market, advertising-based funding also means that comparatively wealthy audiences are favoured in cases when circulation is not the largest in a market; see, eg R Williams, The Existing Alternatives in Communications (London, Fabian Society, 1962) 6; TE Cook, Governing with the News: The News Media as a Political Institution, 2nd edn (Chicago, University of Chicago Press, 2005) 168. 42 BVerfGE 73, 118 (1986) (fourth/Lower Saxony case). 43 BVerfGE 90, 60 (1994) (eighth/Cable Penny case). Advertising by public broadcasters must remain a comparatively minor part of the funding, due to the ‘inherent tendencies’ of advertising ‘to constrict programming and variety’. 44 See, eg Citizens United v Federal Election Commission, 558 US 310 (2010). 45 BVerfGE 74, 297 (1987) (fifth/Baden-Württemberg case). 46 C Witteman, ‘Information Freedom, a Constitutional Value for the 21st Century’ (2013) 36 Hastings International and Comparative Law Review 145, 196. 47 See CE Baker, Media, Markets, and Democracy (Cambridge, Cambridge University Press, 2002); Media Concentration and Democracy: Why Ownership Matters (Cambridge, Cambridge University Press, 2007). 48 Witteman, ‘Constitutionalizing Communications’ (n 24) 141–42. 49 BVerfGE 90, 60 (1994) (eighth/Cable Penny case); among subsequent decisions see in particular BVerfGE 119, 181 (2007) (twelfth broadcasting case). 50 BVerfGE 90, 60 (1994) (eighth/Cable Penny case).

The State of Affairs of Freedom  91 public broadcasters are remaining within their constitutional mandate; and second, whether the funding was assessed ‘correctly and in accordance with the principles of efficiency and thrift’.51 The approach is said to make funding determination ‘a professional rather than a political task’.52 Any change from the result of that process must be appropriately explained by Länder parliaments in order to be constitutional. Despite the task’s characterisation as a professional rather than political one, political battles continued. In 2004, politicians reduced public broadcasters’ funding below the level found to be required by the expert committee. The broadcasters went to court and, in 2007, it overruled the political decision.53 The departure from the expert decision was not sufficiently justified by parliaments and breached the constitutional requirement for adequate funding of public broadcasting.

D.  Internal Pluralism and Broadcasting Councils Broadcasting councils are an important element in the governance of German public broadcasters. The constitutional protection of broadcasting freedom means that broadcaster governance must be independent from state authority, so a ‘normal’ public regulator would be unconstitutional. Article 5 requires diverse council membership, with the aim that programming is not dominated by any particular interest or group.54 As stated in the court’s first broadcasting decision in 1961: ‘Broadcasters of programs must … be organised in such a way that all conceivable groups are able to have influence in their organs and have their say in overall programming.’55 In 2014, the court emphasised that members of the supervisory bodies are ‘advocates of the interest of the general public’ (not their own sectional backgrounds), with the aim being ‘that all important political, ideological, and societal currents, whose diversity … can never completely or representatively be reflected in a pluralistic body, are appropriately given a voice in the overall programming’.56 Typically, state laws set out various ‘socially relevant groups’ from which membership of broadcasting councils is drawn, such as ‘unions, business associations, religious groups, political parties, and social groups of parents, teachers, women, sports fans, artists, and trade groups’.57 It is not intended that members represent the various groups from which they come; rather, they serve as agents for the public interest. (Members cannot be bound by the groups from which they come, and they must have securely regulated tenure.58) The aim is to ‘secure diversity of opinion’ in content.59 51 ibid. 52 ibid. The court also required the legislature to ‘insure the independence of its members … [T]he panel not only should contain no broadcasting representatives, but also … no politicians’. 53 BVerfGE 119, 181 (2007) (twelfth broadcasting case); see, eg Gibbons and Humphreys (n 24) 117; Witteman, ‘Constitutionalizing Communications’ (n 24) 119–20. 54 In general, public broadcasters have ‘broadcasting councils’ and ‘administrative councils’, the latter focused more on financial aspects of management. An overview is provided by I Katsirea, Public Broadcasting and European Law (Alphen aan den Rijn, Kluwer Law International, 2008) 33–39. 55 BVerfGE 12, 205 (1961) (first broadcasting case). 56 BVerfGE 136, 9 (2014) (fourteenth/ZDF Treaty case) [40]. 57 Witteman, ‘Constitutionalizing Communications’ (n 24) 106 and its fn 24. 58 BVerfGE 136, 9 (2014) (fourteenth/ZDF Treaty case). 59 BVerfGE 83, 238 (1991) (sixth/North Rhine–Westphalia case).

92  Andrew T Kenyon This aspect of broadcasting freedom is, not surprisingly, only achieved to a degree. Indeed, it highlights how any attempt to realise free speech will only be partial. The point applies to negative dimensions of free speech as well, but it is perhaps more apparent (to common law readers at least) when thinking about the freedom’s positive dimensions. To borrow from Robert Alexy’s approach to constitutional rights, in its positive dimensions, free speech is a principle that should be realised ‘to the greatest extent possible given empirical and normative constraints’.60 Free speech here is not a binary concept that is present or absent; it is a state of affairs that is approached. It is also the case that councils have remained political in much of their operation: Wolfgang Hoffmann-Riem commented in the mid-1990s that they have ‘in reality emerged as agents of political influence’, with members ‘usually form[ing] coalitions along party lines’ and ‘meeting in so-called cliques’ before council meetings.61 There are also questions about the degree of diversity achieved in practice. ‘Socially relevant groups’ are those that register within Germany’s model of democracy, and there is a tension in seeking diversity through the use of established social groups. The resulting media content may reflect existing mainstream opinion more than the idea of a diverse broadcasting council itself suggests. In addition, caution arises because the legislature has ‘wide discretion’ in determining which groups are relevant.62 Only council composition that was ‘grossly one-sided’ would breach Article 5.63 However, the court has underlined the need to include less organised sectors within councils, as well as the need to meet gender diversity requirements mandated under the Basic Law. As stated in its 2014 decision, the aim must be to include ‘persons from all areas of the community with perspectives and horizons of experience as diverse as possible’. The aim here appears good: ‘smaller groupings which do not per se have access to the media … and perspectives which are not organised in a coherent structure’ must be presented.64 The legislature must effectively counteract the danger of having only conventional majority perspectives present on councils, and must ‘see to it that … a full range of perspectives of diverse societal currents comes into play in a significant way’.65 Where representatives of the state (or close to the state) are included on councils, they must still represent diverse perspectives and must comprise no more than one-third of the membership.66 Limits also apply to membership of any council sub-committees and similar bodies. The court understands these limits as countering the possibility that informal networks of members with links to state authorities or political parties gain de facto influence over a council. But whether that eventuates in practice remains debated.67 Equally notable, given the history elsewhere of political appointments to public broadcasters’ governing bodies,68 is the court’s emphasis on the selection of independent members 60 J Rivers, ‘A Theory of Constitutional Rights and the British Constitution’, translator’s introduction to Alexy (n 2) xvii, xxviii. 61 Hoffmann-Riem (n 20) 124. Similarly, Katsirea (n 54) 38 comments ‘in practice Broadcasting Councils are often dominated by political parties’. 62 Katsirea (n 54) 37. 63 BVerfGE 83, 238 (1991) (sixth/North Rhine–Westphalia case). 64 BVerfGE 136, 9 (2014) (fourteenth/ZDF Treaty case) [39]. 65 ibid. 66 ibid [42], [51]. The dissenting opinion of Paulus J would have gone further and prevented any council membership by members of the political executive, as is the situation in most Länder. 67 See, eg Schulz (n 24). 68 One could consider the political media links of so-called ‘polarised-pluralist’ models; see Hallin and Mancini (n 15). But there are also many examples within more ‘liberal’ models.

The State of Affairs of Freedom  93 – those who are not members of (or close to) state authority. Such appointments must not be subject to state control. In general, specific social groups are entrusted with selecting members, with their recommendations binding in all but exceptional cases with clear justification. Overall, the court’s approach to broadcasting councils appears to be the sort of ‘rough but practical solution’ that Thomas Emerson long ago suggested in relation to positive dimensions of free speech in the US context.69 Or, at least, it is a plausible attempt at supporting diversity through council membership.

E.  Speech Requires a Precautionary Approach The importance of free speech means that the Federal Constitutional Court takes what can be called a ‘precautionary’ approach in seeking to lessen commercial and political pressures on public broadcasting. The term is not used by the court, but I suggest it offers a helpful label for understanding the approach. Free speech’s role in the formation of opinion means it has a position ‘prior to’ constitutional rights in general. This foundational role for speech is even recognised in Germany, where the right to dignity under the Basic Law’s Article 1 takes precedence over, and influences the interpretation of, all constitutional rights. In its classic 1958 Lüth decision, the Federal Constitutional Court stated that free speech is absolutely basic to a liberal-democratic constitutional order because it alone makes possible the constant intellectual exchange and contest among opinions that form the lifeblood of such an order; it is ‘the matrix, the indispensable condition of nearly every other form of freedom’.70

In Lüth, the court quoted US Supreme Court Justice Benjamin Cardozo.71 Using the quoted words underlines the German court’s view that the ‘primacy’ of speech is inherent in democratic constitutional orders, not just the German one. The importance of speech makes the court alive to any possible concentration of power over public opinion. Caution against such influence is warranted because ‘when emerging developments prove to be faulty, they can only be rescinded – if at all – to a certain degree and only with considerable difficulty’.72 That is, speech requires ‘prior’ protection. Because speech is believed to have individual and collective effects, to be implicated in wider social, cultural and political processes, and to underlie the possibility to realise many other rights, speech requires a special form of protection. Precautions must be taken to protect it. The court has continued to apply this reasoning and to maintain that ‘precautions for the protection of journalistic diversity’ – diversity in public speech – must be taken.73 There is nothing about the analysis that is inherently limited to Germany. One aspect of the approach can be seen in the comparatively limited roles given to the executive and legislature in relation to broadcasting freedom, and the relatively active role of the court. The court is, for some commentators, a ‘positive legislator’74 because it is so



69 TI

Emerson, The System of Freedom of Expression (New York, Random House, 1970) 667. 7, 198 (1958). v Connecticut, 302 US 319, 327 (1937). 72 BVerfGE 57, 295 (third Broadcasting case). 73 BVerfGE 136, 9 (2014) (fourteenth/ZDF Treaty case) [36]. 74 L Favoreu and W Mastor, Les cours constitutionelles, 2nd edn (Paris, Dalloz, 2016) 150–53. 70 BVerfGE 71 Palko

94  Andrew T Kenyon precise in its requirements of the actual legislature. Neither executive nor parliament can be left to deal with broadcasting merely as a matter of policy. The free speech issues involved mean the court sets out requirements the legislature must meet: requirements which, in turn, restrict executive power.75 The executive is perhaps the most constrained, but the court also limits the legislature’s room to move. The legislature cannot, for example, limit public broadcasting to its current technological basis.76 To borrow Alexy’s terms again, positive dimensions of free speech, just like negative ones, are too important to be left to a simple parliamentary majority. Similarly, Vincent Porter and Suzanne Hasselbach observe that, constitutionally, German ‘[b]roadcasting and its regulation were too serious to be left in the hands of the politicians’.77

IV.  Free Speech and the State The above points suggest the state has an obligation to pursue conditions under which free speech can be realised. In Germany, the approach has led to a ‘dual’ broadcasting system, in which the public service must be comprehensive, universally available and diverse. And in order to have these qualities, it must be adequately funded through a relatively depoliticised process and have (at least some meaningful) independence from market and political pressures. The analysis has continued with transformations in media across recent decades. The constitutional requirements are ‘not rendered obsolete by the development of communication technology and media markets’.78 These constitutional requirements – based in the protection of free speech – have led the German court to act, rather than leaving matters to markets, regulators, the executive or parliament. In the German context, it is clear that ‘[f]ree and independent media can be the result of state action’.79 Indeed, free speech under the Basic Law requires such action. Even with strong obligations set out by a constitutional court, political battles do not end. For example, after the Federal Constitutional Court’s 1994 decision on public funding, challenges emerged in subsequent funding determinations.80 In 2004, the Länder set out ‘a genuine political rationale’ for limiting the licence fee increase because of economic conditions, other reductions in public spending and consequent hardships for the public.81 But the rationale did not meet the constitutional requirements. Public broadcasters returned to court after the ‘stunning’ deviation of the Länder from the recommended fee increase.82

75 BVerfGE 57, 295 (third broadcasting case). 76 Eg BVerfGE 83, 238 (1991) (sixth/North Rhine–Westphalia case); BVerfGE 119, 181 (2007) (twelfth broadcasting case). 77 V Porter and S Hasselbach, Pluralism, Politics and the Marketplace: The Regulation of German Broadcasting (London, Routledge, 1991) 3. 78 BVerfGE 136, 9 (2014) (fourteenth/ZDF Treaty case) [34]. 79 S Müller and C Gusy, ‘Case Study Report: Does Media Policy Promote Media Freedom and Independence? The Case of Germany’ (2011) 21, www.mediadem.eliamep.gr/findings. 80 HE Meier, ‘Independent Regulatory Bodies and Impossible Jobs: The Failure of German Public Service Broadcasting Regulation’ (2008) 86 Public Administration 133, 138–39. 81 HE Meier, ‘“For a Few Cents Less”? Nested Games in German Broadcasting Regulation’ (2006) 15 German Politics 166, 185. 82 ibid 166.

The State of Affairs of Freedom  95 The example suggests that (one or both of) political self-interest and commercial media lobbying may hamper politicians achieving the independence for public broadcasting that is required by Germany’s constitutional protection of free speech and broadcasting freedom. It suggests the value in courts outside Germany also acting. It is not that courts are disinterested bodies with regard to the media, but they at least have different interests than other major institutional actors. As Eric Barendt noted long ago, judges ‘have no occasion to shape the basic rules of broadcasting law to suit their own ends’.83 That could not be said then of politicians and it could not be said now. The distinction between the operation of courts and party politics underlines the potential value in courts seeking to protect positive dimensions of free speech, just as they seek to protect other constitutional values. That duty on courts is well recognised for restrictions on speech; it is also applicable here. Even so, there are clearly ways in which the German example does not meet its own goals. As well as debates about its effectiveness in practice, it, for example, focuses on public broadcasting and a dual public–commercial system, rather than an approach with a variety of media financing models across different forms of media.84 There are invariably failures – or at least points of ongoing debate – about the protection of negative dimensions of free speech in many democratic legal systems. So concerns about the court’s decisions should not be thought surprising. My interest here, however, is more what the goals are and how they are articulated in the legal decisions.

V. Conclusion The German example shows that courts can act in support of positive free speech, as courts commonly do for negative dimensions of free speech. This is worth emphasising because so often positive dimensions of free speech are seen only in terms of what parliament should enact and what courts should then not strike down. There are many examples of that approach in the US literature, perhaps reflecting the challenge of arguing for positive free speech in that context.85 The First Amendment is overwhelmingly negative (especially under recent US case law). But the First Amendment is only a part of free speech; it should not be equated with the freedom. A constitutional free speech provision can be very different from the First Amendment. It can create requirements for other arms of government to act to support diverse public speech, and those requirements can be judicially enforceable. If structures that affect public communication in turn influence political processes – which appears undeniable – then the ‘architecture’ of public speech must in some sense be set prior to public debate. This is widely accepted in relation to some restrictions on speech –

83 E Barendt, ‘The Influence of the German and Italian Constitutional Courts on Their National Broadcasting Systems’ [1991] PL 93, 115. 84 See further Kenyon, Democracy of Expression (n 1). 85 See, eg JA Barron, ‘Access to the Press – A New First Amendment Right’ (1967) 80 Harvard Law Review 1641; JM Balkin, ‘Some Realism about Pluralism: Legal Realist Approaches to the First Amendment’ [1990] Duke Law Journal 375; Baker (n 47); SH Williams, Truth, Autonomy, and Speech: Feminist Theory and the First Amendment (New York, NYU Press, 2004).

96  Andrew T Kenyon for example, public debate cannot decide to prevent all people of one group from speaking. If it did so, the subsequent public speech would not have a democratically ­legitimating role. (It is often suggested that, in plural democracies, laws have legitimacy through an imagined public authorship of the laws, with public debate offering a communicative form of legitimacy.) Free speech’s positive dimensions include the pluralism and independence emphasised by the German court’s decisions. The precautionary approach illustrated by the court underlines this foundational role of speech and suggests certain requirements for the architecture of public communication. Extending that idea further than the German court has done leads to free speech requiring media of different institutional forms, internal organisation and economic bases, containing speech of diverse content and style, aimed at different ends, creating different and only partially overlapping publics, and seeking to influence (among other things) political decisions.86 Laws could reshape such an architecture of public speech to some degree, but they could not make changes going beyond the requirements of free speech for non-censorship and diversity. They could not undermine the structural underpinnings of public debate that democracy is understood to require. Finally, the example flags matters for state actors elsewhere: courts, legislatures, executives and regulators. The fact that each of these actors is compromised in its ability to act in support of free speech’s positive dimensions is part of the reason why courts should not necessarily leave matters only to the others. The concerns that arise for public speech – and the implications for democratic politics – if courts do not promote a framework for diverse public speech suggests that courts should do more than focus solely on speech as a negative liberty against government. As I have noted elsewhere, that ‘is not to ignore the conservatism of the legal system and many actors within it … Decisions by judges, whatever approach they take to such matters, are likely to reshape the terrain available for political life’.87 The wager is that explicitly addressing positive dimensions of free speech within case law could improve the architecture of free speech compared with not addressing them at all. The German example illustrates important parts of the approach and the reasoning that could be used. Free speech is not merely negative, but requires ‘the state and its agents to establish the conditions necessary for the effective exercise of speech rights’,88 or at least attempt to do so. That makes forms of public media with a mission to provide comprehensive, universally available and diverse content central to free speech. Such media should not be dominated by political or market pressures, or by any particular social group, and its funding must be adequate to that mission and not be politically determined. These qualities mean that freedom of speech is distinct from market freedoms. Media regulation that is independent from state authority is required (and there may well be ongoing attempts to circumvent such independence by political actors). And a precautionary approach is warranted to protect free speech’s foundational constitutional role. Where courts do not act, the idea that freedom of speech should support a diverse media system, including adequate public funding and independence from government and economic control, offers important material for other institutional actors to use, and for scholarly evaluation of their actions or omissions.



86 For

more detail on this approach, see Kenyon, Democracy of Expression (n 1). (n 6) 37. 88 Kommers and Miller (n 17) 449. 87 Kenyon

7 The Collective Speech Rights of Minorities SALLY BROUGHTON MICOVA

Minority language rights are expression-related rights of relatively recent invention. They developed in the later part of the twentieth century as Western democracies recognised that their ‘nation states’ were sometimes made up of several ‘nations’ and languages, and began to institute a variety of protections for minority groups.1 The tradition is arguably somewhat older in Eastern Europe, as the accommodation of ethno-linguistic groups or national diversity was integral to the legal frameworks of the Soviet Union and Yugoslavia.2 In the Western context, the establishment of minority language rights was situated in understandings of rights and democracy, whereas in the Soviet and Yugoslav contexts it was part of collective bargains with the nations constituting those federal states. Although minority language rights are rooted in individual speech rights, at least in the West, they are also collective rights granted to a group of people belonging to a particular national minority, indigenous or even immigrant group. Minority language rights are not simply negative rights protecting individual expression in minority languages from interference; they also involve positive obligations on states to provide education, translation in public services and often publicly supported or state-run media. It is in relation to media where the collective nature of minority language rights is particularly clear, and where these rights are most directly connected to freedom of speech. Henceforth, I refer to these as minority speech rights.3 In this chapter, I argue that there are two distinct categories of justification for protecting them and elaborate how these are grounded in the relationship between the individual and the collective. One stems from the body of speech rights scholarship inspired by the work of Alexander Meiklejohn on the democratic function of speech4 and the need to enable the discursive spaces needed to 1 W Kymlicka, ‘Multiculturalism and Minority Rights: West and East’ (2002) 3(4) Journal on Ethnopolitics and Minority Issues in Europe 1; ‘The Rise and Fall of Multiculturalism? New Debates on Inclusion and Accommodation in Diverse Societies’ (2010) 61 International Social Science Journal 97. 2 For discussion of Soviet language policy, see A Pavlenko, ‘Language Rights versus Speakers’ Rights: On the Applicability of Western Language Rights Approaches in Eastern European Contexts’ (2011) 10 Language Policy 37. 3 This therefore does not cover those rights aimed at preservation of the language (such as education in the minority language) or access to public services (such as translation or bi- or multilingualism in public institutions), or individual rights to use the minority language in courts. 4 In A Meiklejohn, Free Speech and its Relation to Self-Government (New York, Harper, 1948), he sets out clearly the notion of the public as a ‘people’ that has entered into a compact with those governing and argues that the state has an obligation to promote speech. He further develops this view on the first amendment in A Meiklejohn, Political Freedom: The Constitutional Powers of the People (New York, Harper, 1960). There is also a branch that examines the speech rights of collectives, such as organisations and business, but these consider whether those should be considered to have rights to the protection of their self-expression in the same way as individuals rather

98  Sally Broughton Micova achieve participation parity. The other arises from the need to nurture a culture in which an individual can achieve self-realisation. I begin this chapter with a discussion of the basis for minority speech rights in international law and then outline these democratic and selfrealisation functions. I then examine the implementation of these rights in practice in Slovenia and North Macedonia, two small states with numerous minority language groups that not only maintained the old Yugoslav collective tradition, but have also come in line with the conventions and recommendations of the Council of Europe (CoE) and the guidance of the Organisation for Security and Cooperation in Europe (OSCE), both of which emphasise the positive obligations on states. This examination demonstrates the challenge that fulfilling positive minority speech rights can be, particularly for small states. Therefore, I further argue that maintaining the distinction between democratic purposes and the self-realisation function can be used to shape and even prioritise state interventions to enable minority speech.

I.  Rights and Rationales Minority speech rights are grounded in the basic rights set forth in the Universal Declaration of Human Rights (UDHR),5 which states that all people have dignity and rights, that there can be no distinction in the application of these rights based – among other things – on nationality or language, and that everyone has the right to freedom of expression. According to the UDHR, freedom of expression ‘includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers’.6 By defining expression as seeking and receiving as well as imparting, the UDHR recognises it as a ‘two-way street’, and this definition has been carried through in successive international instruments. For example, the Convention for the Protection of Human Rights and Fundamental Freedoms drafted by the CoE in 1950, otherwise known as the European Convention on Human Rights (ECHR), uses the same two-way conception of expression as the UDHR. Though it leaves out the right to ‘seek’, the ECHR defines freedom of expression as the right to ‘receive and impart information and ideas’, and also prohibits discrimination in the exercise of this right, including on the basis of ‘association with a national minority’.7 The International Covenant on Civil and Political Rights (ICCPR), which was adopted by the UN General Assembly in 1966 and came into force 10 years later, includes those rights in the UDHR and additionally sets out basic rights for national minorities. In Article 27, it states: In those States in which ethnic, religious or linguistic minorities exist, persons belonging to such minorities shall not be denied the right, in community with the other members of their group, to enjoy their own culture, to profess and practise their own religion, or to use their own language.8 than considering the societal and democratic function of collective speech. For an example, see M Dan-Cohen, ‘Freedoms of Collective Speech: A Theory of Protected Communications by Organizations, Communities, and the State’ (1991) 79 California Law Review 1229. 5 Universal Declaration of Human Rights (adopted 10 December 1948 UNGA Res 217 A(III) (UDHR). 6 ibid Art 19. 7 Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (ECHR), Art 14. 8 International Covenant on Civil and Political Rights (adopted 16 December 1966, entered into force 23 March 1976) 999 UNTS 171 (ICCPR), Art 27.

The Collective Speech Rights of Minorities  99 The ICCPR, therefore, identifies the individual holder of the right to enjoy his or her own culture and use his or her own language as part of a community of others that share that culture and language. There are no further binding agreements under the auspices of the UN capturing countries worldwide. However, in 1992, the UN General Assembly issued a Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities. Through this declaration, and the conventions and declarations that have been adopted on a regional level mainly in Europe, a consistent rationale for minority speech rights can be seen. Based on a comprehensive analysis of international and European legal instruments and standard-setting documents, Tarlach McGonagle outlines four key elements of the rationale behind protecting expressive rights for minorities: 1. 2. 3. 4.

encouraging democratic participation; creating alternative discursive spaces; creating and sustaining (cultural) identities; and preserving linguistic identity.9

As McGonagle has argued, these rights are premised on the notion that there is a group dimension to certain individual rights, because ‘while the need for a right might be ­individual … the exercise of the right can be conceivably collective and therefore dependent on interaction with others’.10 These four elements identified by McGonagle can be divided into two categories, as the first two are linked to the conduct of democracy and engagement with society, whereas the other two are about identity and culture. Both categories stem from the individual right to expression, but each also has an inherently collective dimension, which is where the enabling obligations on the state come into play.

A.  Discursive Spaces and Participation Parity In the tradition of Milton and Mill, freedom of speech or expression is required in order that society can arrive at truth or best policy. The argument here is not only that truth should be spoken, but that it is arrived at through exchange of countering arguments or views. The assumption made is that better decisions, or truth, are more likely to be reached through uninhibited exchange. This exchange has been conceptualised as taking place within a public sphere in which citizens participate in discourse for the purpose of arriving at these decisions. However, speech rights and even state intervention may be necessary to ensure that there is sufficient parity among the speakers engaged in exchanges within the public sphere. Alexander Meiklejohn is known for elaborating the right to free speech as being in the service of democracy. He justified free speech as being necessary for citizens to get the information they need in order to understand issues and for them effectively to participate in the ‘corporate self-control’ that is modern democratic governance.11 Fundamental to his ­argument was the notion that speech must be public, that policy must be openly 9 T McGonagle, Minority Rights, Freedom of Expression and of the Media: Dynamics and Dilemmas (Cambridge, Intersentia, 2011) 120. 10 ibid 25. 11 Meiklejohn, Free Speech (n 4).

100  Sally Broughton Micova discussed and that views considered dangerous by any government of the time should not be suppressed. Barendt expanded on this citizen participation argument for speech rights by adopting a constitutional conception of democracy that requires that everyone, including members of minorities, be allowed to participate in public discourse and debate. Barendt argues, ‘the rights of minorities to contribute to political debate should be respected, partly because they may have better ideas than those of the elected majority’.12 Though he appears to be thinking mainly of political minorities, Barendt’s argument maintains that whatever majority is in power should not limit or set boundaries on public discourse and that all people must be equally respected in that discourse. Such a vision of citizens equally participating in political discourse to arrive at the best policies was conceptualised by Jürgen Habermas in his elaboration of the public sphere. The Habermasian concept of the public sphere was as a place of rational discourse, within which speakers have a common aim of reaching agreement and coordinating action. He describes the ‘ideal speech situation’ that can allow this to happen based on every participant being able to express views and challenge views freely and equally without interference.13 The public sphere is the location, or discursive space, for political debate, for communication about what action should be taken in society and for the truth-seeking purpose of freedom of speech described by Mill. In describing the bourgeois public sphere upon which he based his concept, Habermas assumed that those participating in the public sphere could set aside their private concerns and social status, and debate as equals. The notion of public speech is fundamental to Habermas’s idea just as it is crucial to Meiklejohn’s view of speech in the service of democracy and better decision-making. The collective of other speakers that make up any given public is therefore necessary for the exercise of this kind of speech within that society. First Amendment scholarship and public sphere theory establish a basis for the role of the collective in the need for a public and confrontation with alternative views for individual speech in the service of democracy. However, even if a politically defined majority is prevented from setting limits on the speech of their opposition and there are no legal obstacles to entering the discursive space of the public sphere, there can still be significant barriers to equal participation for minority groups. As Nancy Fraser argues, social inequalities will permeate the discursive arena, and deliberation will be dominated by powerful groups. It is in her astute critique of the Habermasian bourgeois public sphere that one can find justifications for minority speech rights and for positive obligations on the state stemming from those rights. Introducing the notion of ‘subaltern counterpublics’, she argues that ‘in stratified societies, arrangements that accommodate contestation among a plurality of competing publics better promote the ideal of

12 E Barendt, Freedom of Speech, 2nd edn (Oxford, Oxford University Press, 2005) 20. 13 Habermas’s initial depiction of the public sphere, which appeared in English long after its original publication – J Habermas, The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society, trans T Burger (Oxford, Polity, 1989) – was further developed in his later reflections that acknowledged that some issues may be of a moral rather than purely rational nature, though he seems to have remained convinced of the role of communicative rationality within the procedural bounds he defined through the successive works Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans W Rehg (Cambridge, Polity, 1996) and The Theory of Communicative Action: vol I – Reason and the Rationalization of Society, trans T McCarthy (Boston, MA, Beacon Press, 1984).

The Collective Speech Rights of Minorities  101 participatory parity than does a single, comprehensive, overarching public’.14 According to Fraser, ‘counterpublics’ are necessary discursive places in which the groups that are not dominant in stratified societies can regroup, consolidate and even train for engagement in the wider public sphere. As such, they are not separate silos, but necessary correctives to the imbalances in the wider public sphere of deliberation and decision-making. Fraser is not directly concerned with individual rights to expression within the public sphere, but with equal status and ‘recognition’ for all groups in society. However, the purpose of this recognition is the facilitation of individuals’ participation. Hers is a slightly different view to that of Iris Marion Young, who makes a case for group solidarity above individualism and a ‘group-specific voice’.15 Like Young, Fraser is concerned with justice in the face of pervasive structural inequality, not just for national or ethnic minorities, but for all those who are not part of dominant groups. However, her idea that non-dominant groups need their own discursive spaces, or alternative public spheres, as well as recognition by the wider public is not about consolidation and valorisation of a culturally specific group identity. Her argument is fundamentally about what she refers to as parity of participation for those who have been ‘misrecognised’ by the dominant groups. Rather than assuming that universality is the path to achieving equality and justice, she claims that doing so may require recognition of group distinctiveness, and ‘which people need which kind(s) of recognition in which contexts depends on the nature of the obstacles they face with regard to participatory parity’.16 Therefore, collective counterpublic spheres for those in non-dominant groups need to exist, and be recognised, so that individuals from those groups can contribute on an equal footing in the kind of public discourse and debate referred to by Barendt and Meiklejohn. Minority speech rights can be seen as recognising the distinctiveness of groups in society and the inequalities that exist in non-homogeneous societies. The right to free speech may apply universally to all individuals, but specific speech rights that include obligations on states to ensure adequate discursive spaces for minority groups as distinct collectives within the states in which they live are needed so that minority group members can effectively participate in the governance of that society, on parity with those in the majority.

B.  Individual Self-realisation within Societal Cultures The discussion so far has focused on the role of speech in the service of society, for the best possible decision-making and equal participation in societal governance. As Baker points out, it can be limiting to understand the value of speech only in the societal benefits of its exercise in discussion, rather than in the individual liberty of the speaker.17 He maintains that freedom of speech has two main functions: participation in decision-making and self-fulfilment or self-realisation.18 The individual liberty argument is a common and

14 N Fraser, ‘Rethinking the Public Sphere: A Contribution to the Critique of Actually Existing Democracy’ (1990) 25/26 Social Text 56, 66. 15 IM Young, Justice and the Politics of Difference (Princeton, Princeton University Press, 1990). 16 N Fraser, ‘Recognition without Ethics?’ (2001) 18(2–3) Theory, Culture & Society 21, 31. 17 CE Baker, ‘Scope of the First Amendment Freedom of Speech’ (1978) 25 UCLA Law Review 964. 18 This is most clearly elaborated in ch 3 of his 1989 book – CE Baker, Human Liberty and Freedom of Speech (Oxford, Oxford University Press, 1989) – in which he highlights these two values as being primary from among

102  Sally Broughton Micova compelling one, but the argument that individual self-realisation is at the heart of ­freedom of speech is still directly related to that individual’s position as part of a collective. Unless one is an ascetic relying solely on internal reflection and revelation for one’s self-development, one will require the options presented within a collective for the realisation of personal autonomy. Self-development cannot occur in an intellectual vacuum, so an important goal of speech rights should be to ensure that people’s own ideas are subjected to the kind of ‘tests’ that can occur only when surrounded by a multiplicity of views.19 However, minority speech rights are less about the benefits of confrontation with diversity than they are about the importance of culture generated by the collective in the realisation of individual autonomy. There are some who argue for the preservation of cultural and linguistic diversity for its own sake. Arguably, such a view treats culture and language more as an artefact than as a tool for expression or for the realisation of personal autonomy. Many who take such a stance maintain that diversity of languages and cultures is itself a public good and use that notion to justify special protections for small or endangered language communities.20 There are obvious problems with seeing culture or even language as fixed artefacts. Both culture and language evolve and interact with others. Instead, the cultural and linguistic preservation rationales for minority rights that McGonagle identified in various transnational documents are grounded in the recognition that the exercise of the speech rights of minorities has a group dimension.21 Minority speech rights are intended to help sustain cultural and linguistic identity because individual self-development depends on the existence of collectively formed, though not static, culture. In the ground-breaking elaboration of his liberal theory of minority rights, Will Kymlicka draws a clear connection between ‘group-differentiated’ minority rights and individual liberty. Drawing on Dworkin, he argues that individual freedom involves adopting practices, forming beliefs and developing a conception of the good life from among the options presented and made meaningful to us within a ‘societal culture’.22 He maintains that access to a societal culture is a prerequisite to meaningful individual choice, and this cannot be just any societal culture (that is, that of the majority) because the costs of switching cultures are too high. Acknowledging that a small minority can make a transition, he contends that individuals are deeply connected to their culture; that it is integral to their identity and even self-worth, and therefore access to it must be provided.23 His is not a c­ ommunitarian argument for minority rights that focuses on the value of the cultural community as an

the four identified in the work of esteemed First Amendment scholar Thomas Emerson. Baker convincingly argues that the discovery of truth and achievement of more stable community also listed by Emerson are actually derivative of these two values. 19 J Lichtenberg, ‘Foundations and Limits of Freedom of the Press’ in J Lichtenberg (ed), Democracy and the Mass Media: A Collection of Essays (Cambridge, Cambridge University Press, 1990) 114. 20 Francoic Grin outlines a conceptual model of diversity along these lines in ‘Diversity as a Paradigm, Analytical Device and Policy Goal’ in W Kymlicka and A Patten (eds), Language Rights and Political Theory (Oxford, Oxford University Press, 2003) 169. In the same volume, Idil Boran develops a complementary argument likening linguistic diversity to biodiversity: ‘Global Linguistic Diversity, Public Goods and the Principle of Fairness’ (ibid) 189. 21 McGonagle (n 9). 22 This argument is most clearly explained in W Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford, Clarendon, 1995) ch 5. 23 This argument is decidedly different from the position that cultures and languages need to be protected for their own sake as part of the richness of humanity as a whole, which is similar to the case made for biodiversity.

The Collective Speech Rights of Minorities  103 entity. Instead, Kymlicka makes a case that such rights are necessary to give individuals access to the societal cultures they both want and need in order exercise their personal autonomy.24 As Alan Patten contends: the point of cultural rights is not to guarantee the preservation of any particular culture, but to secure fair background conditions under which people who care about the survival or success of that culture can strive to bring about that outcome.25

This is an inherently practical argument. Neither a culture nor a language can survive without some individuals exercising their agency to make it happen – without some people autonomously choosing to sustain it. Nevertheless, as Patten emphasises, this does not imply indifference on the part of the state or the privatisation of responsibility; instead, it is the role of the state to compensate for the structural inequalities that minority groups face, and the imbalances in the resources and platforms through which culture is maintained and evolves. Patten remarks, ‘[t]he fair solution is not to split the difference but to give each an equivalent bundle of rights and resources with which to pursue the life they value’.26 Linguistic and cultural rights are granted to collectives of individuals to whom a particular language and a unique culture are relevant for their own self-development. Minority speech rights therefore enable individuals from minority cultures, who often speak a different language from the majority, to preserve the societal cultures required for their own self-development. The collective may extend beyond the boundaries of any one state and be defined by shared language and cultural identity. The preservation of the collective use of language and development of culture both relies upon and enables the autonomous agency of those who belong to the collective, but often also requires intervention by the state to help ensure that those individuals have the resources needed to do so.

C.  Expectations of the State There have long been debates about whether the freedom of speech is solely a negative right or if it also obliges the state to undertake measures to ensure that freedom of speech can be exercised.27 Minority language speech issues are a prime example of the type of occasion Barendt refers to in his claim that, although it may be foremost a negative liberty, freedom of speech also ‘reflects values, which may on occasion require promotion through the law’.28 If a society values equality in participation in governance and individual autonomy, as 24 Kymlicka’s idea of multicultural or multinational citizenship based on this argument has been criticised by some as taking an essentialist view of culture. He answers his critics on this point in an EUI paper – W Kymlicka, ‘The Essentialist Critique of Multiculturalism: Theories, Policies, Ethos’, EUI Working Paper RSCAS 2014/59 (2014) – and defends his views in the face of post-multiculturalist arguments here by maintaining that an essentialist or ‘samosas, steel drums and saris’ view of culture misses the point. 25 A Patten, Equal Recognition: The Moral Foundations of Minority Rights (Princeton, Princeton University Press, 2014) 31. 26 A Patten, ‘Moral Foundations of Minority Rights: A Reply to Attas, Bardon, and Gans’ (2015) 12 Jerusalem Review of Legal Studies 91,100. 27 AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379. 28 Barendt (n 12) 36. Barendt also discusses at length the relationship between freedom of speech and the media, concluding that the state must play a role in ensuring that the media serve public discourse, thus providing elaboration of this claim (ch 7).

