Democratic Innovation: Deliberation, Representation and Association

Democratic Innovation

There can be few political questions more vital than the future of democracy. For much of the postwar period, the idea of democracy as rule by the majority through elections has been dominant. Now, however, in the face of ecological, global and technological challenges, this is beginning to unravel. Democratic Innovation offers a tantalizing glimpse of our political future by interrogating the latest ideas aimed at renewing popular power. Featuring new writing by leading European, American and Australian democratic theorists, this book explores the importance of public deliberation in democracies, how effective representation for all might be achieved, and how associations outside government may be at the heart of democracy’s future. The contributors examine issues such as how our political systems can be revitalized; how ordinary people can be empowered when issues are so complex; whether the interests of marginalized groups, or of the natural world, can be effectively represented; and the role which voluntary associations can play in democratic governance. Through clear and rigorous debate, they consider the new institutions and attitudes which will be needed if democracy is to rise to the many challenges confronting it. This original volume offers searching and accessible critiques of the latest thinking on deliberative democracy. It constitutes thought-provoking and challenging reading for students and scholars of Politics. Michael Saward is Reader in Government and Politics at The Open University. His previous publications include The Terms of Democracy.

Routledge/ECPR Studies in European Political Science Formerly edited by Hans Keman, Vrije University, The Netherlands; now edited by Jan W. van Deth, University of Mannheim, Germany on behalf of the European Consortium for Political Research.

The Routledge/ECPR Studies in European Political Science series is published in association with the European Consortium for Political Research – the leading organization concerned with the growth and development of political science in Europe. The series presents high-quality edited volumes on topics at the leading edge of current interest in political science and related fields, with contributions from European scholars and others who have presented work at ECPR workshops or research groups. 1 Regionalist Parties in Western Europe Edited by Lieven de Winter and Huri Türsan 2 Comparing Party System Change Edited by Jan-Erik Lane and Paul Pennings 3 Political Theory and European Union Edited by Albert Weale and Michael Nentwich 4 Politics of Sexuality Edited by Terrell Carver and Véronique Mottier 5 Autonomous Policy Making by International Organizations Edited by Bob Reinalda and Bertjan Verbeek 6 Social Capital and European Democracy Edited by Jan van Deth, Marco Maraffi, Ken Newton and Paul Whiteley 7 Party Elites in Divided Societies Edited by Kurt Richard Luther and Kris Deschouwer 8 Citizenship and Welfare State Reform in Europe Edited by Jet Bussemaker

9 Democratic Governance and New Technology Technologically mediated innovations in political practice in Western Europe Edited by Ivan Horrocks, Jens Hoff and Pieter Tops 10 Democracy without Borders Transnationalisation and conditionality in new democracies Edited by Jean Grugel 11 Cultural Theory as Political Science Edited by Michael Thompson, Gunnar Grendstad and Per Selle 12 The Transformation of Governance in the European Union Edited by Beate Kohler-Koch and Rainer Eising 13 Parliamentary Party Groups in European Democracies Political parties behind closed doors Edited by Knut Heidar and Ruud Koole 14 Survival of the European Welfare State Edited by Stein Kuhnle 15 Private Organisations in Global Politics Edited by Karsten Ronit and Volker Schneider 16 Federalism and Political Performance Edited by Ute Wachendorfer-Schmidt 17 Democratic Innovation Deliberation, representation and association Edited by Michael Saward Also available from Routledge in association with the ECPR: Sex Equality Policy in Western Europe, Edited by Frances Gardiner; Democracy and Green Political Thought, Edited by Brian Doherty and Marius de Geus; The New Politics of Unemployment, Edited by Hugh Compston; Citizenship, Democracy and Justice in the New Europe, Edited by Percy B. Lehning and Albert Weale; Private Groups and Public Life, Edited by Jan W. van Deth; The Political Context of Collective Action, Edited by Ricca Edmondson; Theories of Secession, Edited by Percy Lehning; Regionalism Across the North/South Divide, Edited by Jean Grugel and Wil Hout.

Democratic Innovation Deliberation, representation and association

Edited by Michael Saward

London and New York

First published 2000 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 Routledge is an imprint of the Taylor & Francis Group This ediation published in the Taylor & Francis e-Library, 2003. © 2000 Selection and editorial material Michael Saward; individual chapters the contributors All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Democratic innovation: deliberation, representation, and association / edited by Michael Saward. p. cm. Includes bibliographical references and index. 1. Democracy. 2. Representative government and representation. 3. Associations, institutions, etc. I. Saward, Michael, 1960– JC423.D44175 2000 00-042215 321.8 – dc21 ISBN 0-203-16548-9 Master e-book ISBN

ISBN 0-203-25994-7 (Adobe eReader Format) ISBN 0-415-23442-5 (Print Edition)

Contents

Notes on contributors Series editor’s preface Editor’s acknowledgements

ix xi xiv

Introduction Democratic innovation

3

MICHAEL SAWARD

PART I

Deliberative democracy: advocacy and critique 1 The quest for deliberative democracy

17

JAMES S. FISHKIN AND ROBERT C. LUSKIN

2 Toward deliberative institutions

29

GRAHAM SMITH

3 Deliberation as public use of reason – or, what public? whose reason?

40

T I I NA R ÄT T I L Ä

4 The European Union’s democratic deficit: a deliberative perspective

53

ERIK ODDVAR ERIKSEN

5 Less than meets the eye: democratic legitimacy and deliberative theory

66

MICHAEL SAWARD

6 Discursive democracy vs. liberal constitutionalism JOHN S. DRYZEK

78

viii

Contents

PART II

Representation and deliberation 7 Group representation, deliberation and the displacement of dichotomies

93

JUDITH SQUIRES

8 From theory to practice and back again: gender quota and the politics of presence in Belgium

106

PETRA MEIER

9 Deliberative democracy, ecological representation and risk: towards a democracy of the affected

117

ROBYN ECKERSLEY

10 Ecological constitutionalism and the limits of deliberation and representation

133

MIKE MILLS AND FRASER KING

11 Governance, self-representation and democratic imagination

146

HENRIK PAUL BANG AND TORBEN BECH DYRBERG

PART III

Associations and democracy 12 Active citizenship and associative democracy

161

PIOTR PERCZYNSKI

13 Associative democracy – fashionable slogan or constructive innovation?

172

SIGRID ROßTEUTSCHER

14 Social capital, associations and civic republicanism

184

FRANCISCO HERREROS

15 Deliberative democracy versus direct democracy – plus political parties!

195

IAN BUDGE

Conclusion Variation, innovation and democratic renewal

213

MICHAEL SAWARD

Bibliography Index

221 237

Contributors

Henrik Paul Bang is Director of the Centre for Public Organization and Management of the Copenhagen Business School. His main area of research and publication is democratic theory. Ian Budge has been Professor of Government at the University of Essex since 1976, during which time he has authored or co-authored some twenty books and fifty articles and monographs. He was founder of the annual Summer School on Quantitative Social Analysis and Director of the ECPR 1977–83. His interest in democracy was sparked when he studied under Lasswell and Dahl at Yale, though his interests were more empirical than normative up to the publication of The New Challenge of Direct Democracy in 1996. John S. Dryzek is Professor of Political Science at the University of Melbourne. His most recent book is Deliberative Democracy and Beyond: Liberals, Critics, Contestations. Torben Bech Dyrberg is Assistant Professor in the Centre for Public Organization and Management of the Copenhagen Business School. His research and publication interests are in political theory, especially the concepts of power, justice and democracy. Robyn Eckersley is Senior Lecturer in the School of Political and Social Inquiry at Monash University, Australia. She has published widely in the field of green political theory and is working on a book on green democracy and the state. Erik Oddvar Eriksen is Professor at ARENA, the University of Oslo, and Adjunct Professor at the LOS Centre, University of Bergen, Norway. Recent publications include Kommunikativ Ledelse (Communicative Leadership), and (with Jarle Weigård) Kommunikativ Handling og Deliberativt Demokrati (Communicative Action and Deliberative Democracy). He has edited, among other works, Democracy in the European Union (with John Erik Fossum). His research interest is in social and political theory, with particular emphasis on the European Union. James S. Fishkin holds the Pattinson–Bannister Chair at the University of Texas at Austin, where he is Professor of Government, Law and Philosophy and Director of the Center for Deliberative Polling. He is the author of The Voice of the People: Public Opinion and Democracy, Democracy and Deliberation and The Dialogue of Justice, among other books.

x

Contributors

Francisco Herreros is a researcher in the Centre for Advanced Studies in the Social Sciences at the Juan March Institute in Madrid. He is currently preparing his PhD thesis on the formation of social capital. Fraser King is Lecturer in Politics at Nottingham Trent University. His research interests include democratic theory, with special reference to green political thought, and constitutionalism and democracy, particularly constitutional change and democratization in British politics. Robert C. Luskin is Associate Professor of Government and Research Director of the Center for Deliberative Polling at the University of Texas at Austin. He has written about public opinion, political psychology, voting behaviour, and statistical methods in the American Political Science Review, the American Journal of Political Science, and other scholarly journals. He is a member of the Editorial Board of the American Political Science Review. Petra Meier is a research assistant at the Department of Political Sciences of the Vrije Universiteit Brussels and has worked for the Council of Europe and the European Union Expert Network ‘Women in Decision-Thinking’. She is preparing a PhD thesis on guaranteed representation of social identities (language and gender) in Belgium. Mike Mills is Senior Lecturer in Politics at London Guildhall University. His principal research interests include green political thought, especially the study of green ethics, on which he has published widely. Piotr Perczynski is a post-doctoral researcher at the Department of Political Science, Leiden University and an Assistant Professor at the Institute of Political Studies, Polish Academy of Sciences. He has published on liberalism, political pluralism, functionalism and citizenship, and is currently working on a book on the role of pluralistic citizenship in uniting Europe. Tiina Rättilä is a Junior Research Fellow in Politics at the University of Tampere, Finland. She is currently preparing her doctoral dissertation on the theory and practice of deliberative democracy. She is also engaged in a collaborative research project on styles and aesthetics of political action. Sigrid Roßteutscher is Assistant Professor at the University of Mannheim. She is currently working on an international project on ‘Citizenship, Involvement and Democracy’. Michael Saward is Reader in Government and Politics at The Open University, UK. He is the author of The Terms of Democracy. Graham Smith is Lecturer in Politics at the University of Southampton. Co-author (with James Connelly) of Politics and the Environment: From Theory to Practice, he has published on citizen juries, green political theory and urban politics. Judith Squires is Senior Lecturer in Political Theory at the Politics Department, University of Bristol. She is author of Gender in Political Theory.

Series editor’s preface

Next to the question how to obtain democracy, the most heated controversies in political theory are addressed to the problems and prospects of defending, maturing and expanding existing democratic decision-making processes. This continuous attention to attempts to reform democracies used to focus on problems like the threats of anti-democratic sentiments or the risks of concentrating power, and the conventional solutions usually suggest a reduction of democracy in one way or the other. Limitations on democratic decision making, then, are legitimated by trying to avoid more devastating hazards. Recent debates in political theory resolutely suggest moving in the opposite direction: the only real answer to the difficulties of democracy is to have even more – and perhaps even more complicated modes – of democracy. Democracy cannot be simply defended or maintained; it should be expanded in an innovative way. In the words of the former German Chancellor Willy Brandt, we should be prepared ‘to venture more democracy’ if we want to deal with the problems of democracy. This volume on democratic theory and democratic decision making is not just another addition to the literally thousands of books written on this subject since Pericles defended the exceptional character of decision making in Athens about 2,500 years ago. But it is not the pretensions of an ancient ruler or the problems facing an emerging city state which are the objects of renewed discussions about democracy. The opportunities and complications of democratic decision-making processes in the highly complex and open mass-societies of the late twentieth and early twenty-first century are hardly comparable with the experiences gained in previous eras. Instead of Pericles’ addresses, the works of Alexis de Tocqueville, Joseph Schumpeter, Jürgen Habermas or Benjamin Barber are attacked, praised, unravelled, or utilized. And despite the evident signs of practical disillusionment with aspects of existing of representative democracies – declining voting turnout, severe criticism of parties and politicians – anti-democratic sentiments nowhere play the dominant role in these debates. On the contrary. The idea that the problems of democracy can be dealt with only by expanding democracy obtains ever more support. What leads to deep controversies, however, is the question of what these expansions could and should look like. The unique character of this volume is that all contributors focus on the opportunities for democratic innovations; none of the authors considers expansion to be sufficient.

xii

Series editor’s preface

In his introductory chapter, the editor elucidates the contemporary discussions about democracy and the arrangement of the contributions by focusing on three major topics: deliberation, representation and association. The first part of the volume comprises various discussions about the need for deliberative arrangements in enlightened democracies on the basis of the presumption that democracy is more than just counting heads and that preferences should not only be simply registered but also encouraged and shaped in the course of democratic processes. An instrument already applied is the ‘deliberative poll’, discussed by James S. Fishkin and Robert C. Luskin in their original attempt to apply parts of the Athenian model to societies with millions of citizens. In addition, Graham Smith analyses the chances to include democratic deliberation in institutional settings, while Erik O. Eriksen deals with similar themes with respect to the development of European integration and the European Union. Tiina Rättilä raises a number of questions about the arguments and actors allowed to enter a public debate. The remaining two chapters in this first part take a rather critical stand towards the expected beneficial consequences of deliberation in modern democracies, by focusing on democratic legitimacy (the chapter by Michael Saward) and liberal constitutionalism (the chapter by John Dryzek) respectively. Part two of this volume contains several contributions on the complications of combining deliberative means with the requirements of representation. Several answers to the question ‘who is to do the deliberating?’ are presented. Insights from feminist theory on group representation are discussed by Judith Squires and confronted with the practical complications of implementing gender quota in Belgium described by Petra Meier. A radical view on representation is offered by Robin Eckersley in her suggestion to develop the idea of ‘ecological democracy’ not necessarily restricted to human interests. The cake is cut in a different way by Mike Mills and Fraser King in their endeavour to rescue the positive experiences enjoyed with specific aspects of representative democracy and deliberation by developing the idea of ‘ecological constitutionalism’. The concluding chapter of the second part by Henrik P. Bang and Torben B. Dyrberg consists of a critical and sceptical treatment of the central assumptions about representation and the question who defines the ‘reality of politics’. The set of four essays constituting the third part of the book address the distinctions and overlaps between associative democracy and deliberative democracy. Piotr Perczynski presents a broad overview of the potential contributions of all kind of associations to the development of democratic decision-making processes by distinguishing between self-organizing and state-supported organizations. Drawing on a number of theoretical and especially empirical arguments, Sigrid Roßteutscher is however rather sceptical about the blessings of associational democracy, much of which seem to be ‘fashionable slogans’ instead of thoroughly developed theoretical ideas. Francisco Herreros shows that a distinction between ‘horizontal’ and ‘vertical’ associations is necessary, but that even then much remains to be desired in terms of clarity of the causal interpretations offered in this area. Referring to many points raised in various contributions to this volume, Ian Budge returns to the question about the role and position of political parties in

Series editor’s preface xiii particular as conventional key players in decision-making processes. The weak spots he observes in the request for deliberation is summarized nicely in his provocative depiction of the ‘university seminar model’ of deliberative democracy, leading to a defence of political parties. This volume itself is an excellent example of deliberation and innovation, the need for which is so clearly emphasized in nearly each and every chapter. The authors approach the problems from different perspectives and they weigh the theoretical and empirical evidence available differently. These conflicting assumptions and interpretations are agreeably spelled out by the editor in both his introductory and his concluding chapter. It becomes clear that the decision to present opposing views in distinct contributions in a single volume, has the advantage over rigid attempts to standardize the essays or to include only chapters that follow a similar approach or strategy. For contemporary democratic theory, then, the same conclusion can be stated as for democracy itself: to deal with the apparently growing number of complications and problems, we need more debate and explicit confrontations of different views and approaches. This volume shows in an excellent way what can be gained in democratic theory when the search for innovation and deliberation is based on openness and agreement to disagree. It is to be hoped that the gains for democracy will follow soon. Jan W. van Deth Series editor Mannheim, May 2000

Editor’s acknowledgements

This book originated in a workshop entitled Innovation in Democratic Theory held as part of the 1999 Joint Sessions of the European Consortium for Political Research at the University of Mannheim. That occasion brought together scholars from ten different countries keen to challenge prevailing ideas about democracy; the unusual level and intensity of debate it generated is, we hope, reflected in the chapters that follow. We are grateful to the ECPR for helping us create such a congenial public space for deliberation. Although their work presented in Mannheim is to be published elsewhere, all contributors would like to acknowledge the contributions made to this project by Russell Bentley, Ed Greenberg and Simon Thompson. I would like to thank Deborah Koontz, formerly of Royal Holloway, University of London, and especially June Ayres, of The Open University, for their timely and efficient help in producing the typescript. Michael Saward London July 2000

Introduction

Democratic innovation Michael Saward

The story of democracy is nothing if not a story of innovation. One of the defining features of democracy may well be its restlessness, dynamism and comparative openness to new ideas. This story is now over 2,500 years old (see Dunn 1992) – perhaps much older, if we adopt a less Western-centred view (see Manglapus 19871). Themes in that history are vital to many of the arguments that follow, but the main concern of this book is contemporary, with a focus on key ideas and arguments which right now are reshaping the way we view democracy. Thus the chapters that follow range over the contentious issues of deliberation and its role in democracy, the meaning and scope of representation, and the role that associations can and should play in democratic systems. I shall say more by way of introduction to these pressing themes and the way this book approaches them in a moment; first, some more general comments by way of a rationale for the book as a whole. After many years mired in debates in which ‘participationists’ and ‘elitists’ argued past each other, democratic theory has become a fertile field for innovation. The democratic ideal is being rethought in quite fundamental ways, with new ideas increasingly being linked to new practical institutional designs. Our aim in this book is to reflect this great sense of movement and innovation, while at the same time to influence the agenda in the study of democracy. As such, it is a book about the possible futures of democracy as much as its present character and problems.

Theory and practice This aim is far from representing a disengaged academic undertaking. There is ample evidence that democratic innovation today is by no means limited to the idea or the theory of democracy; political practice, too, shows that in many regions and countries democracy is subject to renewed experimentation. Sometimes new ideas develop first, and at others practical dilemmas prompt innovative thinking. Whatever its detailed roots, the profound and rapid period of reappraisal of democracy we are witnessing at the start of the new century might be traced broadly to the removal of Cold War shackles upon what ‘democracy’ could and should mean to countries and cultures across the globe. The complex relation

4

Michael Saward

between nation, nationalism and democracy has taken on urgent and bloody political importance from Rwanda to ex-Yugoslavia, not forgetting the variety of dilemmas and solutions that the creation of Nunavut and Quebec separatism respectively represent for many places beyond Canada.2 Global and regional pressures, not least the rapid diminishing of states’ autonomous capacity to influence their own economic destiny and the continuing, democratically contested consolidation of the European Union (EU) as a new political form, lead us to rethink democracy both ‘above’ and ‘below’ the level of the nation-state.3 Apparent voter apathy and – as some would have it – voter ignorance force us to rethink the legitimacy of democratic regimes and decisions, and to look for ways in which a more informed citizenry could be fostered through new (and revived older) forms of information, deliberation and association. A new emphasis on (variously) ‘identity’, ‘difference’ and ‘diversity’ in both monocultural and multicultural societies prompts us to ask again a venerable question of democracy: whose interests should be represented politically, and how can their effective representation be achieved? The rise and evolution of environmental consciousness, and the fertility of political ecology as a source of new thinking about democratic politics, has thrown up new ideas on locality and globality alike, and has suggested new mechanisms and institutions to deepen and broaden a recast democracy. Further, a more educated and demanding public in many countries has put direct forms of democracy back onto the agenda – after all, if democracy is a good thing then surely more of it, in more direct forms, is likewise a good thing? Some of these concerns are theory-led (though no less practical for that): academic theorists worry about, and offer solutions to, problems whose political profile remains relatively low. While most governments today must pay at least symbolic attention to ecological constraints, for example, reconfiguring political structures to be more environment-friendly is only just registering on the radar screens of many governmental actors and institutions. Other concerns involve academic theorists struggling to keep up with, and to offer explanations of and guidance for, fast-moving real-world developments, such as the impact of globalizing tendencies on the possibilities for democracy adapting and prospering in the first ‘e-century’. The authors of the chapters in this volume are primarily democratic theorists, responding to and taking forward key debates on the future of democracy, but the need to render theory influential in pressing practical debates is a theme that runs throughout.

Innovation? The phrase ‘democratic innovation’ expresses a critical commitment to democratic values of popular participation and political equality, allied to an urgent imperative for theorists to articulate and analyse new solutions to the problems of democracy. It suggests that there is only one legitimate framework within which to innovate – the democratic one.4 But it also suggests that the framework in question is not wholly fixed in its make-up and dimensions; the elements of democracy,

Democratic innovation

5

such as participation, accountability, responsibility and an ‘open society’, can be interpreted and interlinked in new and sometimes surprising ways to respond to, or to reflect, new circumstances. We do not want to suggest that all innovations discussed in these pages are ‘new’ – that would be ahistorical at best. Classic writers from Madison to de Tocqueville to John Stuart Mill figure throughout what follows. We interpret innovation broadly. It can – and usually does – refer to revived and adapted older ways of thinking about politics and democracy as much as the genuinely new. Direct democracy, for example, is one of the ‘oldest innovations’ in the recorded history of politics, but standard texts on democracy dismiss its relevance today as easily as (for example) John Stuart Mill was able to in Considerations on Representative Government. Deliberation – the apparent contemporary lack of it and the great need for more of it – is the major focus in the book; this too is a theme familiar from the time of Aristotle. Representation is arguably the distinctively modern democratic institution – supposedly the great discovery that allowed ‘democracy’ to transcend its face-to-face origins to become adoptable for modern, industrial nation-states (Dahl 1989) – though the historical and theoretical links between democracy and representation are far more ambiguous than that. So, while there may be nothing altogether new under democratic theory’s sun, the malleability of democracy’s core ingredients enables complex permutations and possibilities; there is much in the menu of historical possibilities to be revived, reshaped and applied to the demands of today and tomorrow with their genuine newness and uncertain trajectories. With that said, I turn to the three major themes of the book by way of a brief introduction to the chapters (I leave detail to the chapters themselves, for we have striven for brevity and clarity throughout).

Deliberation and democracy The deliberative model of democracy has been the dominant new element in democratic theory over the past ten years; indeed the debates around ‘deliberative democracy’ since about 1990 are probably the most sustained and intense exchanges in political theory for many decades. This model arose (variously) out of concern that dominant ‘aggregative’ conceptions of democracy were inadequate: democracy is more than just counting heads, it must involve discussion on an equal and inclusive basis, which deepens participants’ knowledge of issues, awareness of the interests of others, and the confidence to play an active part in public affairs. In part a democratic theory response to abstract dialogical models in Rawlsian and Habermasian political philosophy, deliberative democracy looks to transform given preferences, not merely to design mechanisms to register them. Not surprisingly, a number of influential strands of deliberative democratic theory exist; indeed, different strands are increasingly going their own way (for example, the projects of Fishkin and Dryzek, represented here). Nevertheless, a reasonable stab at a common definition of deliberative democracy is that

6

Michael Saward

of Bohman (1998: 401): ‘Deliberative democracy, broadly defined, is . . . any one of a family of views according to which the public deliberation of free and equal citizens is the core of legitimate political decision making and selfgovernment’. A number of high-profile book-length accounts of this model have recently been published (e.g. Elster 1998c; Bohman and Rehg 1997b; Gutmann and Thompson 1996; Macedo 1999). However, two vital strands are largely missing from these and other recent appraisals: detailed attention to the institutionalization of deliberation, and more fundamental criticism of the wider claims made by the more sanguine deliberative theorists. A number of the chapters which follow address directly these (and other) important gaps in the literature.5 The chapters by Fishkin and Luskin, Smith and Eriksen explicitly link deliberative theory to practical deliberative forums (experimental and otherwise) – such as deliberative polls, citizens’ juries, and more broadly deliberation in EU institutions – pointing out the strengths and limitations of the theory once we think in terms of practical institutional forums. Fishkin and Luskin champion random sampling to gather together a group of citizens to deliberate on policy issues. They argue that ‘deliberative polls’, subject to highprofile and continuing experimentation in Britain, the US and elsewhere, can help to reconcile demands for political equality and deliberation – demands which, in the US at least, they say have come increasingly into conflict in the context of the advance of ‘plebiscitary’ democracy. They also explore the philosophical roots of the deliberative ideal, providing a robust defence of its democratic credentials and potential. Smith too examines deliberative polls, along with citizens’ juries, consensus conferences and other similar mechanisms, in the light of deliberative theory. He draws out the distinctive strengths and weaknesses of these devices, assessing them against key criteria such as inclusiveness and the quality of deliberation. He moves on to pose the vital question: can deliberative democracy be more than this? His response leads to a broader consideration of other potential democratic mechanisms, such as associative democracy and group representation (and thus provides links to major themes later in the book). Deliberative polls and citizens’ juries (and the like) are ‘artificial’ forums, consciously designed to promote forms of deliberation that might enhance decision-making at some level. But the potential impact of deliberation can be traced in other, larger institutions, and indeed in ‘messy’, more ‘organic’ civil society contexts. Eriksen in Chapter 4 shows how deliberative theory can make us look at familiar political problems in a fresh light. He argues that the democratic legitimacy of the EU does not rest solely on the relative power of representative institutions such as the European Parliament, or on the level of citizen participation (as in European elections). It may also be the product of necessary deliberations between the myriad committees of the EU system, and even the much-maligned veto power of member states in the Council of Ministers. These sites of political activity constrain participants to engage in processes of communication which carry a democratic, rationalizing force in themselves. Curiously enough, Eriksen’s argument points us to ways in which deliberation may lessen the urgency of more

Democratic innovation

7

conventional democratic reform, in this case to address the so-called ‘democratic deficit’ in the EU. But deliberative democracy is not without its critics. Rättilä and Saward in Chapters 3 and 5 respectively, and Budge in Chapter 15, take on some of the deliberative model’s most fulsome claims at their roots, questioning (for example) the basic link between deliberation and democratic legitimacy, and the vision of the ‘public sphere’ that model promotes. Rättilä takes to task the prominent American deliberative theorist Seyla Benhabib regarding tensions between (a) wanting open, equal and inclusive deliberation in a broad public sphere, and (b) wanting that deliberation to conform, in some minimal fashion at least, to particular standards of rationality. The latter is difficult to dispense with, but at the same time carries dangers of elitism, according to Rättilä. One potential way out of this impasse is via the view of the public sphere taken in Habermas’s Between Facts and Norms (a work often cited in the book – though not always supportively!), where the public is conceived as a ‘space’ open to all forms of political expression, and is distinguished clearly from the formal parliamentary law-making process with its own canons of deliberative rationality. Through examination of various arguments about the siting of deliberation and the demands of political legitimacy, Saward argues in Chapter 5 that we ought to question the adequacy and strength of widespread claims about how deliberation enhances the legitimacy of democratic outcomes. He also raises concerns about strongly non-deliberative, and possibly elitist and even anti-democratic, strands of thought at the core of prominent deliberative writings. The greater part of Chapter 5 is taken up with a close reading of Joshua Cohen’s article ‘Deliberation and Democratic Legitimacy’ – a key text in the deliberative democracy canon – in an attempt to show significant weaknesses of influential styles of deliberative democratic argument. Budge’s Chapter 15 chimes neatly with Chapter 5 in many ways. Though primarily concerned with the continuing importance of political parties to democratic theory and practice – despite many texts virtually ignoring them – Budge questions sharply the adequacy of the deliberative model of democracy, with (as he puts it) its ‘university seminar’ vision of deliberation. Real politics involves the clash of interests and the counting of heads before it involves (or can involve) such an idealized conception of discursive politics. And parties remain vital to dealing with the enduring challenges of real politics, from coping with the instability of choice mechanisms to the most feasible and desirable future deepening of democracy, along direct democratic lines. Dryzek in Chapter 6 both extends the distinctive vision of discursive democracy that has influenced these debates throughout the 1990s and responds to criticisms by Saward and others of the deliberative model. Dryzek expresses his disappointment at the overly conventional approach of Habermas in his recent work, and is prompted in part by this to distinguish ‘discursive democracy’ more sharply from other variants of ‘deliberative democracy’, elaborating it as a distinctive and radical alternative to ‘liberal constitutionalism’ in particular. Discursive democracy is more about democratization than the formal features of democracy, more focused upon dynamic networks of actors in the public sphere (and across

8

Michael Saward

national borders) pressing states to act than formal features of the state itself. This conception, Dryzek argues, is less vulnerable to the charges Saward and others lay against deliberative democracy. It also demonstrates, he says, that Budge’s criticism of a ‘university seminar’ model misses the mark, at least if it is targeted at his own approach. Thus, the really difficult and vital questions attending democracy and deliberation weave in and out of these chapters. Does deliberation demand new, artificial forums that are more or less ‘controlled’, or ‘uncontrolled’ sites of expressive politics outside and against the state? What impact on traditional democratic politics can we hope, or expect, either to have? Is it really a new model of democracy, as its name suggests, or a set of useful but limited adjuncts to enduring, older models and institutions of democracy?

Representation and deliberation A central part of the debate on deliberation has been the question: who is to do the deliberating? Issues of how diverse interests can or should be included or represented – how we can ensure, democratically, that excluded voices might be heard in the democratic ‘conversation’ – have thrown up fascinating possibilities for the reshaping of that most basic of modern democratic institutions, representation. The second part of the book builds upon feminist and postmodern theory, theories of group representation, and ecological political thought to present a provocative set of arguments covering, among other things, the desirability of a ‘politics of presence’, the representation of ‘difference’ and the excluded, innovative means by which the interests of the non-human world might be brought to bear on democratic governance, and how the alleged shift from government to governance opens up new possibilities for ‘democracy from below’. While the chapters on representation take up classic themes in the literature – the Burkean issue of whether the role of representatives is to be as ‘delegates’ or ‘trustees’, whether anyone can represent anyone or whether (for example) only women can really represent women, the extent to which representation is the core device of a functioning democracy, or one which is subordinate to other constitutional essentials, and the degree to which ordinary people, acting as citizens, might ‘represent themselves’ – they are also crafted in the light of the deliberative challenge to democratic theory orthodoxy. Squires in Chapter 7 finds in Iris Marion Young’s ideas of ‘communicative democracy’ and group representation the major challenges arising from feminist theory for democratic theory. As provocative and challenging as Young’s radically inclusive ideas are, however, Squires’s searching critique uncovers tensions and ambiguities that continuing advocacy of group representation would need to address. She prefaces these arguments with reflections on the key lines of debate in recent feminist theory, offering valuable clarification of such terms as equality, identity and difference and the substantive positions, central to the democratic project today, which they represent. Meier’s discussion in Chapter 8 picks up a number of threads from Squires’s

Democratic innovation

9

chapter to elucidate the influential account of the ‘politics of presence’ in the work of Anne Phillips (1995). After untangling the deceptively simple idea of the politics of presence, Meier reveals the range of complex arguments and difficult judgements inevitably involved in practical representative innovations. Her chapter explores in particular recent Belgian law on gender quotas, placing this new development in the context of a Belgian political culture whose consociationalist history makes it receptive to calls for group representation. Her reflections on the case study tell us a great deal about the desirability and feasibility of specific forms of group representation. Squires and Meier interrogate ideas about who should be represented and in what forms in a way that challenges certain liberal, individualist approaches to what democratic representation is about. Eckersley in Chapter 9 takes this, and her own project of revising democratic categories in the light of political ecology, a step further by asking why only human interests need to be represented politically, or (from a deliberative angle) included in the democratic dialogue. If certain non-human interests are morally considerable, she argues, then epistemological problems of how to include, say, animals in democratic dialogue should not readily be dismissed. Of course, whales cannot vote or talk; but people can speak for their interests, a fact that could readily be institutionalized in democratic systems via an Environmental Defenders’ Office operating according to the canons of the precautionary principle; in this sort of way we might truly approach a ‘democracy of the affected’. Other political ecologists are more sceptical of deliberative democracy’s ‘green’ potential, preferring to constrain democratic procedures more tightly by emphasizing the role of (in this case, ecological) constitutional limits on what governments or majorities can do. While broadly agreeing with the ‘green’ advantages of deliberative models of democracy, Mills and King question whether democrats can rest content with even strong forms of democratic proceduralism given the magnitude of the ecological risks (and therefore the compelling character of attendant ecological values) faced by contemporary societies. So, rather than resting their case on innovative designs to expand the range of interests represented in democratic politics, King and Mills make the case for certain non-majoritarian constraints, drawing upon self-limiting and related constitutionalist threads in the theory of democracy. One great tradition in both classical and modern thinking about democracy is the debunking of the view that representation is even democratically legitimate, let alone democratically desirable. Though they disagree on the grounds, both Dryzek and Budge for example advocate more direct political participation. Bang and Dyrberg offer a particular perspective on such scepticism. They provide a radical analysis of the nature and possibilities of democracy, contrasting ‘mediated’ approaches to democracy with more radical ‘bottom-up’ analysis of civic engagement. In so doing, they draw upon an eclectic range of sources: the mainstream political science literature on governance and networks in addition to the later, conventionalist Rawls and Foucauldian perspectives. The essence of their critique is to shift our focus to new forms of civic engagement – echoing some of the

10

Michael Saward

themes raised in Rättilä’s chapter – and to ask us to examine how people themselves define political reality rather than allowing others, whether representatives or ‘experts’ more generally, to define it for them. One thing at least emerges clearly from these chapters: that any consideration of deliberative democracy will be incomplete without consideration of how inclusive and effective representative structures are (or can be), especially from the perspective of marginalized human interests and groups, and of those who can reasonably be said to represent nature’s interests as well. One might say that the quality of the deliberation can never be so high as to compensate the excluded for their unfair treatment. Of course, the opening chapters also dealt with this theme; Fishkin and Luskin, for example, would advocate random sampling as the democratic representative mechanism par excellence. But theirs is work in democratic theory which locates itself in liberal and governmental terms, and within Anglo-American analytical and empirical traditions (the chapters by Saward, Smith and Budge might also be seen in this way). Other traditions, such as the Habermasian discourse ethics approach that in part animates chapters by Rättilä, Eriksen, Dryzek, and Eckersley, are concerned more with the play of discourse and interests in civil society, and which voices may be heard ‘there’ as well as within traditional representative institutions. The approaches are, of course, by no means exclusive, and in different ways the chapters by (for instance) Eckersley and Bang and Dyrberg show how fruitfully different such perspectives can be blended.

Associations and democracy Globalization and the emergence of diverse modes of governance in various countries in recent years has led many key theorists of democracy to look anew at civil society rather than the state as a key site of democratic practice. At the centre of this trend has been the model of associative democracy. The final part of the book comprises four chapters which between them offer a thorough, combative and multi-faceted appraisal of the revival of the associationalist turn in democratic theory. Rigorous arguments both for and against key associationalist claims in two chapters (by Perczynski and Roßteutscher) is followed by Herreros’s critical account of where the debates on social capital, social trust and associationalism have led us, and what ought to be the future agenda of research in this area. In the final chapter, Budge (as noted above) reminds us why political parties – along with unions, churches and mosques, pressure groups and bowls clubs – cannot be ignored in any comprehensive account of associations and democracy. Perczynski provides an upbeat assessment of the potential contribution of associations to democratic life. Exploring commonalities and contrasts between three especially influential accounts of associative democracy – those of Cohen and Rogers, Hirst, and Schmitter – he draws a major distinction between views of associations that stress their self-organizing or spontaneous nature, on the one hand, and their state-supported or state-sponsored character on the other.

Democratic innovation

11

Questioning the supposed non-democratic character of the latter, as found in the work of Cohen and Rogers for instance, he argues that an ‘imposed artificial solidarity’ is not necessarily democratically objectionable. Perczynski also addresses other common criticisms of associative models, claiming that the latter offer an attractive, participatory vision of democratic citizenship, including a view of how associations can act as ‘schools for democracy’. Roßteutscher in Chapter 13 provides by way of sharp counterpoint a sceptical argument about associative democracy which challenges many of the defences built by Perczynski. Drawing on both the theoretical literature and empirical evidence, Roßteutscher argues that there is a variety of critical contradictions and gaps in the literature on associative democracy, which must lead us to query its coherence as a model of democracy. Macro- and micro-approaches speak past one another; key claims of micro-theories do not sit well with the available empirical evidence; and macro-theories, though prominent and influential, have not yet shown that their claims can bear close empirical scrutiny. Empirical confusion may be compounded by political opportunism: Roßteutscher concludes her reflections with a claim that there is a danger that the popularity of associative models acts as a cover for the dismantling of the welfare state. Broadening the context of the associationalist debate, Herreros examines the multi-layered links between civic republicanism, social capital, social trust, and associations. It is argued that the social capital approach to democracy carries with it the assumption that individual actions are self-centred and calculating. However, the generation of social trust requires something more: virtuous citizens. Herreros argues that certain types of associations – those that are ‘horizontal’ rather than ‘vertical’ in their structure, and ‘political’ rather than ‘social’ in their self-definition – are more effectively able to promote the type of deliberation which can foster civic virtue and social trust. Striking a note of caution, he underlines the point that the causal links between these key concepts are far from clear – echoing a key concern of Roßteutscher’s – and that much is yet to be learned through future conceptual and empirical work. Aside from its broader relevance to key themes across the volume, Budge’s chapter provides a counterpoint to certain core associationalist as well as deliberative themes, stressing the potential of the party-based model of direct democracy (grafting the oldest democratic innovation onto orthodox liberal democratic institutions). It is striking that political parties have played so little role in influential associationalist writings, including that of authors like Paul Hirst who are based in traditions where parties are strong in their organization and distinct in their ideologies. Budge provides a robust defence of the importance of parties to both the deliberative and the associationalist literatures.

Crosstalk It will already be clear from the brief comments above that the chapters in the book are by no means confined to dealing with the sub-theme defined by the section of the book in which they appear. In a great many ways, chapters ‘talk’ to

12

Michael Saward

each other across these divides, which as usual turn out to be artificial (though still useful). For example, all contributors to the sections dealing with representation and association refer to directly (and take forward) the deliberative themes debated in the first section. In terms of individual chapters Fishkin and Luskin argue the case for scepticism about direct forms of democracy advocated by Budge and supported in different ways by Bang and Dyrberg and Saward. Smith picks up and discusses themes raised by Fishkin and Luskin, providing further criteria by which the operations of deliberative polls might be assessed. Eriksen responds to Saward’s critique of deliberative models of democracy by holding that deliberation is more basic than, and prior to, voting, and that it can enhance legitimacy and democracy in EU institutions, while Dryzek explicitly ranges across the themes of the whole book, with a focus on responding to the deliberative model’s critics. Squires and Eckersley build their arguments about representation in part on the critique of the deliberative model’s limitations discussed by Rättilä, Saward, Smith, Dryzek, and others; and Eckersley and King and Mills present contrasting approaches to the incorporation of non-human values and interests in democracy. The chapters by Perczynski, Roßteutscher and Herreros refer directly to each other, in agreement and disagreement, on a number of levels regarding the contested democratic character of associationalism, while Budge’s chapter takes up where deliberative and associative models are relatively silent, noting the nature and importance of that silence. At this point, the chapters must now be allowed to speak for themselves. As a final introductory comment, it is worth noting that, of course, not all important innovations in current thinking about democracy are represented in these pages. Perhaps the most notable omission is a discussion of the ‘cosmopolitan model of democracy’, mentioned earlier. Participatory democracy in the contemporary classic sense, exemplified in the work of Pateman (1970) and MacPherson (1977), is also missing as a discrete topic (though the concerns of these approaches are by no means absent). Direct democracy, too, is arguably under-discussed, though Budge builds his arguments in part around its desirability. It could be argued that the character of the dominant thread in recent democratic theory – the deliberative model – has been such that it has squeezed out more explicit focus on other models. We have taken a middle course, not attempting comprehensiveness regarding democratic innovation – hardly possible within one volume – but focusing closely on debates and models that have animated debate on democracy’s character and future within and across the ten countries and three continents from which the authors are drawn. Given the increasingly rapid pace of technological, economic and cultural change all countries are currently experiencing, it would be surprising if the democracy of 2050 were not to be markedly different to today’s. We can hope and trust that democracy will be neither irrelevant nor discredited; in the meantime we can extend its boundaries as we find them in directions we can best justify to ourselves and others, whether that means ‘back to the future’ or forward to the unknown. We hope this book can make its contribution in that spirit.

Democratic innovation

13

Notes 1 Although Manglapus’ claim that many non-Western cultures practised democracy of sorts, before ancient Athens and since, can be criticized for interpreting scant evidence loosely, it also serves as an important reminder that democratic values are not wholly Western by any means. 2 We opened the ECPR workshop that prompted this book on the day NATO forces entered Kosovo, reminding us of the good fortune of our freedom to deliberate and underlining the critical seriousness of workable innovations by an engaged democratic theory community. 3 The most extensive effort to think through a ‘cosmopolitan model of democracy’ can be found in Held (1995). A range of recent perspectives can be found in Holden (2000). 4 The value of democracy is not a topic I deal with directly here. Most chapters, of course, address the issue in one way or another. Extended discussions on democracy’s justification can be found in Harrison (1993) and Saward (1998). 5 My thanks to Graham Smith, Judith Squires and Fraser King for helping with useful summaries of early versions of a number of chapters.

Part I

Deliberative democracy Advocacy and critique

1

The quest for deliberative democracy1 James S. Fishkin and Robert C. Luskin

The history of democratic reforms is a history of two closely related trends. The first has been toward more direct public consultation. The public has been given a larger and larger role in determining who makes policy decisions and even in deciding directly on policies. Some obvious examples in the US include the spread of mass primaries and relatively open caucuses in candidate selection, the direct election of US Senators, the atrophying of the Electoral College into a mere vote aggregation scheme, the increasing reliance on referenda and initiatives in legislation, and the evolution of public opinion polls into unofficial, advisory referenda. Although the first few of these examples are specific to the US, the same general trend pervades the democratic world. The second trend has been toward broader public consultation. The public being consulted more directly has also grown wider and wider. In the US, the relevant changes have included the broadening of the franchise to include nonproperty owners, women, blacks, and eighteen- to twenty-year-olds; the removal of poll taxes, residency requirements, and other barriers to voting; and positive measures to facilitate registration and voting such as balloting periods of more than one day and motor-voter statutes. Again the rest of the democratic world has been moving along the same curve.

Equality, deliberation and deliberative democracy Generally speaking, these changes have served political equality, the democratic ideal of weighing all citizens’ preferences equally.2 But this pursuit of equality has had the unfortunate effect of undermining another democratic value, deliberation, broadly defined as the serious consideration of arguments and counter-arguments for and against policy alternatives.3 Note, parenthetically, that this notion of deliberation does not absolutely require discussion, face-to-face or otherwise. Theoretically, the arguments and counter-arguments could be raised either through mediated discussion, as by some synchronous or asynchronous on-line exchange, or even without discussion, as by written digests of the full range of arguments that should be considered. Realworld deliberation is a mix – people read, watch, and listen; people ruminate; people discuss. But it does seem safe to say that deliberation quite centrally involves

18

James S. Fishkin and Robert C. Luskin

discussion, and indeed that at least some of the benefits of deliberation would be harder to attain without it. The irony of the institutional changes promoting equality is that they have moved the effective locus of many decisions to a mass public subject to what Anthony Downs (1956) famously termed ‘rational ignorance’. Why should individuals with only one vote in millions spend a lot of time and effort becoming informed about a national policy debate or political campaign? No individual vote or opinion is likely to make any difference. Where deliberation requires immersion in policy-relevant arguments and counter-arguments, the incentives for rational ignorance promote inattention and non-participation. Thus democratic institutions appear to face a choice between politically unequal but deliberative elites and politically equal but non-deliberative masses. Is there some way of combining deliberation and equality? The quest for institutions to accomplish this is the quest for deliberative democracy. Existing democratic institutions based on elected representation may be seen as an effort to achieve the ideal of deliberative democracy by stages. They combine equality at a preliminary, electoral stage with deliberation at a subsequent, decision-making one. The division is not absolute. There are elements of deliberation in elections (as citizens, to whatever usually modest degree, learn about and discuss the choices) and of equality in representative decision-making (one legislator, one vote, at least at final passage). But by and large the equality (such as it is)4 lies in elections and the deliberation in discussions among elected representatives. But this strategy of concentrating deliberation among elected representatives is far from ideal. At the decision-making stage, considerations other than those likely to figure prominently in mass deliberations may carry great weight. Elected officials have been known to pursue their own pet interests or those of powerful individuals or organizations above those of the majority of their constituents. The equality achieved by elections is often subverted. At other times, elected representatives concerned with re-election may simply follow the polls, abandoning any meaningful deliberation. That would not necessarily be bad, if the polls reflected substantial public deliberation. But of course they do not. The results stand in generally unknown and often quite limited relation to the views people would hold if they had the chance of learning, thinking, and talking about the issues (as illustrated below). The decision-making stage, then, is no-win: ignoring mass preferences vitiates equality; following them vitiates deliberation. At the electoral stage, the paucity of deliberation means that there is no great deal of equality (beyond the merely formal) either. Because citizens are often – but varyingly – unengaged and ignorant, not everyone votes. In the US and Switzerland,5 almost half the eligible electorate abstains, even in the most salient national elections. Electorally, non-voters have no voice. In a subtle but important sense, moreover, even voters have unequal voices. Every vote carries equal weight, of course (issues of fraud and apportionment aside). But the paucity of deliberation also means that not everyone votes as he or

The quest for deliberative democracy 19 she would with ‘full information.’ And because information is very unevenly distributed across the electorate, some people approximate their full-information preferences much more reliably than others. Thus while voters’ actual preferences are equally represented, their full-information preferences are not.6 These shortcomings of elected representation have made the quest for deliberative democracy into a quest for a more unitary process capable of embodying both equality and deliberation, a process of deliberation among citizens, not just among legislators or other elected representatives. The key problem is how to overcome the barriers to citizen deliberation in communities numbering in the thousands or millions.

The Deliberative Poll (and other deliberative microcosms) An ancient solution to this problem of scale was the deliberative microcosm selected by lot – in modern terms, by simple random sampling. Random selection gave each citizen an equal probability of participating. The Athenians used this device for citizens’ juries of 500 or more, for legislative commissions, for the Council (which set the agenda for the Assembly), and in a variety of other contexts. By the fourth century BC, legislative commissions of several hundred citizens selected by lot were making the final decisions on legislation (see Hansen 1991 for a systematic account). These ancient examples have lately inspired a range of deliberative microcosms, including citizens’ juries in the US and Britain (Coote and Lenaghan 1997), planning cells in Germany and Switzerland (Dienel and Renn 1995), and consensus conferences in Denmark ( Joss and Durant 1995). The shared idea is to gather citizens together in small groups to discuss policy issues face-to-face, thus providing both opportunity and incentives for the participants to behave more like ideal citizens. For the participants, at least temporarily, it becomes less rational to be ignorant. Instead of being just one vote among anonymous millions, each has one voice among ten or twenty in the small-group discussions and one among only hundreds even aggregating across groups. In addition, the necessity of talking to and in front of one’s fellow participants creates social incentives to inform oneself and think about the issues. The point is to offer the participants’ post-deliberation opinions as an input to politics and policy. To the degree the participants represent the whole public – an important stipulation – the deliberative experience gives the final distribution of opinion a certain normative relevance, even as it renders it highly counterfactual. The wider public remains rationally ignorant, unengaged, ill-informed. In contrast to ordinary polls, showing public opinion as it is, these deliberative microcosms attempt to show public opinion as it would be if its members learned, thought, and talked more about the issues. Our own version of this general idea is the Deliberative Poll. We take a random sample of some relevant population (country, region, city), interview them, and invite them to a single site for a deliberative weekend. We send them carefully

20

James S. Fishkin and Robert C. Luskin

balanced briefing materials, laying out the main policy options and the arguments for and against each. We bring them to the site (all expenses plus an honorarium paid) and set them to deliberating in two alternating modes: small-group discussions led by trained moderators and plenary sessions to put questions composed by the small groups to panels of competing politicians and policy experts. At the end we ask the participants the same questions as at the beginning. While all deliberative microcosms involve deliberation, the Deliberative Poll goes furthest toward preserving equality – as part, in this context, of representation. Two nested sorts of representation, and of equality, are relevant. The first is predeliberation representation of the actual public – demographically, attitudinally, behaviourally, and in terms of knowledge, thought, and involvement. What may or may not be equal here is the probability of being drawn into the deliberative sample. The second sort is post-deliberation representation of a hypothetical, more engaged, thoughtful, and knowledgeable public. What may or may not be equal here, in addition to the probability of being sampled in the first place, is one’s contribution toward determining the post-deliberation results. Good representation of the first sort is a necessary but insufficient condition for good representation of the second. One advantage of Deliberative Polling for post-deliberation representation lies in the nature of the deliberative process and results. Citizens’ juries, planning cells, and consensus conferences are designed to reach some public, collective decision. The Deliberative Poll, in contrast, is designed simply to help its participants inform and clarify their own thinking. We gauge their opinions with the same privately completed questionnaire as at the outset. No one else knows what responses they give. Their opinions are aggregated only statistically, after the deliberations are done. This is essentially a ‘secret ballot’ and has the same advantage vis-à-vis the social aggregation of opinions in citizens’ juries, planning cells, and consensus conferences as the secret ballot does for elections in the outside world: it minimizes the intrusion of other people’s preferences, and thus the inequalities that creep in when some people, inevitably, have greater economic, social, or informational resources than others. The advantage of Deliberative Polling for pre-deliberation representation is still clearer-cut. Equal probabilities of selection require probability, i.e. random, sampling.7 But the participants in consensus conferences are self-selected, and while planning cells have frequently employed random sampling, they have done so only within non randomly selected communities, making their pooled samples, aggregated across communities, non random.8 But representation of both sorts involves not only equality but sample size. Even assuming random sampling, the number of participants must be sufficiently large to lend some assurance that the sample resembles the population – as it is, before deliberation, and as it might be, after. A random sample of several hundred is very unlikely to differ radically from the population. A sample of only ten or twenty, the size of most citizens’ juries and planning cells, may easily do so. Both citizens’ juries and planning cells are therefore much too small to be especially representative. The pooled samples from several parallel planning cells may be large

The quest for deliberative democracy 21 enough but are non random. The result is that there is no great assurance that another jury or cell of the same size, drawn from the same population, and deliberating in the same fashion on the same topic would yield anything like the same results – or, equivalently, that the whole population, divided up into citizens’ juries or planning cells, would do so.9 In Campbell and Stanley’s (1963) terms, the Deliberative Poll is a quasiexperiment, whose one compound treatment consists of everything from the invitation to participate to the last moment of on-site deliberation. In anticipation of the weekend, participants begin to pay more attention to the news, to think more about the issues, and to discuss them more with friends and family. Many of them also read the briefing materials prepared for the event. Perhaps the most important part of the treatment, however, occurs on the weekend itself, in the small-group discussions among the participants and in the dialogue with competing experts and politicians. When the idea was first proposed, one sceptic questioned the necessity for deliberation, arguing that ‘the process of group interaction will produce substantially different results than if the participants consider alternatives and are asked questions in private’ (Traugott 1992: 27). But we believe it important that the participants have the opportunity of hearing others’ views and questions and considering others’ interests. Our ideal citizens would not simply receive and digest political information in private, then respond by voting or otherwise expressing their views without ever having shared or tested them in conversation. Rather, they would talk with a wide variety of other citizens, hearing and weighing opposing points of view in the process of coming to their own considered judgments. They would deliberate. One other important feature needs mentioning: In every Deliberative Poll so far, the proceedings or results have been broadcast to the rest of the community (nationally, in the case of the national Deliberative Polls; regionally, in the case of the regional ones). They have also received coverage in radio and newspaper stories, on internet websites, etc. Such broadcasts and stories give the Deliberative Poll the possibility of making a contribution to the actual practice of democracy. Both citizens and policy-makers can be made aware of the conclusions of the Deliberative Poll and the considerations that weighed with the microcosm in reaching those conclusions. They may help set an example for representative and informed citizen consultation in an age of self-selected viewer call-ins and televised ‘town meetings’ based on convenience samples.