104  Sally Broughton Micova described above, some specific affordances may be needed in order to ensure minorities can access their own public spheres for expression and the receipt of information, and can invest in the development of their societal cultures. Recognising that states cannot simply leave the creation of these opportunities to chance or market forces, international law addresses the fact that respecting minority speech rights entails creating an enabling environment for the exercise of those rights through positive state intervention. The most detailed standards for the protection of minority speech rights have been set within the CoE through the European Charter for Regional and Minority Languages, adopted in 1992,29 and the Framework Convention for the Protection of National Minorities, adopted in 1995.30 Both the charter and the convention set out education and the media as key areas in which states have obligations stemming from the rights of minorities.31 The convention states that ‘the Parties shall adopt adequate measures in order to facilitate access to the media for persons belonging to national minorities and in order to promote tolerance and permit cultural pluralism’.32 The charter requires that states ensure that there is at least one television and one radio programme in public service media, or ‘make adequate provision’ so that their public broadcasters offer programmes, in the regional or minority languages. It states further that the parties must ‘encourage or facilitate’ radio, television and even newspapers, listing the application of financial assistance and provision of training programmes.33 These international agreements place positive obligations on states in relation to minority language media, because the ability of minorities to exercise their freedom to ‘hold opinions and to receive and impart information’34 depends on there being ‘available expressive fora and media’35 for doing so. Media are important for the creation of the alternative discursive spaces or public spheres called for by Fraser. Media in a minority language are also necessary to ‘develop and maintain a sense of a language community’s identity, strengthening its ability to stand up to stronger neighbouring language communities’.36 Having media in a minority’s language can also be a signifier to those outside of the minority group’s legitimacy in the larger society, addressing the problem of recognition within stratified societies elaborated above. Tom Moring has described adequate media provision for the exercise of minority speech rights in terms of the completeness of the media offering available to minorities. Drawing on Kymlicka’s notion of ‘institutional completeness’,37 he sets out a situation

29 Of the 47 member states of the CoE, 25 have ratified the Charter and eight have signed but not ratified it. This may change, as France’s parliament recently passed a constitutional amendment that would allow the ratification of the Charter and began the process towards ratification. 30 Belgium, Greece, Iceland and Luxembourg have signed the Framework Convention, but not ratified it. Andorra, France, Monaco and Turkey have neither signed nor ratified it. All other CoE countries have ratified it. 31 The Charter also details obligations in relation to interaction with courts and with public administration, as well as providing opportunities for artistic and cultural expression. 32 Framework Convention, Art 9, para 4. 33 The Charter, Art 12. 34 Framework Convention, Art 9, para 1. 35 McGonagle (n 9) 571. 36 M McCormack, ‘The Media and Language Maintenance’ in M McCormack and N Hourigan (eds), Minority Language Media: Concepts, Critiques and Case Studies (Bristol, Multilingual Matters, 2007) 52, 54. 37 Kymlicka (n 22). Kymlicka’s institutional completeness refers to the extent to which provision of various services functions in the minority language.

The Collective Speech Rights of Minorities  105 in which ‘there are media platforms available in the minority language for each type of media’.38 He further argues that diversity of ownership and genre are also important in order create a situation of functional completeness, which he describes as the situation in which individuals can live their lives in their mother tongue.39 Based on these recommendations, it would seem that the ideal situation for minority language groups, one that gives them the discursive spaces to house their (counter-)public sphere and nurture their culture, would be one in which minority language media are available in print, broadcast and online forms, with multiple owners or sources of media on each platform. Only a few examples of institutionally complete offerings in minority languages can be found in Europe, such as Catalan in Spain, Welsh in the UK, German in Italy and Swedish in Finland,40 because states deal with minority language in a variety of ways. Drawing on early work by Stephen Riggins, Beata Klimkiewicz sets out four different models to describe the ways that states approach meeting obligations arising from minority speech rights.41 The autonomous model describes states in which minority groups are encouraged to set up their own media, relying partially on advertising and possibly on local government or regional public funds. In both the minority protection model and the anti-discrimination/diversity model there is additionally significant state support for media in minority languages, but the latter also includes specific measures to increase the participation of minorities in mainstream majority language media. Finally, in the assimilation model there is no minority language media. The first three models could all deliver some degree of institutional completeness for enabling the exercise of minority speech rights, but some might only provide the minimum necessary for the fulfilment of the standards set by the CoE Charter and the Framework Convention. In 2003, the OSCE High Commissioner on National Minorities (HCNM) developed guidelines for the practical implementation of these standards.42 These guidelines were based on various human rights instruments, including those of the UN and the CoE, OSCE commitments and previous HCNM recommendations.43 The then Commissioner recommended, inter alia, direct state support for minority language broadcasters, including licensing arrangements or subsidies aimed at encouraging private minority language media, and allocated programming on public service media. Additional guidelines on national minorities and the media in the digital age were published in February 2019.

38 T Moring, ‘Functional Completeness in Minority Language Media’ in McCormack and Hourigan (n 36) 17, 19. 39 See also T Moring and S Godenhjelm, ‘Broadcasting for Minorities in Big and Small Countries’ in GF Lowe and CS Nissen (eds), Small Among Giants: Television Broadcasting in Smaller Countries (Gothenburg, Nordicom, 2011) 177; T Moring and R Dunbar, The European Charter for Regional or Minority Languages and the Media, vol 6 (Strasbourg, CoE Publishing, 2008). 40 Moring (n 38) 19; Kymlicka (n 1). 41 B Klimkiewicz, ‘Media Policy for Ethnic and National Minorities in Poland, the Czech Republic and Slovakia’ in M Sükösd and P Bajomi-Lázár (eds), Reinventing Media: Media Policy Reform in East-Central Europe (Budapest, Central European University Press, 2003) 155. 42 OSCE High Commissioner for National Minorities, Guidelines on the Use of Minority Languages in the Broadcast Media (The Hague, Office of the High Commissioner for National Minorities, 2003). 43 The guidelines mention that all OSCE participating states are bound by the human rights instruments of the UN and most by the standards of the CoE. It also refers to the Hague Recommendations Regarding the Education Rights of National Minorities (1996); the Oslo Recommendations Regarding the Linguistic Rights of National Minorities (1998); and the Lund Recommendations on the Effective Participation of National Minorities in Public Life (1999).

106  Sally Broughton Micova They include additional obligations to support content production across platforms, to encourage language options and due prominence for minority language content on platforms, and to invest in media and information literacy.44 At the time of writing, the newer guidelines were still being disseminated, but the 2003 guidelines were promoted throughout the OSCE region during the mid- to late 2000s and had an influence on media policy in OSCE participating states. This was particularly true in the countries of Central and Eastern Europe and the former Yugoslavia that were revising their legal frameworks for media and for minorities in order to align with the requirements of the EU integration process at that time. In the former Yugoslavia, a minority protection model had been in place that assured relatively complete media offerings for most minorities within the bounds of its partycontrolled media system. After its dissolution and the ensuing wars, each of the successor states established approaches that both drew from the old model and incorporated more autonomy for the new private media and the standards set by the CoE and OSCE.

II.  Minority Speech Rights in Practice in Slovenia and North Macedonia Yugoslavia was a federal state with a highly diverse population and a complex system for affording collective rights to the various groups within it. There was a distinction between those who were considered ‘nations’ (narod) and those who were ‘nationalities’ or national minorities. In 1946, the constitutive nations that formed the state – Slovenes, Croats, Serbs, Montenegrins and Macedonians – were all granted federal units, or republics, while Bosnia and Herzegovina was considered to be a mixed federal unit with a large percentage of Slavic Muslims who were classified as a religious group rather than a nation.45 As national minorities that were indigenously present in the territory, but which also had a ‘kin state’ nearby and therefore were not constitutive nations, the Albanians in Kosovo and Hungarians in Vojvodina were given autonomous provinces. The 1974 Constitution brought changes that gave nations and nationalities equal rights, with consequences for the use of their languages by public institutions, for the education system and for the media.46 Whereas previously only the six republics had their own broadcasters, Radio Television Prishtina (RTP) in Kosovo and Radio Television Vojvodina (RTV) were established shortly after the constitutional change.47 A period of heavy investment in radio and television programming in Hungarian, Italian and other languages in the republics followed this constitutional change. Minority language groups not large or concentrated enough to be in autonomous provinces also had media and education in their own language,

44 OSCE High Commissioner for National Minorities, Guidelines on National Minorities and the Media in the Digital Age (The Hague, Office of the High Commissioner for National Minorities, 2019). 45 V Ortakovski, Minorities in the Balkans (Ardsley, NY, Transnational Publishers, 2000) 230. 46 ibid 230. 47 RTV was established in Novi Sad in 1974 and RTP was established in Prishtina in 1975. Both broadcast in Serbian, as well as Hungarian and Albanian, respectively. RTP was closed after the conflict in 1999. It reopened initially under the auspices of the UN and the OSCE as Radio Television Kosovo, and still has programmes in Serbian as well as other minority languages.

The Collective Speech Rights of Minorities  107 but these were largely groups with status left over from the pre-Yugoslav era. In the ­Socialist Republic of Macedonia, for example, state-run daily newspapers in the Albanian and Turkish languages had been published since before 1950, and there was daily radio and television programming on the republic-level broadcaster RTV Skopje.48 Radio Koper in the Socialist Republic of Slovenia had been broadcasting in Italian since 1949, and was expanded to include television in 1971.49 In Yugoslavia, there was a tradition of collective rights to and state support for media in the languages of national minorities.50 Beginning in 1991, Yugoslavia broke into ever smaller pieces, amidst a great deal of bloodshed. As violence ceased and borders stabilized, each of the successor states joined the CoE and the OSCE, signing up to the conventions and principles that went along with that. Slovenia and Croatia joined the EU in 2004 and 2015 respectively, and the other states are all at various stages of European integration. The Yugoslav heritage in treating minority language groups was combined with the international and European norms taken on by the successor states as they joined the CoE and began the process of accession into the EU. This section focuses on North Macedonia and Slovenia. They are approximately the same size in terms of territory and population, have both been independent states since 1991 and were the regions least involved in the violent conflicts of the early 1990s. Both have multiple minority language populations, and the examination of these two cases shows that fully enabling minority speech rights in these small states presents a complex and costly burden, but also that such states could prioritise enabling discursive spaces for effective participation in the domestic democratic process, as much of the media required for the cultural community necessary for self-realisation could be provided from other sources.

A.  National Minorities in Slovenia and North Macedonia In dealing with minority rights, Western literature makes a distinction between ‘national minorities’, or indigenous minorities, and immigrant communities.51 Often, this involves ascribing more rights, particularly those relating to state investment in media and education, to national minorities than to immigrant communities. The assumptions are that immigrants have chosen to come to the state and therefore can be more expected to assimilate, and that they have a kin state where the language is dominant and therefore are not in need of the same degree of protection. In the former Yugoslavia, this distinction is problematic, both because of the different definitions used during the Yugoslav period and because many of the minorities have been created by the break-up of the former federal state into its smaller independent countries. There was migration within the region due to the war in the early 1990s, but many of the people who now belong to a minority became so when 48 P Saracini et al, Состојбата со медиумите на јазиците на заедниците во Република Македонија [The Situation of Media in the Languages of the Communities of the Republic of Macedonia] (Skopje, The Macedonian Institute for the Media & the Higher School of Journalism and Public Relations, 2012). 49 RTV Slovenia, Radio Television Slovenia in Word and Image (Ljubljana, RTV Slovenia, 2011). 50 Notably, the Roma minority, which had no kin state or tradition of dominance in the region and which was spread out in smaller communities across all the republics, was largely left out of this system. They and some other smaller groups did not fit into the established categories. 51 This has largely been established by Will Kymlicka, one of the leading scholars on multiculturalism, minority languages and minority rights – see Kymlicka (n 22); Kymlicka and Patten (n 20).

108  Sally Broughton Micova the national borders were redrawn. Whether one is considered ‘indigenous’ or a national minority with all of the afforded rights, however, seems to depend solely on which postwar historical moment the constitutional drafters in each country decided to accept as the originating moment.52 The situations in Slovenia and North Macedonia differ significantly from each other. The Slovenian constitution seems to take the end of WWII as its starting point, as it guarantees specific rights to the ‘autochthonous Italian and Hungarian national communities’, and states that in the municipalities in which they reside their languages are also official languages.53 Article 64 of the Slovenian Constitution sets out a list of specific rights, including those to education in their own language and to establish organisations for public media and publishing. A law on the autochthonous communities that was adopted in 1994 further specifies their rights. This includes a right to representation in the governing bodies of the public broadcaster (RTVSLO) and the right to hold the majority in the committee for programmes for the national minorities.54 The constitution addressed the status of the Roma population by stating that it would be established by law; this was achieved with an Act that granted them special rights in 2007. No other minorities are granted specific group rights. In contrast, the North Macedonian constitution reflects the ethnic composition of the state at the time of independence and the results of deals made to end the short internal conflict in 2001. It declares that the country is of the Macedonian people, ‘as well as citizens living within its borders who are part of the Albanian people, the Turkish people, the Vlach people, the Serbian people, the Roma people, the Bosniak people and others’.55 The constitution originally declared Macedonian to be the official language but that other languages could be used in municipalities where national minorities were in the majority. This, however, was changed after the short, violent conflict of 2001 to a complicated system based on a 20 per cent cut-off. This essentially made Albanian the second official language and also made other languages official in particular municipalities. As in Slovenia, the national minorities in North Macedonia are guaranteed education in their mother tongue, and the rights to express and develop their identity. The state is also obliged to guarantee the protection of their ethnic, cultural, linguistic and religious identities.56 In both countries, the constitutional recognition of the national minorities has been reflected in other laws relating to education, local government and the media. This is true of broadcasting in particular, which is highly regulated by law. While the Slovenian constitution only recognises a few national minorities, the North Macedonian constitution recognises many, when in fact the actual populations are both highly diverse. Table 1 shows the available census data that defined the population according to ­nationality at the time that each country’s existing minority language media policies were

52 These were, of course, not arbitrary decisions, but highly political, and were also influenced by the previous arrangements in Yugoslavia and the existence of institutions and services remaining from that time. 53 Arts 5 and 11 of the Slovenian Constitution, The Basic Constitutional Charter on the Independence and ­Sovereignty of the Republic of Slovenia (Official Gazette of the Republic of Slovenia, no 1–4/91I). 54 Self-Governing Ethnic Communities Act (Official Gazette of the Republic of Slovenia, no 65/94), Art 1. 55 Constitution of the Republic of North Macedonia. 56 Art 48, as amended in 2001 Art III.

The Collective Speech Rights of Minorities  109 being established.57 In each case, the category is referred to as ‘nationality’, or in literal translation ‘national belonging’, and the labels are self-declared. The classifications were determined by how individual respondents chose to define themselves; therefore, they do not necessarily coincide with normally accepted national minority groups. Table 1  Breakdown of populations by self-declared nationality or ‘national belonging’58 Slovenia Nationality Slovenes

Macedonia 2002 1,631,363

Nationality Macedonians

2002 1,297,981

Serbs

38,964

Albanians

Croats

35,642

Turks

509,083 77,959

Bosniaks

21,542

Roma

53,879

Muslims

10,467

Serbs

35,939

Hungarians

6,243

Bosniaks

17,018

Albanians

6,186

Vlach

Macedonians

3,972

Others

Roma

3,246

Total

Montenegrins

2,667

Italians

2,258

Others

4,432

Don’t know

20,993 2,022,547

126,325

Refused

48,588

Undetermined

22,141

Total

9,695

1,964,036

There is little in-migration in either of these countries. There was some migration from other parts of the former Yugoslavia during the war in Bosnia (1992–95), but this is surprisingly minimal. The State Statistical Office of Slovenia reported that the increase in the population by just over 50,000 between 1991 and 2002 was only partly accounted for by immigration (28,000), with the rest being due to the legalisation of the status of long-time residents with

57 In both North Macedonia and Slovenia, the most recent available census data that contains a breakdown according to self-declared nationality is from 2002. North Macedonia was due to have a census in 2011; however, this was postponed indefinitely because of the political tensions surrounding the ethnic make up of the country and disagreements between the ruling Macedonian and Albanian political parties. The census held in Slovenia in 2011 was based on administrative records and not a population survey, so has not produced published data on nationality, but instead breaks down the population by citizenship. According to the 2011 data, the total population increased to 2,050,189, of whom 82,746 were of foreign citizenship, with 38,836 of them being citizens of Bosnia and Herzegovina. The numbers of citizens of Serbia and Croatia were only just over 7,000 each, indicating that the Serbians and Croatians in Slovenia are mostly native, with Slovenian citizenship. 58 Sources: Statistical Office of the Republic of Slovenia; State Statistical Office of the Republic of North ­Macedonia.

110  Sally Broughton Micova Yugoslav citizenship.59 The 2002 Census in North Macedonia counted only 5,358 foreigners with residency of over a year, of whom 3,306 were refugees from neighbouring conflicts.60 Therefore, the minority populations portrayed in Table 1 are mostly not immigrants. They mainly represent people who were native or at least resident before each country’s independence. Not all of these self-declared ‘national belonging’ represent a separate linguistic group. Those declaring themselves ‘Muslims’ might speak the language of the majority or perhaps Bosnian.61 In the Slovenian case, the ‘other’ category includes several smaller minorities of fewer than 2,000 members that were delineated in the published census data but are condensed here for the sake of space. As regards the ‘other’ category in the North ­Macedonian case, the published census data category of ‘other’ would include both those who chose not to declare and smaller minorities, such as Croats, Greeks and those declaring as Egyptians.62 Except for the Albanians in North Macedonia, none of the minority groups consist of more than 80,000 members. Each country has multiple linguistically distinct minority groups that number fewer than 20,000 people. In both cases, among those are groups that are constitutionally recognised as national minorities with all the associated rights. Operating under the assumption that any minority language population demands an institutionally complete media, there are groups numbering fewer than 20,000 members that should be served by a diverse selection of media on multiple platforms.

B.  State Enabling of the Exercise of Minority Language Rights In both cases, the history of affording group rights to certain national minorities is still reflected in the services provided by the public service broadcaster (PSB) systems. By 2016, the Slovenian PSB, Radio Television Slovenia (RTVSLO), still had two regional centres operating autonomously and producing television and radio programmes in Italian and Hungarian, as they had done for decades. By then, RTVSLO was additionally producing some radio and television programming in the Romani language on its national ­channels. Macedonian Radio Television (MRT) had condensed its offering somewhat since the ­Yugoslav time to make space for a dedicated parliamentary channel. Nevertheless, in 2016, it still had one national television channel and one national radio channel dedicated to broadcasting in minority languages, each of which mostly contained programming in Albanian, but also programming in five other minority languages.

59 Population Census Results 2002, Slovenia, http://www.stat.si/popis2002/en/default.htm. 60 Census of Population Households and Dwelling 2002, North Macedonia, 21, http://www.stat.gov.mk/ Publikacii/knigaI.pdf 61 There is no space here to enter into a discussion of the historical, political and linguistic controversies around the distinctions between the Serbian, Croatian, Bosnian and Montenegrin languages; this chapter accepts them as recognised by the states as distinct languages. Nevertheless, it is important to note that they are very similar and do not represent barriers to media consumption, but may to some extent in terms of representation in the media. 62 These ‘Egyptians’ do not speak Arabic but are by most accounts Roma who mostly speak the Albanian language, though they may also speak Macedonian or Romani.

The Collective Speech Rights of Minorities  111 Table 2  The revenues and offerings of the Slovenian and North Macedonian PSBs as of 201563 Slovenia (RTVSLO)

North Macedonia (MRT)

Total revenue of PSB in 2015

€127,331,047

€20,881,095 (1,284,187,329 MKD)

PSB TV offering

3 national channels in Slovenian, 2 regional with several hours daily in Italian in Kopar and in Hungarian in Maribor, and a few hours per week in Romani on the national channel

2 national channels in Macedonian (one is Parliamentary Channel) and 1 national channel mostly in Albanian, with 2.5 hours daily in Turkish and 1 hour a day between Romani, Serbian, Vlach and Bosnian, each getting 30 min 3 times per week

PSB radio offering

3 national channels in Slovenian, 2 regional in Slovenian and one regional each in Italian and in Hungarian, a few hours per week in Romani on the national channel

2 national channels in Macedonian and 1 national channel mostly in Albanian, with 5 hours daily in Turkish and 30 min per day in Romani, Serbian, Vlach and Bosnian

PSB online offering

RTVSLO.si portal in mrt.com.mk portal, with one Slovenian, with some content version in Macedonian and a in English separate version in Albanian

The PSBs in these two countries struggled to provide a full range of content genres to their audiences even on their majority language channels.64 This was largely due to their limited financial resources, which are typical of PSBs in small states.65 As Table 2 shows, the budgets for RTVSLO and MRT were quite small, and even though RTVSLO’s was about six times that of MRT, each still operated three national television channels, three national radio channels and more. The Norwegian PSB, NRK, which serves a population of just over twice the size of Slovenia and North Macedonia, also with 3 national television stations, mainly in one language, and 14 radio stations, had revenues of well over €700,000,000 in 2015.66 According to managers and producers in MRT and RTVSLO, their ability to provide an institutionally complete offering in all languages was further constrained by a lack of subject matter or material for content, and those responsible for minority language programmes relied heavily on content from the PSBs in neighbouring countries.67 The capacity of each 63 Sources: Radio Televizija Slovenija, ‘LETNO POROČILO 2015’ (Annual Report 2015) (JP MRT, ‘За МРТ’ (About MRT), http://mrt.com.mk/node/2830; JP MRT, ‘ФИНАНСИСКИ ИЗВЕШТАЈ ЗА РЕАЛИЗАЦИЈА НА ФИНАНСИСКИОТ ПЛАН НА ЈРП МАКЕДОНСКА РАДИО ТЕЛЕВИЗИЈА ЗА 2015 ГОДИНА’ (Financial Report for the Realization of the Financial Plan of JP Macedonian Radio Television for 2015) (March 2016). 64 S Broughton Micova, ‘Born into Crisis: Public Service Broadcasters in South East Europe’ in GF Lowe and J Steemers (eds), Regaining the Initiative for Public Service Media (Gothenburg, Nordicom, 2012) 131. 65 RG Picard, ‘Broadcast Economics, Challenges of Scale, and Country Size’ in Lowe and Nissen (n 39) 43. 66 Statista, ‘Revenues of the Norwegian Broadcasting Corporation (NRK AS) in 2014 and 2015 (in million NOK)’ (Statista, 2016), https://www.statista.com/statistics/619193/revenues-of-the-norwegian-broadcasting-corporation; JA Fordal, ‘A Gigantic Small Broadcaster’ (NRK.no, 3 October 2016), https://www.nrk.no/about/a-gigantic-smallbroadcaster-1.3698462. The amount in NOK was reported as 5.567 billion for 2015. 67 S Broughton Micova, ‘Rights vs Reality: Minority Language Broadcasting in South East Europe’ (2013) 12 Journal on Ethnopolitics and Minority Issues in Europe 54.

112  Sally Broughton Micova of these countries to generate funds and content for public media in general is very limited, which is reflected in the PSBs’ ability to serve minority language populations. In both cases, the provision of content in minority languages on the PSB reflected those named in each country’s constitution. However, whereas in the North Macedonian case this meant that there was some attempt to meet the needs of most of those self-declaring as national minorities in the census data shown in Table 1 through the PSB, in the ­Slovenian case there was a distinct mismatch between the services provided by the PSB and the ‘demand’ for minority language media provision within the population. In some countries, this kind of shortfall might be made up for by the private sector, perhaps with the help of some preferential licensing arrangements or partial subsidies. This was not the case in the Slovenian and North Macedonian markets (see Table 3), though there were no legal barriers to the establishment and licensing of media in minority languages. There were more private media registered in North Macedonia than in Slovenia. Table 3  Private minority language media offerings in Slovenia and North Macedonia as of July 201668 Slovenia Private broadcasting revenues in 2014

N/A

Private TV offerings in minority Several local and regional languages stations have programme windows in the Romani language aimed at the Roma community. No offerings in other minority languages A variety of channels from throughout the region available on multi-channel subscription services

North Macedonia €30,086,504 (1,850,390,000MKD) One national Albanian language station that also has news in Macedonian; local and regional stations covering Albanian, Turkish, Bosnian, Romani and Serbian languages often combined with Macedonian or with each other A variety of channels from throughout the region also available on multi-channel subscription services

Private radio offerings in minority languages

One local radio station in 10 in Albanian and two combined the Romani language, with Macedonian and Serbian some programme windows on other local stations

Newspapers

No domestic minority language newspapers

Three in Albanian and one localised international in Turkish (Zaman)

The advertising budgets in both of these countries have always been quite small. Though official data is not collected by the regulator or collated by the central register in ­Slovenia, managers and advertisers attest to the small amounts available. The OSCE HCNM



68 Source:

Registers of the respective audio-visual media regulators in July 2016.

The Collective Speech Rights of Minorities  113 guidelines suggest that state subsidies should be made available to support private m ­ inority language media. Some state subsidies existed in Slovenia as of 2018 and may still be in place – a competitive fund through the Ministry of Culture – but these were not earmarked for minority language media and have been crucial to the sustainability of many local and regional media in the dominant language. The only domestic private media fully operating in any of the national minority languages in Slovenia was one radio station using the Romani language. The apparent richness of the North Macedonian media landscape in terms of minority language provision is somewhat illusory. Several local television stations are registered as providing some programming in smaller minority languages in addition to Macedonian and/or Albanian. For example, at the time of writing, TV Tri (3), which covers a mountainous region on the western border, was registered as broadcasting in Macedonia, Albanian and Turkish, while TVPlus, in the northern town of Kumanovo near the Serbian border, was broadcasting in Macedonian and Serbian. However, civil society monitoring of the content provided by local and regional stations has shown that it largely consists of music video request shows or similar, through which viewers send greetings to each other, with little or no originally produced content.69 State subsidies were abolished in 2005, but there is evidence that many of these mainly music video media subsist on political advertising during elections garnered through clientelistic relationships. Congratulating one’s nephew on marriage and music may be important parts of selfdevelopment within a societal culture, but they do not represent the kind of speech in the service of democracy imagined by Meiklejohn or the discursive space that will enable the minority group to consolidate itself as the kind of counterpublic described by Fraser. The commercial provision and, by the accounts of those in the minority PSB services, much of the public media provision perhaps satisfied the second category of justifications for minority speech rights by helping to enable the societal culture that members of the minority group need for their self-realisation. However, there are gaps in the provision of the type of media needed to provide discursive space to counterpublics and parity of participation, especially for those minority groups that are not constitutionally recognised and are therefore left out of the public media provision. An important aspect of the commercial media markets in both Slovenia and North Macedonia is the extremely high penetration of subscription multi-channel services. ­ Viewership of foreign channels has been in a range of 15–20 per cent for years, and many of them come from neighbouring countries.70 Channels, publications and websites from Serbia, Croatia, Albania, Kosovo and to some extent Bosnia are highly popular in both Slovenia and North Macedonia. Audience data indicated that by 2015 both minority and majority populations were making use of the media from the kin states of many minority language groups. Not only were minority group members experiencing their societal culture on a regional and linguistic level, but also those cultures were being recognised by Slovene and Macedonian speakers through transborder media and perhaps even that there were some blurry lines among societal cultures in the region.



69 Saracini

et al (n 48). measurements were consistent in these figures from 2010 up to the time of publication.

70 AGB/Neilson

114  Sally Broughton Micova

C.  Discursive Spaces and Societal Cultures in Slovenia and North Macedonia I argued above that states are expected to ensure institutional completeness for media in minority languages in order to provide adequate discursive spaces so that minority language counterpublics can achieve participation parity and to allow for the continued existence and development of societal cultures. In the small yet highly diverse states of Slovenia and North Macedonia, this is a significant challenge. The limited resources and low levels of original content make institutional completeness even in the majority languages difficult to achieve. In the Slovenian and North Macedonian cases, kin states’ media seemed to be playing an important role in nurturing societal cultures for each country’s minority language groups. Serbs and Croats in Slovenia consume entertainment and cultural content from Serbia and Croatia. Albanians and Turks in North Macedonia also meet these needs from foreign media. Actually, Slovenian and Macedonian cultures – though dominant in their respective countries – are among the least well served in terms of media for the development of societal cultures within the region because of the relative size and resources of those linguistic groups. Groups without kin states, such as the Roma and Vlach communities, are minorities with little to no institutionally complete media offering in any of the states in which they live. In Slovenia, and even more so in North Macedonia, they are relatively well provided for, but still lack the critical mass for an enduring and developing societal culture in which individuals can exercise autonomous self-realisation as described by Kymlicka. Having access to media content from neighbouring kin states is not likely to fulfil the information and expressive needs of minority groups for engagement in politics within the society in which they vote and depend on services. For that, they need media that engages them in discourse with a collective in the state in which they live as a counterpublic to the majority as well as with those in the majority. At the time of writing, the Slovenian public media system provided some discursive space for political expression and information in the Hungarian and Italian languages. The domestic commercial market in Slovenia provided no discursive space for receiving and imparting information in a minority language other than Romani, and that provision was minimal. In the North Macedonian media landscape, such discursive spaces were only effectively provided in the Macedonian and Albanian languages, with some minimal provision from public and commercial media in some of the other languages. In neither case was there much diversity of domestic sources of news and information within the minority languages, though arguably recognition had been achieved to some extent in Slovenia for the constitutionally recognised minorities and in North Macedonia for the larger minority language groups.

III. Conclusion Minority speech rights protect the expressive rights of individuals within the collectives in which they live, but these collectives are multilayered. They are intended to put minority groups on a more even footing with those in the majority in the state in which they live in terms of the democratic function of speech, to allow them to consolidate position and information within their own discursive spaces or public sphere, so they can be recognised

The Collective Speech Rights of Minorities  115 and achieve something closer to parity in wider public engagement. They also give those interested in maintaining and developing the societal cultures of minority groups the necessary resources, so that they might continue to exist for the self-realisation of the individuals belonging to those cultures, which may extend beyond state borders. Producing media at a level that can enable speech in the service of democracy, facilitate public spheres for debate and decision-making, and effectively nurture societal cultures requires resources. State intervention is almost always necessary, often in the form of public service media and/or subsidies. States like Canada make distinctions between indigenous minorities, which are granted minority speech rights, and immigrant communities, which are not: a distinction also supported by Kymlicka. Many others do the same. Yet, states such as Slovenia and North Macedonia, which are small but have numerous minorities that could be considered indigenous, cannot easily make such a distinction. The same is true of many even less well-resourced states in Africa that have been left in similar situations through the accident of colonial line-drawing on maps. Attempting to provide adequate media for the exercise of minority speech rights can be an impossible task, even when – as in the North Macedonian case – the legislative framework is highly compliant with international law and the standards set by the CoE and the OSCE. Making a distinction between speech rights for the democratic function of speech and speech rights for self-realisation within a societal culture might be a useful way of re-evaluating state obligations in relation to minority speech rights. In the Slovenian and North Macedonian cases, media from kin states and to some extent commercial media could be relied upon to enable the dynamic societal culture for many of those states’ minority language groups. This could be true in other regions as well. These cases therefore highlight the need for scarce resources to be devoted more towards enabling participation parity in the domestic political life of these states. This requires efforts to provide media spaces for public discourse among a multiplicity of voices from the minority groups within the state and the formation of state-level counterpublics. The 2019 OSCE guidelines emphasise that the ability of a minority group to access foreign media ‘does not diminish States’ obligation to facilitate and support the development and effective operation of media serving national minorities in their own jurisdictions’.71 Greater access to media online and otherwise from a transnational societal culture might mean less state intervention is needed to ensure minority speech rights for self-realisation, but the other function of minority speech rights remains vital. In fact, in the modern media environment in which so many media battle for the attention of individuals, states’ efforts to enable speech in the service of democracy within domestic collectives, such as encouraging community or non-profit media, tailoring rules on prominence and findability, and supporting informative content across multiple platforms, are probably more important than ever.



71 OSCE

HCNM Guidelines (n 44) 20.

116

8 The Positive Right to Freedom of Expression and Party Anonymity in Legal Proceedings MERRIS AMOS

Whilst the positive aspects of some of the rights of the European Convention on Human Rights (the Convention) that have been given further effect by the Human Rights Act 1998 (HRA) are well established in UK national law,1 the positive right to freedom of expression is still very much at a developmental stage. That is not to say that interpreting and applying the right in this way to some hard, and seemingly irresolvable, free speech issues is not possible in theory, or in practice. One such issue, which has been the subject of concern for journalists and many others for some time, is the award of anonymity by courts to parties to legal proceedings. The UK Supreme Court itself has observed that its docket can ‘read like alphabet soup’.2 In 17 per cent of the judgments made by the Supreme Court in 2016, at least one of the parties had been granted anonymity. The figure was even higher in 2010, at 24 per cent of all judgments for that year. By contrast, in 2006, only seven per cent of the judgments of the highest court were anonymised, and in 2002, it was two per cent.3 There is a plethora of legal powers under which a party to legal proceedings can be granted anonymity. Blanket anonymity, where there is no discretion and anonymity is automatically granted, is limited. Criminal and summary proceedings in youth courts are not open to the public and it is prohibited to reveal the name, address or school of any child or young person concerned in the proceedings.4 These restrictions do not apply once the child, or young person, reaches the age of 18.5 Children involved in welfare proceedings are also granted anonymity and it is prohibited to publish any material which is intended or likely to identify such a child.6 A person who makes an allegation that a sexual offence 1 For example, Art 2 (the right to life) not only prohibits life being taken, but also imposes a positive duty on public authorities to protect life, and to investigate deaths where there has been some state involvement. Art 3 imposes a similar duty. See further M Amos, Human Rights Law, 2nd edn (Oxford, Hart Publishing, 2014) chs 7 and 8. 2 Ahmed v HM Treasury [2010] UKSC 1, [1] (Lord Rodger) commenting on the submissions of counsel for the media organisations. 3 These figures concern the judgments of the Judicial Committee of the House of Lords, which was replaced by the Supreme Court in October 2009. 4 Children and Young Persons Act 1933, ss 47, 49. The award of blanket anonymity was found compatible with the European Convention on Human Rights by the European Court of Human Rights in B and P v United Kingdom (2001) 34 EHRR 529. 5 R v Cornick [2014] EWHC 3623 (QB); R(JC) v Central Criminal Court [2014] EHC 1041 (Admin). 6 Children Act 1989, s 97.

118  Merris Amos has been committed has lifelong anonymity, whether the person accused is convicted or not.7 However, proposals in 2010 to provide anonymity to those accused of rape were withdrawn in the light of concerted opposition.8 By contrast, the discretionary power to grant anonymity is very wide and a court may exercise general or specialist powers. In relation to the general power, the Supreme Court has confirmed that courts may grant anonymity to parties via the exercise of a common law power which can develop in response to new circumstances: ‘The application of the principle of open justice may change in response to changes in society and in the administration of justice. It can also develop having regard to the approach adopted in other common law countries.’9 The examples given by the court included exercise of the power where it was necessary in the interests of justice, such as where there were risks to the safety of party or a witness.10 This general common law power has been reinvigorated by the coming into force of the HRA on 2 October 2000.11 It has been held that the courts, as public authorities, are obliged by section 6 of the HRA to impose reporting restrictions, and grant anonymity, in order to protect Convention rights.12 Whilst, in 2014, the Supreme Court held that, despite this change, the common law principle of open justice remains ‘in vigour’, it also appreciated that developments were likely: the starting point in this context is the domestic principle of open justice, with its qualifications under both common law and statute. Its application should normally meet the requirements of the Convention, given the extent to which the Convention and our domestic law in this area walk in step, and bearing in mind the capacity of the common law to develop … although the Convention and our domestic law give expression to common values, the balance between those values, when they conflict, may not always be struck in the same place under the Convention as it might once have been under our domestic law. In that event, effect must be given to the Convention rights in accordance with the Human Rights Act.13

By contrast, in a 2015 judgment, the Court of Appeal held that whenever a court is asked to make an anonymity order, it was necessary to consider carefully whether a derogation from the principle of open justice was strictly necessary. In its view, the approach ‘is the same whether the question be viewed through the lens of the common law or that of the European Convention on Human Rights, in particular articles 6, 8 and 10’.14

7 Sexual Offences (Amendment) Act 1992, s 1. Complainants were first granted anonymity in 1976 following the publication of Report of the Advisory Group on the Law of Rape, Cmnd 6352 (London, HMSO, 1975), which noted the extremely distressing and positively harmful impact of media coverage and the need to further the public interest which demanded that those who perpetrated a crime such as rape were convicted. See further C McGlynn, ‘Rape, Defendant Anonymity and Human Rights: Adopting a Wider Perspective’ [2011] Crim LR 199, 213. 8 UK, Ministry of Justice, Providing Anonymity to those Accused of Rape: An Assessment of the Evidence, Ministry of Justice Research Series 20/10 (London, Ministry of Justice, 2010). 9 A v BBC [2014] UKSC 25, [40]. 10 ibid [38]–[39]. This common law power is also often referred to as the ‘inherent jurisdiction’ of the court. See, eg the reporting restriction orders made pursuant to the inherent jurisdiction of the High Court in Birmingham City Council v Riaz [2014] EWHC 4247 (Fam). 11 The HRA gives further effect to most of the rights contained in the European Convention on Human Rights (ECHR) and Protocol 1 to the ECHR. On the HRA generally, see Amos (n 1). 12 A v BBC (n 9) [60]. See also Re S [2004] UKHL 47, [23]. 13 A v BBC (n 9) [56]–[57]. 14 JX MX v Dartford & Gravesham NHS Trust [2015] EWCA Civ 96, [17].

Positive Freedom of Expression and Party Anonymity  119 The Convention rights which have played the most important role in the grant of party anonymity have been: Article 2, the right to life; Article 3, freedom from torture and inhuman or degrading treatment or punishment; Article 8, the right to respect for private life; and Article 10, freedom of expression. Finally, the specialist powers to grant anonymity on a discretionary basis are very wide, so only a selection of the most important examples are considered here. In criminal and summary proceedings involving children and young people, there is an additional discretionary power for the court to restrict reporting which may lead to the identification of the child or young person, such as naming the parent of a child.15 However, such an order is not automatic, it must be necessary and proportionate,16 and it can be lifted by the court on an application by the media. It is a contempt of court to publish a judgment in a family court case involving children unless either the judgment has been delivered in public or, where delivered in private, the judge has authorised publication. A condition usually imposed is that the published version protects the anonymity of children and family members. If any party wishes to identify him- or herself, they must seek an order of the court.17 Similarly, the general rule is that proceedings of the Court of Protection are held in private but the court may make an order authorising the publication of information or a judgment on the condition that the anonymity of the parties is protected.18 Teachers accused by a pupil at their school that he or she may be guilty of a criminal offence, such as assault or sexual assault, have anonymity, but any person may make an application to a magistrates’ court for an order dispensing with such restrictions.19 The Asylum and Immigration Tribunal may make a direction to secure the anonymity of a party or a witness,20 and a court martial can grant anonymity to defendants before it.21 It will be shown in this chapter that the resulting framework for the grant of anonymity is very unclear. When exercising the judicial discretion to determine whether or not anonymity should be awarded, courts take into account a variety of principles, and current practice is far from consistent. Unlike many other balancing acts, where only the interests of the claimant and respondent are at issue, when considering an award of anonymity, a court must also take into account the various public interests at stake, in particular, the principle of open justice. However, often certain principles are overlooked or subsumed under a detailed discussion of competing Convention rights, such as privacy and freedom of expression. Often the principle of open justice is put at risk, with the rules regarding the grant of anonymity no longer explicit, transparent or consistently applied. The purpose of this chapter is to examine whether or not the application of a positive right to freedom of expression to the current array of law and practice might result in some order being restored. What is presented is not an argument for the positive right to freedom of expression to be applied in each and every instance where party anonymity is considered 15 Children and Young Persons Act 1933, s 39; see Re S (n 12). See also Youth Justice and Criminal Evidence Act 1999, ss 45, 46. 16 R (Y) v Aylesbury Crown Court [2012] EWHC 1149 (Admin). 17 Administration of Justice Act 1960, s 12; ‘Transparency in the Family Courts: Publication of Judgments, Practice Guidance’ (16 January 2014), www.judiciary.uk/wp-content/uploads/2014/01/transparency-in-thefamily-courts-jan-2014-1.pdf. 18 The Court of Protection Rules 2007, rr 90–93. 19 Education Act 2011, s 13, which inserts a new s 141F into the Education Act 2002. 20 Asylum and Immigration Tribunal (Procedure) Rules 2005, r 45(4)(i). 21 The Armed Forces (Court Martial) Rules 2009, r 153.