Deliberative Polling and democratic theory The key theoretical and empirical question to do with Deliberative Polling is how a counterfactual world of informed and thoughtful citizens discussing policy issues would differ from the very different one we know. No mass public anywhere in the world has been continuously and seriously engaged in public issues.10 Between such an ideal public fulfilling the aspirations of democratic theory and real-world publics mired in rational ignorance yawns such a chasm that we have

22

James S. Fishkin and Robert C. Luskin

trouble telling much about what the ideal case would be like. Theorists can only attempt to imagine it. The point of Deliberative Polling is to approach these questions in an entirely different way, moving from the thought experiments of democratic theory to the actual experiments (and quasi-experiments) of social science. From the standpoint of democratic theory such experiments help us approach a problem central to the 2,500 year old dialogue about the desirability of democracy. Advocates of democracy have long faced the challenge that the whole public might not be sufficiently competent to rule. Why not have rule by meritocratic or technical elites instead? Or by only the more educated and the higher strata of society? Scepticism about the competence of ordinary citizens famously motivated Plato’s Republic (although he later qualified this scepticism by his willingness to rely on random microcosms of the citizenry in The Laws). It also motivated J. S. Mill’s (1991) proposal for ‘plural voting’, in effect allotting extra votes to those who could demonstrate their greater wisdom or intelligence. Every society that has invoked the notion of democracy or popular rule has wrestled with this challenge. And every known democracy has employed some combination of public consultation and decision-making by elites. The issue of the appropriate mix, however, remains. In employing a representative microcosm, deliberative polling sheds light on the public’s potential for competent decisionmaking. How far could we trust ordinary citizens, even under unrealistically favourable conditions, to inform themselves and discuss and consider the issues? Deliberative Polling also illuminates a related issue very much at the intersection of democratic theory and political behaviour: namely, what differences it would make to the distribution of policy and electoral preferences if the citizenry knew and cared more about politics. There is now little controversy about claims that most people know little about the details of politics and policy (Price 1999; Luskin 2000). There remains, however, a contention that ill-informed citizens manage by cognitive short-cuts or ‘heuristics’ to approximate their ‘full-information’ preferences (Popkin 1991; Lupia and McCubbins 1998). Such ‘low-information rationality,’ it is claimed, leads to roughly the same preferences as the ‘highinformation’ kind.11 A less-charitable but aggregatively equivalent contention is that while actual and full-information preferences commonly differ, the differences cancel out, with some citizens erring in one direction and some in the other, so that the distributions of actual and full-information preferences are nonetheless the same (Page and Shapiro 1992). We do not doubt that people commonly use cognitive short-cuts in thinking about politics, as about other matters. The question is how far those short-cuts lead to the same preferences as fuller information and more thorough reflection would do. A number of studies using cross-sectional survey data to simulate distributions of preferences if everyone in the sample had the maximum score on some measure of information suggest the existence of some – sometimes sizable – gap between full-information and actual preferences (Bartels 1996; Delli Carpini and Keeter 1996; Luskin and Globetti 1997; Althaus 1998). But these are only simulations: nobody’s information is actually increased. In taking an experimental

The quest for deliberative democracy 23 approach to the same question, Deliberative Polling, by contrast, provides measurements of what happens to the distribution of preferences when given individuals are actually made more informed. The results suggest that more informed and engaged citizens would indeed frequently hold different preferences and cast different ballots.12

Some results We have now conducted fifteen Deliberative Polls in varied contexts. In Britain, we have conducted five national Deliberative Polls in collaboration with the television network Channel 4 and the London research institute Social and Community Planning Research (now the National Centre for Social Research), on crime (1994), Britain’s future in Europe (1995), the monarchy (1996), the economic issues in the British general election (1997), and the future of the National Health Service (1998). In each case, a national random sample was transported to a single place (the Granada Television Studio in Manchester) and, after small-group discussions and dialogue with competing experts and politicians (usually the relevant cabinet ministers and shadow ministers), completed the same questionnaire as in their homes at the time of recruitment. In the US, we employed the same basic model at the National Issues Convention in January 1996. A national random sample of 466 eligible voters from all over the country assembled in Austin, Texas to hold a four-day-long dialogue on three major issues: the economy, America’s role in the world, and the state of the American family. The participants got to put questions to panels of policy experts, several of the Republican presidential candidates, and Vice President Al Gore. The event received nine-and-a-half hours of broadcast coverage on PBS (hosted by Jim Lehrer), was seen in all by some ten million viewers, and formed the subject of roughly six hundred newspaper articles. In the State of Texas, we have supervised eight deliberative polls on utility issues in Corpus Christi, Abilene, Shreveport (in Louisiana but covering adjacent areas of Texas), El Paso, Houston, Beaumont, Amarillo and Dallas. These had their origin in the Public Utility Commission’s requirement of ‘public participation’ for integrated resource planning by electric utilities and have been conducted with the Commission’s co-operation. The utility companies, in complying with this requirement, faced a dilemma. If they conducted polls of the conventional sort, the public would have so little information about the issues that the results would not be particularly meaningful.13 If they held ‘town meetings’ of the familiar sort, they would draw only self-selected and highly unrepresentative participants, notably tending toward representatives of organized interests. If they conducted focus groups, they would also be relying on non random (if less-predictably biased) samples, which would in addition be too small to be very representative even if randomly drawn. The Deliberative Poll, with its harvest of considered opinions from a representative sample of the population of the service area, proved a nice solution.14 In Australia, together with Issues Deliberation Australia, the Australian National

24

James S. Fishkin and Robert C. Luskin

University, The Australian newspaper, the survey firm Newspoll, and the Australian Broadcasting Corporation (ABC), we conducted a national Deliberative Poll two weeks before the national referendum of 7 November 1999 on the proposal to make Australia a republic, severing its remaining ties to the British crown. In both preparing the briefing materials and selecting the expert panels, we had the co-operation of the official ‘yes’ and ‘no’ committees for the referendum. The sample, numbering 347 Australian voters, met in the historic Old Parliament House in Canberra. The proceedings were broadcast nationally on the ABC and formed the basis of a segment of the highly-rated commercial television programme ‘Sixty Minutes.’ The data permit a variety of analyses. By comparing the actual participants to those initial interviewees who did not make it to the on-site deliberations and to census and other demographic data, we can demonstrate that the participant samples are highly representative, demographically, attitudinally, and in all other respects. By comparing the responses to knowledge questions in the initial interview to the responses to the same knowledge questions immediately after the on-site deliberations, we can demonstrate that many of the participants learn something about the issues. From note-taking and transcripts, we can say that contributions to the small-group discussions were, quantitatively, at least, surprisingly equal. Some participants talked more than others, of course, but relatively few said nothing, and differences one might have expected, as between men and women, were relatively small. And by comparing responses in the initial interview to responses to the same policy questions immediately after the on-site deliberations, we can demonstrate that, on many issues, many participants change their views. The magnitude of the learning and attitude change that can occur is suggested by the results of the recent Australian Deliberative Poll, whose participants registered increases of as much as fifty or sixty points in knowledge of basic facts about the constitution and the referendum proposal and increased their support for the referendum proposal by twenty points, from 53 to 73%. The voting intentions of a more fully informed electorate turned out to be very different from those of the actual electorate (which rejected the referendum by a small margin). These are admittedly among the largest information gains and attitude changes we have seen, but notable results of both sorts are extremely common, and these are not the only spectacularly large ones. To cite two more examples: the percentage of our 1997 British sample intending to vote for the Liberal Democrats in that year’s general election increased by twenty points. Across the eight electric utility Deliberative Polls, the percentage of our samples willing to pay extra to see the utility make greater use of wind or solar power increased by an average of thirtytwo points.15 The influence on politics and policy has been generally harder to assess, but the utility polls have been a major exception. The results have been incorporated into the integrated resource plans filed with the Public Utilities Commission. Based on the results of the first three polls, the utility companies have made the largest investments in renewable energy in the history of the state, an outcome well-

The quest for deliberative democracy 25 received by stake-holder groups like the Environmental Defense Fund, as well as by the companies themselves.16 Recently, the Commission has announced that it would move to ‘green pricing’ state-wide, based on the Deliberative Polling results.17

More on the role of deliberation The Deliberative Poll fulfils some of the functions of Mill’s (1991: 116–18) ‘Congress of Opinions,’ the most important role he hoped representative bodies might serve: Where every person in the country may count upon finding somebody who speaks his mind as well or better than he could speak it himself – not to friends and partisans exclusively but in the face of opponents, to be tested by adverse controversy; where those whose opinion is overruled feel satisfied that it is heard, and set aside not by a mere act of will, but for what are thought superior reasons . . . [the Congress is] when properly constituted a fair sample of every grade of intellect among the people which is at all entitled to a voice in public affairs. If by ‘fair sample’ we mean one that does not depart too strikingly from the whole population, and by ‘intellect’ we mean not intellectual capacity (of which we have no direct gauge) but cognitive complexity about politics, our deliberative samples have been ‘fair samples’ not only of intellect but of every demographic, attitudinal, and behavioural characteristic we can measure.18 On any given point, most people should be able to find participants who speak their mind, and frequently as well or better than they could speak it themselves. And it should be obvious, to outside viewers as well as participants, that the final distribution of opinions stems from what the participants collectively have seen as ‘superior reasons,’ after having heard competing arguments voiced and responded to. The deliberative process seems to involve not only the acquisition of purely factual information but also some greater degree of mutual understanding. Toward the beginning of the National Issues Convention weekend an 84-year-old conservative white man from Arizona was heard during the dialogue on family issues to question whether one of his fellow small-group members, an AfricanAmerican woman on welfare from New York, really ‘had a family.’ A family, he said, requires ‘a mother and a father and children’ in the same household. At the end of the weekend, he came up to her and asked ‘what are the three most important words in the English language? They are “I was wrong.” ’ After four days of discussion, he had come to understand that she did have a family. She had become more than a fleeting, stock image on television. She had become a real person, with real needs and interests. This change of view is of a kind we have observed quite a number of times in our small-group discussions. It reflects a gain in empathy or mutual understanding. It also reflects learning, of a particular kind. The participants gain

26

James S. Fishkin and Robert C. Luskin

abstract, affectively-cool information about budgetary statistics, unemployment rates and the like. But they also gain information about the circumstances, interests and views of others quite differently situated from themselves. This story raises an issue about the counterfactual public whose views the Deliberative Poll is trying to capture. If the public as it otherwise is were simply to increase its frequency of political discussion, our 84-year-old white conservative Arizonan would be exceedingly unlikely to meet our African-American New Yorker on welfare, much less engage her in a discussion of welfare policy. Yet the deliberative dialogue is enriched by such exchanges. People learn from the articulation of interests very different from their own when they speak across social cleavages, class differences and geographical boundaries. The possibility of such dialogues is an important part of the opportunity to behave more like ideal citizens. Given adequately sized samples, random sampling ensures (with extremely high probability) a microcosm of the interests that need to be articulated – and responded to – in any serious deliberation on policy issues. In the first British Deliberative Poll, on crime, the small group discussions revealed that the sample contained criminals – just as any sizable random sample would do. And the criminals enriched the dialogue. They had actually been to prison. They had some first-hand (if anecdotal) sense of what motivates people to commit crimes, and of what might deter them from doing so. Similarly, when some of the participants in the recent British deliberative poll on the future of the National Health Service suggested that the NHS should no longer invest in trying to save low-birth-weight babies, another participant volunteered that he had been a low-birth-weight baby himself, while another mentioned that he had had two low-birth-weight children who died, despite enormous efforts by the NHS. The deliberative microcosm represented the people who pay, the people who get treated, the people who have to live with the results of any public policy priorities that might be set. In bringing them all together, Deliberative Polling permitted them all to become informed in a dialogue in which their concerns could be answered by others with different points of view and in which they could all come to their own considered judgments.

Conclusions Deliberative Polling embodies a vision of democratic possibilities. Almost uniquely, it combines high degrees of both political equality and deliberation. As compared with other deliberative bodies, it embodies the equal chance of all citizens to be selected and to participate. As compared with conventional polls, it shows opinions that are more considered, less top-of-the-head. It is thus the kind of public voice that merits special attention. If democracy requires democratic consultation about the benefits and burdens borne by the people, why not consult the people under conditions where everyone’s voice counts equally and where they can all learn about, consider and discuss the issues? The application of this Athenian model to the modern large-scale nation-state

The quest for deliberative democracy 27 can illustrate the possibilities for improved democratic consultation, influence the public dialogue on politics and policy, and shed light on the deliberative capacities of mass publics. It is no panacea, and we are only beginning to understand its practical merits and limitations. But we offer it as both a contribution to democratic decision-making and a research programme, as a means both of increasing democratic deliberation and of determining its consequences.

Notes 1 An earlier version of this paper appeared in The Good Society, Vol. 9, no. 1, Sept. 1999, pp. 1–14 and is reproduced by kind permission. The Deliberative Polling Project and our thinking about it owe much to our collaborations with Roger Jowell, Norman Bradburn, Ken Rasinski, Alison Park, Dennis Thomas, and Will Guild, among others, while the statistical analyses to which we allude have benefited from the assistance of Christopher Bratcher, Han Dorussen, Dennis Plane, Jeffrey Ladewig, and Nedim Ogelman. Fishkin has filed a trademark on the term ‘Deliberative Polling’, and the Center for Deliberative Polling at the University of Texas at Austin, of which Fishkin is Director and Luskin Research Director, is receiving fees from the trademark to fund research. 2 Political processes may be said to embody political equality when the decision process gives every citizen an equal chance of casting the decisive vote. 3 By ‘serious consideration’ we mean at minimum that there should be: (a) sincere and open-minded participation in a process in which arguments are offered and evaluated on their merits; (b) a level of completeness with which arguments voiced by proponents of a position are responded to by those with rival views; (c) a degree of attentiveness and mutual respect toward the arguments and concerns of other participants; and (d) a degree of accuracy in the factual claims made. All these criteria are continua, of course, but any deliberative process should exhibit them to at least some reasonable degree. 4 For some discussion of the limits of formal political equality see Fishkin (1991: 29–34). 5 Interestingly, these are the two countries that have probably developed the most extensive efforts at direct public consultation, in the form of numerous elections and referenda. 6 The same points apply to polls, even well conducted ones. Not everyone drawn into the sample gets interviewed (or responds to given questions), and those who are interviewed (or do respond) have very unequally considered opinions. True, polls based on probability sampling serve equality far better than do town meetings, viewer call-ins, and other manifestations of mass opinion based on self-selection or other sorts of nonprobability sampling. True, the best academic surveys have response rates exceeding virtually all turnout rates (although these are not usually the surveys politicians see). But much inequality remains. 7 Since not everything that gets called ‘random sampling’ is in fact random sampling, it may be worth being more explicit: at bottom, random sampling involves the equivalent of numbering every potential sample member and choosing the sample by drawing correspondingly numbered cards from a perfectly shuffled deck containing one card for every potential sample member. ‘Stratification’ or ‘clustering’ may add complications, but that is the basic idea. The investigator surrenders all control over the selection of sample members to a random number generator. Choosing haphazardly from potential sample members who are handy or present themselves is not the same thing. The surrender of control is incomplete. Deliberately trying to select a representative sample, containing about the right proportions of men and women, young and old, etc., is still less random. That, in its more systematic versions, is what is known as ‘quota sampling.’ Only true random sampling carries assurances of unbiasedness (in the sense described in note 9) and justifies statements about ‘margins of error’ or ‘statistical significance.’

28

James S. Fishkin and Robert C. Luskin

8 Consensus conferences have typically selected panels of 10 to 14 participants from volunteers responding to newspaper advertisements (see Grundahl 1995, esp. p. 33). Planning cells in Switzerland have employed nominations by interest groups (see Dienel and Renn 1995: 133). 9 Regardless of sample size, the use of random sampling is enough to guarantee that, over hypothetically repeated sampling from the same population, the mean of the sample mean (of any characteristic whatever) would be the corresponding population mean. If, for example, the population mean age is 43.2 years, one sample may have a mean age of 40.2 years, another a mean of 45.7 years. The sample mean will never exactly equal the population mean. More occasionally, a sample may even have a mean that is much further off, say of 52.8 or 30.6. But, given random sampling, the average of the sample means, over infinitely repeated sampling, is exactly 43.2. The sample mean, given random sampling, is an unbiased estimator of the population mean. The problem with small samples is that the sample mean varies widely around the population mean, so that, in any given sample, the sample mean can easily fall very far from the corresponding population mean. 10 Ackerman (1991, section 1) has described ‘constitutional moments’, during which he believes the American public has seriously deliberated. Although we find it quite plausible that the degree of public deliberation varies, present-day survey evidence from around the world would lead us to wonder how large any increase in deliberation across the whole public could be. 11 The terms come from Popkin (1991). For a similar position see Lupia and McCubbins (1998). 12 For an overview of this whole issue, see Luskin (2000). 13 The centrepiece of Converse’s (1964) seminal discussion of ‘non-attitudes’ or phantom opinions was a question about the government’s role in electric power. 14 After a time, the Public Utilities Commission amended its rules to require informed public participation with representative samples in an ‘interactive’ process. 15 For more on these various Deliberative Polls and their results, see Fishkin (1997); Fishkin and Luskin (1999); Luskin, Fishkin, and Jowell (1996); Luskin and Fishkin (1998); Luskin, Fishkin, Jowell, and Park (1999); and Luskin, Fishkin, and Plane (1999). 16 Environmental Defense Fund, press release 27 February 1997. 17 Deregulation has now brought a halt to this particular series of Deliberative Polls. 18 This is not to say that there have been no statistically-significant differences between participants and ‘non-participants’ (those interviewees who did not attend the deliberative weekend). There have been some, and some of them have been substantively non-trivial. But none has been truly large, nor any larger than might be expected of sample-population differences in ordinary media polls.

2

Toward deliberative institutions1 Graham Smith

Deliberative democracy is fast establishing itself as a new orthodoxy in contemporary democratic theory – as the comments in this volume by friends and critics alike make clear.2 Whilst advocates have addressed questions of the constitutional rights and principles that are necessary for the emergence of a deliberative polity, few (apart from Fishkin and Luskin, previous chapter) have engaged in the detailed task of institutional design. Accounts of deliberative democracy and the analysis of institutional arrangements can take a number of forms. Elsewhere in this book, the focus is on the nature of civil society (Rättilä) and supra-national structures (Eriksen). This chapter initially considers the growing interest in innovative democratic experiments, such as deliberative opinion polls, citizens’ juries and consensus conferences, before briefly discussing some of the tensions that may emerge in attempting to institutionalize democratic deliberation in other institutional settings. Experiments with ‘citizen forums’3 share a number of features: a cross-section of the population is brought together for three to four days to discuss an issue of public concern; citizens are exposed to a variety of information and hear a wide range of views from witnesses who they are able to cross-examine; and the fairness of the proceedings is entrusted to an independent facilitating organization. There is an assumption that ordinary citizens are both ‘willing and able to take important decisions in the public interest’ (Coote and Mattinson 1997: 4). Fishkin and Luskin provide an outline of the deliberative opinion poll process in Chapter 1. Since 1994, they have run polls in a variety of contexts in both the US and the UK. In conjunction with television and other media services, deliberative opinion polls have tackled public issues such as national candidate assessments, crime and health provision. More recently, public utilities in the State of Texas have begun using deliberative opinion polls to fulfil their requirement for public participation in resource planning (Fishkin 1997). Citizens’ juries have been run and promoted since the 1970s in both Germany (where they are known as planning cells) and the United States (Stewart et al. 1994; Dienel and Renn 1995; Crosby 1995).4 It is in Germany that they have had most political impact, with a range of government bodies and agencies commissioning planning cells on a number of different policy issues, such as planning, energy and transport policy, and agreeing to take into account their recommendations in future decisions.

30

Graham Smith

Recently a number of citizens’ juries have been run in the UK, usually in conjunction with health authorities and local government – two institutions which are frequently criticized for failing meaningfully to involve their local populations in the decision-making process. Consensus conferences have been run regularly since the 1980s by the Danish Board of Technology as a means of incorporating the perspectives of the lay public in the assessment of new scientific and technological developments that raise serious social and ethical concerns. Experiments with consensus conferences have also been conducted in The Netherlands and the UK, although without the level of media and public interest or political impact observed in Denmark ( Joss and Durant 1995). In almost all cases, advocates of these innovations argue they should not be seen as an alternative to existing representative structures, but rather as a complement to them. At one and the same time such mechanisms are seen as providing decisionmakers with the informed views of citizens and responding to the perceived democratic deficit in contemporary society – the growing distance between the lives, experiences and attitudes of citizens and the decisions made in their name. Although all three models share many features, most significantly the desire to create the conditions under which democratic deliberation can occur among citizens, there are some important differences. First, there are variations in the number of citizens who participate. Each deliberative opinion poll has involved between 250 and 600 citizens. In comparison, a citizens’ jury or consensus conference will only involve between 12 and 25 citizens.5 Second, to select citizens deliberative opinion polls and citizens’ juries use some form of random sampling procedure. However, because of the relatively small size of most citizens’ juries, citizens are often selected using stratified random sampling to ensure that different demographic, and at times attitudinal, criteria are fulfilled. The sheer size of a deliberative opinion poll means that stratification is unnecessary.6 Consensus conferences differ in that volunteers are recruited through advertisements and make written applications from which the panel is selected on the basis of sociodemographic criteria. Thus the first stage of the procedure is self-selecting. However there is an element of self-selection in both citizens’ juries and deliberative opinion polling, given that citizens who are randomly selected still have a choice whether to accept the invitation to attend. The final important distinction rests with the outcome of the different models. With juries and consensus conferences, citizens come to collective decisions after a period of deliberation and provide a series of recommendations as a group. By comparison, at the end of the deliberative opinion poll process the individual views of citizens are recorded. As the name suggests, there is a post-deliberation opinion poll. We shall return to some of the implications of these differences between the models as the analysis progresses.

Are citizen forums deliberative institutions? Deliberative opinion polling was specifically developed by Fishkin in light of his commitment to the ideal of deliberative democracy (Fishkin 1991, 1997) and

Toward deliberative institutions 31 connections have begun to be made between deliberative democratic theory and other innovations such as citizens’ juries and consensus conferences (Smith and Wales 1999, 2000; Renn et al. 1995). How can we judge the extent to which these institutional designs embody deliberative democratic ideals? These citizen forums are briefly assessed below against four criteria: inclusiveness, the quality of deliberation, citizenship, and legitimacy. These criteria have been chosen to highlight different aspects of deliberative democratic theory and are as applicable to the analysis of micro-experiments, such as citizen forums, as they are to largerscale manifestations of deliberation.7 Inclusiveness Ideally, deliberative democratic institutions would entitle each citizen to participate in decision-making processes and have an equal right to put forward reasons, to express and challenge needs, values and interests (Cohen 1989; Benhabib 1996a). Voices should not be excluded; parties have an equal right to be heard. It is difficult to imagine an institutional design that was able to embody fully such an ideal and quite clearly the vast majority of a population is excluded from direct participation in citizen forums. However deliberative opinion polling, citizens’ juries and consensus conferences all attempt to approximate the ideal of inclusiveness by aiming for a broadly representative jury selection which is able to draw on a wide range of experiences and backgrounds. Quite simply, the larger the number of citizens randomly selected, the more statistically representative they are of the socio-demographic characteristics of the wider population – as such deliberative opinion polling and the German method of running a number of planning cells are more likely to approach the ideal of statistical representation. Unlike smaller juries and consensus conferences, it is unnecessary to use stratified selection processes since with large numbers of citizens, no voice or perspective is systematically excluded.8 However, even though not systematically excluded, small minorities may still not achieve presence or voice and mechanisms may need to be introduced to ensure their inclusion in citizen forums. Deliberation For deliberative democrats, dialogue is deliberative to the extent that ‘interactions are egalitarian, uncoerced, competent, and free from delusion, deception, power and strategy’ (Dryzek 1990b: 202). Deliberative institutions aim to embody a form of political dialogue which promotes mutual understanding, where actors are ‘motivated to resolve conflicts by argument rather than other means’ (Warren 1995: 181). Citizen forums are all carefully designed to create the conditions within which democratic deliberation can emerge. Citizens are exposed to a wide range of information, have the opportunity to question witnesses and are able to reflect on the experiences and perspectives of fellow citizens with varying backgrounds. Even in larger-scale deliberative opinion polls, much of the discussion

32

Graham Smith

between citizens occurs in small groups – the small size of groups and the time that citizens spend together help to create an environment in which stable expectations and relations of trust can be fostered between participants (Fishkin 1991: 92–3; Stewart et al. 1994: 10). Citizens in juries and consensus conferences will often break down into even smaller groups to ensure that everyone has the opportunity to be involved in discussions. Suspicion of strategic action on the part of citizens is lessened given that these forums do not ‘represent an opportunity for advancement, promotion or election’ for those involved (Dienel 1996: 114). Further, much of the deliberations take place away from the public gaze. The absence of an audience means that citizens are not as subject to pressures to stick rigidly to predeliberative positions, but are more likely to reflect on different viewpoints in light of new evidence and reasoned arguments (Elster 1998b: 109). Facilitators or moderators have an important role to play in both ensuring inclusiveness and encouraging an ethos of mutual respect during deliberations. They are alert to the manner in which witnesses might try to manipulate the citizens when offering evidence or answering questions and also to the potential domination of confident and outspoken citizens (Davies and Sang 1997). Citizenship Deliberative democratic theory stresses the importance of an active conception of citizenship. As Warren argues, ‘democracy works poorly when individuals hold preferences and make judgements in isolation from one another, as they often do in today’s liberal democracies. When individuals lack the opportunities, incentives, and necessities to test, articulate, defend, and ultimately act on their judgements, they will also be lacking in empathy for others, poor in information, and unlikely to have the critical skills necessary to articulate, defend, and revise their views’ (Warren 1996: 242). Citizen forums embody an active conception of citizenship, one in which the experiences and judgements of citizens are brought into the public domain and where citizens are encouraged to develop a disposition of mutual respect and understanding toward other citizens. Again though, such an opportunity is directly experienced only by the relatively few citizens who are involved in the forums. This issue will be taken up briefly in the next section. Important evidence is emerging from these experiments which shows that citizens take their role seriously and are willing to reflect on different evidence and experiences. In the jury process in the UK, for example, one of the first things that the citizens typically do is spend time drawing up ‘rules of conduct’ which emphasize the need to respect others and to listen to their arguments (Smith and Wales 1999: 303). Evidence is emerging that, over the period of the forums, citizens become better informed and many of their preferences and judgements change (Stewart et al. 1994; Fishkin 1997; Mayer et al. 1995). There is also some indication that citizens are more civically minded and active well after the process is ended (McIver 1997; Coote and Lenaghan 1997; Dienel 1989). Empirical backing is therefore emerging for the theoretical

Toward deliberative institutions 33 claim made for the transformative and educative power of democratic deliberation. Legitimacy For deliberative democrats, the legitimacy of decisions will be increased if they are made after a period of deliberation (Manin 1987; Gutmann 1996). Sceptics have understandably contended that deliberative institutions attached to the state will be systematically undermined or subverted by decision makers, ‘undertaken with co-optation of potential troublemakers in mind, or as a veneer for decisions reached independently by conventional political means’ (Dryzek 1992: 34). Advocates of citizen forums are extremely sensitive to such accusations of bias and manipulation. There is a recognition that it is in the initial selection of the issue to be addressed and the choice of relevant information and witnesses – before any citizen deliberations – that the potential for agenda setting and the mobilization of bias is highest. The independence of the facilitating organization is fundamental here9 and organizers are fastidious in their attempt to draw together a range of stakeholders to help select relevant questions and evidence. In the consensus conference model, the citizen panel is brought together for preparatory weekends where they have the opportunity to be involved in the selection of expert witnesses and key questions. Again, in the citizens’ jury process, citizens are typically given the opportunity to call new witnesses as they deliberate and learn about the issues under consideration. There are also concerns that commissioning bodies ‘cherry-pick’ only the favourable recommendations and judgements from citizen forums. This problem is somewhat ameliorated in citizens’ juries in Germany and the United Kingdom where a pre-jury contract is drawn up between the independent facilitating organization, the commissioning body and the jurors requiring the commissioning body to either act on the jury recommendations or to give reasons why it has decided not to act.10 Finally, the limited number of citizens directly involved in these forums means that awareness and discussion of a forum’s deliberations amongst the wider population is an important element of establishing democratic legitimacy. Forums need to be part of on-going public deliberations and opinion-formation. Equally, wider public recognition acts as a potential counter to the manipulation of recommendations by political elites. Danish consensus conferences are afforded a high political and media profile, particularly because they are held in the parliament building and on topical issues of public concern. They appear to play a significant role in the public debate on controversial scientific and technical developments and have at times had important political repercussions (Klüver 1995: 44). Similarly, deliberative opinion polls have often achieved relatively high levels of media attention. Publicity is an important virtue of democratic institutions (Luban 1996) and the media are typically given access to plenary sessions and to the final report or findings. This provides an important link between the deliberations in the forums and wider public opinion.

34

Graham Smith

The wider project of deliberative design Clearly then, there are features of deliberative opinion polls, citizens’ juries and consensus conferences that are appealing to deliberative democrats. However, if these innovations are indeed examples of deliberative institutions, there are aspects of their structure which may raise questions about the wider project of deliberative design. In particular, these forums highlight a tension between representation and deliberation; the potential ‘fragility’ of deliberation in larger forums; and the role of decision rules in deliberative design. What lessons can be learned from these experiments for the development of other deliberative institutions? Are there limits to the institutionalization of democratic deliberation? Deliberation and representation One of the strengths of these various citizen forums lies in creating the conditions under which citizens are able to reflect on a range of evidence and perspectives and to alter their judgements accordingly. Even though some advocates argue that they realize a ‘microcosm’ model of representation (Fishkin 1997; Crosby 1995) – they ‘mirror’ the wider population’s values and attitudes – it is important that citizens are not seen as representing ‘people like them’ in any strong sense. No group of citizens can accurately mirror all the standpoints and views present within the wider community and there is a danger of creating false essentialisms – citizens who share similar socio-demographic characteristics do not necessarily share the same views and attitudes. For example, women in these forums should not be seen as ‘representing’ all women in the wider community. An emphasis on representation may undermine the ideal of active citizenship and the emergence of democratic deliberation. The primary task of citizens within these innovative forums should be understood in terms of deliberation rather than representation (Abramson 1994: 141). As Abramson argues with regard to legal juries: We do not want to encourage jurors to see themselves as irreconcilably divided by race, selected only to fill a particular racial or gender slot on the jury. Yet we do want to encourage jurors to draw upon and combine their individual experiences and group backgrounds in the joint search for the most reliable and accurate verdict. The difference is subtle but real. (Abramson 1994: 11) If indeed there is a tension between forms of representation and democratic deliberation, what does it mean for the wider project of deliberative design? Within contemporary political theory there is much interest in questions of recognition, which, as Squires and Meier in their different ways show in this volume, often manifests itself in calls for group representation for the politically marginalized. The tension between representation and deliberation haunts the work of advocates of group representation, such as Young and Phillips, who are also

Toward deliberative institutions 35 committed to forms of deliberative democracy (Young 1990, 1996; Phillips 1995). It may be an onerous task in designing democratic institutions to achieve both the goal of institutionalizing group representation for unique and often marginalized social groups, and the goal of deliberative democracy, which would require representatives to engage in a process of deliberation where perspectives may be transformed (Rosenblum 1998: 343–6). Both difference theorists and deliberative democrats wish to ensure voice and presence for all perspectives, especially those of the politically disenfranchised, but would group representatives have the freedom or desire to engage deliberatively? Representing the perspective and interests of a social group and being able or even willing to enter into a deliberative process may be in dynamic tension. The same question cuts across the models of associative democracy that are examined in detail in the final section of the book. Cohen and Rogers, for example, are explicit that their conception of an enhanced role for associations embraces deliberation (Cohen and Rogers 1995a). However, as Mansbridge rightly suggests, deliberation needs to be institutionalized not only between elites but also between elites and rank and file, and among the rank and file within associations; otherwise ‘a narrowly self-interested citizenry will eventually throw out its public-spirited representatives’ (Mansbridge 1995: 143–4). Relatively few associations even begin to approach such an internal institutionalization of deliberation, thus potentially compromising the deliberative potential of associational engagement in decisionmaking processes. There is no question that the institutionalization of group representation and associational engagement provides certain goods in democratic polities, whether in the form of the developmental and self-transformative impact on individuals, contributions to the public sphere, giving voice to the marginalized, or affecting decision-making processes (Rosenblum 1998; Warren 1998). But this is not to escape the fact that institutional design will have to acknowledge and respond to possible tensions between forms and practices of representation, and the ideal of democratic deliberation. Is small beautiful? The experiments with citizen forums place a high premium on creating the institutional conditions for deliberation to occur. Engagement between citizens and between citizens and witnesses is carefully structured to mitigate strategic manipulation. In particular, deliberation occurs in small groups, away from the wider public gaze and under the guidance of moderators. This may point to the necessity of creating a protective space for democratic dialogue to emerge. What implications might this have for deliberative institutional design? First, can democratic deliberation only occur in face-to-face situations where there are a small number of participants? Advocates of citizen forums certainly highlight this feature of their design (Stewart et al. 1994: 10; Fishkin 1991: 92–3) – mutual understanding and respect is taken as a function of the size of the forum. There is a central role played here by moderators who employ techniques to

36

Graham Smith

encourage deliberation, such as breaking into small group discussions to ensure that the process is not dominated by outspoken and confident individuals. There is evidence that in less structured environments, face-to-face dialogue may subvert deliberative democratic ideals. For example, in her studies of town meetings, Mansbridge has raised concerns that face-to-face interaction may encourage the suppression of conflict (Mansbridge 1983: 276–7). We shall return to the question of whether forums might generate a false consensus below, but it is important to recognize the difference between a town meeting, where the social pressures of the local community come to bear on participants, and a citizen forum where the participants are randomly selected, do not know each other and are not under the same social pressures. What then of other forms of interaction between actors, beyond such small-scale, face-to-face contact? We have already raised concerns about how representation might affect deliberation: does size also place limitations on the ability of legislative chambers to approximate the deliberative ideal, for example? Dynamics of large assemblies tend to differ markedly from the protected sphere of deliberation within citizen forums. So Elster for one argues that debate in larger assemblies tends to be unsystematic and lacks coherence, ‘dominated by a small number of skilled and charismatic speakers . . . who count on rhetoric rather than argument’ (Elster 1998b: 107). This is not to argue that democratic deliberation can only occur in highly structured small-group settings. However, if more large-scale processes of deliberation are to be institutionalized, then thought needs to be given to how deliberation can be ‘protected’ and enhanced, and the form and qualities of mediating structures between citizens, associations and other institutions. This clearly relates to a further important feature of the design of citizen forums – the absence of an audience during periods of deliberation. Again, this may throw up further dilemmas for institutional design. As already mentioned, being away from the public gaze means that citizens are less likely to feel they must represent and adhere to a particular position or perspective. In some ways though, this could be seen as contradicting the deliberative democratic ideal of ‘free and open deliberation’. The need to ‘protect’ the sphere of deliberation from the pressures of the wider community is in tension with the virtue of public openness. A balance is struck whereby many sessions are private, but the media and observers have access to sessions where evidence is presented and witnesses questioned, and at the end of the process when the results are made public. But again, what does this mean for the wider project of deliberative design? The pressure of the public gaze relates back to the problematic relationship between representation and deliberation. Public openness is at one and the same time a defence against strategic manipulation and a potential barrier to participants developing a deliberative disposition. Deliberation, consensus and decision rules Arguably, the most significant difference between deliberative opinion polls on the one side and citizens’ juries and consensus conferences on the other, is the

Toward deliberative institutions 37 ‘outcome’ of the processes: deliberative opinion polls end with individual responses to a range of policy-relevant questions; citizens in juries and consensus conferences come to collective decisions in the form of recommendations. This difference raises interesting questions about decision rules and the role of consensus in deliberative democratic institutions. Consensus is a contested goal for deliberative democrats. Is it a regulative ideal of deliberation (Cohen 1989), or a dangerous ideal in the name of which conflict is suppressed (Young 1996; Mansbridge 1983)? As noted above, in citizens’ juries moderators are particularly alert to the way individuals can dominate the agenda and define a consensus (Davies and Sang 1997). Room is typically made to accommodate majority/minority decisions. Where there is no consensus, space is required so that remaining areas of conflict and disagreement can be expressed and clarified. However, on a few occasions, participants in citizens’ juries have expressed concern that moderators have pushed for consensus at the expense of allowing citizens to work through their differences (Smith and Wales 1999). This pressure may be even more of a problem in consensus conferences. As a project manager from the Danish Board of Technology stresses: ‘every effort is made to attain the greatest consensus between the lay-panel members on the actions to be recommended. Minority opinions should be allowed only when the process reveals very wide differences of opinion’ (Grundahl 1995: 37). There is a fine line between the search for consensus and the suppression of conflict. Much of this is avoided in deliberative opinion polls, given that the outcome of polls is the individuals’ post-deliberation opinions on a range of questions. Only after the poll are these preferences aggregated and statistically manipulated. In consensus conferences and citizens’ juries, concerns are raised about the process of how recommendations are generated, but their strength is that the citizens themselves are collectively engaged in this process and are able to debate and reflect on how their differing opinions and values should be balanced in any recommendations. With deliberative polls, however, it is a third party that aggregates the individual preferences.11 Where citizens’ juries and consensus conferences tell us something about how citizens might balance competing opinions and values, and provide us with clear recommendations, in deliberative opinion polls that job is left to an analyst. Citizens in deliberative opinion polls are less likely to feel any pressure toward consensus; however they are not given the opportunity to exercise the type of political judgement required in coming to collective political decisions. This raises questions about the use of decision rules in deliberative democratic institutions. In some ways, the very idea of a deliberative decision-making institution exhibits a certain paradoxical quality – tensions may exist between the need for a decision and the institutionalization of deliberation (see the comments of Saward and Budge in later chapters). A decision implies the end of a discursive process. But deliberation is in principle ongoing – excepting universal consensus, there is no obvious end-point to the process of mutual understanding, reasoned dialogue, persuasion and judgement. On the other hand, politics requires decisions: there is a temporal limit to debate. The ‘economy of time’ haunts

38

Graham Smith

political institutions (Beetham 1992b). Here we witness tensions between the goods of mutual understanding and efficiency (Chambers 1995: 241). At some point, excepting consensus, a decision needs to be made and preferences aggregated. We have already raised concerns about the practice of a third party aggregating preferences in deliberative opinion polls, but deliberative democrats should not be opposed to aggregation as such, only to the aggregation of nondeliberative preferences (Habermas 1996c: 304; Miller 1993a). After a process of deliberation, the nature and justification of the majoritarian principle is altered: ‘the procedure preceding the decision is a condition for legitimacy, which is just as necessary as the majority principle. It is the conjunction of these elements that creates legitimacy’ (Manin 1987: 359). Again though, Chambers raises important concerns that the prospect of closure ‘will always place a constraint on the discursive process. The closer and more final is that point of closure, the more participants will be motivated to act strategically rather than discursively. . . . A realistic model of deliberative democracy must concede that decision rules in large democracies will always place constraints on constraint-free dialogue’ (Chambers 1995: 255). The introduction of any decision rule may well have adverse affects on the quality of democratic deliberation within institutions.

Where now for deliberative democracy? Where does this brief foray into the institutionalization of democratic deliberation leave deliberative democracy? At one extreme, the tensions and problems discussed could be seen as indicative of the paradoxical nature of any attempt to develop deliberative spaces within the institutions of the state: they will always be manipulated and subverted by political elites (Dryzek 1992).12 At the other extreme, one could imagine a radical vision in which citizen forums provide the model for decision-taking bodies; the idea of a ‘jury democracy’, for example, bears some resemblance to Burnheim’s vision of ‘demarchy’, where positions of authority are decided through random sampling procedures (Burnheim 1985).13 More realistically, what the discussion of these experiments highlights is that they could (and perhaps should) have a role to play in contemporary democracies – that an increased use of such forums could both offer citizens a meaningful way of participating in policy-making processes and a way of increasing the democratic legitimacy of decisions. Additionally, although the preceding discussion raises a number of possible barriers to the further institutionalization of democratic deliberation, it does not close the door on the design or emergence of other forms of deliberative spaces within and between institutions. Democratic theorists need to think creatively about the interplay between different forms of democratic institutional design. This analysis also points to a recognition that although deliberation is an essential element of democratic legitimacy, it is but one among many goods in a democratic polity. It must be balanced and judged against other ideals and goods, such as group representation, social justice and efficiency.

Toward deliberative institutions 39

Notes 1 This chapter draws on an analysis of citizens’ juries developed with Corinne Wales (Smith and Wales, 1999; 2000). The author would like to thank Susan Stephenson, Michael Saward and the participants at the excellent Mannheim ECPR workshop on Innovation in Democratic Theory for their comments and suggestions on earlier versions of the chapter. 2 For a useful typology of different approaches within deliberative democratic theory, see Blaug (1996). 3 The term ‘citizen forum’ will be used in this chapter, since there is no recognized term that embraces this type of institutional innovation. 4 For further references to citizens’ juries in Germany, the United States and the United Kingdom, see Smith and Wales (1999). 5 In Germany, Dienel (who developed the idea of planning cells) often runs a number of planning cells concurrently and/or in series. To date, the largest project involved 500 citizens from all over Germany. 6 Similarly when planning cells are run concurrently or in series, thus increasing the number of citizens involved, Dienel simply uses random sampling. 7 For an alternative set of criteria for judging deliberative institutions based on the ideals of ‘competence’ and ‘fairness’, see Renn et al. (1995). A more developed defence of inclusiveness, deliberation and citizenship as criteria of analysis can be found in Smith and Wales (2000). 8 There is evidence that the citizens selected are, according to certain socio-demographic characteristics, statistically representative of the communities from which they are drawn. The typical practice in consensus conferences of advertising for volunteers and then requiring citizens to send in application letters could well act as a barrier to some elements of the population. It is interesting that the citizen panel for the most recent UK consensus conference on radioactive waste management was chosen using a stratified random selection process. See www.ukceed.org. 9 The practice of some local authorities in the UK of running their own citizens’ juries on controversial public policy issues, thus avoiding the costs of engaging an independent facilitator, raises questions about the legitimacy of any recommendations. 10 This practice was originally developed by Dienel (Stewart et al. 1994: 47). 11 Dienel faces a similar problem of collating results when running a series of planning cells on the same issue. 12 On this view, incipient deliberative designs can only emerge in the public sphere, independent from, and in confrontation with, state power. 13 However, Burnheim’s conception of ‘statistical representation’ differs from that of citizen forums in that the individuals chosen within a representative sample are taken to be representatives of particular interests, thus introducing tensions between representation and deliberation. Such a vision also leaves a number of questions unanswered, including issues of democratic accountability and efficiency.

3

Deliberation as public use of reason – or, what public? whose reason? Tiina Rättilä

It is curious that the theory of deliberative democracy which relies so much on the beneficial effects of public deliberation (e.g. see Manin 1987; Elster 1998a; Fearon 1998) should pay so little attention to the concept of the public itself. At first sight this omission may not look like a pressing problem. After all, deliberation is sometimes referred to as collective conversation whereby the participants or citizens reason together publicly about their common norms, institutions and policies. Why would this clear-from-the-start commitment of deliberative politics to public openness not suffice? At a closer look, however, it becomes clear that unless deliberative democracy is able to provide a well-grounded theory of the public that defines and generates conditions for participatory politics, it runs the risk of being adjudged just another elitist model that rationalizes politics away from – what to theorists may look like – the unsophisticated political commentary of ordinary citizens over their selfinterested needs (see Saward, this volume). Deliberative democracy might still produce epistemically superior, wonderfully reasoned collective decisions (see Estlund 1997), but it could not unproblematically justify itself as a viable alternative to competing participatory models (such as republicanism or associative democracy), as critics have repeatedly pointed out (Sanders 1997; Young 1996; Fraser 1992). Deliberative democracy is not a coherent theoretical whole, however, and some theorists have more to say about the public and its potential than others. Broadly speaking, we can distinguish two orientations. The first, and perhaps the dominant one, sees deliberation mainly as a rule-governed process aimed at making binding collective decisions (Cohen 1989; Elster 1998b; Fearon 1998; Mackie 1998). This orientation makes two simple, in its own understanding empirical, assumptions about the role of public openness within such processes. One is that it has a ‘civilizing’ effect on the ways in which participants reason and conduct themselves when appearing in, or before, the public. According to the second, ‘public opinion’ acts as a constraining factor in political decision-making that otherwise is ruled by sectoral and group-based interests. This approach tends to see the public as an ‘audience’; that is, as individual receivers and consumers of public communication who influence political decision-making mainly (or solely) through aggregated, calculated and polled

Deliberation as public use of reason 41 opinions. Consequently, the proponents of such a view are not much concerned with how citizens could be empowered to act as active creators of and contributors to public discourses and the political agenda. The second orientation, which we could call the civil society approach, pays explicit attention to the public sphere as a locus for wide and effective citizen participation (Habermas 1989, 1992, 1994 and 1996c; Benhabib 1992b and 1994). It typically makes three assumptions about the desirable characteristics of a public political sphere: It should be (a) autonomous, independent of the influences of coercive institutions and strategic action (Habermas 1996c: 444); (b) inclusive, allowing access to all concerned (or ‘affected’) citizens (Benhabib 1994); and (c) rational, organizing its communicative processes as rational–critical discourse under conditions of ‘egalitarian reciprocity’. As formulated by Benhabib (1992b: 89), the latter include that each participant must have an equal chance to initiate and continue communication; that each has an equal chance to make assertions, recommendations, and explanations; that each has equal chances to express her wishes, desires, and feelings; and that each must be free to thematize those power relations that in non-discursive contexts would constrain the free articulation of opinions and positions. In this essay I shall be concerned with the second approach, focusing especially on the models of the deliberative public developed by Seyla Benhabib and Jürgen Habermas. I shall argue that while their contributions are important, even vital, to the credibility of the theory of deliberative democracy, their assumptions and conceptualizations are problematic (although I am going to suggest that Habermas’s recent ideas offer a theoretical advance on Benhabib’s model – or, for that matter, on his own original theory). All three assumptions (about autonomy, inclusion, and rationality) are questioned and qualified in the course of the discussion. My claim is that the kind of discursive rationality deliberative theorists entrust with public discourse cannot easily comply with conditions for equal and inclusive participation. The question that deliberative democrats must address seriously is this: are the implications for democratic participation of a rationalized discourse ones that the proponents of the theory are willing to live with – even if these evoke an embarrassing flavour of (what may be seen as) outdated elitism? Or, should deliberative democrats reconsider their commitments to rationality to allow for more inclusive and efficacious participation – even if this means that the epistemic aspirations and the proceduralist norms of the model need be relaxed? I shall try to put the case for the latter.

I Let me begin with a short review of Benhabib’s theory and those points I find important and convincing in her argument. When delineating the basic characteristics and commitments of the deliberative public, Benhabib draws distinctions between it and the liberal and the republican models (1992b; see also Habermas 1994). What she criticizes in the latter two is that both place unnecessary restrictions on the substance of political discourse

42

Tiina Rättilä

(later it will be argued that in her own model Benhabib ends up reproducing similar constraints). Benhabib makes two criticisms of the liberal (or ‘legalistic’) understanding of the appropriate scope and uses of public reason. First, liberals seem to assume that in pluralistic societies participating actors already know prior to public deliberations what their disagreements are (going to be) about, and, accordingly, are able to avoid taking up disputed questions on the political agenda. The problem with this solution (‘the method of avoidance’ as Rawls calls it in Political Liberalism, 1993) to the ‘fact of pluralism’ is that the distinction between issues of justice and those of the good life cannot be decided in advance by some moral geometry. Rather, as Benhabib points out, it is often the process of unconstrained public dialogue itself that helps define and politicize the nature and implications of the issues that are being debated. A second limitation of the liberal model is that it often conceives of political relations too narrowly along juridical ones, positing the principle of neutrality as the defining criterion according to which the scope and goals of political action should be judged. However, while neutrality certainly is one of the core prerequisites of the modern legal system, politics and democratic struggles are about something else. All struggles against oppression in the modern world, Benhabib argues, have begun by redefining what had previously been considered private, non-public, and non-political issues as matters of public concern, as issues of justice, and as sites of power that need discursive legitimation (1992b: 84; see also Habermas’s critique 1996c: 308–14). In this respect, the women’s movement, the peace movement, ecology movements, and new ethnic-identity movements follow a similar logic. Though this is a good point, I will later highlight the difficulties Benhabib has in explaining democratic struggles within her own model. Following a different line of thinking about politics and the public Arendt, too, comes to place restrictions on what public action should be about. In seeking to secure an autonomous sphere for performative and agonistic politics that resists the repressive power of the ‘social’ and the modern disciplinary institutions, Arendt (1958) is tempted to exclude from political attention all those phenomena that help reproduce such ‘constatives’ (see Honig 1992). Consequently, questions related to needs, identity, body, reproductive work, economics and technology, are not the proper stuff of political action in Arendtian terms. Here a forceful objection to Arendt’s account is that in modern capitalistic societies such constriction of the proper domain of political action is, in effect, undemocratic and unjust. Benhabib points out that different action types, such as work and labour, can become the locus of public space and action when reflexively challenged and placed into question from the standpoint of the asymmetrical power relations governing them (1992b: 80). Defining the public agenda in advance is futile also in the sense that, even in Arendt’s own terms, the empowering effect of ‘acting together’ is to put new and unexpected items on the political agenda. In this other (‘associational’) conceptualization of the public, Arendt develops something like a procedural concept of the public, Benhabib argues. Here, what is important is not so much what public discourse is about as the way in which it

Deliberation as public use of reason 43 takes place: through power-free reasoning between free and equal citizens about their common good (1992b: 80–1). Therefore, from the standpoint of the procedural public, neither the distinction between the social and the political nor the distinctions among work, labour, and action are notably relevant (1992b: 80–1). The discourse model of the public sphere, Benhabib contends, is able to avoid such limitations on public reason, as well as to suggest an alternative political solution to the problem of pluralism, because it starts with a different kind of understanding of politics and rationality. Following closely the position of Habermas (see Habermas 1995 and 1996c: 322–8), Benhabib, like many other deliberative theorists, assumes that the basically egalitarian structure of communication aimed at reaching understanding and consensus can be translated into a democratic method (that is, to norms of practical discourse) by which the members of the society can govern themselves rationally. In this view democracy is taken to be an open discursive process (Kulynych 1997), the rules and outcome(s) of which are never predetermined (Benhabib 1994: 33). Moreover, because outcomes are not predetermined, and there are no received norms and values that due to some traditional power constellations privilege some citizens and groups over some others (Habermas 1994: 4), there is a good chance that all citizens can accept the deliberative model as the most justified one, and endorse its procedural norms. As long as they do so, citizens can enter the public domain, each with his or her own ‘doctrines’, preferences and interests. Benhabib states that the advantage of viewing political participation as discursive will formation is that public dialogue need not be judged by how ‘neutral’ participants’ arguments are, but by the procedural criteria of practical discourse. Public spaces come into existence whenever and wherever citizens affected by social and political norms engage in discourse, governed by the norms of egalitarian reciprocity, to evaluate their validity. In effect, there may be as many publics as there are controversial general debates about the validity of norms (1992b: 87). However, Benhabib criticizes Habermas for making a number of distinctions that are not compatible with the principle of publicity as democratic participation (1992b: 88). For instance, the discourse model of ethics is said to be about norms as opposed to values, about generalizable interests as opposed to culturally interpreted needs, about questions of justice as opposed to questions of the good life, and so on. Again Benhabib objects to such limitations: ‘If the agenda of the conversation is radically open, if participants can bring any and all matters under critical scrutiny and reflexive questioning, then there is no way to predefine the nature of the issues discussed as being ones of justice or of the good life itself prior to the conversation’ (1992b: 89) (cf. Michelman 1997). So far so good. However, my concern is that Benhabib’s commitment to the conditions of discursive rationality, especially to the expectation that deliberation should transform preferences and bring about generalizable standpoints, may turn against her best participatory and emancipatory intentions. This is the focus of the next section.