120  Merris Amos by a court, but for such a right to be used to cut through the disarray and support an argument for better regulation in an area which is already heavily regulated. New regulation would ideally remove the confusion, utilise consistent principles and ensure that the award of party anonymity becomes far more consistent, predictable and reflective of potential harm. The chapter is in three parts. In the first part, a definition of the positive right to freedom of expression in theory, and in the case law of UK courts and the European Court of Human Rights (ECtHR), is set out. In the second part, the current problems with the award of party anonymity are explained. And in the final part, it is determined how a positive right to freedom of expression could be used to establish much more principled, accessible and clear regulation in this area.

I.  The Positive Right to Freedom of Expression There is considerable enthusiasm in the scholarship for a positive right to freedom of expression, in particular, a positive right to freedom of expression which would support more state regulation to facilitate freedom of expression, rather than less. Lichtenberg has noted that whilst there is the non-interference principle within the idea of free speech there is also, what she has labelled, the ‘multiplicity of voices’ principle. In her view, the ‘purposes of freedom of speech are realized when expression and diversity of expression flourish’.22 She connects this to the underlying arguments for freedom of expression: All of the arguments for freedom of expression, with the possible exception of the argument from self-expression, demonstrate the centrality not of speech simply but of discussion, debate, diversity of ideas and sources of information. They point to the multiplicity of voices as their central and unifying theme.23

Similarly, Kenyon has observed that if the rationales said to underlie free speech are to be fulfilled, the ‘absence of government restrictions on speech is insufficient’.24 He continues: some state action in support of speech appears to further classic rationales for free speech. It is not ‘harmful’ action. The action is of a relatively minimal type that has been uncontroversial in many countries during the era of mass media.25

Gibbons also makes the argument for more regulation in support of free speech, noting the need for sufficient ‘checks and balances’ to enable citizens to control undue exercises of power and pointing out that we already recognise state intervention ‘in many areas of social policy on this basis’.26 As observed by Steyn and Slarks: A right of access to information concerning state decisions and policy can empower society to hold the state to account, removing a shroud of secrecy surrounding potential human rights abuses. In this sense, information rights can be seen as necessary in order to make other rights effective.27 22 J Lichtenberg, ‘Foundations and Limits of Freedom of the Press’ (1987) 16 Philosophy & Public Affairs 329, 334. 23 ibid 348. 24 AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379, 379. 25 ibid 391. 26 T Gibbons, ‘Free Speech, Communication and the State’ in M Amos, J Harrison and L Woods (eds), Freedom of Expression and the Media (Leiden, Martinus Nijhoff, 2012) 19, 44. 27 K Steyn and H Slarks, ‘Positive Obligations to Provide Access to Information under the European Convention on Human Rights’ (2012) 17 Judicial Review 308, 308–09.

Positive Freedom of Expression and Party Anonymity  121 The support in the scholarship is also reflected in both the jurisprudence of the ECtHR and UK courts. In 2012, the ECtHR, confirming once again that positive duties can arise under Article 10 of the Convention, expressed the elements of the duty as follows: in determining whether a positive obligation exists, regard must be had to the fair balance that has to be struck between the general interest of the community and the interests of the individual, the search for which is inherent throughout the Convention. The scope of this obligation will inevitably vary, having regard to the diversity of situations obtaining in the Contracting States, the difficulties involved in policing modern societies and the choices which must be made in terms of priorities and resources. Nor must such an obligation be interpreted in such a way as to impose an impossible or disproportionate burden on the authorities … Lastly, in determining whether there is a positive obligation under Article 10 of the Convention, the Court takes into account the nature of the freedom of expression in question, its capacity to contribute to a debate on a topic of public interest, the nature and scope of the restrictions imposed on that freedom, the existence of alternatives and the weight of any competing rights of third parties or of the public in general.28

But despite these encouraging words, applications claiming that a positive right to freedom of expression has been infringed are rarely considered by the ECtHR. The most important area of impact, for the purposes of this chapter, has been on the right of access to information.29 Although Article 10 provides that the rights ‘shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority’, it was held in Leander v Sweden that in the circumstances under consideration, Article 10 did not ‘confer on the individual a right of access to a register containing information on his personal position’, nor did it ‘embody an obligation on the Government to impart such information to the individual’.30 As reflected in the judgment of the UK Supreme Court in Kennedy,31 however, ECtHR jurisprudence on this question is evolving. In Guseva v Bulgaria,32 the applicant complained that the mayor’s refusal to provide her with the information she sought was in breach of her freedom to receive and impart information. The ECtHR confirmed that Article 10 of the Convention does not guarantee a general right of access to information. However, it also stated that it had consistently emphasised that Article 10 guaranteed not only the right to impart information, but also the right of the public to receive it. It pointed out that where the applicant was an individual journalist and human rights defender, the gathering of information was an inherent protected part of press freedom. Similar considerations applied to NGOs involved in matters of public interest.33 Here, it concluded that the gathering of information with a view to its subsequent provision to the public fell within the applicant’s freedom of expression, and a violation of Article 10 was found as the interference was not prescribed by law: the applicant, a member of an association active in the area of animal defence, sought access to information about the treatment of animals in order to exercise her role of informing the public on this matter of general interest and to contribute to the public debate. Furthermore, the existence of the applicant’s right of access to the information sought was recognised both in the domestic legislation and in three judgments which ordered the Mayor to provide the information to her.34 28 Frăsilă v Romania, ECtHR, app no 25329/03, unreported, 10 May 2012, [55]. 29 See generally A Scott and A Burke, ‘The Access to Information Dimension of Positive Free Speech’, this volume, ch 4. 30 (1987) 9 EHRR 433, [74]. 31 Kennedy v Charity Commission [2014] UKSC 20. 32 ECtHR, app no 6987/07, unreported, 17 February 2015. 33 ibid [36]–[38]. 34 ibid [41].

122  Merris Amos In Magyar Helskini Bizottság v Hungary,35 the Grand Chamber held that it was not prevented from interpreting Article  10(1) as including a right of access to information.36 It held as follows: Article  10 does not confer on the individual a right of access to information held by a public authority nor oblige the Government to impart such information to the individual. However, as is seen from the above analysis, such a right or obligation may arise, firstly, where disclosure of the information has been imposed by a judicial order which has gained legal force … and, secondly, in circumstances where access to the information is instrumental for the individual’s exercise of his or her right to freedom of expression, in particular ‘the freedom to receive and impart ­information’ and where its denial constitutes an interference with that right.37

In the present case, it concluded that the information sought by the applicant NGO from the relevant police departments was necessary for the completion of the survey it was undertaking on the functioning of the public defenders’ scheme in order to be able to contribute to discussion on an issue of ‘obvious public interest’. Therefore, it concluded that the denial of access by the authorities ‘impaired the applicant NGO’s exercise of its freedom to receive and impart information in a manner striking at the very substance of its Article 10 rights’.38 A violation of Article 10 was found. Another impact of the ECtHR’s development of this right, also of relevance to this chapter, has been on the distribution of information. Although the ECtHR has held that Article 10 does not impose on a state a positive obligation to collect and disseminate information of its own motion,39 where non-state actors are prevented from distributing information or ideas, positive measures of protection may be required. For example, in Appleby v UK, the applicants alleged that they had been prevented from meeting in the town centre, in a privately owned shopping mall, to impart information and ideas about proposed local development.40 The ECtHR held that whilst Article 10 did not require the creation of rights of entry to private or publicly owned property, where the bar on access to property had the effect of preventing any effective exercise of freedom of expression, a positive obligation could arise for the state to regulate property rights.41 Here, no violation of Article 10 was found, as the applicants could obtain individual permission to distribute leaflets in some areas or employ alternative means of campaigning. The ECtHR has also used the Article 10 positive right to freedom of expression to protect journalists and publishers from violence and intimidation,42 and to promote and protect media pluralism. In its judgment in Manole v Moldova, the ECtHR confirmed that in the field of audiovisual broadcasting, Article 10 imposes a duty on states to ensure: first that the public has access through television and radio to impartial and accurate information and a range of opinion and comment, reflecting inter alia the diversity of political outlook within 35 Grand Chamber, ECtHR, app no 18030/11, 8 November 2016. 36 ibid [149]. 37 ibid [155]. 38 ibid [180]. 39 Guerra v Italy (1998) 26 EHRR 357, [53]. Violation of the positive duty under Art 8 was found as the national authorities did not take the necessary steps to ensure effective protection of the applicants’ Art 8 rights. 40 (2003) 37 EHRR 38. 41 ibid [47]. A corporate town where the entire municipality was controlled by a private body was given as an example. 42 Ozgur Gundem v Turkey (2001) 31 EHRR 49.

Positive Freedom of Expression and Party Anonymity  123 the country and secondly, that journalists and other professionals working in the audiovisual media are not prevented from imparting this information and comment.43

It also confirmed that whilst there was no obligation to create a public broadcasting system, if a state had done so, ‘domestic law and practice must guarantee that the system provides a pluralistic service’. In the ECtHR’s view: it is indispensable for the proper functioning of democracy that it transmits impartial, independent and balanced news, information and comment and in addition provides a forum for public discussion in which as broad a spectrum as possible of views and opinions can be expressed.44

Here, it found that during the relevant period there was a significant bias towards reporting on the activities of the president and the government, with insufficient opportunity for representatives of the opposition parties to gain access to television to express their views. There was also evidence of a policy of restricting discussion or mention of certain topics. The ECtHR concluded that the state had failed to comply with its positive duties and a violation of Article 10 was found.45 There has been much less consideration of Article 10 positive duties in claims brought in the UK under the HRA. In some early judgments, it was held that Article 10: prohibited interference with expression rather than requiring its facilitation;46 did not require the public subsidy of free speech;47 and did not entitle anyone to make a free television broadcast.48 Apart from in the context of access to information (which is considered below), instances similar to those where the ECtHR has found there to be a positive duty have not actually occurred in the UK. Journalists and media proprietors are not often attacked or physically intimidated, and there is relatively healthy media pluralism. The positive duty arising under Article 10 which has generated the most HRA litigation is access to information, although it has been confirmed twice by the UK Supreme Court that Article 10 creates no general right of access to information.49 The most recent relevant judgment was Kennedy v Charity Commission.50 The background to that case was that a journalist at The Times had sought disclosure of information under the Freedom of Information Act 2000 concerning three inquiries conducted by the Charity Commission. The Charity Commission had declined to disclose such information, relying upon the absolute exemption contained in section 32(2) of the HRA. It was argued that, pursuant to Article 10, this section should be read down in accordance with section 3 of the HRA or declared incompatible under section 4. The Court of Appeal had found against Kennedy, although at first instance the Information Tribunal had found in his favour.

43 ECtHR, app no 13936/02, unreported, 17 September 2009, [100]. 44 ibid [101]. 45 See also Centro Europa 7 SRL v Italy, ECtHR, app no 38433/09, unreported, 7 June 2012. The ECtHR has also considered: (i) protection from dismissal for the expression of particular opinions: Fuentes Bobo v Spain (2001) 31 EHRR 50; Palomo Sánchez v Spain (2012) 54 EHRR 24; and (ii) the duty to implement the freedom of expression judgments of the ECtHR or national court: Verein Geren Tierfabriken Schweiz v Switzerland (No 2) (2011) 52 EHRR 8; Frăsilă (n 28). 46 Persey v Secretary of State for Environment, Food and Rural Affairs [2002] EWHC (Admin) 371, [122]. 47 R (Smeaton) v Secretary of State for Health [2002] EWHC (Admin) 886. 48 R (Prolife Alliance) v BBC [2003] UKHL 23. 49 Sugar (deceased) v BBC [2012] UKSC 4; Kennedy (n 31). 50 Kennedy (n 31).

124  Merris Amos The majority of the Supreme Court, dismissing Kennedy’s appeal, held that ECtHR jurisprudence on the issue was confusing, and even if Article  10 afforded access to information in these circumstances, its requirements were fulfilled by national law. By contrast, the dissenting judges found that a right to require an unwilling public authority to disclose information could definitely arise under Article  10. Lord Carnwath in particular was keen to stress that national courts could adopt their own interpretation of Article 10.51 To date, UK courts have not explicitly referred to how a positive right to freedom of expression could impact on the award of anonymity. This could include a duty to notify media organisations when an order for anonymity is sought52 or to ensure that the chilling effect of costly legal proceedings is offset by the courts themselves in the absence of media representatives. In its judgment in AP,53 for example, the Supreme Court considered the continuation of an anonymity order in control order proceedings, although no submissions were invited from the media and the media did not seek to intervene. Whilst the Supreme Court maintained that all parties favoured the continuation of anonymity, this was not conclusive, and the Supreme Court was aware of the absence of submissions from the media.54 The Supreme Court considered it sufficient that the media could later make an application for the order to be set aside, should it become aware. It did, however, note that an improved procedure may be possible and desirable.55 Steyn and Slarks have criticised the ECtHR’s (and, by implication, the UK courts’) approach to Article  10 as piecemeal, arguing that it is characterised by an avoidance of the big questions such as the ‘principles and policy underlying information rights under the Convention’.56 It is, however, possible to discern some overarching principles which are in keeping with the theoretical justifications for positive free speech, in particular the promotion of Lichtenberg’s ‘multiplicity of voices’ ideal. For example, in Guseva, the ECtHR recognised the right of an NGO engaged in a matter of public interest to have access to information so it could inform the public and contribute to the public debate. And in Manole, it held that Article 10 imposes on states a duty to ensure that the public has access to impartial and accurate information and a range of opinions and comment; it viewed television and radio as ‘a forum for public discussion in which as broad a spectrum as possible of views and opinions can be expressed’.57 It is therefore not entirely without merit to consider the possible application of Article 10, interpreted and applied in this way, to the current framework for the award of party anonymity.

II.  The Award of Party Anonymity The norm in the UK used to be that judicial proceedings were held in public and the parties were named in judgments. Names were also given in newspaper reports and in the 51 ibid [219]. 52 In A v BBC (n 9), anonymity was granted on an application for interim suspension of deportation. The media was not notified and there was no media representation at the hearing. 53 Secretary of State for the Home Department v AP (No 2) [2010] UKSC 26. 54 ibid [17]. 55 A v BBC (n 9) [68]. 56 Steyn and Slarks (n 27), 318. 57 Manole (n 43) [101].

Positive Freedom of Expression and Party Anonymity  125 law reports.58 In 1913, determining an appeal against an order for exclusion of the public and restraint of publication of details in divorce proceedings, the House of Lords confirmed the existence of the common law principle of open justice. In his judgment in this case, the Lord Chancellor, Viscount Haldane, stated as follows: But unless it be strictly necessary for the attainment of justice, there can be no power in the court to hear in camera either a matrimonial cause or any other where there is a contest between the parties. He who maintains that by no other means than by such a hearing can justice be done may apply for an unusual procedure. But he must make out his case strictly, and bring it up to the standard which the underlying principle requires … he must satisfy the court that by nothing short of the exclusion of the public can justice be done. The mere consideration that the evidence is of an unsavoury character is not enough, any more than it would be in a criminal court, and still less is it enough that the parties agree in being reluctant to have their case tried with open doors.59

In his concurring judgment, Lord Shaw explained the rationale for the principle, essentially a plea for all information to be available and for all voices to be heard: If the judgments … were to stand, then an easy way would be open for judges to remove their proceedings from the light and to silence for ever the voice of the critic and hide the knowledge of the truth. Such an impairment of right would be intolerable in a free country, and I do not think that it has any warrant in our law.

At the time, three exceptions to the principle of open justice were possible: in suits affecting wards; in lunacy proceedings; and in those cases where secrecy – such as a trade secret – was the essence of the action. In the Lord Chancellor’s view, whether the present state of the law was satisfactory was a question not for the courts, but for the legislature. In 1926, Parliament did intervene, by passing the Judicial Proceedings (Regulation of Reports) Act 1926, which restricted the freedom of the press to report any indecent matter the publication of which would be ‘calculated to injure public morals’. Over time, exceptions to the principle of open justice have continued to grow. As noted above, there is blanket anonymity, where there is no discretion and anonymity is automatically granted,60 and discretionary anonymity, where a court may exercise general or specialist powers.61 Also, as previously noted, the general common law power has been reinvigorated by the coming into force of the HRA on 2 October 2000.62

A.  Principles Applied to the Discretionary Award of Anonymity Whilst the application of a positive right to freedom of expression is also relevant to those instances where blanket anonymity prevails, the main focus of this chapter is the 58 Ahmed (n 2) [22] (Lord Rodger). 59 Scott v Scott [1913] AC 417. 60 These include: criminal and summary proceedings in youth courts (Children and Young Persons Act 1933, ss  47, 49); children involved in welfare proceedings (Children Act 1989, s 97); and a person who makes an­ allegation that a sexual offence has been committed (Sexual Offences (Amendment) Act 1992, s 1). 61 These are also very wide, and include: judgments in family court cases involving children (Administration of Justice Act 1960, s 12; ‘Transparency in the Family Courts’ (n 17)); proceedings in the Court of Protection (The Court of Protection Rules 2007, rr 90–93); teachers accused by a pupil at their school of a criminal offence (Education Act 2002, s 141F); the Asylum and Immigration Tribunal (Asylum and Immigration Tribunal (Procedure) Rules 2005, r 45(4)(i)); and courts martial (The Armed Forces (Court Martial) Rules 2009, r 153). 62 The HRA gives further effect to most of the rights contained in the ECHR and Protocol 1 to the ECHR. On the HRA generally, see Amos (n 1).

126  Merris Amos ­ iscretionary award of anonymity. When exercising this discretion, courts take into account d a variety of principles. The first principle almost always considered is the principle of open justice, and it is this principle which has the closest relationship to the provision of information and multiplicity of voices ideal. The Supreme Court has confirmed and explained this principle as follows: It is a general principle of our constitutional law that justice is administered by the courts in public, and is therefore open to public scrutiny. The principle is an aspect of the rule of law in a democracy … society depends on the courts to act as guardians of the rule of law … Who is to guard the guardians? In a democracy, where the exercise of public authority depends on the consent of the people governed, the answer must lie in the openness of the courts to public scrutiny.63

In the Report of the Committee on Super-Injunctions,64 open justice is described as a ‘fundamental principle of the common law’, a ‘sacred part of the constitution of the country’ and as an ‘aspect of freedom of speech’.65 The principle of open justice is also reflected in Article 6 of the Convention. Its importance has been explained by the ECtHR as follows: The public character of proceedings before the judicial bodies referred to in Article 6(1) of the Convention protects litigants against the administration of justice in secret with no public scrutiny. It is also one of the means whereby confidence in the courts, superior and inferior, can be maintained. By rendering the administration of justice visible, publicity contributes to the achievement of the aim of Article 6(1) namely a fair trial, the guarantees of which is one of the fundamental principles of any democratic society, within the meaning of the Convention.66

In the criminal and civil justice contexts, open justice facilitates the proper functioning of the courts and its participants, ‘ensuring protection of the right to a fair trial; [it] boosts public confidence in the courts and reduces the likelihood of vigilantism’; and it provides an educative function, ‘improving public debate relating to criminal and legal matters’. It can also enable publicity regarding suspects, and enable ‘greater monitoring of the police and prosecuting authorities’.67 Such justifications apply equally to the civil justice context. Despite the importance of the principle, however, difficulties have arisen in formulating the test to apply when exceptions are contemplated. Originally, only exceptions that were ‘strictly necessary’, rather than simply ‘necessary’, were countenanced. ‘Strictly necessary’ is also the wording utilised in Article 6, although it has been suggested that the test of strict necessity only applies to possible exceptions not mentioned in Article 6 itself.68 The HRA has muddied the waters, and courts are now clearly confused between a test of strict necessity and a test of necessity which encompasses the test of proportionality.69

63 A v BBC (n 9) [23]. See also Re S (n 12) [30] (Lord Steyn). 64 ‘Report of the Committee on Super-Injunctions: Super Injunctions, Anonymised Justice and Open Justice’ (London, Committee on Super-Injunctions, 2011). 65 ibid 8. 66 Fazliyski v Bulgaria 35 BHRC 583, [64]. 67 McGlynn (n 7) 204. See also A Zuckerman, ‘Common Law Repelling Super Injunctions, Limiting Anonymity and Banning Trial by Stealth’ [2011] Civil Justice Quarterly 223, 224. 68 It has been suggested that the restrictions set out in Art 6 must be a proportionate response to a pressing social need. DJ Harris, M O’Boyle, EP Bates and CM Buckley, Harris, O’Boyle and Warbrick: Law of the European Convention on Human Rights, 3rd edn (Oxford, Oxford University Press, 2014) 434. 69 The confusion is illustrated by the Report of the Committee on Super-Injunctions (n 64) 12, 15, where the Committee alternated between a proportionality test and a strict necessity test.

Positive Freedom of Expression and Party Anonymity  127 The second principle almost always considered by a court determining a request for anonymity is freedom of expression. This is considered in two ways: as directly linked to the principle of open justice and as a freestanding right. It has been held that the media has an important role in scrutinising the administration of justice and as the conduit of information about particular proceedings which may be of public interest.70 Where freedom of expression is considered as a freestanding right protected by Article 10 of the Convention, an order for anonymity is seen as an interference with this right.71 According to the Supreme Court, stories about particular individuals are ‘simply much more attractive to readers than stories about unidentified people’, and ‘article 10 protects not only the substance of ideas and information but also the form in which they are conveyed’.72 The viability of newspapers and magazines has also been taken into account.73 In its judgment in Ahmed, the Supreme Court held that the possibility that some sectors of the press ‘abusing their freedom to report cannot, of itself, be a sufficient reason for curtailing that freedom for all members of the press’. But it is important to recall that there is no absolute right to freedom of expression in this context. Whilst freedom of expression must be respected, its application must be tempered by the knowledge that the media can behave very badly indeed. As Bohlander observes, elements of the media are incentivised by ‘making money’ and feeding the ‘salacious appetites of tabloid consumers for being dished the dirt on their fellow citizens’.74 The example concerning Sir Cliff Richard, the police and the BBC demonstrates that it is not just the tabloid press at whom accusations can be levelled.75 The Leveson Report into the culture, practices and ethics of the press confirmed many of the media’s worst excesses.76 The third and final principle taken into consideration by a court considering an award of anonymity is the risk of harm that disclosure of a person’s identity may cause to ‘the maintenance of an effective judicial process’ or to the ‘legitimate interests of others’.77 The former justification has enjoyed a renaissance in recent years with the judgment of the Supreme Court in A v BBC, which was a challenge to a deportation order where the claimant had been convicted of sexual offences against his stepchild.78 The lower courts had concluded that anonymity was necessary as, in the absence of anonymity, there was a real risk that A’s identity and history as a sex offender would be publicised. Contrary to Article 3 of the Convention, this publicity may expose him to vigilante behaviour in his country of origin, to which he was to be deported. The Supreme Court concluded that the publication of A’s identity would have subverted the tribunal’s decision to authorise his deportation, which had been based on an assessment that there was no real risk of a violation of Article 3 if his identity was not published: ‘the publication of A’s identity would therefore have frustrated 70 A v BBC (n 9) [49]. 71 Ahmed (n 2) [35]. 72 ibid [63]. 73 ibid [63]. See also Re S (n 12) [34] (Lord Steyn). 74 M Bohlander, ‘Open Justice or Open Season? Should the Media Report the Names of Suspects and­ Defendants?’ (2010) 74 Journal of Criminal Law 321, 323. 75 See further A Travis, ‘Sir Cliff Richard Raid by Police Was “Utterly Inept” Say MPs’, The Guardian, 24 October 2014, www.theguardian.com/uk-news/2014/oct/24/cliff-richard-raid-police-south-yorkshire-bbc. 76 BH Leveson, An Inquiry into the Culture, Practices and Ethics of the Press, vol II, HC 780 (London, Stationery Office, 2012) Pt G, ch 4, [2.39]. 77 A v BBC (n 9) [41]. 78 ibid.

128  Merris Amos the judicial review proceedings before the court … [i]ndeed, the entire proceedings since at least 2007 would have been rendered largely pointless’.79 In A v BBC, the decision to depart from the principle of open justice was based on the interests of justice, to protect A’s safety and to maintain the authority and impartiality of the judiciary.80 Most claims for anonymity, however, concern harm to the ‘legitimate interests of others’, and this is now almost exclusively concerned with the harm which will flow from the possible breach of an individual’s Convention rights. The Supreme Court has confirmed that states are obliged by Articles 2 and 3 of the Convention to have a structure of laws in place which will help to protect people from attacks on their lives or from assaults;81 in practice, however, very few anonymity orders are made on this basis.82 What is now most common is for an order for anonymity to be based primarily upon Article 8 of the Convention, the right to respect for private life, particularly as the scope of Article 8 also includes at least some aspects of an individual’s reputation.83 Where Article 8 is engaged, making an anonymity order addressed to the press is ‘one of the ways that the United Kingdom fulfils its positive obligation under article 8 of the Convention to secure that other individuals respect an individual’s private and family life’.84 However, Article 8 does not grant an absolute right to respect for private life. Interferences are possible provided these are in accordance with the law and necessary, in a democratic society, for the protection of the rights and freedoms of others.85 In its application, this principle is riven with difficulties. One of the most serious is the failure of the courts to apply a rigorous standard of proof to allegations of possible harm. The 2011 recommendations of the Committee on Super-Injunctions, that derogations from the principle of open justice be strictly necessary, established through clear and cogent evidence, and subject to intense scrutiny, are rarely utilised in practice.86 A further problem is the scope which has been afforded to Article 8, the right to respect for private life. A question often overlooked when it is sought to protect information, such as a name, is whether or not the information is actually private.87 It is possible to argue that where an administrative decision is contested in the courts, as it was in AP, privacy has actually been waived. The extension of the scope of Article 8 to encompass reputation has brought further complications. For example, in Ahmed, each of the appellants had been informed that the Treasury had reasonable grounds for suspecting that he was, or might be, a person who facilitated the commission of acts of terrorism. They were designated under Article  4 of the ­Terrorism (United Nations Measures) Order 2006 and subject to freezing orders. Each denied the ­allegations and maintained that they were of good character. Pursuant to a power in the Order, the Treasury decided that their identity should be treated as ­confidential. 79 ibid [73]. 80 ibid [75]–[76]. 81 Ahmed (n 2) [27]; A v BBC (n 9) [49]. 82 See, eg Venables v News Group Newspapers [2001] 2 WLR 1038. 83 Ahmed (n 2) [42]; Re S (n 12) [27]. See further T Aplin and J Bosland, ‘The Uncertain Landscape of Article 8 of the ECHR: The Protection of Reputation as a Fundamental Human Right?’ in AT Kenyon (ed), Comparative Defamation and Privacy Law (Cambridge, Cambridge University Press, 2016) 265. 84 Ahmed (n 2) [42]. 85 A v BBC (n 9) [54]. The so-called ‘super-injunction’ was a direct result of the application of Art 8 of the ECHR in this context. See further Zuckerman (n 67). 86 See, eg A v BBC (n 9). Contrast the judgments in Ahmed (n 2) [14] and Riaz (n 10) [148]. 87 R (Willford) v Financial Services Authority [2013] EWCA Civ 674, [9].

Positive Freedom of Expression and Party Anonymity  129 However,  it  could not explain to the court why it had decided this and subsequently reversed its position. The Supreme Court concluded that as the right to protection of reputation fell within Article 8, publication of the fact that these individuals were subject to these measures was within the scope of Article 8.88 It is difficult to understand why the fact that someone has been subject to what effectively amounts to a civil penalty is considered to be an aspect of private life when in most instances anonymity is not afforded to those accused, charged or convicted of a crime.89 Furthermore, the extension of Article 8 to the protection of reputation in this context is not in keeping with the jurisprudence of the ECtHR. The ECtHR has held that a person’s reputation forms part of his or her personal identity and psychological integrity, and therefore falls within the scope of private life. But in order for Article 8 to come into play, the attack on personal honour and reputation must attain a certain level of gravity and be in a ‘manner causing prejudice to personal enjoyment of the right to respect for private life’.90 Almost all of the relevant ECtHR jurisprudence concerns attacks on reputation, usually through lies and malicious insinuation, rather than penalties or restrictions based on reasonable suspicion. Other judgments of the ECtHR indicate that, rather than accepting that penalties based on reasonable suspicion constitute a sufficient interference with reputation to engage Article 8, it is more appropriate for Article 8 to offer protection to those who have been accused, but not charged or convicted, of a crime.91

B.  Where Is the Information Informing a Multiplicity of Voices? Although a court considering an award of discretionary anonymity starts from the point of preserving open justice, as outlined above, the principles then taken into account – particularly the application of Article 8 – have resulted in a legal system where often information is suppressed and voices silenced without due regard to the state’s duty to ensure that freedom of expression is upheld. This has been of particular concern to journalists. In an article in June 2016,92 journalist Louise Tickle wrote an opinion piece about the ban on reporting family court proceedings following the murder of six-year-old Ellie Butler by her father: The ban on reporting means decisions made by judges, as well as evidence given and processes followed – sometimes extremely poorly – by local authority social work departments in care cases, are simply not subject to any sort of scrutiny … Reporting of what went on in court may not have saved Ellie. But grave decisions that affect the future of vulnerable children and entire families should not be allowed to evade public view in the name of privacy. … There are wider issues at stake too. When decisions are being made that affect a family’s future for ever – such as non-consensual adoption, which is now driven by a significantly ideological policy push from the government – preventing detailed reporting becomes a convenient way for central and local government and the judiciary to avoid being scrutinised or held to account.

88 Ahmed (n 2) [42]. 89 Mr B v Revenue & Customs Commissioners [2014] UKFTT 256 (TC), [93]. 90 A v Norway, ECtHR, app no 28070/06, unreported, 9 April 2009, [64]. 91 Apostu v Romania 38 BHRC 485. 92 L Tickle, ‘The Ellie Butler Case Shows Why Family Courts Must Be Opened Up’ The Guardian, 21 June 2016, www.theguardian.com/commentisfree/2016/jun/21/ellie-butler-family courts-opened-public-scrutiny-judges.

130  Merris Amos Problems with the current system can also be illustrated by an analysis of the Supreme Court judgment in AP. Here, a control order had been imposed on AP in 2008 under the Prevention of Terrorism Act 2005. Such orders were made where the Secretary of State had reasonable grounds for suspecting the individual was or had been involved in terrorismrelated activity.93 AP brought various legal proceedings to challenge the order, and the Supreme Court, in its June 2010 judgment, quashed the residence requirement part of the order which required AP – who was from London – to live 150 miles away.94 The ­anonymity order was made at the outset of proceedings in the Administrative Court and before the Special Immigration Appeals Commission, and was never challenged. This judgment of the Supreme Court concerned whether the anonymity order should cover publication of the 2010 judgment. The Supreme Court concluded that AP’s right to respect for his private and family life, and the need to protect him from the risk of violence, justified maintenance of the anonymity order. No submissions were invited from the media and there is no evidence that any representatives of media interests sought to intervene. Both AP and the Secretary of State wished for anonymity to continue. This was despite a number of countervailing considerations. The identity of AP as someone subject to a control order, and then subject to deportation on national security grounds, was clearly a matter of public interest. The use of control orders was extremely controversial and, whilst it is appreciated that in some instances anonymity is appropriate to protect the subject of the control order, granting anonymity in each instance can result in a situation where little is known about the type of person subject to a control order, in particular, their background, the area where they live and what they are suspected of. Even if the potential for the media to draw unwarranted attention to AP was taken into account, this had to be balanced against the fact that AP was not the first person to be subject to a control order.95 Even though the media did not appear at the proceedings, a number of interests it may have raised should have been taken into account. For example, there was no clear and cogent evidence that the use of control orders generally would be put in jeopardy by refusing anonymity to controlees. Furthermore, there was no evidence that extreme hardship would result if AP were not granted anonymity. There were no credible threats of violence against him. Whilst there were reports of community tensions and racist attacks on members of the Muslim community in the town where AP was required to live, there was no evidence that the level of threat was different to that presented by any other place. Furthermore, Article 8 was not breached. On the assumption that Article 8 was even engaged, the lifting of anonymity would reveal a restriction imposed, based on reasonable suspicion, which was necessary, in the interests of national security and for the prevention of crime. It is difficult to see how the removal of anonymity would result in a breach of Article 8 whilst imposing a control order does not. The exception to open justice here does not appear to have been strictly necessary or based upon cogent evidence subject to intense scrutiny by the court, and the anonymity order should have been lifted. 93 Prevention of Terrorism Act 2005, s 2. 94 AP (n 53). 95 Control orders were made against 52 people during the lifetime of the 2005 Act. See further D Anderson, Control Orders in 2011: Final Report of The Independent Reviewer on The Prevention of Terrorism Act 2005 (London, Stationery Office, 2012).

Positive Freedom of Expression and Party Anonymity  131

III.  Applying the Positive Right to Freedom of Expression The justifications given for open justice closely match those supporting the positive right to freedom of expression, and the following quotation from the Supreme Court’s judgment in A v BBC explains how the two are inextricably linked: The connection between the principle of open justice and the reporting of court proceedings is not however merely functional. Since the rationale of the principle is that justice should be open to public scrutiny, and the media are the conduit through which most members of the public receive information about court proceedings, it follows that the principle of open justice is inextricably linked to the freedom of the media to report on court proceedings.96

It is therefore difficult to understand why the principles applied by courts considering a discretionary award of party anonymity are in their current state. Sources of information informing public discussion and debate are, in many instances, being suppressed. But the experience of the UK is not an isolated example; other countries have experienced comparable difficulties. In its 2009 report,97 the New Zealand Law Commission concluded that a number of changes were required to the national law concerning anonymity ‘to place greater emphasis on the principle of open justice and to ensure that the grounds on which it may be departed from are transparent, explicit and consistently applied’. Subsequently, the broad discretion was reduced and the grounds on which anonymity could be granted were set out in legislation.98 In the UK, whilst there are a number of laws and principles regulating the award of party anonymity, what is needed – as was the case in New Zealand – is transparency and consistency, and a positive right to freedom of expression could be the catalyst by which this is achieved. Rather than the present piecemeal approach, this would support overarching regulation where the starting point would be the principle of open justice. Where exceptions are requested, in accordance with the recommendations of the Committee on Super-Injunctions, the onus must be on the claimant to establish, through clear and cogent evidence, that this is strictly necessary. The court must subject this request to intense scrutiny with the objective of keeping derogations from open justice to an absolute minimum. Assumptions about the media must be supported by evidence, and the possibility of bad behaviour on the part of the media must be given due consideration where there is evidence that its conduct may cause harm. Courts should also give rigorous consideration to the interests of the media and to freedom of expression, even if the media does not make submissions. Where it is claimed that the disclosure of identity may harm the maintenance of an effective judicial process, or the interests of others, extreme hardship should be the test utilised. Clear and cogent evidence of the extreme hardship must be required, and the court should subject this to intense scrutiny. A claim of a possible breach of Convention rights should not be allowed to trump all of the other rights and interests at stake. Courts should also think more carefully about the scope of Article 8 in this context in particular, whether 96 A v BBC (n 9) [26]. 97 New Zealand Law Commission, Suppressing Names and Evidence, Report 109 (Wellington, Law Commission, 2009) 8. 98 See Criminal Procedure Act 2011 (New Zealand), ss 200–04.

132  Merris Amos the ­claimant has waived privacy by bringing the matter before the courts, and whether the protection of reputation really outweighs restrictions and penalties based upon reasonable suspicion of wrongdoing. Whilst it is beyond the scope of this chapter to consider, it might also be the case that overarching regulation of party anonymity might secure anonymity for those who have not previously been granted it, including those who have been accused of a crime or arrested (but not yet charged), or defendants in sexual offence prosecutions.99 In recent years, all new proposals for blanket anonymity have failed. In 2002, the Home Affairs Committee report concerning historic allegations of child abuse recommended that persons accused of such abuse should be granted anonymity.100 This recommendation was rejected by the government, which claimed that there was no evidence underpinning a number of the Committee’s conclusions.101 In 2010, the Ministry of Justice published Providing Anonymity to those Accused of Rape: An Assessment of the Evidence,102 but, following considerable opposition, the proposal was withdrawn.103 Efforts to secure discretionary anonymity for such individuals through the application of human rights law have also generally been unsuccessful,104 although there has been a recent development as a result of the High Court judgment in Richard v BBC.105 Entertainer Cliff Richard argued that a police investigation and search of his home were matters in respect of which he had a legitimate expectation of privacy against the police and the BBC, which had reported on the investigation and search. The court concluded that whether or not there was a reasonable expectation of privacy in a police investigation was a fact-sensitive question and not capable of universal answer.106 It concluded on the authorities, and as a matter of general principle: As a general rule it is understandable and justifiable (and reasonable) that a suspect would not wish others to know of the investigation because of the stigma attached. It is, as a general rule, not necessary for anyone outside the investigating force to know, and the consequences of wider knowledge have been made apparent in many cases (see above). If the presumption of innocence were perfectly understood and given effect to, and if the general public was universally capable of adopting a completely open- and broad-minded view of the fact of an investigation so that there was no risk of taint either during the investigation or afterwards (assuming no charge) then the position might be different. But neither of those things is true. The fact of an investigation, as a general rule, will of itself carry some stigma, no matter how often one says it should not.107

However, it was careful to point out that this did not mean there was always a right to privacy, as there may be reasons why, in a given case, no expectation arises or is displaced.

99 The anonymity for defendants accused of rape introduced by the Sexual Offences (Amendment) Act 1976 was repealed in the Criminal Justice Act 1988. 100 UK House of Commons, Home Affairs Committee, The Conduct of Investigations into Past Cases of Abuse in Children’s Homes, Fourth Report of Session 2001–02, HC 836 (London, Stationery Office, 2002) [99]. 101 See further P Rumney and R Fenton, ‘Rape, Defendant Anonymity and Evidence-Based Policy Making’ (2013) 76 MLR 109, 113. 102 Above n 8. 103 See further Rumney and Fenton (n 101) 100; McGlynn (n 7) 200–02. 104 See, eg PNM v Times [2017] UKSC 49. 105 [2018] EWHC 1837 (Ch). See also Khan v Bar Standards Board [2018] EWHC 2184 (Admin). 106 Richard v BBC (n 105) [237]. 107 ibid [248].