44

Tiina Rättilä

II Deliberative democrats typically want two things of public deliberations: that participants commit themselves to unconstrained give-and-take argumentation where all can freely express their views, opinions and preferences; and, second, that the ‘logic’ of public dialogue in the course of the deliberative process transforms initial self-regarding preferences into generalizable, other-regarding positions (Miller 1993a; Warren 1992). The assumption is that articulating views in public encourages interlocutors to avoid narrow, self-regarding interests. Therefore, public deliberation has a ‘civilizing’ effect (see Elster 1998b). Benhabib states: [t]he very procedure of articulating a view in public imposes a certain reflexivity on individual preferences and opinions. When presenting their point of view and position to others, individuals must support them by articulating good reasons in a public context to their co-deliberators. This process of articulating good reasons in public forces the individual to think of what would count as a good reason for all others involved. One is thus forced to think from the standpoint of all involved for whose agreement one is ‘wooing’. (Benhabib 1994: 33) In fact, this assumption is taken a step further to postulate that the whole rationale of deliberation is bound to its transformative function. Equitable communication is good as such, but not enough to make deliberation rational. Rationality, on the other hand, is essential for deliberative theorists because it is coupled with legitimacy – where the latter is understood to refer to something like ‘agreement by all to reasons arrived at in a deliberative process’. This bind between the two is tight. Why? To justify itself, the deliberative model needs to show that it improves legitimacy in some significant way compared to competing models such as, say, liberalism. This is accomplished by defining legitimacy in the aforementioned way, which implies that deliberation is a sine non qua condition of legitimacy. This reasoning would, of course, be self-referential without some mediating variable. So, here rationality steps in as a bridging principle between the two, and the equation is ready. Deliberation is desirable because it enhances the rationality of decision-making processes, and rationality – defined through the transformative function – improves democratic legitimacy by producing norms and policies ‘agreed to by all’ in reasoned deliberations by free and equal citizens. Now, the crucial question we can pose to deliberative democrats is: could these features of deliberation, along with the expectation of preference transformation, compromise in actual discourses the terms of egalitarian reciprocity; that is, the possibility of all participants to ‘initiate and to continue communication, to make assertions, recommendations, and explanations, to express their wishes, desires, and feelings, and to freely thematize those power relations that in non-discursive contexts would constrain the free articulation of opinions and positions’? There

Deliberation as public use of reason 45 are at least two arguments we can make about the possible negative effects of rationality defined this way. First, Benhabib assumes that reasons evoked in public discourse should be articulated and substantiated so they can be accepted by ‘all’. But ‘all’ are bound, under modern conditions of pluralism, to hold conflicting beliefs, opinions, and interests. We can therefore assume that a commitment to finding shared reasons implies that interlocutors should not articulate and express particularized experiences, identities, and differences (especially ones that are controversial in the community; see Charney 1998: 102). If this is so, Benhabib might contradict herself in promoting, on the one hand, the right of all social actors freely to voice their concerns in public, and requiring, on the other, that public arguments be based on generalizable reasons. Even if Benhabib is able to hold on to the formal conditions of egalitarian reciprocity, the logic of discourse may still restrict the range of arguments that can be put forward and defended in public. If Benhabib wants to avoid reproducing liberal constraints on public discourse, she should state more clearly how these two conditions (unconstrained communication and transformation of preferences) are compatible. My second concern is related to the first. Consider Benhabib’s conjecture that her model of the deliberative public allows different kinds of communicative inputs into the dialogue. When discussing Rawls’s account of public reason, she notes: While there is little doubt that the Rawlsian principle of public reason expresses a governing limit for the coercive power and public accountability of the major institutions of a liberal-democratic society, consider also what is missing from it. All contestatory, rhetorical, affective, impassioned elements of public discourse, with all their excesses and virtues, are absent from this view. Public reason is not freely wielded public reasoning, with all the infuriating ideological and rhetorical mess that this may involve (1994: 37) This position raises the question that has especially concerned feminist theorists: how does communicative freedom fit with the reason-articulating function of deliberation?1 And in case it does not, which one has to go? Benhabib would like to think that it fits. Remember, however, the commitment of the discourse model to finding and expressing reasons acceptable for all: ‘[n]obody can convince others in public of her point of view without being able to state why, what appears good, plausible, just and expedient can also be considered so from the standpoint of all involved. Reasoning from the standpoint of all involved not only forces a certain coherence upon one’s own views, it also forces one to adopt a standpoint which, Hannah Arendt, following Kant, had called the “enlarged mentality” ’ (Benhabib 1994: 33; see also Benhabib 1996a). It is quite plausible to think that this prerequisite may discourage attempts to bring ‘contestatory, affective and infuriating sentiments’ into deliberative discourse. Or, if communication really is unhindered and all kinds of sentiments can be expressed, there is no guarantee

46

Tiina Rättilä

that initial preferences will be transformed into common standpoints. Note in this context the views of Schudson, who argues that ‘what makes conversation democratic is not free, equal, and spontaneous expression but equal access to the floor, equal participation in setting the ground rules for discussion, and a set of ground rules designed to encourage pertinent speaking, attentive listening, appropriate simplifications, and widely apportioned speaking rights. The primal American political conversation was carefully structured so that it could be genuinely deliberative [sic!] and genuinely democratic. It thus had to be anything but spontaneous [sic!]’ (Schudson 1997: 307). We can now return to Benhabib’s earlier argument about the failure of liberals to account for the emergence of democratic struggles. Her assumption there was that the latter can be better conceptualized and explained through the model of the deliberative public – which posits, as we noted, that public spaces can emerge whenever ‘affected’ citizens come together to engage in open and inclusive discourse about the validity of common norms and policies. In the light of the critical points presented above, this assumption now needs to be qualified. Namely, Benhabib cannot explain convincingly the rise of not-yet-public issues into the public political agenda through the assumptions of the discourse model. Let me illustrate this difficulty by referring again to the experience of women. Benhabib has suggested that feminist critiques of the exclusionary implications of the liberal public sphere can, in fact, benefit from the conditions and rules provided by the Habermasian discourse model to create public space(s) for critically reflecting on and transforming gender-biased politics (1992b). But can they really, or how far can they? Fraser has made the interesting case that because communicative action is defined procedurally, it is particularly unsuited for addressing issues of speech content (cited in Kulynych 1997: 325). Consequently, in most deliberative accounts the relationship between procedure and content, which is at the core of feminist deconstructions of language and discourse, goes unnoticed and unthematized. Kulynych illuminates this point by noting that ‘[a] procedural approach can require that we accommodate all utterances and that we not marginalize speaking subjects. It cannot require that we take seriously or be convinced by the statements of such interlocutors. In other words, a procedural approach does not address the cultural context that makes some statements convincing and others not’ (1997: 325). It might even be argued, then, that the achievements of women’s movements may not have come about as a result of arguing within the discourse model but as a result of these movements having successfully posed and thematized their concerns outside it, through, for instance, ‘performative’ acts and demonstrations that have exposed the limits of the dominant rationalized discourse with its normalizing and universalized images of ‘women’s issues’ (see Kulynych 1997: 336–41). Recently, many writers have called attention to the problems and limits inhering in deliberative communication that is thought to privilege disembodied, dispassionate, formal, and general speech over other communicative styles (Fleming 1995; Young 1996; Kulynych 1997). A critical reader may, however,

Deliberation as public use of reason 47 wonder why there should be such debate about ‘communicative styles’ when it is a simple fact that in democratic talk we have to back our opinions with reasons; without the force of reason democracy would become meaningless. If safeguarding legal and political impartiality requires that some communicative freedom be lost, we should simply let it go, our critic might conclude. Without denying the value of reason-giving as a core principle of liberal justice, we can still object to the argument in at least two ways. For example, we could refer (simply but effectively) to freedom of expression as at least as influential a principle of liberal constitutionalism. Deliberative democrats accept this, of course, but would add that the two are not incompatible. On the contrary, this is exactly what deliberative theorists postulate: that ‘reason’ emerges from free expression. However, in developing their schemes for enhancing the rationality of decision-making processes, deliberative theorists sometimes seem to forget that freedom of expression cannot be compromised very far, not even in the context of promoting rational and effective decision-making. Failing to take seriously the implications of this principle and its egalitarian spirit for democratic practices provides a powerful weapon for critics. The second objection raises this question: can deliberative participation address real social problems effectively? Feminist and postmodern critics have their doubts. Kulynych argues that as long as we define the purpose of participation in terms of influence, privacy, legitimacy, and self-development (i.e. through traditional categories), we will be unable to see how political action can be effective in the contemporary world (1997: 318). Deliberative processes are not helpful in exposing and addressing issues of power because it does not necessarily, or typically, appear as visible relations and hierarchies. Rather, it works through the ‘disciplinary technologies’ within the practices of everyday life that produce new objects and subjects of knowledge, incite and channel desires, generate and focus individual and group energies, and establish bodily norms and techniques (Sawicki 1991, cited in Kulynych 1997: 319). In this context, civil society and the public sphere cannot be thought of as autonomous. What citizens are up against here are not just, or even mainly, such external forces as coercive institutions (e.g. the state apparatus) or self-interested politicians, but power mechanisms that permeate the whole structure of their life-contexts. Under such conditions, it is not necessarily meaningful, and certainly not effective, to act with respect to the state – instead, citizens need to find innovative ways of resisting the productive power of, say, consumerism, the media, and technology. These are the kind of struggles that the deliberative theory is badly equipped to address. If the discursive model of democracy and the public is constructed upon certain exclusions, and if these revolve around particular conceptions of rationality, the meaning(s) of the latter need be reconsidered. Here we can turn, ironically perhaps, back to Habermas for fruitful suggestions as to how such a task could be pursued. In what follows, I argue that Habermas’s recent work, notably in Between Facts and Norms (1996c), shows how we could address the contradictions in the theory of deliberative democracy.

48

Tiina Rättilä

III Since the publication of The Structural Transformation of the Public Sphere (1989) Habermas has, in the context of his general theory of communicative action, developed a detailed theory of the modern forms of rationality and (re)evaluated ways in which they condition publicity and the political public sphere in contemporary (liberal) society. The central difference between Habermas’s old and new vision is that he currently sees the potential for publicity that is critical and powerfree in the unconstrained and informal communication processes of civil society. In Between Facts and Norms, Habermas presents a thorough conceptual and sociological analysis of the infrastructural features of the public sphere(s), where ‘public’ is understood as a special kind of communication structure rooted in the multiplicity of overlapping communicative and associational networks of society (1996c: 360). With this theoretical shift Habermas (at least partly) moves the focus of political participation from institutionalized politics to those democratic processes that serve as the necessary background for law- and policy-making. Habermas understands the public sphere in terms of a (dialogical) ‘space’, not as an institution, an organization or a system. It is a ‘social phenomenon just as elementary as action, actor, association, or collectivity’ (1996c: 360). In their least complicated form, publics emerge in simple, episodic encounters between social actors. They become more complex and abstract the more these encounters and interactions expand to constitute larger networks of communication within and between associations and other actors of the society. Here, ‘the public sphere can best be described as a network for communicating information and points of view’ (1996c: 361). More specifically, the political public sphere refers to processes of informal opinion- and will-formation uncoupled from decision-making institutions and operating in an inclusive but unstructured network of overlapping subcultural publics. It plays an important role as a mediator between civil society and the administrative power of the procedurally regulated public sphere (the parliament and the judiciary; Habermas 1994: 8). The special ‘function’ of informal deliberation is to generate public discourses that ‘uncover topics of relevance to all of society, interpret values, contribute to the resolution of problems, generate good reasons, and debunk bad ones’ (1992: 452). Informal public opinions act as a political ‘signal’, communicating and thematizing problems to be further processed by the political system. In both Benhabib’s and Habermas’s models we are working with the same three basic ideals: (a) both understand democracy as a discursive process by which citizens can govern themselves rationally and legitimately; (b) both take public discourse to be an open, uncoerced process of communication that allows free change of opinions and arguments; (c) both assume that legitimate norms and decisions can emerge only through a procedure that respects the norms of egalitarian reciprocity; and (d) both share an overall idea(l) about the role of the public in deliberative politics. The assumption is that (discursive) participation through free publics is a necessary background condition for the legitimation of norms and of the political process itself.

Deliberation as public use of reason 49 Despite these similarities, what is interesting here is that Habermas’s recent conceptualizations take his model in a different, more radical direction – leaving Benhabib currently more ‘Habermasian’ than Habermas himself. This claim will be addressed in the next section.

IV As noted above, Habermas (like Arendt) understands the public sphere as a social space that is opened up between social actors when they engage in a dialogue. It is not an institution with established structures, rules, practices, competencies and roles. Rather, it is characterized by its ‘performative’ character: it comes into existence in and through the acts of the interlocutors and, in order to persist, it needs to be reproduced in a sequence of performances (it does not ‘stay there’ automatically after being generated). Now, let me be explicit about the significance of this definition. I would like to claim that it makes a difference whether we conceive of the public as a space or, like Benhabib, as a particular type of political process (even though she occasionally uses the expression ‘space’), or, in the case of the decision-making approach, as an audience. The problems of the latter have already been explained. In Benhabib’s model, the public ‘space’ is quickly transformed into a process that is governed by certain procedural rules and has certain predefined goals. It sets the terms for ‘how things should go’ and how participants should act. It therefore also defines in advance what the right and acceptable kind of politics ‘looks like’, closing off spaces for ‘politics by other means’, so to speak. The clear implication is that non-discursive acts and expressions that take place outside the process are not appropriate political acts, or at least, not ones that need to be taken seriously by the participants in the deliberative discourse. Because of this structured picture of a discursive process, it is difficult to conceptualize and account for those unexpected elements of politics that appear outside rational and ‘normal’ discourses – acts that resist such boundaries in the first place. On the other hand, when we conceive of the public as a contingent space, we open it up to different and unpredictable forms of action – what it will become depends on how the participants themselves utilize the space. We might say, then, that actors create a political public when they enter the space and use the opportunity to voice their own concerns, bring up new problems, criticize norms, and the like. This conceptualization of the public gives rise to an agent-oriented perspective, enabling meaningful, efficacious political action ‘from below’ (see Bang and Dyrberg, this volume). By stressing the ability of all citizens to create publics, and to make things public (by utilizing politically those spaces they enter) – instead of assuming that there is only one ‘big’ public (e.g. constituted by institutions and practices of mass media), which facilitates entry to some and denies it to others – we can encourage and empower citizens to become participants in public discourses and political processes. Habermas is realistic enough to think that public spheres do not guarantee democratic politics. Since public spaces are contingent and unpredictable, nondemocratic acts and consequences are quite possible. To say that public spheres

50

Tiina Rättilä

are open and inclusive is not to say that they are automatically egalitarian, or that the public opinions they generate could produce impartial law. Therefore, Habermas wants to keep the distinction between the free publics of civil society and the democratic (especially law-making) institutions, much as in liberaldemocratic countries constitutions are made difficult to amend. Law-making institutions and procedures represent sorting and testing mechanisms that are expected to transform ‘opinions’ into universalizable reasons. In accord with this distinction, Habermas defines two kinds of deliberative politics: (a) the proceduralized deliberations of decision-making institutions; and (b) informal deliberations as they evolve from the free opinion- and will-formation processes of the free publics. Habermas contends that democracy needs both, addressing explicitly the relations and the mediating processes between the two levels of deliberative politics. This enables him to respond to the question often asked of deliberative models, namely where do they leave the traditional representative politics? Habermas’s solution is to leave it where it is. Another point where Habermas differs from Benhabib (when pushed a bit) concerns the rationality of deliberative processes. In the understanding of most deliberative theorists, including Benhabib, rationality is related to the procedural conditions and the transformational effects of the deliberative process. I have argued that there are problems with this conceptualization. Now Habermas too thinks of informal deliberations as discourses, ‘more or less’ fulfilling the procedural conditions of practical discourse (as unconstrained communication about problems and common norms aimed at arriving at legitimate decisions). But there is a difference between the strict proceduralism of formal deliberations and the relaxed proceduralism of informal deliberations. Habermas’s view of informal deliberation in the political public sphere – free, unregulated, disparate communications by participants in anonymous publics that present, thematize and criticize issues and norms – potentially radicalizes the way rationality is perceived within such discourse. In this perspective rationality is not taken to be a quality of an actor, such as his or her capacity to offer good reasons in public debate. Rather, rationality is a feature of a certain communication structure. The emancipatory potential of this formulation lies in the fact that rationality in this sense cannot be used as an exclusionary principle, or to justify an assumption that some are more capable than others of presenting good, generalizable reasons (see Saward, Chapter 4, and the response by Dryzek, Chapter 5, on issues of deliberation and exclusion). Another potentially empowering implication is that this conception of rationality does not predefine the ways and styles of acting politically. On the contrary, allowing different ways of voicing and ‘performing’ political views, preferences and desires is indispensable for opening political processes to wider participation. Not all citizens can or want to engage in structured processes; many are motivated to keep out of them, though they are not unmotivated politically. This should not be taken to mean that they could not take part, or that they are not ‘rational’. Habermas’s model implies the possibility of extending the idea of democratically meaningful communication to include different kinds of political manifestations,

Deliberation as public use of reason 51 as can be read from this comment: ‘From the perspective of democratic theory, the public sphere must amplify the pressure of problems, that is, not only detect and identify problems but also convincingly and influentially thematize them, furnish them with possible solutions, and dramatize them in such a way that they are taken up and dealt with by parliamentary complexes’ (1996c: 359). As Kulynych sees it, Habermas’s demand that public discourses be both attention-catching and innovative, convincing and dramatizing, requires more than mere rational argumentation. It requires a kind of political action that can effectively disrupt cultural common sense and provide alternative ways of thinking and acting (1997: 327). Habermas’s restated theory of the public sphere opens up new perspectives for critically examining the deliberative model of democracy. He challenges, if unintentionally, deliberative theorists to reconsider their conceptualizations of reason and rationality, as well as the implications of such definitions for democratic participation. However, Habermas still leaves a number of problems for critics to ponder.

V It can be argued that the clear distinction in Habermas’s revised theory between formal decision-making and informal participation results in a built-in tendency to keep ‘high politics’ away from the direct influence of citizens. Apparently this is warranted because the representative mechanisms of the political system satisfy the requirement of democratic inclusion. However, there is much evidence from this book alone of profound concern about the adequacy of traditional styles of democratic representation, from feminist, ecological, associative and direct democratic perspectives at any rate (see respectively the chapters by Squires, Meier, Eckersley, Roßteutscher and Budge). Perhaps Habermas leaves democracy too much as it is? He does not really aim to provide incentives for participatory citizenship. He merely shows how participation ‘normally’ figures in the functions of the public sphere. For instance, Habermas states that ‘the success of public communication is not intrinsically measured by the requirement of inclusion – but by the formal criteria governing how a qualified public opinion comes about’ (1996c: 362). Hence, it seems that the innovative potential of his theory for developing new institutional designs and other opportunities for extending participation is rather limited. Indeed, it seems that the overall argument in the book lacks critical potential. The theory says too little about how different democratic struggles challenge prevailing institutions, norms and relations of power (see the comments of Dryzek in this book, for the reaction of one prominent and radical theorist who has been influenced by Habermas). Habermas does, of course, assume that they do so, but this leaves ambiguous the emancipatory and critical potential of discursive participation. There is a further problem here. Kulynych points out that traditional concerns of political science about unequal resource distribution and its impact on abilities to act are mitigated in Habermas’s broad definition of discursive participation,

52

Tiina Rättilä

which confines citizen participation to an anonymous sphere where only ‘virtual’ presence is required (1997: 321). While we may agree that the ‘virtual’ participatory acts of scattered publics may sometimes influence social opinion, such occasional political successes can hardly make up for the unequal distribution of resources, and the different incentives to participate that stem from them. The question we can pose for Habermas is, why confine problem-solving discourses to institutionalized politics only? Why not extend them to other, concrete and localized contexts, and allow ‘real’ citizens to participate in them? In the final analysis, it seems that Habermas is still too much occupied with system-theoretic conceptualizations – an approach that inevitably blinds analysis to real experiences and meanings on the micro-political level. There sometimes seems to be confusion in Habermas’s account of how much of discourse ethics to preserve, and how much sociological analysis to incorporate in the theory. The problem is that the two will not agree conveniently to meet at an optimal point. I have attempted to press Habermas’s argument to show how his account of the public sphere could be utilized to empower wider participation.

VI It is rather paradoxical that although deliberative democracy lays great stress on the issue of legitimacy, requiring in strong terms that it should ‘result from the free and unconstrained public deliberation of all about matters of common concern’ (Benhabib 1994: 26), theorists rarely ask a number of questions: what does participation actually mean for the participating citizens?; how does it affect their understanding and experience of the political process?; how do they see their own role in the legitimation process?; what is meaningful political action, from their perspective? Citizens are conceived in rather passive, inanimate terms by deliberative theorists. As Sanders points out, ‘democratic citizens as described in these theories seem to live on another planet: they are devoid of race, class, and gender’ (1997: 353). This formalized picture of participation as deliberation fails to grasp participation in its different forms and guises, missing ‘the real, although much more humble, opportunities for citizens to “take part” in their own “governance” ’ (Kulynych 1997: 336). I have tried to show how this constricted view of political action and participation is built into the way theorists typically structure the deliberative process. If deliberative theorists genuinely are motivated to develop deliberative democracy into a form of participatory democracy, they should start by reconsidering and reconceptualizing their understanding of reason, rationality and the public.

Notes 1 Faced with this dilemma, some theorists privilege rational argumentation (e.g. Schudson 1997: 307). Others, notably feminists theorizing difference, are concerned to guarantee participation for all voices, and propose varied forms of communication as models of deliberation (see Fraser 1992; Young 1996; Gould 1996).

4

The European Union’s democratic deficit A deliberative perspective Erik Oddvar Eriksen

Introduction In national settings it is held that the solidarity and trust required for integration to come about are provided for by common sentiments in civic associations and primordial communities. Intergovernmentalists who hold that the European Union (EU) is merely an entity for solving the (given) problems of the Member States are faced with a problem in so far as the EU is becoming a far-reaching polity, in territorial, functional and legal terms. The EU is a puzzle for theories that take preferences as given, or that see integration as dependent upon civic values and commonalities. Further, (neo-) functionalists are faced with explanatory problems, because functional efficiency does not justify outcomes but is itself in need of legitimation. There is thus a need for another perspective on integration, in particular one that makes space for the impact of democratic legitimation. The EU’s striving for openness, the strengthening of the Parliament, the enhanced status of the European Court of Justice, the forthcoming enlargement of the Union, and its stand on human rights indicate that a broader set of assessment criteria are required than the ones built around efficiency and utility. Through measures aimed at redistribution, through regulation of social, environmental and health policies, and through police and judicial co-operation, the EU affects the daily lives of people all over Europe. It impinges upon their lives as customers, as employers and employees, as citizens. These undertakings raise both questions of identity – who are we? – and questions of fairness, the latter because both Member States’ and citizens’ interests are affected profoundly by decision making activity of the EU. Political powers that distribute goods and burdens, imposing sacrifices on some and providing opportunities for others, are in need of justification (Beetham 1992b). In one way or another, a polity claims legitimacy: it claims that the support it obtains is somehow deserved. In a modern post-metaphysical context only a democratic procedure, and the manner in which it incorporates and deals with interests and values, can warrant this claim. It is the procedure itself, and how it fosters democratic processes, that bears the burden of legitimation in a modern state, and this also goes for the EU. Democracy is the only legitimate form of governance! To understand and assess the nature of the EU requires a conception

54

Erik Oddvar Eriksen

of legitimation that extends well beyond that provided by any economic calculus. Conventional theories conceive of the EU as merely a problem solving unit, and of legitimacy in terms of performance and efficiency, i.e. as ‘output-oriented’ legitimacy (Scharpf 1999: 6ff ). The enlargement of the EU’s tasks well beyond the immediate interests of the Member States demonstrates the need for more direct, or ‘input’, legitimacy. I would like to address the prospects for such a concept of legitimacy from a discourse-theoretical deliberative perspective on democracy, because the prevailing view of the democratic deficit of the EU is founded on a restricted set of assessment standards. They are derived either from liberal-aggregative or civic republican concepts of democracy, and it may be argued that the standards should be reset (Majone 1998: 6). In this chapter I seek to show that discourse theory represents a promising theoretical alternative. I will first provide a brief outline of the interpretation of the discourse theoretical perspective that informs the analyses.

Voting or arguing? It will be clear from the introduction and the preceding chapters in this book that the theory of deliberative democracy has received renewed attention in political analysis. This theory maintains that majority rule ‘is never merely majority rule [because] counting of heads compels prior recourse to methods of discussion, consultation and persuasion’ (Dewey 1927: 207). Increasingly, political scientists have paid attention to the way discussion is required to facilitate collective decision-making. The voting mechanism can not stand alone, but presupposes discussion as an additional and supportive device, to foster decision-making. Empirically speaking, there is a great deal of communication going on in decisionmaking, as claims and proposals require verbalization and justification. Logically speaking, there is no assurance that aggregation of exogenous preferences will produce robust results, as was shown by Arrow (1951) and Riker (1982). Even though preferences may be rational and transitive the resulting social rankings are fundamentally arbitrary; majority decisions do not represent ‘the will of the people’ (Shapiro 1996: 34). This goes for majoritarian as well as proportional procedures. The chances are ‘that we get either less or more than the one and only “will of the people,” or its collective preference’ (Offe 1997: 92). Aggregation of preferences is therefore not enough to legitimate political decisions. Pure voting results command little respect unless they are supported by reasons that tell us in whose interest they are cast or to what kind of good they contribute. However, the point about deliberative democracy is not just that discussion is needed in order to reach binding decisions. No model of democracy can do without discussion. It is an efficient means to represent preferences and to compensate for asymmetric information, and so contribute to better decisions (Elster 1998b; Johnson 1993; Cohen 1997b; Fearon 1998). The claim on behalf of deliberative democracy is stronger than pointing to the way discussion is essential for pooling private information in order to form a consistent opinion and to make rational

The European Union’s democratic deficit

55

means–end calculations. The claim is that public communication is needed in order to legitimize outcomes vis-à-vis the citizenry. Thus, democracy is the only way to reach collective decisions in a justifiable way. The governors may be dismissed by the governed at regular intervals and in between are liable to public criticisms. Democracy is a (or the) way to secure right results – and is, thus, a legitimation principle. Only deliberation can get political results right as it entails the act of justifying the results to the people who are bound by them, something which is secured by decision-making governed by deontological norms that in themselves command respect – democratic procedures, including public deliberation and voting, and human rights. In a political context it is required that people ground their beliefs and justify their claims. Justification may take different forms, but from a discourse-theoretical perspective a certain structure of argumentation or reason-giving is required for legitimacy to be achieved. In order to justify claims within a political setting an actor has to compare his claims and beliefs to those of others, and enlarge his position so that it becomes compatible with things that others say, or hold to be proper or correct. He or she has to speak from the position of the generalized other, i.e. from a neutral and disinterested perspective in which all interests and values are taken into consideration (cf. Mead 1962). The concept of communicative rationality sheds light on speech acts which involve actors attempting to achieve mutual understanding. By arguing in relation to intersubjective, ideal standards of rightness, participants can reach an agreement and a base for judging reasonable choices. Parties try to talk themselves into consensus by mutually respecting justified norms and validity claims (Habermas 1984: 392). Even though appeals to norms in real discussions are liable to deception, the fact that parties at least tend to be hypocrites – they have to pay homage to norms in order to achieve agreement – testifies to the importance of norms (Elster 1998b: 110). However, it also demonstrates, I believe, that strategic communication is parasitic on authentic communication, because falseness or nonsense is logically secondary to truth and sincerity. What is more, in this approach deliberation is not dependent upon pre-existing virtues or normative presuppositions – such as ‘reasonable individuals committed to deliberation’ – as is the point in the variants of deliberationist theory that Michael Saward addresses in the following chapter. In the discourse-theoretical version of deliberative democracy it is contended that only by adhering to the formal rules of argumentation, and due to the logic of performative contradictions, the participants may reach common understanding and consensus on normative questions. Common interests and civic virtues are not the necessary inputs in the process but the outcome of collective deliberations properly conducted. Arguing, then, not voting or bargaining, is the currency of democracy. Further, in politics the social fabric is made up of parties that give and respond to reasons. An actor is not able to act in politics unless he or she is able to drum up some sort of support. It takes co-ordination through communication to be able to act because in one way or another agreements have to be reached (alliances have to

56

Erik Oddvar Eriksen

be formed, collective priorities set) – which takes sincere communication. Some members of the polity have to agree on some goals or projects at some time in order to get things moving at all. This is not to say that prevailing disagreement – due to hostility, conflicts, and dissimilar preferences over outcomes – does not necessitate decision-making devices such as voting, instruction and bargaining. The point is rather that non-deliberative procedures presuppose well conducted processes of argument, not least because they must be justified – argued for – in order to be employed ( Johnson 1998; Gutmann and Thompson 1996). This is also the reason why referendums cannot be the basic principle of democracy (see the chapters by Saward and Budge below, which argue in favour of direct democracy). They themselves depend upon prior decisions built on processes of argument. Discourse theory does not neglect or disapprove of voting. The point is rather that voting depends upon fair deliberative processes in order to produce robust results. How, then, to understand the modern constitutional state?

Private and public autonomy The modern western state is premised on the rights of the individual. Modern societies are pluralistic and conflict-ridden but their constitutions make it possible for different groups and subgroups to live together under a common law. This is so because contending parties have to resolve their conflicts through the medium of law. Rights offer individuals protections and entitlements. Politics and law are the official means of solving problems and resolving conflicts in modern societies. Citizenship and nationality – demos and ethnos – are conceptually decoupled in the modern state and the state is seen as neutral regarding different conceptions of the good life. The discourse theoretical concept of deliberative democracy sits very well with supranationalism because it decouples citizenship and nationhood, and conceives of the constitution as a system for accommodating difference. Modern states allow for cultural diversity, as there is a right to non-participation. Thanks to the legal structure and to the modern idea of citizenship, which means to rule over one’s equals and to be ruled in turn by one’s equals, the democratic constitutional state makes solidarity between strangers possible. Actually, they are not nation states – Volksnationen – founded on an ethical Sittlichkeit or cultural substance, but rather nations of citizens integrated by common laws and legal procedures. The democratic constitutional state (Rechtstaat) does not – in principle – need another basis than the recognition of and trust in the very procedures that ensure participation in collective goal-formation and that make peaceful conflict resolution possible (Habermas 1996b: 189): The democratic constitutional state institutionalizes the conditions necessary for integration through self-legislation. This practice is anchored in the medium of law, for it simultaneously secures the private autonomy of the individual by certain protective rights, and secures the public autonomy of the individual by a right to participation. Human rights and democracy secure the private and the public autonomy of the citizens respectively. However, individual rights themselves have to be subject to deliberation by

The European Union’s democratic deficit

57

citizens in order to be rendered valid. The democratic process, which itself is legally constituted, is the source of legitimation. Democracy is conceived of as a set of argumentative or communicative presuppositions and procedural conditions. Applied to the political institutions of modern societies the following reading of the principle of popular sovereignty emerges: ‘Only the principles of the guaranteed autonomy of the public spheres and competition between different political parties, together with the parliamentary principle, exhaust the content of the principle of popular sovereignty’ (Habermas 1996c: 171).

The notion of the public sphere The public sphere provides the type of deliberative arrangement appropriate to discourse theory. It is a common space for free, unrestricted communication secured by legal rights to freedom of expression and assembly, where problems are discovered, thematized and dramatized, fostering opinion-formation and the formation of ‘wills’ that are to be acted upon by decision-making agencies (see the discussion in the previous chapter). The pessimistic view of democracy in the EU maintains that because there is a lack of collective identity, the prospect for a viable European public sphere is rather bleak (Scharpf 1994: 220). There is little agreement on common interests or values, and different languages and national cultures make opinion-formation and coherent action unlikely. According to Scharpf, there are ‘triple deficits . . . the lack of a pre-existing sense of collective identity, the lack of Europe-wide policy discourses, and the lack of Europe-wide institutional infrastructure that could assure the political accountability of office holders to a European constituency’ (Scharpf 1999: 187). There is a civic-republican or communitarian string to this critique, which conceptualizes politics as people reasoning together about the common good. This process of reasoning is seen as something quite different from the discussion of private concerns, i.e. it appears as if a common will prevails from the outset. This view presupposes a homogeneous culture and a united people deliberating and deciding in public on issues of common concern. It portrays the public sphere as distinct and stable, a place where enlightened and equal citizens can assemble to discuss public matters. This is the resurrected concept of res publica handed down from the Greeks, where citizens meet to deliberate in the Agora before decisions are reached in the Equalise. The Greek model of the public sphere which (for example) Hannah Arendt (1958) deploys presupposes a homogenous political community (Benhabib 1992b: 90ff ). A volonté genérale – and a collective subject, such as a nation or class, capable of action – is possible because citizens are equal and share common values and notions of the public interest. In this model there is no distinction between deliberation and decision-making, between opinion-formation and will-formation. Consequently, this conceptualization does not capture the way the modern public sphere is institutionalized in opposition to government, rendered possible by the fact that citizens have rights that can be used against the state.

58

Erik Oddvar Eriksen

There are many public spheres in modern states and they are not confined to national borders. There are subaltern counterpublics and there are overarching publics transcending limitations of time and space, made possible by new media technologies and audio-visual spaces. There are local publics, regional, national and international publics, and general publics; intermediate and semi- and quasipublics (Habermas 1996c: 374ff; Cohen and Arato 1992; Fraser 1992). Although the structure of the media, commercial interests, and the evolution of new forms of communication make this a complex topic, it is fair to say that the more publics, the more debate, the more democracy. Fewer voices are excluded and more questions are asked; more publics provide more possibilities for testing the legitimacy of power, enabling criticism of hegemonic truths, and forcing decision makers to provide more general or universalistic justifications. Central to the discourse theoretical notion of the public sphere is a distinction between opinion-formation, which is the domain for the public sphere, and willformation, which is the domain for decision-making units within the political system. Publics do not act, for they possess no decision-making agency; the public may be seen to act in revolutionary moments, but generally in public spheres it is only possible to deliberate and form opinions about what should be done. In pluralistic and complex societies public opinion is ‘anonymous’ – dispersed into the network of communication itself, with no power to govern. Thus, Habermas (1996c) maintains that popular sovereignty has to be located in the interplay between institutionalized and non-institutionalized bodies for deliberation and decision-making. On this view, the public sphere in Europe is not totally missing. There are new European audio-visual spaces – newspapers (Financial Times, the Economist), television (Euro-News, BBC World, DW, CNN), internet, the use of English as a ‘first language’ – and cross-border social movements and identity politics (Schlesinger and Kevin 2000) that are conducive to a European public discourse. The public sphere should not be seen as existing prior to or independent of decision-making agencies but as emerging in opposition to them – as a vehicle to test the legitimacy of legal provisions and as a counterweight to governmental power. This view of the emergence of the public sphere is based on the contention that the state originated, more or less, through war or brute force (Tilly 1975: 42). Only subsequently was state authority democratized, i.e. subjected to the rule of law. First came the state, then came democracy. Collective identity has to be made rather than merely discovered. It is from this assertion that the contention ‘no European demos without a European democracy’ is derived. As noted, it is the interplay between free and open debate in non-institutionalized publics and institutionalized debates in the political system which together secure the presumption of rational opinion- and will-formation. Popular sovereignty is to be secured by a ‘two-track’ model. According to Habermas, democratic procedures constitute a context of justification, because they provide the reference point for decision-making and negotiation that make clear which norms and goals are to be realized.

The European Union’s democratic deficit

59

Processes of legitimation Institutionalized discourses in representative bodies are necessary not only in order to narrow down alternatives for decision-making and voting, but also in order to reach more enlightened opinions about the public interest or the common good. They improve decision-making. Representation contributes to refining and enlarging opinions by making them the focus of attention for chosen members of the demos. The representatives are placed in larger contexts in which the interests of many constituencies must be taken into consideration: structurally they are put in a position to make more reflective choices. As James Madison contended, representation contributes to political rationality by lifting elected members of the community out of local, parochial settings. In national or federal settings the representatives have to take into consideration different interests and perspectives in order to justify particular claims; thereby, they can reach more reasonable and legitimate decisions. Representation may be seen as a precondition for political rationality because it secures institutional fora in which elected members of constituencies peacefully and co-operatively can seek alternatives and solve problems and resolve conflicts on a broader basis (Sunstein 1988a; Weigård 1995). From the point of view of discourse theory, however, to be legitimate both the principles of representation and the outcome of institutionalized deliberations must undergo scrutiny in an open rational debate in which all citizens are free to participate. One may also call upon such devices as consensus conferences, citizens’ juries, and deliberative opinion polls to foster public deliberation on topical issues (see the discussions by Fishkin and Luskin and Smith earlier in this book). Even if there is some credibility to the thesis of the non-existence of a European public sphere, there is no lack of representative bodies in the EU. There are bodies for representing Member State citizens (the European Parliament), Member States (Council of Ministers, European Commission), and regions (Committee of the Regions). Regarding the notion of democratic deficit, the argument now takes on a different character. It is often maintained that the deliberations in these bodies adhere to the logic of power rather than to the logic of argument; self-interest rather than the common interest dominates. It is the large Member States that have the most votes, and it is economic-functional interests, with considerable lobbying and bargaining resources, whose voices are most easily heard. Decisions are reached by experts rather than elected representatives, and laws are passed with little debate, transparency or publicity – it is an elite game (Middlemas 1995: 612). The representative system and the institutionalized discourses do not live up to democratic ideals. The EU system on the whole is, on this reading, occupied by the special interests of big business and the ideology of free markets (Lodge 1989; Traxler and Schmitter 1995; Andersen and Burns 1996). On another reading, however, the EU already possesses a constitution and laws that establish legal governance structures. The political process is legally constrained and the European Court of Justice, which has not formally adopted a

60

Erik Oddvar Eriksen

European Bill of Rights, claims to be guided by constitutional rights, international conventions and in particular the European Convention on Human Rights (Bellamy and Castiglione 1998: 171). Small countries and rural interests are also systematically overcompensated in the voting formula of the Council of Ministers, while unanimity is required on a whole range of issues, in effect giving Member States veto power, and thus putting them on an equal footing (akin to the principle of equality of states). Veto-power is the main barrier to supranationalism, but on the other hand it represents a constraint on discourse that induces communicative rationality: when parties can block outcomes, actors have an incentive to convince others by employing ‘rational’ arguments. They cannot rest content with convincing a mere majority; because they must reach a consensus they are compelled to make arguments convincing to all. This is not merely to stipulate positive effects of vetoes; it amounts to a proposition that can be tested. Do decision-making bodies in which voting is an option fare better with regard to reaching broadly acceptable solutions, than bodies in which a consensus has to be reached? Moreover, vetoes or qualified majority rules may protect against professionals and lobbying. Joseph Weiler (1991) holds that the veto of each Member State is ‘the single most legitimating element’ of the integration process in the EU. The emphasis placed on consensus in the EU is interesting also because representatives and delegates from Member States are brought together in many kinds of deliberating fora and arenas. They meet in committees, in boards and councils, in order to resolve conflicts and find solutions to common problems in a co-operative manner. This interaction is governed by legal norms and procedures, which means that parties have to raise claims to correctness even if they are subjectively following their own interests (cf. Alexy 1989: 219ff ). Regulatory acts establish missions and objectives for the bodies, and legal rules both enable and constrain action of their members in order to secure fair and reasonable results. There are for instance procedures for consultation, co-operation, and co-decision-making, and the objectives of the Commission are to identify the European interest, to consult as widely as is necessary and to respect the principle of subsidiarity. The Treaty of Amsterdam strengthened the role of that procedure, which requires the greatest amount of deliberation and reason-giving, namely co-decision.1 The heightened role of the European Parliament in the EU’s decision-making structure has led to more inter-institutional deliberation. The Parliament’s decision-making powers have been increased and its scope of action widened, making it a co-legislator with the Council on thirty-seven different types of issues. Co-decision will increasingly be perceived as the standard procedure in legislative matters. Consultation or co-operation are likely to be considered more as exceptions (Nentwich and Weale 1998: 8). There is considerable onus on consentbased legitimation in the very nature of EU decision-making, in that virtually all its decision-making procedures are based on extensive inter-institutional deliberation. The type of inter-state interaction generally referred to as intergovernmentalism may be described as strategic, and as conflict resolution through bargaining (Moravcsik 1998; Scharpf 1999). The question of whether the EU is

The European Union’s democratic deficit

61

something more – representing a deeper form of integration2 – can be addressed by examining the ability of Member States to act in concert, to achieve consensus on conflicting issues and to form a common will in the face of common problems. If the EU is to be something more than an arrangement for inter-state co-operation, it has to be able to act on a collective basis: Member States must have an open mandate from their constituencies in order to be able to identify genuinely common interests. Forming a common will is conditional on the ability to act communicatively rationally. Actors with different interests or preferences will need to put agreement-seeking ahead of self-interest maximization. The outcomes of bargaining between strategic alliances or of majority voting reflect not a common will but merely the will of the strongest or of the many. Such results do not have the quality needed to regulate interests, because they command little respect in a moral sense. What is more, they are also held to be politically difficult in a scheme of voluntary co-operation such as the EU where a common identity is lacking. Why should the winners be willing to compensate the losers? Why should the minority accept sacrifices imposed by the collectivity? For such benevolent actions to come about, for a collectivity to be able to employ majority voting as a means to solving problems and resolving conflict, an established legitimate political order and prior bounds of trust and solidarity are needed (Offe 1998; Lord 1998). The question of whether or not the EU is a supranational democratic polity can fruitfully be examined from the point of view of interaction modes. In short, is interaction made up of Member States competing for the biggest share, or is it also constituted by some real notion of common interest and shared identity? In order to examine the quality of the interaction process in representative and collegial bodies, the conceptual strategy of realism or the methodology of economics is insufficient – because non-egoistic commitment or alteration of preferences is needed to bring about integration. The achievement of solidarity and trust may be seen to hinge on the possibility of communicative action, which, as mentioned, denotes the practice of establishing mutual understanding. My contention is that the nature of co-operation may be of a different kind. Some institutions embody rules and procedures that facilitate communicative action while others foster strategic interaction. In any research programme there is a need for criteria to distinguish strategic from argumentative processes.

Bargaining or arguing? If the critique of the public sphere is informed by the evaluative scheme of civic republicanism or communitarianism, the critique of institutionalized negotiation is informed by the evaluative scheme of rational choice or the economic theory of democracy, seeing democracy primarily as a question of fair aggregation of preferences, and negotiations as strategic interaction (Moravcsik 1998). Bargaining is a kind of speech act, but the objective for each of the parties is to obtain a maximum outcome for their own interests, and to obtain this end they have to be able to persuade their counterparts that, if necessary, they have the resources

62

Erik Oddvar Eriksen

needed to force their interests through (Elster 1992: 15). In bargaining processes parties employ speech acts strategically in order to achieve results. They may even misrepresent their preferences and deceive one another by paying homage to virtue and social norms (Elster 1998b). The EU system is marked by consensus rather than by conflict in decisionmaking processes; majority voting is not often used in collective decisions (Laffan 1996). A simple distinction between unanimity and voting will not suffice to arrive at a conclusion about the quality of interaction. Agreements may, even when parties have veto power, be of a different kind. It may be a compromise, which is the typical (positive) outcome of bargaining. That usually means none of the parties gets exactly what he or she wants, but each regards the result as better than no outcome at all. How much the various actors have to deviate from their opening position depends on their bargaining power. Parties will have different arguments for consenting. Another possibility is rational consensus, which ‘. . . rests on reasons that convince all parties in the same way . . .’ (Habermas 1996c: 166). Communicative rationality, in the present sense, means trying to establish consensus by using reasons – employing arguments – and in so far as agreements are backed by mutually acceptable arguments one may say that there has been some sort of deliberative process going on. A transition to a higher level of abstraction – discourse or principled arguing – where participants examine what lies in the equal interest of all concerned, is not necessary, I believe, for reaching binding and normatively valid decisions in cases involving political controversy. The existence of formally correct procedures, and of rights and the knowledge of and respect for others’ strong evaluations, is often involved in actual decision-making and sheds light on the fact that decisions come about that are neither compromises nor rational consensuses. In a deliberative process one sees the force of a good argument working to accommodate some points of view and values which make further co-operation possible. The participants do not agree on all matters, or on all premises of a conclusion, but the discussion makes the parties aware of the values at stake and of how the situation may be described from another angle. This means that there are many situations where the force of the better argument influences decision-making without resulting in a qualified consensus. These agreements neither rest upon a pure convergence of interests nor are they negotiated compromises between contending parties. They are communicatively achieved working agreements not void of normative quality, because they are supported by someone’s good arguments (Eriksen 1999). Democratic legitimacy results from an open, public deliberative process, but the reasons that convince many need not convince all (Manin 1987; Bohman 1996). In addition, the minority may understand and respect the reasons provided by the majority and accept the result because it was reached by a fair process, which among other things gives the losers the possibility of being winners next time. However, as the ultimate test of the legitimacy of collective decisions, the rational consensus unavoidably provides the standard, because unless all the affected parties have consented in a free debate one cannot know whether or not the decision reached

The European Union’s democratic deficit

63

is correct. This mode of communicative interaction resulting in working agreements may shed some light upon the workings of committees in the EU.

Comitology and deliberation Implementation of legislative acts by the Commission is assisted by hundreds of committees of experts from the Member States. Comitology is an EU expression designating this practice. The Council has been rather reluctant to confer implementing powers on the Commission – they have, as Joerges and Neyer put it, not been ‘willing to loosen their intergovernmental grip on the implementation process in favour of supranational institutions. In more constructive terms, the Comitology decision rejects the idea of supranational central implementation machinery headed by the Commission, and thus indirectly forces national governments into a co-operative venture’ (1997: 277). Comitology, an EU invention, may be understood as an institutional response to efficiency and legitimacy requirements. The Commission needs unbiased information and expert knowledge as well as lay opinions (and/or representatives of non-governmental bodies), in addition to loyalty and support from representatives of national governments in order to be able to implement measures effectively. Several hundred committees are active in the implementation of Council decisions and enjoy extensive freedom of discretion. Broad participation and fair decision-making rules enhance legitimacy, but what is peculiar to comitology – contrary to other international committees – is that these committees are centrally involved in making decisions that are binding on domestic governments. Comitology might be seen as a new political order that may repair the democratic deficit, and because Member State and disinterested parties are included in the decision-making process it also contributes to deliberative supranationalism. We then have to ask: what conditions do the undertakings of the committees have to comply with in order to achieve such good outcomes? My proposal is that they have to take on a communicative style if they are to remedy the alleged democratic deficit or if they are to contribute to democratic supranationalism. The problem of bargaining and voting procedures is that they encourage processes of give-and-take, pork-barrelling, log-rolling, etc., that do not change opinions or involve enlarging or refining perspectives; there is no moulding of a common rational will. The most that is to be expected is agreement on what the parties cannot subscribe to. In a way it signals that the discussion has reached deadlock. It also indicates that the parties have accepted an outcome, not because it is optimal but because of the resources and power relations involved. Each participant would ideally like another and better outcome for themselves, but can live with the agreement that has been obtained. From the point of view of the actors it is a suboptimal solution.3 Arguing, by comparison, is marked by the way in which discussion helps to mould preferences and to alter standpoints. In case of conflict at least one of the contenders has to change position in order to establish agreement. If common problems are to be solved, agreement on collective goals is needed, i.e. moulding

64

Erik Oddvar Eriksen

of a common will is required. When strategic rational actors change their views, it is only to strike a better bargain: they are not moved by the force of the better argument, but by the prospects of success. Which logic the deliberations in fact adhere to is an empirical question. Whether they adhere to the logic of bargaining or to the logic of arguing cannot be settled in advance. Joerges and Neyer (1997), however, conclude after having studied the working of food committees that different interests and affected parties are taken into account and thereby actually meet the congruence criteria, i.e. that those who are affected should also be those responsible. They contend that interaction takes on a communicative rather than a bargaining intergovernmental style. Reports from the deliberations of the working groups that COREPER (Committee of Permanent Representatives) and the Council rely on in preparing the deliberations of the Council also point in this direction. ‘Several authors argue that the national civil servants involved in these working group meetings are exposed to a spirit of co-operation and mutual understanding, to an esprit de corps (Sasse et al. 1977; Pag 1987; Wessels 1991; Westlake 1995; Kerremans 1996) . . .’ and, further: ‘Obviously, informal communication is intense in the working groups manned by full-timers. These working groups also seem to enjoy a common leadership of core members. Among these core members the non-state institutional actors clearly are more than primi inter pares. They generally are the hub of the informal communication network. The principal fact emerging from the data is that our respondents appear to have adopted a common attitude to their different partners’ (Beyers and Dierickx 1998). There is a distinct supranationalist dimension to the way the interaction between experts takes place, for they have to pay more attention to expertise, how to reach agreement, and to ‘get the work done’ (Beyers and Dierickx 1997: 454; Egeberg 1999; see further Eriksen and Fossum 2000). Even though the undertakings in committees may adhere to the logic of communicative action, and as such possess a certain normative quality, their democratic legitimacy hinges on whether the principles of representation, along with the results, can withstand critical scrutiny in open public debate. Nevertheless: ‘European committees cannot simply be classified as the agents of a bureaucratic revolution. Rather, with all its sensitivity for the modern complex of risk regulation and for the intricacies of internationalized governance within nonhierarchical and multi-level structures, the committee system may be argued to possess a normative, if underformed, character of its own; or, more precisely, to operate within a novel constitutional framework informed by the notion of “deliberative supranationalism” ’( Joerges and Everson 2000: 164).