Positive Freedom of Expression and Party Anonymity  133 Here, it concluded Cliff Richard had a reasonable expectation in relation to his investigation and search of his home against the police and the BBC.108

IV. Conclusion The positive right to freedom of expression has been justified and promoted both in theory and in practice. Whilst currently underdeveloped in UK case law, there is no reason why it could not operate as a useful organising principle to secure better regulation in this difficult and complex area. The ECtHR has already held that Article 10 of the European Convention on Human Rights can be used to support public access to information so that as broad a spectrum of views and opinions as possible can be expressed. Rather than a piecemeal approach, what is necessary now is for proposals for reform of this complex area to take this into account and move forward. New regulation would do much to ensure that the award of party anonymity is more principled, and more predictable, whilst upholding the core values underpinning freedom of speech. Only more comprehensive regulation could ensure that competing Convention rights are balanced and that Article 8, the right to respect for private life, is kept within appropriate limits. The framework for reform is already there; it remains to be seen whether or not there will be any positive movement.



108 ibid

[252]. See also ERY v Associated Newspapers Ltd [2016] EWHC 269 (QB), [2017] EMLR 9.

134

9 Positive Free Speech and Public Access to Courts JUDITH TOWNEND

Even the most casual observer of courts in the UK, whether through crime dramas or news reports, will likely have some awareness of the guiding principle that for justice to be done, it must be seen to be done. But if one digs beneath that popular and well-cemented phrase, understood as critical to the rule of law, it becomes clear that open justice takes many different forms, and is in practice a complex principle to administer and enforce. Clearly, the UK courts have a positive obligation in law to provide information about their activities, but to what extent?1 Just how much, and by what method and why, are questions that are contested and open for interpretation. This chapter uses the positive free speech framework and normative arguments forwarded in this collection to address these questions and to suggest how the judiciary and courts service should be publicly communicating what happens in court. Accepting the premise that there are positive duties on the state that ‘insist[s] that everyone be able to ­exercise their rights’2 in order to ‘support diverse speech environments’,3 and that positive free speech is a ‘central element existing alongside concerns about legal limitations on speech’,4 it proposes a way forward for further enhancing positive free speech in a court setting. The chapter first sets out an explanation of the way in which courts are accessed, at both the theoretical and the practical level. It then uses two case studies, a recent terrorism trial and a public inquiry, to examine further the way that courts are accessed in practice. Lastly, it considers the relevance of the government’s ‘open government’ and ‘open data’ policy agenda.5 When taken as a whole, this account suggests that there are existing negative and 1 This chapter focuses on courts in England and Wales and at a UK level (eg UK Supreme Court); it does not attempt to assess the situation in Scotland and Northern Ireland. All websites accessed in July 2019 unless otherwise stated. 2 S Fredman, Human Rights Transformed: Positive Rights and Positive Duties (Oxford, Oxford University Press, 2008) 30. 3 AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379, 391. 4 ibid 380. 5 This chapter develops case studies discussed in two previous articles: J Townend, ‘Leveson Online: A Publicly Reported Inquiry’ (2013) 10 Ethical Space 14 and H Irving and J Townend, ‘Censorship and National Security: Information Control in the Second World War and Present Day’ [2016] History & Policy, www.historyandpolicy. org/policy-papers/papers/censorship-and-national-security-information-control.

136  Judith Townend positive free speech elements to reporting and accessing courts in English law, but that the judiciary and courts service need to do more to protect an important positive liberty for all citizens.

I.  Free Speech in the Courts A.  Open Justice Principle in Theory and Practice The root of open justice runs deep, and it has grown and developed over centuries during which there have been dramatic technological and social shifts. In fact, it came about ‘almost by historical accident’ as an evolution of the public gatherings at court in the Middle Ages.6 Over time, the principle was succinctly articulated as an essential component of justice and an important means of ensuring a fair trial, which is protected under Article 6 of the ­European Convention on Human Rights (ECHR), as well as in the common law. The open justice principle has served a number of purposes relating to court access.7 For Robertson and Nicol, ‘trials derive their legitimacy from being conducted in public’.8 They, and other scholars, offer a range of associated benefits of this public justice: it discourages perjury and protects against judicial error; it protects the parties from an unfair trial; and it helps maintain public confidence in the courts. These factors can be described as part of the accountability rationale. Robertson and Nicol also suggest factors which are less well interrogated in judicial decisions: reporting ‘enhances public knowledge and appreciation of the workings of the law’, and it ‘permits the revelation of matters of genuine public interest’.9 These, I suggest, can be described as the education rationale, particularly important at a time when cuts to legal aid have created a growing number of litigants-in-person, and when many civil claims are resolved out of court owing to the high financial and personal cost of defending or bringing claims. The education justification is extended with the proposition, acknowledged by the courts in some instances,10 that the media needs to know particular details of cases to attract readers, and less tenably, to ensure news organisations’ financial survival.11 Finally, there is the deterrence rationale: the argument, usually made in debates over the naming of individuals, that publicity of proceedings deters the parties and the wider public from committing future offences; as part of this process, the public and press are able to judge the named individuals involved. This is often described as ‘naming and shaming’, although it is also arguable that deterrence can be achieved even with anonymisation. Open justice in the Middle Ages may have been motivated by the logic of this third category, but it is perhaps the least convincing rationale for complete openness of proceedings.

6 G Robertson and A Nicol, Media Law, 5th edn (London, Penguin, 2008) 463. 7 The open justice principle is so well embedded it is often accepted uncritically. A notable exception is the critique by J Jaconelli, Open Justice: A Critique of the Public Trial (Oxford, Oxford University Press, 2002). 8 Robertson and Nicol (n 6) 464. 9 ibid 464. 10 eg ETK v News Group Newspapers Ltd [2011] EWCA Civ 439, [13]. 11 See M Amos, ‘The Positive Right to Freedom of Expression and Party Anonymity in Legal Proceedings’, this volume, ch 8.

Positive Free Speech and Public Access to Courts  137 As  the  presiding judge found in a recent and rare case, when upholding the anonymity of the two convicted teenage defendants who were found guilty of murdering 39-year-old Angela Wrightson: ‘it is arguable that no further deterrence is necessary or, if it is, the naming of the individuals will add little to the fact that those responsible have been brought to justice, been convicted and been sentenced’.12 The rationales for open justice are wide-ranging, but the justifications for derogations from the principle – at least in theory – are more concise. According to the common law, for courts to hear a case in private, it must be deemed that there is a serious possibility that open proceedings would render impractical or frustrate the effective administration of justice: for example, by deterring prosecution of a case.13 This will trump any competing argument, including the public interest served by open reporting. Open justice, justified in the ways described above, is often perceived as meaning complete transparency: that is, everything that takes place in the courtroom should be made public. In practice, justice ranges from opaque to translucent to transparent, depending on the types of proceedings and court or tribunal – there are many shades of open justice. Opacity or translucence is achieved through a range of automatic and discretionary ­restrictions. In some cases, public access may be denied through the decision to move in camera, that is, into closed proceedings. This is common for parts of sensitive national security cases, for courts hearing privacy injunction applications and in courts such as the Investigatory Powers Tribunal and the Special Immigration Appeals Commission, which hold hearings with no press or public permitted. In some courts and cases, access may be allowed but reporting is restricted. By way of illustration, in the Court of Protection, a court which makes decisions on cases involving people deemed to lack mental capacity, a transparency regime permits regular access to the public and media with restricted reporting;14 in the criminal courts, sexual offence cases can be attended and reported, but victims of a wide range of sexual offences are given lifetime anonymity. A variation of this model allows the postponement of reporting of certain information: for example, only very restricted details of allocation and sending hearings in the magistrates’ courts can be reported at a preliminary stage. More problematically, in certain courts and cases, a more limited class of the public are allowed to attend. This is the model used in the family and youth courts, where accredited members of the media may attend private hearings closed to the public, and special reporting restriction regimes apply.15 Beyond this range of automatic restrictions, the court can postpone or prohibit

12 ‘Sentencing Remarks of Mr Justice Globe, Leeds Crown Court 7 April 2016’, [57], www.judiciary.gov.uk/ wp-content/uploads/2016/04/sentence-f-d-1.pdf. 13 See AG v Leveller Magazine [1979] AC 440; Guardian News and Media Ltd & Ors v Incedal [2014] EWCA Crim 1861, [17]. 14 A transparency pilot originally ran from 29 January to 31 July 2016, which was then extended and made permanent under Practice Direction 4C supplementing Part 4 of the Court of Protection Rules 2017, effective from 1 December 2017. 15 G Millar and A Scott, Newsgathering: Law, Regulation and the Public Interest (Oxford, Oxford University Press, 2016) [10.17]–[10.24]. Among other senior judges, the former President of the Family Division, Sir James Munby, has championed a perceived need to ‘open up’ the family justice regime. He has contended that ‘nothing short of radical reform will enable us to rid ourselves of the relentlessly repeated and inevitably damaging charge that we operate a system of private – some say secret – justice’: see J Munby, ‘The Family Justice Reforms’, speech delivered 29 April 2014, London, https://www.judiciary.uk/announcements/the-family-justice-reforms-remarks-by-sirjames-munby/.

138  Judith Townend other details – such as individuals’ names – being reported on a case-by-case basis. These automatic and discretionary restrictions may protect the administration of justice and a fair trial (by not prejudicing proceedings, for example), and rights associated with child welfare, and private and family life. As will be shown in the case studies, approaches to digital openness vary greatly, not only between, but also within, different categories of courts. Some courts have been far more proactive in publishing details of proceedings and decisions. Others rely almost entirely upon third party providers, such as the official law reports and the press, to communicate their proceedings. The predominant approach to open justice and access to information taken by the judiciary and Her Majesty’s Courts and Tribunals Service (HMCTS), an executive agency of the Ministry of Justice, is protection of a neutral or negative principle, one of non-interference: non-censorship of proceedings (most courtrooms can be visited in person), and a judge may lift restrictions upon application of a third party, such as a newspaper. A positive liberty is also present, if often secondary: in many instances, there is a positive application of open justice, in which the court proactively notifies parties of potential restrictions and publishes details of its work online. Obligations to safeguard the identity of witnesses or journalistic sources may also be seen as part of a positive liberty for those it protects. However, this positive approach, in which the court perceives a positive obligation to notify and inform, is not consistent or reliable across the courts service, and is vulnerable to error and oversight.16 The notion that the public should be able to attend court, or have proceedings reported to them via a proxy such as the press, accords well with the freedom of expression concept under Article 10 ECHR, which provides a public right to receive and impart information and ideas.17 Arguments in favour of increased access have often relied on Article 10. It is, however, the common law principle of open justice which has proved the more reliable weapon in the arsenal in a UK context, as there are limits to how Article 10 ECHR and its domesticated version under the Human Rights Act have been used positively to protect free speech.18

B.  Right to Report Courts Most usually, Article 10 arguments have been articulated with regard to the right to report court proceedings. This builds on Lord Denning’s understanding of press reporters as the ‘watchdog of justice’, that ‘every member of the public must be entitled to report in the public press all that he has seen and heard’.19 This remark, made in 1955, seems oddly broad for its time: the choice of the word ‘report’ rather than ‘read’ corresponds with the ‘­imparting’ component of Article 10, rather than the ‘receiving’ element. At that time, very few members

16 This is an issue that has been dealt with extensively in the family courts and Court of Protection. See, eg A Healthcare NHS Trust v P & Q [2015] EWCOP 15. 17 See Amos, this volume, ch 8, who discusses the courts’ treatment of freedom of expression in relation to party anonymity as directly linked to the principle of open justice and also as a freestanding right. 18 For a fuller discussion of the European and UK jurisprudence, see A Scott and A Burke, ‘The Access to ­Information Dimension of Positive Free Speech’, this volume, ch 4; Amos, this volume, ch 8. 19 A Denning, The Road to Justice (London, Stevens & Son, 1955) 64.

Positive Free Speech and Public Access to Courts  139 of the public would have the means of communicating what they had observed in the public press. Reading it in the twenty-first century, it makes more sense: it suggests that anyone attending court has an entitlement to report the proceedings – which they can now easily do, via their electronic device on freely available social media and blogging platforms. Whether that would have been Lord Denning’s intention is open for discussion. Despite Lord Denning’s broad understanding of this public entitlement, court reporting has largely been confined to a specialist – if diminishing – breed of individuals in the twentieth and twenty-first centuries: the reporters producing the official law reports; and journalists employed by agencies, magazines, newspapers and broadcasters as court reporters and legal correspondents. Beyond this, reporting may be said to include any lists, results and transcripts published by the courts or a private provider, and written judgments, increasingly made available freely online. Online technology has allowed lawyers increasingly to participate in reporting, whether through news items and commentary on their firm’s or chamber’s website and blogs, or on their own personal blogs. It has also facilitated members of the public to report court proceedings directly, without requiring the cooperation of a professional outlet. Despite this liberating technology, court reporting has remained oddly traditional, with a few exceptions, such as the freelance journalist Peter Jukes’s remarkably committed live tweeting of the phone hacking trials for seven months in 2014. The Criminal Practice Directions on tweeting from court make a distinction between representatives of the media and ‘legal commentators’ on the one hand,20 and ordinary members of the public on the other.21 The latter category must seek special permission to tweet, whereas the former may tweet from open court proceedings without making a specific application. Nonetheless, a surge in regular citizen reporting is yet to occur, perhaps because it would be time-consuming and expensive to undertake for someone not employed to do so. It is worth noting that Jukes’s tweeting was facilitated by a well-publicised crowd-funding initiative. A more dramatic social development, which is proving an ongoing challenge for the court and the Attorney General’s Office, is the social media interaction around the trial, whether involving parties and their families or the broader public. Much of this social media conversation is harmless, but some publications on social network services can threaten to prejudice a trial to a serious degree. In the original trial of two anonymised teenage girls for the murder of Angela Wrightson, for example, prejudicial Facebook activity around the case led to a retrial and a wide-ranging reporting restriction preventing media organisations from linking or posting reports of the case on social media.22 Despite these events, a response by the government to its consultation on the impact of social media on criminal trials claimed that social media does not currently pose a threat to the criminal justice system and that the aforementioned case was not representative.23 More recently, Stephen 20 The Lord Chief Justice was perhaps thinking of legal bloggers when drafting the original guidance, although the courts do not yet appear to have considered the definition of ‘legal commentator’, which could be variously interpreted. 21 Criminal Practice Directions 2015 [2015] EWCA Crim 1567. See generally Millar and Scott (n 15), [10.80]–[10.83]. 22 R (on the application of British Broadcasting Corporation and others) v F & D [2016] EWCA Crim 12. 23 Attorney General’s Office, ‘Response to Call for Evidence on the Impact of Social Media on the Administration of Justice’ (5 March 2019), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/ attachment_data/file/783627/Call_For_Evidence_05.03.2019_v2.pdf.

140  Judith Townend Yaxley-Lennon, the former leader of the English Defence League, better known as Tommy Robinson, was found to have committed a contempt of court after he breached a reporting restriction imposed under section 4(2) of the Contempt of Court Act 1981, by livestreaming a video from outside the public entrance to the court and aggressively confronting and filming some of the defendants as they arrived at court.24

C.  Right to Receive Court Information As in other discussions on free expression, the right to receive information is often overlooked in favour of a preoccupation with a right to report. This can be partly explained by the courts’ passive and reactive approach. Challenges to restrictions are usually brought by media organisations with the financial and legal resources to do so, and their concern tends to be with their right to report what they have observed in court and material that they already hold. On occasion, however, the right to information has been explicitly asserted. In GNM v Westminster Magistrate’s Court, the newspaper sought access to documents in extradition proceedings, partly relying on Article 10. Although the appeal court allowed access (breaking ‘new ground’ in the application of the open justice principle, though not the nature of the principle itself), Lord Justice Toulson’s leading judgment was based on the common law rather than European Court of Human Rights (ECtHR) jurisprudence, although he acknowledged that the latter may be seen as supportive of the overall reasoning.25 The judgment also illuminates the lack of clarity from the ECtHR on a right to information; as noted elsewhere in this collection, ECHR jurisprudence is evolving on the question of a positive obligation to provide information.26 In November 2016, the Grand Chamber of the ECtHR held back from recognising a freestanding right to receive information, confirming that ‘the right to receive information cannot be construed as imposing on a State positive obligations to collect and disseminate information of its own motion’. It did, however, recognise that a positive obligation arises in two types of cases: first, where disclosure of the information has been imposed by an enforceable judicial order; and secondly, where access to the information is instrumental for the individual’s exercise of his or her right to freedom of expression and where its denial constitutes an interference with that right.27 In determining whether a denial of access to information constitutes an interference with an applicant’s freedom of expression rights, the following criteria should be considered in a given case: (i)  the purpose of the information request; (ii) the nature of the information sought; (iii) the role of the applicant; and (iv) whether the information is ‘ready and available’.28 While UK challenges such as GNM v Westminster Magistrate’s Court and Kennedy v ­Charity Commission29 could be placed within the ‘negative free speech’ category as attempts by individuals and organisations to remove an obstacle (unsuccessful in the case of Kennedy),

24 HM Attorney General v Yaxley-Lennon [2019] EWHC 1791 (QB), [4]. 25 R (on the application of Guardian News and Media Ltd) v City of Westminster Magistrates’ Court [2012] EWCA Civ 420, [88]. 26 See Scott and Burke, this volume, ch 4; Amos, this volume, ch 8. 27 Magyar Helsinki Bizottság v Hungary, app no 18030/11 (ECHR, 8 November 2016) [156]. 28 ibid [157]–[170]. 29 Kennedy v Charity Commission [2014] UKSC 20.

Positive Free Speech and Public Access to Courts  141 they could also lead to a positive obligation on courts and other organisations to provide certain information. In this way, GNM has led to a ‘default position’ in which the media is entitled to have access to documents that have been placed before a judge and referred to in the course of proceedings.30 Prior to this case, access to certain court documents was possible, relying on the Criminal and Civil Procedure Rules, but this ruling extended and embedded the position, proving practically useful for journalists in particular. Most recently and importantly, and further entrenching the decision in GNM, the UK Supreme Court has upheld a Court of Appeal ruling that it was within a court’s inherent jurisdiction to allow non-parties inspection of a wide range of documents placed before the court or referred to in civil proceedings.31 The availability of documents does not guarantee access to all the British public, however. To give a specific example, Part 5 of the Civil Practice Directions permits members of the public, for a prescribed fee, to obtain statements of case in civil litigation from civil courts, including the Queen’s Bench of the Royal Courts of Justice.32 However, a number of obstacles have meant that this available data has not been readily accessible, though the introduction of electronic filing in 2019 has somewhat improved access.33 Most importantly – and this remains so under the new system – the costs are prohibitively high, especially if an ordinary member of the public, or small organisation, wishes to access more than one or two claims at a time. For claims lodged before the introduction of e-filing,34 the court must be visited in person, which may inhibit requesters with a disability, or those unable to travel to London during office hours because of other commitments or expense. For older claims made prior to the introduction of e-filing, a final obstacle remains in locating a claim number with which to make an application for case documents; this process involves searching through a large ring binder of case listings, unless the number is retrieved by another means.35 I have previously remarked that it is as if this court’s data, publicly available by law, is stored in an open filing cabinet with no drawer handles or labels.36 The e-filing system improves this to some degree, but costs remain high, for no obvious reason, given that the court staff are no longer burdened with the physical labour of locating and copying documents. The availability of orders that restrict reporting is another concern for journalists and other actors. Plans to create a database of such orders for the media were abandoned owing to reportedly ‘eye-watering’ costs that would be charged to the media by a private contractor.37 As with the statements of case, court orders should be publicly accessible on

30 ‘Reporting Restrictions in the Criminal Courts’ (Judicial College, 2015) 29, www.judiciary.gov.uk/wp-content/ uploads/2015/05/reporting-restrictions-guide-2015-final.pdf. 31 Cape Intermediate Holdings Limited v Dring [2019] UKSC 38; Cape Intermediate Holdings Ltd v Dring (Asbestos Victims Support Group) [2018] EWCA Civ 1795. 32 ‘Practice Direction 5a – Court Documents – Civil Procedure Rules’, www.justice.gov.uk/courts/procedurerules/civil/rules/part05/pd_part05a. 33 Lord Burnett, ‘Speech by Lord Chief Justice: Launch of Electronic Filing in the Queen’s Bench Division’ (15  April 2019), www.judiciary.uk/publications/speech-by-lord-chief-justice-launch-of-electronic-filing-in-thequeens-bench-division/. 34 According to communication with court personnel. 35 It may be known and shared by the parties of a case or accessed via a paid-for service, such as Lawtel. 36 J Townend, ‘Closed Data: Defamation and Privacy Disputes in England and Wales’ (2013) 5 Journal of Media Law 31. 37 Anon, ‘Law Commission Calls for Reporting Restrictions Website with Paid-For Access’, Press Gazette, 26 March 2014.

142  Judith Townend request, but this usually requires a physical visit to the court. Furthermore, if a journalist or member of the public has not attended every day of a trial, they may be unaware of any discretionary reporting restrictions that have been put in place (they would be expected to know the automatic ones, which raises a different issue about access to legal education in this regard). In 2014, the Law Commission recommended the creation of a portal where members of the public would be able to find out if a postponement order under section 4(2) of the Contempt of Court Act 1981 had been made.38 No such system has yet been developed, despite the government’s financial commitment to digital court services and the ongoing programme of court reform and digitalisation,39 and in any event it would be partial only in terms of its coverage of the range of discretionary restrictions that may be imposed. The small press office for the judiciary supplies journalists with reporting restriction orders from time to time. It cannot cover all UK courts’ activity, however, and may not consider inquiries from other members of the public – such as researchers and legal commentators – with a legitimate desire to report court proceedings. To give one example, Lucy Reed, a specialist family barrister and blogger, who writes for a range of media and legal publications, but who is not an ‘accredited’ journalist as understood by the family courts, found it extremely difficult to access the reporting restriction orders associated with a series of family judgments related to the high-profile murder of a child.40 The judgments had been published online, but had subsequently been removed. Helpfully for Reed, they were made available by the judiciary press office on request, along with details of a reporting restriction order that had been imposed on some, but not all, of their contents. She wished to check whether any other relevant reporting restriction orders were in place before writing about the judgments. She was advised to contact the court directly; it replied, stating: ‘all family proceedings are confidential, unless you are a party in the matter we cannot disclose any information to yourself ’. It advised her to make an application to the court to seek the information; this is a process that would likely incur a fee of around £200. She commented on the failings of a system that did not allow non-parties or non-accredited journalists to check the reporting restrictions in a case that was the subject of much public debate: ‘what about the self-employed journalist or the blogging citizen journalist … how are responsible bloggers to know what is and is not prohibited?’41 Happily for Reed, following her charity’s application to the Family Procedure Rules Committee, the court has now introduced a pilot which extends access to a limited class of legal bloggers, including qualified lawyers in practice, working for an educational charity or employed by a higher education institution.42 There is similarly mixed practice around the supply of court lists. Some lists are made available through a private service called Courtserve, either for free or for a ­subscription 38 Law Commission, Contempt of Court (2): Court Reporting, Law Com No 344, HC 1162 (London, Stationery Office, 2014). 39 See www.gov.uk/guidance/the-hmcts-reform-programme. 40 This account is based on personal correspondence with Reed and the Transparency Project (of which I am a member) and Reed’s blog, www.pinktape.co.uk. 41 L Reed, ‘What Price Transparency?’, Pinktape (blog), 25 June 2016, www.pinktape.co.uk (original emphasis). 42 Author disclosure: as a member of the Transparency Project charity core committee, I was involved in the drafting of this proposal. See www.justice.gov.uk/courts/procedure-rules/family, in conjunction with Family ­Practice Direction 36J – Pilot Scheme: Transparency (Attendance at Hearings in Private).

Positive Free Speech and Public Access to Courts  143 fee. Some courts, such as the Royal Courts of Justice, publish their cause lists online. ‘Full’ lists, which contain names, addresses and full charges, are made available to journalists by individual courts, and such access is ensured by a protocol agreement between HMCTS and two media representative groups, the News Media Association and the Society of Editors.43 These full lists may be made available to local independent bloggers as well as those working for traditional news organisations, but there is no consistent cross-court policy, which HMCTS attempted to address with new guidance issued in 2018.44 In October 2013, amendments to Criminal Procedure Rules 2013 to permit online publication of all court lists were approved by the Criminal Procedure Rule Committee, but this has not yet resulted in any major change to practice.45 With regard to the outcomes of cases, no substantial collated results data is made available to either journalists or the wider public. As will be discussed below, if such data were made available to the public, important legal and ethical questions would need to be addressed through public consultation. Some internal collation does take place which informs regular statistical publications, but the data available publicly is very limited – so much so that even the Ministry of Justice had to ‘estimate’ the number of defamation trials that had taken place in a given year for its impact assessment on the Defamation Bill in 2012.46 Beyond this, written judgments are either published on courts’ or the main judiciary website and/or on the third party charitable site BAILII. Not all courts have consistent procedures on this, and moreover many criminal and civil cases – especially those in the lower courts – do not result in a written judgment at all. Sentencing remarks in criminal cases may be made publicly available by the judiciary press office, particularly in cases attracting significant media attention, but again there is no consistent practice on this. A notable exception is the UK Supreme Court, which, as well as sending its judgments to BAILII, tracks all the stages of its cases online. The Freedom of Information Act 2000 would have provided an ideal opportunity to put a carefully drafted positive obligation into statute, and to clarify the type of court data that should be proactively disseminated, but in fact the opposite occurred. Section 32 of the Act provides a complete exemption for court records, meaning the legislation – with its positive obligation on public authorities to provide information to the public – permits only a limited type of information to be requested about judicial and court processes and activity. To summarise, there is a positive duty for the judiciary and courts service to make available various court materials to the public, but in practice this is not sufficiently fulfilled. The public’s right to information is hampered by cumbersome and expensive systems that require physical access to court. Furthermore, complicated systems of hybrid access which usually favour journalists over other types of requesters are inadequate in a contemporary digital publishing environment. The failings of the system were exposed dramatically in the trial of Erol Incedal. 43 G Vassall-Adams, Reporting Restrictions in the Criminal Courts (London, Judicial College, 2015) 29. 44 See HMCTS, ‘Guidance to Staff on Supporting Media Access to Courts and Tribunals’ (24 October 2018), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/750526/ HMCTS_Media_Guidance.pdf. 45 Vassall-Adans (n 43) 29. 46 ‘Impact Assessment’, IA No MOJ 145 (Ministry of Justice, 1 April 2012) [2.230], 51, www.parliament.uk/ documents/impact-assessments/IA12-009.pdf.

144  Judith Townend

II.  Case Study in ‘Negative Free Speech’: The Incedal Case In 2013, a journalist spotted an unusual advance notice of a reporting restriction on the court noticeboard. If granted, it would allow an entire criminal trial to be conducted behind closed doors, with no press or public present. The media mobilised a joint response to the court, although the presiding judge, Nicol J,47 decided to hold the entire trial in camera. Such measures, deemed necessary to protect the effective administration of justice and national security, would be enforced by court order and ministerial certificates.48 A consortium of 12 media organisations appealed. The Court of Appeal agreed with the presiding judge that there was ‘a significant risk – at the very least, a serious ­possibility – that the administration of justice would be frustrated were the trial to be conducted in open Court’.49 While maintaining restrictions, it relaxed them in an unusual and hitherto unknown arrangement, with three levels of access dependent on the nature of daily proceedings. Stage 1, as it was referred to in court, was in open court, and allowed the public and the media to attend and report freely (in sessions including the swearing in of the jury, the judge’s introductory remarks, verdicts and sentencing). Stages 2 and 3 were heard in camera, with the judge and court staff, the jury, the defendant, police, prosecution and defence lawyers, and – for Stage 2 only – also a small number of ‘accredited’ journalists present. The accredited journalists were permitted to take notes during the Stage 2 p ­ roceedings, but could not carry those notes out of court. Journalists have described how they were asked to lock their mobile phones inside a soundproof box and put their ‘secret’ notebooks in a safe. Additionally, they were not allowed to disclose what they had observed except to nominated legal representatives for the purpose of obtaining legal advice. According to The Guardian’s Ian Cobain, journalists were instructed that this should only take place in a closed room in which mobile phones were switched off, with no notes taken.50 Importantly, the court also allowed the defendants to be named as Mounir RarmoulBouhadjar, who was charged with possessing a terrorist document (count three), and Erol Incedal, who faced the same charge (count two) and another of preparing an act of ­terrorism (count one).51 The unusual reporting arrangements commenced in October 2014. The second defendant, Mounir Rarmoul-Bouhadjar, pleaded guilty to possessing a terrorist document. At trial, Erol Incedal was found guilty of the same offence, but the jury failed to reach a verdict on the other charge. He was eventually acquitted at a second trial, held under the same reporting arrangements.52 47 The same Nicol who co-authored Media Law (n 6), a text which makes a powerful case for the benefits of open justice. The irony was made much of by the columnist Nick Cohen, who wrote: ‘Secret Justice Will Lead to Paranoia – and It’s Not Very British’, The Guardian, 14 June 2014. 48 Such certificates are often understood as part of the law on public interest immunity (PII), but, as contended by Paul Scott, they are in fact used as a distinct mechanism for closing court from public view, ‘part of a body of case law on in camera trials which has at times struggled to differentiate itself from that on PII’: PF Scott, ‘An Inherent Jurisdiction to Protect the Public Interest: From PII to “Secret Trials”’ (2016) 27 King’s Law Journal 259. 49 Incedal (n 13) [31]. 50 I Cobain, ‘Why Is the Crux of the Incedal Case a Secret? You’re Not Allowed to Know’, The Guardian, 26 March 2015. 51 Counts two and three: an offence contrary to the Terrorism Act 2000, s 58 (collection of information); and count one: an offence contrary to the Terrorism Act 2006, s 5 (preparation of terrorist acts). 52 Rarmoul-Bouhadjar and Incedal were sentenced to imprisonment for three years and three and a half years, respectively.

Positive Free Speech and Public Access to Courts  145 There is no official record of how the trial divided into each stage, but the BBC reported that 10 hours of evidence were heard in public, 28 hours were heard in private with ‘accredited’ journalists present, and 30 hours were heard in private without any press or public attending.53 This three-tier system led to the case being widely characterised by media organisations as a ‘secret trial’. Despite having privileged access, the ‘accredited’ journalists made clear their discomfort with the nature of the trial in articles and tweets. There was also concern about the manner in which security was controlled. The London Evening Standard claimed that one journalist ‘was stopped from leaving court even though he had taken no notes’, and that ‘a second was threatened with arrest at his home when he had not even been in court’.54 The reasons for the decisions taken during the Incedal case remain unclear, as The Times illustrated in its report on the verdict of the second trial, in a dramatic fashion: ‘a law student jailed yesterday for possession of a bomb-making manual was a ■■■■■■■■■■■■■■■■’.55 The media has not been able to report the prosecution’s main argument, the evidence that led the jury to find Incedal not guilty of preparing an act of terrorism, or the reason why these matters have been concealed. Following the trial, various organisations sought to report the ‘core issues’ in the private sessions, but their initial applications were rejected. This decision was upheld in the Court of Appeal in February 2016, with the court finding that a departure from the principle of open justice was necessary for justice to be done. While supportive of the trial judge’s decision, the justices raised concerns about procedure adopted, and also about the deficiency of information on which they had to rely. They observed that normally they would look to previous decisions to determine issues which were similar. In this case, however, relevant ‘closed judgments’ were ‘not retained within the court files or, as far as we have been able to ascertain, in any specified place within the court’. The judges requested that a working group be set up to advise the court on further action; a Practice Direction on closed judgments has since been published specifying a process for the retainment of closed judgments in certain types of proceedings.56 The case was characterised by an approach that emphasised ‘negative free speech’. It saw the courts do little to facilitate public access and understanding. Indeed, events surrounding this case exposed systematic issues in the courts’ managing of access. Access was only secured on the intervention by media organisations and with the court lifting its initial restrictions, rather than the court proactively considering and protecting speech from the outset. The case also highlighted three key practical issues. First, there is no coordinated and systematic administrative procedure that guides the publicising of reporting restrictions and their application. Although notice of the restriction was advertised on a noticeboard, general knowledge of what was proposed was still reliant on a journalist being present in

53 D Casciani, ‘Erol Incedal: The Trial We Couldn’t Report’, BBC News, 26 March 2015. 54 P Cheston, ‘Secret Terror Trial Ends in Farce as Student Is Cleared of Targeting Blair’, London Evening­ Standard, 26 March 2015. 55 S O’Neill, ‘Terror Trial Evidence Must Remain Secret’, The Times, 2 April 2015. 56 Guardian News and Media Ltd & Ors v R & Incedal [2016] EWCA Crim 11; Lord Chief Justice and Senior P ­ resident of Tribunals, ‘Practice Direction – Closed Judgments’ (14 January 2019), www.judiciary.uk/ announcements/practice-direction-closed-judgments/.

146  Judith Townend court and alert to the irregularity of the situation. It appears that there was no proactive notification of any media organisation through other channels in this case. Second, decisions can be arbitrary. The resource and experience of legally represented media organisations allowed them to bring a challenge and to secure ‘accredited’ status for their journalists. The court failed to consider the rights of smaller forms of media, such as specialist and online-only platforms. Furthermore, it may have been valuable to have non-media observers present during the Stage 2 proceedings, who would play a different watchdog role in holding the judicial process to account and whose reporting would not have been constrained by a media organisation’s agenda (which is necessarily guided by the newsworthiness of particular details). Lawrence McNamara has suggested that there is a strong case for giving legal professional bodies and specialist NGOs a similar status as independent observers.57 Thirdly, there is a paucity of material on transparency decisions themselves, with previous cases not properly documented in the court system. (This also occurred with the granting of so-called super injunctions and anonymised privacy injunctions before a new procedure was introduced in 2011 following a report by the Master of the Rolls.58 It has since transpired, however, that the data is not necessarily complete, and the collection system is being reviewed by the judiciary.59) The way in which the publicity of this trial was handled has been widely criticised, and not only by the media. The Lord Chief Justice, Lord Thomas, said that ‘there ought to be very much clearer guidelines and rules’ in future. He stated that the press should be able to ‘see such material as can be shown to them so they know what the argument is about’ and hoped that a ‘proper way of dealing with [anonymised defendants]’ could be developed.60

III.  Case Study in ‘Positive Free Speech’: The Leveson Inquiry In contrast to court and tribunal proceedings, public inquiries have sometimes taken a much more proactive approach in protecting judicial and executive accountability and supplying information to a public audience. More could be done to improve the formats of digital evidence and reports to make them more searchable and accessible to public audiences, but in general they show how judicial proceedings can be shared with a wide audience, albeit that public inquiries are distinct from other types of court proceedings and instigated to investigate particular events that have caused or are capable of causing public

57 L McNamara, ‘Secret Trials – a Little Transparency, a Lot to Worry About’, UK Human Rights Blog, 12 June 2014, https://ukhumanrightsblog.com/2014/06/12/secret-trials-a-little-transparency-a-lot-to-worry-about-lawrencemcnamara. 58 Committee on Super-Injunctions, Super-Injunctions, Anonymised Injunctions and Open Justice (London, Master of the Rolls, 2011). 59 J Townend, ‘Where Did All the Privacy Injunctions Go? A Response to the Queen’s Bench “Media List” Consultation’, Inforrm’s Blog, 31 May 2017, https://inforrm.org/2017/05/31/where-did-all-the-privacy-injunctionsgo-a-response-to-the-queens-bench-media-list-consultation-judith-townend. 60 ‘Transcript of Lord Chief Justice’s Annual Press Conference 2014’ (12 November 2014) 4, www.judiciary.gov. uk/wp-content/uploads/2014/11/lcj-transcript121114.pdf.

Positive Free Speech and Public Access to Courts  147 concern. The Leveson Inquiry, established by the Prime Minister, David Cameron, in July 2011, broke new ground in the way it was communicated online, and was the first inquiry under the Inquiries Act 2005 that properly utilised online space. It was also the first to be accompanied by a continual stream of third party social media commentary.61 The Leveson Inquiry was bound by the terms of section 18 of the Inquiries Act 2005, which states that an inquiry must allow public access to inquiry proceedings and­ information: (1) subject to any restrictions imposed by a notice or order under section 19, the chairman must take such steps as he considers reasonable to secure that members of the public (including reporters) are able: (a) to attend the inquiry or to see and hear a simultaneous transmission of proceedings at the inquiry; (b) to obtain or to view a record of evidence and documents given, produced or provided to the inquiry or inquiry panel.

A narrow reading of section 18(1) could mean merely facilitating physical access at the court rather than via a website. Lord Justice Leveson’s approach indicated a more expansive interpretation, with timely and extensive communication of the inquiry online. A bespoke website made a range of material available during the inquiry’s proceedings: information about the inquiry, rulings by Lord Justice Leveson (on core participants or admissibility of evidence, for example), transcripts of evidence and written evidence (once formally read in). Most importantly, proceedings could be watched live or at a later date online, except for a few rare examples in which anonymity or special provisions applied. This opened the hearings to people unable to travel to court. The courtroom itself had limited public and media access, and so an overflow Annexe (a marquee in the Royal Courts of Justice grounds) was opened, with sections for the media and the public to watch the proceedings from several screens. This had both pros and cons: while it allowed journalists to work in perhaps a more comfortable and relaxed environment, and view a scrolling transcript, they lost court-side immediacy. Journalists were also limited by the fact that they did not have access to the documents as they were discussed in the inquiry. The document appeared on the screen in the Annexe, but as one reporter covering the inquiry explained, ‘you were trying to take down the document at the same time as you were trying to listen to people being questioned’.62 The inquiry was also communicated via Twitter, although not in an official capacity; the inquiry’s own Twitter account only published one update between July 2012 and the report’s publication in November. It was used for posting notifications about updates to the website, rather than interacting with online followers. The surrounding conversation was diverse and interactive. Largely uninhibited by the strict contempt restrictions that would generally apply in court proceedings, a range of commentators – including campaigners, journalists, bloggers and academics – reported and commented on what was taking place. Sometimes this activity played a key part in extending the ‘public watchdog’ role of the

61 While the Sir John Chilcot’s inquiry into the Iraq War, which heard evidence from 2009 to 2011, was also broadcast live and tweeted about, it was not a public inquiry under the Inquiries Act 2005. 62 W Turvill, ‘Reporting the Reporters: Interviews with the Journalists Who Covered Leveson’, Press Gazette, 6 November 2012.

148  Judith Townend press, drawing attention to key passages and sharing important information in the public interest. At other times, it was flippant and trivial. For example, one of the inquiry counsel gained ­notoriety – partly due to further promotion by the mainstream media – during an episode in which Twitter users mocked her for apparently gazing at witness Hugh Grant during his evidence, tagging their updates ‘#womanontheleft’.63 The Leveson Inquiry recognised the distinct role of reporters and journalists, but also the part of the broader public audience. This reflected section 18(1) of the Inquiries Act 2005, the wording of which indicates that reporters are a special sub-set of the public. In the inquiry, reporters were given special privileges: a small number had guaranteed access to press seats in courtroom 73, a portion of the Annexe was reserved for the press, and they received communications from a dedicated press officer. It is likely that the timing of the report’s release on 29 November 2012 to a room of invited journalists was chosen to suit a broadcast media schedule. Perhaps to avoid partisan national newspapers dominating the coverage with their take on the findings, Leveson LJ announced the report in a statement at lunchtime, which gave the broadcasters, regulated by rules of impartiality, a chance to run their reports in the afternoon and evening television and radio news bulletins, alongside the newspapers’ online coverage but ahead of the next morning’s front pages. In these ways, priority was given to national news media actors, but this was not at the cost of public access. After the report was released, hyperlinks to the report’s four volumes and the executive summary were published on the website. Since the conclusion of part one of the inquiry (the government has decided that part two will not proceed), the website materials and video have been preserved by the National Archives. Furthermore, on more than one occasion, wider public – rather than institutional media – interests were given the greater weight. When media organisations challenged the admission of evidence from anonymous journalists represented by the National Union of Journalists, Lord Justice Leveson found in favour of the latter’s Article 10 rights.64 The online communication – both during and following the Leveson Inquiry – is to be commended, although there are ways in which it could have been made even more accessible. The transcripts and report were published in cumbersome non-machine-readable formats, which limited their accessibility. Third parties have attempted to make the vast and rich outputs more useable, although this has required a significant investment of time. A service offered by the civic organisation ‘mySociety’ has produced search-friendly HTML versions of the transcripts, which cross-reference related material.65 This allows, for example, a user easily to search all the oral evidence for a particular phrase (eg ‘data protection’) or by witness, which would not be possible with the original formats. Robert Sharp, a writer and free speech campaigner, also converted the report into HTML to allow better functionality; his version cross-references parts of the multi-volume report with hyperlinks, allowing for improved reader comprehension and navigation.66 A third offering is Discover Leveson, a searchable public archive, published by the journalism department at Kingston University.67



63 See

Townend (n 5) 19. (on the application of Associated Newspapers Ltd) v Rt Hon Lord Justice Leveson [2012] EWHC 57 (Admin). 65 Available at leveson.sayit.mysociety.org. 66 Available at http://leveson.robertsharp.co.uk. 67 Available at https://discoverleveson.com/. 64 R

Positive Free Speech and Public Access to Courts  149 Unlike the Incedal trials, which took place two years later, the Leveson Inquiry was predominantly characterised by an approach of ‘positive free speech’, where access was not only secured by the intervention of third parties challenging restrictions; the inquiry proactively considered, promoted and protected speech from the outset. Although the evidence and report could have been made even more accessible to online readers, the chair considered public access beyond the terms of the Inquiries Act, and gave the public audience an opportunity to access materials directly without relying on a mainstream media proxy. In this way, public accountability and the public right to receive information was not only served through passive policies of access, or reactive decisions. It is important to note, however, that Lord Justice Leveson went beyond what he was required to do by statute. The openness of an inquiry is still very much at the discretion of the chair, and a less diligent chair may not have made such great efforts to publicise materials effectively. Clearer guidance for inquiries would give better protection to this type of ‘positive free speech’. It would help inquiry chairs navigate questions over digital publications: for example, the nature of material that should be published and when it should be published became important questions during the Azelle Rodney Inquiry in 2012.68 It could also recommend publication methods to inquiry chairs that would enable more userfriendly formats than enormous PDFs, and make explicit the arrangements for archiving proceedings in the National Archives. Finally, it should be remembered that a public inquiry is distinct from ordinary court proceedings, and this has aided the greater openness of inquiry proceedings. A public inquiry will have greater financial resources to allow the appointment of a dedicated press office and special audiovisual and online technology arrangements. Additionally, while an inquiry may be alert to any potential contempt arising from the evidence (affecting an ongoing related trial, for example) it is free from many of the contempt of court issues that must be dealt with in the courts when publicising their activities.

IV.  Open Data from the Courts The obligation to open proceedings can be understood as a positive measure, but as the case studies above illustrate, this can take many different forms, depending on the type and particular circumstances of the proceeding, the attitude and interpretation of the presiding judge, and interventions by the parties involved. The criteria for openness are somewhat hazy. How might more clarity be achieved? One solution might be to look to the government’s policy on open government and its standards on open data. Four UK Open Government National Action Plans have promised to improve public access to data. The government committed to identify and publish, in the period 2016–18, core data assets to create ‘a high quality national information infrastructure, making government data more secure and easier to find, store and access’, and to make better use of data assets. It promised to ‘encourage and support data-driven techniques in policy and service delivery across government departments and encourage the better 68 Transcript of the oral hearing of 4 September 2012, webarchive.nationalarchives.gov.uk/20150406091509/ http://azellerodneyinquiry.independent.gov.uk/transcripts/269.htm.

150  Judith Townend use of open data in the economy and civil society’.69 Its latest plan ‘recognises the need to deepen previous commitments and improve the quality of the data that has been released’.70­ Separately, the Ministry of Justice’s Single Departmental Plan for 2015–2020, now withdrawn, promised to: ‘open up as much data as possible overall across the MOJ and the National Offender Management Service to enable researchers, academics, charities, the media and the wider public understand what is happening in the criminal justice system’.71 Notably, this specific commitment is not present in the latest Single Departmental Plan; its ambition for data and transparency is drafted in vaguer terms.72 On the surface, these strategic ambitions seem ideal vehicles through which to enforce a positive free speech agenda. A commitment to open data means publishing data that can be accessed, shared and reused; such publication should be made ‘in an open format, [that] is machine readable and is published under a license that allows for free reuse’.73 Specific court data could be opened up in these formats under such an obligation, releasing it from the control of private third parties and allowing the public to access data directly without relying on imperfect media proxies. For some data, this might be the right approach. If it is appropriate for the data to be accessible through search engines and to be reused by any third party – corporate or nonprofit – then the courts and Ministry of Justice should certainly consider collecting and publishing data as open data. However, it may be inappropriate for some forms of nonanonymised data. For example, if HMCTS started to collate complete and accurate sets of listings and results data in the criminal courts, which may include individuals’ names and addresses, should this be published as open data? Offender rehabilitation rights and the so-called ‘right to be forgotten’, or erasure, recognised in data protection law, would need to be seriously considered before doing so. There is a valid discussion to be had on what types of data should be open data, and whether some data sets should be licensed under particular conditions. In other words, the data could be made accessible without being published as open data. There may be alternative licensing regimes for data sets, or ways of centrally controlling data access that would allow appropriate access control in the context of some court data. Whatever formats are determined as appropriate for different data sets, there is a timely opportunity for government and the judiciary to improve and overhaul the gathering and collection of court data with a view to different forms of publication, as the government moves forward with its national plan.74 69 UK, Cabinet Office, ‘UK Open Government National Action Plan 2016–18’, policy paper (12 May 2016), www. gov.uk/government/publications/uk-open-government-national-action-plan-2016-18/uk-open-governmentnational-action-plan-2016-18. 70 Department for Digital, Culture, Media and Sport, ‘UK National Action Plan for Open Government 2019–2021’ (12 June 2019), www.gov.uk/government/publications/uk-national-action-plan-for-open-government-2019-2021/ uk-national-action-plan-for-open-government-2019-2021. 71 Ministry of Justice, ‘Single Departmental Plan: 2015 to 2020’ (19 February 2016), www.gov.uk/government/ publications/moj-single-departmental-plan-2015-to-2020/single-departmental-plan-2015-to-2020. The specific mention of ‘criminal justice’ is significant. Civil matters have been given much less attention than crime in the government’s agenda on open data. 72 Ministry of Justice, ‘Ministry of Justice Single Departmental Plan 2019–2022’ (27 June 2019), www.gov.uk/ government/publications/ministry-of-justice-single-departmental-plan/ministry-of-justice-single-departmentalplan-3. 73 A Acuña, ‘A Simple Intro to Open Data’ (4 November 2013), https://data.blog.gov.uk/2013/11/04/a-simpleintro-to-open-data/. 74 Author disclosure: as a consultant to the NGO Spotlight on Corruption, a member of the Open Government Network, I have advised on options for a draft commitment to open justice in the next action plan (2021–2023).

Positive Free Speech and Public Access to Courts  151

V.  Conclusion: Rethinking Access to Court Information Using the conceptual framework for positive free speech developed in this collection and at a workshop that preceded its publication, this chapter has shown the ways in which positive and negative free speech is present in the UK court system at both the theoretical and the practical level. Accepting the proposition that positive free speech is beneficial to a functioning democracy, it posits that the state should extend its positive free speech obligations in this domain, and more adequately fulfil those that already exist. In view of the changing media landscape (including a decline in specialist legal coverage), and the digital tools available, the courts should make more proactive positive interventions to protect speech and, in tandem, open justice. It is at the practical level that free speech is most at risk; the existing provisions in common law and statute – which can be understood as providing a positive obligation on courts – allow for practice that can significantly inhibit an individual’s right to access information in the following ways. First, the system is overly reliant on the proactivity and interventions of print and broadcasting media organisations. It cannot be assumed that media concerns represent public concerns, and the media may choose not to report or intervene in cases where the output is not obviously ‘newsworthy’, or may focus on acquiring information that serves media interests (eg being able to name child defendants) and neglect other sorts of information which may fulfil other civic interests (eg legal argumentation). Additionally, there is a wellremarked decline in systematic and specialist court reporting by the traditional media owing to a reprioritising of resources. If this pattern continues, the public will need other means through which to access information from court. This could be provided by academic and not-for-profit initiatives, which do not currently enjoy the status and access of the press in court processes. Improving and lowering the costs of access to court documents would offer access for those otherwise inhibited by their financial circumstances or by a disability. Secondly, the publicity system is often inconsistent and arbitrary, with inadequate documentation of its decisions on transparency, as was exposed by the trials of Erol Incedal. Precise guidelines are needed, informed by consultancy among the different parties, or their representatives, involved in court processes (eg lawyers, victims, witnesses, the media, academic researchers, third sector campaigners) and the wider public, on the way in which different sorts of court data should be handled and proactively communicated to public audiences. Any system should not be ‘one-size-fits-all’, but rather should be appropriately designed for different types of civil and criminal proceedings. Such an approach could include implementing a well-designed digital system for sharing information about intended or active reporting restrictions, further developing the Law Commission’s proposed model. A more clearly defined procedure for deciding on the appropriate publicity for a case, including digital publication of materials, would also help protect the effective administration of justice, and legitimate ECHR rights associated with, for example, child welfare, family and private life, and national security. It could also offer greater accountability of court decisions relating to transparency of proceedings. A passive and reactive system of non-censorship does not provide adequate free speech or open justice protection; more positive free speech, delivered through obligations on the judiciary and courts proactively to communicate what happens in court, would serve a broadly defined public interest and enable fuller participation in civic life.

152

10 Hiding the Truth in the Shadow of the Law? Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts ANDREW SCOTT

Confidentiality provisions, non-disclosure agreements or gagging clauses have played a role in a number of significant recent public scandals. Such clauses were implicated, for example, in the News International phone-hacking scandal,1 in the masking of allegations of serial sexual harassment and abuse perpetrated by Harvey Weinstein,2 and in the obscuring of the Mid-Staffordshire Health Trust debacle.3 In the penumbra of such scandals, questions regarding the legitimacy of the use of such provisions have come to be asked with increasing persistence. Obligations of confidence can serve as a significant disincentive for those bound by them to communicate important truths to a wider public. They can provide a basis for enjoining publication, and their breach can sound in contractual damages.­ Potential liability will extend in equity to any journalist – or other third party – who acquires

1 A confidentiality clause in a contract settling a claim for misuse of private information lay at the heart of the phone-hacking scandal. Notoriously, having first taken the view that it was very weak, News Group Newspapers settled the first claim for invasion of privacy by way of voicemail interception brought against it – by the Chief Executive of the Professional Footballers’ Association, Gordon Taylor – for £425,000 plus substantial costs. This was notwithstanding the fact that at that time the highest award made by the courts in a claim for misuse of private information reached only a little beyond ten thousand pounds. The view of the Culture, Media and Sport Committee was that this was ‘clearly done to buy silence, avoid further damaging publicity and to avert further civil claims over phone-hacking – fruitlessly, as it turned out’ – see News International and Phone-Hacking, Eleventh Report of Session 2010–12, HC903 (London, Stationery Office, 2012) [119]. A truth-obscuring confidentiality agreement was also reached by News International with Glen Mulcaire – see generally, N Davies, Hack Attack: How the Truth Caught Up with Rupert Murdoch (London, Chatto & Windus, 2014) 16–25; J Hanning and G Mulcaire, The News Machine: Hacking – the Untold Story (London, Gibson Square, 2014). 2 J Kantor and M Twohey, ‘Harvey Weinstein Paid Off Sexual Harassment Accusers for Decades’, New York Times, 5 October 2017; M Garrahan, ‘Harvey Weinstein: How Lawyers Kept a Lid on Sexual Harassment Claims’, Financial Times, 23 October 2017; Editorial, ‘Weinstein Case Shows a Need to Curb Gagging Clauses’, Financial Times, 24 October 2017. 3 R Francis, Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry, HC947 (London, Stationery Office, 2013); R Francis, Freedom to Speak Up: An Independent Review into Creating an Open and Honest Reporting Culture in the NHS (London, Department of Health, 2015). Numerous other arrangements of this type have been shown to have limited the freedom of many whistleblowers to highlight systemic failings in health and social care over time – see, eg F West and A Parsons, Briefing Note: Gagging Clauses in the NHS (London, Public Concern at Work, 2011); Whistleblowing Commission, Report on the Effectiveness of Existing Arrangements for Workplace Whistleblowing in the UK (London, Public Concern at Work, 2013).

154  Andrew Scott the information while knowing that it is subject to an obligation of confidence. This powerful disincentive can be generated even where a person is wrong to believe that he or she might face legal repercussions should the information in question be disclosed. The truth can be hidden in the shadow of the law. A related, but distinct, anxiety is that the concerns of the immediate party to the contract may be met through compensation, with that compensation being provided, in part, in exchange for the acceptance of the confidentiality clause in the settlement. Alternatively, such a person may be browbeaten into accepting confidentiality clauses in return for more beneficial treatment under a contract than might otherwise be offered (for instance, the promise of positive references). Silence can be bought. The risk, then, is that any potential for interested third parties – other persons directly affected, regulators or the wider public for instance – to learn of the events or circumstances that gave rise to the immediate dispute may be eliminated. The public can be left designedly ignorant of potentially important insights into criminal, negligent or abusive behaviour or systemic malpractice on the part of powerful interests. The aim of this chapter is to take up such complaints, especially insofar as they impugn the behaviour of public authorities. The assumption is that, in the context of positive free speech and the associated right of access to public information, it can never be legitimate for public authorities to deploy such clauses solely for the purposes of securing ‘nondisparagement’ in pursuit of ‘reputation management’. This starting point generates a quandary, however. Given the nature of the circumstances in which such agreements are entered, the incentives generated for the parties involved and the limited opportunities to assess and challenge confidentiality clauses that are available to third parties, it is not at all clear how, or by whom, the illegitimate use of such provisions can be effectively regulated so as to deter their use or obviate them when they have been agreed. The options open to control the use of confidentiality clauses by private parties are fewer still. Having introduced the phenomenon of ‘gagging clauses’, the paragraphs that follow consider the extant operation of the law where confidentiality obligations become contentious, note the conditioning intent of a small number of legislative interventions, posit the option of judicial review of public sector contracting initiated by third parties to contracts and set out the fledgling moves towards installing reporting and oversight regimes in respect of some categories of public sector compromise agreement. Having assessed the efficacy of existing responses to the perceived problem, the chapter concludes with further reflection on the possible roles of reporting obligations, regulatory bodies and general public education in this regard.

I.  The Prevalence of Confidentiality Agreements Non-disclosure obligations are commonplace; they are regularly included in contracts entered into for a wide range of purposes by a wide range of parties. Explicit confidentiality clauses may be found, for example, in employment contracts, procurement contracts and dispute settlement contracts.4 There is a wide variation in the form that a particular 4 Dispute settlement contracts might be entered for the purposes of employment severance, or as part of the resolution of personal injury claims or claims for invasion of privacy.

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  155 ­ bligation of confidence might take. Obligations might arise from the implied terms of a o contract; from one of many express promises in a contract which are given in exchange for an undifferentiated amount of consideration; from an express term (among others), but with a specific amount explicitly added in exchange for the duty of confidence; or from a standalone non-disclosure agreement running alongside a main settlement contract. All such agreements involve the voluntary assumption of limitations on one’s freedom of speech. The aims underpinning the generation of obligations of confidence are normally entirely legitimate.5 A company may wish to protect business secrets should an employee take up a new position with – or a customer switch business to – a competitor; an employer may wish to prevent the circumstances that have resulted in a settlement becoming known to others because no liability has been accepted and there is a fear that unmeritorious ‘copycat’ claims may be brought; both parties may wish to facilitate the avoidance of protracted legal proceedings; and there may be myriad other reasons why the communication of knowledge obtained in the course of a contractual relationship may be fairly restricted.6 The scope of any given clause is largely in the hands of the individual drafters. Some such clauses will do no more than protect trade secrets and equivalent information from disclosure by former employees and others who have access to it. Such aspirations are understandable and sensible. Where the circumstances in which a contract is entered are contentious, however, the obligations created by confidentiality clauses invariably go further. Generally, they will include the preclusion of future discussion of the existence and terms of the agreement. The parallel here with the much-criticised super-injunction are clear.7 Moreover, the drafting of clauses is often vague and imprecise.8 Parties made subject to such under-determined obligations of confidence can easily be left uncertain as to what can and cannot be disclosed to third parties. The questionable, seemingly deliberate, use of confidentiality clauses by both public authorities and private bodies to ‘manage reputations’ has become a recognised problem. Signal instances have received significant attention. Nevertheless, there is limited empirical evidence available on the use of compromise agreements and confidentiality obligations. There is no systematic evidence on the extent of use by private parties; the evidence regarding use by public authorities that does exist is far from comprehensive. In 2013, the National Audit Office analysed a relatively small sample of public sector compromise agreements arising from the employment context better to understand the policy and practice underpinning public sector use of confidentiality clauses.9 Taking a cue, the Public Accounts Committee then undertook an inquiry

5 EC Tippett, ‘Companies Need Confidentiality Clauses – But Not to Muzzle Sexual Abuse Victims’, The Conversation, 22 November 2017. 6 For instance, in BBC v HarperCollins Publishers Ltd [2010] EWHC 2424 (Ch), the purpose was to prevent disclosure of the identity of the anonymous racing driver ‘Stig’, who played a role in a segment of the programme Top Gear. Ben Collins, the driver, had authored an autobiography the publication of which the BBC sought to enjoin. 7 See generally Committee on Super-Injunctions, Super-Injunctions, Anonymised Injunctions and Open Justice (London, Master of the Rolls, 2011). 8 Francis, Freedom to Speak Up (n 3) [69]: ‘[Some clauses] contained restrictions that seemed unnecessarily draconian, and I can appreciate how individuals might think they were “gagged”. This is a hindrance to ­transparency. Greater care needs to be taken in the drafting of confidentiality clauses, which should only be included if they are genuinely in the public interest. All settlement agreements should be available for inspection.’ 9 Comptroller and Auditor General, Confidentiality Clauses and Special Severance Payments, HC130 (London, Stationery Office, 2013). The NAO sampled 50 agreements across four government departments (health, defence,

156  Andrew Scott focused on the use of compromise agreements, confidentiality clauses and special severance payments by public authorities in the context of the termination of employment.10 Both reports were pre-empted by an earlier study undertaken by the Northern Ireland Audit Office (NI Audit Office) into the use of non-disclosure obligations in the settlement of ­clinical negligence disputes.11 Other inquiries into particular events have also noted the role played by confidentiality agreements in managing associated publicity. The National Audit Office (NAO) found that almost all agreements included terms intended to prevent discussion of the fact that the agreement had been made and the nature of the terms of the agreement. Around half of the agreements analysed included obligations not to disclose confidential information obtained during the course of employment, and/ or not to publish derogatory, defamatory or disparaging statements about the employer.12 None of the agreements analysed by the NAO extended explicitly to impede subsequent discussion of the underlying circumstances of the dispute at hand. Indeed, given that the agreements reviewed by the NAO derived from the employment context, to have done so could have been contrary to statutory protections for whistleblowers.13 Nevertheless, the NAO also found that those persons who had been made subject to such clauses often understood that they had in practice been prevented from making any public statement concerning the circumstances of the dispute; that they had ‘felt gagged’.14 Such perceptions may be attributable to the relative imprecision of the language used by the drafters of such clauses, and to the personal and cultural contexts in which agreements are negotiated and signed. Both the language and the context may sometimes be deliberately contrived. For instance, seven out of every eight non-disclosure agreements studied by the NAO did not directly reference the continuing right to make disclosures protected by the Public Interest Disclosure Act 1998. Moreover, all of those that did were drawn from one and the same particular employment context (health). Indeed, it has been argued explicitly that: often the heavy handedness of the lawyers representing the employer leads the individual to believe they have either signed a gagging clause or have been effectively gagged and they are too scared to subsequently speak up and take their concern outside of the organisation even to an appropriate regulator.15

In its report, the Public Accounts Committee expressed ‘deep concern’ about the use of compromise agreements including confidentiality clauses in the specific context of education, and communities and local government). According to Treasury data, in the three years to March 2013, the Treasury approved 1053 special severance payments totalling £28.4 million. The true number and value of payments across the public sector is unknown. The Treasury does not approve payments by, for example, local government, the police, the BBC or private sector providers of public services. The NAO published a further report later in 2013 – see Comptroller and Auditor General, Confidentiality Clauses and Special Severance Payments: Follow-Up, HC684 (London, Stationery Office, 2013). 10 House of Commons, Public Accounts Committee, Confidentiality Clauses and Special Severance Payments, Thirty-Sixth Report of Session 2013–14, HC477 (London, Stationery Office, 2013). 11 Northern Ireland Audit Office, Compensation Payments for Clinical Negligence, NIA 112/02 (Belfast, Stationery Office, 2002). 12 Confidentiality Clauses (n 9) 7. 13 Employments Rights Act 1996, s 43J (as amended by Public Interest Disclosure Act 1998, s 1). s 43J provides that any confidentiality clause in an employment or compromise agreement ‘is void insofar as it purports to preclude the worker from making a protected disclosure’. 14 Confidentiality Clauses (n 9) 7–8, app 4. 15 See also West and Parsons (n 3) [8].

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  157 e­ mployment dispute resolution in the public sector.16 It reported that 88 per cent of public sector employment compromise agreements included confidentiality clauses.17 One fear expressed by the committee was that a lack of transparency, oversight and proper accountability over the use of such legal devices had allowed authorities to avoid unwelcome publicity and reputational damage.18 The committee recognised that the use of confidentiality clauses could be appropriate in some circumstances, but warned that they had in fact ‘been used inappropriately to deter former employees from speaking out about serious and systematic failures within the public sector, for example, in patient care or child safety’.19 It  was explicit that ‘confidentiality clauses have been used in compromise agreements to cover up failure’, and that in such cases ‘the employer’s [authorities’] interest may have masked the wider public interest’.20 In 2002, the NI Audit Office reviewed the overwhelming majority of the 326 clinical negligence claims that had been settled by the providers of health and personal social services in the three years to 2000–01. It noted that in a number of cases where agreements were made out-of-court, confidentiality clauses had been inserted into the terms of the settlement.21 It found that this was done mainly at the instigation of the defendants in the actions. The NI Audit Office considered the use of such clauses to be ‘questionable’, and in the majority of cases ‘inappropriate’ given that payment of public monies was involved.22 It welcomed assurances given by the NI Department of Health that immediate action would be taken to prevent such clauses from being included in settlements henceforth, and noted the department’s acknowledgement to the Public Accounts Committee that such clauses were inappropriate.

II.  The Legality of Confidentiality Agreements The obvious forum in which the legality of a confidentiality clause or non-disclosure agreement might be tested is the court. Consideration by a judge might be triggered should disclosure occur prompting legal action against the perpetrator for breach of the contractual duty of confidence. Alternatively, the person subject to the obligation might apply to the court to seek a declaration that the clause or agreement did not restrict an intended disclosure. Facilitation of a lawful disclosure might be achieved, therefore, either through a narrowing interpretation of the restrictive clause or through a determination that the clause was void. This may be done either on the basis of contractual or equitable principles, or by reference to other legal restrictions on a contracting party’s freedom of contract derived from freedom of information legislation or whistleblowing protections. 16 Public Accounts Committee (n 10) 3. 17 ibid 6. 18 The Committee was also concerned at the notion that taxpayers’ money was being misused unaccountably. 19 Public Accounts Committee (n 10) 3. 20 ibid 6. 21 The precise proportion of settlements in the sample in which confidentiality clauses had been included was not stated in the report. Notably, however, when a non-statistical subset of 20 cases was drawn from the year 1998–99 to analyse the factors that influenced the outcome of cases, it was reported that a confidentiality clause with no admission of liability had been included in every instance – see Northern Ireland Audit Office (n 11) [3.17]. 22 ibid [32] and [3.18].

158  Andrew Scott A number of commentators have suggested that overbroad confidentiality clauses would not be legally valid.23 While such comments are superficially reassuring, quite why this might be the case is rarely fully explained. Moreover, and unfortunately, the ramifications of any finding that a given clause or contract was unlawful, and hence void, may not be altogether reassuring. The legal and practical position of a person who wishes to communicate an important truth notwithstanding the fact that he or she is subject to a gagging clause is neither straightforward nor enviable. He or she may find themselves in effectively the same position – by dint of having either to pay contractual damages or to reimburse monies paid in exchange for the promise of confidentiality under the law of restitution – irrespective of whether the case was won or lost.

A.  Contractual and Equitable Bases for Obligations of Confidence The starting point for an analysis of obligations of confidence is the contract itself. Equity will also shape the nature and extent of obligations of confidence; it will inform the interpretation of contractual obligations of confidence.24 Hence, a court’s consideration of a putative breach of contract by way of disclosure will be largely equivalent to that undertaken for the stand-alone equitable claim. The assessment rests upon a threefold test. This requires, first, that the information in question has the ‘necessary quality of confidence’ about it;25 secondly, that such information has been ‘imparted in circumstances importing an obligation of confidence’;26 and thirdly, that there has been an ‘unauthorised use’ of that information to the detriment of the party who imparted it.27 In the Spycatcher case, Lord Goff added that there are three ‘limiting principles’ on the scope of any duty of confidence.28 The first two of these limitations are arguably no more than the converse of the first limb of the tripartite test. They relate, respectively, to information that is already in the public domain and to trivial or anodyne information. The third limiting principle discussed by Lord Goff concerns the scenario where it is in the public

23 See, eg the comments of Geoffrey Robertson QC (n 59 below) regarding the legal validity of the Zelda Perkins-Harvey Weinstein agreement on the BBC 2 Newsnight programme, 19 December 2017. 24 Of course, equity can also protect confidences even in the absence of any pre-existing legal relationship and where there is no contract to create an obligation of confidence through its express or implied terms: Prince Albert v Strange (1849) 1 Mac & G 25, (1849) 1 De G & Sm 652; Saltman Engineering Co Ltd v Campbell Engineering Co Ltd (1948) 65 RPC 203. Duties of confidence can also come about through the operation of statute. A large number of statutes impose or entail obligations of confidence on public authorities not to disclose information provided by individuals under statutory compulsion: see P Coppel, Information Rights: Law and Practice, 4th edn (Oxford, Hart Publishing, 2014) [26-018]; T Aplin, L Bently, P Johnson and S Malynicz, Gurry on Breach of Confidence: The Protection of Confidential Information, 2nd edn (Oxford, Oxford University Press, 2012) ch 13. Statutory ­obligations of confidence are also regularly imposed on private individuals: for example, by the Civil Procedure Rules 1998, the Data Protection Act 1998 and the Official Secrets Act 1989. 25 See Aplin et al (ibid) chs 5–6. 26 ibid ch 7. 27 The classic statement of this test was that offered by Megarry J in Coco v AN Clark (Engineers) Ltd [1968] FSR 415, 419. As noted by Lord Goff in Spycatcher (Attorney General v Guardian Newspapers Ltd (No 2) [1990] 1 AC 109, 281–82), the question of whether it is necessary to demonstrate within this that detriment has in fact been or is likely to be occasioned by the unauthorised use remains moot. 28 Spycatcher (ibid) 282. In addition, it can sometimes be important whether there has been any grant of consent by the claimant to disclosure – see Aplin et al (n 24) [14.05]–[14.16].

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  159 interest that the information in question should be disclosed.29 This principle imports public policy into the determination of claims for breach of confidence. Confidentiality obligations generated by way of contract can be express or implied.30 In the absence of an express ‘confidentiality clause’, a duty of confidence will very often ­nonetheless arise as an implied term of a given contract. For instance, it is an implied term of an employment contract that an employee will refrain from disclosing certain types of information acquired as a result of his or her employment. For the duration of the employment, this is coupled with an implied duty of loyalty or fidelity to create a broad duty of confidence.31 Following the end of the employment relationship, the implied duty of confidence will persist with regard to certain types of information: ‘trade secrets’, as broadly understood.32 Implied terms will arise less commonly, and less restrictively in contexts other than that of employment, such as with compromise agreements arising from personal injury. Notwithstanding the likelihood that where a contract exists there will be an implied contractual duty of confidence, parties regularly emphasise such obligations through the use of explicit ‘confidentiality clauses’, or even stand-alone confidentiality contracts that run alongside the main contracts entered between parties.33 It is very likely that an express contractual duty of confidentiality will carry more weight than an implied duty in any balancing against the right to freedom of expression, although much will depend on the individual circumstances of a given case.34 The extent of an express duty may be wider than that generated by implication.35 The nature of the pre-existing relationship between the parties can be of real significance to the outcome of a case.36 Aside from the legal ramifications, the practical purpose of including an express clause will often be to emphasise the prospect that there will be financial implications for the source should any disclosure be made in breach of putative obligations of confidence. Whether given information has the ‘necessary quality of confidence’ about it is an objective question. It rests on whether a reasonable person would understand the information to be confidential in all the circumstances.37 This can be informed by accepted norms – ‘usage and practices’ – in the given context.38 The mere fact that a person who conveys 29 See Aplin et al (n 24) ch 16. 30 The range of possibilities was well illustrated in the case of BBC v HarperCollins Publishers Ltd [2010] EWHC 2424 (Ch), [20]–[27]. In a series of contracts between the BBC and a service company, an explicit confidentiality clause was sometimes included and sometimes not. The judge considered that when an explicit clause had not been included, the service company remained bound by implied terms. Moreover, while the driver himself was not a party to the contracts in question, he too was bound by an equitable duty of confidence. 31 The duty of fidelity entails a ‘confidence arising out of the mere fact of employment, the confidence being … that the servant shall not use, except for the purposes of service, the opportunities which that service gives him of gaining information’: Merryweather v Moore [1892] 2 Ch 518, 524. 32 Faccenda Chicken Ltd v Fowler [1986] 3 WLR 288. The case and rule are particularly relevant in the context of restraints on trade. 33 Indeed, the courts have long strongly encouraged such practice; see Printers & Finishers Ltd v Holloway [1965] 1 WLR 1, 6; AT Poeton (Gloucester Plating) Ltd v Horton [2001] FSR 14 (CA). 34 Associated Newspapers Ltd v HRH Prince of Wales [2006] EWCA Civ 1776, [69]; Campbell v Frisbee [2002] EWCA Civ 1374, [22]; Viagogo Ltd v Myles [2012] EWHC 433 (Ch), [45]. 35 In Balston Ltd v Headline Filters Ltd [1987] FSR 330, 348, Scott J concluded that confidential information that could not be protected by an implied term could – in principle – be protected by a suitably limited express covenant. 36 Ash v McKennitt [2006] EWCA Civ 1714, [15]. 37 Lancashire Fires Ltd v SA Lyons & Co Ltd [1996] FSR 629, 656 (CA). 38 Thomas Marshall (Exports) Ltd v Guinle [1979] Ch 227, 248 (Sir Robert Megarry VC).

160  Andrew Scott i­nformation indicates that it is to be treated as being confidential does not necessarily entail that it will be recognised as such.39 Moreover, not all information obtained during the course of a confidential relationship may itself be properly described as confidential.40 There can be real difficulties in determining whether particular information that has been obtained is covered by a general duty. For this reason, again, the courts have encouraged the use of express contractual clauses to bring greater clarity. Conversely, it is clearly the case that trivial information does not possess the requisite confidential quality. In Spycatcher, Lord Goff explained that any duty of confidence ‘applies neither to useless information, nor to trivia’.41 Hence, there can be no obligation of confidence in respect of anodyne details of the canteen menu, the colour of the office carpet or of other ‘trivial tittle tattle’.42 Similarly, and as a matter of logic, information that is already widely known cannot be the subject of a claim for breach of confidence. Equity does nothing in vain. As Lord Goff explained in the Spycatcher case, ‘once [information] has entered what is usually called the public domain … then, as a general rule, the principle of confidentiality can have no application to it’.43 That said, the parameters of the ‘public domain’ concept are uncertain. Prospective liability for a putative breach of confidence can be set aside where there is a public interest in the disclosure of the information concerned.44 This factor can be sufficient to negate both contractual and equitable obligations. It is not possible exhaustively to delimit the range of scenarios in which there might be a public interest in disclosure: ‘the circumstances in which the public interest may override [a claim] are probably not capable of precise categorisation or definition’.45 Insight can be drawn from the wide array of previous decisions regarding the concept in the areas of copyright, defamation, breach of confidence and misuse of private information.46 In general terms, however, the circumstances in which a public interest argument may prevail in a claim for breach of confidence can be summarised under three overlapping headings: the public interest in disclosure of iniquity; the public interest in the public not being misled; and the public interest in the disclosure of matters of public concern.47 An important additional factor has a bearing on cases involving public authorities. In any democracy, there is a general public interest in promoting the accountability of government, and of state institutions more broadly. This is recognised by the courts, and is coherent

39 Worsley v Cooper [1939] 1 All ER 290; Re Dalrymple [1957] RPC 449; Re Galley Ltd’s Application [1959] RPC 141. 40 Ash (n 36) [15] and [18]. 41 Spycatcher (n 27) 282. 42 Coco (n 27) 421 (Megarry J). 43 Spycatcher (n 27) 282. See also, Douglas v Hello! Ltd [2005] EWCA Civ 595, [105] (Lord Phillips MR): ‘once information is in the public domain, it will no longer be confidential or entitled to the protection of the law of confidence’. 44 Spycatcher (n 27) 282. This embraces the concept that there is ‘no confidence in iniquity’. It is moot whether this is because the public interest obviates the confidential quality of the information concerned, or because the public interest factor outweighs the recognised confidentiality. 45 Hyde Park Residence Ltd v Yelland [2001] Ch 143, 172 (Mance LJ). 46 See G Millar and A Scott, Newsgathering: Law, Regulation and the Public Interest (Oxford, Oxford University Press, 2016) ch 2. 47 Coppel (n 24) [25-024]–[25-027].

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  161 with both the positive freedom of speech and the principle of open government.48 As Lord Goff put it in Spycatcher, ‘in a free society there is a continuing public interest that the workings of government should be open to scrutiny and criticism’. Hence, the weight of the public interest in cases involving attempts by state authorities to claim for breach of confidence is generally greater than that in cases involving only private citizens. Governmental bodies are not entitled to use breach of confidence claims to prevent disclosures about themselves and their activities unless there is a particular and sufficient countervailing public interest which requires publication to be restrained.49 This idea has been developed more fully in comparable Australian jurisdictions in the context of claims for breach of confidence. In A-G (UK) v Heinemann Publishers Pty Ltd, for example, McHugh JA explained that: the relationship between the modern State and its citizens is so different in kind from that which exists between private citizens that rules worked out to govern the contractual, property, commercial and private confidences of citizens are not fully applicable where the plaintiff is a government or one of its agencies. Private citizens are entitled to protect or further their own interests, no matter how selfish they are in doing so … But governments act, or at all events are constitutionally required to act, in the public interest. Information is held, received and imparted by governments, their departments and agencies to further the public interest. Public and not private interest, therefore, must be the criterion by which equity determines whether it will protect information which a government or governmental body claims is confidential.50

In Commonwealth v John Fairfax and Sons Ltd, Mason J had gone further in this line of thought: when equity protects government information it will look at the matter through different spectacles … it can scarcely be a relevant detriment to the government that publication of material concerning its action will merely expose it to public discussion and criticism. It is unacceptable in our democratic society that there should be a restraint on the publication of information relating to government when the only vice of that information is that it enables the public to discuss, review, and criticise government action … unless disclosure is likely to injure the public interest, it will not be protected … the court will not prevent the publication of information which merely throws light on the past workings of government … so long as it does not prejudice the community in other respects. Then disclosure will itself serve the public interest in keeping the community informed and in promoting the discussion of public affairs. If, however, it appears that disclosure will be inimical to the public interest because national security, relations with foreign countries or the ordinary business of government will be prejudiced, disclosure will be restrained.51

This approach mirrors that adopted by Lord Widgery CJ in Attorney-General v Jonathan Cape Ltd.52 It has also been affirmed regularly as representing the law of confidence in

48 See further A Scott and A Burke, ‘The Access to Information Dimension of Positive Free Speech’, this volume, ch 4. 49 There is clearly room for debate on the question of how far this limitation extends in terms of which entities are covered. See generally Coppel (n 24) [25-028]–[25-029]. 50 A-G (UK) v Heinemann Publishers Pty Ltd (1987) 10 NSWLR 86, 191; 75 ALR 353, 454. See further Coppel (n 24) [25-028]. 51 (1980) 147 CLR 39, 51. 52 [1976] QB 752.