Conclusion When we address the EU system from a discourse-theoretical perspective we become aware that democratic legitimacy does not rest solely on representative institutions for aggregating preferences, or on citizens’ participation in decisionmaking. Both informal and formal institutions are needed, and procedures for both deliberation and decision-making are required. The procedures themselves

The European Union’s democratic deficit

65

guarantee only the right to participation, not the quality of discursive processes. No procedure in itself guarantees just results. These can be obtained only if the argumentative process manages to examine critically the quality of the reasons put forward by the contenders. A more positive theoretical programme is possible, for some institutional rules and practices favour one type of behaviour while others favour others. Certain procedural norms and institutional settings are required to ensure that participants take on a communicative and not a strategic attitude. The preceding discussion may help us to understand why the EU survives all the criticism levelled at it, and endures as a governmental structure capable of making binding decisions, despite all its obvious defects. The EU must be seen as something that is constantly evolving, still ‘under construction’. The EU is ‘not just a state of affairs’, but rather ‘a process of integration . . . that requires normative justifications’ (Beetham and Lord 1998b: 32). The system is in constant change and it is the process and how it is conducted, through procedures, formal agreements, public debate, referendums, and so on that makes for integration. Accordingly, I propose we seek the actual legitimacy of the EU in the process of integration itself and the way it gives the parties a due hearing. The EU is labelled post-national because it is an entity sui generis with sufficient independence to require a more direct mode of legitimacy, beyond that conferred upon it by the Member States. ‘What is emerging in Europe is a multi-levelled, highly fragmented system in which policy “develops” but is beyond the firm control of any single authority’ (Pierson and Leibfreid 1995: 433). It is a system which to a large extent is premised on preferences and interests, not as exogenously fixed or set but rather as endogenously shaped through complex patterns of interaction. Preferences are moulded and reinforced in the political process. Hence there is a structurally and procedurally induced possibility of overriding national self-interest, which is supportive of and also requires communicative processes. This also helps to shed light on the normative aspect of the EU puzzle. Because the EU is not a fixed entity, it should not be evaluated on the basis of established notions of democratic legitimacy, which up till now have been derived from entities such as the city and the nation state.

Notes 1 However, in the second and third pillars – in the Common Foreign and Security Policy, and Justice and Police Affairs Co-operation – only the less arduous consultation procedure is employed. 2 Several conceptions have been proposed in an effort to capture the post-national nature of the EU. These include: supranationalism (Weiler 1995), deliberative supranationalism ( Joerges and Neyer 1997), a mixed polity (Bellamy and Castiglione 1997), an imperfect state (Middlemas 1995), Condomino or Consortio (Schmitter 1996), and a quasi-federal ‘regulatory state’ (Majone 1993; 1996). 3 Nevertheless, they may even support it with rational arguments: considering the circumstances, a compromise is in fact ‘a first-best outcome’ unless you waste some of the resources in the process of bargaining. Of course, you could achieve better results if you had more power, more resources, were more talented; but that is wishful thinking. See also Eriksen and Weigård (1997).

5

Less than meets the eye Democratic legitimacy and deliberative theory Michael Saward

Deliberative models have dominated democratic theory for the past ten years. This chapter argues that, in some prominent versions at least, these models are fundamentally flawed. Criticism of the deliberative model is not new; however, few critics challenge the most basic assumptions of that model, preferring to argue on the ground set out by the deliberationists themselves. My aim is to suggest that deliberative theory lacks basic coherence and consistency, and that it contains worrying elitist threads. The major shortcomings that I focus upon are: the models’ claims to legitimacy; their reliance on metaphor; and their exclusiveness. I proceed by examining, in turn, legitimacy, the siting of deliberative democracy, the work of Cohen, and the Schumpeterian problem (especially as it is raised in the work of Offe). First, a word about the context of this critique. My present task is a negative one: elucidating the flaws in deliberative democracy. For my part, the more positive side of the story lies in regarding direct democracy as the best, most defensible aspiration for the democratic future. If we, as democrats, are to take citizenship seriously and treat citizens equally, then democratic legitimacy can only properly be conceived of as voting for the outcomes one prefers in the context of an ‘open society’. Of course, appropriate deliberation is a good thing, useful within a larger framework facilitating direct participation and decision. However, that deliberation is democratically secondary, a (difficult, complex) matter of tweaking more fundamental features of a defensible conception of democracy (see Saward 1998 for an elaboration of this view, and Budge 1996 and in this volume for compatible arguments).

Deliberative democracy and political legitimacy The deliberative model’s proponents – even some relatively critical ones – display a profound confidence in its theoretical and political importance. In my view, this confidence is displayed most clearly in the connections commonly made between deliberation and political legitimacy. To cite some prominent examples: Manin (1987: 351–2) argues that ‘the source of legitimacy [of political decisions] is not the predetermined will of individuals . . . , but rather the process of its formation, that is, deliberation itself ’; Bohman and Rehg (1997a: ix), in their introduction to

Democratic legitimacy and deliberative theory

67

a major volume on this model, state baldly that ‘[b]roadly defined, deliberative democracy refers to the idea that legitimate lawmaking arises from the public deliberation of citizens’; Joshua Cohen, in what may be the most-cited article in the deliberative canon, is even more forthright when he claims that ‘outcomes are democratically legitimate if and only if they could be the object of free and reasoned agreement among equals’ (1989); and Benhabib argues that ‘legitimacy in complex democratic societies must be thought to result from the free and unconstrained public deliberation of all about matters of common concern’ (1996a: 68). These are major, powerful claims. Are they warranted?

Voting and aggregation Even the most convinced deliberationists concede that voting (and some version of majority rule) will play a role in deliberative democracy. Cohen notes – briefly, en passant – that ‘[e]ven under ideal conditions there is no promise that consensual reasons will be forthcoming. If they are not, then deliberation concludes with voting, subject to some form of majority rule’ (1989: 23). The concessions, when they come, are so muted that we could forgive Przeworski his comment that ‘deliberation theorists . . . wish away the vulgar fact that under democracy deliberation ends in voting’ (1998: 141; see also Miller 1993a; Elster 1998a: 14). Regarding voting as a significantly lesser source of political legitimacy than deliberation (of a certain type) is mainstream for deliberationists and highly marginal in more full-blown and influential accounts of political legitimacy. I suggest that its mainstream status is in large part the product of the blinkered, artificially self-contained character of the ‘deliberative model’, which in turn is the product of the terms in which a good deal of influential deliberative theory is produced (see the detailed discussion of the example of the work of Cohen, below). Voting and elections are subsumed by deliberationists under the ‘aggregative’ model of democracy. This aggregative model variously involves: atomistic (liberal) individualism; secret voting on the basis of ‘pre-given’, ‘non-deliberative’ preferences; self-interested voting; and the absence of consideration of the ‘common good’. It is standardly presented as a black-and-white contrast to the deliberative model, with one model unambiguously good, the other bad. However, there is a sleight of hand involved in this. For example, Benhabib (1996a) begins her analysis with ‘the deliberative model of democracy’. Tagging ‘democracy’ to ‘deliberative’ by initial stipulation immediately brackets off liberal, individualistic and majoritarian – ‘aggregative’ – ways of thinking about democracy. It does so in a way that implies that the deliberative model is self-sufficient – that deliberation of a certain sort could indeed be the major ingredient in an ideal but fully functioning and practical democratic system. Whatever the merits of the specific features of the deliberative models they advocate, writers such as Benhabib and Cohen not only under-rate considerably the basic fact that majority votes must decide democratic outcomes where, even after deliberation, views conflict; more broadly, they do not appear to take sufficiently on board the fact that the modern state is inevitably, structurally

68

Michael Saward

hierarchical, secretive, and unequal in the resources it grants to participants in and against its processes. While deliberation may well be an ideal part of democratic processes, it can only be a part; it can be seen neither as the totality of a democratic decision-making process, nor appropriately practised in any and all key institutions in a democratic system. In short, this is not properly a ‘model of democracy’, but rather a desirable ingredient within a larger theory of democracy; an element within a more encompassing, more complex system of democratic structures – many of them ‘aggregative’ – rather than a self-contained substitute for some other, inherently separate model of democracy. The overdrawn contrast between aggregative and deliberative models also lumps together factors that operate to the detriment of voting as a legitimating mechanism in democracies. The key point here is that voting, and preparing to vote, in elections or referendums need not be tarred with the brush of ignorant, isolated individualism. Can we not ‘deliberate’ in private – by reading, listening and thinking; by mulling over desirable candidates or outcomes? Responding to the maximalist claims about legitimacy that many deliberationists make, Fearon asks: ‘does deliberation [have] to involve discussion rather than being a solitary affair? Surely it is possible to deliberate privately, weighing reasons and arguments in a mental dialogue, even if this might not be as consequentially effective as deliberation via discussion?’ (1998: 61). In other words, could not circumstances quite different from, and more realistic than, idealized free, reasoned, uncoerced deliberation among equals in public, be an important contributor to political legitimacy? Indeed, I will suggest that key deliberationists themselves rely in their arguments upon the reasoning capacities of abstract, isolated individuals; implicitly, these writers provide ‘aggregationists’ with much unintended support. So, deliberation-as-legitimacy is mainstream for deliberationists in part because of under-defended assumptions about what can correctly be loaded into an ‘aggregative’ category, and in part due to the linguistic slipperiness involved in positing the deliberative model as a self-sufficient model of democracy.1

The broader view of political legitimacy Deliberation of a certain kind may well contribute to the legitimacy of outcomes; it is not my intention to argue that the deliberationists have nothing useful to say on the matter. But surely a three-dimensional view of what may produce legitimacy must take (much) more into account. The absence of recognition of this basic point in the deliberationists’ writings highlights the curiously hermetic nature of deliberative theory. Without wishing to go into this point too far, let me suggest Beetham’s (1992a) account of ‘the legitimation of power’ as a suitable counterpoint. Beetham argues, in my view convincingly, that political power can be legitimized by (1) its being exercised according to legally valid rules, (2) the grounding of those rules in terms of shared beliefs (about the source of authority and the structuring of the system so it might serve the common good), and (3) its being the product of express consent. Each of these is no doubt vital, but imagine in particular that the third criterion is absent – there is little evidence of

Democratic legitimacy and deliberative theory

69

(genuine, equal, universal) express consent, that is, actual consent by at least a majority of those subject to the laws of a political community. Ultimately, can democratic legitimacy exist absent fair popular voting on the basis of universal adult suffrage, in principle regardless of the extent to which the preferences or interests that inform people’s votes are shaped by deliberative procedures? If deliberationists answer no – if they give the democratic answer – they undermine their grand claims about deliberation producing legitimacy. If they answer yes, then profound questions must be asked about the democratic credentials of ‘deliberative democracy’. This is a vital theme, to which I shall return in a brief discussion of the work of Claus Offe in a later section. There might be a deliberationist response to this challenge, along the following lines: in advanced societies, democracy can and should mean more than the aggregation of preferences, however vital the latter remains as a democratic baseline requirement. Preference aggregation is the first step on the democratic ladder, deliberative forums a later, compatible, desirable development. This view has a reasonable hue about it, but note that it is not one that, to my knowledge, deliberative theorists have articulated. Even then, thin ice abounds. This amounts to an argument that non-democracies, and as yet unconsolidated democracies, must innovate as ‘we’ did before they can innovate as we ought. There are powerful echoes here with the arguments of John Stuart Mill in Considerations on Representative Government (1975), who argued that certain backward peoples were not yet ready for ‘real’ democracy (and even at ‘home’, plural voting should be employed to factor in differential degrees of readiness for active democratic citizenship). Powerful reminders too of the parochial nature of much deliberative theory despite the universal-sounding language of its proponents (in the literature, only Gutmann and Thompson (1996) explicitly acknowledge that they are writing about processes appropriate to one country, the US). In sum, the surprising parochialism, assumed universality and assumed superiority of the ‘deliberative model’ feed into a highly inadequate account of legitimacy – an account that lies at the heart of the deliberationists’ concerns. But that is not all.

Differential legitimacy According to the deliberationists, political legitimacy is (above all) a product of deliberation. But what does it mean to ‘deliberate’? Cohen writes that: When properly conducted . . . democratic politics involves public deliberation focused on the common good, requires some form of manifest equality among citizens, and shapes the identity and interests of citizens in ways that contribute to the formation of a public conception of the common good. (1989: 19) This conception may represent an ‘ideal’, which we can only try to ‘mirror’ as best we can in real procedures. Presumably, ‘fully legitimate’ policy (or candidate

70

Michael Saward

election) would result from an accurate mirroring of the ideal in practice. But how then do we account for less-than-fully legitimate outcomes, which are the product of less-than-ideal deliberative procedures or forums? On the presumption that no deliberative theorist would want to say that a real procedure that is more than a marginal deviation from the ideal results in clearly non-legitimate outcomes, we must be prepared to entertain the notion of differential degrees of legitimacy flowing from differing degrees of approximation of actual procedures to the ideal. But how many degrees, how many categories of semi-legitimacy, must we allow between the near-ideal and the wholly non-deliberative? Who is to decide? The answer must be that it is a subjective matter, akin in some ways to efforts in democratic theory to pin down measures of preference intensity. Given its subjectivity, we can expect reasonable disagreement on the measures (even if we do not demand of them categorical precision; cf. Dryzek, this volume), raising major doubts about basic ideas of democratic legitimacy. And all that is premised upon a single dimension of degrees of differential approximation to the ideal. If, more realistically, we allow for multiple dimensions (as suggested in Cohen’s remarks, quoted above), we find ourselves in even murkier depths. Consider: deliberation involves discussion, but what sort? Among how many people? On what range of issues must it be repeated? Over what time frame? Must opinions change? (How much appropriate deliberation took place in the long, slow-moving voting queues across South Africa in its first democratic elections in 1994?) Each question would have many answers, some ‘better’ and some ‘worse’ from a deliberative perspective. Tracing the potential variants from answers to one question to answers to the others will rapidly result in hundreds if not thousands of possibilities. Thus, on a realistic view, the deliberationist claims about democratic legitimacy descend into unworkable meaninglessness.

Siting deliberative democracy Even there, I have set aside one crucial question: where does or should deliberation occur? Within this volume, we have discussions of deliberation in (among others): especially constructed micro-forums such as ‘deliberative opinion polls’ and citizens’ juries (Fishkin and Luskin, Smith); supra-national committee networks (Eriksen); civil society broadly speaking (Rättilä); and associations (Herreros). For Rawls (1993), ‘public reason’ is best exercised by the Supreme Court. Others, such as Mansbridge (1996: 57) and Benhabib (1996a), evoke a broad ‘public’ sphere of ‘protected enclaves’ or ‘subaltern counterpublics’. In the words of Mansbridge, Interest groups, political parties, and social movements, as well as churches, workplaces, ad hoc political collectives, and consciousness-raising groups, provide different forms of protected enclaves, in which members legitimately consider in their deliberations not only what is good for the whole polity but what is good for themselves individually . . . and for their group. (1996: 57)

Democratic legitimacy and deliberative theory

71

Table 5.1 A typology of deliberative forums Deliberative forum

Formal

Informal

Representative

A Parliament and linked institutions such as Select Committees; deliberative opinion polls linked to referendums or initiatives?

B Deliberative opinion polls which are not statesponsored; citizens’ juries; some ‘focus groups’

Non-representative

C Supreme or High courts with constitutioninterpreting functions; cabinets in appointive systems (e.g. US)

D Associations (statesponsored or otherwise); political parties (statefunded or otherwise, especially in multi-party systems); ‘protected enclaves’; ‘subaltern counterpublics’; ‘discursive designs’

Benhabib (1996a) also reminds us that traditional parliaments must be a key part of the picture. Most interestingly, Cohen goes part way down a similar route. In his earlier comments on deliberative democracy, Cohen cites publicly-funded political parties as the desirable potential forums for deliberative democracy in practice; parties, he contends, can provide the more open-ended arenas needed to form and articulate the conceptions of the common good that provide the focus of political debate in a deliberative democracy. . . . The question is how we can best approximate the deliberative conception. And it is difficult to see how that is possible in the absence of strong parties, supported with public resources. (Cohen 1989: 32) Later, he writes how the use of public power to ‘encourage the development of the right kinds of secondary association’ could facilitate the creation of new ‘deliberative arenas’, or ‘schools for deliberative democracy’ (Cohen 1996: 110–13). The issue of siting raises many fascinating theoretical and institutional issues. I confine myself here to comments that pertain to democratic legitimacy; if deliberation is to make a (the?) major contribution to political legitimacy, then the prominence and inclusiveness of the site(s) for deliberation are clearly vital. First, consider a simple typology of possibilities (see Table 5.1). The formal or informal dimension concerns whether or not the forum has any constitutionally stipulated political function such that the outcomes from an institution’s deliberations become, or must formally be taken into account in the making of, government policy. The representative or non-representative dimension concerns whether or not some plausible claim to be broadly representative of the wider population of the ‘master’ political community is tenable (in either elective or statistical terms).

72

Michael Saward

Among more influential deliberative democrats, for example, Rawls sites deliberation in C, Fishkin primarily in B, and Cohen, Benhabib and Dryzek in D. Those who make the most expansive claims about political legitimacy site deliberative forums in Box D. But institutional designs in Box D are not decisive of the content of government policy and can make no significant claims to be representative of the wider community (unless via some democratically dubious claim of ‘virtual representation’). To focus on Cohen’s views, neither political parties nor secondary associations as sites for deliberation (focused on the public good, etc.) can possibly provide the level and depth of legitimating force the whole thrust of his theory requires. Parties are partial – they represent members’ views, and are animated by sectional interests (one need not be an out-and-out Downsian to accept this). As far as I can see, the only type of political party that could in theory approach the role Cohen asks of it would be the single party in a one-party state; in the words of Julius Nyerere, ‘. . . where there is one party, and that party is identified with the nation as a whole, the foundations of democracy are firmer than they ever can be where you have two or more parties, each representing only a section of the community’ (quoted in Nursey-Bray 1983: 105). Much more should be said on this point, no doubt. However, I shall merely put the question: if deliberation is so vital to legitimacy, why not formalize it and put it on a representative basis? Surely any reasonable claim to legitimacy rests upon such moves. This is not to say that informal, societal deliberation is not a good thing; by and large, I do not see how it could be otherwise. However, despite the crucial nature of deliberation to democracy according to the model’s advocates, the non-state is preferred to the state (Dryzek 1990) and the local and partial to the general (Barber 1984; Cohen 1989 and 1996). This in turn raises the key issue that leads us to the next section.

The undermining metaphor: a critique of Cohen In my view, to put the point too bluntly, the foundations of the ‘deliberative model’ are in fact non-deliberative. I am not referring here to the temporal argument that any deliberative procedure must in practice have non-deliberative origins (see Michelman 1998) – I take that to be self-evident and uncontroversial. Rather, that a certain style of political theorizing – Rawlsian theorizing – provides both the unacknowledgeable architecture and the metaphorical force for the deliberative model. One upshot of this is that the deliberation that is general, inclusive, universal, equal, face-to-face, and the source of political legitimacy is only expressible as such because it is purely hypothetical, and it is so because the Rawlsian ‘original position’ (Rawls 1972) – apparently now an historical curiosity among theoretical devices, and downgraded in significance by Rawls himself (Rawls 1993) – lives on, all the more powerful for its being unsaid. Further, the basis of the definition of the original position in the Rawlsian idea of ‘reflective equilibrium’, and the closely linked notion of the ‘four-stage sequence’, provide further unacknowledged metaphors at the heart of the architecture of the

Democratic legitimacy and deliberative theory

73

deliberative model which (respectively) (a) undermine claims that proper deliberation must be collective, and (b) highlight the intrinsic elitism and dislike of politics that characterize the ‘deliberative model of democracy’. These are large claims, and I now turn to close examination of features of the influential work of Cohen in an effort to make good on them. In his oft-cited essay ‘Deliberation and Democratic Legitimacy’ (1989), Cohen places at the core of his model of deliberative democracy the idea that people are ‘committed’ to deliberation: ‘. . . the aim of ideal deliberation is to secure agreement among all who are committed to free deliberation among equals . . . ’ (1989: 23); a commitment to deliberation ‘carries with it a commitment to advance the common good and to respect individual autonomy’ (1989: 23). Where does this commitment come from? Clearly it is crucial, since on the face of it those who are uncommitted stand in danger of playing no part in deliberative democratic procedures. The answer is that it comes from a barely-disguised original position, in the form of a ‘formal conception of deliberative democracy’, whose explicit differentiation from the original position serves only to highlight its original-position-like qualities and status in Cohen’s work. Let us look in a little more detail at this argument. One way of cementing into place the commitment levels any fully deliberative democracy would require is to argue a motivational case: if people become accustomed to ‘presenting reasons’ in deliberative forums, then their commitment will be enhanced. If and when people feel more committed in this way, they will be less likely strategically to misrepresent their view in the deliberative forum (1989: 24). However, I suggest this particular argument depends on hope rather than conviction, and that it cannot do the work Cohen needs it to do for his strong view of commitment. Accordingly, Cohen must fall back on his alternative (and main) argument for (universal) commitment. But this alternative argument assumes universal commitment merely by stipulating it. Cohen does this as part of his ‘formal conception of deliberative democracy’ – a spelling out of an ‘intuitive ideal’ of a deliberative association (1989: 21). In this formal conception, Cohen stipulates that the members of an association share ‘a commitment to co-ordinating their activities within institutions that make deliberation possible and according to norms that they arrive at through their deliberation’ (1989: 21). What is the status and origin of this ‘formal conception’? Consider: Cohen begins his piece by citing Rawls. Drawing the view that ‘When properly conducted . . . democratic politics involves public deliberation focused on the common good, requires some form of manifest equality among citizens, and shapes the identity and interests of citizens in ways that contribute to the formation of a public conception of the common good’ (1989: 19) from A Theory of Justice, Cohen endorses Rawls’ ‘informal’ argument for ordering political institutions, which holds that as far as is feasible actual arrangements should ‘mirror’ the ideal conditions of the original position. He then appears to distance himself from this derivation of key features of deliberation from the original position. He writes that the three key features he gets from Rawls

74

Michael Saward comprise elements of an independent and expressly political ideal that is focused in the first instance on the appropriate conduct of public affairs – on, that is, the appropriate way of arriving at collective decisions. And to understand that ideal we ought not to proceed by seeking to ‘mirror’ ideal fairness in the fairness of political arrangements, but instead to proceed by seeking to mirror a system of ideal deliberation in social and political institutions. (1989: 20)

In other words, this is not mirroring the arrangements of the original position, but rather mirroring a different, deliberatively democratic, ideal. My suggestion is that this move in the argument does not work – and it is important to see that it does not. By invoking Rawls and the original position, by deriving directly his key features of deliberation from Rawls, by talking within the theoretical mind-set of A Theory of Justice (‘mirroring’ of ‘ideals’ that are ‘intuitive’), by not specifying the status of his intuitive ideal except by immediate comparison with Rawls’s original position, by expecting that the very status of his construct as an intuitive ideal makes it theoretically compelling – Cohen betrays the origins of his ‘formal conception’. Further, the construction of the details of the formal conception gives the game away: it posits, for example, an ‘independent association’ that is to ‘continue into the indefinite future’ (you do not know your generation behind the veil of ignorance?); self-interest does not exist, or if it does it is wholly subsumed under common commitments; the parties recognize each other as equals. More broadly, the ideal deliberative model acts as a metaphor for the original position in its ‘nowhereness’, its face-to-face assumption, the ‘clean slate’ quality of what it must consider, the fact that people are ironed-out, bloodless and neutral, and that all people are ‘there’ nonetheless. In short, Cohen’s formal conception and the ideal deliberative procedure derived from it gain whatever theoretical power they have by invoking key features of the original position, by being presented in juxtaposition with the original position, and by doing the work of the original position. It is the original position, hypothetical, non-deliberative (all are the same, therefore all are one – the Rousseauian Lawgiver?) unless solitary-deliberative, and – above all – an escape from politics whether deliberative or otherwise. Cohen’s deliberative model represents democratic theory in Rawlsian thrall. Rawls, via the device of the four-stage sequence, effectively rules important, constitutional, issues out of ordinary politics; Cohen appears to invert the fourstage sequence by placing a democratic ideal (the formal conception) in place of the original position, but as I have suggested he does not in fact do this. The architecture of the argument remains the same – the formal conception substituting for the original position. Therefore, Cohen (in line with the Rawlsian four-stage sequence) too leaves little space for ordinary citizens to get their hands dirty with difficult, important political questions. What is unsaid can be as eloquent as what is said; Cohen can hardly avoid the solitary deliberation of reflective equilibrium, though given the effort to distance himself from Rawlsian techniques it is perhaps not surprising that he leaves the status of his intuitive ideal floating, unsupported by some functional substitute for reflective equilibrium.

Democratic legitimacy and deliberative theory

75

Why does this matter? First, it tells us that much of the supporting framework Cohen employs is non-deliberative – possibly anti-deliberative. Second, it vastly narrows the scope for deliberation, a point which (alongside those made earlier) sits very uncomfortably with the idea that deliberation is the source of legitimacy in politics. Third, it leaves Cohen with no real body which could possibly mirror his ideal – we cannot all meet and deliberate face-to-face – which helps to explain the hasty dive from the great heights of the theory to the advocacy of deliberation in political parties and secondary associations (see above). And fourth, if the foundations for ‘commitment’ are inadequate, it opens Cohen’s model to major charges of elitism, and anti-democracy: are the uncommitted – the unconvinced – to be excluded? On this point the mask rarely slips, but when it does it is revealing, as in an admission in Cohen’s later work: ‘Perhaps an ideal deliberative procedure is best institutionalized by ensuring well-conducted political debate among elites’ (1996: 107). Many of the considerations I have sought to sketch here point to the fact that if democratic theory is in Rawlsian thrall it stands in danger of undermining itself: Rawls’s political interest is . . . reflected in the virtual absence of any discussion of democratic processes. He has less to say about the working of deliberative assemblies than he does about economic institutions and the agencies of the state that regulate them. (Esquith and Peterson 1988: 322) Further, as Walzer comments, this may be because the Rawlsian style is that of the ‘withdrawing and retiring’ philosopher. Particular people in particular democratic communities will – and will have the right to – decide on the uses of the philosopher’s systems. Walzer writes: ‘[t]he philosopher himself . . . is the only actual inhabitant of the ideal commonwealth, the only actual participant in the perfect meeting’ (1981: 389). To my mind, this describes at one and the same time both Rawls’s original position and Cohen’s ‘formal conception’2.

Deliberative democracy and exclusion: the Schumpeterian question As is well known, Schumpeter (1976) advocated a minimalist model of democracy; on his view, ordinary citizens so lacked a capacity for rationality and autonomous thought that, realistically, democracy could only mean elite rule and occasional choices among competing elites by citizens. A key charge thrown back at Schumpeterians has been: if the people are so irrational that it would be foolish and dangerous to have them participate politically more than the bare electoral minimum, how is it that they are worthy of having the vote at all? Ian Budge (1996, and this volume) reminds us eloquently that arguments against direct democracy – or, for present purposes, for less democracy – stand in danger of becoming arguments against democracy of any sort. It is here that the final theme I wish to develop against deliberative models of

76

Michael Saward

democracy comes in. The logic of these models is exclusive rather than democratically inclusive. Claus Offe (1997), in one of the most effective discussions of deliberation and democracy, expresses considerable concern about the exclusive implications of the deliberative model. He is prepared to accept arguments that we might discount people’s preferences in democratic politics where those preferences would not serve the welfare of the people concerned. However, he argues, it is quite another thing to downgrade, exclude or ignore preferences because of their origins, rather than their content: No doubt, as it is imprudent in terms of individual and collective welfare to pursue preferences that violate standards of welfare, government is free to ignore or to actively try to change such preferences. But is this also true regarding preferences which are quite neutral in terms of their welfare effects, but just happen to differ from those that disinterested and virtuous citizens might adopt as the outcome of a deliberative collective judgement? I doubt it. (1997: 98). He goes on to argue that ‘the division of the universe of human preferences into those that are “prepolitical” and those that originate from “citizens” in the fullest sense of the word’ is both exclusive (anti-egalitarian) and arbitrary (1997: 98). Just what do deliberative democrats propose to do with non-deliberative preferences – and to the political rights of those who persist in holding them? At a basic democratic level, if you exclude arguments you exclude people. Perhaps this is one reason deliberationists do not talk much about voting; they are not too keen on the idea that (mere, non-reflective, non-deliberative, prepolitical preferencebased) votes should be decisive. But surely equal votes of equal value for equal citizens is a non-optional democratic baseline? Surely, too, because we cannot rely on a deliberative consensus, we must recognize that ‘the everyday institutions of democratic rule such as voting are . . . the heart of democracy, for they define how the umpire operates’ (Gaus 1997: 234)? And, contrary to the deliberationist thrust, it is voting that authorizes government composition and action (Przeworski 1998: 142).

Conclusion Much more would need to be said fully to substantiate the critique offered here; to that extent my comments are suggestive. That said, the arguments do suggest that the ‘deliberative model of democracy’ has involved some careless, overblown claims about democratic legitimation; that core threads of that model rely on a metaphorical architecture which serves directly to undermine a range of core claims; and that (to say the least) advocates have paid insufficient attention to exclusivist, rather than democratically inclusive, implications of the model. Perhaps the overarching point to be made in this context is that there is no such thing – there can be no such thing – as ‘the deliberative model of democracy’. Deliberation, of a certain kind, is of course a desirable feature of a healthy,

Democratic legitimacy and deliberative theory

77

functioning, dynamic democratic system. But such systems – to state what ought to be obvious – also require much more: constitutional structures; formal (and to some degree hierarchical) organizations; voting and other decision mechanisms that can be decisive in the last instance; and institutionalized equal respect for all citizens regardless of (for example) their willingness or even capacity to engage in deliberation or other distinctive forms of political participation. Does this mean that democratic decisions can be wrong, ill-informed, misguided, reflective of nondeliberative preferences? Of course. Perfectionist desires cannot make that fact go away. In the end, as Dahl says, democracy is a ‘gamble that a people, in acting autonomously, will learn to act rightly’ (Dahl 1989: 192). One last point: surely a key response to deliberative concerns about ill-informed preferences lies in inclusive voter education, rather than in the elitist, exclusive threads of the ‘deliberative model’. Life-long learning, citizenship education in schools, new ‘enabling’ institutions, creative use of the media – these are some of the routes by which democracy might be brought to the people. I see no reason why, for example, the deliberative polls of Fishkin and Luskin (Chapter 1) should not play a part. Once we have set aside the ‘deliberative model of democracy’, we can begin to take seriously how structured deliberation between real citizens might be a revitalizing element of a more democratic future.

Notes 1 Key deliberationists are subject to the criticism that Sartori (1987) effectively aimed at ‘participatory democrats’ such as Peter Bachrach, namely the contrast of a real system (in our case, aggregative) with an ideal one (deliberative). The ideal will inevitably come out shining in any such contrast, less for its substantive superiority and more for the way the contrast is set up in the first place. 2 Clearly, this critique of Cohen only works as a critique of deliberative democracy more generally if Cohen’s approach is in key senses representative of the broader canon. Fishkin (1991, and this volume), for example, makes fewer claims about legitimacy, and so is less vulnerable to this critique. I would argue that Benhabib (1996a) is vulnerable, with her expansive claims about deliberation and legitimation. Though more would need to be said to substantiate these points, I hope to have shown at least that many deliberative democrats have a case to answer.

6

Discursive democracy vs. liberal constitutionalism John S. Dryzek

Understanding the deliberative turn For better or for worse, the theory of democracy came in the 1990s to be dominated by a deliberative approach, as is confirmed by the attention paid to deliberation in the other chapters of this volume. This swift conquest is remarkable given the question mark that still hangs over deliberation when it comes to perhaps the central question of democracy: how do collective decisions get made? I shall argue that in the breathless rush to advance deliberation, a key distinction has been lost. There are in fact two different views of democracy that have been subsumed and intermingled under the deliberative heading. In this chapter I intend to disentangle these two views, criticize one, defend the other, and show how this distinction helps make sense of some of the defences, illustrations and criticisms of deliberative democracy that appear in other chapters of this volume, as well as in the field more generally. The two views, I label respectively liberal constitutionalist deliberative democracy and discursive democracy. Part of the reason for the rapid advance of deliberative democracy lies in its accommodation with key aspects of liberal constitutionalism – which, in this liberal era, makes deliberative democracy safe for dominant forms of politics. The terms ‘deliberative democracy’ and ‘discursive democracy’ are now used virtually interchangeably in the literature – including, on occasions, by myself. But given that I coined the term ‘discursive democracy’ (Dryzek 1987a; 1990a), I now claim proprietary rights. I seek to recover its meaning as a more insistently critical orientation to democracy – and especially to the institutions of the liberal state.1 But before pursuing this distinction and its implications, let me say exactly what I think sets deliberation apart from other forms of communication, and so helps to define deliberative democracy in general. Deliberation has informational, argumentative, reflective, and social dimensions. The informational component involves bringing new facts, interpretations and perspectives to the awareness of others. The argumentative component enables questioning of the logical consistency of the positions of others, uncovering of premises and assumptions, clarification whether disagreement is a matter of conflicting interpretations of facts or values, and elucidation of the interdependence

Discursive democracy vs. liberal constitutionalism

79

of issues. The reflective component arises inasmuch as individuals are induced to think about their own positions by listening to others, in the knowledge that these positions must be justified in terms that others can accept. The social component arises because deliberation creates a situation in which people talk and listen to others, and so may come to recognize their common membership of a social group (for a more formal treatment of these conditions, see Dryzek and List 1999). These four dimensions indicate that though deliberation involves argument, it can also admit other kinds of communication – and this, I will show, is helpful when it comes to replying to at least some of its critics.

Deliberative democracy’s assimilation to liberal constitutionalism So much for deliberation: what about deliberative democracy? The standard formulation is that in a deliberative democracy, collective decisions are legitimate to the extent that individuals have the right or capacity to participate in deliberation in decisions that affect them, such that these decisions must be justified in terms that these individuals are capable of accepting on reflection (see, for example, Cohen 1989). It should be stressed that there is no compulsion to deliberate: it is only the right and capacity that is at issue here, which individuals may choose not to exercise. Still, framed in this way, there seem to be obvious affinities between deliberative democracy and participatory democracy. In this light, it is perhaps surprising that by the late 1990s deliberative democracy had been for the most part taken under the wing of liberal constitutionalism, in which democracy is indirect and representative (this is one aspect of what Bohman 1998, terms ‘the coming of age of deliberative democracy’). This development is more surprising still given that liberal democracy by definition deals in the aggregation of preferences defined prior to political interaction, while deliberative democracy requires the transformation of preferences in interaction (see the distinctions drawn by Miller 1993a and Warren 1992). How could this reconciliation with liberal constitutionalism happen? How could what began as a radical challenge to existing forms of democracy end up being assimilated to these forms (and so add further proof that liberalism is the most effective vacuum cleaner in the history of political thought)? Part of the answer is that liberalism is a remarkably flexible and accommodating doctrine. Still, it is not just accommodation with some vague ‘liberal democracy’ – itself an uneasy compromise between liberalism and democracy – that is at issue here. Rather, the accommodation has occurred in relation to a much more well-defined subset of liberal thinking: that which is concerned with the specification of constitutional principles and rules. The reconciliation turns on the fact that the standard formulation of deliberative democracy in terms of legitimacy still begs the question of how collective decisions actually get made. Liberalism, or rather liberal democracy, has a set of institutions that provide a ready answer. And, as I will now show, there are three ways in which liberal constitutionalists can reason about these institutions in deliberative terms.

80

John S. Dryzek

First, deliberative principles can justify liberal rights. Notably, Joshua Cohen (1996) tries to specify liberalism’s principal rights and equalities on the grounds that they are necessary to ‘free public reasoning among equals’ (p. 99) (see also Gutmann and Thompson 1996). The rights that follow are freedom of expression, freedom of religion (out of respect for those who cannot be reached by argument), and equal political rights (because deliberation must be capable of persuading all, so all must be in a position to be persuaded, as opposed to coerced). Second, liberal constitutions can be interpreted as devices to protect a deliberative realm. The more standard liberal approach treats constitutions as necessary first and foremost to protect a private realm from the intrusions of government and other individuals. A deliberative interpretation, in contrast, emphasizes the public realm that constitutions protect. Walzer (1991) interprets the Bill of Rights in the US Constitution in these terms. Rawls (1997: 772), Bessette (1994), and Uhr (1998) are among those who stress deliberation in the legislature; but emphasis can also fall upon the courts (Rawls 1993: 231 believes the US Supreme Court is the best place to look for public reason), and public life more generally. Third, constitution-making itself can be seen as a deliberative process. Ackerman (1991) and Estlund (1993) are among those who believe that liberal politics is normally aggregative and partisan, but that on special occasions, notably those attending the writing of constitutions, deliberation should enter. Deliberative capacity is treated by Ackerman as a scarce resource, which must be reserved for great crises of state, or great opportunities. Similarly, for Rawls (1993) deliberation is not to be the normal mode of government, but should only concern constitutions and matters of ‘basic justice’ (later I will address the issue of whether Rawls is truly a deliberative democrat; this itself has major implications for the adequacy of a liberal constitutionalist assimilation of deliberation). However, other liberals believe that deliberation is more widely applicable; for example, Gutmann and Thompson (1996) want it to be applied to all public issues featuring deep moral disagreement. What is wrong with the assimilation of deliberative democracy to liberal constitutionalism? First, the incompleteness problem, concerning how collective decisions get made, that motivates a search for some authoritative institutional structure in which to house deliberation is only a problem to the extent democratic theorists are expected to provide complete models of democracy. Elsewhere, I have argued that democratic theorists should forget models of democracy, and emphasize instead processes of democratization (Dryzek 1996). Realistically, no real-world political system is ever going to adopt a theorist’s model wholesale; far better, then, for theorists, including deliberative theorists, to locate and interrogate the spaces for democratization that really do exist. A second reason to reject assimilation is liberal constitutionalism’s implicit assumption that constitutional structure fully or primarily determines the kind of politics that occurs. This assumption is manifestly false. Liberal states all operate within capitalist political economies, which are home to both material and discursive forces that can determine policy outcomes irrespective of the constitutional structure. When, for example, it comes to maintaining the confidence of

Discursive democracy vs. liberal constitutionalism

81

financial and capitalist markets, all states must produce policies of pretty much the same sort – otherwise they will be punished by disinvestment and capital flight. Other agents of distortion include the instability and arbitrariness associated with aggregative conceptions of democracy, highlighted by social choice theorists (though deliberation can curb these problems; see Dryzek and List 1999), and the effective suppression of alternative voices by supposedly neutral liberal norms of dialogue. The latter possibility is highlighted by difference democrats, addressed in this collection in the chapters by Squires and Meier. Finally, the de facto policymaking process may bear little resemblance to what is specified in the constitution; for example, corporatist systems that have set up mechanisms for policy formulation and implementation that bypass both parliament and the public bureaucracy.

Discursive democracy as a critical alternative For all these reasons it is important to resist the assimilation of deliberative democracy to liberal constitutionalism, and to recover discursive democracy as an alternative that is more insistently critical not just of liberal constitutions but, more importantly, the state structures and political economy of which these constitutions are just a part. Discursive democracy does not take constitutional structure at face value. It is, in addition, aware of the need for deliberation within a public sphere that may find itself in confrontation with the state, as well as within state structures. Beyond its critical orientation to the liberal state, discursive democracy is expansive in the kind of communication it allows and promotes, refusing confinement to reasoned, measured, gentlemanly argument. Elsewhere (Dryzek 2000), I argue that discursive democracy should also be pluralistic in promoting communication across difference without dismantling difference, questioning in its attitude to established traditions, transnational in its extension beyond state boundaries to a system where there is no constitutional framework, green in its openness to communication with non-human nature (see also Eckersley in this volume), and flexible in its responsiveness to the changing limits upon and opportunities for further democratization. This discursive approach was once associated with the main non-liberal theoretical root of discursive democracy, critical theory, and in particular Habermas’s (1984; 1987) account of communicative action. In these terms, communicative rationality describes the extent to which social and political interaction is free from domination, coercion, manipulation, and strategizing, and engaged in by competent actors. The connection to the conditions of authentic deliberation is straightforward. Those influenced by this account (for example, Benhabib 1996a; Bohman 1996; Chambers 1996; Dryzek 1990a) were careful not to locate the main potential for authentic deliberation within the institutions of the liberal state, but rather in a public sphere that may encompass state actors, but may also be constituted in large part by social movements and actors in confrontation with the state. Habermas’s own historical archetype was the early bourgeois public sphere in Europe (Habermas 1989).

82

John S. Dryzek

This connection has been weakened by the recent defection of Habermas himself and Habermasians such as Bohman (1998). Now, Habermas is not yet a liberal constitutionalist in the American mould. He retains an emphasis on the public sphere as the main site for the generation of public opinion, which is in turn central to democratic authenticity. This sort of emphasis is put to good use in this volume in Eriksen’s elaboration of a new way to think about remedying the European Union’s democratic deficit – though, significantly, Eriksen proclaims Habermas’s (and by implication his own) ‘liberal position’. Aside from the continued emphasis on the public sphere, in Between Facts and Norms (Habermas 1996c) we see a model of democracy that is otherwise quite consistent with the liberal constitutionalist model, in particular because a liberal constitution is regarded as necessary to protect and nurture the opinion-formation capacities of the public sphere. What happens next is that ‘Informal public opinion-formation generates “influence”; influence is transformed into “communicative power” through the channels of political elections; and communicative power is again transformed into “administrative power” through legislation’ (Habermas 1996a: 28). This is actually a very conventional model of democracy; not so distant from a cardboard cut-out, civics-textbook version of how democratic government should work. It is a bit like Lowi’s (1969) ‘juridical democracy’ in which administrators behave precisely according to the instructions they receive from a deliberative legislature that represents public opinion in all its majesty. Habermas justifies this model in part through acceptance of the facts of complex, differentiated societies in which face-to-face direct democracy is no longer a possibility. But there are other facts about the contemporary world that he now downplays: in particular, the fact that states are in large measure defined by the imperatives that they must follow, which in turn flow from the politicaleconomic structure in which they are embedded. He has turned his back on extraconstitutional agents of distortion of democracy – but also upon extraconstitutional agents of democratic influence which, I would argue, are central to discursive democracy. Ouch! The liberal constitutionalist and discursive categories are not mutually exclusive and collectively exhaustive when it comes to mapping the territory of deliberative democracy. Rather, they represent two tendencies, which can be scrambled in the works of particular theorists (for example, Eriksen and the later Habermas). Nor should it be assumed that the difference is necessarily that liberal constitutionalists seek deliberation in state institutions (notably courts and legislatures), while discursive democrats seek deliberation outside the state in the public sphere. Rather, the key difference is that discursive democracy problematizes the democratic potential of the state in the way liberal constitutionalism does not. So discursive democracy accepts that sometimes authentic deliberation within state structures may be possible; but that often, perhaps most of the time, it is not. Consider, in this light, the deliberative opinion polls and citizen juries discussed in the chapters by Fishkin and Luskin, and Smith. Deliberative polls fall outside liberal constitutional structure; indeed, it is the democratic weaknesses of this structure that justify its enhancement by deliberative polls to begin with.

Discursive democracy vs. liberal constitutionalism

83

Deliberative polls embody a non-liberal kind of representation, based on a more republican view of the deliberative capacities of ordinary citizens. Republicans might, however, prefer the results to be summarized by the deliberators themselves, rather than by the analyst of the poll data. Citizens’ juries too are justified by the intrinsic weaknesses of democracy within the liberal constitution. Smith is right to characterize citizens’ juries as indicative of a deliberative alternative to liberal democracy. This distance between deliberative polls and citizens’ juries on the one hand and liberal constitutionalism on the other is perhaps secure to the degree they remain advisory, rather than authoritative, policy-making bodies. Perhaps the key feature that sets discursive democracy apart from liberal constitutionalism is a sense in which the former is in the end truer to the democratic project. Central to the very idea of democracy is insistent search for more and better democracy (formally, this can be on any one of three dimensions: the extent of effective franchise; the scope of the issues under popular control; and the authenticity of that control). In reaching an easy accommodation with liberal constitutionalism, and so helping to justify the institutions of the (democratically problematic) liberal state, liberal constitutionalism prematurely forecloses on the search for other paths that might further the democratic project – and indeed, perhaps even further it more effectively. Liberal constitutionalist deliberative democrats do not of course rule out democratic innovation, and they generally seek to secure and deepen the democratic qualities of liberal institutions. But they leave themselves very little room here, in light of both the long-established tension between liberal and democratic precepts, and the constraints on the degree of democracy attainable within the contemporary liberal state to which I have alluded.

Enter the critics A further reason to resist the assimilation of deliberative democracy to liberal constitutionalism is that this move depletes the imaginative resources necessary to confront critics of deliberative democracy. Here I do not have the space to confront the entire range of such critics, so I will demonstrate this point through reference to the preceding chapter by Saward, on the grounds that this constitutes one of the most trenchant and incisive critiques of deliberative democracy now available. My point here is not that the discursive variant I have outlined is an allpurpose defence against every criticism that could be pressed against deliberative democracy; that would of course be illogical in the absence of comprehensive knowledge of all criticisms that could be made. I merely want to suggest that discursive democracy enables a better response to some key lines of criticism that have now been pressed – by Saward and others. Saward’s charges are as follows. 1

Deliberation is not self-sufficient as a model of democracy because it cannot say how collective decisions should be made; thus we must fall back on more familiar aggregative mechanisms, such as voting.

84

John S. Dryzek

2

Democratic legitimacy, to be democratic, requires equal votes for all regardless of the extent to which they choose to participate in deliberation. Deliberative theory gives no guide as to how much deliberation, and of what type, must occur for democratic legitimacy to be achieved. Deliberative democrats site deliberation in very different venues: parliament, deliberative polls, high courts, and informal associations. Some of these are undemocratic (courts); others do not contribute to legitimacy (informal associations). Rawlsian theorizing, deployed by Cohen among others, underpins deliberation; but that theorizing is ultimately non-deliberative; it leads deliberation into an elitism that denigrates the deliberative capacities of ordinary people, and an ultimately anti-political attitude that as such cannot speak in a useful way to the real world of politics. Deliberative democracy is repressive and exclusive because it excludes the holders of non-deliberative preferences from political equality, not recognizing their political rights. This is closely related to the second point, above.

3 4

5

6

Saward’s first charge actually stacks the deck a bit against deliberation by assuming that democracy must be a matter of counting heads. He also assumes that counting heads is not a problem – but, as social choice theory has shown, it is a massive problem; indeed, there is no popular will independent of the mechanism designed to ascertain it. At this point we can step back and ask whether democracy does indeed require counting heads. From the point of view of discursive democracy, I would argue that a logically complete alternative exists based on a conceptualization of intersubjective communication in the public sphere as a matter of the contestation of discourses. Examples of such contestation might involve industrialism, environmentalism, and sustainable development; patriarchy and various kinds of feminism; realism and idealism in international affairs. In this light, public opinion can be interpreted as the provisional outcome of this contestation, as transmitted to the state through a variety of discursive means. I lack the space here to develop this alternative (see Dryzek 2000 for a full account). Voting need not, then, be central to democracy; and the alternative need not be the highly problematic idea of consensus. Eriksen is quite right to reject consensus as an ideal for real-world deliberative democracy; but, as he points out, we can still seek workable agreements (or what Sunstein 1997 calls ‘incompletely theorized agreements’) instead. Consensus requires that individuals agree on a course of action for identical reasons; workable agreements allow for continued difference as to why a course of action is desirable, though in a deliberative democracy the justifications adduced have to be accepted as legitimate by other participants, even if they are not shared.

Different kinds of communication Saward’s second and sixth criticisms relate strongly to the more widely-shared criticism that deliberative democracy is too elitist and exclusive; it is too much like

Discursive democracy vs. liberal constitutionalism

85

a gentlemen’s club. Relatedly, some difference democrats believe that deliberation excludes the communication styles that would advance the interests of oppressed minorities; I will respond to that line of criticism shortly. Saward himself is not so concerned with communicative style, but rather with more straightforward exclusive tendencies. Notably, those who choose not to deliberate are excluded (point two); those who do participate, but whose preferences turn out to fail the ‘deliberative’ test by being inconsistent with (say) Rawlsian versions of public reason, are also excluded (point six). Thus his criticisms two and six are underwritten by the proclaimed foundational nature of Rawlsian public reason for deliberative democracy (point five) – or at least something very close to it, if not Rawls’s own precise specification. This connection suggests that a more expansive view of what constitutes deliberation might enable effective response to Saward’s points two, five and six. Liberal constitutionalists like Rawls are constricted not just in the venues they specify for deliberation, but also in the kinds of communications that are admissible. Communication must consist of argument couched in neutral terms, at least when it comes to matters of the constitution and ‘basic justice’ (once these structures are in place, politics can then feature self-interest and strategizing). A more expansive notion of deliberation requires only that communication induce reflection on preferences in a non-coercive fashion. This does not mean that ‘anything goes’ in terms of communication; but it does mean that the bar is lowered substantially when it comes to the preferences that can pass the ‘deliberative’ test (point six). An expansive definition of deliberation would therefore allow not only argument in varied terms, but also rhetoric, humour, emotion, testimony or storytelling, even gossip. The stipulation that communication be non-coercive excludes domination through the imposition of power, manipulation, indoctrination, propaganda, deception, threats (of the kind that define bargaining), and the imposition of ideological conformity. Mere expressions of self-interest are unlikely to pass the test of inducing reflection in others – but they could, if tied to more general (but not necessarily universal) principles. For example, a polluter who claimed that environmental regulations hurt his profits could expect to find little resonance in fellow-deliberators – unless, perhaps, he tied his case to more general principles concerning the sanctity of private property (which might not convince anyone but might at least induce reflection). A more expansive conception of the kinds of admissible communication also enables deliberative democrats to respond to the criticisms of difference democrats – or, as Squires prefers to call them in her chapter, ‘diversity democrats’, such as Iris Young. Part of the problem in replying to critics such as Young is that, as Squires rightly notes, it is unclear whether Young proposes a communicative ethics of difference or just a variation of communicative impartiality. Still, the charge of the difference democrats remains that deliberation seeks homogeneous participants committed to particular standards of rational, impartial argument; and to the extent this is true, excludes the particular experiences of the oppressed in supposedly neutral ways that in reality favour the interests of the advantaged.

86

John S. Dryzek

But if we redefine deliberation in terms of any kind of communication that induces reflection on preferences in non-coercive fashion, the force of the difference democrats’ critique dissolves; we can, indeed, agree with Young (1997) that group difference is actually a resource for deliberation. Indeed, a model of the public sphere in terms of the contestation of discourses, central to discursive democracy, is quite compatible with difference democracy. One should, however, still apply critical standards, or communicative ethics, within that sphere: communications are to be welcomed if they can pass two tests. First, as I have already indicated, they must be non-coercive; second, they must be capable of connecting the particular to the general (otherwise, there is no appeal across difference). For an example of the latter test, consider the story of a refugee from ethnic cleansing: such stories have power to the degree they speak to universal ideals of humanity, as opposed to the particular interests of an ethnic group locked in a struggle with another group. Now, even after these expansions, Saward might still want to claim that the capacity and desire to deliberate is not universally or equally distributed. Fair enough; but then neither is the desire (say) to vote. People can be forced to vote (as in Australia), but they cannot be forced to deliberate – but that is only as a result of the non-coercive element built into the definition of deliberation. This brings us back to Saward’s third criticism: that it is not clear how much and what kind of deliberation is sufficient for democratic legitimacy to be achieved. Setting aside the issue that such lack of quantitative precision is the normal condition of political theory, the best way of answering this charge is through reference to the character of the democratic project as I characterized it earlier: that intrinsic to the idea of democracy is the search for more democracy. In this light, models of democracy, still less models of democracy with quantitative requirements for the amount of deliberation that must occur for a decision to be legitimate, are less interesting than processes of democratization. And it is quite easy to recognize both advances and retreats in democratic legitimacy, even without the specification of a precise quantitative threshold as sought by Saward.