162  Andrew Scott England,53 and has clear parallels with restrictions on the freedom of government entities or political parties to sue in libel.54 Importantly, whatever its nature, it is not enough for a defendant merely to demonstrate that there is a ‘public interest’ in the subject matter of a particular disclosure. Rather, in cases involving private parties, the public interest in disclosure must outweigh the private interests of the party who has contracted for confidentiality. In all cases, the public interest in disclosure must also outweigh the general countervailing public interest in protecting confidences entered between parties: ‘the test to be applied … is not simply whether the information is a matter of public interest but whether, in all the circumstances, it is in the public interest that the duty of confidence should be breached’.55 Where interpersonal obligations have been expressly and freely adopted by way of contract, the public interest in upholding the duty of confidence will be correspondingly strong.56 Moreover, the immediate expression rights of the parties – as opposed to the interests of the wider public – to the contract may be less weighty in the balance.57 Even should the balancing exercise favour disclosure, the legitimate extent of disclosure – that is, the categories of persons to whom the information can lawfully be disclosed – can also be qualified by the strength of the countervailing factors. Under the Human Rights Act 1998, the undertaking of this balancing exercise by a court must align at least with Article 10 of the European Convention on Human Rights (ECHR), although – insofar as other rights are relevant in the circumstances of the case – it may see a judge employ the two-stage methodology for the balancing of ECHR rights.58

53 See, eg Spycatcher (n 27) 150–52, 203, 218, 221–22, 257–58, 270 and 283. 54 Derbyshire County Council v Times Newspapers [1993] AC 534. See generally R Parkes and A Mullis (eds), Gatley on Libel and Slander, 12th edn (London, Sweet & Maxwell, 2013) [8.20]. Arguably, the same logic explains the introduction of anti-SLAPP legislation in an increasing number of jurisdictions, albeit that in that context the logic extends into the domain of limiting private power to restrict criticism. 55 HRH Prince of Wales (n 34) [67]–[68]. See also Viagogo Ltd v Myles [2012] EWHC 433 (Ch), [35]. The public interest in the maintenance of confidentiality provided for by contract was explained and emphasised by Tugendhat J in Raab MP v Associated Newspapers Ltd [2011] EWHC 3375 (QB), [60]: ‘the court must also have regard to the effect upon third parties of any order it may make. Confidentiality agreements are very common in the settlement of legal disputes. It is the policy of the law to encourage settlements of disputes by agreement. Confidentiality is required to achieve this, not only in the form of confidentiality clauses in settlement agreements, but also in procedures such as marking letters “without prejudice”, in mediations, and, of course, in the form of legal professional privilege. All of these advance the public interest for all litigants. Newspaper publishers make use of confidentiality agreements in the settlement of disputes as much as any other litigants. If a party minded to settle litigation knows that any confidentiality agreement included in the terms of settlement is liable to be set aside … then that would tend to make parties less willing or able to settle disputes.’ 56 It may be thought that this is seldom the case, however; that the normal case is that of unequal bargaining power which may unduly restrict the rights and interests of the weaker party to the contract (invariably the individual employee, the injured party or whomever). 57 This will not necessarily be the case. It can be argued, for instance, that the law permits ‘journalists [to] enter into agreements which they fully intend to repudiate further down the line’, eg when taking up employment with the subject of a journalistic investigation in order to access information on working practices that might not otherwise be accessible – see G de Wilde, ‘Viagogo Ltd v Myles: Investigative Journalism and Confidentiality’, Inforrm’s Blog, 1 June 2012. 58 As prescribed by Lord Steyn in Re S (a child) [2004] UKHL 47, [17]. In Viagogo (n 34), Hildyard J appeared to consider that the desire to protect reputation was not a powerful factor in the mix (at [35] and [79]). This may have been due to the fact that the case concerned a company, and hence corporate reputation, which may not be protected under Art 8 ECHR. Even if it proves a weak factor in the immediate circumstances of a given case, however, personal reputation will be relevant and its presence will entail a two-way balancing as prescribed by Lord Steyn. For the undertaking of a balancing exercise in a case involving the dismissal of a public official on account of his whistleblowing, see Guja v Moldova (2011) 53 EHRR 16. See also Marchenko v Ukraine (2010) 51 EHRR 36.

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  163

B.  Other Legal Principles that Obviate Overbroad Confidentiality Clauses In addition to limitations in the law of confidentiality, two further legal principles restrict the freedom of public authorities to utilise overbroad confidentiality clauses for the purposes of reputation management.59 First, a public authority is not able to ‘contract out’ of its obligations under the Freedom of Information Act 2000. It cannot describe information that it holds as confidential de novo in a contract, and thereby ensure that such information cannot be released either under the 2000 Act or by the other party to the contract. The ‘tying-up’ of information by way of contract becomes most effective, therefore, only when the information in question arguably could not be released under the 2000 Act (for instance, due to its consisting of personal information or its coverage by some other freedom of information exemption). Secondly, under the Public Interest Disclosure Act 1998, a confidentiality clause in a compromise agreement cannot legally be used to prevent a person from making a ‘protected disclosure’. The 1998 Act is an unusual animal: a Private Member’s Bill that reached the statute book.60 It is intended to protect employee whistleblowers from detriment at the hands of the employer on the grounds that the worker has made a protected disclosure. It did this by introducing a new Part IV A into the Employment Rights Act 1996. Section  43B(1) of the 1996 Act limits the subject matters of protected disclosures to disclosures of the facts that: (a) a criminal offence has been committed, is being committed, or is likely to be committed, (b) a person has failed, is failing, or is likely to fail to comply with any legal obligation to which he is subject, (c) a miscarriage of justice has occurred, is occurring, or is likely to occur, (d) the health or safety of any individual has been, is being, or is likely to be endangered, (e) the environment has been, is being, or is likely to be damaged, or (f) information tending to show any matter falling within any one of the preceding paragraphs has been, is being, or is likely to be deliberately concealed.

These heads are not restricted to events in the UK,61 but they do not cover whistleblowers who commit an offence by their disclosure.62 Section 43J of the Employment Rights Act 1996, as amended by section 1 of the 1998 Act, provides that any confidentiality clause in an employment contract or employment severance contract that purports to prevent the making of a ‘protected disclosure’ is void. Notably, such protection is limited to the employment context. In any event, it is also 59 A third principle might also be relevant, although currently in abeyance. Speaking on the BBC 2 Newsnight programme on 19 December 2017, lawyer Geoffrey Robertson QC noted that formerly it had been unlawful for any person to interfere with the communication of information on criminal conduct to the police. He recommended reintroduction of the repealed law. It may be that the common law offence of perverting the course of justice already adequately fills the perceived gap. 60 For an overview, see D Pyper, ‘Whistleblowing and Gagging Clauses: The Public Interest Disclosure Act 1998’, Briefing Paper CBP 7442 (London, House of Commons Library, 2016). In July 2013, the Department for Business Innovation and Skills issued a call for evidence on the operation of the Act: ‘The Whistleblowing Framework’, BIS/13/953 (London, Department for Business, Innovation and Skills, 2013). 61 s 43B(2). 62 For instance, an offence under the Official Secrets Act 1989 – see s 43B(3).

164  Andrew Scott relatively circumscribed by the narrow definition of the public interest that characterises protected disclosures. While the definition arguably encapsulates the core aspects of the public interest concept, it does not extend as far as the concept has been taken in other areas of the law. Any broader protection deriving from the common law or ECHR jurisprudence that reached beyond that afforded by the Public Interest Disclosure Act 1998 would presumably also be available in the employment context.63 In Cream Holdings Ltd v Banerjee, Lord Nicholls rejected an argument to the effect that disclosures by employees should be governed exclusively by the express statutory provisions of the 1998 Act. He pointed out that the ‘graduated protection afforded to “whistle blowers” by [section 1 of the 1998 Act] … were intended to give additional protection to employees, not to cut down the circumstances where the public interest may justify private information being published at large’.64 The possibility that the 1998 Act may provide more limited protection than the common law has been recognised from the outset.65

C.  Ramifications of ‘Successful’ Challenge to a Confidentiality Clause While the outcome of any given case will be reliant on its specific facts, it is likely that the courts would facilitate and not censure disclosure of information on any matter of significant public importance, especially insofar as that information concerned malpractice or underperformance in public office. However, the ramifications that would flow from a successful challenge to a confidentiality clause – or defence of litigation alleging a breach of contract – may not be universally positive. One potential concern should an agreement – or elements thereof – be struck down as unlawful is that ‘once a contract has been found to be ultra vires, it must in general be unravelled’.66 In principle, any monies paid under the contract in exchange for the limitation on freedom of speech could then be liable to be repaid under the law of restitution. Ultimately, the result for the party to the contract may be the same as that where the contractual clause was upheld and the party sued for breach of contract. The extent to which monies would have to reimbursed, in practice, may depend on the particular form of provision used in the contract. Different scenarios can be envisaged. If an obligation of confidence was created in a stand-alone contract which ran alongside a main settlement contract, then the illegality of the clause would mean that the whole purpose of the contract would be negated, and in principle all monies paid thereunder would have to be reimbursed. In the middle ground is the contract in which multiple obligations are prescribed, but where a particular sum is specified as being payable on account of the acceptance of an obligation of confidence. In either of these two circumstances, repayment

63 The legal protection afforded under the Act does not introduce a general right for all whistleblowers: protected disclosures are limited to six categories where specific procedures for disclosure have been pursued. 64 [2004] UKHL 44, [24]. 65 Y Cripps, ‘The Public Interest Disclosure Act 1998’ in J Beatson and Y Cripps (eds), Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams (Oxford, Oxford University Press, 2000) ch 17; S Palmer, ‘Human Rights: Implications for Labour Law’ (2000) 59 CLJ 168, 195. 66 ACL Davies, ‘Ultra Vires Problems in Government Contracts’ [2006] LQR 98, 98.

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  165 would be required unless the sum attached to the promise of confidentiality was so significant and disproportionate that it might be read as a penalty clause.67 In some circumstances, conversely, repayment may not be required. For instance, if a settlement contract comprised the placing of multiple different obligations on each party, including the payment of compensation, but no particular sum was attached specifically to the promise to maintain confidentiality, then reimbursement may not be required. The wider purpose of the contract – dispute resolution – could still be served. This would also likely be the scenario with implied or explicit terms imposing obligations of confidence in an employment contract. In all of these scenarios, there is also the ‘social realism’ factor to be considered. The practical risk that a public authority or private party may seek to recoup monies paid must be reduced by the likelihood that it would suffer further reputational harm from acting in such a fashion. This would compound the negative exposure generated by the original attempt at reputation management through the use of a gagging clause. Magnanimity may prove the preferable course of action. Recalcitrance is also possible.68

III.  Regulating the Use of Confidentiality Agreements It seems clear that many confidentiality clauses or agreements would not prevent parties from making disclosures on matters of public interest. It is something of a paradox, then, that very general express confidentiality clauses that purport to apply to any disclosure appear to be widely included in contracts notwithstanding this limited legal effect. The strict efficacy of the legal provision, however, is often not the point. Individual parties to an employment dispute or a medical negligence action cannot be expected to understand the finer details of the rules of contract and equity. They will likely consider themselves bound by the duty of confidence expressed in the contract irrespective of the legal reality. Many such purported restrictions may never be challenged.69 A prospective source may not come forward should he or she mistakenly believe themselves to have been subjected to

67 Under the ‘penalty rule’, a payment of money for a breach stipulated in the contract is unenforceable against the offending party if it is an exorbitant alternative to common law damages – see generally the joined cases of Cavendish Square Holding BV v Talal El Makdessi and ParkingEye Ltd v Beavis [2015] UKSC 67. As one commentator has put it, ‘the trick, then, is for the employer to identify a sum of money referable to confidentiality which is high enough to disincentivise the worker from speaking out but not so high as to appear disproportionate to the loss that the employer might suffer from a breach of confidentiality. Easier said than done’ – see J Laddie, ‘Gagging Clauses for Whistleblowers: Worth the Paper They’re Written On?’, Inforrm’s Blog, 26 February 2013. 68 In a notable illustration, the United Lincolnshire Hospitals NHS Trust warned its former Chief ­Executive, Gary Walker, that if he went ahead with a planned interview with the BBC Radio 4 Today programme he would be in breach of a compromise agreement and that he might be forced both to repay the settlement he had received and to cover the Trust’s legal costs. This threat was widely reported; Walker proceeded with the interview (14 February 2013) – see generally J Farragher, ‘Gary Walker Whistleblowing Case: Time for an NHS Culture Change?’, Personnel Today, 18 February 2013; J-P Ford Rojas and R Alleyne, ‘NHS Whistleblower Faces Ruin after Speaking Out about Patient Safety’, Daily Telegraph, 15 February 2013. 69 It might be added that in circumstances where a disclosure is made on a matter that is in the public interest, very often it might be expected that no attempt to enforce a confidentiality clause would then be made by the other party for fear of further exacerbating the loss of reputation attendant on the exposure of wrongdoing – see West and Parsons (n 3) [10].

166  Andrew Scott binding and comprehensive obligations of confidence. He or she may fear the loss of their employment or some compulsion to return monies paid should they speak out.70 Erroneous expectations can generate powerful and often effective disincentives. Confidentiality clauses may sometimes be used to encourage silence as a matter of fact rather than as a matter of effective law. Alternatively, a prospective source – having been well-compensated for personal losses – may be satisfied with his or her lot and prove indifferent to the wider ramifications of their knowledge or experiences. In either case, any third party interest in the publicising of such error, malpractice, criminality or systemic failings as has become apparent to the aggrieved party may be disregarded. From a societal perspective, this is clearly undesirable. An obvious question arises as to how such ‘expedient’ practice might be curtailed in circumstances where the immediate party to the compromise agreement or other similar contract is not likely to challenge its legitimacy in court. Notwithstanding the extant restrictions on the deployment of confidentiality clauses, the ability otherwise to have their use held to be unlawful could be of significant value to newsgatherers and the wider public. This is especially the case where the immediate parties to the contract are unwilling to challenge its validity. It is important to assess, therefore, whether – both within and outside the employment context – public authorities might be disbarred by law from deploying confidentiality clauses in contracts that they enter into insofar as such provisions limit the freedom of counterparties to discuss elements of the public body’s policy or performance. As regards the propensity of public authorities to utilise overbroad clauses, two options present themselves. The first relies on the argument that it is always unlawful in an administrative law sense for a public body to include an overbroad confidentiality obligation in any contract, with the result that an application for judicial review to challenge the making of the contract may be made by any person with requisite standing.71 This would extend the opportunity to challenge the legitimacy of a confidentiality clause in court beyond the immediate parties to a contract. Similar arguments would be considered in this context as would be the case if a private law claim based on the contract was brought before the court. The second option would involve the invigoration of a standing regime that notionally oversees the use of such clauses. Few, if any, options present themselves with regard to the restriction of the use of such clauses in contracts between private parties. In that case, overbroad confidentiality clauses will continue to be effective unless challenged or ignored by persons made subject to them.

A.  Administrative Law Option: Judicial Review of Public Authority Contracting As noted above, there is a strong principle against the bringing of a claim for breach of confidence by public authorities to prevent disclosures about themselves and their a­ ctivities 70 Similarly, a former employee may fear the withdrawal of contractual commitments to provide positive e­ mployment references. 71 Journalists and similar organisations are sometimes afforded standing to make applications for judicial review – see, eg R v Felixstowe Justices, ex parte Leigh [1987] QB 582, [1987] 2 WLR 380; R (on application of Privacy International) v HMRC Commissioner [2014] EWHC 1475 (Admin), [77].

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  167 unless some strong countervailing public interest is at stake.72 Moreover, many public authorities are barred altogether from bringing an action in defamation.73 It would seem entirely appropriate for the same logic that applies in those contexts also to bear on other exercises of state power. Indeed, such thinking was reflected across each of the speeches made in the recent freedom of information case of Kennedy v Charity Commission.74 In that case, the Supreme Court developed a previously unheralded common law principle of open government, with Lord Mance grounding this principle in the appreciation that ‘­information is the key to sound decision-making, to accountability and development; it underpins democracy and … the press, NGOs and individuals concerned to report on issues of public interest [depend upon it]’.75 Similar ideas are reflected in the emergent Convention jurisprudence on the putative Article  10 right to receive information from public authorities.76 It is at least arguable that this principle is applicable generally across all the various activities of government. Whether on the basis of Article 10 jurisprudence or by reference to an incipient common law principle, then, public authorities may be constrained in their freedom to enter contractual agreements that are intended to prevent disclosures about themselves or their activities in the absence of some specific public interest justification for so doing. The essential argument would be that by entering a contract a public body would be exercising a form of public power. The legal capacity of a public authority to enter a contract can be based on either a statute that governs the area of activity concerned or the inherent capacity of the Crown as a bearer of corporate personality.77 Such exercise of public power would be subject to review against the principle of ultra vires, and against ECHR rights and/or common law principles. The fact that public authorities are bound to act in accordance with ECHR rights even when exercising private law powers seems evident from the Human Rights Act 1998. It has also been confirmed by the courts.78 Deployment of an overbroad confidentiality clause would be in breach of section 6(1) of the Human Rights Act 1998 insofar as the clause served unnecessarily to restrict the Article 10 right. Assessment against the common law principle would be more poignant still given the recognition by the Supreme Court in Kennedy that officials’ ‘unwillingness to disclose information may arise through habits of secrecy or reasons of self-protection’, and the emphasis placed on the interests of the wider public and the needs of a democratic polity.79 Presumably, when considering the lawfulness of a confidentiality clause against ECHR rights or common law principles, the aim of the court would be to limit any necessary interference with the will of the parties to the contract to the minimum degree possible to

72 Spycatcher (n 27) 283 (Lord Goff). 73 See Parkes and Mullis (n 54) [8.20]. 74 [2014] UKSC 20. See further Scott and Burke, this volume, ch 4. 75 Kennedy (ibid) [1]. 76 See further Scott and Burke, this volume, ch 4. 77 C Turpin, British Government and the Constitution: Text, Cases and Materials, 5th edn (London, Butterworths, 2002) 620. Some authors refer to this second source of power, imprecisely, as an exercise of ‘prerogative’ power – see Davies (n 66) 102, citing T Daintith, ‘Regulation by Contract: The New Prerogative’ (1979) 32 Current Legal Problems 41; M Freedland, ‘Government by Contract and Public Law’ [1994] PL 86, 91–92. 78 Clift v Slough Borough Council [2009] EWHC 1550; Guja (n 58). 79 Kennedy (n 74) [1] (Lord Mance).

168  Andrew Scott ensure compliance with the rights or principles in question. Hence, insofar as it was possible, contractual obligations should be interpreted by the court so as to exclude any conflict. Very broad or vague terms should be interpreted so as not to preclude speech on a matter of public interest. Only when such accommodation of the rights or principles at hand was not possible should the court move to strike out a provision as unlawful or to declare a contract unlawful and hence void. A third party who sought to challenge the making of a contract by way of an application for judicial review would have to take into account the potential ramifications for the party who had entered into the contract with the public authority. Any monies paid under a void clause or contract would be liable to be repaid under the law of restitution. Of course, sympathy for such a person may be limited, but equally it could be very real. The potential efficacy of a regime of this nature is limited by the restricted ability of third parties to attain knowledge of particular compromise agreements. There will be occasions, for instance, when a journalistic investigation uncovers aspects of a story but where further investigation is frustrated by a potential source being bound by a confidentiality agreement.80 In such circumstances, the option for a third party to the contract to challenge its legal validity would appear to be desirable.

B.  Oversight Regime for Confidentiality Clauses in Public Sector Contracts Irrespective of whether the use of over-general confidentiality clauses by public authorities is unlawful ab initio, but more especially if it is, it would seem desirable for an effective oversight regime to supervise and, where necessary, pre-empt the use of such legal devices. At present, financial reporting rules require central government departments and their agencies to seek Treasury approval in advance of making a special severance payment. It is also mandatory for public bodies to publish details of special severance payments in their annual reports and accounts. There is ample evidence, however, that such mechanisms are not currently serving a coherent function in terms of deterring misuse of confidentiality clauses in reputation management strategies. It is not clear whether public authorities are under any obligation to report on payments made in settlement of other disputes, such as claims for personal injury. In 2013, the Public Accounts Committee undertook an inquiry focused on the use of compromise agreements, confidentiality clauses and special severance payments by public authorities.81 At the outset of its inquiry, and notwithstanding the requirement to notify and seek approval from the Treasury, the Public Accounts Committee (the Committee) identified a ‘lack of any meaningful central oversight’.82 It noted that ‘neither the ­Treasury

80 Consider the factual circumstances of Raab MP v Associated Newspapers Ltd [2011] EWHC 3375 (QB), a case in which the defendant sought to have a third person who might be able to provide them with information supporting a justification defence in libel proceedings released from a confidentiality clause before it was required to serve its justification defence. 81 Public Accounts Committee (n 10). 82 ibid 5.

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  169 nor the Cabinet Office review the confidentiality clauses contained within compromise agreements’.83 The Committee also found that practice in respect of the obligatory reporting of details of special severance payments in annual accounts is inconsistent, and, indeed, that nondisclosure is permissible under the Treasury’s Government Financial Reporting Manual.84 It recommended that the Treasury should ‘revise the reporting requirements in the [Manual] to ensure the mandatory and consistent disclosure of special severance payments in public sector annual accounts’.85 The Committee formally called on the Treasury to inform it of the measures that it planned to take to encourage wider public sector bodies to act in accordance with the revised guidance, and on whether such compliance could be made mandatory.86 It  noted further that there is currently no requirement for private sector providers that deliver public services to comply with any guidance. During the course of the Committee’s inquiry, the Cabinet Office undertook to prepare new guidance for the civil service on the appropriate use of these legal devices. It also indicated that it would institute improved monitoring processes, and undertook to publish an annual consolidating report for Parliament. The Committee insisted that this guidance should include notes on the governance arrangements that should be in place to approve agreements.87 It also recommended that, as regards compromise agreements and payments relating to instances of whistleblowing (as understood in the Public Interest Disclosure Act), the guidance should require public sector organisations to: secure prior approval from the Cabinet Office; set out standard terms and conditions to be used in such agreements; and secure approval from the Cabinet Office before departing from these standard terms.88 Implicitly, the government agreed with these suggestions,89 but no action of general application has been taken on the matter to date. This prompted the Committee to reiterate its recommendation and to state an expectation that revised guidance would be published, and that this should require public sector bodies to seek Cabinet Office approval for compromise agreements.90 It remains unclear whether action will be taken by the Cabinet Office in this regard. Notably, in the wake of several high-profile instances of institutional failings and of the imposition of gagging clauses in severance agreements in the health sector,91 the Department of Health has provided updated guidance and model provisions for contracts.92 The stated aim of that guidance is to ensure that all compromise agreements explicitly reference the right to make a protected disclosure, thereby emphasising the individual’s freedom to 83 ibid 6. 84 ibid 7. The Committee reported that freedom of information requests relating to compromise agreements have sometimes been refused. 85 ibid. 86 ibid 8. 87 ibid 5. 88 ibid 6. 89 HM Treasury, Treasury Minutes, Government Responses on the Thirty Fifth to the Forty Fifth Reports from the Committee of Public Accounts: Session 2013–14, Cm 8847 (London, Stationery Office, 2014). 90 House of Commons, Public Accounts Committee, Whistleblowing: Ninth Report of Session 2014–15, HC593 (London, Stationery Office, 2014) 7 and [29]. 91 See, eg Francis, Report (n 3). 92 Department of Health, Hard Truths: The Journey to Putting Patients First, Cm 8777 (London, Stationery Office, 2014) [2.15]–[2.16]. See also Department of Health, Patients First and Foremost, Cm 8576 (London, Stationery Office, 2013) [2.35]–[2.42].

170  Andrew Scott do so. This regime could stand as a valuable model for other parts of the public sector. Manifestly, it must be supported by a cultural shift among those authoring such contracts, and in institutional cultures more broadly.93 However, ‘culture change [is] not achieved overnight’.94 The absence of any body with effective ‘cross-government responsibility for driving improvements in whistleblowing arrangements’ remains a significant concern.95 There is also room for concern that the new NHS guidance remains limited by reference to the Public Interest Disclosure Act 1998, and to the employment context more generally. There is no obvious reason why an effective oversight regime should apply to severance payments only. It should apply equally to other contracts entered by public bodies, including compromise agreements that determine tortious claims and conceivably also to more routine agreements, such as procurement contracts.96 All such contracts provide opportunities for officials improperly to frustrate the proper focusing of public attention on institutional failings.

IV. Conclusion A series of scandals that have come to light over the past decade have provided increasing levels of insight into the dangers that non-disclosure agreements and confidentiality clauses in other contracts pose to the integrity of public sphere processes. Especially insofar as such instances impugn the conduct of public authorities, the positive freedom of speech and the associated principle of access to state information insist on a more broadly based response in furtherance of democratic constitutionalism. The court-based response based upon the law of contract and equity can offer only a measure of reassurance that confidentiality clauses will not be wrongly exploited. It relies upon the activism of immediate parties to contracts, people who very often face significant disincentives to initiating action and who stand to lose much irrespective of the outcome of proceedings. A commitment to free speech and the proper accountability of state bodies require a more developed response. The Freedom of Information Act 2000 and the Public Interest Disclosure Act 1998 each serve to condition the conduct of public authorities, installing the principle that state-held information on matters of public importance should be readily accessible. The practical impact of such legislative interventions is questionable, however. While a good number of cases involving apparent imposition of workplace s­anctions

93 In pursuit of such a cultural shift in the NHS, the Department of Health has tasked a Chief Inspector of Hospitals to judge whether trusts actively promote the benefits of openness and transparency as part of ­inspections – Department of Health, Hard Truths, ibid [2.15]. Furthermore, Sir Robert Francis QC undertook a further review focused on this theme – see Francis, Freedom (n 3). On confidentiality clauses, he highlighted that good practice entailed that such should not be automatically included in settlement agreements; that they should be the CEO to confirm consistency with the public interest in transparency when used, and they should be ‘written in plain English’ (207). 94 Public Accounts Committee (n 90) [23]. 95 ibid [26]. 96 The former category of agreements can be seen as responses to exceptional events, and hence are similarly ripe for an approvals regime. The proper approach to contracts of the latter type may be the prescription of standard form contracts, deviations from which might be subject to approval.

Addressing the Misuse of Confidentiality Clauses in Public Authority Contracts  171 following protected disclosures reach the Employment Appeal Tribunal each year, the evidence suggests that the use of confidentiality clauses remains habitual and – in some cases – designedly cloaking. Obligatory public reporting of the inclusion of restrictions on speech is currently partial and piecemeal, and is not met with holistic scrutiny. Expectations of transparency entail that confidentiality agreements entered by public authorities should never include restrictions on the freedom of the other party to discuss the mere fact that an agreement had been entered into. There is a profound public interest in knowledge of how public monies are dispensed. Indeed, positive free speech would tend to require that all such agreements should be adequately reported in a public manner, whether in a dedicated register or in annual reporting. Alongside the imposition of reporting obligations on public authorities, the depth of concern regarding non-disclosure agreements and confidentiality clauses requires unremitting effort better to educate the public on what is acceptable. Any such pedagogic initiative might also be profitably oriented towards lawyers themselves, in which context the option of supporting knowledge development with professional regulatory oversight of professional misconduct should also be explored. Only through the dedication of sustained attention to the problems sometimes generated by confidentiality agreements might the requirements of free speech in this context be met. In this respect, it is encouraging that, in light of the #MeToo movement, the Women and Equalities Select Committee of the UK House of Commons has dedicated significant effort to understanding the role of such practices in the context of a broader consideration of sexual harassment in the workplace. It remains to be seen whether this initiative, instigated in the wake of egregious abuses of the legal options available to close down public criticism, can garner a sufficient momentum to achieve substantive change.

172

11 Speaking and Governing through Freedom of Access to Environmental Information MÉLANIE DULONG DE ROSNAY AND LAURA MAXIM

The principle of freedom of access to sanitary and environmental information produced by industry for regulatory purposes should enable citizens, NGOs, scientists and governments to check data and reassess substance toxicity risks. Through this, it supports opportunities to speak and to participate in the governance process. Under freedom of information laws, the ability to improve transparency, independently reassess risks and exercise free speech relies on the availability of data. However, access to environmental information is limited by companies’ property rights, a source of legal exception to the principle of free access. Restrictions within environmental and sanitary information laws show commonalities with the field of information commons, in terms of access enclosure (trade secrets, confidentiality, intellectual property, data protection) and the conditions of producing effective free speech depending on the reusability of information (open data, open formats). This chapter examines the limits imposed by environmental law and regulatory practices on free speech, and considers institutional changes and the integration of open science and open data principles to improve both the governance of information produced by industry and its availability for the public to develop analysis and speech. Access to environmental information was developed as a tool for governing environmental risks, and in particular chemical risks, in reaction to what was considered the failure of ‘command and control’ policies. The objective was to reinforce non-governmental stakeholders’ information and hence power resources and capabilities to act to limit chemical risks by enabling attention-focusing free speech (for example, through public campaigns and press articles). Proponents of access to information-based environmental policies argued that such access could serve as an instrument to control pollution, because firms would want to reduce voluntarily their pollution figures in response to public scrutiny.1 Through access to information, governments intended to create a closer-to-level playing field for societal self-regulation. They intended to facilitate scrutiny of industry activities by NGOs, citizens and the press, and encourage changes in industry practices in response to information-based societal criticism. Access to information is thus a major landmark in the switch in environmental regulation towards the limitation of direct governmental

1 JT Hamilton, Regulation through Revelation: The Origin, Politics and Impacts of the Toxic Release Inventory Program (Cambridge, Cambridge University Press, 2005).

174  Mélanie Dulong de Rosnay and Laura Maxim action on the private sphere, and the empowerment of other stakeholders to act on the industry through public speech. This policy trend joined that on public participation, which developed through the 1980s in response to repeated controversies over chemical pollution, and the exponential development of computing and digital technologies that facilitated the diffusion of masses of information to large audiences. Highlighting the role of state regulators and society in science, the field of chemical risks provides an illustration of the asymmetry of power between industry and public health agencies or private stakeholders. The chemical industry is lobbying both legislators and regulators, the recipients of toxicity studies which condition access to markets, so that it can provide as little information as possible. The process of producing and communicating studies is expensive, and companies claim information on substances could reveal confidential and proprietary business information.2 Therefore, they ask legislators and regulators respectively to develop and implement restrictions on the amount and scope of information industry has to deliver, and on limiting access and reuse by third parties. Open science principles conflict with laws restricting access to results and information on health for financial interests3 and competitiveness.4 The lifecycle of data reveals possible bottlenecks at various stages: production by industry, collection by regulator, access by independent actors, reusability by regulator and independent actors, communication of results. The framework of negative and positive liberty is an effective instrument to guide our analysis of access to information throughout this lifecycle.5 The first section of this chapter examines how theories of positive free speech can be applied and can contribute to environmental legislation on access to information. We provide a critical assessment of the legal framework of access to information and its institutional settings in section II, before explaining which limits to free speech are impeding access and reuse in section III. In section IV, we consider conditions and options for the exercise of positive free speech towards a multiplicity of ‘voices’,6 also allowing a balance between transparency and other values and interests to produce better science supporting the development of safe products protecting the public interest.7 The chapter applies this discourse to the production of information on chemical risks, a field where both lobbyists’ lawyers8 and academics reviewing case law9 usually argue on the 2 M Bronckers and Y van Gerven, ‘Legal Remedies under the EC’s New Chemicals Legislation REACH: Testing a New Model of European Governance’ (2009) 46 CML Rev 1823. 3 J Conrad, ‘Open Secrets: The Widespread Availability of Information about the Health and Environmental Effects of Chemicals (2006) 69(3) Law and Contemporary Problems 141. 4 S Jasanoff, ‘Transparency in Public Science: Purposes, Reasons, Limits’ (2006) 69(3) Law and Contemporary Problems 21–45. 5 I Carter, ‘Positive and Negative Liberty’ in EN Zalta (ed), The Stanford Encyclopaedia of Philosophy, Summer 2018 edn, https://plato.stanford.edu/archives/sum2018/entries/liberty-positive-negative. 6 J Lichtenberg, ‘Foundations and Limits of Freedom of the Press’ (1987) 16 Philosophy and Public Affairs 329; AT Kenyon, ‘Assuming Free Speech’ (2014) 77 MLR 379. 7 Jasanoff (n 4). 8 Conrad (n 3); Bronckers and van Gervens (n 2). 9 See, eg J-L Laffineur, ‘First ECJ Ruling on REACH: Choosing Registration over Exemption’ (2010) 22 Journal of Environmental Law 135; L Squintani, ‘Case Law of the Court of Justice of the European Union and the General Court: Reported Period 15.3.2015–31.8.2015’ (2015) 12 Journal for European Environmental & Planning Law 379; L Squintani, ‘Case Law of the Court of Justice of the European Union and the General Court: Reported Period 31.8.2015–31.12.2015’ (2016) 13 Journal for European Environmental & Planning Law 99.

Speaking and Governing through Freedom of Access to Environmental Information  175 boundaries of the right to access,10 on legal exceptions to the duty to produce information and on legal exceptions to the right of access, as discussed in section III.

I.  Freedom of Access to Information as a Basis for Speech In this section, we explain how both the ‘absence of obstacles, barriers or constraints’ and the ‘possibility of acting’ are applied to access to information.11 Sections II and III will present the legal framework and obstacles to access, and section IV will present experiences and possible options to develop autonomy and speech. Progress can be achieved not only by decreasing legal limitations (as double negative liberty) on access to information impeding the production of speech, but also by actively supporting (as positive liberty) the production of appropriate information that could be expected to lead to good decisions.12 As theorised by Lichtenberg for freedom of press,13 free speech is limited and constrained by the property rights of others, while the ability to publish is the condition of exercise of positive free speech. The principle of freedom of access developed to enable positive free speech is indeed facing restrictions to access based on property rights held by industry. As ‘free speech is commonly seen in negative terms as a limitation on government action that restricts speech’,14 counter-lobbying for the absence of exceptions or barriers, which can be seen as a negative liberty, is only a first step to improve access and positive liberty. Another way to envision free expression and reframe the debate on access to environmental information in positive terms is not only based on the absence of restrictions on access to the material. The possibility of autonomy, beyond the unrestricted availability of relevant data for effective reuse, requires ‘an obligation to act’ on the part of the government,15 understood in our case study as legal obligations for both the legislator and an empowered regulator to support free speech. The principle of access to information is established in international conventions, and European and national laws. In principle, the public may access information held by public authorities and previously produced by industries and provided to health and safety agencies for regulatory purposes, or to authorise or deny market access to substances based on toxicology studies. Consumers and workers may also directly access certain information held by industries. Freedom of information, one of the foundations of the right to access, together with the right to health, implies the capacity to check reports and data independently, and to reassess risks to health and the environment. As underlined by Gelber’s interpretation of Nussbaum’s ‘capabilities approach’, being able to have good health and to exercise political speech are items on a list of 10 capabilities.16 According to a negative approach to freedom of speech, 10 J Wates, ‘The Aarhus Convention: A Driving Force for Environmental Democracy’ (2005) 2 Journal for ­European Environmental & Planning Law 2. 11 Carter (n 5). 12 C Abbot and M Lee, ‘Economic Actors in EU Environmental Law’ (2015) 34 Yearbook of European Law 26. 13 Lichtenberg (n 6). 14 Kenyon (n 6) 379. 15 ibid. 16 K Gelber, ‘Nussbaum’s Capabilities Approach and Freedom of Speech’ in F Panzironi and K Gelber (eds), The Capability Approach: Development Practice and Public Policy in the Asia-Pacific Region (London, Routledge, 2012) 38 (quoting Nussbaum’s 2006 list).

176  Mélanie Dulong de Rosnay and Laura Maxim the government should only ‘refrain from interfering’: in our case, it would not devise access to information legislation, and would allow the market to decide the amount and level of information which was provided without endangering property rights and innovation. A positive approach for Gelber would, on the contrary, create public policy that ‘enhances’ the opportunity for a range of entities to speak. Peer review of data, methodologies and results is an essential part of the scientific process. This is all the more so in a context of conflict of interest, since the industry sponsors studies to assess the risk of its own products and activities.17 This democratic process, connected to the principle of reproducibility of scientific results, is deemed to allow the public to review the regulatory process, the study in question and the data underlying it. Hence, the public should be in a situation to exercise free speech and be able to alert industry, media, NGOs, public opinion and/or public authorities of possible health and environmental concerns about products or activities. In order to be effective, freedom of expression, relying on freedom of information or the public right to know, requires informed participation.18 The industry is expected to deliver useful data in meaningful ways once the government has created a positive obligation: Although the phrase the right to know is a useful generic designation, it is an inadequate description for the legal rights and obligations that govern the communication of workplace obligation on toxic substances. A person will not have a meaningful right to information unless someone else has a corresponding duty to provide that information.19

Ashford and Caldart further distinguish the duty to disclose, the duty to generate, the right of access and the duty to inform. Access on demand (pull) is different from online availability (push), with the latter being closer to a positive liberty. And because of restrictions based on copyright, the right of access does not necessarily include the right to reuse. If the right of reuse is not explicitly included in legislation on access to environmental information (or a generic law on freedom of access to and reuse of public sector information), actions of positive communication, such as republishing the results of toxicity studies or performing additional research mining several databases, can be impaired by copyright, legislative or contractual database rights of producers of information. The construction of the right of access can clash with a negative space, created by exceptions to exclusive rights of the producers and owners of information. Law does not appear to be ensuring appropriate access, since open science is restricted by a phenomenon known as data sequestration.20 Online availability can also be seen as useless or dangerous according to some stakeholders, especially in light of potential ‘big data’ use of raw data for commercial competition, while the data to be made available is not about commercial information but toxicity. These stakeholders will favour the approach of disclosing on demand (not as a positive obligation) in order to know who is asking.

17 D Michaels, ‘Foreword: Sarbanes-Oxley for Science’ (2006) 69(3) Law and Contemporary Problems 1. 18 Jasanoff (n 4). 19 NA Ashford and CC Caldart, Environmental Law, Policy, and Economics: Reclaiming the Environmental Agenda (Cambridge, MA, MIT Press, 2008) 775. 20 Michaels (n 17).

Speaking and Governing through Freedom of Access to Environmental Information  177

II.  Assessment of the Legal Framework for Access to Information The current European legal framework for access to information on health and environmental risks is scattered across: (i) sectoral regulation for health and environment (in Europe, the REACH Regulation,21 the Conventions of Aarhus22 and Tromsø,23 and the Directive on access to environmental information24); (ii) the Regulation on access to public sector information (PSI)25 and freedom of information; and (iii) legislation on copyright, personal data and industrial property. These pieces of legislation have different legal foundations (human rights, the right to know, safety and protection of human health and environment, but also innovation and competitiveness) and enforcement schemes. This can make it difficult for agencies, NGOs and citizens to understand which regime to invoke and how to use it. Small NGOs, s­ cientists and citizens also report difficulties in terms of accessing data to be able to exercise free speech and review or comment on such data, such as studies performed by the industry. As a consequence, public involvement in the governance of risks within agencies has been identified as limited compared to the involvement of industry.26 The effectiveness of access to information is determined by specific regulatory and institutional contexts. The entanglement of multiple laws does not clarify the situation or simplify the processing of making claims.

A.  Environmental Law in the USA While the focus of this chapter is European law, an incursion into US law helps to outline the general model that has subsequently been influential worldwide. The first high-profile regulation having access to information at its core was the Emergency and Community Right to Know Act of 1986. This was adopted in the USA two years after a release of methyl isocyanate at the Union Carbide chemical plant in Bhopal, India. The accident killed more than 2000 people. Industrial facilities were required to report yearly to the US Environmental Protection Agency the quantities of their toxic releases of certain substances. This information was then published in the USA’s first publicly accessible database of its type: the Toxic Release Inventory (TRI). Analysis of this regulation showed that access to information was effective and successful in reducing pollution only when the larger context included many other favourable factors. 21 Regulation (EC) No 1907/2006 of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) [2006] OJ L396/1. 22 United Nations Economic Commission for Europe (UNECE) Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters, 1998. 23 Convention of the Council of Europe on Access to Official Documents, 2009 (CETS 205). 24 Directive 2003/4/EC of 28 January 2003 on Public Access to Environmental Information and Repealing ­Council Directive 90/313/EEC [2003] OJ L41/26. 25 Directive 2013/37/EU of 26 June 2013 amending Directive 2003/98/EC on the Re-Use of Public Sector ­Information [2013] OJ L175/1. 26 Abbot and Lee (n 12).