The role of the public sphere Let me turn now to Saward’s fourth criticism, pertaining to the problematic character of the venues in which deliberation might be sought. Discursive democrats can accept his criticism of deliberation in high courts – especially in light of all the restrictions that courts impose upon admissible kinds of communication. The insistently critical component of discursive democracy is well-placed to exercise vigilance over the democratic qualities of all institutions of state, legislatures and discursive designs as well as courts, and also to conduct comparative analyses of the prospects for deliberative authenticity in these institutions and the more ‘informal’ public sphere. Saward would presumably respond that deliberation in the latter venue cannot contribute to democratic legitimacy because there is no connection to the collective decisions that need to be legitimated to begin with.

Discursive democracy vs. liberal constitutionalism

87

However, a connection can be made. The public sphere may be separate from the state, but it is almost by definition oriented to the state and its actions. Moreover, deliberation in the public sphere can exercise influence over the content of public policy. Habermas has long recognized this influence, despite his more recent (1996c) stress on elections as the main means of transmission. How is such influence exercised? By rhetoric; by changing the terms of political discourse in ways that can pervade the understandings of state officials; perhaps by creating cultural change; by creating fear of instability; perhaps even as a result of good arguments being articulated and then heard by public officials. The institutions of the liberal state, or any other kind of state, do not just operate in the formal way their constitutional ‘hardware’ suggests. In addition, their operations, and the content of the policies they produce, must also be understood in terms of the discursive ‘software’ – discourses in the Foucauldian sense – that pervades them, as Foucault himself and his followers have long stressed. While Saward’s concern is that deliberation in the public sphere cannot easily contribute to democratic legitimacy because it is severed from collective decision making, Rättilä worries that ‘deliberation’ in the public sphere is not actually very deliberative: that it must inevitably fall short of ideals of openness and inclusion, especially because many people are not members of deliberating associations. She feels too, in criticizing Benhabib’s notion of ‘an anonymous public conversation’, that it is hard to apply critical standards of any sort to this elusive kind of interaction; indeed, it is hard for participants and observers alike to know whether what is going on is actually deliberative and democratic. Rättilä is right that many conceptions of deliberation – and, I would suggest, especially those associated with liberal constitutionalism, and more especially still those that stress ‘public reason’ – sit uneasily with what we observe of the public sphere in action in the real world. However, I do not feel that a greater emphasis on ‘performative’ action really speaks to the problem. It is not easy to see how performance relates to the generation of public opinion, still less how it connects with collective decision. What I said about admitting any kind of communication that induces reflection on preferences in non-coercive fashion does, however, apply to performative action too – so there is no reason to banish it. My point is simply that an exclusive emphasis on such action would be inadequate. A better response to these problems would I believe take as its point of departure the idea that public opinion in the public sphere is generated in the contestation of discourses, and both the content and outcomes of that contestation can be transmitted to the state in a variety of ways. Such contestation takes us far from the ‘university seminar’ characterization of deliberation portrayed by Budge in his contribution to this volume. Now, the mere fact of contestation does not signify democracy, least of all deliberative democracy. It can be engaged in unequally, manipulated by spin doctors, public relations experts and propagandists, and prove vulnerable to the imposition of religious and ideological orthodoxies. So the key question from the point of view of deliberative democracy is: who controls the terms of this discourse, and on what basis? We can speak of a relatively

88

John S. Dryzek

democratic interplay of discourses to the degree it is engaged in by a broad variety of competent actors.

Networks of democracy How might this sort of engagement be institutionalized? Here, the network form of organization as celebrated by Schlosberg (1999) in the context of the US environmental justice movement merits attention (see also Torgerson 1999: 148–54). The basic idea of a network is that it is a bottom-up form of organization, constituted in the first instance by a variety of actors with very different backgrounds and local experiences – but with some issue in common. The issue might be biopiracy, which currently receives opposition from a transnational network, or toxic waste dumping, or pollution from oil refineries. A network has no central organization (save perhaps for an informational clearing-house), so it is hard for hierarchical leadership to develop and monopolize or control negotiation. The very fact of diversity across network participants, combined with the absence of any central leadership to impose principles and strategies, means that the network has to embody norms of openness, equality, respect and reciprocity in order to function. Sometimes these norms are formalized in network principles (for an example from the Southwest Network for Environmental and Economic Justice in the USA, see Schlosberg 1999: 128). Thus decentralized and discursive control of the terms of communication within the network, and of the terms with which the network engages the broader world, are not contingent features of particular networks; they are integral to the very idea of a network. They can even be extended to interactions with those the network opposes. The discursive contests that networks engage in are not organized centrally, but instead emerge from the different experiences of network participants. In the case of the network against biopiracy, for example, the network’s stance is an outgrowth of a variety of local experiences: farmers in Ethiopia who confronted scientists from the University of Toledo, indigenous peoples trying to protect their rainforest’s resources in Central America. Encompassing activists and communities from many countries, this network relies on a variety of particular experiences and ecological knowledges; these enable construction of what biopiracy means in practice and exploration of the ways it can be resisted. The biopiracy network has been quite successful in reframing the terms of international debate on this issue – which previously turned on the more innocuous-sounding ‘bioprospecting’. Similarly, the networks of the US environmental justice movement have successfully brought issues concerning the distribution of environmental hazards across racial, ethnic and social class lines to public attention – and into public policy practice.

Conclusion Deliberative democracy is not just one thing. I have argued that the key issue is not whether to embrace or reject deliberation: in fact, it is hard to imagine

Discursive democracy vs. liberal constitutionalism

89

democracy of any sort without some deliberation, somewhere. Rather, we should ask who gets to deliberate, where they should do it, what kinds of communications are admissible, what kinds of institutions – including informal ones such as networks – are conducive to authentic deliberation, and what kinds of connections can be made between deliberation and collective decision. I have argued for some particular answers to these questions that I believe define a defensible approach to discursive democracy. While allowing that authentic deliberation within and about the constitutional structure of liberal states can sometimes occur, discursive democracy does not remain confined to the constitutional surface of political life, adopting instead a more critical approach to the liberal state and its political economy. This escape enables discursive democracy to further the democratic project in ways unavailable to liberal constitutionalism – in a variety of locations, and in a variety of modes of communication.

Notes 1 To be more precise about conceptual origins: in 1985 Terence Ball suggested the title ‘Discursive Designs’ for a paper I was writing on critical theory and political institutions. Discursive designs then became central to discursive democracy.

Part II

Representation and deliberation

7

Group representation, deliberation and the displacement of dichotomies Judith Squires

The emergence of a model of deliberative democracy is perhaps one of the most significant recent innovations in democratic theory. Yet this deliberative model of democratic theory has received an ambivalent reception amongst feminist political theorists. Although it appears to some to offer invaluable theoretical resources for engaging with central feminist concerns regarding democratic inclusion, it generates amongst others a profound scepticism concerning its ability to recognize difference. The relation between deliberative democracy and feminist theory is ambivalent, then, not least because feminist theory is itself a contested terrain. If one focuses on the feminist literature itself it becomes apparent that the most significant innovation regarding democratic theory here is the emergence of arguments for group representation. The feminist theorist who has come to be most closely associated with such arguments is perhaps Iris Marion Young. Significantly, Young is also a cautious advocate of a modified model of deliberative democracy. I propose to explore the detail of Young’s attempt to negotiate a form of deliberative democracy modified by group representation, taking her project as indicative of the more general issues surrounding the attempt to synthesize two distinct, and at times antithetical, bodies of thought. I hope to explain why Young is critical of deliberative democracy but nonetheless endorses a (modified) form of it. The explanation, I suggest, arises from the particular form of feminist theoretical frame that she adopts, which I label a strategy of displacement. Young tells us that ‘. . . a politics that aims to do justice through public discussion and decision making must theorize and aim to practice a third alternative to both a private interest competition and one that denies the reality of difference in public discussions of the common good. This third way consists in a process of public discussion and decision making that includes and affirms all particular social group perspectives in the society and draws on their situated knowledge as a resource for enlarging the understanding of everyone and moving them beyond their own parochial interests’ (Young 1997: 399). Is this third way convincing? Do the concepts of social groups and objective judgement that she develops offer a compelling displacement of the dichotomies between interests and identities, and between impartiality and particularity? And, if so, does this offer a way of

94

Judith Squires

integrating feminist arguments for group representation with deliberative accounts of democracy?

Feminist theory When considering democratic innovations arising within feminism it is useful to grasp two central features about current feminist theory. It is, first, not a unified body of literature and can best be understood as falling into three distinct camps, which can be characterized by pursuit of strategies of ‘inclusion’, ‘reversal’ and ‘displacement’. It has in recent years also tended to focus on ontological rather than advocacy issues. Let me briefly say something about each. Those pursuing a strategy of inclusion aim to include women in a politics from which they are currently excluded. They usually aspire to impartiality, conceive of people as autonomous and espouse an equality politics. They are often labelled liberal feminists. Those pursuing a strategy of reversal aim to reconfigure the political as currently conceived such that it becomes more open to their gendered specificity. They usually adopt an interpretative methodology, talk of ‘Woman’ or ‘women’ and espouse a difference politics. They are often labelled radical, maternal or cultural feminists. Those pursuing a strategy of displacement aim to destabilize the apparent opposition between the strategies of inclusion and reversal. They usually adopt a genealogical methodology, speak of subject positions and of gendering (as a verb) rather than gender (as a noun), and espouse a diversity politics. They are often labelled postmodern or post-structuralist feminists. The strategy of inclusion seeks gender-neutrality; the strategy of reversal seeks recognition for a specifically female gendered identity; and the strategy of displacement seeks to deconstruct those discursive regimes that engender the subject (see Ferguson 1993 and Di Stefano 1990 for similar typologies). This last strategy of displacement has had profound implications for the nature of debate within gender theory. Prior to its impact it was common to find feminist theory characterized by a clear opposition between those who would endorse and extend dominant values to all irrespective of gender, and those who would challenge and reverse dominant values from a specifically female perspective. The advocate of displacement, by contrast, argues that whether gender justice was thought to entail the extension or reversal of dominant norms it actually manifests a tendency to echo that which it sought to oppose. Both operate, in different ways, within a dichotomous framework generated by established power networks. The second general claim that I would make about feminist theory is that there has been a tendency in recent years to focus on questions of subjectivity rather than on issues regarding normative political proposals. If we adopt Taylor’s distinction between ontological and advocacy issues, we can see that much feminist theory has been preoccupied with the former. Ontological issues are concerned with the question of what ‘you recognize as the factors you will invoke to account for social life’; advocacy issues concern ‘the moral stand or policy one adopts’ (Taylor 1989: 159). As Taylor argues (in relation to communitarian theses about identity), whilst such debates are purely ontological, they do not amount to

Group representation and deliberation 95 an advocacy of anything. ‘What they do purport to do, like any good ontological thesis, is to structure the field of possibilities in a more perspicuous way. But this precisely leaves us with choices, which we need some normative, deliberative, arguments to resolve’ (Taylor 1989: 161). The ‘turn to culture’ and the shift from the social sciences to the humanities in feminist theorizing has meant that these normative, deliberative arguments have been somewhat marginalized of late (Barrett and Phillips 1992). The undeniable benefit of the focus on subjectivity has been the creation of ‘a new political imaginary centred on notions of “identity”, “difference”, “cultural domination”, and “recognition” ’ (Fraser 1997: 11). The concern, frequently articulated by those engaged in these debates themselves, is that much recent feminist theorizing has become overly entrenched and antagonistic regarding issues of ontology and worryingly sketchy and inattentive regarding issues of advocacy. As Fraser says, ‘if the politics of the deconstructive theorists have at times been simplistic, it is probably because of the difficulty that arises from trying to deduce a normative politics of culture from an ontological conception of identity and difference’ (Fraser 1997: 183). Young’s attempt to use a strategy of displacement to negotiate a theory of democracy is clearly not simplistic in this way. But it does contain tensions which indicate that more work needs to be done in this area.

Interests, identities and group representation Turning our attention to the issue of democratic theory we can, I think, discern the impact of these two general features of current feminist theorizing. The most significant innovation in democratic theory within the feminist literature seems to me to be the concentration on issues of group representation. One can argue for group representation from each of the three frames detailed above, but the detail of one’s proposals and the justification given for them will vary according to whether one adopts a strategy of inclusion, of reversal or of displacement. Young argues for group representation as a strategy of displacement. In order to understand the detail of her account, it is important to locate it in relation to other recent feminist theory. This enables us to see both the shared commitment to group representation (not found so readily in more mainstream democratic theory) and the distinct form of defence offered for this form of democratic structure. A brief consideration of the work of Virginia Sapiro, Irene Diamond and Nancy Hartsock, and finally Anne Phillips should serve to establish that group representation is a shared preoccupation and that it is endorsed by advocates of inclusion (as with Sapiro), reversal (as with Diamond and Hartsock) and a synthesis of the two (as with Phillips). Sapiro explores the claim that women share particular experiences and have common ‘representable interests’ (Sapiro 1998: 164). To assess whether women are an ‘interest group’ and, if so, what interests they have, Sapiro proposes that one considers both women’s ‘objective situation’ and their consciousness of their own interests. Her claim is that women may have a different social position from that of men and therefore have interests to be represented, without being conscious of

96

Judith Squires

these differences (Sapiro 1998: 167). This is politically significant because, contrary to the Burkean notion of paternalistic representation of the interests of others, political systems are – Sapiro notes – not likely to represent previously under-represented groups ‘until those groups develop a sense of their own interests and place demands upon the system’ (Sapiro 1998: 167). Moreover, if the interests in question are not clear and pre-formed, but are still in the process of being uncovered via processes of consciousness-raising, it will then be more difficult to distinguish between the represented and the representative. In these circumstances women would seem to be best placed to advocate the interests of women. In contrast, Diamond and Hartsock argue against casting women as simply another interest group among many, and against arguing that fairness requires that women promote their interests within the existing political system equally with all other such interest groups. This, they claim, underplays the distinctive and radical challenge posed by the recognition of women’s experiences and political ambitions. It also overlooks the new political and methodological questions raised by their position: ‘if the inclusion of women into politics threatens the most basic structures of society, one cannot fit their concerns into the framework of interests’ (Diamond and Hartsock 1998: 193). The very language of interests, they argue, emerges with and then perpetuates the division of labour that creates the ideal of rational economic men seeking to maximize their satisfactions. They propose that it be replaced by more encompassing categories of analysis which more adequately capture the range of human emotions, such as needs. Their basic resistance to Sapiro’s focus on interests is that it implies that the issue of women’s fair representation is an issue of inclusion: that women are seeking to catch up with men (Diamond and Hartsock 1998: 197). In direct contrast they seek recognition that female experience inverts that of the male and forms a basis on which to expose masculine values as fundamentally flawed (Diamond and Hartsock 1998: 195). Including questions of reproduction and sexuality into the political process will transform the very concept of the political, eroding the public/private distinction and, presumably (though they do not state this directly), undermining the current system of representative democracy in favour of a more participatory one. Nonetheless, within the confines of the current representative system they are clear that ‘only women can “act for” women in identifying “invisible” problems affecting the lives of large numbers of women’ (Diamond and Hartsock 1998: 198). In short, they reject Sapiro’s strategy of inclusion in favour of a strategy of reversal. They too argue for group representation, but the group in question is conceived as an identity group. This debate between strategies of inclusion and reversal (between the more effective representation of women’s interests within interest-group pluralism and the transformation of the representative system itself) is now subject to various feminist strategies of displacement. Both Young and Phillips take us beyond the either/or of previous debates. Phillips aims to synthesize the interest-based and identity-based approaches in a manner that she claims offers a less problematic basis for arguing for the representation of women. Young, who has a greater

Group representation and deliberation 97 methodological commitment to displacement than does Phillips, shares many of Phillips’s arguments for group representation, but eschews the language of pragmatic synthesis in favour of deconstruction. Phillips’s ‘rather commonsensical’ solution is to use both the terms ‘interests’ and ‘needs’ together (Phillips 1995: 73). There is a long history to this tension, as Phillips notes (Phillips 1995: 72). Both positions, she claims, have their strengths and weaknesses. ‘Interests can sound rather grasping and competitive, but it does at least serve to remind us that there may be conflicts between different groups. Need has more obvious moral resonance, but it originates from a paternalist discourse which lends itself more readily to decision by experts on behalf of the need group’ (Phillips 1995: 73). The first may be overly individualistic but there are contrasting worries that the second may be overly assimilatory. In an attempt to synthesize the best of these approaches and provide a firmer normative basis from which to consider the arguments for the increased representation of women, Phillips proposes a ‘politics of presence’. A politics of ideas is Phillips’s term for a politics that focuses on policies and a representation that focuses on people’s beliefs and interests. Fair representation is, on this ideas-based model of politics, realized in the ongoing responsiveness of representatives to those they are representing. The accountability of representatives to their electorate is therefore paramount. As long as they are responsive, it matters little who the representatives are: ‘the messages will vary, but it hardly matters if the messengers are the same’ (Phillips 1995: 6). A politics of presence, on the other hand, is Phillips’s term for a politics that also considers the role of the messengers themselves. Fair representation, on this conception of politics, requires that the overly cerebral concentration on beliefs and interests be extended to recognize the normative and political significance of the identity of the representatives. The gender (and any other social identity deemed politically significant) of the representative is here considered to be ‘an important part of what makes them representative . . .’ (Phillips 1995: 13). This endorsement of a politics of presence is controversial. Phillips considers three central objections. The first two originate from within the interest-based model of politics and the third from the participatory democracy model of politics (there being internal divisions within these camps as to whether group representation is the best way forward). First, there is the argument that such a politics poses a threat to national unity and leads to a ‘balkanization’ of the polity, or a politics of the enclave, in that it encourages intransigence rather than cohesion. Second, there is a concern that it undermines the basis for political accountability in that it is much harder to define clearly what a social group, as opposed to an interest group, really wants (what its interests are and whether they are being pursued). The third objection, made not by advocates of interest-based politics but by civic republicans and deliberative democrats, is that this is yet another capitulation to representative politics (albeit group-based rather than individualistic), which detracts from the pursuit of a truly inclusive and participatory politics of the common good (Phillips 1995: 21–4). We could also add a fourth objection, which arises from a more deconstructive concern that any

98

Judith Squires

institutionalization of group identity will work to reify and normalize identities in a manner that might then be used to re-subordinate the group in question. It is the second concern about accountability that Phillips takes to be the most serious in relation to debates about the political representation of women. As accountability ‘is best understood in relation to the politics of ideas’, it is essential that a politics of ideas is not jettisoned altogether in any move towards a politics of presence (Phillips 1995: 56). Accordingly, she argues that ‘It is in the relationship between ideas and presence that we can best hope to find a fairer system of representation, not in a false opposition between one or the other’ (Phillips 1995: 25). She also takes seriously the third concern about participatory democracy, and argues that arguments for group representation are at their strongest when placed in the context of wider arguments for participatory democracy (Phillips 1995: 145–65). The fourth objection, arising from the deconstructive approach to gender issues, is one that Phillips does not consider in detail. This pragmatic defence of group representation modifies the mechanisms of representative democracy in order to secure a greater parity of presence for women and moves ‘in close parallel with arguments for more participatory democracy’ (Phillips 1995: 190). Theorists grounded in a tradition of deliberative and participatory democracy are, Phillips feels, best placed to develop arguments for a politics of presence and for group representation, which avoid the pitfalls of the overly narrow arguments for group representation based on the traditions of interest-group pluralism or identity-based politics. Notably, Phillips thinks that Young’s particular vision of group representation ‘avoids most of the pitfalls in appealing to shared experience as an automatic guarantee . . . it makes no claims to essential unities or characteristics; it recognizes the potential diversity and disagreement within any social group; and it provides some basis for the accountability of representatives to those they might claim to represent’ (Phillips 1995: 54). Young has a clear commitment to a strategy of displacement. Hers is perhaps the most sophisticated and detailed attempt to articulate a feminist democratic theory from this perspective. It is also an attempt to integrate two features of recent democratic theory that are usually assumed to be mutually exclusive: group representation and deliberation. This, I think, is a particularly significant and under-explored feature of Young’s democratic model, and one that generates many interesting insights for both deliberative and feminist theorists.

Social groups Young advocates a form of democratic theory that is based on a ‘politics of difference’. She claims that existing electoral and legislative processes are ‘unrepresentative’ in the sense that they fail to reflect the diversity of the population, leading her to demand that a certain number of seats in the legislature be reserved for the members of marginalized groups. This call is made on the assumption that under-representation can be overcome only by resorting to guaranteed representation and that representing difference requires constitutional guarantees of group participation within the parliamentary system. Groups who

Group representation and deliberation 99 have suffered oppression need guaranteed representation in order that their distinct voice can be heard. Her claim is that a just polity requires the participation and inclusion of all groups, only secured by differential treatment for oppressed groups. This rejection of the assimilationist ideal is based in a belief that attachment to specific traditions, practices, language and other culturally specific forms is a crucial aspect of social existence. Young argues that ‘a democratic public should provide mechanisms for the effective recognition and representation of the distinct voices and perspectives of those of its constituent groups that are oppressed or disadvantaged’ (Young 1990: 184). These mechanisms will involve three distinct features. First, the provision of public resources, which will be used to support the self-organization of group members, ‘so that they achieve collective empowerment and a reflective understanding of their collective experiences and interests in the context of the society’ (Young 1990: 184). Second, the provision of public resources to enable the group to analyse and generate policy proposals in institutionalized contexts, and the formal requirement that decision-makers show that they have taken these perspectives into account. Third, group veto power regarding specific policies that affect a group directly, ‘such as reproductive rights for women’ (Young 1990: 184). Rather than transcending particularity Young proposes that ‘. . . attention to social group differentiation is an important resource for democratic communication’ (Young 1997: 385). But social groups are not to be confused with interest groups. Young is keen to point out that these groups should be understood in relational, not essentialist, terms. The social groups she would positively recognize in her vision of differentiated citizenship are products of social relations and are therefore fluid and intersecting. Young employs a relational interpretation of difference: ‘. . . the social positioning of group differentiation gives to individuals some shared perspectives on social life’ (Young 1997: 385). Social groups are neither ‘any aggregate or association of persons who seek a particular goal, or desire the same policy . . .’ nor ‘a collective of persons with shared political beliefs’ (Young 1990: 186). This attempt to unsettle the dichotomy between interest groups and identity groups is indicative of her commitment to the strategy of displacement. It is not always clear though that she succeeds in distinguishing her preferred concept of social groups from the interest and identity groups she rejects. Some commentators have confused her position with identity politics (Elshtain 1995), others argue that Young’s vision is actually nothing more than a rearticulation of interest-group pluralism (Mouffe 1992: 369–85). But these are both charges that Young rejects. She rejects identity groups, insisting that what makes a group is not internal to the attributes and self-understanding of its members but the relation in which it stands to others (Young 1997: 389). Much of the writings about a politics of difference has itself explored ‘problems with a politics of difference as the positive assertion of group identity, and has often itself argued against a politics of identity’ (Young 1997: 397). On the other hand she claims that her vision of group representation is fundamentally different from interest-group pluralism in that it promotes public discussion and decision-making rather than the pursuit of

100

Judith Squires

pre-defined interests (Young 1990: 186–90). In other words, she avoids the problems of essentialism and unaccountability, which arise with identity-group politics, only to be charged with adopting a form of interest-group pluralism that is itself problematic. Her attempts to distinguish her position from interest-group pluralism then propel her to invoke norms of just deliberation that arise from a deliberative democracy framework which she has, on other occasions, criticized for being overly universalistic. Her argument is that interest groups simply promote their own interests in a political marketplace, with no reference to a conception of social justice or the common good. The social groups argued by Young to require special representation are, on the other hand, defined with reference to a specific vision of justice which generates criteria for assessing social oppression and hence criteria for establishing which groups require such representative guarantees. This vision of justice offers guidance not only about which groups require special representation rights, but also about how they should act in the political realm. A distinction is made between demands stemming from self-interest and those stemming from justice: ‘the test of whether a claim upon the public is just or merely an expression of self-interest is best made when those making it must confront the opinion of others who have explicitly different, though not necessarily conflicting, experiences, priorities and needs’ (Young 1990: 186). In other words, the engagement in deliberation with other social groups differentiates a just political dialogue from a simple expression of instrumental interest. Deliberation and group representation are drawn together in Young’s distinctive model of democracy. Her vision is significant in that it represents an attempt to incorporate elements of recent deliberative and feminist theories, focusing attention on issues currently marginalized in the respective literatures. There are undeniable benefits to such a strategy, yet certain tensions remain. In her attempt to defend a model of group representation as a strategy of displacement rather than inclusion or reversal, Young develops a notion of social groups. To do this she needs to appeal to a conception of justice that privileges deliberation: but this presents a problem. The model of deliberative democracy that she invokes assumes an ideal of impartiality that she has vocally rejected. In order to get around this paradox she appeals to a notion of objective judgement. Whereas the concept of social groups displaces the apparent dichotomy between interest groups and identity groups, the concept of objective judgement displaces the apparent dichotomy between impartiality and particularity. The notion of social groups is more developed than that of objective judgement, which we shall consider next. But the former is not, I think, viable without a fully articulated defence of the latter.

Objective judgement Young is openly hostile to the ideal of impartiality, which she depicts as offering a conception of the self as ‘dispassionate, abstracting from feelings, desires, interest, and commitments that he or she may have regarding the situation . . .’ (Young

Group representation and deliberation 101 1990: 100). She claims that it denies difference in three ways. It denies the particularity of the situation. It seeks to ‘eliminate heterogeneity in the form of feeling’. It reduces ‘the plurality of moral subjects to one subjectivity’ (Young 1990: 100). On the other hand, the ideal of community or shared subjectivity also denies difference. On this conception people become fused, mutually sympathetic, understanding one another as they understand themselves. This, she claims, denies difference in the sense of the basic asymmetry of subjects. Neither option is adequate. Both are generated by an ideal of impartiality, which dichotomizes reason and feeling. Young claims that the ‘modern’ or ‘Enlightenment’ ideal of impartiality seeks to reduce differences to unity ‘by abstracting from the particularities of situation, feeling, affiliation, and point of view’ (Young 1990: 97). The ideal entails a vision of the public realm as attaining the universality of a general will by jettisoning all particularity to the private sphere. In practice, this was achieved by the exclusion from the public realm of those groups perceived to embody particularity ‘especially women, Blacks, American Indians, and Jews’ (Young 1990: 97). In the pursuit of a single, universal set of principles to govern the public realm, complex difference is necessarily repressed, paradoxically creating dichotomy instead of unity. If the citizen is understood to be a ‘universal reasoner’, detached and impartial, he or she must abstract from the ‘partiality of affiliation, of social or group perspective, that constitutes concrete subjects’ (Young 1990: 100). The result is that members of minority social groups are either excluded from citizenship or included only to the extent to which they are able to repress the particularity of their identity. Given that the universalist ideal ‘continues to threaten the exclusion of some’, argues Young, ‘the meaning of “public” should be transformed to exhibit the positivity of group difference, passion and play’ (Young 1990: 97). Young finds both the liberal and the civic-republican traditions equally guilty of projecting ‘an ideal of universal citizenship’ (Young 1998: 401). She, like Habermas, warns against simply replacing an impoverished conception of liberal citizenship with a civic republican conception, which provides the notion of the general will (Habermas 1996a). The civic-republican tradition may stand in ‘critical tension’ with individualistic contract theory, but shares a common commitment to universalism. And this commitment necessarily entails the exclusion of all groups that threaten to explode the unity of the polity (Young 1998: 404–5). One would expect a displacement theorist to argue that Habermas’s politics, formulated in the name of critical theory and an emancipatory project, excludes otherness. The model of deliberative democracy built on Habermasian discourse ethics is not sympathetic to the notion of alterity. Yet it is precisely this notion that underpins Young’s defence of group representation. Young states that the ‘ideal of impartial moral reason corresponds to the Enlightenment ideal of the public realm of politics as attaining the universality of a general will that leaves difference, particularity and the body behind in the private realms of family and civil society’ (Young 1990: 97). Contra this ideal, she suggests that ‘the ideal of the civic public as expressing the general interest – the impartial point of view of

102

Judith Squires

reason – itself results in exclusion. By assuming that reason stands opposed to desire, affectivity and the body, this conception of the civic public excludes bodily and affective aspects of human existence’ (Young, 1990: 109). Yet, intriguingly, Young does not reject discourse ethics altogether. We have seen that she needs it to sustain her definition of social groups. There is something interesting about the fact that Young advocates both group representation and deliberation, and holds them to be compatible, not competing, democratic mechanisms. She argues that even though Habermas ‘seems unwilling to abandon a standpoint of universal normative reason that transcends particularist perspectives’ he ‘has gone further than any other contemporary thinker in elaborating the project of a moral reason that recognized the plurality of subjects’ (Young, 1990: 106). Her critique of deliberative democracy clearly emerges from her rejection of the ideal of impartiality. But, as republican theorists have noted, the attempt to distance herself from the individualism and instrumentalism of interest-group pluralism actually propels Young towards an endorsement of a form of impartiality and deliberation that she finds unacceptable in republicanism (Miller 1993b: 16). Indeed, as Phillips notes (Phillips 1995: 147), whereas Young’s initial formulation of her argument for group representation relied on heavily criticizing deliberative democracy, she now uses this framework, in a slightly modified form that she labels communicative democracy, to defend her own vision of group representation (Young 1996: 120–36). She recommends that the forms of communication considered significant in debates about justice be extended to include greeting, rhetoric and storytelling. Greeting entails non-linguistic gestures, such as smiles and handshakes, which bring bodies into communication. Rhetoric entails humour, wordplay, images and metaphors, which bring desire into communication. Storytelling entails narratives that exhibit subjective experience and evoke sympathy, which brings experience into communication (Young 1996: 129–32). In short, she claims that there is an alternative to moral theory founded on impartiality, and this is labelled communicative ethics. Young is convinced that dialogue between social groups must take place. Such dialogue across difference serves three functions. First, it motivates claimants to express their proposals as appeals to justice rather than expressions of mere selfinterest. Second, confrontation with different perspectives teaches individuals the partiality of their own. Third, expressing, questioning and challenging differently situated knowledge adds to social knowledge. ‘Public critical discussion that includes the expression of and exchange between all relevant differentiated social perspectives transforms the partial and parochial interests and ideas of each into more reflective and objective judgement. . . . Judgement is objective in this sense when it situates one’s own particular perspectives in a wider context that takes other perspectives into account as well. Objectivity in this sense means only that judgement has taken account of the experience, knowledge, and interests of others. Such objectivity is possible only if those particular perspectives are expressed publicly to everyone’ (Young 1997: 402). There are, Young allows, conditions required to reach a just decision (which she declines to label a revised ideal of impartiality), but these are weaker than those

Group representation and deliberation 103 proposed in discourse ethics. They are simply a recognition of significant interdependence, a commitment to equal respect for one another, and agreement on procedural rules of fair discussion and decision-making (Young 1996: 126). These conditions are, she claims, much thinner than that of shared understanding. In other words we find that by incorporating a conception of communicative ethics into her model of group representation, Young aims to distinguish her vision of democracy from both interest-group pluralism and universalist civic republicanism. To do so she develops the concepts of social groups and objective judgement. Whether this ideal of communicative ethics really stands in opposition to impartiality, or simply articulates a form of impartiality that attempts to be attentive to alterity, is however worth considering. For instance, Young’s response to the question of how interest-seekers will be transformed into citizens who attend to the claims of others is sketchy. She addresses the dilemma by quoting Hannah Pitkin, who argues that interest- group competition draws us into politics because ‘we are forced to find or create a common language of purposes and aspirations . . . we are forced to transform “I want” into “I am entitled to”, a claim that becomes negotiable by public standards’ (quoted in Young 1990: 107). Young then goes on to add: ‘In this move from an expression of desire to a claim of justice, dialogue participants do not bracket their particular situations or adopt a universal and shared standpoint. They only move from self-regarding need to recognition of the claim of others’ (Young 1990: 107 italics added). The crucial question of precisely how this transformation occurs is not directly considered, nor is the possibility that this might effectively be an appeal to impartiality. As a result Young’s notion of objective judgement appears unstable. The instability arises from the attempt to synthesize the competing demands of group representation, which is necessarily partial, and deliberation, which aspires to be impartial, into a single model of democratic practice. My sense is that Young can only resolve this tension if she accepts that her defence of group representation is itself grounded in a commitment to impartiality. Acknowledgement of this fact would allow feminist theorists who are pursuing a strategy of displacement to be more sympathetic to the theoretical underpinnings of political liberalism. There is value then in being forced, as we are in the work of Young, to confront this theoretical instability.

Conclusion I have argued that the most significant innovation in democratic theory to emerge from the recent feminist literature is the endorsement of group representation. Feminist arguments for group representation are made from three distinct frames: inclusion, reversal and displacement. Young adopts a strategy of displacement to argue for group representation. Her conception of a group is neither an interest nor an identity group. She displaces this dichotomy with appeal to a relational notion of social groups. She also seeks to displace the dichotomy between impartiality and particularity by introducing a notion of objective judgement. Significantly the former move is much more compelling than the latter. For it is in

104

Judith Squires

the area of ontological debate, rather than in advocacy issues, that feminist theory has developed the most useful critical resources. The literature on what objective judgement, as opposed to impartiality or particularity, might involve remains under-developed. If a feminist strategy of displacement is to engage fully with advocacy as opposed to simply ontological issues, this needs to be addressed. The literature on deliberative democracy has, in contrast, focused on questions of justice as impartiality, but has largely failed to acknowledge the challenge of alterity or to consider more practical issues of equal access to institutional decision-making procedures. Attempts to locate deliberation within the democratic polity vary (see Saward’s chapter, above, for one mapping exercise). There are significant differences in the literature. Self-proclaimed deliberative democrats disagree as to whether deliberation should be located within representative or unrepresentative arenas. Proposals for informal but representative models of deliberative democracy are found in the advocates of deliberative opinion polls and citizens’ juries (see respectively Fishkin and Luskin, and Smith, this volume). Proposals for informal and unrepresentative deliberation are found in advocates of associations (Cohen and Rogers 1995a; Benhabib 1996a; Dryzek 1990a). Of those commonly categorized as a deliberative democrat only Rawls appears to propose locating deliberation within a formal institution: the supreme court. Deliberative democrats largely avoid locating deliberation within the formal, representative structures of democracies because they fear that the constraints imposed by such institutions will render the conditions for deliberation unsustainable. They purposely locate deliberation outside the formal representative arenas of politics, aiming to inculcate the skills of deliberation more widely in society (see the discussions by Rättilä and Dryzek, above). What they do, paradoxically, is to turn their attention to the everyday practices of democracy that early second-wave feminists did so much to prioritize. They do this at a time when feminist democratic theorists seem to have moved on and shifted their sights to the formal institutional elements of democracy. The traditions of feminist and democratic theory are not yet unified, indeed they seem to be resolutely out of sync. The optimistic response to this discontinuity would be to recommend that democratic theorists pay more attention to recent developments in feminist theory in order to learn from its political insights and theoretical innovations. Young is unique in proposing that deliberation be located within the formal and representative institutions of decision-making. In so doing she is perhaps the most optimistic of deliberative democrats. Her attempt to integrate group representation and deliberation into the formal procedures of democracy is particularly challenging given that one of the most frequently voiced concerns about group representation is that it encourages intransigence rather than cohesion. Deliberation on the other hand emerges as a valued democratic ideal precisely because it is viewed as a way of countering the inevitable intransigence that emerges from the representative form of democracy. With the exception of Young deliberative democrats have proposed that the representative and deliberative aspects of democracy be kept procedurally and institutionally distinct. There are

Group representation and deliberation 105 clear pitfalls in Young’s more integrated vision, which may ultimately render it unrealistic, but it nonetheless issues a provocative challenge to advocates of both group representation and deliberation. If Young’s particular vision is unstable, what alternative relation between representation and deliberation is envisaged? Advocates of group representation need to be more explicit about how representatives might balance the competing demands of accountability and justice. Deliberative democrats need to be much more explicit about the structural role and institutional location of deliberation within the decision-making process.

8

From theory to practice and back again Gender quota and the politics of presence in Belgium Petra Meier

If representative democracy is not representative enough, two options for radical reform suggest themselves: increased reliance on a mechanism of random sampling (see Fishkin and Luskin, and Smith, this volume), or the establishment of very precise rules on who has to be represented. Random sampling might provide a solution to overcome the selectivity inherent in many electoral systems. As long as the sample is large enough, random sampling leads to what Pitkin (1967) calls descriptive representation, reflecting the precise composition of the population, without having to deal with the difficult normative issue of why specific social identities1 have to be present. If we accept that random sampling would not involve a problem of accountability of the incumbent towards the citizens,2 there remains the issue of the functioning of the political process. If extensively and formally used, random sampling would severely undermine political parties. Parties not only place political problems in a broader societal and ideological framework (see Budge, this volume). They also provide structures to ‘bundle’ individuals and to establish a certain continuity. Random sampling might provide a solution for the issue of representativeness in itself. The idea is appealing and experiments lead to exciting results, but it is no short-term alternative to the existing parliamentary representative establishment, given the radical change involved if descriptive representation were to become the norm. Standardly, representation is organized and channelled through political parties. A randomly sampled group of individuals representing the nation has no link with the actual system and it is unreasonable to believe that the existing parliamentary assemblies will be eradicated and based on a completely different selecting mechanism. Furthermore, random sampling does not provide a solution for a major problem of actual representative institutions: the under-representation of structural minorities. Random sampling will lead to no more than a symbolic presence of minorities in parliament in proportion to their number. If the voice of minorities has to be heard, then the establishment of very precise rules on who has to be represented in what number is a more potentially effective solution. A growing body of literature deals with the issue of why and how to represent structurally marginalized social identities, often focusing on the deliberative aspect of the political process and on the need for members of the marginalized group to be present within it (Kymlicka 1995; Phillips 1995; Williams 1998; Young 1990).

Theory to practice and back again

107

For instance, Phillips argues that the politics of ideas, dominant in modern liberal democracies, has been challenged and that there is a shift away from ideas to identity. She pleads for a fairer system of political representation where presence, conceived as a combination of both ideas and identity, is the central issue. This literature is mainly Anglo-American, which means it evolved in societies with electoral systems that tend to have strong polarizing effects and to that extent to be rather unrepresentative. This chapter does not provide an exhaustive analysis of that literature. Rather, I start at the other end, looking at a concrete case of representing social identities. No matter how important theory may be, models of democracy should not be reified. Democracy evolves in the concrete, not the abstract. Analysing the Belgian gender quota law of 1994, I argue that guaranteeing the representation of social identities is a commonly used mechanism in the Belgian political system. It shows that what seems to be a very recent current in political theory has long been at the heart of the Belgian conception of political representation. There has been no shift from ideas to identity, or to the notion of presence; a combination of ideas and identity have long underlain the Belgian system of representation. The Belgian example illustrates how a society concretely deals with ‘presence’ without problematizing the representation of social identities. In that respect it is a useful case to back up the Anglo-American literature defending political ‘presence’ based on social identities. Next to this I argue that, even if there is no shift from ideas to identity, there is a shift within the concept of social identity. Before turning directly to the Belgian case, let me say a few words on the literature generally.

Arguing for enhanced representativeness One of the main features of modern democracies is their representative character. There may be elements of direct participation in many systems, but they are very much secondary to indirect participation. Citizens send their representatives to parliament on a regular basis. Given this lack of direct participation, the issue of adequate representation is all the more important, and has received a great deal of attention in political theory. For some time the key question has been: on what grounds should we select those who represent and rule? Universal suffrage formally solved this problem, leaving the choice to the individual citizen. The rather select character of most legislative assemblies has, however, put the issue back onto the agenda. This has led to a search for mechanisms to enhance the representative character of politics. Especially interesting is the development of theoretical foundations explaining the need for various social identities to be present in parliament. In this chapter I focus on Phillips’s politics of presence (1995), because her argument for presence is based on the activity of representation itself. Phillips states that traditionally a distinction has been made between two concepts of representation, the ‘politics of ideas’ and the ‘politics of presence’, i.e. the representation of opinions or beliefs and the representation of social characteristics or identities. Liberal democracy mainly relies on a concordance of voters’ and their representatives’ ideas, and the representatives’ opinions and

108

Petra Meier

beliefs are detached from his or her personal experience and characteristics. The politics of ideas is challenged by the recognition that certain groups are politically marginalized because of their social identity, e.g. in terms of sex, race or ethnicity. Phillips rejects both a pure politics of identity and a pure politics of ideas in order to overcome this marginalization. Whereas the latter seemed no adequate answer to these problems, the former does not in itself provide an attractive alternative (see Squires, previous chapter in this book). The issue of political equality can only be dealt with in terms of a politics of presence, understanding and respecting the relationship between ideas and identity. The mistake generally made is that the two concepts of representation are conceived as antagonistic. ‘Most of the problems, indeed, arise when these two are set up as exclusionary opposites: when ideas are treated as totally separate from the people who carry them; or when the people dominate attention, with no thought given to their policies and ideas’ (Phillips 1995: 25). It cannot be assumed that opinions, beliefs and interests shaped in that framework are objectively given and exist independently of those who hold them. One point is that personal experiences and opinions shape ideas. The other point is that interests are formulated in the political process. The more clear-cut interests are, the less it matters who defends them. If voters could choose among issues that were precisely defined beforehand it would not matter whom the representatives are as long as they defend the voters’ interests. If issues have to be shaped or even detected in the process of political decision-making, and if the personal experience or opinion of those doing it influences this, then identity plays a role. Here the notion of descriptive representation, which is irrelevant in the setting of a politics of ideas, enters the picture and the priority of a politics of ideas is questioned. To address criticisms aimed at identity-based representation, Phillips emphasizes the point that the need for presence is no plea for an essentialist form of group representation. Individuals do not need to be represented by someone sharing their social identity. Presence is a prerequisite to influencing and changing the political agenda, and the presence of representatives with a specific personal experience or social identity is a supplementary guarantee for the representation of certain interests. In that sense a politics of presence does not undermine the deliberative process as such. On the contrary, deliberation is central to Phillips’s argument. The politics of presence is emancipatory to the degree that it fosters the intervention of formerly excluded social groups in the framing of issues, since challenging the way issues are framed can mean challenging prevailing social relations and structures. It is precisely this emancipatory aspect which prevents the locking of representatives into a static, essentialist identity. The evolution of the definitions of interests and the potential for societal change arising out of this undercut an essentialist definition of identity.

Mechanisms to improve representativeness Many societies have put in place mechanisms designed to enhance the representativeness of parliamentary assemblies. These mechanisms tend to correct

Theory to practice and back again

109

imbalances by guaranteeing political representation to specific social groups. Traditionally such guarantees focus on regional, cultural or ethnic identities, paying special attention to the protection of minorities. Reserved seats such as those established for the Maori people in New Zealand (Fleras 1985), the sectarian composition of the Lebanese chamber (Lijphart 1986) and affirmative gerrymandering for Afro-Americans and Hispanics in the United States (Kymlicka 1995) are examples. More recent is the trend to apply gender quota to electoral lists, either by political parties, as in Scandinavia (Sundberg 1995), or through a legislative measure, as in the Argentinian gender quota law ( Jones 1996). In the (former) socialist regimes and in young democracies specific attention is also paid to criteria such as age; Uganda, for example, reserves a fixed number of seats for young representatives (Broere 1998). These mechanisms facilitate access to or guarantee presence in parliamentary assemblies for specific social identities. In a report on mechanisms to facilitate women’s access to parliamentary assemblies, the Interparliamentary Union (IPU 1992; 1997) distinguishes between quota, reserved seats and appointment, but these categories are not watertight. Reserved seats are a form of quota (Dahlerup 1998), and both may involve elections (for instance the debates on double lists in France in order to achieve parity democracy; see Mossuz-Lavau 1998). There is, however, an important difference between quotas applied to electoral lists and reserved seats in the sense that the latter guarantee a specific outcome. Therefore, I prefer to distinguish between the levels at which guarantees of representation are applied rather than between the mechanisms themselves. Speaking in terms of political representation, and more precisely parliamentary representation, one can distinguish mechanisms applied to the electoral process and those applied to the electoral outcome. The first concern the proportion of candidates running for legislative seats, the second the proportion of incumbents actually present in the legislature. While mechanisms applied to the electoral process theoretically do not determine who will be elected so as not to violate the meaning of electing,3 mechanisms applied to the electoral outcome can guarantee a fixed numerical presence of specific social identities in parliament. Therefore, I also distinguish between mechanisms facilitating representation and those guaranteeing representation. The first might facilitate the presence of certain social identities, but do not contain a commitment in terms of outcomes. Certain mechanisms applied to the electoral outcome can, however, guarantee a predetermined (minimum) result. Table 8.1 maps out the different mechanisms.

Belgium as a laboratory for guarantees of representation4 Having schematized different mechanisms, let us now take a closer look at the Belgian case. Constitutional reforms in 1970, 1980, 1988 and 1993 turned Belgium into a federal system, comprising a federal upper and lower house, and five regional parliaments. The federal upper house has fixed ratios of seats for the two major Flemish and French-speaking linguistic communities, to be allocated

110

Petra Meier

Table 8.1 Guarantees of representation Mechanisms facilitating presence

Mechanisms guaranteeing (minimal) presence

Mechanism applied to the electoral process

• Quota applied to electoral lists

• (Theoretically non-existent)

Mechanism applied to the electoral outcome

• Changing size and boundaries of circumscriptions

• Reserved seats

• Neutralizing the list vote in a proportional list system

• Bi-nominal tickets imposing the choice of candidates with a specific social identity • Appointing incumbents with a specific social identity

through direct election and appointment. One more seat is reserved for a nominated representative of the German-speaking minority. The parliament of the Brussels Capital Region contains various mechanisms to facilitate the presence of Flemish incumbents. The assembly comprises a relatively large number of representatives in order to leave more room for the (smaller) Flemish parties. Furthermore, the number of votes won by the various parties are pooled at the level of the linguistic community so as to maximize the potential number of seats for the Flemish minority (Mares 1999). Also to be mentioned is the recently heightened debate on reserved seats for the Flemish in the parliament of the Brussels Capital Region and in the Brussels local councils. At both levels the Flemish have about 15 per cent of the seats and are, numerically speaking, not under-represented. Nonetheless, some Flemish politicians claim a guaranteed Flemish representation in the Brussels local and regional representative assemblies, especially in the context of granting a voting right to EU (and eventually other non-Belgian) residents. Fearing that the new electorate would mainly vote for French-speaking parties, Flemish politicians are unwilling to apply the EU directive awarding EU residents a right to vote in local elections in their country of residence without guarantees for a Flemish presence. However, research does not confirm that this fear is well-founded (Bousetta and Swyngedouw 1999). Similar guarantees are requested at the regional level, mainly through Flemish fears of the electoral strength of second- and third-generation naturalized migrants. Leaving aside the issue as to what extent the Flemish request can be defended, the point is that this is an attempt to obtain guarantees of representation based on language. The traditional major parties – the Christian Democrats, the Socialists and the Liberals – have long respected balances between the major organizations in civil society belonging to the same ideological pillar when it comes to the composition of their electoral lists (De Winter 1988). These balances are not formalized, and

Theory to practice and back again

111

the younger parties pay less attention to them, in some degree due to the fact that they are part of a less well organized social subculture or ideological pillar. More recently, parties have given attention to social identities in terms of age, ethnicity and sex. The Flemish ecological party, Agalev, applies a non-formalized quota for young candidates. Most other parties also show their interest in young people (though little complementary interest is shown to senior citizens). Belgian nationality is still a prerequisite to stand for elections and, therefore, the notion of ethnicity refers only to Belgian citizens of foreign origin. During the 1994, 1995 and 1999 elections candidates with foreign origins were encouraged to stand. This is partly an electoral strategy to attract the votes of second- and thirdgeneration Turkish and Moroccan migrants who obtained Belgian nationality and in its wake the obligation to vote (Van de Putte 1997). Even if it is not inspired by a concern to adequately represent society, this strategy reflects the idea that the electorate identifies with candidates sharing (part of) their social identity. Several parties applied a gender quota until a quota law over-ruled such initiatives in 1994. The Flemish Christian Democrats were the first to introduce a gender quota for their electoral lists. From 1974 onwards at least one woman had to obtain an eligible place on the lists. In 1993 a one-third gender quota was rejected in favour of a rule gradually raising the percentage of women candidates to 50 per cent. Both the Flemish and the French-speaking Greens have long applied a 50 per cent quota. This quota took the form of a zipper list, i.e. an alternation of candidates of both sexes over the whole list. The other parties did not have any rules to enhance the presence of women, considering such mechanisms to be incompatible with their ideological or philosophical positions (Van Molle and Gubin 1998). The existing rules had little binding force. They guaranteed a minimal presence of women and not an equal representation of both sexes, even where this was required. Therefore, Belgium adopted a law imposing a gender quota for electoral lists in 1994. Parties in various countries apply a gender quota to their electoral lists, but few countries enforce it by law. Belgium is the only one in Western Europe, after Italy dropped its gender quota law in 1995 (Guadagnini 1998). Other examples are Argentina, Bolivia, Brazil, Ecuador, Peru ( Jones 1998), and the Democratic People’s Republic of Korea (IPU 1997). The Belgian law stipulates that electoral lists may contain a maximum of twothirds of candidates of the same sex. Parties that do not manage to fill up at least one-third of the places with candidates of the under-represented sex, have to leave these places open. Otherwise the lists do not get approved. The initially proposed version of the law not only specified a maximum of two-thirds of candidates of the same sex for the entire list but further defined such a quota for the sum of the safe and combative seats, as well as of the first successor’s place.5 Due to disagreement among the Ministers on the exact contents of the law, a working group was set up comprising the party chairmen of the governing coalition. This group considered that the law interfered too much in the parties’ strategic privilege to compose the lists and dropped the stipulation of how to spread candidates of both sexes over the electoral lists (Carton 1995). During the parliamentary debates on the law, amendments were introduced in order to define such a spreading6 but

112

Petra Meier

none of them were passed (Uyttendaele and Sohier 1995). The quota was first applied during the June 1999 elections. Even if the law did not stipulate where to place women on the lists, there had been so much debate on the issue that parties were careful to put forward female candidates for safe seats, and the proportion of women running for parliament rose considerably. Albeit less important, a rise of the proportion of women elected was also achieved. So the Belgian gender quota law is, at least in the short run, an effective mechanism to facilitate women’s presence in parliament. This overview shows that the idea of facilitating or guaranteeing presence is not alien to the Belgian political context. All formal mechanisms guaranteeing presence concern social identity in linguistic terms, but so far they are only applied in the federal upper house. The mechanisms applied to the parliament of the Brussels Capital Region also concern the electoral outcomes, but they only facilitate presence. Recently, there have been calls for formal guarantees of the presence of the Flemish minority in Brussels. All the other rules mentioned operate at the level of the electoral process. Parties have long paid attention to the representation of various intermediary organizations belonging to their ideological pillar, but this concern is not necessarily translated into formal agreements. More recently parties pay attention to new forms of social identity, focusing on female or young candidates or on candidates of a non-Belgian origin. Gender is the only social identity for which formal rules exist. What makes the gender quota so interesting is the fact that rules existing at the party level were raised to and formalized at the legislative level.