178  Mélanie Dulong de Rosnay and Laura Maxim In the case of TRI, it depended, among other things, on the knowledge of the programme among companies, the degree to which environmental information was circulated to investors, the political colour of the state and federal political majorities at a given time, the likelihood of collective action against polluters, the awareness of the community bearing the risks and even fears about the use of information by terrorists.27 As expected, the information has been used by environmental groups to exercise free speech and pressure firms for reductions in pollution, and by legislators for changes in environmental policies. Another example of early access to information-based chemical policy is the Toxics Use Reduction Act (TURA). Adopted in the state of Massachusetts in 1989, this scheme aimed to change production processes and to reduce the use of toxic substances, and hence the quantities of toxic wastes produced, by companies. Its ambition was to reduce the quantity of toxic wastes in Massachusetts by 50 per cent within 10 years, using a strategy of public communication of information provided by companies and technical support to companies to identify possible improvements in their production processes. The scheme comprised positive actions to generate public speech and awareness. Industry had a unique responsibility to identify the use of a range of substances listed by the public authority, and – when a given threshold was exceeded – to submit to the Department of Environmental Protection a plan for waste reduction. This policy was clearly a success: having involved more than 943 companies, it led to a 40 per cent reduction in the quantity of toxic wastes between 1990 and 2005 and saved millions of dollars for the companies involved (eg through limiting operating costs by reducing both raw material and waste disposal/treatment costs). However, the ‘right to know’ did not play its expected role in TURA. Even if, theoretically, any individual could obtain the data collected from the industry, no one successfully used this mechanism.28 The public did not play its expected role of applying pressure to motivate companies, and the factors that most probably led to success were the economic interest of companies in reducing the costs of managing their waste and the personal relationships that public administrators created with the local industry. This concurs with the findings in relation to the TRI, which related the effectiveness of access to information to a larger political process, in which many other favourable factors needed to be involved.

B.  In Europe: REACH Institutional Settings to Produce and Share Data Since the very beginning, REACH was conceived as a regulation that would produce information about chemical risks. The ‘existing and new chemicals’ regulatory framework that preceded REACH required a risk assessment for any ‘new’ chemical released to the market. For the 100,106 ‘existing’ chemicals (those put on the market before 1981 and listed in the European Inventory of Existing Commercial Chemical Substances), the public authorities had to determine whether there was a need for examination and, if so, to conduct the

27 Hamilton (n 1). 28 D O’Rourke and E Lee, ‘Mandatory Planning for Environmental Innovation: Evaluating Regulatory Mechanisms for Toxics Use Reduction’ (2004) 47 Journal of Environmental Planning and Management 181.

Speaking and Governing through Freedom of Access to Environmental Information  179 necessary assessment. From 1993 to 2001, however, only 141 high-volume chemicals were singled out for risk assessment and the procedure was accomplished for only 27. By 2004, only four further assessments were implemented under Community legislation. REACH was thus intended to address to the enormous informational gap created by the underdevelopment of the earlier regime. Limited financial resources were the main reason for the failure of the ‘existing and new’ chemicals regulation before REACH. REACH sought to change this by requiring a company that would benefit commercially from selling a substance, also to pay the costs of testing its risks to humans and the environment. Hence, REACH led to the placing of the burden of proof on industry: the ‘no data, no market’ principle. Whereas the principle of placing the burden of proof on industry seems both pragmatic (for reasons of availability of information about the substance) and equitable (for the distribution of the testing effort between public and private arenas), however, the institutional setting created for producing and sharing data is at the core of the current difficulties of accessing information. This hampers the implementation of REACH in line with its original ambition. This institutional setting gives all the responsibility for data production to the industry, even if other options might have been possible, such as the charging of a public agency with gathering the information and requiring financial contribution from the industry.29 Alternative approaches could have better secured accountability, a recurrent problem identified by both the i­ ndustry and NGOs.30

III.  Limits to Access to Information The law is not effective in ensuring appropriate access to and reuse of information, as identified by Michaels31 and explained by the authors in a previous short paper.32 In this section, we identify limits to the right of access, which in turn impair the exercise of freedom of speech.

A.  Procedural Limits Beyond the issue of conflicts of interest (industry is asked to produce ‘objective’ data about the risks of its own substances, data which is also used to regulate the industry), data production by industry creates several problems related to access to information. First, the regulatory authorities in charge of implementing REACH are dependent on the format in which data is provided by industry. For example, regulators may have access to the abstracts of studies performed by the industry but not to the full studies. For a correct appreciation of

29 European Environmental Agency, Science and the Precautionary Principle: Lessons for Preventing Harm: Late Lessons from Early Warnings, vol II (Copenhagen, European Environmental Agency, 2013). 30 See Bronckers and van Gervens (n 2) and Abbot and Lee (n 12), respectively. 31 Michaels (n 17). 32 M Dulong de Rosnay and L Maxim, ‘L’ineffectivité du droit d’accès à l’information environnementale sur les risques chimiques’ (2012) 64(3) Hermès: La Revue 149.

180  Mélanie Dulong de Rosnay and Laura Maxim the quality of data provided by industry, however, access to the full study is most often vital. The current procedure – in which authorities have to write a letter of demand to an industry actor, which has no obligation to respond – places the authorities in a weak position relative to industry players. Industry can choose whether or not to provide the full study. This opens the way to a negotiation, one which is necessary if authorities are to perform their regulatory role but in which industry has no motivation to provide the information demanded. Furthermore, negotiation can contribute to creating undue proximity between industry and its regulator. Indeed, REACH promotes the principle of ‘no data, no market’, but neither ‘no good data, no market’ nor ‘no full data, no market’ is embodied in this principle. ‘Some’ data, be it partial and of low quality (sometimes studies several dozen years old are submitted by industry), is still ‘data’ and grants access to the market, unless the authorities are capable of reversing the evidence. Paradoxically, the regulation thus creates a situation in which the public authorities are disadvantaged in a power balance with the industry, as they cannot regulate substances that are potentially dangerous based on available information without being able to produce a sufficient level of proof of danger. Yet, to produce such levels of proof, public authorities need access to the widest information possible, including full industry studies. Secondly, when data is lacking, a public authority cannot simply demand a new study from industry. Rather, it has to demonstrate, often based on very scarce information only, that existing studies are insufficient and that producing new ones is both scientifically justified and proportional to allow a decision. Despite the original declared ambition of reversing the burden of proof by requiring producers and consumers to communicate information on toxicity, in reality REACH continues to put the burden of demonstrating the risk of a substance on regulators. Thirdly, high-quality information is also essential for limiting the impacts of chemicals on human health and the environment. Poor-quality scientific results can support poorquality decisions that fail to identify and limit risks that are nevertheless real. Most of the registration dossiers submitted by industry, however, do not respect the regulatory requirements. A report evaluating the registration dossiers for substances with a production volume of at least 1000 tons showed that 58 per cent of the screened dossiers were deficient in terms of the REACH information requirements.33 This was mainly because the test material did not correspond to the registered substance, and studies were not conducted in accordance with the appropriate guidelines. Only one dossier was found to be compliant, while no firm conclusion was reached in relation to 42 per cent of the dossiers. In other words, who produces the information, the institutional setting for its transmission from private to public stakeholders, the quantity, the quality and the specific form of the environmental and health information are all directly relevant to guaranteeing effective reuse. The general public is not the only stakeholder for whom such access is challenging. In REACH, public authorities need good-quality and complete information owned by the industry in order to regulate, but lack the regulatory means to obtain such information. The current regulations on access to information focus on access by the public to information

33 A Springer, H Herrmann, D Sittner, U Herbst and A Schulte, REACH Compliance: Data Availability of REACH Registration. Part 1: Screening of chemicals > 1000 tpa (Berlin, Bundesinstitut für Risikobewertung, 2015).

Speaking and Governing through Freedom of Access to Environmental Information  181 held by public authorities. But access by public authorities to information on health and environmental risks – directly relevant for public health but produced and owned by private stakeholders – is also essential. It represents the condition sine qua non both for the protection of public health and for the access of the public to information on risks.

B.  Access to Information Depends on its Communication Format and Tools Three forms of direct communication to the public have been envisaged in REACH: first, on demand from the European Chemicals Agency (ECHA); secondly, online on the ECHA’s website, which contains some freely available information from the registration dossiers; and thirdly, on demand from the industry. Any consumer can ask any supplier if their products contain Substances of Very High Concern (SVHC) that are listed on ECHA’s Candidate List. These substances have been identified as carcinogenic, mutagenic or toxic for reproduction, as persistent and bioaccumulative, or as warranting similar concern. Suppliers are legally obliged to provide an answer with at least the name of the SVHC, and information allowing safe use of the article, within 45 days, free of charge, if the substance is contained in a proportion of more than 0.1 per cent in the article in question. These two forms of direct communication are examined in turn here. EU citizens should have access to information about chemicals to which they may be exposed, in order to allow them to make informed decisions about their use of chemicals. A transparent means of achieving this is to grant them free and easy access to basic data held in the Agency’s database, including brief profiles of hazardous properties, labelling requirements and relevant Community legislation including authorised uses and risk management measures.34

For the ECHA’s website, whereas the objective is to communicate information to the general public, in practice the form of this information and the functionalities of the tool render it difficult. The tool is adapted to a level of understanding of a scientifically highly educated ‘public’, able to understand technical toxicological and physicochemical data, mastering the chemicals denominations, and having a clear idea about the link between certain substances and their use in daily life. Furthermore, the public targeted by the tool is expected to understand technical English. The ECHA has attempted to improve its communication of chemical information. Since 2016, presentation of information on up to 120,000 chemicals is structured in three layers: InfoCard, brief profile and detailed source data. The InfoCard is a dissemination tool that is intended to make the technical information published in the ECHA’s substance databases more accessible to the general public, but the information still remains highly technical and synthetic. For the example of bisphenol A, a substance which has been highly debated worldwide for more than 20 years, the InfoCard does not include any harm prevention measures available to consumers, but only those suggested by manufacturers and importers that are oriented to their own employees and downstream users (ie c­ ompanies or individuals using chemicals in their industrial or professional activities).



34 REACH

(n 21) Art 117.

182  Mélanie Dulong de Rosnay and Laura Maxim Figure 1  Partial screen shot of results for the search of the substance ‘bisphenol A’ on the ECHA’s home page (11 May 2017)

The Brief Profile summarises non-confidential data on substances held in the ECHA databases, but the website warns that this information is automatically generated and therefore may not be fully up to date or complete. Indeed, the data included is only that submitted by the registrant company. As for access to information on demand directly from the suppliers of articles containing Substances of Very High Concern, a report produced by European Environmental Bureau showed that it remained largely unapplied.35 Of 158 ‘right to know’ requests sent to 60 brands in Europe, 50 per cent remained entirely unanswered, while over 75 per cent received answers that did not fulfil REACH requirements.

C.  Access to Information Depends on the Power Balance that is in Place Besides the generic framework for access to environmental information stipulated in the Aarhus convention to support ‘environmental democracy’,36 each specific regulation creates the conditions for the exercise of this right. As discussed above, REACH is a typical example of compromise-based policy. It prolongs the trend in the environmental regulation, and in self-regulation in general, to influence industry’s practices through continuous negotiation and persuasion rather than constraints (which are hard to implement in any event).

35 European Environmental Bureau, The Fight to Know? Substances of Very High Concern and the Citizens’ Right to Know Under REACH (Brussels, European Environmental Bureau, 2010). 36 Wates (n 10).

Speaking and Governing through Freedom of Access to Environmental Information  183 Behind the apparently strict rules of REACH, the complexity of chemical risks and the unequal capacities of public authorities and industry to produce information lead to implementation that strongly depends on communication and agreement between these two parties. This has two consequences for access to information for NGOs and the public. First, as compared with relationships with other stakeholders, there is a privileged relationship between public authorities and industry. Without communication with industry, concrete implementation of the regulation is impossible. Secondly, public authorities are overloaded due to the complexity of REACH, and have little time remaining for other duties such as communication to the public. In addition, some authorities may consider time invested in public communication to be less useful and effective in terms of public health than their main mission, which is the implementation of the regulation. Even with its more than 500 pages, REACH cannot cover all the potential situations that may appear during its implementation, often dependent on the substance addressed and the industry concerned. Therefore, a pragmatic ‘jurisprudence’ comes into being during implementation. Informal rules based on experience, agreed between the public and private stakeholders, become standards for further similar situations and substances. Access to such informal rules and information thereon is possible only for stakeholders that are directly involved in the process of implementation, such as employees of, and experts working with, public authorities (ECHA, health and environmental agencies of Member States, national ministries) and the industries concerned. In addition to public access, some NGOs can be granted access to information directly through procedures specifically dedicated to participation in ECHA’s work. Such interaction with stakeholders is directed to accredited NGOs, which, in terms of exercising access to information, have advantages in terms of direct contact with ECHA and access to informal information produced during the implementation process. The downside of such participation is that NGOs that are accredited by ECHA for participation have to work within the logic of compromise.37 Potentially, this may limit their propensity to use information for frank criticism and effective free speech. Whereas access to information increases the capability of non-governmental stakeholders to act, the initial objective of controlling industry action through the empowerment of such stakeholders remains a myth. The forces in place are largely unbalanced;38 the human competences and financial resources available to industry to produce and control information are significantly greater than those of NGOs.39 Furthermore, access by industry to public regulators is significantly easier, given that regulation makes industry the main discussion partner for governmental health agencies.

D.  Substantial Limits Alongside procedural limits on rights of access to and reuse of information, a series of substantive limits also impede such engagement. These include the confidentiality of business information, successful contentions that information obligations serve as barriers to



37 See

http://echa.europa.eu/about-us/partners-and-networks/stakeholders. example, of the list of stakeholders accredited to work with ECHA, 71% are industry associations. 39 Abbot and Lee (n 12). 38 For

184  Mélanie Dulong de Rosnay and Laura Maxim trade, and the nature and contours of information that is required to be disclosed under regimes.

E.  Confidential Business Information Confidentiality is another obstacle that limits the scope of information that can be accessed. As explained in REACH: [only] non-confidential information on chemicals is to be available, for example to allow those exposed to chemicals to take decisions as to the acceptability of the associated risks. Some information is accessible free of charge on the Agency’s website, other information on request. However, the Agency may not disclose confidential company data.

The first exception to the general principle of access to sanitary and environmental information is stated in all access to information laws that foresee exceptions for confidentiality, trade secrets and intellectual property, without defining these notions or providing guidance for agencies on interpreting what may fall within the category of confidential business information. Thus, industry players have the option of claiming the protection of confidentiality and intellectual property rights to dismiss demands of access to information and/ or to oppose the publication or further distribution of information they are compelled to provide to regulatory agencies for evaluation and authorisation purposes.40 In practice, confidential data and data protected by intellectual property can be withheld by the industry, or redacted from documents delivered to the state agency and/or the public. Agencies have a duty to protect confidential business information. The ECHA website publishes evaluation decisions in ‘non-confidential versions of the decisions originating from compliance checks and examination of testing proposals’. Hence, there is a risk of abuse of the confidentiality exception, which companies may interpret expansively and apply to environmental and health data without commercial interest or risk of competition (for example, to the results of a toxicity test).

F.  Restriction of Trade Notwithstanding its seeming laxity, REACH is the subject of stringent criticism by free trade advocates. Implementing its complex obligations is said to discriminate against non-EU manufacturers and non-EU World Trade Organization members.41 Chemical regulation requiring the disclosure of information is seen as a cost for industry, in the same way as they lobby against obligations such as information labelling requirements under consumer law. Similarly, in the field of data protection, the right to privacy is also seen as a barrier to trade by industry. Producing information and providing access to this information is assimilated to a nontariff barrier to trade. This categorisation was seen in a leaked joint document prepared 40 Bronckers and van Gervens (n 2). 41 LA Kogan, ‘REACH Revisited: A Framework for Evaluating Whether a Non-tariff Measure Has Matured into an Actionable Non-tariff Barrier to Trade’ (2013) 28 American University International Law Review 489.

Speaking and Governing through Freedom of Access to Environmental Information  185 by the American Chemistry Council and European Chemical Industry Council for the purposes of the Trans-Atlantic Trade and Investment Partnership (TTIP) free trade agreement negotiations, and positive obligations may end up being diminished.42

G.  Nature and Contours of Information Sanitary and environmental information produced by industry to fulfil legal obligations to safety agencies to evaluate product risks before market approval include reports and underlying toxicological data. The breadth of the information to be produced and released to the public is subject to interpretation and negotiation. For example, studies may be based on too small a sample or be of too limited a duration, or results may be analysed using an inadequate statistical tool. The analysis of toxicology studies related to pollution and human health requires access to the experimental protocols used in the studies. The assumptions in these protocols are crucial for evaluating possible bias or flaws in the study, and the need to rerun alternative experiments. Also, the regulatory requirements themselves – which resulted from negotiations between stakeholders and not from ‘purely scientific’ input – can be a source of uncertainty and even scientific flaws. For instance, in chemical risks assessment, research identified that a small dose rather than a large dose will be poisonous;43 therefore, the dosages chosen in testing will significantly influence the results. The level of information provided by the industry depends on the amount of the substance produced or imported. More information must be provided when higher tonnages of the substance are produced or imported, based on the assumption that higher tonnage means greater exposure and hence greater potential effects.44 The information that might be relevant for science (small quantities can be harmful) therefore contradicts such quantity-based legal requirements, which assume that larger quantities are necessarily more problematic. Conflicts about the contours of the information to be communicated are also frequent. The first landmark case law on REACH helped define the scope of exceptions for polymers.45 Other rulings helped to shape the contours of the communication duties to ECHA and to consumers under REACH.46 Seeking access to information on several products individually does not have the same impact as asking for information on a complex end product as a whole. In the ClientEarth case, two NGOs sued ECHA to force it to release tonnage information and exact production quantities in order to pressure companies to adopt safer alternatives and to help consumers make informed choices. The European Court of Justice concluded, however, that the ‘ECHA was entitled to take the view that the disclosure of the precise tonnage of the substances registered would have undermined the commercial ­interests of the persons concerned’.47 42 American Chemistry Council–European Chemical Industry Council joint proposal enhancing US–EU ­chemical regulatory cooperation under TTIP: see http://ciel.org/Publications/CH_Pro.pdf; M Karlsson, ‘TTIP and the Environment: The Case of Chemicals Policy’ (2015) 1 Global Affairs 21. 43 European Environmental Agency, Chemicals in the European Environment: Low Doses, High Stakes? (­Copenhagen, European Environmental Agency, 1997). 44 Laffineur (n 9). 45 ibid. 46 Squintani (n 9). 47 T-245/11 ClientEarth v European Chemicals Agency (ECHA) [2015] ECJ.

186  Mélanie Dulong de Rosnay and Laura Maxim

IV.  Institutional Changes to Develop Capacities The two previous sections revealed institutional, epistemological and legal limits on access to information and their consequences for the possibility, the effectiveness and the quality of subsequent public speech. In this section, we examine possible changes to the governance of risk-based access to information which could implement a positive approach to free speech, beyond the mere negotiations of its limits. In the two parts of this section, we consider possible action points which require a positive action from the government: • Developing the role of regulatory agencies and third parties to address some of the obstacles identified in the previous sections, as the process of maintaining secrecy is easily abused.48 • Applying open data and open science principles to environmental law.

A.  Developing the Role of Regulatory Agencies and Third Parties Unbalanced institutional settings and the definition and interpretation of confidentiality have both been identified as internal limits to effective access to information. Accountability for providing misleading information to the regulator and the public could be developed and a person could be designated to evaluate studies and methods, and to justify claims to confidentiality.49 This proposition, inspired by the quite different field of accounting regulation, would have some similarities with the possible ways to develop positive free speech as theorised by Gelber: seeking to limit harmful speech (understood as misleading information), if confidentiality, which is often claimed to maintain secrecy, is misused and therefore used in a harmful way.50 Meaningful and effective legislation on access to sanitary and environmental information could define confidentiality and commercial interest, and develop mechanisms – perhaps using third parties – independently to assess the legitimacy of claims of confidentiality for business information. If access to information is envisioned as a positive right, exceptions should be interpreted narrowly. An alternative option would be to entrust intermediaries – trusted third parties – to anonymise reports or to remove confidential information from them. Techniques for anonymising parties in lawsuits could be developed for confidential business information, as often it is not the commercial aspect of the substance which is of interest in assessing the risk. The interpretation of data and the ability of citizens to process it are keys to the assessment of the risk of substances. Increasing the power of agencies to communicate results and to mediate information on risks (as media, whistleblowers and academics would do) is a solution which could be endorsed by industry. It would be up to Member States to



48 Michaels 49 ibid.

50 Gelber

(n 17).

(n 16).

Speaking and Governing through Freedom of Access to Environmental Information  187 relinquish power to regulators such as ECHA,51 and to increase their budgets to transform them into such watchdogs to inform the public.

B.  Importation of Open Data and Open Science Principles into Environmental Law The double influence of open copyright policy and citizen science participatory projects can contribute to the development of positive free speech in the field of environmental law. Yet, unlimited transparency is not necessarily the best way forward. The field of science and technology studies provides a useful counterbalance to openness, participation evangelism and techno-idealism. First, the process of transforming data into relevant information and improving the capacities of citizens to speak and NGOs to challenge regulatory decisions is not straightforward. More data, even if it is better data, is not immediately transformed into actionable knowledge. Secondly, excessive transparency can be a source of continuing doubt: political decisions could be postponed, while facts are questioned over and over again. Jasanoff theorises that some ‘black-boxing’ – relying on Latour – is essential for robust claims, avoiding the ‘manufacturing of uncertainty’ for political ends.52 The mere right of access, or the right to read, does not amount to the right to mine databases and make new scientific findings from data produced by others (equivalent to a right to read by computers). Indeed, copyright can limit the effectiveness of freedom of information laws, in terms of reusing information obtained under them, beyond reading it. The 2013 revised directive on public sector information53 introduced a right to reuse documents which can be accessed under national access to documents laws. It only invites Member States to make more documents available in machine-readable and standard open formats: there are no compulsory provisions regarding the nature and the contours of information to facilitate scientific processing and peer-review by NGOs. Nevertheless, the experience of copyright licensing under generic freedom of information laws suggests that freedom of access to environmental information laws should include provisions to allow effective reuse and data processing, require open formats and prohibit, or make void, restrictive terms of use by databases producers. There would be a risk of having access to data, but not being able to interpret it and not being able to make it speak to the public. A mere right of access, giving an illusion of transparency, is not sufficient. A right of reuse is also needed. Access to information, allowing reports to be read, is a necessary but insufficient condition for the exercise of speech. NGOs, and potentially citizens, need to be able to reproduce, republish and mine databases. According to open science principles, a genuine open access allowing others to reproduce experiments and to verify results requires legal and technical accessibility.54 According to

51 Bronckers and van Gervens (n 2). 52 Jasanoff (n 4); B Latour, Science in Action: How to Follow Scientists and Engineers through Society (Cambridge, MA, Harvard University Press, 1987). 53 Directive 2013/37/EC (n 25). 54 M Dulong de Rosnay, ‘From Free Culture to Open Data: Technical Requirements for Open Access’ in D  ­Bourcier, P Casanovas, M Dulong de Rosnay and C Maracke (eds), Intelligent Multimedia: Sharing Creative Works in a Digital World (Florence, European Press Academic Publishing, 2010) 47.

188  Mélanie Dulong de Rosnay and Laura Maxim open data advocates, the right to read and the right to process and perform data mining55 and meta-analysis requires open, downloadable data formats. Government can also nudge or prompt citizens to reuse the open data it retains. The French Minister of Environment’s 2016 ‘roadmap’ proposed to create a general supervisor for data within the ministry. This new role is expected to develop a culture of data, both within the ministry and within the public, and to empower citizens through the organisation of hackathons to analyse, visualise, process, mine and otherwise exploit and communicate available data to the public.56 Crossing boundaries across public policy and law is essential for developing the best practices for data producers to provide data which can be reused in meaningful ways.

V. Conclusion Access to information should allow civil society to send critical feedback to industry on the risks of its substances and hence help regulators in managing those risks. The practice is far more complicated: access to information would at the first sight appear to be granted by existing legal texts, but numerous opportunities exist in practice for reducing access. Our analysis supports the conclusion that the practice of access to environmental information fails to reach its full potential, namely contributing to the protection of public health in Europe. Policy debates rarely move along the lines from negative to positive speech, understood as the ‘architecture’ or conditions which allow speech to occur. Such an approach could also be applied to other heavily lobbied fields, such as intellectual property, where more time is spent legislating on the scope of exceptions than on the development of policies to support the development and release of information. In the line of the introducing trusted third parties to receive business information and remove confidential information from it, a prospective approach to resolving conflicts on the disclosure of information could consist in applying commons-based governance models to the governance of information on chemical risks. Mandatory data sharing about substances within Substance Information Exchange Forums with other industry members working on similar substances is a mechanism which has raised opposition from the industry as the governance was not consensually negotiated by all members.57 Instead, such forums could be managed as ‘common pool resources, with mechanisms allowing collective benefits sharing, monitoring of usages, surveillance of free riders and a range of sanctions’.58 This approach has been conceptualised for microbial data.59 Sharing biological material and data on chemical risk are equally challenging, as they contain potential competitive 55 P Murray-Rust, ‘The Right to Read is the Right to Mine’, Open Knowledge Foundation Blog, 1 June 2012, https://blog.okfn.org/2012/06/01/the-right-to-read-is-the-right-to-mine. 56 L Monnoyer-Smith, Rapport de préfiguration de la fonction de superviseur général des données du ministère de l’environnement (Paris, Ministère de l’Environnement, de l’Energie et de la Mer, 2016). 57 Bronckers and van Gerven (n 2). 58 E Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge, Cambridge University Press, 1990). 59 JH Reichman, PF Uhlir and T Dedeurwaerdere, Governing Digitally Integrated Genetic Resources, Data, and Literature: Global Intellectual Property Strategies for a Redesigned Microbial Research Commons (Cambridge, Cambridge University Press, 2016).

Speaking and Governing through Freedom of Access to Environmental Information  189 information. Managing to communicate data which is useful for science, without disclosing the commercial-only value, requires the protection of participants’ commercial opportunities. But this has to occur in a manner that does not amount to intellectual property and claims of confidentiality disrupting the level of access needed to perform some types of research and toxicity control. Commons-based pool arrangements for science are capable of supporting both commercial and public interests, enhancing the opportunities of all parties to speak and produce meaningful data, and could be considered as desirable governance arrangements to manage information on chemical risks.

190

BIBLIOGRAPHY Abbot, C and Lee, M, ‘Economic Actors in EU Environmental Law’ (2015) 34 Yearbook of European Law 26. Ackerman, JM and Sandoval-Ballesteros, IE, ‘The Global Explosion of Freedom of Information Laws’ (2006) 58 Administrative Law Review 85. Alexy, R, A Theory of Constitutional Rights, trans J Rivers (Oxford, Oxford University Press, 2002). Amos, M, Human Rights Law, 2nd edn (Oxford, Hart Publishing, 2014). Ananny, M, Networked Press Freedom: Creating Infrastructures for a Public Right to Hear (Cambridge, MA, MIT Press, 2018). Anderson, D, Control Orders in 2011: Final Report of The Independent Reviewer on The Prevention of Terrorism Act 2005 (London, Stationery Office, 2012). Aplin, T and Bosland, J, ‘The Uncertain Landscape of Article 8 of the ECHR: The Protection of Reputation as a Fundamental Human Right?’ in AT Kenyon (ed), Comparative Defamation and Privacy Law (Cambridge, Cambridge University Press, 2016) 265–90. Aplin, T, Bently, L, Johnson, P and Malynicz, S, Gurry on Breach of Confidence: The Protection of Confidential Information, 2nd edn (Oxford, Oxford University Press, 2012). Arriaza Ibarra, K, ‘Management and Organization of Public Service Media Companies: Basic Concepts Related to Efficiency and Failure in European Public Service Media, with Two Case Examples (Germany and Spain)’ in K Arriaza Ibarra, E Nowak and R Kuhn (eds), Public Service Media in Europe: A Comparative Approach (Abingdon, Routledge, 2015) 147–69. Ashford, NA and Caldart, CC, Environmental Law, Policy, and Economics: Reclaiming the Environmental Agenda (Cambridge, MA, MIT Press, 2008). Bakan, J, Just Words: Constitutional Rights and Social Wrongs (Toronto, University of Toronto Press, 1997). Baker, CE, ‘Scope of the First Amendment Freedom of Speech’ (1978) 25 UCLA Law Review 964. —— Human Liberty and Freedom of Speech (Oxford, Oxford University Press, 1989). —— Media, Markets, and Democracy (Cambridge, Cambridge University Press, 2002). —— Media Concentration and Democracy: Why Ownership Matters (Cambridge, Cambridge University Press, 2007). Balkin, JM, ‘Some Realism about Pluralism: Legal Realist Approaches to the First Amendment’ [1990] Duke Law Journal 375. Barendt, E, ‘The Influence of the German and Italian Constitutional Courts on Their National Broadcasting Systems’ [1991] Public Law 93. —— Freedom of Speech, 2nd edn (Oxford, Oxford University Press, 2005). Barnett, S and Townend, J (eds), Media Power and Plurality: From Hyperlocal to High-Level Policy (Basingstoke, Palgrave Macmillan, 2015). Barron, JA, ‘Access to the Press: A New First Amendment Right’ (1967) 80 Harvard Law Review 1641. —— Freedom of the Press for Whom? The Right of Access to the Mass Media (Bloomington, University of Indiana Press, 1973). Berlin, I, Two Concepts of Liberty: An Inaugural Lecture Delivered before the University of Oxford on 31 October 1958 (Oxford, Clarendon, 1958). —— ‘Two Concepts of Liberty’ in I Berlin, Four Essays on Liberty (Oxford, Oxford University Press, 1969) 118–72. —— ‘Introduction’ in Four Essays on Liberty (Oxford, Oxford University Press, 1969) ix–lxiii. Birkinshaw, P, ‘Freedom of Information and Openness: Fundamental Human Rights?’ (2007) 58 Administrative Law Review 177. Bishop, CA, Access to Information as a Human Right (El Paso, LFB Scholarly, 2012). Blair, T, A Journey (London, Hutchinson, 2010). Blasi, V, ‘The Checking Value in First Amendment Theory’ (1977) 2 American Bar Foundation Research Journal 521.

192  Bibliography Bohlander, M, ‘Open Justice or Open Season? Should the Media Report the Names of Suspects and Defendants?’ (2010) 74 Journal of Criminal Law 321. Boran, I, ‘Global Linguistic Diversity, Public Goods and the Principle of Fairness’ in W Kymlicka and A Patten (eds), Language Rights and Political Theory (Oxford, Oxford University Press, 2003) 189–209. Boyron, S, The Constitution of France: A Contextual Analysis (Oxford, Hart Publishing, 2013). Breyer, S, Active Liberty: Interpreting a Democratic Constitution (Oxford, Oxford University Press, 2008). Brighouse, H, ‘Democracy and Inequality’ in A Carter and G Stokes (eds), Democratic Theory Today: Challenges for the 21st Century (Cambridge, Polity, 2001) 52–72. Bronckers, M and van Gerven, Y, ‘Legal Remedies Under the EC’s New Chemicals Legislation REACH: Testing a New Model of European Governance’ (2009) 46 Common Market Law Review 1823. Broughton Micova, S, ‘Born into Crisis: Public Service Broadcasters in South East Europe’ in GF Lowe and J Steemers (eds), Regaining the Initiative for Public Service Media (Gothenburg, Nordicom, 2012) 131–47. —— ‘Rights vs Reality: Minority Language Broadcasting in South East Europe’ (2013) 12 Journal on Ethnopolitics and Minority Issues in Europe 54. Calliess, C, ‘Dimensions of Fundamental Rights: Duty to Respect versus Duty to Protect’ in H Pünder and C Waldhoff (eds), Debates in German Public Law (Oxford, Hart Publishing, 2014) 27–42. Carolan, E and Spina, A, ‘Behavioural Sciences and EU Data Protection Law: Challenges and Opportunities’ in A Alemanno and A-L Sibony (eds), Nudge and the Law: What Can EU Law Learn from Behavioural Sciences? (Oxford, Hart Publishing, 2015) 161–78. Carter, I, ‘Positive and Negative Liberty’ in Edward N Zalta (ed), The Stanford Encyclopaedia of Philosophy (Summer 2018 edition) https://plato.stanford.edu/archives/sum2018/entries/liberty-positive-negative. Chandler, JA, ‘A Right to Reach an Audience: An Approach to Intermediary Bias on the Internet’ (2007) 35 Hofstra Law Review 1095. Clayton, R, ‘The Empire Strikes Back: Common Law Rights and the Human Rights Act’ [2015] Public Law 3. Cobain, I, The History of Thieves: Secrets, Lies and the Shaping of a Modern Nation (London, Granta, 2016). Cohen, J, ‘Money, Politics and Political Equality’ in A Byrne, R Stalnaker and R Wedgwood (eds), Fact and Value: Essays on Ethics and Metaphysics for Judith Jarvis Thomson (Cambridge, MA, MIT Press, 2001) 47–80. Conrad, J, ‘Open Secrets: The Widespread Availability of Information about the Health and Environmental Effects of Chemicals (2006) 69(3) Law and Contemporary Problems 141. Cook, TE, Governing with the News: The News Media as a Political Institution, 2nd edn (Chicago, University of Chicago Press, 2005). Coppel, P, Information Rights: Law and Practice, 4th edn (Oxford, Hart Publishing, 2014). Couldry, N, Why Voice Matters: Culture and Politics after Neoliberalism (London, Sage, 2010). Cripps, Y, ‘The Public Interest Disclosure Act 1998’ in J Beatson and Y Cripps (eds), Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams (Oxford, Oxford University Press, 2000) 275–87. Crossman, R, ‘The Real English Disease’, New Statesman, 24 September 1971. Curran, J, Media and Democracy (London, Routledge, 2011). Curtis, MK, ‘Democratic Ideals and Media Realities: A Puzzling Free Press Paradox’ (2004) 21 Social Philosophy and Policy 385. Daintith, T, ‘Regulation by Contract: The New Prerogative’ (1979) 32 Current Legal Problems 41. Dan-Cohen, M, ‘Freedoms of Collective Speech: A Theory of Protected Communications by Organizations, Communities, and the State’ (1991) 79 California Law Review 1229. Davies, ACL, ‘Ultra Vires Problems in Government Contracts’ [2006] Law Quarterly Review 98. Davies, N, Hack Attack: How the Truth Caught Up with Rupert Murdoch (London, Chatto & Windus, 2014). Delany, H and Carolan, E, The Right to Privacy: A Doctrinal and Comparative Analysis (Dublin, Round Hall, 2007). Delgado, R, ‘First Amendment Formalism Is Giving Way to First Amendment Legal Realism’ (1994) 29 Harvard Civil Rights-Civil Liberties Law Review 169. Dryzek, JS and Dunleavy, P, Theories of the Democratic State (Basingstoke, Palgrave, 2009). Dulong de Rosnay, M, ‘From Free Culture to Open Data: Technical Requirements for Open Access’ in D Bourcier, P Casanovas, M Dulong de Rosnay and C Maracke (eds), Intelligent Multimedia: Sharing Creative Works in a Digital World (Florence, European Press Academic Publishing, 2010) 47–66. Dulong de Rosnay, M and Maxim, L, ‘L’ineffectivité du droit d’accès à l’information environnementale sur les risques chimiques’ (2012) 64(3) Hermès: La Revue 149. Durham, WC, ‘General Assessment of the Basic Law: An American View’ in P Kirchhof and DP Kommers (eds), Germany and Its Basic Law: Past, Present and Future (Baden-Baden, Nomos, 1993) 37–63.

Bibliography  193 Dworkin, R, Sovereign Virtue: The Theory and Practice of Equality (Cambridge, MA, Harvard University Press, 2000). Emerson, TI, The System of Freedom of Expression (New York, Random House, 1970). Estlund, D, ‘Political Quality’ (2000) 17 Social Philosophy and Policy 127. European Environmental Agency, Chemicals in the European Environment: Low Doses, High Stakes? (Copenhagen, European Environmental Agency, 1997). —— Science and the Precautionary Principle: Lessons for Preventing Harm: Late Lessons from Early Warnings, vol II (Copenhagen, European Environmental Agency, 2013). European Environmental Bureau, The Fight to Know? Substances of Very High Concern and the Citizens’ Right to Know under REACH (Brussels, European Environmental Bureau, 2010). Fallon Jr, RH, ‘Post on Public Discourse under the First Amendment’ (1990) 103 Harvard Law Review 1738. Favoreu, L and Mastor, W, Les cours constitutionelles, 2nd edn (Paris, Dalloz, 2016). Fiss, OM, ‘Free Speech and Social Structure’ (1986) 71 Iowa Law Review 1405. —— The Irony of Free Speech (Cambridge, MA, Harvard University Press, 1996). Folami, AN, ‘Using the Press Clause to Amplify Civic Discourse beyond Mere Opinion Sharing’ (2013) 85 Temple Law Review 269. Fraser, N, ‘Rethinking the Public Sphere: A Contribution to the Critique of Actually Existing Democracy’ (1990) 25/26 Social Text 56. —— ‘Recognition without Ethics?’ (2001) 18(2–3) Theory, Culture & Society 21. Fredman, S, Human Rights Transformed: Positive Rights and Positive Duties (Oxford, Oxford University Press, 2008). —— Comparative Human Rights Law (Oxford, Oxford University Press, 2018). Freedland, M, ‘Government by Contract and Public Law’ [1994] Public Law 86. Freedman, D, The Contradictions of Media Power (London, Bloomsbury, 2014). Freedman, D, Curran, J and Puttnam, D, A Future for Public Service Television: Content and Platforms in a Digital World (London, Goldsmiths, 2016) www.futureoftv.org.uk. Gearty, C, On Fantasy Island (Oxford, Oxford University Press, 2016). Gelber, K, ‘Nussbaum’s Capabilities Approach and Freedom of Speech’ in F Panzironi and K Gelber (eds), The Capability Approach: Development Practice and Public Policy in the Asia-Pacific Region (London, Routledge, 2012) 38–52. Gibbons, T, ‘Free Speech, Communication and the State’ in M Amos, J Harrison and L Woods (eds), Freedom of Expression and the Media (Leiden, Martinus Nijhoff, 2012) 19–44. —— ‘Active Pluralism: Dialogue and Engagement as Basic Media Policy Principles’ (2015) 9 International Journal of Communication 1382. Gibbons, T and Humphreys, P, Audiovisual Regulation under Pressure: Comparative Cases from North America and Europe (Abingdon, Routledge, 2012). Gould, CC, Rethinking Democracy: Freedom and Social Cooperation in Politics, Economy, and Society (Cambridge, Cambridge University Press, 1988). Graber, MA, Transforming Free Speech: The Ambiguous Legacy of Civil Libertarianism (Berkeley, University of California Press, 1991). Grimm, D, ‘The Protective Function of the State’ in G Nolte (ed), European and US Constitutionalism (Cambridge, Cambridge University Press, 2005) 137–55. Grin, F, ‘Diversity as a Paradigm, Analytical Device and Policy Goal’ in W Kymlicka and A Patten (eds), Language Rights and Political Theory (Oxford, Oxford University Press, 2003) 169–88. Grzeszick, B, ‘The “Serving” Freedom to Broadcast: Subjective versus Objective Dimensions of a Fundamental Right’ in H Pünder and C Waldhoff (eds), Debates in German Public Law (Oxford, Hart Publishing, 2014) 75–95. Habermas, J, The Theory of Communicative Action: vol I – Reason and the Rationalization of Society, trans T McCarthy (Boston, MA, Beacon Press, 1984). —— Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans W Rehg (Cambridge, Polity, 1996). —— The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society, trans T Burger (Oxford, Polity, 1989). Hallin, DC and Mancini, P, Comparing Media Systems: Three Models of Media and Politics (Cambridge, Cambridge University Press, 2004).