Recognizing gender as part of social identity When we consider this case from the point of view of the concept of representation, several questions arise. The introduction of the law can be understood as a sign that the Belgian government, the parliamentary assemblies and the political parties all consider representation in terms of gender to be important. But what does this new mechanism to facilitate women’s presence tell about the prevailing concept of representation in Belgium? There and elsewhere, how important is the representation of identities compared to the representation of ideas? Does the gender quota law indicate that our concept of political representation is changing, that we are moving from a politics of ideas to a politics of presence? Or does it mean that within a politics of presence the notion of social identity is changing? In order to answer these questions let us look at the parliamentary debates on the gender quota law, what they tell us about the concept of representation underlying the law, and how this relates to the prevailing concept of representation. Authors such as Marques-Pereira (1998) and Paye (1997) who have analysed the parliamentary debates on the quota law agree that there was a general consensus on the importance of enhancing the presence of women in politics; even for the conservative parties there was no real cause for debate. A gender-balanced (be it 50–50 or a threshold) political representation was considered likely to lead to better representation, in terms both of form and of content. More women in

Theory to practice and back again

113

politics reduces the gap between the political representatives and the citizens, and enhances the representative character of democracy, which in turn increases the legitimacy of the legislative assemblies. The debates clearly show that Belgian political society conceives sex as a social identity requiring specific political representation. Instead of focusing on the relevance of sex as a social identity to be present in parliament, the debates turned on the issue of whether quota would be an adequate tool to increase female presence. This debate might seem less interesting from the point of view of the concept of representation, but arguments on the adequacy of the tool might encompass suppressed arguments against the relevance of sex as a social identity. These arguments either considered the law to be too weak or they defended another mechanism, such as the neutralization of the list vote. The Greens consider quotas as such to be the adequate mechanism but nonetheless rejected the law. Perceiving equality in terms of outcome they argued that the quota law was not far-reaching enough. They defended a 50 per cent quota, in the form of a zipper list. Such a precise definition of the quota would, given the Belgian proportional list system, indeed guarantee a minimal electoral outcome (namely in all the cases where a party wins more than one seat per constituency). The Liberals as well as the Flemish regionalists and extremeright parties rejected any structural mechanism to guarantee the representation of social identity in terms of sex. The Liberals preferred the neutralization of the list vote, which means that only the votes obtained by the individual candidates and not the list order are taken into consideration when it comes to the allocation of seats to the candidates of a list. Such a mechanism would not only restore the full weight of the voter’s choice. It would also enable the representation of other social identities without having to establish an endless list of quotas. This argument was not meant to express disagreement on the relevance of gender in terms of political representation. It rather reflected the liberal conception of equality and of the role of the state (Marques-Pereira 1998). Finally, the Socialists and Christian-Democrats on both the Flemish- and French-speaking sides defended the law. Framing the whole discussion in terms of formal versus substantial equality and the role that the state should play in this context, the debate did not question the concept of political representation as representation of identities. Why was political representation in terms of gender so easily admitted? Paye (1997) seeks the answer in the history of the Belgian state and nation, its consociative character and pillarized structure. It is not simply that the plurality of social identities is a key element of the Belgian political scene. In addition, it is recognized that social identities can only be represented by those who share that social identity. Even if the degree of pillarization of Belgian political and civil society is decreasing, it still implies the conception of citizenship as embedded in various particular social identities and the need for a balanced representation of these identities. Arguing for a representation of both sexes from the point of view of their numerical equality and their different social identity relates directly to the basic feature of a consociative society, namely the sharing of power between the various recognized social identities. Paye’s argument is interesting because it links

114

Petra Meier

the issue of representation to the broader Belgian political model. Phillips (1993) considers consociationalism to be one of the ‘mainstream’ traditions dealing with a form of identity politics. Contrary to Paye she argues that the model of consensus democracy only takes into account groups or segments of society embodied in political parties. Other segments are neglected, because consociationalism mainly deals with the accommodation between political elites and not with democracy itself. Hooghe (1998), analysing the way in which the Belgian system has reacted to the feminist and green new social movements, refines this idea. He concludes that it integrates the ideas of these movements but tends to exclude their political actors. It is true that the feminist party set up in the 1970s, and the seniors’ party of the beginning of the 1990s, did not break through electorally, whereas the issues of representation in terms of sex and age have been taken up by other parties. Leaving aside the question as to why identity in terms of gender and age is channelled within the existing parties and not by specific parties, as well as the discussion on the precise degree of openness of the Belgian consociational model, the gender quota law reflects three points. It symbolizes the point that political representation in terms of social identity is conceived as a legitimate claim. It reveals the will to institutionalize such claims. And it shows that the notion of social identity is not fixed.

Social identity, guarantees of representation and presence Belgium is an example of a society where political representation in terms of presence, as an interplay of ideas and identity, has always been important. The presence of specific groups, representing various social identities, is a basic characteristic of the Belgian system of political representation. The issue of linguistic presence has been explicitly institutionalized (with the exception of the case of Brussels). The notion of presence is also distinguishable at the party level itself. Within the framework of a broad but coherent ideological setting diverging groups persist which claim and receive specific representation. What is evident is that the notion of the features that should be politically represented is changing. Presence in linguistic, socio-economic, religious, and philosophical terms is being challenged by presence in terms of sex, age and ethnicity. The presence of the existing features is not necessarily losing importance. Even if it is only at an informal level, existing agreements to guarantee representation are still respected. But alongside them new features are emerging. Sex, age and ethnicity are receiving a growing amount of attention in the framework of political representation, even if sex is the only feature for which formal rules have been set up. Phillips (1995) mentions that the accentuation of difference in terms of sex, race and ethnicity is taking place at a moment when absolute differences are losing weight or are becoming less apparent. She seeks the answer to this paradox in the increasing importance attached to identity: ‘The attention currently directed to sexual and other kinds of difference cannot be understood just in terms of an

Theory to practice and back again

115

absolute or growing difference. More precisely, it reflects a shift in political culture and claims, where people who may be significantly less different than at some point in the past come to assert a stronger sense of themselves and their identities’ (Phillips 1995: 12). I would subscribe not so much to the idea that people develop a stronger notion of identity, but that the perception of (social) identity is changing. Bennett (1998), albeit in a different context, states that the notion of collective identities is being replaced by what he calls a more personal identity. Important structural changes at the socio-economic level and the insecurities emerging from a changing social organization of work make it increasingly difficult for citizens to identify with broader political projects. Instead they focus on more personal projects and it is at this level that they identify themselves in the political arena. Returning to the Belgian scene, this idea of a shift from a collective to a personal identity rather than the idea of a strengthened sense of identity helps to explain the current focus on new features emerging in the shaping of political representation. Where does all this lead us? What is the appeal of the Belgian gender quota law beyond the issue of gender or beyond the Belgian context? The Belgian example questions the so-called dominance of a politics of ideas. At least in the Belgian case it could be said that there is no shift of a politics of ideas to a politics of presence. Presence is a characteristic that has long been inherent to the Belgian system of political representation. The Belgian gender quota law is a further example of this prevailing conception of political representation. The Belgian case shows that in certain political cultures the notion of presence is not only accepted but also institutionalized, without endless discussions on the danger of essentialist group representation, on the lack of accountability or on loss of the content of the act of representing. It is evident that the specific historic context and development of the nation and of its institutions plays a key role, and that the Belgian example is a specific case. However, analysing systems in more detail might reveal that a certain notion of group representation underlies many systems. A politics of presence might be a feature of more political systems than opponents of group representation would like to believe. The Belgian example also shows that the notion of social identity to be represented is flexible. What should be kept in mind and where further research might be useful is to what extent guarantees of representation that arise out of a specific need for presence are conceived in absolute terms or whether they are conceived as temporary mechanisms that can be adapted to evolving needs.

Notes 1 I prefer this term to what is often referred to as social groups. 2 Randomly sampling representatives involves the acceptance of the following assumptions: the presence of every person likely to be selected makes sense, and every person likely to be sampled is capable of occupying the function of representative. Given the fact that representation then focuses on the notion of who you are and that everybody’s selection is equally likely, the issue of accountability becomes redundant. 3 However, good quota rules as mechanisms applied to the electoral process can have a

116

Petra Meier

determining impact on the final composition of parliamentary assemblies (cf. Jones 1998). 4 The present chapter only deals with mechanisms facilitating or guaranteeing parliamentary presence. It leaves aside other rules in Belgian constitutional and administrative law assuring the representation and rights of groups in political/philosophical and linguistic/territorial terms in several governments or in the federal public administration. 5 In other words, the places on the list that are certainly or likely to lead to a term of office, either by getting elected or by taking over the mandate of an incumbent switching to another function. 6 The first two candidates should be of a different sex or no three successive candidates should be of the same sex.

9

Deliberative democracy, ecological representation and risk Towards a democracy of the affected Robyn Eckersley

Introduction The intensification of economic globalization has generated a growing disjuncture between those who enjoy the benefits of economic modernization and those who suffer the negative ecological and social ‘side-effects’.* While Ulrich Beck has suggested that ‘. . . poverty is hierarchical, while smog is democratic’ (Beck 1995: 60, his italics), the environmental justice movement and Third World political ecologists have pointed to the many ways in which racial minorities, the poor, and heavily indebted developing countries tend to suffer a disproportionate share of the ecological and social problems generated by economic modernization. Accordingly, the green political challenge raised by the ecological crisis involves the double challenge of not only finding ways of reducing ecological and social risks wherever possible but also finding ways of minimizing the unfair externalization and displacement of risks onto innocent third parties in space and time. The history of modern grassroots environmental activism and the broader green movement has been, among other things, a history of attempts to address the problems of risk-generation and risk-externalization and displacement by seeking to extend and deepen democracy (e.g. Paehlke 1980). In the course of pursuing the cause of environmental protection, the ecology movement has sought to improve the quality and free up the flow and availability of information to affected parties; challenge the entrenched power of technocratic and corporate elites; create more transparency in policy-making and administration; and encourage more citizen participation in economic and environmental planning and decision-making. In so doing, the ecology movement has brought new issues and concerns on to the policy-making agenda, it has introduced new ways of framing and defining environmental policy problems, and it has challenged the assumptions and framework of those policy professionals who manage risk assessment (Torgerson 1997). Ulrich Beck has also highlighted the ways in which the ecology movement has been in the vanguard of challenging the ‘relations of definition’ – the structures of authority that define, assess and manage risks. As Beck puts it, ‘Industrial society has created a “truncated democracy”, in which questions of the technological change of society remain beyond the reach of political-parliamentary decisionmaking’ (Beck 1998: 342, his italics). Yet Beck (along with Giddens 1994) is a relative

118

Robyn Eckersley

late-comer to the growing line of green political theorists who have called for a new model of ‘ecological democracy’ that provides a post-positivist, socially and ecologically inclusive model of risk assessment. A good deal of the growing literature on environmental policy and risk assessment has also underscored the centrality of democratic dialogue to ecological reform at all levels of governance (Dryzek 1987b, 1992; Fischer 1993; Lafferty and Meadowcroft 1996b; Leiss 1990; Torgerson 1997; Wynne 1996). This literature also suggests how the multiple democratic deficits concerning accountability and control may be addressed by, among other things, new forms and styles of political communication, which bring together as many disparate players as practicable (including culprits and beneficiaries, experts and lay people) into an open and constructive dialogue that seeks a broad social agreement about alternative development paths. In this chapter, I explore critically the extent to which deliberative democracy might serve as a viable candidate for such a post-positivist, socially inclusive (and I would add, trans-boundary) model of environmental decision-making and risk assessment. Such an exploration is intended not only as a contribution to the critical debates on deliberative democracy, but also as a contribution to fleshing out the enticing, but still vague, notion of ecological democracy.

Ecological democracy: an ambit claim My critical exploration of deliberative democracy proceeds on the basis of a very simple, but ultimately very politically challenging, ambit claim for ecological democracy: all those potentially affected by risks should have some meaningful opportunity to participate or otherwise be represented in the making of the policies or decisions which generate such risks. This formulation is centrally informed by, but ultimately does not strictly insist upon, the moral argument that persons and communities should not be subjected to avoidable risks without their free and informed prior consent (I say ‘avoidable risks’ in order to rule out natural environmental disasters involving no or negligible human agency). That is, I am not insisting that all those potentially affected by risks must always actually reach a prior consensus about whether or not to proceed with risk-generating activity, since this is not always practicable (for large populations) or possible (for future generations). Nonetheless, I shall be arguing that it is at least intuitively plausible to maintain that if those representatives who do engage in decision-making with risk implications proceeded as if all affected were present, informed and capable of raising objections, then this would encourage an orientation that is both (a) riskaverse and (b) also concerned to avoid the unfair externalization of risks, thereby addressing what I have identified as the ‘double challenge’ of ecological democracy. Now, at first blush, there is nothing obviously new or ‘ecological’ about this formulation of democracy, which already resonates with the deliberative ideal of democracy. However, what makes this formulation both new and ecological is the accompanying argument that the opportunity to participate or otherwise be

Deliberative democracy, ecological representation and risk 119 represented in the making of risk-generating decisions should literally be extended to all those potentially affected, irrespective of social class, geographic location, nationality, generation, or species. This ecological extension of the idea of a ‘democracy-of-the-affected’ is intended to be inclusive and ecumenical, incorporating the concerns of environmental justice advocates, risk society sociologists and ecocentric green political theorists. Indeed, environmental democracy may be best understood not so much as a ‘democracy-of-the-affected’ but rather as a ‘democracy-for-the-affected’, since the class of beings entitled to moral consideration in democratic deliberation (whether infants, the infirm, the yet-to-be-born, or non-human species) will invariably be wider than the class of those who are actually alive and physically and intellectually capable of engaging in democratic deliberation (namely, morally competent citizens). As an ideal, ecological democracy must necessarily always contain this representative dimension, which poses a direct challenge to Habermas’s procedural account of normative validity that runs as follows: ‘According to the discourse principle, just those norms deserve to be valid that could meet with the approval of those potentially affected, insofar as the latter participate in rational discourses’ (Habermas 1996c: 127, my emphasis). In relation to all those subjects lacking communicative competence, my ecological formulation replaces the italicized words ‘insofar as’ with ‘as if ’. I do not propose to rehearse here the basis for this green critique of the anthropocentric, neo-Kantian foundations of Habermas’s discourse ethic (for more extensive critiques, see Dryzek 1990b; Eckersley 1990, 1999; Dobson 1993, 1996b; Krebs 1997). For the moment, my concern is merely to raise this first challenging dimension of ecological democracy.

The community-of-fate The second challenging dimension of ecological democracy has to do with the way in which the demos is re-conceived as no longer fixed in terms of people and territory. The ‘ambit claim’ argues that in relation to the making of any policy or decision entailing potential risks, the relevant moral community must be understood as the relevant ‘community-of-fate’, tied together not by common passports, nationality, blood line, ethnicity, or religion but rather simply by the potential to be harmed by the particular proposal, and not necessarily all in the same way or to the same degree. For example, in the case of a proposal to build a large dam, the community-of-fate might be the relevant watershed, irrespective of the location of state territorial boundaries. For a proposal to build a nuclear reactor, the spatial community-of-fate might be the better part of an entire hemisphere, spanning continents and oceans (the community-of-fate would also extend almost indefinitely into the future, encompassing countless generations). For a proposal to release genetically modified organisms into the environment, the relevant communities-of-fate might be variable and not contiguous in space or contemporaneous in time. In each case, the affected community would typically include both present and future human populations and the ecosystems in which they are embedded. Moreover, the boundaries of such communities would rarely be

120

Robyn Eckersley

determinate or fixed but instead possess more of the character of ‘spatial-temporal zones’ with nebulous and/or fading edges. The foregoing ‘ambit claim’ for ecological democracy raises complex moral, epistemological and institutional challenges. It is morally challenging because it loosens the requirement of moral reciprocity that is basic to the Kantian tradition of moral reasoning and conventional notions of citizenship by seeking to extend democratic consideration to a somewhat indeterminate community-of-fate. It is epistemologically challenging because it asks those who can participate in democratic deliberations to search for meaningful, practical and parsimonious ways of ‘representing’ the interests of those who cannot – through no fault of their own – represent themselves (most notably, future generations and non-human others). It is institutionally challenging because it rejects territorially fixed structures of democratic rule and suggests the need for new, flexible institutions and procedures that are capable of mapping onto the complex and variable contours of ecological problems and the human and non-human communities they affect. The primary question for consideration in this discussion is: to what extent can deliberative democracy meet these challenges? The subsidiary question – to what extent should the ambit claim for ecological democracy be qualified or trimmed back to accommodate counter-challenges and practical exigencies – will also be considered along the way.

The intuitive green appeal of deliberative democracy Many green political theorists have turned to deliberative democracy out of dissatisfaction with ‘actually existing liberal democracy’.1 The primary appeal of deliberative democracy is that it eschews the liberal paradigm of strategic bargaining or power trading between self-interested actors in the marketplace in favour of the paradigm of unconstrained egalitarian deliberation over questions of value and common purpose in the public sphere. That is, the conditions of ‘undistorted’ and ‘other-regarding’ communication characteristic of the deliberative democratic ideal are defended as more likely to lead to the prudent protection of public goods (such as environmental quality) than the ‘distorted’ and ‘strategic’ political communication that is characteristic of liberal democracies (e.g. Dryzek 1987b, 1990b; Goodin 1996). Public-spirited political deliberation is the process by which we learn of our dependence on others (and the environment) and the process by which we learn to recognize and respect differently situated others (including non-human others and future generations). It is the activity through which citizens consciously create a common life and a common future together, including the ecosystem health and integrity which literally sustains us all. Deliberative democracy has a long pedigree, reaching back to Athens and including the long tradition of civic republicanism as well as more recent innovations in critical theory, such as Jürgen Habermas’s discourse ethic.2 However, for our purposes, I will unceremoniously lump these various strands together and then single out three mutually constitutive features that together encapsulate what

Deliberative democracy, ecological representation and risk 121 I take to be the core ideals and appeal of the deliberative model: unconstrained dialogue, inclusiveness and social learning. Unconstrained dialogue The requirement that dialogue be unconstrained or free is a requirement that nothing but the ‘force of the better argument’ be allowed to sway the participants in the dialogue. Dialogue becomes constrained and distorted to the extent to which participants are swayed by considerations other than rational argument (such as by implicit or explicit force, deception, bribery, or the authority and status of the speaker rather than the content of what is said) or when insufficient time is allowed for deliberation over the meaning and consequences of putative facts or proposed norms. The requirement of free dialogue also necessarily encompasses the requirement of publicity. Dialogue is constrained when information is withheld or ‘misinformation’ is spread. It is also constrained when persons affected by proposed norms are denied an opportunity to participate or be represented in the dialogue. Inclusiveness Deliberative democrats typically enlist the requirement of impartiality as an essential requirement of deliberative dialogue, since the point of deliberation is to weed out purely partial or self-interested arguments in favour of arguments that can be defended as acceptable to all. However, in the light of postmodern scepticism towards the very possibility of impartial thinking, the notion of inclusiveness or ‘enlarged thinking’ perhaps better describes the ‘other-regarding orientation’ that is expected of participants (while also avoiding the debate about the possibility of impartiality). Enlarged thinking – or what Hannah Arendt (1961) calls ‘representative thinking’ – refers to the imaginative representation to ourselves of the perspectives and situations of others in the course of formulating, defending or contesting proposed collective norms. This idea of inclusiveness is derived from the more fundamental moral notion of respect for the autonomy of others. Indeed, the discursive democratic requirement that, if agreement is to be reached with others, individuals must defend their proposed norms in terms that may be acceptable to others is the very mechanism which steers deliberation away from merely selfish arguments towards generalizable ones. Social learning The social learning dimension of deliberative democracy flows from the requirement that participants be open and flexible in their thinking, that they enter a public dialogue with a preparedness to have their preferences transformed through reasoned argument. This is, to some extent, a restatement of the requirement of free or unconstrained dialogue, whereby participants are moved to change their position by the force of best/most appropriate reasoned argument rather than

122

Robyn Eckersley

extraneous considerations. However, this feature also highlights what is typically defended as one of the great strengths of deliberative democracy, viz. its educative or social learning potential. That is, openness and flexibility on the part of deliberators makes it possible for them to make decisions that are adaptable and self-correcting in the face of new circumstances, new information, and new or revised arguments. This is also the sense in which deliberative democracy is contrasted with purely aggregative models of democracy (such as voting or opinion polling), whereby individual preferences are merely added together but not challenged and debated, and where no communication takes place between preference holders. (Of course, it is also possible to precede aggregation with deliberation; the contrast here is merely between deliberative/communicative and non-deliberative/isolationist preference-formation). These three features of deliberative democracy – unconstrained dialogue, inclusiveness and social learning – arguably make deliberative democracy especially suited to dealing with complex and variable ecological problems and concerns. In particular, such a model privileges generalizable interests over private, sectional or vested interests, thereby making public-interest environmental advocacy a virtue rather than an heroic aberration. It invites reflexivity, selfcorrection and the continual public testing of claims. Such critical testing and questioning from the perspective of differently situated others is crucial to arresting and reversing what Habermas (1971) once called ‘the scientization of politics’, the process whereby the lay public cedes ever greater areas of system decision making to technocratic (including scientific, professional, corporate, and bureaucratic) elites. The continual critical and public testing of normative claims, including norms embedded in scientific claims, also makes it possible to expose and subject to scrutiny the assumptions, interests and world-views of technocratic policy professionals, politicians and corporate leaders. In the field of risk assessment, the deliberative model is likely to generate a more long-range, inclusive and risk-averse orientation rather than an after-the-fact damage limitation approach. Such a posture necessarily arises when one asks the question: would all those potentially affected by proposed risk-generating practices, rich and poor, citizens and ‘aliens’, now and in the future, consent to such risks if they were fully informed of the potential consequences? Moreover, it is a short (albeit very provocative) step to extend the ‘other-regarding’ orientation of deliberative democracy to include, at least notionally, non-human species as imaginary partners in communication in the course of (other-regarding) dialogue. According to Robert Goodin (1996: 844), we may think of this kind of ‘democracy as a process in which we all come to internalize the interests of each other and indeed of the larger world around us’. That is, when the circle of moral considerability is widened to the maximum to include all potentially affected persons and social and ecological communities, then the very possibility of externalizing ecological costs onto innocent third parties is foreclosed. In the terms of the double challenge of ecological democracy, then, deliberative democracy, prima facie, appears promising. Not only is it likely to generate a risk-averse orientation, it is also likely to guard against unfair displacement of risks onto innocent third parties. Such an

Deliberative democracy, ecological representation and risk 123 orientation provides a welcome move away from the utilitarian framework of cost– benefit analysis (which permits the sacrifice of the interests of minorities, those lacking ‘preferences’, and the discounted future in favour of present majorities) towards a more inclusive orientation that strives to find ways of mutually accommodating (rather than trading off ) the needs of the present and the future, the human and the non-human.3 In short, a case can be made that deliberative democracy is especially suited to making collective decisions about long-range, generalizable interests, such as environmental protection and indeed sustainable development. Moreover, because it does not confine itself to the citizens and territory of a particular polity it may be understood as a transnational form of democracy that is able to cope with ‘fluid boundaries’ (Dryzek 1999: 44). It also has the capacity to accommodate the complexities and uncertainties associated with ecological problems; include and evaluate both expert scientific and vernacular understandings of ecological problems; and identify and evaluate risks in socially and ecologically inclusive ways. Above all, deliberative democracy may be defended as the best model for reaching mutual understandings about common norms, and the quest to create an ecologically sustainable society is fundamentally a normative concern and only secondarily a technical matter. As Barry emphasizes, the concept of sustainability ‘needs to be understood as a discursively “created” rather than an authoritatively “given” product’ (1996: 116). And so ends the case for the affirmative. How might sceptics respond to the foregoing theoretical defence of this highly idealized model of deliberation?

The critique of deliberative democracy In evaluating deliberative democracy as a candidate for a post-positivist, socially and ecologically inclusive model of political communication, everything turns on how we understand the character of the claims made by proponents of this ideal. In particular, are we to understand deliberative democracy as: •

• •

a counterfactual regulative ideal (rather than a blueprint to be strictly adopted) that provides a critical vantage point from which we might impugn particular deliberations and decisions that have fallen short of the ideal; an aspiration, and source of inspiration, for ongoing democratic reconstruction; and/or a democratic model or a blueprint of political deliberation for the wellordered organization/community/society?

I shall be arguing that deliberative democracy is at its strongest when understood in the first and second senses and at its weakest when understood in the third sense and, further, that the critique of deliberative democracy as blueprint does not detract from the force of the defence of deliberative democracy as regulative ideal and aspiration/inspiration for democratic reform. Nonetheless, the critique of deliberative democracy as blueprint does place the onus on defenders of

124

Robyn Eckersley

deliberative democracy to demonstrate how and to what extent deliberative democracy (green or otherwise) might be successfully institutionalized. Now the practical strength of deliberative democracy as a counterfactual ideal is that it provides a potent critical vantage point from which to unmask unequal power relations and the political actors who sanctify them; identify issues and social groups that are excluded from public dialogue; and sift out genuinely public interests from merely vested private interests. However else one wishes to defend deliberative democracy, I take this ‘critical vantage point’ argument to constitute its unimpeachable core. Indeed, it is implicitly invoked even by those critics of deliberative democracy who claim it is too dispassionate, rationalist, disembodied, masculine, and western/Eurocentric in its orientation in insisting only on certain modes of rational, critical argument (e.g. Young 1997). Such criticisms, in pointing to different modes of political communication such as greeting, rhetoric, story-telling/testimony and performative action (such as satirical street theatre) that appear to be excluded from overly rationalistic ideals of deliberative democracy, presuppose at least a structurally similar evaluative standpoint to that of deliberative democrats. That is, critical postmodernists effectively join with critical modernists in defending the ideal of free and equal human subjects determining their own individual and common destinies in circumstances that are free from explicit or implicit coercion. Without this ideal, there would be no basis upon which to mount such a critique. While there is certainly room to argue for a widening of what should count as valid or appropriate political communication, this is an immanent critique that does not in itself impeach the critical normative orientation of deliberative democracy: freedom as non-domination. In any event, deliberative democracy seems well capable of absorbing Young’s arguments. As Bohman (1998: 410) has put it, deliberative democracy cannot ignore different styles of political communication ‘without threatening social co-operation in deliberation itself ’. Moreover, if we adopt Dryzek’s formulation (in this volume) of deliberation as communication that induces reflection on preferences in a noncoercive fashion, then we leave room for a wide variety of modes of political communication. My green case for deliberative democracy (as outlined above) effectively employs and extends deliberative democracy in exactly this way – as a regulative ideal of free communication against which we may impugn the legitimacy of communication that can be shown to be less than free. Here, ‘less than free’ can also include insufficiently inclusive in those circumstances where affected parties are not given ‘a voice’ in the deliberations. This, then, is one (critical) sense in which I would maintain that deliberative democracy is able to ‘serve’ the ambit claim for ecological democracy. However, it would be politically unsatisfactory to rest the argument here. In the move from regulative ideals and political critique, on the one hand, to practical institutional reform, on the other hand, many problems have to be negotiated. These problems arise because, as Johnson (1998: 173) has noted, it is foolhardy to make ‘heroic assumptions’ about the motivations of political actors in democratic deliberation, especially when considering institutional re-design. That is, in a world where power disparities are ever-present, it is, to put it mildly, naïve to expect

Deliberative democracy, ecological representation and risk 125 policy-makers always to be so virtuous and patient as to put the public good ahead of their own interests, concerns and identities, and genuinely listen to, and accommodate, all opposing viewpoints in the course of political dialogue and decision-making. As Said, in a spirited critique of the discourse ethic, has put it, the ‘scrubbed, disinfected interlocutor is a laboratory creation’, which bears no relationship to real political discourse (Said 1989: 210; see also Sanders 1997). Moreover, the idealizing force of the deliberative model as blueprint is not especially helpful when it comes to real-world institutional design and political decision-making where time, information, knowledge, and other constraints abound. Clearly, if we are to do justice to the marginal and dispossessed (including those who cannot represent themselves), and if we are to achieve feasible outcomes, then political procedures and institutions must not be formulated in the philosophical laboratory (where power disparities are absent) but in the real world, where power disparities, distortions in communication and other pressures are ever-present. Moreover, if it is accepted that there is a multiplicity of genres of speech and argument, which may be traced to (among other things) different linguistic and cultural backgrounds, then one might also challenge the normative presupposition of a shared, implicit telos towards mutual understanding in political dialogue, especially in multicultural polities. In such complex and diverse polities, we can expect disagreement to be the rule rather than the exception, and we can also expect that such disagreement will not necessarily always be ‘reasoned’ or ‘reasonable’. Indeed, on many moral, religious and philosophical questions (such as the abortion debate), we can expect disagreement to be a permanent state of affairs. However, the foregoing criticisms should not be taken to render the regulative ideal ineffectual, since without the ideal we would have no normative basis upon which to impugn any political communication or decision. Moreover, this regulative ideal has enduring value not only as a critical vantage point but also as a constructive vantage point, serving as the source of inspiration for ongoing creative institutional design. In any event, recent work on deliberative democracy has been increasingly preoccupied with practical concerns about disagreement, feasibility, social complexity, and institutionalization.4 Indeed, Bohman, after an extensive survey of such work, has declared that: ‘Tempered with considerations of feasibility, disagreement and empirical limits, deliberative democracy has now “come of age” as a practical ideal’ (1998: 422). For example, many advocates of deliberative democracy have turned their attention away from the counterfactual ideal of deliberation and towards the actual processes of deliberation in an effort to develop a more dynamic understanding of the relationship between ideals and practices. While all deliberative democrats may prize consensus, they neither assume nor expect it in every case; instead they have offered a framework for understanding and dealing with difference and disagreement. For example, Gutmann and Thompson (1996) have argued that the fact of persistent disagreement is hardly a reason for abandoning deliberative democracy. Rather, they suggest that it highlights its great virtue, since its procedural requirements (which

126

Robyn Eckersley

they identify as reciprocity, publicity and accountability)5 still make it superior to other methods for resolving conflicts. It can, for example, better facilitate the search for ‘an economy of moral disagreement’.6 Similarly, Dryzek (this volume) has defended ‘workable agreements’, which are consistent with Cass Sunstein’s ‘incompletely theorized agreements’, which Sunstein argues ‘represent a distinctive solution to social pluralism’ and ‘a crucial aspect of the exercise of reason in deliberative democracies’ (Sunstein 1997: 115). Such agreements are agreements on outcomes and narrow or low-level principles on which people can converge from diverse foundations; they are concerned with particulars, not abstractions. Sunstein also suggests that agreements of this kind are well suited to the need for moral evolution. More generally, Bohman (1998: 422) has concluded that ‘most feasible formulations of the ideal of deliberative democracy require the check of empirical social science’. Building on this insight, and turning now to the challenge of institutional design, it is important to acknowledge that what I have singled out as the core ideals of the deliberative model – free or unconstrained dialogue, inclusive/ enlarged thinking and social learning – sometimes have to be actively cultivated rather than assumed to exist before deliberation, or assumed always to arise in the course of deliberation.7 When one considers those forums where something approximating genuine deliberation tends to take place (Quaker meetings; university tutorials – at their best, not their worst; reading groups; well-facilitated public meetings in the town hall; citizens’ juries; deliberative opinion polls and consensus conferences), it is either because of a pre-existing, deep-seated mutual understanding that engenders the necessary mutual respect (the Quaker meetings); a shared culture of critical discourse (the reading group or tutorial); and/or because the forum is carefully contrived and managed as a deliberative microcosm to facilitate free dialogue. Critical explorations of the group dynamics of conventional juries provide a sobering reminder that we cannot always expect unconstrained deliberation to arise ‘naturally’, even in small groups (see for instance Sanders 1997). In this respect, if deliberative democracy is to be understood as a ‘school for democracy’, then citizens and their political representatives sometimes need to be actively schooled in deliberative democracy before it is likely to take hold and flourish beyond the kinds of enclaves that I have listed. So what are we to make of these small-scale experiments in deliberative democracy, which seek to find a middle way between ‘politically unequal but deliberative elites and politically equal but non-deliberative masses’ (Fishkin and Luskin, this volume)? The fact that most deliberative microcosms have been, for the most part, merely advisory does not detract from the educative experience enjoyed by the participants, and the broader public in those cases where the deliberations are broadcast on national television, such as in deliberative polls (Fishkin 1995). Nor should the fact that such experiments are very expensive, time consuming and demanding of the participants detract from the value of such a consciousness-raising exercise. Nonetheless, these points are sobering for those who might wish to build more extensive deliberative institutions in both the state and civil society to counteract distortions in political communication stemming

Deliberative democracy, ecological representation and risk 127 from manifestly unequal power relations. Herein lies the greatest practical and institutional challenge to deliberative democrats (green or otherwise): can deliberative democracy be made to work in those areas where it most counts? This is an important question, because it is rarely in the interests of those elites with the political, economic and/or cultural power to control ‘the relations of definition’ – corporate managers, cabinet ministers, party power brokers, media magnates, managers of scientific research establishments – to conform to the deliberative protocol, since they typically have other, more efficient and potent mechanisms of ‘force’ at their disposal other than the force of the better argument to enable their will to prevail. Now defenders of deliberative democracy would doubtless respond to this challenge by pointing out that a vibrant and vigilant civil society, well-educated and well-versed in deliberative discourse, would keep such elites ‘on their toes’ by means of critical questioning, exposing and moral shaming. But there can be no guarantees that elites will always capitulate, a fact that underscores the precarious ‘force’ of critical reason (green or otherwise) when set against more conventional forms of explicit and implicit force. While this acknowledgement should not detract from the moral force of deliberative democracy as regulative ideal/critical vantage point, it nonetheless invites consideration of how the regulative ideal of deliberative democracy might serve as a reconstructive vantage point for ecological democrats who are interested in changing the world. Here it is necessary to explore what kinds of institutional devices might go some way towards redressing the unequal power relations that routinely thwart fulsome risk assessment on behalf of the ‘community-of-fate’. In addressing this question of how to ‘unrig’ the communicative context, let us accept that the idealized and demanding conditions of deliberative democracy are aspirational, and therefore can only ever be approximated (rather than fully realized) in everyday politics; alternative decision rules other than consensus may need to be applied to conclude what might otherwise be interminable debate; and, as Young (1996; 1997) argues, cultural and social differences should be considered a resource for public reason, rather than as divisions which public reason must somehow transcend.

Linking deliberation and representation Without offering an exhaustive response to the challenging question of institutional reform, I want to suggest a number of institutional ‘devices’ (some familiar, some less familiar) that might help to bring into fuller view the community-of-fate. Now we have already noted that it is neither possible nor practicable for all affected parties literally to deliberate together en masse. Indeed, we have seen that ecological democracy must necessarily contain a representative element if it is to function as a democracy-for-the-affected, including future generations and nonhuman species.8 Accordingly, the question of political representation emerges as a crucial issue in both the theory and practice of deliberative democracy. The institutional devices suggested below rest on the argument that fulsome ecological deliberation is more likely with more fulsome representation. The essence of the

128

Robyn Eckersley

argument is that while it is impossible to orchestrate a meeting of the entire ‘community-of-fate’, we can at least devise forms of political representation (along with appropriate procedures and decision rules) which serve to widen and deepen the horizons of those who are actually engaged in the making of risk-generating decisions. In particular, risk-generating and risk-externalizing decisions are less likely to survive policy-making communities and legislative chambers that are more rather than less inclusive in terms of class, gender, race, region and so on, and especially so when the deliberators are obliged to consider the effects of their decisions on social and ecological communities both within and beyond the formal demos. Such procedures would, in effect, serve to ‘redraw’ the boundaries of the demos to accommodate the relevant affected community in every potentially riskgenerating decision. (Such procedures also offer an alternative to the multiplication of regional and international governance structures advocated by cosmopolitan democrats such as David Held (1995), which introduce their own democratic deficits.) We have already noted the dangers of assuming virtue (ecological or otherwise) on the part of political representatives and policy-makers. One of the reasons why this is a dangerous assumption is that it fails to grapple with the unavoidable epistemological and motivational limitations of all forms of political representation (see Gargarella 1998: 261–2). That is, political representatives may find it difficult or impossible to understand or imagine the perspectives of all differentlysituated others (whether because of lack of personal experience of ‘the other’, lack of information, misinformation, or scientific uncertainty), in order to formulate norms that may be acceptable to such ‘others’. Moreover, representatives may not possess the necessary motivation to treat the situation, concerns and interests of differently-situated others on an equal par with their own. More generally, as feminist difference theorists (or, following Squires’s terminology in this volume, diversity theorists) have pointed out, all political arguments, however wellintended, cannot be entirely detached from the experience, cultural and class background, and material interests of their proponents. These insights apply as much to the arguments of public-spirited environmentalists as they do to corporate elites. One way of responding to these epistemological and motivational limitations that are inherently associated with political representation is to insist that political representation be as diverse as possible. In short, deliberative democracy must be representative in a double, reflexive sense. It must encourage enlarged thinking, and it must also provide for enlarged (i.e. diverse) representation on the understanding that it is dangerous always to trust in the ‘political imagination’ of the chosen or privileged few (whether Burkean, Madisonian, green, or otherwise). That is, diverse representation facilitates enlarged thinking by minimizing the problem of a narrow band of elites ‘second-guessing’ (benignly or otherwise) the concerns and interests of differently-situated others, especially minority groups. In the language of Anne Phillips, the ‘politics of ideas’ must be supplemented with a ‘politics of presence’ (Phillips 1996). This argument is also broadly consistent with Iris Marion Young’s neo-Habermasian conception of ‘communicative

Deliberative democracy, ecological representation and risk 129 democracy’, which criticizes both the liberal and civic republican ideal of impartiality and relies instead on group representation as a ‘strategy of displacement’ (see Squires, this volume). Diverse representation provides one means of confronting, displacing and ultimately stretching the political imagination of representatives, thereby going some way towards correcting the exclusionary implications of the knowledge and motivational deficits associated with political representation. From the point of view of environmental justice advocates, ensuring the ‘presence’ of minority or disadvantaged groups in legislative assemblies and environmental policy-making communities (via such measures as ‘balanced tickets’ and proportional representation electoral systems) will go some way towards preventing the unfair displacement of ecological and social costs onto those minority communities (see Meier, in this volume, on the representation of women). The adoption of multi-member electoral systems would also increase the likelihood of green parties gaining a formal presence in parliaments. But how might we apply these lessons to the problem of ‘representing’ those who literally cannot represent themselves? If it is accepted that communicative competence is arbitrary from a moral point of view, then we ought to accept second-best solutions for realizing these moral claims since finding an approximate form of representation is better than providing none at all. Accordingly, while it is much more difficult and sometimes even hazardous to engage in ‘representative thinking’ on behalf of non-human others compared to differently-situated human others, it is nonetheless a task we ought to attempt from the broader standpoint of environmental justice (contra Habermas, see Eckersley 1999). It is precisely this striving to adopt the standpoint of differently-situated human and non-human others that serves to steer deliberation towards finding inclusive, common ground between human and non-human communities. The remaining problem then is a practical one: how to redress the epistemological and motivational problems that we will expect to encounter in ‘representing’ non-human others and future generations? Turning first to the epistemological problem, improving the quality and free flow of knowledge about ecological problems by means of, say, mandatory stateof-the-environment reporting, along with comprehensive and cumulative (as distinct from merely project-specific) regional environmental impact assessments are familiar mechanisms that would go some considerable way towards improving the knowledge base of the general public and environmental policy-makers in particular. (However, as we shall see, further mechanisms are necessary to deal with the inevitable gaps, limitations, controversies, and uncertainties in both scientific and vernacular understandings of environmental problems.) With respect to the motivation problem, the constitutional entrenchment of the precautionary principle so that it became obligatory in all decision-making and policy-making involving potential ecological risks would provide a highly effective and parsimonious means of forcing more systematic consideration of potential environmental impacts on differently-situated others, including impacts on the interests of future generations and non-human species. The Rio Declaration formulation of the precautionary principle provides that ‘Where there are threats

130

Robyn Eckersley

of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation’ (principle 15). Adding the words ‘to present and future human and non-human communities’ after the words ‘irreversible damage’ would head off narrow, anthropocentric interpretations of this decision rule, which effectively provides a presumption against decisions that carry serious or irreversible environmental risks (such as species extinction, climate change, nuclear fall out). The decision rule also serves as an evidentiary rule in placing the onus of proof on the proponent to prove the absence of such risks for human and non-human communities, now and in the future. The case for constitutional entrenchment of this decision rule may be justified on grounds of justice: that special constitutional protection is required precisely because there are no other direct and formal mechanisms to ensure the representation of the non-human world, future generations and ‘aliens’ living beyond the polity, even though they may be adversely affected by decisions made within the polity. The reciprocal observance of this decision rule by all nation states would provide one powerful surrogate mechanism for institutionalizing ecological and social responsibility towards the relevant community-of-fate in those cases where that community transcends state jurisdictional boundaries. Finally, a case can also be made that diverse cultural representation in policy communities and legislative chambers should also be supplemented by specialized environmental advocacy. The ‘presence’ of green parties obviously goes some way towards realizing this claim, but green parties are typically poorly resourced and politically marginal when compared to mainstream parties backed by wealthy and vested ‘producer’ interests (chiefly capital but also labour).9 There are also countless other ways in which environmental advocacy might be institutionalized in structures of national and global policy-making and governance to complement (as distinct from replace) the advocacy of the environmental non-government organizations (NGOs) and grassroots community initiatives in civil society. At the local and national levels, the establishment and proper resourcing of an independent Environmental Defenders Office, staffed by a multidisciplinary team and charged with responsibility for environmental monitoring, political advocacy and legal representation would go some way towards ensuring that more systematic attention is directed towards ecological problems. At the international level, the formal inclusion of major international environmental NGOs in multilateral negotiations over environmental treaties and declarations as well as in the governance structures of international financial and trade organizations (such as the World Bank, the IMF, and the GATT/WTO) would also provide a more systematic and fulsome representation of affected ecological communities in different environmental policy domains.

Conclusion The institutional innovations sketched above are illustrative only. My main concern has been to defend deliberative democracy as being highly conducive to

Deliberative democracy, ecological representation and risk 131 ecological democracy but also to argue that inclusive deliberation demands more inclusive forms of representation if ecological justice is to be done. This is not a case of rigging the system in favour of the environment, as some liberal democrats might wish to argue, but rather a matter of improving the conditions of dialogue while also redressing major power imbalances in political communication and representation. Implicit in this argument is the claim that no model of democracy can ever be accepted as being merely procedural. All models of democracy are derived from more fundamental norms which, in turn, reflect and shape the character and boundaries of the moral and political community that we seek to cultivate. Of course, we can expect the struggle to reform political institutions in ways that promote greener representation (and hence greener deliberation) to encounter strong resistance from those who stand to lose present advantages. All of this suggests that the ambit claim for ecological democracy is likely to remain mostly that, and the deliberative ideal which serves it is likewise likely to remain, for the most part, aspirational. However, it is precisely because this ambit claim functions as a regulative ideal that we ought not ‘trim its sails’, since that would undermine its enduring value as both a critical vantage point on the present and as an ongoing source of inspiration for the reform of our democratic institutions.

Notes * A special thank you to Peter Christoff, Mark Edmondson, Michael Saward and especially John Dryzek for insightful comments on earlier formulations of this chapter. 1 See, in particular, the contributions to Mathews (1996) and Doherty and de Geus (1996). According to Bohman (1998: 400), ‘deliberative democracy begins with the standard practices of liberal democracy’. 2 Although Habermas (1996a and 1996c) has recently situated his particular model of discursive democracy between liberal democracy and republicanism, it still remains much closer to civic republicanism in spirit. 3 These criticisms are partly offset by recent attempts by environmental economists to include option values, bequest values and existence values in the cost-benefit calculus (see, for example, Pearce, Markandya and Barbier 1989). 4 See, for example, Dryzek (1990 and 1996); Gutmann and Thompson (1996); and Habermas (1996c). For a recent survey of some of these (and other) works, see Bohman (1998). 5 For Gutmann and Thompson, reciprocity, publicity and accountability ‘address an aspect of the reason giving process: the kind of reason that should be given, the forum in which they should be given and agents to whom they should be given’ (1996: 52). 6 See also Thompson (1999: 124). 7 I wish to thank Mark Edmondson for first bringing to my attention this less than fully investigated problem. That is, will deliberation founder if there is no mutual respect between the parties at the beginning of deliberation, or will mutual respect develop during the course, or as a result, of deliberation? What non-deliberative factors might account for the absence or presence of mutual respect before, during or after deliberation? 8 Sunstein has argued that deliberative democracy is necessarily representative ‘on the theory that direct democracy is less likely to be pervaded by reasons’ (1997: 94). 9 Dennis Thompson has suggested the establishment of forums in which representatives could speak for the ordinary citizens of foreign states while also considering the views of

132

Robyn Eckersley

international organizations, possibly formalized by a Tribune for Noncitizens (1999: 121–2), thereby correcting the bias of commercial and government-to-government negotiations. In a similar vein, Andrew Dobson (1996b) has defended the provocative idea of formal ‘proxy representation’ of both non-human animals and future generations in representative assemblies of deputies elected from the environmental sustainability lobby.

10 Ecological constitutionalism and the limits of deliberation and representation Mike Mills and Fraser King Robyn Eckersley’s chapter (above) serves to illustrate the fact that it has become central to the green or ecological view of democracy that it should be based primarily on a form that is deliberative in character. This will, it is argued, not only secure the promotion of green values, but will also enable self-development and the excavation of deeper societal ethical norms. However, the nature of environmental risks and, indeed, the green values that often identify them, are not necessarily best served or resolved through deliberation alone. While deliberation does have its benefits, to focus on deliberation without considering its constitutional context, is to miss much of the political benefit of constitutional change. We suggest that constitutionalism (a much-neglected feature of ecological democratic thinking) provides a mechanism whereby ecological values and pervasive societal risks may be reconciled. This chapter argues that a preoccupation with deliberation is not unproblematic. We suggest a greater emphasis on constitutional engineering as a more fruitful mechanism for securing ecological goals within a broadly liberal democratic framework. The purpose of constitutionalism in this context is to resolve a number of the difficulties with deliberative forms of democracy. In particular, we claim that deliberation is limited in its ability to deal effectively with issues of risk and hence we suggest that boundaries need to be set on the deliberative (or any other democratic) process itself. In this way the selfbinding nature of democracy is retained and, importantly, questions of severe ecological risk are not subject to the arbitrary outcomes implicit in the deliberative process. Even models of democracy which are ostensibly ecological and seek to adjust the representative mechanism by the innovative use of the proxy (Dobson 1996b) do so in ways which simply replicate, rather than overcome, the problems implicit in the representative model. So, whilst sympathetic to the distinctive position adopted by Eckersley, we wish to emphasize the importance of the constitutional dimension in innovative ecological visions of democracy.

Constitutionalism, democracy and ecological political thought Constitutionalism provides the (generally obscured) background to contemporary political life and political discourse in much of the western world. Much has been

134

Mike Mills and Fraser King

written about defining and understanding constitutionalism but we shall rest with the formula developed by Elster, namely that: Constitutionalism refers to the limits on majority decisions; more specifically, to limits that are in some sense self-imposed. The limits can take a variety of forms. They can be procedural or substantive; they can block or merely slow down the process of legislative change . . . a written constitution . . . among other things, prescribes complicated procedures for changing the constitution itself, including delays, qualified majorities and the like. Lower level clauses can be either procedural or substantive. They include detailed regulations of the electoral process and of assembly voting, as well as guarantees for various individual rights, civil as well as political. (Elster 1988: 2) On this definition we might conclude that democracy (taken to mean the mediated or unmediated will of the people) is incompatible with constitutionalism.1 Such a view is contingent on a scheme of democracy that must be ephemeral and shortlived. We will not enter the substantive basis of these arguments, suffice to say that any democratic system worthy of the name must be logically bound by itself: constitutional frameworks seem the most appropriate in this sense (Dahl 1989; Saward 1998a). Taken in this way, political ecologists might find more comfort than they imagined from traditional liberal-democratic sources. Constitutions thus constrain majority decisions that threaten the basis of democracy and set the legitimate boundaries of ethical discourse and policy scope. As a consequence, it is inconceivable that a scheme of deliberative democracy could exist in a constitutional vacuum. The key point which flows from this is that constitutionalism provides a means of institutionalizing environmental rights, providing for the legal standing of non-human interests, and the institutionalization of the precautionary principle, all in a way that is not rigid and inflexible. Constitutional safeguards appear to be an obvious starting point for an ecological vision of democracy because, while not preventing deliberation, they can set the normative boundaries within which deliberation can take place. As the chapter proceeds it will become clear that the primary weakness of the deliberative case is that it holds that democratic legitimacy requires the priority of process over outcome.

Ecological values and ecological democracy The distinctive feature of ecological values is the way in which they employ a relational metaphysic. They help us to see humans as part of broader bio-systems, the harmony or balance of which humans have the capacity to affect adversely. This leads to the attribution of intrinsic value to nature as well as humans, although authors differ as to what is to be valued (higher mammals; all sentient beings; non-sentient life, etc.). It seems central to ecological values that the ultimate complexity of the universe is, if not unknowable, highly elusive and we should be wary of claims to knowledge

Ecological constitutionalism, deliberation and representation 135 which suggest otherwise. In short, a logical outcome of the relational view is that there is a heightened perception of uncertainty, hazard and risk, and a strong ethical constraint on actions which may adversely affect bio-systems. Guiding principles to policy would, therefore, be bound by such things as reversibility, protecting the vulnerable or precaution. In each of these cases there is an acknowledgement that we are uncertain about the consequences of our actions and should act with some humility in the face of limited knowledge (see, for example, Goodin 1983) To be a political ecologist is to accept that we inhabit a moral system broader than ourselves. This suggests that political ecologists will adopt a particular view of the social and political process which is consistent with models of democracy that are essentially inclusive and participatory. Not only is this viewed as being consistent with the development of ‘the self ’ in the world but the process of democracy is seen as essential for the realization of ecological goals (Barns 1995: 103; Paelhke 1995; Janicke 1996). Assertions are made not only about democratic forms but about the foundations of such forms. Bronwyn Hayward (1995: 220–1) suggests that the commitment to a participatory form of democracy is consistent with such core ecological views as self-determination, distrust of technocracy and of the inflated role of ‘experts’, and the transformative effects of political participation, along with ecological conceptions of what the ‘good life’ actually entails. Participatory theories of ecological democracy seek to prioritize some version of discursive, communicative or deliberative democracy. Such theories would normally include some combination of the following elements: decentralization; the public expression of ‘reasoned’ argument; open political structures that allow popular access and scrutiny; new means of participating that do not rely upon formal state institutions (for example, citizens’ juries); the incorporation of environmental concerns into basic constitutional and policy arrangements (e.g. environmental rights; animal rights; the precautionary principle); the opening up of dialogue between the different levels of the state; and improving the flow of information within the political system and between the state and citizenry (see, for example, Barns 1995; B. Hayward 1995; Dryzek 1995; Smith 1999). This model, or variations of it, now tends to be posited within a framework which accepts the ‘new realism’ (Lafferty and Meadowcroft 1996a; Eckersley 1996; Saward 1996; Smith 1999), that is, one which looks to work within the basic structures of the liberal democratic state. Smith (1999) suggests that there are two main features of deliberation which ought to benefit the efforts of political ecologists. First, public deliberation gives ‘equality of voice’, and second, the nature of discursive democracy is more open to the type of ethical reasoning that underpins most ecological thinking. Dryzek (1995) suggests that liberal political institutions are less likely than deliberative ones to be ecologically rational because the latter are more responsive to the complexities and uncertainties that environmental problems tend to expose. The reason for this is that the process of deliberation, by its very nature, allows for ‘negative feedback’ to be captured in the decision-making process and as a consequence improves the flow of information to decision-makers.

136

Mike Mills and Fraser King

Recent debates, however, have suggested that this model needs to be unpacked before claims to be genuine democrats will stand up to scrutiny. In particular, it has been claimed that some ecological values, despite an overt commitment to democracy, do not seem to be as democratic in character as is sometimes suggested.

The critique of ecological democracy The debate over ecological democracy centres largely on the work of Saward (1993) and Goodin (1992), in which it is argued that political ecologists tend to view the democratic process simply as a means to an end and consequently the value of democracy is contingent on it being able to deliver ecological policies: To advocate democracy is to advocate procedures, to advocate environmentalism is to advocate substantive outcomes: what guarantees can we have that the former procedures will yield the latter sorts of outcome? (Goodin 1992: 168) Furthermore, the values which tend to support such arguments (the intrinsic value of nature, for example) do not, at first glance at least, seem to support conventional views of democracy. By placing such value upon non-human interests the very ethical basis of democracy (equality, popular control, rights) seems to become subordinate to other values of an ecological nature (reversibility, precaution, etc.). It is, for example, not too difficult to argue the case for political equality within a particular species, but new (though not insurmountable) problems arise if we extend that principle to other species. In short, political ecologists are accused of undermining both the principles and the preconditions of democracy. Ecological responses begin by suggesting that such criticisms often rely upon the presumption that political ecologists use an exclusively utilitarian ethic (Goodin 1992). Eckersley (1996) has pointed out that it is equally justified to view them as taking a deontological position on the environment – that species, habitats, resources, and so on should be protected or promoted (or left alone) for their own sake. Equally, to take an entirely utilitarian view would not be consistent with views on the transformation of the individual. Similarly, Dobson (1996a) argues that the link between democracy and environmentalism can be made in two ways. First, as an argument from principle: that certain principles – such as autonomy – should be given voice in a democratic polity. Second, as an argument from preconditions: that a precondition of democracy is an environment capable of sustaining life – a point strongly made by Dryzek (1995). Both, in Dobson’s view, are necessary and neither is necessarily harmful to the democratic process itself. Importantly, neither betray a naïve instrumentalism or utility. Sagoff (1988) – himself a liberal – argues strongly that liberal concerns that environmentalists will undermine liberal rights (one of the preconditions of democracy) are largely unfounded because environmental issues tend not to be of the type which challenge those particular rights. Of course, they may conflict with other values but this does not make them undemocratic.

Ecological constitutionalism, deliberation and representation 137 The response, then, has been to anchor instrumentalism in values perceived to be compatible with democracy and which would secure the self-binding nature of democracy itself. The normal qualification to the democratic process bringing into political being values other than democratic ones is that they should not prevent the continuation of democracy, and are allowed to find political form only so long as the contingency of their acceptance is understood. Consequently, it is only the simplest of questions which can establish the contingent compatibility of ecological values with democracy: ‘have we chosen to take these ecological values as an aspect of the good life?’ and, if we have, ‘does the ability to change our mind still exist?’ In practical terms there does not seem to be any intrinsic reason why certain areas of political activity might not be subject to ecological values if this was generally agreed to provide, for example, a more ethical or more beneficial outcome. We could replace cost–benefit analysis with the precautionary principle; ‘ring-fence’ some areas of policy (perhaps those relating to sentient beings) as being the subject of extra democratic values prior to those in other policy areas (say, recreation); regulate manufacturing methods on the basis of revised views of the risks, hazards and uncertainties involved, and so on. Notice, though, that at least some of the ecological agenda appears best served by means which are broadly constitutional in character, rather than by solely deliberative or innovative views on representation more generally. So, ecological thinking has tended to emphasize deliberation but has also wanted to ensure political outcomes as well. It seems that the accusation that these two things are not compatible has led to a retreat from being ‘outcome oriented’ towards resting upon some form of deliberation as the only viable (or ethical) option left. It is this we take issue with because we believe that the use of constitutional mechanisms can facilitate a stronger regard for ecological outcomes without challenging the essential elements of democracy. It will become clearer as the chapter progresses why constitutional mechanisms may be preferable to deliberation and/or representation. As an example, we take below the arguments of Dobson (1996b) to illustrate why constitutional mechanisms avoid the complexities of using representative processes to further ecological ends.

Representing nature by ‘proxy’ Although deliberation is overwhelmingly the preferred model of political ecologists, innovative work on representation has also been done. In the next section we shall look more closely at the contrasts between deliberative and representational models, but the example given by Dobson (1996b) is a good introduction to these contrasts. Dobson suggests that environmental constituencies, which he identifies as nonnationals, future generations and other species, could and should be represented by ‘proxy’. To do so would not, he argues, involve compromising democratic legitimacy or the principle of the equal consideration of interests. Dobson’s

138

Mike Mills and Fraser King

representative formula, derived from Birch (1971), states that representation ‘denotes an agent or spokesman who acts on behalf of his principal’. This is argued to concur with the three general arguments in favour of the democratic representation of interests. The first springs from a consideration of subjective interests, the second from the principle of autonomy and the third from the logic of anticipated reactions – representatives will act in the interests of those they represent because they need to submit themselves for re-election. Two of Dobson’s cases are particularly pertinent for our concerns: the representation of future generations and the representation of other species. Dobson argues that it would be possible to isolate a lobby, drawn from those groups that advocate sustainable development, which could feasibly be regarded as having the interests of future generations at heart. The democratic objection to the use of such ‘experts’ in the policy process is trumped, in Dobson’s view, by the dominant value of securing an equal representation of interests. He maintains that as long as debates are played out under democratic conditions the argument for a proxy still holds because it meets any of the arguments for democratic representation given above. The case for the proxy representation of other species, although potentially weaker in Dobson’s view, would still fulfil the criteria for the equal representation of (in this case non-human) interests. Most of these arguments relate to the similarities of characteristics between human and non-human animals. On these grounds proxy representation for animals is argued to fit the general criteria of representing subjective interests and effective representation as proxies would be held accountable to the proxy community. The use of a proxy to represent the interests of nature is an innovative and compelling scheme to ‘green up’ liberal democracy. But, as with deliberation, significant questions remain about (a) how much benefit would accrue if the scheme was instituted, and (b) the implications of the model for other interests, besides those highlighted by political ecologists. Crucially, the problem of outcomes remains. Under Dobson’s proposals, outcomes will be highly dependent upon the balance of representation of the varying constituencies. This is not the case where constitutional constraints are deployed instead. Second, if the three principles of democratic inclusion can be applied meaningfully to interests other than those now represented, it is presumably also true that other interests are entitled to representation by proxy. Architecture, art, landscape – in fact, any aesthetic, cultural, or artistic form which cannot represent itself – is also entitled to representation. This in itself is not an objection to Dobson’s argument but it does suggest that the broader these interests are, the less compatible with ecological values these outcomes may be. Equally, it should be noted that there are interests which may be described as under-represented at present (minorities defined on the basis of their religion, culture, sexuality, region, ethnicity and so on) who may argue for a prior claim to representation. In common which much ecological thinking, Dobson’s argument does rather presuppose changes in our perceptions of what or who is morally considerable; indeed, it relies upon such a change. In contrast, our argument, while still relying

Ecological constitutionalism, deliberation and representation 139 upon democratic sanction and changes in public opinion, simply requires an acknowledgement that risks exist and they should be avoided, but does not require its constituency to be defined in advance. Hence it is both more effective than the reliance upon representation and more flexible than the simple demarcation of constituencies. So, while adjustments to the standard representational model are difficult, though worthy and challenging, the commitment to deliberation remains within ecological democratic thought and it is to this that we now return.

Deliberative democracy The factor that distinguishes deliberative theorists is their contrast between the process of merely aggregating preferences (usually in the form of votes) with the transformation of preferences by virtue of argument. This distinction between aggregative and transformative models of democracy is true for all models of deliberation that appear in the existing literature, although the reasons behind a faith in, and a commitment to, deliberation and/or communication differ. Deliberative models imagine an active citizenry oriented toward reaching a consensus or the common good, who address collective goals, aims and decisions by a process of public discussion. Deliberation encourages the public articulation of preferences toward publicly justifiable goals. In this scheme, individuals and organizations present their arguments in a forum of free and open dialogue and are subsequently challenged by others who test, by a process of public reasoning, the arguments given. The deliberative process continues until agreement about the common good can be reached. In this sense deliberation draws people away from an atomized private means of will formation toward a public defence of their claims, which are accepted or rejected only by the force of the better argument. Taken together this amounts to the following claims. First, deliberative forms of democracy have a better chance of incorporating ecological values. Second, communication enables far better informed decisions because, third, unlike aggregative mechanisms they are not reliant upon the private, untested vote. Fourth, deliberation encourages/results in consensus, and finally the nature of deliberative institutions and/or the process of deliberation itself transforms preferences.2 These are weighty and challenging claims and we shall deal with each in turn. First there is the question of values. Habermas and Cohen are clearly concerned with locating a procedural ethic which leaves the values that may be deliberated upon discretely factored out. However, in some models of deliberation there is a commitment to locating the common good using values implicit in the public culture of political communities (Fishkin 1991; Barber 1984; Sunstein 1993). It is doubtful whether any democratic forum, deliberative or otherwise, is value-free and it would appear best to proceed on the basis that particular values will always inform particular political processes. Factoring in decision rules such as the precautionary principle is not necessarily impossible, but may be problematic depending on which particular form of deliberation one opts for. More

140

Mike Mills and Fraser King

pertinently, the values present in the relevant community may not include the types of ecological values that we describe above. As Smith (1999) argues, political ecologists can have no guarantees aside from the fact that they will have a voice. This does not seem to us a firm foundation for the promotion and achievement of vital ecological values. Second, the conditions for deliberation demand a level of communicative competence which appears to be quite narrowly constituted. This restriction to all reasonable claims that may be regarded as reasonable by others (Cohen 1996), leaves little room for manoeuvre for political ecologists. Habermas in particular restricts this rationality to communicatively competent persons. Dryzek (1995) and Eckersley (this volume) offer a far more expansive notion of communication, which seems much more helpful in terms of expanding the moral community and the form of moral weight we place on reason or rationality. This may help the promotion of ecological values in so far as the circle of communicative rationality is widened, as this would tend to place appropriate restrictions on deliberative outcomes. However, this in itself provides no guarantee against the possibility of acts destructive of nature or dangerous to human well-being. Third, there is the question of removing aggregation from the democratic process. The general theme of the deliberationist case is to contrast the aggregative with the transformative. But what is actually meant by removing the aggregative character of democratic decision-making? Saward (1998a, and this volume) has expressed concern about these constructions and argues that, surely, at some point a decision has to be made. Cohen (1998) rightly contends that any intelligent model of decision-making needs discussion (regardless of the level), but by the same token any credible model of democracy has, at some point, to aggregate preferences. Whilst the concern for a deeper process of public reasoning is laudable, this seems to take place wholly within existing aggregative structures (Saward 1998a: 64–5). Fourth, the deliberative case normally rests upon the claim that a consensus or agreed decision is the outcome of the deliberative process (though see Dryzek, above, for various ways of looking at the status of outcomes of deliberation). This conclusion does rather rely on the premise of leaving subjective preferences behind being true or reasonable. We find this claim less than compelling. Dobson argues that there can be no guarantee that free and equal conversations will result in greater value or regard for non-human nature (Dobson 1993: 198). The point could be taken further still. Bronwyn Hayward (1995) follows Young (1995) in suggesting that reasoned argument itself would tend to privilege some interests over others. Notwithstanding this, consensus tends to be reached far more easily in societies marked by high levels of homogeneity. Since very few modern industrial societies conform to this characteristic, why would deliberation not produce polarization as opposed to agreement? It seems also that this point is central to the claim that deliberation has the potential to transform preferences. We do not doubt that in some circumstances preferences can be transformed by face-to-face dialogue; however, there is a degree of doubt about how far this may extend, and how lasting such a trans-

Ecological constitutionalism, deliberation and representation 141 formation may be. The underlying assumption seems to be that in aggregative systems people do not leave their subjective preferences behind when they vote, but during deliberation they will. We suggest that at the very least the possibility of the public conflict of subjective preferences still remains and that these may not lead, as deliberationists believe, to consensual outcomes. Similarly, models of deliberation offer little opportunity for many people to participate – at least in the more formal, ‘artificially’ constructed models – so even if preferences are transformed, this need not indicate a broader, societal change of consciousness or preference. This point is part of the issue of the siting of deliberation, which Saward maps out in Chapter 5 above. Although deliberation may give political ecologists more voice and access to the decision-making process it does not necessarily connect those values to the democratic process. By contrast and more weakly, it simply allows for the case to be made. The rejection of deliberative forms does not necessarily mean that the liberal democratic paradigm is exhausted for political ecologists. Saward (1998b) has suggested that the task for ecological theorists is to imagine what an ecological state might look like. For Saward, the ecological state must be conceptualized within liberal democratic norms; we are happy to go along with this and suggest that the more productive path for political ecologists is to use the vehicle of the constitution to further ecological goals.

Deliberation, risk and values Let us return to ecological values and to a relational metaphysic. We said earlier that this is the fundamental insight of ecological thinking and the basis upon which all other values are built. Commonly, this is expressed through the concept of risk. Risk encapsulates if not the full metaphysic then certainly the tendency to recognize interdependence within and between species. The concept of risk is therefore a useful one as far as political ecologists and deliberation are concerned because if the deliberative process is one which appears able to deal adequately with issues of risk (in the ecological sense) then it is reasonable to suggest that ecological goals and ecological values will be similarly well-served. Needless to say, the converse might also be true. Beck (1992), though not a political ecological theorist by our definition, suggests that current risks have a particular character: they are pervasive, that is to say global, irreversible, incalculable, uninsurable, of unlimited reach, dependent on scientific detection, and an unintended consequence of the normal operation of our society. Such risks would include those associated with civil and military nuclear capabilities, the effects of inorganic pesticides and herbicides, genetically modified foods, the ingestion or inhalation of toxic substances from vehicle emissions and chemical plants, as well as things like global warming and the hole in the ozone layer. This would be very much in accord with ecological thinking, but of course there are also more local and perhaps more tangible forms – road safety, the siting of large industrial plants, health and safety at work, the local use of agro-chemicals, food hygiene, the disposing of household or industrial waste,

142

Mike Mills and Fraser King

and so on, are the primary local manifestations of these more pervasive risks. Beck’s position suggests that part of the political response to these risks should be the development of a more deliberative means of decision-making which will circumvent the obvious difficulties associated with markets, ‘experts’ and the state. But from where does this faith in the deliberative process come? Could a transition to deliberation resolve problems of risk-taking? In the literature on risk there is a strong argument which suggests that people’s perceptions of risk are in part determined by their perceptions of whether risks are justly distributed or justly arrived at (Douglas 1985). Perceptions of acceptable risk are tied to the fairness of the risk to which people are exposed. It is a short step from this to suggesting that a just social and political system would be one which actually lowered the aggregate level of societal risk because typically risks are not distributed fairly (the poor, for example, are invariably exposed to higher risks). This equality principle fits very well with views on the democratization of decisionmaking – to be open, accountable, responsive, and to value equality is, on some interpretations, to be just and fair. So, it is not difficult to see how the democratization of the decision-making process might be seen as a means of lowering risks – we would be more aware of the risks that existed; we would understand that a fairer exposure to risk could generally be achieved only if more people experienced less risk; we would be clearer on expert and commercial input into both political and economic decision-making, and so on. Equally, there is empirical evidence to suggest that moves in the direction of democratization do seem to have ecological benefits. Not only do environmental political issues tend to push decision-making in a deliberative direction, but where this does happen, policy is ‘better’ as a consequence. Janicke (1996: 77) argues that ecological questions have fostered political modernization in countries such as Japan, The Netherlands, Sweden and Germany. Further, the changes he cites (more participation, decentralization, self-regulation, intervention, and dialogue structures) all have a deliberative feel to them. Political modernization may indeed involve changes to current political practice, and greater deliberation may be one of these changes. However, it is also true that much of the benefit highlighted by Janicke was simply reform of aggregative systems – the same may be said of some of the proposals of Beck. It seems to us that it is the conditions of deliberation which are producing the political benefits in these instances rather than the practice. This is a process within an essentially aggregative system, a system which retains those core structures within which broader or more pervasive social values may be embedded and around which it is possible to anchor otherwise heterogeneous political demands. If Douglas is right, and culture is central to the ability of societies to resolve risk questions, then cultural differences in multicultural societies serve as a barrier to the resolution of risk-based conflict because they indicate a heterogeneous valuebase. Campbell (1985), for example, shows clearly that amongst experts the willingness to give advice in the face of uncertainty is not primarily a function of expertise itself, but of other values. Our preference is to see pervasive and

Ecological constitutionalism, deliberation and representation 143 occasionally chronic ecological problems not as issues of deliberation alone but rather as issues of constitution where extra-democratic values, bound by the minimal conditions of democracy, set the boundaries for the working of aggregative and deliberative systems alike. While it may be that the retention of subjective preferences helps to reduce the risks which any individual is willing to subject themselves to, this may not help the more general difficulty of reducing risks to others given the pervasive nature of the risks themselves. It is this shift from working with individual subjective preferences to taking others into consideration that is the major weakness of deliberative models. As far as contemporary risks are concerned it is simply not feasible to leave these issues to individuals’ subjective preferences. This is not an argument for authoritarianism, but rather one meant to suggest that a suitably qualified democratic (and deliberative) process may suit political ecologists’ purposes better than they think. It is in the nature of risks that we need certain values even to define them; accordingly, constraining the deliberative process in line with certain overriding values will be necessary in order to identify risks, hazards and uncertainties. Indeed, Douglas is clear that culture is central to the identification of risks and to the resolution of the conflicts that arise from them. It is the institutions and conventions of societies which indicate who should be consulted, what a hazard looks like, how we assign responsibilities, and so on (Douglas 1985: 68). It is difficult to see therefore how a deliberative process could aspire to neutrality, how it could avoid recognizing intrinsic cultural values or aggregating diverse value systems in multicultural societies. Moreover, it will often be the case, we suggest, that those values which are necessary to define social and ecological goals will exist in addition to those necessary for democracy. It will certainly entail consideration of future generations and may well include the well-being of non-humans (as might be the case in deliberating over factory farming, for example). Rather than skew the democratic process, we follow Sagoff (1992) in believing that the decision-making (deliberative) process is, in fact, skewed when the existence of extra-democratic values is denied. The question is not whether values should be factored in, but how. But there is also an ethical argument in favour of constraining the deliberative process (and the democratic process more broadly). Tim Hayward (1997: 67–85)3 makes the point that no act is one of moral indifference – we must always reflect to establish the consequences of our actions. If we have moral obligations to others then it will always be the case under conditions of uncertainty or risk that we are obliged to take the least risky option because we do not know what the consequences of doing otherwise might be. While the arguments for deliberation cited by Bronwyn Hayward (above) all have a strong ethical basis, here we have an equally compelling argument suggesting that regardless of the deliberative process there are certain things we cannot or should not do. It is perfectly possible to hold experts accountable, to exercise self-determination, to be transformed by political participation, and to experience our own sense of the good life without deliberating on all things, all of the time. We may choose

144

Mike Mills and Fraser King

to ring-fence some policy areas, or to constitutionalize values (such as those represented by the precautionary principle) in order to set clearly the boundaries of deliberation and, at most, appeal to values other than those of deliberation itself. To us, political ecologists have little to lose, and much to gain, from relying less on the deliberative process to produce the outcomes they would like. There are other mechanisms, compatible with the self-binding nature of democracy, to which they may look. The strength of constitutionalism from an ecological point of view is its essentially prescriptive character – unlike deliberation, which by definition leaves political outcomes relatively open. Most deliberative models are content to claim that the values which deliberation unearths are those that are common to the public political cultures in which deliberation takes place. However, this depends upon extended deliberation actually excavating the values that are central to ecologists. They can only do so, it seems to us, if those values are in some measure integrated into the constitutional order of the state. As we argued above in the section on risk, these are not issues of limited interest but relate to the all-pervasive character of ecological risks and hazards. In many ways this is tied to what a coherent vision of the ecological constitutional state might look like, and the notion of the state playing a principal role in fashioning the value system is something that traditionally political ecologists have shied away from. However, political ecologists ought not be concerned that constitutionalism is potentially a ‘top–down’ solution. As we have already argued, democracy itself is meaningless without a strong bind to prevent outcomes that threaten its preconditions. If democracy is to be meaningfully connected to ecological values (which the process of deliberation does not necessarily facilitate) a constitutional clause (like, say, the precautionary principle – O’Riordan and Cameron 1994) or a set of environmental rights (Hayward 1994) takes us more strongly in that direction. Indeed, a commitment to a ‘bottom–up’ process of decision making is likely to be enhanced by a political system not engineered in an ecological vacuum. This does not of course mean that there is nothing to be decided or indeed that deliberation is not necessary. Rather, it means that the nature and context of the deliberative (or indeed any democratic form) process has to be guided by a set of values that have more ecological certainty about them than do discursive forums. In other words, the boundaries of deliberation have been set by the constitutional order. The criticisms that have already been made of the deliberative process cannot so readily be made against constitutionalism. The two key claims that have been refuted – that deliberation increases the ecological voice and enables preference transformation – can be easily incorporated into a political system if ecological values are constitutionalized, for example, in the form of environmental rights or the precautionary principle. Again, this should not be seen as a restatement of eco-authoritarianism, but rather as a development of liberal democracy. This is of course to be taken with a dose of realism: we are not using the constitution as a means of finding an ‘ecological state’ but rather something more dynamic and practical – a more ‘ecological’ state that is also democratic.

Ecological constitutionalism, deliberation and representation 145

Conclusion We have argued that the ecological commitment to deliberative democracy may not secure the benefits that political ecologists suppose. There is a strong critique to be made of deliberation which suggests that many of its promises need to be supplemented by a commitment to constitutional considerations such as environmental rights and the precautionary principle. This conclusion is predicated upon a consideration of risk. Risk reflects what might be seen as the key ecological value (a relational metaphysic) and because of this we can imagine the deliberative model having a disaggregating effect as far as the accommodation of ecological and social values are concerned. We have suggested that some aspects of the aggregative system – notably those associated with constitutionalism – are still more favourable to ecological values.

Notes 1 The perceived conflict between democracy and constitutionalism, usually played out with reference to Madison and Jefferson in the debates over the US constitution, has plagued discussion of the two subjects. 2 Eckersley touches crucially on many of these points; see the previous chapter. 3 To be fair, Hayward does not believe that this principle can be applied to ‘the environment’ but, at the most, other species: for him there need to be ‘morally relevant similarities’ (1997: 70) between agents for the argument to be intelligible.

11 Governance, selfrepresentation and democratic imagination Henrik Paul Bang and Torben Bech Dyrberg

The relationship between individual and community is a classical theme in democratic theory. To search for innovations in democratic theory calls for re-thinking this relationship. In modern political theory this relationship has predominantly been referred to as the terrain of, and the challenges to, government. The theoretical set-up has mostly been centred on the oppositions between public/private and state/civil society. These oppositions have been related to discussions of the nature of politics in general and democratic politics in particular. The typical oppositions have been those between the common good versus neutrality, the instrumental versus the normative, power versus autonomy, and ‘thick’ versus ‘thin’ conceptions of individual and community. In recent years the individual/community dispute has turned in the direction of a discussion of how to improve deliberative democracy and civic democratic engagement. The practical interest of political analysis is no longer exclusively concerned just with improving the quality of governmental decisions and actions. The expertise provided by political analysis has turned to the task of how to bring the people back into politics and the making of policy. As Fishkin puts it, ‘If we believe in democracy, can we somehow get citizens who are more prepared to exercise public responsibilities’ (Fishkin 1997: 14)? The new empowerment ambition of political theory and research does not fit into the opposition between individualism and collectivism, and the understanding of politics as a trade off between instrumental success and normative integration, self-interest and the common good, efficient political steering and citizenship. There is no longer an abyss between public decision-making and ‘the politics of amateurs, where every man is compelled to encounter every other man without the intermediary of expertise’ (Barber 1984: 214). Experts are not a threat to citizens, but are instead conditional for establishing ‘a form of connectedness to the system that expresses our collective political identity’ (Fishkin 1997: 44). In contrast to this kind of elite-staged empowerment – where deliberation and civic engagement aim at creating an informed understanding of the way elites handle public concerns – we have empowerment from below, where the participants define the aims of civic engagement and deliberation on their own terms. In such a radical bottom–up perspective there is no inherent opposition between expertise and lay involvement. Lay-actors can draw upon experts in their

Governance, self-representation and democratic imagination 147 definition of the situation, but it is up to them to decide the conditions of the exit and re-entry of expertise. The distinction between these two perspectives forms part of a shift from government to governance. By this we mean a context in which ‘the system’ is no longer unified, but refers to an ensemble of political relations of autonomy and dependence between elites and lay actors crossing established boundaries between public/private and state/civil society (March and Olsen 1995; Rhodes 1997). Governance throws doubt on traditional ways of conceiving of democracy in terms of stable and efficient government in the state, on the one hand, and dialogue, consensus-building, emancipation, and voluntary organizing in civil society, on the other. Within a governance perspective, voluntary organizing, for example, can either be part of expert systems, aiming at making political steering more efficient and responsive, or part of lay involvement aimed at creating identity, self-governing and free public spaces. We would thus be sceptical of Fishkin’s argument that the alternatives in present-day politics amount to ‘rational ignorance’ versus ‘an engaged and informed public’ that is able to ‘to think seriously and fully about public issues’ (Fishkin 1997: 41, 43). The problem with this way of addressing the issue is that the only alternative to expert-induced engagement is the passivity of ordinary people. This in turn implies that expertise attains the status of a meta-principle of democracy by bridging efficiency and legitimacy, and by initiating public deliberation and civic engagement from above. There is no room within this conception for a politics of the ordinary. The possibility of connecting lay action and experience with expertise in terms of the ordinary is thus ruled out from the outset. However, Barber’s alternative is no better: it is effectively the mirror-image of Fishkin’s argument. The point is not to get rid of experts to achieve ‘strong democracy’, where ‘the citizen is by definition a we-thinker’, who appears ‘when “masses” start deliberating, acting, sharing, and contributing’ (Barber 1994: 214, 215). What we need is a model of deliberation that allows for self-governance and identity-formation on the part of both experts and lay actors, and which moreover recognizes that co-governance between them requires acceptance of their differences with regard to their individuality and commonality. Rawls’s political liberalism points in the direction of a form of deliberation that avoids dichotomizing expertise and lay involvement. Even though Rawls is associated with a liberal government perspective, his arguments do carry important insights as to how to connect deliberation from below and from above in a context of governance. The ‘reasonable society is neither a society of saints nor a society of the self-centred. It is very much a part of our ordinary human world, not a world we think of much virtue, until we find ourselves without it’ (Rawls 1993: 54). Contrary to Barber, reasonable society is a matter of both individuality and commonality; and contrary to Fishkin, expertise is claimed to originate from ‘our ordinary human world’. For Rawls the context of deliberation is not an elite construct, just as it is not simply the domain of amateurs. Rather, it is rooted in a politics of the ordinary.

148

Henrik Paul Bang and Torben Bech Dyrberg The rational experts I

Individuality

II

Public reason

IV

Commonality

III The ordinary amateurs

Figure 11.1 Democratic options

‘In justice as fairness there are no philosophical experts. Heaven forbid! But citizens must, after all, have some ideas of right and justice in their thought and some basis for their reasoning. And students of philosophy take part in formulating these ideas but always as citizens among others’ (Rawls 1993: 174–5). The connection is the reasonable that mediates between the rational and the ordinary. Figure 11.1 shows four democratic options. Liberalism emphasizes the connection between individual autonomy and the expertise of the rational choice (I). Republicanism focuses on the relation between appropriate leadership and the common good (II). Communitarianism stresses the link between the common good and the civic involvement of amateur citizens (III). Rawls opens the door to (IV) by insisting that expressive individuality is constitutive of being an ordinary citizen. Fishkin seems to represent a fusion of (I) and (II), whereas Barber relates more explicitly to (III). We would suggest that democracy be viewed full-circle and within a perspective of democratic governance rather than of democratic government. In the following we shall, first, specify what is involved in the shifts from government to governance. Then we will identify some of the threats and challenges facing democratic governance today. Here we focus on two new political identities, namely those of demo-elites (I, II), and everyday makers (III, IV) which combine individuality and commonality, privileging either rational experts or ordinary amateurs. This brings us to Rawls’s political liberalism and his public reason argument. We shall in particular look at the relation between public reason and political community as the basis for building political capital rather than social capital.

Governance, self-representation and democratic imagination 149

Identifying the shifts from government to governance Governance identifies a new political reality where politics and policy are framed in more or less enclosed networks that restructure and intensify the connections between power, knowledge and politics (Rose 1996): Networks operating on the global, regional and local levels challenge the ‘pluribus unum’ of the nation-state. In the wake of these trends, the ideals of a sovereign centre and a sovereign people are thrown into question. Nation and people cannot alone frame politics, but are instead identified as political problems that give rise to new ways of drawing political frontiers. New ways of conducting politics occur in new locations. Politicians, administrators and organized interests engage in processes of negotiation in networks, often on an informal basis. At times they function in the shadow of the state, at others they make the state function in their shadow. The blurring of the division of labour between these agencies shakes the hierarchical relationship between them. The Weberian model of bureaucracy as a rationally unified and hierarchical entity is fragmented, hollowed out or made polycentric as multiple institutions and ad hoc settings. Authority is no longer merely hierarchical, but is negotiated horizontally at various levels. Public administration is politicized, and is organized less and less according to strict procedures, and more and more in terms of practical estimates and in relation to concrete projects. Parliamentary debates are increasingly subordinated to what hegemonic elites manufacture as the imperatives of ‘necessary policies’. The elite triangle of power–knowledge–politics presents a threat to both the theory and practice of representative democracy. It affects the structuring of political identification; the extent to which lay actors want to get involved in politics; and the legitimacy of representative democracy. It is moreover important for democratic theory that political agenda-setting by political parties and parliament is largely taken over by other agencies, notably the media, government, administration, organized interests, voluntary organizations, and multinational companies, which interact in various issue and policy networks. Civil society cannot be seen in terms of the good life whose resistance to the system safeguards its integrity. Social movements and voluntary organizations face difficulties in portraying themselves as sources of emancipation. They are more and more woven into the politics of a negotiated economy crossing established boundaries between state and civil society, thus creating a grey zone between public and private.

Threats and challenges facing democratic governance What do these changes from government to governance imply for politics and democracy? Five tendencies seem important here. First, politics cannot be confined within specific settings, but has to be seen as a network including both formal and informal components. What is significant is

150

Henrik Paul Bang and Torben Bech Dyrberg

not distinctions such as state/civil society or public/private, but how to distinguish the political from the non-political, and the public from the non-public. Political domination is not only an issue of class power or state coercion, but also of exclusions from elite-governed networks. Second, the distinction between representative democracy in the state and participatory democracy in civil society breaks down. Citizen activists and social movements have increasingly turned into ‘expert citizens’ who are drawn into governance networks through which policy programmes are designed and implemented. And experts become more and more committed to empower lay actors for the sake of equipping them with knowledge of and responsibility for public policy. Third, the efforts of balancing effectiveness and legitimacy no longer exclusively concern the state or citizens struggling against it. Instead, stress is put on dialogue and co-operation between leaders, employees and users in institutions. This has led to a weakening of the state’s ability to conduct efficient steering and to frame political identification. Civic engagement is politicized and oriented less towards the state and more towards a search for identity and the possibility of exercising political power in local institutional settings. Fourth, rank and file activists react to the professionalization of voluntary organizations by insisting on defining the nature and intensity of their involvement themselves. They do not want to be engaged on a full-time basis, but when and as long as they feel like it and find it relevant. They engage as much from enjoying their ability to make a difference (power) and to develop themselves as for moral and collective reasons. They often resist being committed to a common good, but want instead to be part of a political community that allows for differences. Finally, visions of the democratic public as a uniform political morality disintegrate as a regulative ideal. Agreement on rational procedures and identification with comprehensive doctrines are replaced by what is considered more basic to politics, notably that of recognizing and accepting difference. Ideas of the common good are replaced by common concerns, which are neither thick nor thin, but simply ordinary; the private disappears as the site of a univocal self and is deconstructed as a plurality of non-public selves. How can we know that these five tendencies are actually operative? They are partly confirmed by findings in the Danish research programme ‘Democracy from Below’. The centrepiece of these studies has been the changes in political identity and civic engagement experienced in an inner-city neighbourhood in Copenhagen – traditionally a stronghold of left-wing parties and grassroots movements. Two new threats to democracy are identified in these studies, which connect to the enrolment of voluntary organizations in elite networks. The first concerns the exclusion of lay actors and their conventional wisdom from elite deliberation and negotiation. The second deals with the tendency of lay-activists to turn away from the ‘big’ politics of experts in governance networks and to concentrate on ‘small’ projects related to daily life in their locality (Bang and Sørensen 1997; 1998a, b, c).

Governance, self-representation and democratic imagination 151 The first threat is that there are tendencies to discredit lay actors, tendencies which derive from the privileging of expertise and the struggle for influence in networks composed of leaders from public, private and voluntary organizations. These elite networks show signs of closure and self-sufficiency. Their relative autonomy within the state and from the state, and particularly from the government, has produced the belief that there is a close relationship between the autonomy of elites and the principles of democracy. According to Etzioni-Halevy (1993): •

•

•

elites do not constitute a coherent block or unified structure. They should rather be regarded as relatively autonomous entities more-or-less loosely coupled in networks of governance; elites need not always belong to the established elite. Elites from social movements and voluntary organizations can make a significant political difference even when having no access to the negotiated consent of the established elites; and elites can facilitate the building of public spaces and democratic identities. They can do so even though strong interest groups or other key elites do not back them. Civic virtue and rhetorical skills can compensate for a lack of legal or material resources.

Relative elite autonomy is the principle of democracy. It is the basis of civil association, civic culture and the public sphere, as well as of the rules, procedures and power of the state. Democratic governance by elites is the combined result of relations within and between elites, between elites and government, and between elites and the public. These forge governance structures facilitating self-, co- and hierarchical governance. The second threat is that one reaction to the demo-elitist credo is the retreat of lay actors from party politics. This retreat is even indicative of a new type of activist, who has been termed the ‘everyday maker’ (Bang and Sørensen 1997), and has also been found in Swedish research (Montin 1998). Everyday makers resist the professionalization of civic engagement. They regard it as a matter of course to have the right and the power as laypersons to cope with day-to-day political problems in a conventional and practical manner. They want to do things by themselves, where they are, on their own terms and for their own purposes. Their image of democratic political community is a politics of the ordinary, which is ‘weak’ enough to allow for determinations by both elites and non-elites, and which is ‘thin’ enough to provide space for their various ideas of the good. Everyday makers have thus come to defy ideologies and types of activism that are full-time. They engage when they have the time and feel like it, and solely for the sake of solving concrete problems. Bang and Sørensen have classified the everyday maker on the basis of their own statements in twenty-five extended interviews: •

Do it yourself: ‘If we can make it run by ourselves, it would be fine to do it ourselves, wouldn’t it?’

152 •

•

•

•

•

•

Henrik Paul Bang and Torben Bech Dyrberg Do it where you are: ‘It’s important that you’re active where you are; and this is very political. I’m not too crazy about the headlines. In reality, I don’t think they are particularly significant.’ Do it for fun, but also because you find it necessary: ‘I have tried to find some things which combined what I thought would be fun to do and what I found necessary. So I’ve tried to combine it, you see. And this I think is quite proper, so that one is not just trying to polish one’s halo, but is also getting something personal out of it.’ Do it ad hoc or part-time: ‘It is exciting to learn how one can access and try to influence some things and situations’. But I am ‘not one of these fulltime grassroots who are always going on weekend trips to discuss something.’ Do it concretely rather than ideologically: ‘It’s . . . a personal thing, but I find that I become more engaged and more active, when it’s something more concrete’. ‘I have nearly no relationship to party politics’. ‘I vote for a party, because it’s the way it’s set up when I go to the ballot box when there’s an election.’ Do it responsibly and trust yourself: ‘Well, the responsibility I take for my life will probably show the influence I have. If I’d do something else, then I could move, or I could do things so that my situation became different. You can’t avoid feeling pressured by this and that. That’s life.’ Do it by looking at expertise as an other rather than as the enemy: ‘I don’t consider my work in the day-care board as determining for the way the institution is run. Well, there is a budget. This we cannot touch. The frame is given, and then I also think that the staff has greater competence. We’re heard, and that’s OK.’

Studies of everyday makers and demo-elites show how experts and lay-actors become increasingly interwoven in issue and policy networks that cannot be confined within the boundaries between public/private, state/civil society and national/international. It would be futile to set out to mediate these oppositions by trying to re-establish the boundaries between state and civil society, and the division of labour between politicians, bureaucrats and citizens. Elites and lay actors do not require the presence of the state for mediating between private and public interests. They simply presume a political logic for holding each other accountable to the practical deliberation that makes the difference between what can be regarded public as against non-public. Then how are we to cope with the threats of exclusion and retreat from ‘big’ politics that result from demo-elites and everyday makers? We need new approaches to democracy that neither make political decisionmaking the business of elites in the state nor make citizenship the prerogative of amateurs in civil society. In a network society, politics has no unified or univocal centre of publicly assessable control of political power. Nor does it carry any ‘grand narratives’ of domination and emancipation with clear-cut political frontiers between power and resistance (Foucault 1980: 142, 1981: 95–6).

Governance, self-representation and democratic imagination 153 Rawls provides a way of mediating the tension between expertise and conventionality. Hence, we turn to his concept of public reason, which disentangles democratic theory from the dualism of perfectionism/anti-perfectionism, the rational/the normative and power/autonomy. This makes it possible to rethink the nature of politics as well as the relation between political authority and political community. It should also be possible to sketch how a Rawlsian approach undermines protective and elitist models of democracy. This in turn calls for rethinking the parameters of democratic theory from a non-elitist perspective.

Demo-elites/everyday makers and Rawls’s political liberalism The relation between elites and everyday makers is an endogenous political relation around which democracy revolves. Rawls’s political liberalism provides a way of discussing the difference between public and non-public governance without lapsing into old dichotomies between power and knowledge. We would wish to make various key claims about Rawls’s theory, as follows: •

•

•

it makes most sense if public reason is understood as referring to developing and sustaining reasonably balanced relations of autonomy and dependence between political authorities and ordinary political agents in a democratic political community; it breaks with mainstream liberal views of the political as an independent sphere and a neutral level above society, without endorsing a communitarian position that sees the political as a medium for realizing the common good; it shows that protective democracy is responsible for producing the oppositions between perfectionism/anti-perfectionism, normativity/rationality and power/autonomy by denying the possibility of viewing politics as a dimension of social relations, which can be both public and non-public.

Rawls’s political liberalism can be seen as being based on three interconnected claims, which advance beyond the parameters of the liberalism/communitarianism debate, and which are implicit in the civic engagement adumbrated by everyday makers. First, the distinction between anti-perfectionism and perfectionism does not have to be coined in terms of neutrality and the common good. Political authority has a power of its own that conditions the appropriate use of public reason. It is in this respect we can talk about recognition and acceptance of difference as the bottom line of democratic politics. Second, individual reason and conduct vis-à-vis others in a community do not have to be viewed in terms of individualism/egoism and collectivism/altruism. Instead, the relation between public reason and comprehensive views can be seen as an articulation of individuality and communality (Foucault 1988; Strong 1994). Third, individual and collective freedom can never be freedom from political power, but can be obtained only by engaging in common political concerns, which

154

Henrik Paul Bang and Torben Bech Dyrberg

are neither ‘thin’ nor ‘thick’, but first of all conventional or ordinary. This in turn requires mutual recognition and acceptance of difference. These three claims point away from the typical reading of public reason as geared towards a narrow focus on state, government and the enforcement of law. In Political Liberalism Rawls contends that ‘[t]he nature of public reason will be clearer if we consider the differences between it and non-public reason. First, there are many non-public reasons and but one public reason’ (Rawls 1993: 220). The argument in favour of ‘one public reason’ is not that there are no politics outside state and government, or that individual autonomy demands that we should protect ourselves from political power. Rather, it is that ‘we may over the course of life come freely to accept, as the outcome of reflective thought and reasoned judgement, the ideals, principles, and standards that specify our basic rights and liberties, and effectively guide and moderate the political power to which we are subject’ (Rawls 1993: 222; see also Burt 1994: 139–42). Rawls points to the relation between the democratic political constitution and democratic political culture. This implies looking at the relation of the state or government to the demos as a whole in which notions such as authority, reasonability, tradition, confidence, reciprocity, and trust play a significant role (Rawls 1993: 86, 163). Public reason is not simply about the political regime and its democratic constitution. It is, instead, intended ‘to connect a particular understanding of freedom and equality with a particular conception of the person thought to be congenial to the shared notions and essential convictions implicit in the public culture of a democratic society’ (Rawls 1993: 369). Rawls’s argument is congenial to a network point of view discussing the potential of, and the threats posed by, the new political identities of demo-elites and everyday makers. The task is to expose the power of their political networking across established boundaries of state and civil society to public reason. It is this possibility of keeping non-state or non-governmental political forces accountable to public reason that is opened up by Rawls’s ‘political turn’.

The public reason argument For Rawls, democracy requires much of its citizens: they must decide and act reasonably; regard each other as free and equal; be ready to grant one another fair terms of co-operation in light of what they consider appropriate; and agree to act on these terms even when the latter conflict with their own interests, provided that all others do so as well. Although relative elite autonomy is important for democracy, pluralism cannot safeguard democracy because the latter demands that elites are citizens first and experts only second. Whatever elites choose for making and implementing efficient collective decisions, these are binding only if they are publicly available for deliberation by everybody (Rawls 1997: 773). The idea of public reason starts ‘by looking to the public culture itself as the shared fund of implicitly recognized basic ideas and principles’ (Rawls 1993: 8). This implies an idea of a democratic political community as integrated by the recognition of common political principles rather than solely by constitutional

Governance, self-representation and democratic imagination 155 essentials or comprehensive views. In such a politically and democratically constituted world, reasonable people will always disagree. This is so because public reason is itself a source of conflict (Rawls 1993: 44; Burt 1994: 133), and because it can never be dissociated from comprehensive views. Rawls’s innovation in democratic theory lies in the fact that public reason is not confined to the dualisms of perfectionism/anti-perfectionism, individualism/ communitarianism and power/freedom. This means that (a) power is constitutive of freedom; (b) individuality and commonality mutually condition each other; and (c) long-lasting controversies are at the heart of an overlapping consensus of ideals. Public reason ‘is a way of reasoning about political values shared by free and equal citizens that does not trespass on citizens’ comprehensive doctrines so long as those doctrines are consistent with a democratic polity’ (Rawls 1997: 807). Rawls’s argument is congenial for dealing with the democratic challenges posed by the emergence of new ways of conducting politics, which are marked by profound and lasting tensions between the expert politics of elites and the everyday politics of laypersons. The endogenous view of democratic politics with regard to both decision-making and civic engagement implies that political problems cannot be adequately grasped in terms of oppositions such as those between individual preferences versus common interests and the rational versus the normative. These oppositions are insufficient to deal with the questions of building democratic political communities in the face of ‘the fact of pluralism’. Rawls’s public reason is geared toward a political conventionality, which is recognized and accepted as an integral part of social and personal life (Gunnell 1998; Strong 1994). This requires a new focus for discussing the relation between individuality and communality that does not conceive politics as a trade-off between economic and sociological categories (Bang and Dyrberg 1997: 52–4).

Public reason and political community Rawls rarely uses the term ‘political community’, and then only negatively as a reference to a comprehensive doctrine. However, to speak of political community one does not have to engage in ‘Gemeinschaftschwärmerei’. Easton (1965b: 177), for example, defines it as: ‘that aspect of a political system that consists of its members seen as a group of persons . . . who are drawn together by the fact that they participate in a common structure and set of processes, however tight or loose the ties may be’. Political community thus conceived comes close to what can be experienced in the case of everyday makers. Here it does not matter whether the members form a social community in the sociological sense of a group of people who have common interests, a widespread sense of social solidarity and share comprehensive views. People may lack a sense of belonging together in terms of a common good, as the separation of the ‘high politics’ of demo-elites and the ‘low politics’ of everyday makers suggest. Yet, despite controversies, exclusions and separations they may still share the awareness that ‘as long as they are part of the

156

Henrik Paul Bang and Torben Bech Dyrberg

same political system, they cannot escape sharing in or being linked by a common division of political labor’ (Easton 1965b: 178). Rawls’s idea of public reason would gain from being related to Easton’s concept of political community. It could help underpin the view that to speak of public reason as free-standing is to speak of the democratic regime as constituted by the political interaction between political authorities and lay actors. At first this may seem to contradict Rawls’s argument that ‘justice as fairness does indeed abandon the ideal of political community’ (Rawls 1993: 201), which points in the direction of regarding politics as simply a means for individual or collective ends. However, Rawls just as clearly denies this instrumental view, stating that ‘in the well-ordered society of justice as fairness citizens do have final ends in common’ (Rawls 1993: 202). Moreover, he is adamant in stressing the importance of the ordinary, the conventional, civic traditions, political capital, and so on, which are crucial in speaking about political community. It should thus be feasible to conceive of a political community that pays allegiance to the political values of liberal democracy without invoking an idea of the common good, whilst it at the same time accepts the existence of incommensurable differences of values, practices, interests and forms of life.

The building of political capital Rawls’s view of democratic political community as the terrain for developing, accumulating and nourishing political rather than social capital finds a parallel in the political practice of everyday makers. Rawls’s capital is not social in the sense of taking place outside the reach of politics, nor does it merely serve to keep the state effective and responsive. Instead, it is a political capital, which is accumulated within the political community, and is not made possible by people outside who ‘demand more effective public service, and [who] are prepared to act collectively to achieve their goals’ (Putnam 1993: 182). ‘The term capital is appropriate in this connection’, says Rawls (1993: 157n), ‘because these virtues are built up slowly over time and depend not only on existing political and social institutions (themselves slowly built up), but also on citizens’ experience as a whole and their knowledge of the past. Again, like capital, these virtues depreciate, as it were, and must be constantly renewed by being reaffirmed and acted from in the present’. They call attention to the existence of a relatively autonomous democratic political community. In such a community the liberty of equal citizenship is rooted in a civic tradition that connects the institutional and the personal dimensions of politics and cultivates our sense of fairness. What is publicly reasonable can only be determined in deliberations and struggles that revise it on an ongoing basis in a democratic public in which the expert public is considered different from but, nonetheless, interrelated with the lay public. For Rawls politics is seen as having a specificity of its own, and public reason as being freestanding vis-à-vis the reasoning anchored in comprehensive doctrines. If these two requirements are met we can accept (or reject) a political authority on

Governance, self-representation and democratic imagination 157 political grounds without endorsing or violating comprehensive views. For people can accept a political authority for all sorts of reasons, which becomes evident if their specific comprehensive reasons for this acceptance are put behind a veil of ignorance (Rawls 1993: 25; see also Easton 1953: 132–4; Easton 1965a: 50; 1965b: 207, 292–3). By articulating political theory and politics we can get an idea of what it means to say that it is difference – rather than unsustainable pleas for neutrality and independence – that underpins the anti-perfectionism of public reason. Everyday makers contain elements of political solidarity, which is compatible with even marked cultural differences. However, they have to be reconnected with ‘big’ politics to flesh out Rawls’s argument that the reciprocal recognition and acceptance of difference is modelled upon the idea of the overlapping consensus.

Final comments To search for innovations in democratic theory requires a view of the individual/ community relationship that reflects the transition from government to governance. Political governance is becoming more and more dependent on the abilities of local and global political actors and institutions to do things by and for themselves. The importance of this shift lies in its dislocation of traditional parameters of democratic theory, which assumes that politics has a particular location (the state) and a particular raison d’être (to exercise power over people and things). We have argued that a Rawlsian perspective is particularly well-suited to meet the challenges confronting democratic theory in the light of the shift from government to governance. The reason is that it gives full weight to the specificity of politics by focusing on the building of political capital in the context of the relationship between political regime and political community. It is this view of politics from ‘within’ that renders it necessary to interpret the anti-perfectionism of public reason in terms of common concerns based on the recognition and acceptance of differences. To engage in common concerns requires a change of authority from hierarchy to negotiation in and between organized expert systems that pay due respect to the pressures and needs for selfand co-governance from below. This in turn requires that freedom, far from being conceived as a flight from politics, as in most liberal and emancipatory discourses, presupposes an active involvement in politics. The figure of the everyday maker illustrates this. It draws attention to the need to empower lay actors as capable and knowledgeable political individuals who can speak and act for themselves and each other, but who also accept and are at least minimally committed to the systems that express and administer their collective political identity.

Part III

Associations and democracy

12 Active citizenship and associative democracy Piotr Perczynski

In the 1970s and 1980s the debate on citizenship seemed to be in its death throes. However, the last decade has proven that, to paraphrase Mark Twain, the news of the death of citizenship has been grossly exaggerated. A significant number of new theories of citizenship is being constructed, developed and reviewed.1 The reason for this renewal is that the notion has proved to be extremely useful in the current climate of political fragmentation, individualization and pluralization on the one hand, and political integration on the other, by providing a way to deal with the relations of the individual to larger (national or supranational) political entities (Kymlicka and Norman 1994: 352–4; Van Gunsteren 1998: 8–10). In uniting Europe, it was largely the introduction of ‘European citizenship’ (or, more accurately, citizenship of the European Union) by Article 8 of the Maastricht Treaty in 1993 which invigorated the debate, since European citizenship introduced new theoretical and practical problems (such as split or multiple loyalties) for a large part of the European population (Balibar 1996; Meehan 1993; Weiler 1999). Citizenship is a concept that is difficult to pin down, at the national as well as the supranational level, because it is derived from different political, cultural and legal traditions arising in different areas. Moreover, we can look at citizenship from different perspectives. In this chapter I am not focusing on narrow, ‘passive’ citizenship, which is understood just as a legal status, and hence is often used as a synonym of membership in a democratic nation-state and which ‘tends to mean nationality, the right to a passport’, etc. (Kaldor 1996: 22). Neither am I concentrating on the ‘social citizenship’ approach, which assumes that civil and political rights have already been established, and therefore concentrates on social rights, an approach closely connected to the now classical formulations of T. H. Marshall (Marshall 1950: 47–74; Van Gunsteren 1998: 106–9). Instead, I focus here on the active political citizenship understood as relationship between the individual and the political community, which stresses not only entitlements but also some obligations to participate in public life on different levels (Kymlicka and Norman 1994: 354–5; Dekker 1994: 11–12). The oldest, classical understanding of active citizenship comes from republicanism, which emphasizes the ‘double function’ of citizenship: governing and being governed at the same time. In addition, it underlines civic virtues such as sacrifice for

162

Piotr Perczynski

the community and willing obedience to law (Heater 1990; Riesenberg 1992). Civic virtues are here of martial (‘heroic’) nature (Filipowicz 1997: 113–140). Modern republicans preserve egalitarianism and civic virtues, but they stress – apart from the duty to obey law – the voluntary character of the obligations (as opposed to the obligations that any state, including non-democratic ones, imposes on its members). In this neo-republican approach, the civic virtues concern more the ‘un-heroic’, private part of human life (Van Gunsteren 1998; Walzer 1994). The idea of sacrifice for community as we find it in the classical republican theory has now been replaced by care for ‘the other’, referred to as ‘other-regard’, a form of civic consciousness that crucially involves taking the interests of fellowcitizens into account and, by so doing, making it possible to live together in the ‘community of fate’. Other new ideas in neo-republican theory are visible in the attitude towards the group. Rather than being anxious about the tyranny of (sub-)communities and (especially ‘deep’) groups, as the classical republican theory is, neo-republicanism tries to deal with probable frictions by ‘organizing plurality’. Neo-republicanism focuses on the process of societal change rather than aiming towards a pre-defined goal – the shape of future ‘republic’ is unknown and unpredictable. Active citizenship is realized by actually practising it; as Van Gunsteren puts it, neorepublicanism ‘concentrates on the actual situated exercise of citizenship’ (1998: 27). The more transformed and ‘privatized’ the republican virtues are, and the more positively modified the perception of the communities and groups within society, the more this modern version of republicanism – to paraphrase Jane Mansbridge – rightly uses the prefix ‘neo’.