194  Bibliography Hamilton, JT, Regulation through Revelation: The Origin, Politics and Impacts of the Toxic Release Inventory Program (Cambridge, Cambridge University Press, 2005). Hanning, J and Mulcaire, G, The News Machine: Hacking – the Untold Story (London, Gibson Square, 2014). Harris, DJ, O’Boyle, M, Bates, EP and Buckley, CM, Harris, O’Boyle and Warbrick: Law of the European Convention on Human Rights, 3rd edn (Oxford, Oxford University Press, 2014). Held, D, Models of Democracy (Stanford, Stanford University Press, 1987); 3rd edn (Cambridge, Polity Press, 2006). Held, V, Rights and Goods: Justifying Social Action (New York, Free Press, 1984). Hennessy, P, Whitehall (London, Secker & Warburg, 1989). Heun, W, The Constitution of Germany: A Contextual Analysis (Oxford, Hart Publishing, 2011). Hindman, M, The Myth of Digital Democracy (Princeton, Princeton University Press, 2009). Hins, W and Voorhoof, D, ‘Access to State-Held Information as a Fundamental Right under the European Convention on Human Rights’ (2007) European Constitutional Law Review 114. Hirschmann, NJ, The Subject of Liberty: Toward a Feminist Theory of Freedom (Princeton, Princeton University Press, 2002). Hitchens, L, Broadcasting Pluralism and Diversity: A Comparative Study of Policy and Regulation (Oxford, Hart Publishing, 2007). Hocking, WE, Freedom of the Press: A Framework of Principle, A Report from the Commission on Freedom of the Press (Chicago, University of Chicago Press, 1947). Hoffmann-Riem, W, Regulating Media: The Licensing and Supervision of Broadcasting in Six Countries (New York, Guilford Press, 1996). Hogan, GW and Whyte, GF, JM Kelly: The Irish Constitution, 4th edn (Haywards Heath, Tottel, 2003). Honohan, I (ed), Republicanism in Ireland: Confronting Theories and Traditions (Manchester, Manchester University Press, 2008). Hood, C and Heald, D (eds), Transparency: The Key to Better Governance? (Oxford, Oxford University Press, 2006). Horwitz, RB, ‘The First Amendment Meets Some New Technologies: Broadcasting, Common Carriers, and Free Speech in the 1990s’ (1991) 20 Theory and Society 21. Hutchinson, AC, ‘Talking the Good Life: From Free Speech to Democratic Dialogue’ (1989) 1 Yale Journal of Law and Liberation 17. Irving, H and Townend, J, ‘Censorship and National Security: Information Control in the Second World War and Present Day’ (2016) History & Policy, www.historyandpolicy.org/policy-papers/papers/censorship-andnational-security-information-control. Jaconelli, J, Open Justice: A Critique of the Public Trial (Oxford, Oxford University Press, 2002). Jasanoff, S, ‘Transparency in Public Science: Purposes, Reasons, Limits’ (2006) 69(3) Law and Contemporary Problems 21. Karlsson, M, ‘TTIP and the Environment: The Case of Chemicals Policy’ (2015) 1 Global Affairs 21. Katsirea, I, Public Broadcasting and European Law (Alphen aan den Rijn, Kluwer Law International, 2008). Keane, J, The Media and Democracy (Cambridge, Polity Press, 1991). —— The Life and Death of Democracy (London, Simon & Schuster, 2009). —— Democracy and Media Decadence (Cambridge, Cambridge University Press, 2013). —— Power and Humility: The Future of Monitory Democracy (Cambridge, Cambridge University Press, 2018). Keller, P, European and International Media Law (Oxford, Oxford University Press, 2011). Kenyon, AT, ‘Assuming Free Speech’ (2014) 77 Modern Law Review 379. —— ‘Who, What, Why and How: Questions for Positive Free Speech and Media Systems’ in M Edström, AT  Kenyon and E-M Svensson (eds), Blurring the Lines: Market-Driven and Democracy-Driven Freedom of Expression (Gothenburg, Nordicom, 2016) 29–39. —— ‘Positive Free Speech: A Democratic Freedom’ in A Stone and F Schauer (eds), Oxford Handbook on Freedom of Speech (Oxford, Oxford University Press, forthcoming). —— Democracy of Expression: Positive Free Speech and Law (Cambridge, Cambridge University Press, forthcoming). Klimkiewicz, B, ‘Media Policy for Ethnic and National Minorities in Poland, the Czech Republic and Slovakia’ in M Sükösd and P Bajomi-Lázár (eds), Reinventing Media: Media Policy Reform in East-Central Europe (Budapest, Central European University Press, 2003) 155–84. Knight, C, ‘Article 10 and the Right of Access to Information’ [2013] Public Law 468. Kogan, LA, ‘REACH Revisited: A Framework for Evaluating Whether a Non-tariff Measure Has Matured into an Actionable Non-tariff Barrier to Trade’ (2013) 28 American University International Law Review 489.

Bibliography  195 Kommers, DP and Miller, RA, The Constitutional Jurisprudence of the Federal Republic of Germany, 3rd edn (Durham, Duke University Press, 2012). Kymlicka, W, Liberalism, Community and Culture (Oxford, Oxford University Press, 1989). —— Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford, Clarendon, 1995). —— ‘Multiculturalism and Minority Rights: West and East’ (2002) 3(4) Journal on Ethnopolitics and Minority Issues in Europe 1. —— ‘The Rise and Fall of Multiculturalism? New Debates on Inclusion and Accommodation in Diverse Societies’ (2010) 61(1) International Social Science Journal 97. —— ‘The Essentialist Critique of Multiculturalism: Theories, Policies, Ethos’ (2014) EUI Working Paper RSCAS 2014/59. Laffineur, J-L, ‘First ECJ Ruling on REACH: Choosing Registration over Exemption’ (2010) 22 Journal of Environmental Law 135. Latour, B, Science in Action: How to Follow Scientists and Engineers through Society (Cambridge, MA, Harvard University Press, 1987). Lemley, BK, ‘Effectuating Censorship: Civic Republicanism and the Secondary Effects Doctrine’ (2002) 35 John Marshall Law Review 189. Levy, R and Orr, G, The Law of Deliberative Democracy (Abingdon, Routledge, 2017). Lichtenberg, J, ‘Foundations and Limits of Freedom of the Press’ (1987) 16 Philosophy & Public Affairs 329. —— ‘Foundations and Limits of Freedom of the Press’ in J Lichtenberg (ed), Democracy and the Mass Media: A Collection of Essays (Cambridge, Cambridge University Press, 1990) 102–35. Mac Síthigh, D, ‘From Freedom of Speech to the Right to Communicate’ in ME Price, SG Verhulst and L Morgan (eds), Routledge Handbook of Media Law (London, Routledge, 2013) 175–91. MacCallum Jr, GC, ‘Negative and Positive Freedom’ (1967) 76 The Philosophical Review 312. Madison, J, The Writings of James Madison, ed G Hunt, vol 9 (New York, Putnam, 1910). Mason, A, ‘The Relationship between Freedom of Expression and Freedom of Information’ in J Beatson and Y Cripps (eds), Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams (Oxford, Oxford University Press, 2000) 225–38. Matheson, D, ‘No Moderator Needed: A Liberty Tradition Right to Broadcast Advertorials’ (2006) 33 Hastings Constitutional Law Quarterly 255. McCormack, M, ‘The Media and Language Maintenance’ in M McCormack and N Hourigan (eds), Minority Language Media: Concepts, Critiques and Case Studies (Bristol, Multilingual Matters, 2007) 52–68. McGlynn, C, ‘Rape, Defendant Anonymity and Human Rights: Adopting a Wider Perspective’ [2011] Criminal Law Review 199. McGonagle, T, Minority Rights, Freedom of Expression and of the Media: Dynamics and Dilemmas (Cambridge, Intersentia 2011). —— ‘Positive Obligations Concerning Freedom of Expression: Mere Potential or Real Power?’ in O Andreotti (Council of Europe Task Force for Freedom of Expression and Media) (ed), Journalism at Risk: Threats, Challenges and Perspectives (Strasbourg, Council of Europe Publishing, 2015). Meier, HE, ‘“For a Few Cents Less”? Nested Games in German Broadcasting Regulation’ (2006) 15 German Politics 166. —— ‘Independent Regulatory Bodies and Impossible Jobs: The Failure of German Public Service Broadcasting Regulation’ (2008) 86 Public Administration 133. Meiklejohn, A, Free Speech and its Relation to Self-Government (New York, Harper, 1948). —— Political Freedom: The Constitutional Powers of the People (New York, Harper, 1960). —— ‘The First Amendment is an Absolute’ [1961] Supreme Court Review 245. Michaels, D, ‘Foreword: Sarbanes-Oxley for Science’ (2006) 69(3) Law and Contemporary Problems 1. Millar, G and Scott, A, Newsgathering: Law, Regulation and the Public Interest (Oxford, Oxford University Press, 2016). Mills, T, The BBC: Myth of a Public Service (London, Verso, 2016). Moon, R, The Constitutional Protection of Freedom of Expression (Toronto, University of Toronto Press, 2000). Moran, C, Classified: Secrecy and the State in Modern Britain (Cambridge, Cambridge University Press, 2013). Morange, J, La liberté d’expression (Brussels, Bruylant, 2009). Moring, T, ‘Functional Completeness in Minority Language Media’ in M McCormack and N Hourigan (eds), Minority Language Media: Concepts, Critiques and Case Studies (Bristol, Multilingual Matters, 2007) 17–33.

196  Bibliography Moring, T and Dunbar, R, The European Charter for Regional or Minority Languages and the Media, vol 6 (Strasbourg, CoE Publishing, 2008). Moring, T and Godenhjelm, S, ‘Broadcasting for Minorities in Big and Small Countries’ in GF Lowe and CS Nissen (eds), Small among Giants: Television Broadcasting in Smaller Countries (Gothenburg, Nordicom, 2011) 177–200. Müller, S and Gusy, C, Case Study Report: Does Media Policy Promote Media Freedom and Independence? The Case of Germany (2011) 21, www.mediadem.eliamep.gr/findings. Nagle, E, ‘Keeping its own Counsel? The Irish Press Council, Self-Regulation and Media Freedom’ (2009) 20 Entertainment Law Review 93. Napoli, PM, Social Media and the Public Interest: Media Regulation in the Disinformation Age (New York, Columbia University Press 2019). O’Neill, O, A Question of Trust, BBC Reith Lectures 2002 (Cambridge, Cambridge University Press, 2002). O’Rourke, D and Lee, E, ‘Mandatory Planning for Environmental Innovation: Evaluating Regulatory Mechanisms for Toxics Use Reduction’ (2004) 47 Journal of Environmental Planning and Management 181. Ortakovski, V, Minorities in the Balkans (Ardsley, NY, Transnational Publishers, 2000). Örtenhed, K and Wennberg, B (eds), Press Freedom 250 Years: Freedom of the Press and Public Access to Official Documents in Sweden and Finland – A Living Heritage from 1766 (Stockholm, Sveriges Riksdag, 2016). Ostrom, E, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge, Cambridge University Press, 1991). Palmer, S, ‘Human Rights: Implications for Labour Law’ (2000) 59 Cambridge Law Journal 168. Parkes, R and Mullis, A (eds), Gatley on Libel and Slander, 12th edn (London, Sweet & Maxwell, 2013). Pasquale, F, Black Box Society: The Secret Algorithms that Control Money and Information (Cambridge, MA, Harvard University Press, 2015). Patten, A, Equal Recognition: The Moral Foundations of Minority Rights (Princeton, Princeton University Press, 2014). —— ‘Moral Foundations of Minority Rights: A Reply to Attas, Bardon, and Gans’ (2015) 12 Jerusalem Review of Legal Studies 91. Pavlenko, A, ‘Language Rights versus Speakers’ Rights: On the Applicability of Western Language Rights Approaches in Eastern European Contexts’ (2011) 10 Language Policy 37. Peters, B, Public Deliberation and Public Culture: The Writings of Bernhard Peters, 1993–2005, ed H Wessler, trans K Tribe (Basingstoke, Palgrave Macmillan, 2008). Pettit, P, On the People’s Terms: A Republican Theory and Model of Democracy (Cambridge, Cambridge University Press, 2012). Picard, RG, ‘Broadcast Economics, Challenges of Scale, and Country Size’ in GF Lowe and CS Nissen (eds), Small among Giants: Television Broadcasting in Smaller Countries (Gothenburg, Nordicom, 2011) 43–56. Porter, V and Hasselbach, S, Pluralism, Politics and the Marketplace: The Regulation of German Broadcasting (London, Routledge, 1991). Post, R, ‘Meiklejohn’s Mistake: Individual Autonomy and the Reform of Public Discourse’ (1993) 64 Colorado Law Review 1109. —— ‘The Constitutional Status of Commercial Speech’ (2000) 48 UCLA Law Review 1. Post, RC, ‘The Constitutional Concept of Public Discourse: Outrageous Opinion, Democratic Deliberation, and Hustler Magazine v Falwell’ (1990) 103 Harvard Law Review 601. —— Constitutional Domains: Democracy, Community, Management (Cambridge, MA, Harvard University Press, 1995). Potschka, C, Towards a Market in Broadcasting: Communications Policy in the UK and Germany (Basingstoke, Palgrave Macmillan, 2012). Rawls, J, Political Liberalism (New York, Columbia University Press, 1993). Redish, MH and Lippman, G, ‘Freedom of Expression and the Civic Republican Revival in Constitutional Theory: The Ominous Implications’ (1991) 79 California Law Review 267. Redish, MH and Mollen, AM, ‘Understanding Post’s and Meiklejohn’s Mistakes: The Central Role of Adversary Democracy in the Theory of Free Expression’ (2009) 103 Northwestern University Law Review 1303. Reichman, JH, Uhlir, PF and Dedeurwaerdere, T, Governing Digitally Integrated Genetic Resources, Data, and Literature: Global Intellectual Property Strategies for a Redesigned Microbial Research Commons (Cambridge, Cambridge University Press, 2016). Ringhand, LA, ‘Concepts of Equality in British Election Financing Reform Proposals’ (2002) 22 Oxford Journal of Legal Studies 253.

Bibliography  197 Rinner, WJ, ‘Maximizing Participation Through Campaign Finance Regulation: A Cap and Trade Mechanism for Political Money’ (2010) 119 Yale Law Journal 1060. Rivers, J, ‘A Theory of Constitutional Rights and the British Constitution’, translator’s introduction to R Alexy, A Theory of Constitutional Rights, trans J Rivers (Oxford, Oxford University Press, 2002) xvii–li. Roberts, A, Blacked Out: Government Secrecy in the Information Age (Cambridge, Cambridge University Press, 2006). Robertson, G and Nicol, A, Media Law, 5th edn (London, Penguin, 2008). Rowbottom, J, ‘Property and Participation: A Right of Access for Expressive Activities’ [2005] European Human Rights Law Review 186. —— Democracy Distorted (Cambridge, Cambridge University Press, 2010). —— ‘Animal Defenders International: Speech, Spending, and a Change of Direction in Strasbourg’ (2013) 5 Journal of Media Law 1. Rumney, P and Fenton, R, ‘Rape, Defendant Anonymity and Evidence-Based Policy Making’ (2013) 76 Modern Law Review 109. Saracini P, et al, Состојбата со медиумите на јазиците на заедниците во Република Македонија [The Situation of Media in the Languages of the Communities of the Republic of Macedonia] (Skopje, The Macedonian Institute for the Media & the Higher School of Journalism and Public Relations, 2012). Schaeur, F, ‘Hohfeld’s First Amendment’ (2008) 76 George Washington Law Review 914. —— ‘Positive Rights, Negative Rights, and the Right to Know’ in DE Pozen and M Schudson (eds), Troubling Transparency: The History and Future of Freedom of Information (New York, Columbia University Press, 2018) 34–51. Schudson, M, The Good Citizen: A History of American Civic Life (New York, Martin Kessler, 1998). —— The Rise of the Right to Know: Politics and the Culture of Transparency, 1945–1975 (Cambridge, MA, The Belknap Press of Harvard University Press, 2015). Schulz, W, ‘How to Control Informal Practices by Formal Law? The German Constitutional Court’s Ruling on the Independence of the Public Service Broadcaster ZDF from the State’ (2015) 7 Journal of Media Law 145. Sedley, S, ‘Information as a Human Right’ in J Beatson and Y Cripps (eds), Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams (Oxford, Oxford University Press, 2000) 239–48. Snell, R, ‘The Kiwi Paradox: A Comparison of Freedom of Information in Australia and New Zealand’ (2000) 28 Federal Law Review 575. Springer, A, Herrmann, H, Sittner, D, Herbst, U and Schulte, A, REACH Compliance: Data Availability of REACH Registration. Part 1: Screening of Chemicals > 1000 tpa (Berlin, Bundesinstitut für Risikobewertung, 2015). Squintani, L, ‘Case Law of the Court of Justice of the European Union and the General Court: Reported Period 15.3.2015–31.8.2015’ (2015) 12 Journal for European Environmental & Planning Law 379. —— ‘Case Law of the Court of Justice of the European Union and the General Court: Reported Period 31.8.2015–31.12.2015’ (2016) 13 Journal for European Environmental & Planning Law 99. Stein, L, Speech Rights in America: The First Amendment, Democracy, and the Media (Urbana, University of Illinois Press, 2006). Steyn, K and Slarks, H, ‘Positive Obligations to Provide Access to Information under the European Convention on Human Rights’ (2012) 17 Judicial Review 308. Stubbs, R, ‘Freedom of Information and Democracy in Australia and Beyond’ (2008) 43 Australian Journal of Political Science 667. Sunstein, CR, ‘Free Speech Now’ (1992) 59 University of Chicago Law Review 255. —— Democracy and the Problem of Free Speech (New York, Free Press, 1993). Swift, A, Political Philosophy, 2nd edn (Cambridge, Polity, 2006); 3rd edn (Cambridge, Polity, 2013). Syvertsen, T, Enli, G, Mjøs, OJ and Moe, H, The Media Welfare State: Nordic Media in the Digital Era (Ann Arbor, University of Michigan Press, 2014). Taylor, C, Sources of the Self: The Making of the Modern Identity (Cambridge, MA, Harvard University Press, 1989). Taylor, G, ‘Diversity Within Government Broadcasters’ in G Gounalakis and G Taylor (eds), Media Diversity Law: Australia and Germany Compared (Frankfurt, Peter Lang, 2016) 221–45. Tiilikka, P, ‘Access to Information as a Human Right in the Case Law of the European Court of Human Rights’ (2013) 5 Journal of Media Law 79. Townend, J, ‘Closed Data: Defamation and Privacy Disputes in England and Wales’ (2013) 5 Journal of Media Law 31. —— ‘Leveson Online: A Publicly Reported Inquiry’ (2013) 10 Ethical Space 14.

198  Bibliography Turpin, C, British Government and the Constitution: Text, Cases and Materials, 5th edn (London, Butterworths, 2002). Tushnet, M, Advanced Introduction to Freedom of Expression (Cheltenham, Edward Elgar, 2018). Valcke, P, Sükösd, M and Picard, RG (eds), Media Pluralism and Diversity: Concepts, Risks and Global Trends (Basingstoke, Palgrave Macmillan, 2015). Vassall-Adams, G, Reporting Restrictions in the Criminal Courts (London, Judicial College, 2015). Vincent, D, The Culture of Secrecy: Britain 1932–1998 (Oxford, Oxford University Press, 1997). Wates, J, ‘The Aarhus Convention: A Driving Force for Environmental Democracy’ (2005) 2 Journal for European Environmental & Planning Law 2. Webster, JG, The Marketplace of Attention (Boston, MA, MIT Press, 2014). Weinstein, J, ‘Participatory Democracy as the Central Value of American Free Speech Doctrine’ (2011) 97 Virginia Law Review 491. West, F and Parsons, A, Briefing Note: Gagging Clauses in the NHS (London, Public Concern at Work, 2011). Whistleblowing Commission, Report on the Effectiveness of Existing Arrangements for Workplace Whistleblowing in the UK (London, Public Concern at Work, 2013). Williams, R, The Existing Alternatives in Communications (London, Fabian Society, 1962). Williams, SH, Truth, Autonomy, and Speech: Feminist Theory and the First Amendment (New York, NYU Press, 2004). Witteman, C, ‘Constitutionalizing Communications: The German Constitutional Court’s Jurisprudence of Communications Freedom’ (2010) 33 Hastings International and Comparative Law Review 95. —— ‘Information Freedom, a Constitutional Value for the 21st Century’ (2013) 36 Hastings International and Comparative Law Review 145. Worthy, B, The Politics of Freedom of Information: How and Why Governments Pass Laws That Threaten Their Power (Manchester, Manchester University Press, 2017). Wragg, P, ‘Free Speech Rights at Work: Resolving the Differences Between Practice and Liberal Principle’ (2015) 44 Industrial Law Journal 1. Wu, T, ‘Is the First Amendment Obsolete’ (2017) Knight First Amendment Institute, Emerging Threat Series, https://knightcolumbia.org/content/tim-wu-first-amendment-obsolete. Youn, M, ‘First Amendment Fault Lines and the Citizens United Decision’ (2011) 5 Harvard Law and Policy Review 135. Young, IM, Justice and the Politics of Difference (Princeton, Princeton University Press, 1990). Zuckerman, A, ‘Common Law Repelling Super Injunctions, Limiting Anonymity and Banning Trial by Stealth’ [2011] Civil Justice Quarterly 223.

INDEX Aarhus Convention (1998), 177, 182 access to courts see public access to courts access to information: court proceedings reporting, 137–40 right to receive, 140–3 distribution of information, 122 ECHR, 138 ECtHR case law, 44, 47–8, 49, 50, 61–3 environmental information, 173–89 FOI see freedom of information legislation open government, 3, 7, 44, 51–61, 62, 149, 160–1, 167 positive free speech and, 43–64, 122, 175–6 property rights and, 173, 175, 176, 184, 187 right to know, 43–5 theory, 45–7 United Kingdom, 44–64 common law principle, 50, 51–61 FOI Act, 43–4, 51, 59–60 Good Law Project, 58–60 HMRC, 55–8 Ingenious Media, 57–8 Kennedy, 32, 44–55, 62–3, 121, 123–4, 167 post-Kennedy, 60–1 Privacy International, 55–7 Alexy, Robert, 92, 94 American Chemistry Council, 185 anonymity see party anonymity in legal proceedings Ashford, NA, 176 Australia: breach of confidence, 161 Austria, 31, 32, 48 BAILII, 143 Baker, C Edwin, 90, 101 Barendt, Eric, 100, 101, 103 Barron, Jerome A, 12 BBC, 38, 39, 127, 132–3 Berlin, Isaiah, 13 Bhopal disaster (1984), 177 bisphenol A, 181–2 Blair, Tony, 45 Blasi, Vincent, 46 Bohlander, Michael, 127 Bosnia Herzegovina, 106 Brexit, 59

broadcasting see mass media Burke, Abbey, 7 business secrets, 155, 159, 173 Caldart, Charles C, 176 Cameron, David, 147 Canada: minorities, 115 Carolan, Eoin, 4, 9 censorship: free expression and, 83 private censorship, 28–30 social pressures, 5, 30 chemical industry: environmental information, 174–89 Cheston, Paul, 145 child abuse, 132 Cobain, Ian, 141 Cohen, Joshua, 26–7 commercial speech, 15 communication rights: free expression and, 66–7 Irish Constitution, 66–7 communicative equality: abuse of state power, 39–40 assumption, 11 digital media, 38–9 election campaigns, 28, 34, 40 impossibility, 18, 27 judicial promotion, 30–3 mass media, 36–8 minority speech rights and, 99–101 overview, 7–8 positive free speech and, 25–41 private censorship, 28–30 promoting equal opportunities, 33–9 public spaces, 34–6 subsidies, 28, 34, 35, 36–8, 39 substantive political equality, 26–8 compromise agreements, 156–7, 168–70 confidentiality: clauses see confidentiality clauses environmental information and, 183–4, 186 confidentiality clauses: damages and, 154 effect, 153–4 FOI legislation and, 157, 162

200  Index form, 154–5 implied terms, 159 legality, 157–65 contract and equity, 158–62 usage and practice, 159–60 limiting principles, 158–9 managing reputations, 155 objectives, 155 prevalence, 154–7 public contracts, 153–71 judicial review, 166–8 open government and, 160–1, 167 oversight regime, 168–70 practice, 155–7 regulating, 165–70 public domain information and, 158, 160 public interest and, 158–9, 160–1, 164, 165, 170 scandals, 153 scope, 155, 158 successful challenges: repayments, 164–5 terms, 156 whistleblowing and, 157, 163–4, 169, 170 contempt of court, 119, 140, 142, 147, 149 copyright, 64, 160, 176, 187 Council of Europe see also European Convention on Human Rights Former Yugoslavia and, 107 language rights, 98, 104 on media pluralism, 19–20 Tromso Convention (2009), 177 counterpublics, 100–1, 113, 114, 115 Court of Protection, 137 courts: access to information and, 44–64 open justice see open justice party anonymity see party anonymity in legal proceedings promotion of positive free speech, 30–3 public access to see public access to courts role, 6–7, 84 Courtserve, 142–3 Croatia: EU membership, 107 data protection, 148, 150, 184 Davis, David, 59 defamation, 74, 143, 160, 167 democracy: access to information and, 43 communication and, 11–12, 20 communicative equality and, 26–7 fake news and, 22 free speech and, 25 Ireland, 68–70 minority speech rights and, 97–101 monitory democracy, 8

open government and, 160–1, 167 overview, 8 Denning, Alfred, 138–9 digital media: big data, 176 communicative equality, 38–9 court proceedings, 139–40, 141, 142, 143 Leveson Inquiry, 146–9 discrimination: denying access to speech, 18, 31 Dworkin, Ronald, 102 election campaigns: communicative equality, 28, 34 spending caps, 28, 34 state funding, 34, 40 Emerson, TI, 12 employment: compromise agreements, 156–7, 168–70 implied duty of confidence, 159 positive free speech and, 30–1 protected disclosures, 163–4 environmental information: access to, 173–89 Aarhus Convention (1998), 177, 182 basis of speech, 175–6 European Union, 178–85 institutional change, 186–8 limits, 173, 179–85 objectives, 173–4 United States, 177–8 asymmetry, 174 chemical industry, 174–89 EU limits to access, 173, 179–85 business confidentiality, 183–4, 186 ClientEarth, 185 communication methods, 181–2 free trade issues, 184–5 intellectual property, 184, 187 nature of information, 185 power balance, 182–3 procedural, 179–81 institutional change capacity development, 186–8 open data principles, 187–8 regulatory agencies, 186–7 equality: communicative see communicative equality political equality and expression, 26–8 European Chemical Industry Council, 185 European Chemicals Agency (ECHA), 181–5 European Convention on Human Rights: fair hearing, 126, 136 free speech (Article 10), 30–3 access to court proceedings, 138–9 access to information, 44, 47–50, 59–63

Index  201 confidentiality clauses and, 162 positive free speech, 7, 30–3, 121–3 two-way street, 98 party anonymity in legal proceedings and, 118–20 privacy rights, 119, 128–9 European Court of Human Rights: access to information, 44, 47–8, 49, 50, 61–3 court proceedings, 140 exhaustion of domestic remedies, 60 positive free speech, 7, 30–3, 121–3, 124 privacy rights, 129 European Union: access to environmental information, 177 Aarhus Convention (1998), 177 limits, 179–85 REACH, 177, 178–85 access to public sector information, 177 Croatian and Slovenian membership, 107 on media pluralism, 20 TTIP, 185 Facebook, 38, 139 fair hearing: ECHR, 126, 136 open justice and, 136 public hearing, 126 fake news, 22, 38 Fiss, Owen M, 17, 75 France: environmental information, 188 Fraser, Nancy, 100–1, 104, 113 free association: Ireland, 68 free expression see also individual countries; individual treaties; specific aspects international treaties, 98–9 negative v positive, 1–5, 12–14 party anonymity in legal proceedings and, 119 positive see positive free speech strong conception: basis, 14–16 traditional concepts, 12–14 freedom of information legislation: see also access to information case law, 59–60 confidentiality clauses and, 157, 162 copyright and, 187 court records exception, 143 United Kingdom, 43–4, 51, 59 gagging clauses see confidentiality clauses Gelber, Katharine, 175–6 Germany: broadcasting freedom, 83–96 adequate non-political funding, 89–91 broadcasting councils, 91–3

case law, 87–96 pluralism, 91–3 precautionary approach, 93–4, 96 commercial broadcasting, 89 Constitution free speech, 86–7, 94 human dignity, 93 Constitutional Court, 86 broadcasting case law, 87–95 court role, 6–7, 84, 95–6 democratic corporatism, 86 free expression Constitution, 86–7, 94 positive dimension, 88–9, 95–6 state and, 94–5 public broadcasting adequate non-political funding, 89–91, 94 dual system, 89, 94, 95 Gibbons, Thomas, 3, 5, 7, 8, 120 Google, 38 Grant, Hugh, 148 group solidarity, 101 Habermas, Jürgen, 100 Hallin, Daniel, 86 Hasselbach, Suzanne, 94 hate speech, 30, 74 Hoffmann-Riem, Wolfgang, 92 Hungary: access to information, 61 ‘ideal speech situation’, 100 identity: minority speech rights and, 99, 102 Incedal case, 143, 144–6, 151 India: Bhopal disaster (1984), 177 individualism, 15, 68, 101 information see access to information; environmental information inquiries, 146–9 intellectual property, 173, 184, 187 International Covenant on Civil and Political Rights (ICCPR): free expression, 98–9 Investigatory Powers Tribunal, 137 Ireland: binary model of free expression citizenship, 70–3 civic discourse, 71–81 consequentialism, 69–70, 72, 75, 80 content-based restrictions, 73–4 democratic value, 69–70, 72 implications, 67–79 limitations of civic value, 75–6 limits, 74–5 media role, 75–9 public v private discourse, 67–9, 80 Republican model, 70–1

202  Index Constitution democracy, 72 free association, 68 free expression, 66–7 Republicanism, 65, 70–1 right to communicate, 66–7 court role, 7 free expression binary model, 67–79 civic discourse, 71–81 Constitution, 66–7 personal v private spheres, 67–9, 80 positive free speech, 79–81 Post Office monopoly, 66 press regulation, 78 trade unions, 68 Italy, 47–8, 105 Jukes, Peter, 139 Kenyon, Andrew, 11, 71, 120 Klimkiewicz, Beata, 105 Kosovo, 106 Kymlicka, Will, 102–3, 104, 115 language rights see minority speech rights Leveson Inquiry, 127, 146–9 Lichtenberg, Judith, 12, 120, 175 MacCallum, Gerald, 13 McGonagle, Tarlach, 99 McNamara, Lawrence, 146 Madison, James, 43 Mancini, Paolo, 86 mass media: advertising, 90 communicative equality, 36–8 Germany, 87–96 impartiality and, 36–7 international publications, 38 mass media age, 88, 120 plurality laws, 37–8 political inequality, 27, 29 right to reply, 33 subsidies, 39 media see digital media; mass media media pluralism: communicative equality and, 37–8 implementing active pluralism, 21 positive free speech and, 19–23 medical negligence disputes, 156, 157 Meiklejohn, Alexander, 12, 18, 69, 97–8, 99–100, 101 #MeToo movement, 171 Michaels, David, 179

Mid-Staffordshire Health Trust, 153 Mill, John Stuart, 99, 100 Milton, John, 99 minority speech rights: alternative discursive spaces, 99–101 communicative equality, 99–101 democracy and, 99–100 European minority languages, 105 group solidarity, 101 individual self-realisation, 101–3 international treaties, 104 justification, 98–106 meaning, 97 media provision, 104–6 Slovenia and North Macedonia, 106–14 national minorities, 107–10 public platforms, 110–13 state obligations, 103–6 Mollen, Abby M, 69 Moring, Tom, 104–5 multiple actors: overview, 6–7 mySociety, 148 naming and shaming, 136–7 National Audit Office, 155–6 New Zealand, 78, 131 News International, 153 News Media Association, 143 Nicol, Andrew, 136 non-disclosure agreements see confidentiality clauses North Macedonia: immigration, 109–10 language rights, 106–14 discursive spaces, 114 public platforms, 110–13 national minorities, 107–10 Northern Ireland Audit Office, 156, 157 Nussbaum, Martha, 175 online media see digital media open government, 3, 7, 44, 51–61, 149, 160–1, 167 open justice: common law principle, 140 deterrence rationale, 136–7 education rationale, 136 open data from courts, 149–51 party anonymity and, 118–19, 125–31 requirements, 46–7 scholarship, 136–7 UK derogations, 137–8 Organisation for Security and Cooperation in Europe (OSCE): Former Yugoslavia and, 107 language rights, 98, 105–6, 112–13, 115

Index  203 parliaments: role, 6 party anonymity in legal proceedings: ECHR and, 118–20 New Zealand, 131 positive free expression and, 131–3 United Kingdom child abuse, 132 control orders, 130 development, 124–9 discretion, 118, 119, 125–9, 137–8 family proceedings, 129, 137, 142 sexual offences, 117–18, 127, 137 suppressing information, 129–30 youth courts, 117, 137 Patten, Alan, 103 phone hacking, 139, 153 platforms: access criteria, 17–18 access to court proceedings, 138 implementing active pluralism, 21 media pluralism, 19–23 minority speech rights, 104–5 partisan restrictions on content, 18–19 practicalities of access, 16–19 provision, 11–23, 123 public provision, 28 public v private, 16–17 Porter, Vincent, 94 positive free speech: access to information, 43–64, 122, 138, 175–6 basis, 14–16 communicative equality, 25–41 court role see courts ECtHR case law, 7, 30–3, 121–3, 124 examples, 3–4 Germany, 83–96 media pluralism and, 19–23 negative v positive dimensions, 1–5, 12–14 party anonymity in legal proceedings, 131–3 platforms see platforms pluralism, 19–23, 123 protection of journalists and publishers, 122–3 public access to courts, 135–51 scholarship, 120–1 terminology, 2–3 Post, Robert C, 71, 75 privacy rights: ECHR, 119, 128–9 environmental information and, 184 party anonymity in legal proceedings and, 119, 128–9, 130, 131–2 privatisations: public spaces, 35–6 property rights, 173, 175, 176, 184, 187

public access to courts: court lists, 142–3, 150 digital media, 139–40, 141, 142, 143 Incedal case, 143, 144–6, 151 inquiries, 147 Leveson Inquiry, 146–9 methods, 138, 151 open data from courts, 149–51 open justice principle, 136–8 outcomes of cases, 143 positive free speech and, 135–51 reporting rights, 138–40 restricted reporting, 137–8 orders, 141–2 rethinking, 151 right to receive court information, 140–3 sentencing, 143 terrorist trials, 144–6 public contracts: confidentiality clauses see confidentiality clauses open government, 160–1, 167 public sphere: Habermas, 100 Redish, Martin H, 69 Reed, Lucy, 142 Richard, Cliff, 127, 132–3 Riggins, Stephen, 105 right to life: party anonymity in legal proceedings and, 119, 130 Robertson, Geoffrey, 136 Robinson, Tommy, 140 Rodney, Azelle, 149 Rowbottom, Jacob, 5 Russia: electoral communications, 32 Schauer, Frederick, 12 Schudson, Michael, 43 Scott, Andrew, 7 search media, 7 Sharp, Robert, 148 Slarks, Hannah, 120, 124 Slovenia: EU membership, 107 immigration, 109–10 language rights, 106–14 public platforms, 110–13 national minorities, 107–10 social media: civic discourse, 79 communicative equality and, 18, 38–9 court reports, 139–40 fake news, 22, 38 Leveson inquiry and, 147 online shaming, 30

204  Index public watchdogs, 62 speakers and audiences, 7 social pressures, 5, 30 Society of Editors, 143 solidarity, 101 Soviet Union: language rights, 97 Spain: Catalan language, 105 speakers and audiences: overview, 7 Special Immigration Tribunal, 137 Steyn, Karen, 120, 124 structural conditions, 7, 8–9 subsidies, 2, 28, 35, 36–8, 39, 105, 112–13, 115, 123 Sunstein, Cass, 75 super Injunctions, 126, 146 terrorism: reporting court proceedings, 144–6 Thomas, Lord, 146 Tickle, Louise, 129 torture/inhuman treatment, 119 Toulson, Lord, 64 trade secrets, 155, 159, 173 Trans-Atlantic Trade and Investment Partnership (TTIP), 185 Tromso Convention (2009), 177 Twitter, 38, 79, 147–8 Union Carbide, 177 United Kingdom: access to information, 44–64 common law principle, 50, 51–61 FOI Act, 43–4, 51, 59, 143 Good Law Project, 58–60 HMRC, 55–8 Ingenious Media, 57–8 Kennedy, 32, 44–55, 62–3, 121, 123–4, 167 post-Kennedy, 60–1 Privacy International, 55–7 charities, 51–2 control orders, 130 court role, 7, 32 distribution of information, 122

election campaign spending, 28, 34 Human Rights Act, 54 inquiries, 146–9 lobbying, 39–30 mass media, 36–7 party anonymity in legal proceedings, 117–18 child abuse, 132 development, 124–9 discretion, 118, 119, 125–9, 137–8 family proceedings, 129, 137, 142 sexual offences, 117–18, 137 youth courts, 117, 137 positive free speech, 121, 123–4 public access to courts, 135–51 public protests, 35 super-Injunctions, 126, 146 terrorist trials, 144–6 Welsh language, 105 United Nations: ethnic minorities and, 99 United States: access to information, 43 elections, 33, 34 environmental information, 177–8 free speech case law, 93 court role, 95 legal development, 5 market driven, 90 negative formulation, 12, 65, 80 realism, 9 TTIP, 185 Universal Declaration of Human Rights (UDHR): free expression, 98 Weinstein, Harvey, 153 whistleblowing, 157, 163–4, 169, 170, 186 Witteman, Christopher, 90, 91 Wrightson, Angela, 137, 139 Young, Iris Marion, 101 Yugoslavia: minorities, 97, 106–10