Democracy for citizenship The ‘double function’ of active neo-republican citizenship – governing and being governed simultaneously – is closely related to the two main types of democracy: representative and participatory democracy (Held 1993; Dekker 1994: 12–14; Pateman 1970). The borders between the two types of democracy are far from precise (see Budge, this volume). For example, participatory, active elements are present in less mechanistic variants of representative democracy. The process of voting for representatives must be regarded as a participatory action in itself, just like consultation of representatives by the represented (Andeweg 1985: 96–105). Likewise, one could list representative mechanisms in several models of participatory democracy (such as special ‘functional chambers’ of parliament in functional models). Both types of democracy are relevant for different aspects of active (neo-republican) citizenship. However, because of the stress in this variant of citizenship on practising it, and because of the very nature of its double function, different models of the participatory type of democracy are especially important to analyse. One of the models that promises the most in this respect is associative democracy. The first reason for this assertion is that the associative model relates to the classic idea of perceiving associations as ‘schools of citizenship’. Another reason is

Active citizenship and associative democracy

163

the potential inclusiveness of associative democracy: the elements of two other major models of participatory democracy – direct and deliberative – are present in associative democracy. Crucial to the notion of citizenship and clearly apparent in associative democracy is also the problem of split loyalties and, closely connected to that, the importance of other-regard (or ‘external solidarity’). Below I shall discuss this in more detail.

Main designs of associative democracy The term ‘associative democracy’ is sometimes used in a very broad way by theorists to encompass many political or social designs that posit groups or associations as a point of departure, such as in communitarian, social capital or civil society theories (see also Roßteutscher, this volume). Being defined so broadly, associative democracy makes a very convenient target, for example when being criticized for inconsistencies. In my opinion, the term ‘associative democracy’ should be applied only to designs of participatory democracy in which key importance is attached to the individual participation taking place in the context of self-governing interest groups or associations,2 which in turn have some sort of democratic structure. Within the framework of society, those groups can inform, compete and co-operate in order to achieve their goals (i.e. the goals of their members). By doing so, they can also directly or indirectly provide the society with additional means of representation (many of the key interests of individuals are represented through the groups they are members of ) and regulation or governance (associations can play key roles in implementing certain government policies). Territorial representation could be seen as a special case of this type of functional representation: territorial interests are represented by territorial groups, i.e. the groups of individuals belonging to (living in) a certain area. Summing up, we could say that associative democracy is a model of participatory democracy based on self-governance of internally democratic, voluntary and functional groups (for a similar definition see also Streeck 1995: 188). The idea of associative democracy originates from different sources, which is one of the reasons why its character, the type of associations it involves, and their functioning in society differ from author to author. The designs of associative democracy were developed (quite independently) by several political theorists. One of the first contemporary designs came from Paul Hirst. He formulated in the second half of the 1980s his own original plan for associational reconstruction of democracy, by expanding and developing ideas of guild socialism and English political pluralism of the 1920s (Hirst 1989, 1993, 1994).3 Hirst preserves economic aspects of the pluralist doctrine and guild ideology, but he broadens the focus to make the model applicable to modern political systems. He envisions associations as spontaneously arising, democratically legitimated, selfgoverning, and (especially) voluntary (1993: 12, 1994: 19, 1997: 17). The state should have minimal power over their structure, and more broadly over what groups ought to exist. In fact, Hirst prescribes that the relations between authorities and the associations should take place on the local or regional

164

Piotr Perczynski

level, with (if possible) the omission or bypassing of the national state level (1994: 39). The role Hirst envisions for associations is quite broad. He sees them as taking active part not only in informing the government and representing interests of their members, but also in actual governance (1997: 17). They are to take over some of the important functions of the central government, especially as far as providing of services is concerned. Because of the complicated and highly individualized character of modern services, a democratic and voluntary association is (because of its scale/size, and short distance to the consumers) much better suited to provide them than the modern state, although for those who prefer it the latter can still remains a provider of services. Hirst believes that the functioning of such groups would be a solution for the shortcomings of both the liberal capitalist economy4 and parliamentary representative democracy, vulnerable to the dangers of the ‘tyranny of majority’. He considers associative democracy as a ‘vital supplement’ – rather than a substitute – for both (1994: 42, 1997: 24). In fact, he states that associative democracy can strengthen representative institutions – and thus representative democracy – because freeing them from an overload of activities allows them to concentrate better on their main functions, ‘to provide society with a framework of basic laws to guide social actors; to oversee forms of public service provision to hold public officials accountable; and to protect the rights and interests of citizens’ (1997: 18). We find a similar idea of supplementing representative democracy with associative democracy in the work of Joshua Cohen and Joel Rogers, who build their design on radically transformed corporatist ideas where interest groups play a central part. Like Hirst, they see the beneficial character of a certain type of group, namely internally democratic associations in which there is clear accountability of group leadership to members (1995a: 48). They mention various functions that such groups are capable of performing, and by so doing contribute to ‘egalitarian-democratic order’. Like Hirst, in their view associations can take an active part in ‘social governance’. The role of associations as ‘problem solvers’ is underlined – they are not only to help to formulate public policies but also to execute them. Closely connected to this function is that of providing information. Although some tasks are still reserved for the state – which plays a much larger role than in Hirst’s design – associations can provide the state (much better than state agencies) with accurate information on ‘member preferences, the impact of proposed legislation or the implementation of existing law’. This function becomes more important because, according to Cohen and Rogers, the state is today heavily involved in regulatory tasks and constantly extends its regulatory reach to more diverse and technically complex areas and rapidly changing processes. The function that affects the overall democratic system most directly is the function of ‘equalizing representation’ (1995a: 43). Associations can represent interests that are not sufficiently organized through ‘existing territorial politics based on majority rule’ (1995a: 43). The function that links Cohen and Rogers’s design of associative democracy most closely to the notion of citizenship is the function of civic education.

Active citizenship and associative democracy

165

Following a famous argument of de Tocqueville, Cohen and Rogers stress the importance of associations as a particular type of school where their members can learn in practice how democratic norms work (1995a: 43–4). To these two prominent designs of associative democracy we could add the (neo-corporatist) proposals of Philippe Schmitter. Like the previously discussed authors, Schmitter perceives his design as an addition to representative democracy. Associations in his view are ‘modern’ interest groups. Such interests are understood in the broadest sense – almost any organization claiming to represent ‘causes’ or ‘rights’ could fit it. According to Schmitter, these associations already perform a variety of public functions in the dual sense that they affect the public with their actions and they carry out policies, often subsidized by the state. Schmitter envisions a design in which people could choose more directly which interests concern them the most and which associations could best represent them. In order to gain a ‘semi-public’ status, the associations would have to meet several conditions of which the most important are – again – an internal democratic structure and a non-profit character. Schmitter proposes a fixed ‘Charter’ (or general provisions) for all groups able to play a role in his schema, which describes their internal structure, their rights and the means to check on them from outside. Schmitter mentions for example a guarantee for democratic procedures, a guarantee that public authorities will not intervene in internal deliberations and choices, and a commitment to full public disclosure of associational revenues and expenditures (1995: 175). Only associations who would fulfil their duties according to those general provisions could gain and maintain a semi-public status and take part in ‘voucher voting’. For this, Schmitter proposes that every individual gets a fixed amount of vouchers, with which he/she can ‘vote’ for associations that have the semi-public status, by giving them one or more of the vouchers. In this setting, being a member is not a requirement for voting for the association, and likewise, membership carries no obligation to vote. Schmitter sets out quite detailed arrangements concerning the distribution of the vouchers; his idea is that they could be administered jointly with the yearly income tax. However, unlike with paying taxes, people would not be forced to participate; in the end, funds would be distributed in the proportion to the ‘vote’ of those who wished to participate – just as in normal elections. By introducing what one could call ‘voucherism’, Schmitter offers a distinctively democratic mechanism, parallel to the electoral mechanism of representative democracy. But the links between participatory and representative democracy are, because of the scrutiny provisions, clear and strong, as in the two aforementioned designs. Here Schmitter envisions, similarly to Cohen and Rogers, a more active and positive role of the state, or rather of the government, based on concerns arising from traditional representative democracy. Clearly, there are significant differences between these key associative designs. In particular, the ideal role for the state for Cohen and Rogers and Schmitter is substantially greater than for Hirst. However, there are important common elements. All three designs are offered as supplements to representative democracy and deal only with internally democratic associations. Of course, by ‘imposing’

166

Piotr Perczynski

democratic conditions on groups, the model of associative democracy excludes a great variety of groups from taking part. This exclusion is criticized by many theorists, among other things for ignoring the reality in which many groups are not of a democratic nature (Bader 1998a, b; Mansbridge 1995; Young 1995). However, within the designs, the existence of other types of group is not denied; they are simply not considered to be associations able to contribute to associative democracy. In fact, reducing the number of relevant associations in this way also serves pragmatic goals, such as effectiveness of negotiations among associations (Immergut 1995: 206) and between associations and the state, as well as the overall efficiency of the ‘double democratic’ – associative and representative – system.5

Relevance of associative democracy to citizenship All three designs of associative democracy discussed above have the potential to enrich considerably the participatory component in citizenship. This may seem surprising, as the designs of both Hirst and Cohen and Rogers are sceptical about the very notion of citizenship.6 They perceive it as an outdated notion in a modern society. In fact, Cohen and Rogers claim that their associative democracy promotes what used to ‘[march] under the banner of “citizenship” ’ (1995b: 262), but, apparently, does not march there anymore. Hirst is generally critical about what he calls the ‘republican-citizenship standpoint’, because he, like Cohen and Rogers, views it as irrelevant in a modern society (1997: 82). Although Schmitter considers citizenship a ‘crucial property intrinsic to all types of democracy’, and even accuses Cohen and Rogers of ignoring the notion (1995: 183), he does not show how associative democracy could contribute to it. Nevertheless, we can identify important reasons why associative democracy as a model of participatory democracy is highly compatible with the notion of active citizenship. The educational project is traditionally an important ingredient of citizenship theories (Heater 1990; Dekker 1994). Modern republicanism is quite sceptical about the chances for success in learning how to be a ‘good citizen’ via traditional theoretical methods of instruction (Van Gunsteren 1998: 81–90). The advantage of civic education within associative democracy designs is that, rather than consisting of such methods, education is implicitly embedded in the practical, real-life setting of self-governing and (most importantly) democratic associations. Through dealing with the existing procedures within associations, members can learn the difficult art of exercising their rights and fulfilling their duties towards the group. In this way, the group can be perceived as a ‘school of citizenship’. The idea of learning citizenship by practising it in (and via) the association, obtains an extra dimension when we consider the idea of European citizenship, in which split and multi-layered loyalties have to be taken into account. In associative democracy, multiple – often conflicting – loyalties are present almost by definition, as individuals are often members of various associations at the same time, and in any case members of the inclusive ‘community of fate’. Those associations can be sub- or supra-associations of each other or completely distinct associations, or even something in between: overlapping in some ways, but distinct in others. The

Active citizenship and associative democracy

167

infamous ‘democratic deficit’ in the European Union is mainly linked to the disfunctioning of representative democracy on the European level, such as the limited power of the European Parliament (the directly chosen element of the system), and to the omnipotence of the Council of Ministers and the Commission (see Eriksen, this volume, for a more optimistic view based on a deliberative perspective). But equally absent on this level and crucial from the point of view of European citizenship are participatory mechanisms, which makes the idea of such citizenship a passive, empty and artificial notion. It seems certain that the only type of European citizenship conceivable (and for most people, the only type desirable) for the near future is some sort of multi-layered citizenship, as national citizenship will have to co-exist with supranational citizenship. Associations could in fact also serve as ‘schools’ for exactly this multi-layered aspect of the European, pluralist citizenship, for they always operate within multi-layered systems. However, democratic character is a necessary but not a sufficient condition in rendering associations relevant for ‘teaching citizenship’. Even the most democratic groups can be particularistic and hostile to the society outside the association. The authors of the schemes of associative democracy discussed here are well aware of this point. The necessary ‘social element’ within the group is, in the associative democracy debate, often called ‘other-regard’ or ‘external solidarity’. All of the designs include certain proposals in this respect. Schmitter tries to achieve some other-regard by imposing certain ‘externally oriented’ conditions, forms and rules for semi-public associations, in his ‘Charter’. These include prohibitions against the advocacy of violence, racism and other forms of criminal behaviour. Another condition would be openness, understood as, for example, a commitment to full public disclosure and accepting all individuals as members. Hirst is also aware of the particularism and potential hostility of associations, but he is faithful to the idea of the spontaneous and ‘organic’ character of associations, and believes that associations will voluntarily interact, co-ordinate and co-operate (1995: 112). Civic consciousness appears in neither Hirst’s nor Schmitter’s vocabulary. For Cohen and Rogers, on the other hand, this virtue is required for maintaining the conditions of popular sovereignty, political equality and distributive equity – three norms of democratic governance which are fundamental in their design (1995a: 38). They also pay the most attention to the notion of other-regard. In fact, we could say that their whole design takes the possible danger of overly particularistic (or ‘anti-social’) groups as an important point of reference. It makes their design especially attractive, as it does not overlook the problems and weaknesses of associationalism, but on the contrary concentrates on solving them. Their proposed solution is that the state shapes (‘curbs’) the associations, in order to make them more public spirited or other-regarding. By encouraging co-operation between, and co-ordination of, interests, less-narrow group programmes and greater awareness of interdependence are also encouraged. Cohen and Rogers describe several features of qualitative variation that associations should possess, and propose a judgement system in which the exact ‘tuning’ of the variables needed for achieving a certain goal – those needed in a particular kind of association – is left to the central state authorities. Quite conventional tools

168

Piotr Perczynski

of public policy are used: taxes, subsidies and legal sanctions (1995a: 44). Variables that may be tuned in this way could include ‘encompassingness’ (or coverage of group membership relative to affected populations), and characteristic modes of interaction with other groups. Because associations meeting these conditions are not narrowly organized, ‘the solidarity of their memberships approaches a social solidarity’ (1995a: 48–9).

Conclusion As mentioned before, both types of democracy – participatory and representative – are essential to citizenship. The associative democracy designs discussed here do not even attempt to address the problem of the great number of individuals who, because of circumstances or exercise of free will, do not participate in any association and so do not enjoy the protection and other benefits that association members enjoy. In an associative democracy setting they are protected and ‘taken care of ’ simply by the representative ‘part’ located above and controlling the participatory ‘part’. In Cohen and Rogers’s and Schmitter’s designs, this control is exercised more directly by the governmental agencies emanating from the representative system, which evaluate the conditions and performance of the associations. In the case of Hirst’s design, the legal system, maintained by traditional representative means, plays a similar role, creating a framework or ‘rules of the game’ for the associative, participatory part. All designs present a specific variant of connecting the two types of democracy. Proposals included in the particular designs of associative democracy could make not only a contribution to the development of a theory of citizenship but also to the theory of democracy in general. I have attempted to show that associative democracy is a model of participatory democracy with very distinctive links to the ideal of active citizenship. One of the main advantages of associative democracy is that it incorporates aspects of other participatory models, and so provides for participation on different levels and in different ways. For example, facets of deliberative democracy are clearly present in associative democracy designs, as associations provide concrete arenas of deliberation (in fact, the overall associative system could also be seen as such an arena – though see Rättilä, this volume, for a sceptical view). Decision-making in the democratic association is, because of its scale, naturally exercised by means of arguments offered ‘by and to participants’, which is, according to Elster, a principal facet of a deliberative model (Elster 1998a: 8). It is not surprising that in one of his articles Cohen writes that the very notion of deliberative democracy is ‘rooted in the intuitive ideal of democratic association in which the justification of the terms and conditions of association proceeds through public argument and reasoning among equal citizens’ (1997a: 72). In the design he has developed with Rogers, precisely this type of association is considered. From this point of view, the difference between the groups advocated in, for example, deliberative polling (see Fishkin and Luskin, this volume) or citizens’ juries (see Smith, this volume) and the groups considered in associative democracy is only that the latter are more organic

Active citizenship and associative democracy

169

or spontaneous (even in the most corporatist designs). One can also recognize features of direct democracy within the self-governing democratic associations, where members are taking active part in the face-to-face decision-making and implementation processes within the association. The problem of the ‘transmission’ of democratically taken decisions – highly problematic in most deliberative designs – does not create a tension for associative democracy, as the decisions affect in principle only the members of the association itself. In associative democracy there is also no ‘ideological problem’ that is faced by some deliberative democracy designs hostile to the simple ‘counting heads’ idea. Votes could and often should be decisive as far as decision making in the association is concerned. The internal debate among the theorists involved in developing the associative democracy designs, which heated up particularly in the first half of the 1990s, focused crucially on the issue of artificiality versus naturalness (spontaneity) of groups, or, in other words, top–down (via the state) versus bottom–up (i.e. societal) approaches to associations. In my opinion, a certain degree of both artificiality and spontaneity is needed for any association in order to function well in the society. Bader is probably right that what is needed is a ‘skilful combination’ of the state approach and the societal approach towards associations (1998b: 19). That combination should, in my view, depend especially on the stage or phase which the associative project has reached. Following Cohen and Rogers’s reasoning, we could agree that the initial stage, during which a new public or semi-public association is establishing itself, would probably call for greater support from the state (government), for example in providing suitable conditions for the development of other-regarding attitudes. Conversely, the more developed the whole associative system and the more associations it contains, the less direct interference of the state is needed and desired. Because the representative democracy system reserves the right to control to itself, there is (even in the most developed stage) always a balance or combination of ‘spontaneity’ and the ‘top–down’ approach. However, the system should aim towards being less constrained, because otherwise groups might lose their natural character, which (and here I agree with Hirst) is their biggest asset and the basis of their robustness. Nevertheless, the establishing of other-regard remains a fundamental problem of associative democracy, and one that has not received due attention in the ‘internal deliberations’ between the theorists. There was a debate about otherregard, triggered by Cohen and Rogers’s proposals, but it concentrated on accusations of imposing artificial solidarity from above (Hirst 1995), to which the reply was that imposed solidarity is better than no solidarity at all (Cohen and Rogers 1995b). In my opinion, this answer leaves untouched the real dilemma: protecting particular interests versus maintaining external solidarity. The discussion would be helped by having more concrete proposals to solve this problem, coming from different existing or (potential) new designs of associative democracy. It is not only associative democracy designs that have something to offer here, but also other projects in which groups – not necessarily restricted to the type present in the associative democracy designs – play an important role. One

170

Piotr Perczynski

of them could be the social capital proposals, closely related to the republican tradition (see Herreros, this volume), in which the generation of social trust is one of the mechanisms transmitting social capital created inside the association to the rest of the community. Also here a certain selection of groups (although less rigid than in associative democracy) has to take place – it is clear from the work of social capital theorists that there exist associations that could produce something opposite to ‘external solidarity’ (Putnam 1993: 146–8). Additionally, some of the communitarian projects may offer associative democracy a more ‘external’ or civic dimension. The emphasis in the communitarian proposals of Amitai Etzioni on education and ‘personal and social responsibility’ – or in other words societal duties – within the group, could provide another fruitful link to associative democracy (1995: x, 23–53, 89–115). The call for more openness in the associative democracy model to other associational proposals may sound paradoxical in the light of my earlier request not to extend the usage of the term ‘associative democracy’ to these very theories. However, such openness could be the only chance for the development, and perhaps even the survival, of the associative democracy model, so long as the supplementary character of the model and the democratic and self-governing nature of its associations is preserved. This is needed not only for the application of associative democracy to active citizenship, but, in my opinion, also for the theory of democracy in general, providing that associative democracy model is perceived not as the model of democracy but, like direct and deliberative democracy, an ‘ingredient in democracy’s future’ (see in this volume Saward’s similar reservations as far as deliberative democracy is concerned). Because society will benefit from the participation of citizens in democratic, voluntary and selfgoverning associations, this challenge should be taken up.

Notes 1 For an overview of modern theories of citizenship see Heater 1990; Kymlicka and Norman 1994; Van Gunsteren 1998. 2 Most of the authors of associative democracy designs use the terms ‘association’ and ‘group’ interchangeably and I do the same here. 3 English political pluralism is a political doctrine developed in the 1920s mainly by G. D. H. Cole and H. Laski, under the strong influence of both British socialism and liberalism (Perczynski 1995, 1998). Because of its strong functional aspect and hostility toward representative democracy it differs from the current better-known American variant which stresses group competition within the traditional system of representative democracy. However, the two variants have many similarities and it is educational to compare in this respect two texts of Robert Dahl (1947, 1989). 4 Radical anti-collectivism has influenced the associative proposals of Mathews, who, even more than Hirst, focuses on the economy. His associative democracy is to be understood as ‘collective activity via existing associations of workers and citizens’ (Mathews 1989: 12) who are directly involved ‘in shaping the economy’ (114–15). Likewise the imaginative associative design of Sabel focuses on the economy (Sabel 1995). 5 Apart from the criteria imposed on associations, we find in Schmitter’s design another exclusion mechanism aimed against the inefficiency of the system: a threshold for new associations, comparable to the well-known threshold in some parliamentary election

Active citizenship and associative democracy

171

systems. Only with a certain minimum number of vouchers could candidates acquire the status of semi-public association (Schmitter 1995: 180). 6 They use the term ‘citizen’ solely to refer to individual members of the state (apparently because of the lack of a better term). In his associative democracy design, Mathews analyses Marshall’s social rights in the context of citizenship – not surprising, given his design has a distinctly economic character (Mathews 1989: 5–9).

13 Associative democracy – fashionable slogan or constructive innovation? Sigrid Roßteutscher

There is widespread agreement amongst writers of highly diverse provenance: the existing system of political representation is deficient and thus no longer able to respond adequately to the novel challenges of a changing social and economic global order.1 What should be done? As is clear from this volume, theorists of democracy offer diverse models of deliberation in particular. A second answer, popular among politicians and intellectuals alike, is simply to strengthen the role of intermediary associations, in the conviction that this might cure modern democracy’s most urgent maladies (see the discussion in Perczynski’s chapter, above). Ever since de Tocqueville, this healing power is said to be an inherent feature of associational life: people learn how to trust, developing a stronger sense of community; interests are mediated in a more reliable manner; rates of political participation increase; civic virtues are taught; and, as a result, social integration is assured once more. At the core of the concept of ‘associative democracy’ is the assumption that democracy is functionally and/or normatively dependent upon a vibrant associational life; associations are seen as a necessary, if not a sufficient, condition of a functioning democracy. It is this crucial linkage between associations and democracy which provides the substantial coherence between otherwise widely divergent theoretical accounts. Theories of social capital, of pluralist interest mediation, of (neo-) corporatism, of communitarianism and civil society, or of participatory democracy have little in common apart from the strong belief that associations contribute greatly to the ‘functioning’ of democracy. The ‘functions’ of associations, however, are defined in very different terms. As outlined in Figure 13.1, five different functions of association in theories of associative democracy can be found: (i) interest mediation; (ii) providing political legitimacy; (iii) aiding decision-making and implementation; (iv) acting as a school of democracy; and (v) social integration. Hypothetically, the relationship between associations and all these functions is causal and positive in nature. Unfortunately, most theories concentrate upon single functions in a highly selective manner: pluralist and corporatist macro-theories being exclusively concerned with associations’ impact on interest mediation, legitimacy and decision-making, whilst micro-theories of trust and civic virtue emphasize the individual, almost (socio-) psychological effects of associability. Moreover, the concern of macro-theories is

Associative democracy – slogan or innovation? 173

Interest mediation

State legitimacy

MACRO

The ‘proper’ democratic institutions

Decision-making

Direct

School of democracy The ‘proper’ democratic citizens

MICRO Indirect

Trust, integration

Figure 13.1 Functions of associations in diverse theories of associative democracy

the design of appropriate institutions, whilst micro theories focus upon the character and virtue of the individual. As a result, there is hardly any substantial overlap between different theoretical constructions. In juxtaposition, this general relationship of mutual but neutral ignorance covers highly contradictory assumptions.

Associations in ‘mass society’: social pacifiers or schools of democracy? In one way or another, protagonists of communitarianism, social capital and civil society claim that active participation in associations contributes to the acquisition of democratic skills and virtues, making it a necessary condition for political activity. However, pluralist theory assumes that the contrary is true: integration in organizations fosters social peace and political demobilization. This assumption is in large part indebted to pluralist theory’s historical anchorage in the concept of ‘mass society’: where people are not securely related to a plurality of independent groups, they are available for all kinds of adventures and ‘activist modes of intervention’ in the larger society (Kornhauser 1959: 37). In other words, through organization elites control and steer the political behaviour of citizens, whilst institutional integration protects political elites from the uncontrolled and irrational desires and demands of non-elites. Hence, overlapping memberships and the resulting cross-pressures will lead to political passivity and thus to democratic stability (e.g. Lazarsfeld et al. 1948/1968). The counter-position is evident in the thesis of ‘civic voluntarism’ that emphasizes the ‘learning’ of certain civic skills, which in turn facilitate and encourage

174

Sigrid Roßteutscher

political participation: ‘In this way, the institutions of civil society operate, as de Tocqueville noted, as the school of democracy’ (Verba et al. 1995: 366). JeanJacques Rousseau was the original promoter of the idea that participation will create virtues, i.e. that participation is not dependent upon pre-existing attitudes or personality features but that through participation such features can be created. Indeed, as Evans and Boyte in a well-known reformulation of Rousseauean ideals suggested, social organizations create the necessary ‘free spaces’, ‘in which people are able to learn a new self-respect, a deeper and more assertive group identity, public skills, and values of co-operation and civic virtue’ (1992: 17f.). Barber attacks explicitly pluralist theory’s concept of ‘thin democracy’, which promotes ‘politics as zoo-keeping’ (1984: 3). In many respects, Barber’s principles are typical in participatory theory: (i) the priority of politics over social and private concerns; (ii) a belief in the community-creating impact of collective action; and (iii) a deep conviction that people are indeed willing and able to learn and to participate in politics. In contrast to participatory theory’s civic optimism, communitarian (such as Etzioni 1993) and social capital arguments (such as Putnam 1995a, b) are quite defensive. They view associations as bulwarks against social disintegration and deteriorating democratic and economic conditions. As a result, the present debate implies a return to the ideals of participation so prominent in the 1960s and 1970s. The context, however, has changed significantly. Three decades ago, political participation was a means for the ‘empowerment’ of ordinary citizens, a step towards more political equality. Nowadays, participation is seen mainly as a mean to prevent further decline in social trust, and the loss of community. In short, the recent debate squares very well with traditional pluralist thinking. More importantly, there is the curious problem that pluralist theory promotes social participation to prevent political activism, whilst micro-theories of trust, civic virtues, and participation emphasize the politicizing impact of social integration. Note that it is not the function of social integration which is contested; it is institutional integration’s impact on the political behaviour of the ‘integrated’ which constitutes the problem. In other words, the debate centres on whether or not political participation as such should be encouraged and whether, and to what extent, associations can perform that role.

The solution? Associative democracy In recent years, pluralist theory has experienced both harsh criticism and radical reformulation. It was Cohen and Rogers’s, Hirst’s and Schmitter’s merit to revive a highly ‘questionable’ (Zimmer 1996: 60) concept of interest mediation and representation which, for different reasons, increasingly lost ground in comparison to participatory, direct or deliberative concepts of democracy. This reformulation was radical in the sense that it took pluralist theory’s premise of a plural but ‘voluntary’ system of interest mediation, and turned it into a plural but ‘compulsory’ system of interest mediation (though Hirst is an exception). As a result, ‘new’ pluralism puts even more emphasis on associations than did ‘old’ pluralism. ‘New’

Associative democracy – slogan or innovation? 175 pluralist arguments are based upon three assumptions. First, democracy depends chiefly upon the mediation and representation of interests via associations. Second, the existing system of organized interests is deficient (as Schmitter (1994: 160) puts it, ‘of all the things that do not work well in contemporary democracies, [it] must be rated among the worst’). Third, and consequently, the system should be reformed thoroughly. As Cohen and Rogers (1992: 426) argue, if the ‘right sorts of associations do not arise naturally’, government action should ‘supplement nature with artifice’. In other words, whilst traditional pluralist theory assumed that existing interests will automatically and naturally find an association to voice and mediate their concerns, Cohen and Rogers assume that associations can and should be constructed deliberately, depending less upon ‘natural’ interests than on certain normative criteria. According to Cohen and Rogers, proper associations resemble the traditional ‘social partners’; large and relatively encompassing organizations with accountable but powerful leaders and significant means of sanctioning their members. Moreover, there should be a centralization of authority in group decision-making and a strong relationship between state and association (1992: 428f.). Evidently, the theory of ‘associative democracy’ celebrates a type of organization that is harshly criticized because of its authoritarian character, its high susceptibility to bureaucratization, its decreasing potential for social integration, its inability to teach civic skills, and its low internal possibilities for participation.2 Moreover, it is a type of organization that is most drastically exposed to processes of individualization and, thus, most significantly suffers from decreasing public acceptance and a declining ability to recruit volunteers (Dechamps 1993: 82). Blatantly, the ‘new’ concept of associative democracy promotes the ‘old’ concept of corporatist interest groups. This overt contradiction is no accident. Despite its label, the theory of associative democracy is in fact not primarily a democratic theory. On the contrary, its main interest is to ‘enhance government competence and improve economic performance’ (Cohen and Rogers 1992: 430). Accordingly, Cohen and Rogers devise a complex model of governance where associations are centrally involved in national, regional, and local processes of decision-making and implementation, with localized groups’ main task being the enforcement and administration of previously determined policies (1992: 438). This conception clearly reveals the functional and streamlining impetus of ‘associative democracy’. In fact, Hirst is right when he claims that Cohen and Rogers’s conception of the state is not particularly democratic, proposing as it does a state which is no longer ‘subject to citizen sovereignty and the majority principle’ (Hirst 1992: 476). Although the concept might offer the desired ‘serious alternative to the Keynesian welfare state’ (Cohen and Rogers 1992: 430), its democratic status is highly uncertain. In a revised version Cohen and Rogers emphasize the civic, deliberative, and solidarity-breeding effects of associative involvement (1994: 152ff.). In other words, they partially shift their focus from a functional macro-perspective to the micro-world of involved individuals. However, the one-sided and uncritical presentation of alleged individual benefits of participation nourishes doubt as to

176

Sigrid Roßteutscher

whether this shift of perspective is any more than a loose addition of evidence supporting the grand functional scheme of associative democracy. At any rate, Cohen and Rogers’s arguments rest on shaky ground. There is a stark contrast between the individual functions celebrated in this later version – arenas of deliberation and co-operation, school of democracy, competence, trust, otherregardingness – and the centralized, semi-public, leadership-guided associations promoted previously. This contrast will prove difficult to reconcile. Some (though not all) of these remarks apply also to Schmitter’s concept of associative democracy. Like Cohen and Rogers he believes in the basically democratic and thus desirable role of interest groups, and he seriously dislikes the existing pluralist system. Unlike Cohen and Rogers, whose interest is in political and economic competence, Schmitter’s concern is with social inequality (1994: 160). Schmitter’s key suggestion is to realize ‘secondary citizenship’ through obligatory associational involvement and a state administered system of vouchers. Interest associations would have semi-public status; they would be financed through compulsory contributions, and funds would be redistributed by means of citizen vouchers (Schmitter 1994: 163). Through a ‘coercive levy’ everybody would be forced to participate – or at least to contribute financially – and the initial advantages of the privileged would largely disappear. Accordingly, Schmitter’s concept of associative democracy advocates a revision of ‘weak’ democracy, which is significantly less unequal than the traditional ‘thin’ democracies of a pluralist or corporatist nature (1994: 169). Schmitter’s proposal is clearly policy-oriented and less ambitious with regard to democratic theory. It would, however, in practice as well as in theory, mean a substantial change in the nature of contemporary democracy. Although Schmitter uses the term ‘secondary citizenship’, the role of associations appears to be of primary importance. The great majority of contemporary democracies are based upon voluntary citizen participation, compulsory voting being a rare exception. Hence, a coercive system of associative interest representation might easily become traditional politics’ ‘bigger’ brother, particularly in countries with low voter turnout and little interest in party politics. In short, Schmitter’s policy reform suggests a substantive and qualitative change in the nature of contemporary democracy, going far beyond the original aim of diminishing social inequalities. Moreover, there is a further danger in replacing party politics by associative politics. Political parties provide clear clues for evaluating politics and policy proposals, with the left–right continuum being the most important. ‘New’ pluralists’ associations cannot provide such clues and, as a result, electors, deprived of simplifying devices, might be confused and large numbers might either distribute vouchers randomly or abstain completely (see Budge’s chapter in this volume for further defence of the continuing importance of political parties to democracy). Hirst is the only advocate of the ‘political doctrine’ (1997: 2) of associative democracy who professes his faith in the old pluralist ideal of voluntarism, rendering his views decidedly hostile toward the state-centred versions of both Cohen and Rogers and Schmitter (Hirst 1994: 37ff.). Hirst is an outspoken

Associative democracy – slogan or innovation? 177 advocate of a ‘small’ state, a state which is society’s great tax collector but which does not provide services for the most part. Instead, Hirst promotes an expanded version of ‘pillarization’: a society in which different ‘pillars’ – ‘environmentalists, feminists, ethnic communities, even conservative religious groups’ (1997: 36) – would themselves ‘deliver services in areas like health, education, and welfare’. The state, once again ‘a minimal public power’ (1997: 33), would finance these self-governing associations depending upon their membership strength: the more members there were, the more money an association would receive. That said, voluntarism is seriously undermined by the fact that associations are society’s major providers of services; all those who want health care, all those who want schooling for their children, all those who want any kind of welfare are, in effect, forced to join. In other words, associationalism à la Hirst implicitly forces compulsory participation. What, then, is the democratic status of such a vision? There are two ways of responding to this question. First, Hirst’s scheme can be seen as predominantly cosmetic or symbolic. In this case, associative democracy is not particularly novel. In comparison with countries such as Germany, where charity organizations of both the Catholic and Protestant churches and the working class have long been centrally involved in the provision of welfare services, Hirst would only introduce some additional ‘players’ (feminists, environmentalists, etc.). However, differences between services would be minor because the state would remain the most powerful player, providing and educating the personnel and setting the standards for both curricula and the extent and nature of services to be provided. Associative self-government would thus take place within very strong limits. If, however, Hirst promotes ‘true’ self-government, where associations are ‘the primary means of both democratic governance and organizing social life’ (1994: 25), democratic prospects are indeed very doubtful. If finance depends upon strength of membership, very few associative pillars would be capable of designing and running adequate systems of health, education or welfare. If, on the other hand, Hirst wants all value communities, independent of their numerical strength, to regulate their own affairs, the resultant society would be outright anarchic, with countless and presumably incompatible systems competing with each other. This is the real dilemma confronting associationalists: they either, like Cohen and Rogers and Schmitter, embrace a compulsory concept with a strong interventionist state, or, like Hirst, run into the danger of promoting a voluntaristic society which is either highly unequal or torn apart by institutional anarchism.

The counter position: associations’ undemocratic nature Associationalists from Putnam to Hirst share the conviction that associations are inherently democratic and that democracy cannot survive without them. There is, however, a distinct, albeit so far ignored, counter position to this belief. Its origins date back to the French Revolution which, based upon Rousseau’s ideal of the volonté géneral, deliberately did not promote freedom of association, claiming

178

Sigrid Roßteutscher

that mediating organizations undermine individual freedom and foster alienation. This scepticism survived in the classic republican critique of pluralism. Republicans have a strong suspicion of intermingling the ‘social’ and the ‘political’. In Arendt’s view, apolitical associations threaten both the purity and rationality of politics. If the social and the private become political, politics will deteriorate into a ‘rule of the administration’ and public discourse will retreat behind the ‘pull, pressure and the tricks of cliques’ (Arendt 1958: 203). This tradition of republicanism thus ‘proposes institutional reforms that aim to insulate arenas of collective choice from the pressures of particular interests’ (Cohen and Rogers 1992: 406). Even the almost commonly shared conviction about associations’ integrative function has been harshly refuted. The world of local communities and associations, as Milofsky (1988a: 20ff.) indicates, is a world in which societies degrade from ‘open’ to ‘closed’ systems. The communitarian quest for more community and associational integration thus results in a stronger dependency upon social organization, the price of integration being less freedom and more social control. Lack of individual freedom and intellectual inspiration through organizational integration is undoubtedly the core of Max Weber’s critique of associational life. By 1910, Weber had already characterized ‘contemporary man’ as an ‘associational being to a terrifying, never anticipated extent’ (1924: 442). According to Weber, clubs and associations breed and foster the passivity of obedient subjects. It was little wonder that monarchs and authoritarian governments, in general, protect and favour associational activity (Weber 1924: 445). In short, there is no argument without an explicit counter-argument. Whereas Taylor (1991: 57) describes a strong and powerful net of social organizations as ‘the’ precondition of democracy, Arendt warns of the totalitarian progress of the ‘social’. Whereas Barber emphasizes the activating impact of associations and their contribution to grassroots politics, pluralist theorists stress their depoliticizing and passivity-cultivating impact. Whereas communitarians hope for an increase in civic virtues and a sense of community, Weber turns our attention to the authoritarian and anti-democratic effects of organizational integration.

An empirical perspective If theory cannot agree upon the basics, maybe empirical evidence can contribute to solving the puzzle? The concept of associative democracy embraces a particular image of mankind, man being an ‘organised man’ (Schuppert 1997: 120). Having said that, European citizens correspond to this ideal in a very different manner, making associative democracy in some countries almost a ‘true’ description of reality. In others, however, man is predominantly ‘unorganized’, with voluntary associations playing only a marginal role. The Scandinavian democracies are highly associative, whilst in southern Europe only small minorities are involved. Considering levels of active participation – any kind of involvement that goes beyond mere membership – differences between nations, however, are small, associational activity reaching nowhere more than a tiny part of the population

Associative democracy – slogan or innovation? 179 (for comparative figures see for example Goul Andersen 1996; van Deth and Kreuter 1998). Clearly, evidence suggests that associative democracy is a theoretical construct, having little in common with empirical reality.

Direct impact: ‘school of democracy’ Low levels of overall participation are not sufficient to contradict positive claims about associations’ impact on democracy or political participation. Older studies did indeed find a consistently positive impact of social organization: association members are ‘better’ democrats, have more information about politics, show a higher interest in public affairs, and are themselves politically active to a higher degree. Conclusions drawn from this evidence were straightforward, voluntary organizations being ‘the most important foundation’ of democracy (Almond and Verba 1963: 320ff.). Contemporary quantitative research persistently and virtually unambiguously shows a positive relationship between social and political participation, i.e. socially active people tend to be politically active as well (among many examples, see Parry et al. 1992). There is, however, hardly any difference between nominal and active members of associations (Verba et al. 1995: 339), this indeed being ‘one of the unresolved mysteries of voluntary activity literature’ (Newton 1997: 6). Mystery or not, the modest difference between passive and active involvement is undoubtedly a strong argument against the assumption that civic virtues and participatory skills are inculcated through active participation in social organizations. It is unclear whether there is a causal relationship at work here at all. Indeed, empirical research shows that there are certain ‘joiner’ personalities – people who are (hyper-)active both socially and politically (Smith 1977), many of whom do not really differentiate between social or political activities. Besides, the generally positive relationship between social and political participation is a rather weak relationship (van Deth 1996: 392). Indeed, socio-economic variables reduce the impact of social organizations significantly. Nevertheless, social activists are indeed ‘better’ democrats, more tolerant and better informed. As Verba and his collaborators argue, ‘while the process exacerbates political inequality, it may enhance the quality of political discourse and democratic governance’ (1995: 507). However, social participation’s impact seems to be limited to the area of conventional or institutionalized patterns of political participation; with regard to ‘alternative’ or non-institutionalized modes of political action, membership in associations is at best irrelevant (see van Deth 1996: 403; Anderson 1996: 115ff.). A positive relationship between membership in associations and election turnout is in fact confirmed for many different countries (see Abowitz 1990; van Deth 1992; Anderson 1996). Comparative research on the basis of micro-data, moreover, demonstrates that there is indeed a modest positive relationship between social participation and protest activities (Dekker et al. 1997: 224). On the basis of macro-data, by contrast, the same authors conclude that the relationship is reversed; organizational integration being negatively related to both ‘the level and

180

Sigrid Roßteutscher

radicalness of protest’ (229). Evidently, micro- and macro-data speak two different languages. What can be learned from this brief excursus into empirical research? To be sure, associative democracy is a project that does not square too well with empirical evidence. In many respects, modern ‘man’ is an ‘unorganized’ man, active participation in voluntary associations being in fact limited to minorities in all countries of the industrialized world. Moreover, these active minorities are no representative sample of the population. They tend to be highly educated and economically well off.3 Besides, it is unclear whether there is a causal relationship between social and political participation, and the assumption that voluntary associations are a fertile training ground for participatory skills and civic virtues is hardly supported by the evidence. Finally, voluntary associations seem to be allied to conventional politics, strengthening particularly party and voting-related activities. As a result, associative democracy is a project that supports the notion of ‘old’ politics, leaving almost no traces on the many facets of ‘new’ politics. This last point should be considered seriously by authors who relate associative democracy to concepts of ‘green’ or ‘direct’ democracy. Empirically speaking, there is not much ground on which such a relationship could be built.

Indirect impact: societal integration There is, however, evidence that associations’ impact on democracy is, as claimed by Putnam and many communitarians, indirect, fostering social integration and general trust. However, it is debatable whether integration and trust have the supposed positive impact on civic virtues and political participation: how does integration into one specific organizational context contribute to integration into society as a whole? Why should associations matter that much, if, from an individual point of view, time spent with associations is rather marginal compared to time spent with family, work colleagues, local community, neighbours, or circles of friends?4 Does organizational integration lead, as Putnam suggests, to general trust? Or is trust a precondition for joining social organizations? Finally, does trust promote activity or, as pluralists hope, passivity? Indeed, Crenson’s (1983) seminal study of neighbourhoods questions all aspects of Putnam’s simple formula that equates social participation with integration, integration with trust, and trust, finally, with political participation (see Putnam 1995a). As Crenson indicates, neighbourhoods with many social conflicts – neighbourhoods which are poorly integrated internally – show a higher level of community activities and involvement in local politics. Second, distrust between neighbours encourages participation, trust in fact leads to passivity; simply, ‘the more distrust, the more action’ (Crenson 1983: 170, 178). Only in one instance do integrated neighbourhoods perform more activity than conflict-prone neighbourhoods: if action is directed against an external ‘enemy’, such as bureaucrats in town administration, or particular neighbourhood groups who have moved to the neighbourhood only recently and disturb previous arrangements or behavioural codes (Crenson 1983: 254ff.). These results correspond nicely with

Associative democracy – slogan or innovation? 181 social psychology’s old ‘truism’ that there is no integration without exclusion, no in-group formation without definition of out-groups.5 Other studies indicate, however, that house ownership and length of residency in a specific area are positively related to political participation (e.g. Verba et al. 1995: 455). This evidence supports Putnam’s thesis, but at the same time, it casts doubts on the supposed causal relationship. Indeed, integration does not seem to be the result of community activity or involvement in associations. On the contrary, integration seems to be prior to associative engagement. Similar scepticism prevails concerning the expansion from group-specific trust to generalized social trust. Moreover, whether an organization teaches trust depends largely on the features of that organization. Frequently, as several Putnam critics indicate, group members primarily learn conformity to organizational rules and norms and social obedience (e.g. Levi 1996: 49ff.). Moreover, there is some evidence that the case of trust is equivalent to the case of integration, trust not being the result of organizational activity but its precondition. In other words, ‘the trusting join’ (Newton 1997: 6). More generally speaking, trust is a ‘product of democracy’ (Muller and Seligson 1994: 646). If this is correct, the entire debate about ‘the strange disappearance of social capital’ (Putnam 1995b) would be turned upside down! It is not that a decline in social capital endangers democracy, but that the democratic ‘malaise’ results in declining trust, thus decreasing social participation. To summarize, what has been said about the validity of the ‘direct impact’ thesis can be repeated here, the available empirical evidence leading to highly ambivalent conclusions. Relationships are weak and their causal nature is often entirely unclear. As critics of the ‘trust’ paradigm indicate, we know far too little about the impact of different types of organizations. A positive relationship between integration, trust, and political activity might be a valid conclusion for one type of organization and an utterly wrong one for another.

A contribution to democracy? The validity, usefulness, and fruitfulness of the concept of associative democracy depends crucially upon the fact that it can demonstrate that associations fulfil certain democratic functions. The fact that associations fulfil important social functions is not sufficient. Social integration – no matter how important – is a pre-requisite for the health of any political system, democratic or not. Protagonists of ‘associative democracy’ have to demonstrate convincingly that associations contribute to the psychological well-being of individuals, protect society against anomie and social alienation, create some sense of community, and that they are not only a functional aid to government performance and economic growth. Clearly, this is no plea for the theoretical insignificance of social integration. But as long as democratic theory cannot demonstrate that associational life is a substantive contribution to democracy, Max Weber’s scepticism must prevail. If, on the other hand, reforms might even ‘suggest a new form of political– constitutional order’ (Cohen and Rogers 1994: 138), then a clearer presentation

182

Sigrid Roßteutscher

of evidence is required. Problems are abundant. Micro-theories address a certain type of question and give a certain type of answer, whilst macro-theories concentrate upon issues of an entirely different nature. In other words, there is not one discourse about the democratic function of associations, but many different sub-discourses, which, up to now, talk past one another. Considering this mutual ignorance, it is surprising to see how different approaches with different methods and different concerns come to a very similar conclusion: associations are vital for the survival of democracy. If one broadens the perspective and considers political and public discourses as well, one must acknowledge an even further-reaching consensus. Associations play a crucial role in the political debate about a re- or deconstruction of the traditional welfare system, and voluntary work in associations is discussed as a substitute or complement to unemployment benefits.6 However, if one scratches the surface a bit, this apparent consensus quickly falls to pieces. In fact, the presumption of the democratic usefulness of associations remains virtually the only common denominator in an otherwise extremely incoherent field. If one asks why associations are so vital to democracy, a host of different and overtly contradictory answers emerge. As I indicated, radically different perspectives with different questions explain a part of this puzzle; lack of communication explains another. More important though, one cannot even find a common, uncontested, understanding of what an association is, not to mention of what a ‘proper’ democratic association should look like. There are, in fact, as many definitions and conceptions as there are authors in the field. Indeed, corporatist associations or pluralist theory’s interest groups have very little in common with the local self-organized units promoted by communitarians. Similarly large is the difference between Barber’s political grassroots organizations and Putnam’s apolitical choral societies. There is even less compatibility between Cohen and Rogers’s promotion of the traditional ‘social partners’ and, for instance, Evans and Boyte’s conception of associations as ‘free spaces’ in which participants ‘learn’ certain (almost socio-psychological) ‘habits of the heart’. Presumably, much of this confusion could be clarified if one explored the specific contributions of specific types of association. However, such research is virtually non-existent. We do not even know whether and if, to what extent and in which respect, the concrete organizational context matters at all. Things would look less bleak if theory would at least acknowledge existing empirical findings. As I demonstrated, there is a quite impressive body of research concerning the microeffects of associative activity. These findings, however, do not always support theoretical assumptions; on the contrary, empirical evidence clearly suggests that we should not push the theoretical argument about associations’ contribution to democracy too far. With regard to macro-effects, the situation is worse, not much empirical evidence being available.7 Considering the state of the art, the discourse on associative democracy has still to demonstrate that it is more than a passing academic fashion. The present debate about associative democracy and associations’ role in democracy has contributed to a further inflation of theoretical and normative arguments, a ‘collection of untested prejudices, at best hypotheses claiming plausibility’ (Bühler et al. 1978:

Associative democracy – slogan or innovation? 183 145). This situation is entirely unsatisfactory, in particular considering the fact that the concept of associative democracy increasingly penetrates public and political discourses about the future of the welfare state. Indeed, there is a danger that the concept of ‘associative democracy’ will be (ab-)used in public discourse in order to justify cuts in welfare provision, with social organizations being promoted as cheaper substitutes for state activity. If the discourse on ‘associative democracy’ does not want to become a fig-leaf for particularized interests in budgetary cuts and the dismantling of the welfare state it has to provide much stronger arguments.

Notes 1 Many friends and colleagues have provided valuable comments on earlier versions of this chapter. Most particular thanks are due to Jan van Deth (Mannheim), Michael Saward (Open University), William Maloney (Aberdeen), Meindert Fennema (Amsterdam), Piotr Perczynski (Leiden), and Lars I. Andersson (Lund). 2 For references on ‘large’ bureaucratic associations’ deficiencies, see for example Zimmer (1996: 44ff.), Milofsky (1988b: 184f.). 3 That social activists originate from privileged strata is an often repeated result of empirical research (Newton 1976; Conway 1991; Verba et al. 1995). 4 Newton calculated that Dutch citizens spend about 8 per cent of their leisure time in clubs and associations (Newton 1997: 4f.; similarly for Germany, see Schwarz 1996: 261). 5 E.g. Katz (1985), Bloom (1990). Responding to Putnam, Portes and Landolt demonstrated that group ties were deliberately mobilized in order to keep ‘outsiders’ out (1996: 19). 6 For example, see Tony Blair’s ‘welfare to work’ programme, and similar concepts put forward by Claudia Nolte, Germany’s former Minister of Family, Senior Citizens, Women and Youth, or an expert report commissioned by the regions of Bavaria and Saxony (Kommission für Zukunftsfragen 1997). 7 An exception is the present Johns Hopkins project on the Third Sector which collects data on non-profit organizations in a wide range of different countries (see Salamon and Anheier 1994). One can hope that this data base will provide the material to test some of the core arguments of macro-theories.

14 Social capital, associations and civic republicanism Francisco Herreros

Social capital is a concept widely used in recent years to explain a variety of phenomena, from the effectiveness of democratic institutions to the formation of human capital. There are different definitions of social capital available in the literature, but all of them agree that social capital implies ‘a set of institutionalized expectations that other actors will reciprocate cooperative overtures’ (Boix and Posner 1996). Authors of the social capital research paradigm often relate social capital to civic republicanism. The best-known defence of this link is found in Making Democracy Work, where Robert Putnam traces the origins of the differences in social capital among Italian regions up to the eleventh century and the rising of the city republics of northern Italy.1 The presence of these civic traditions is equated with the presence of social capital: virtuous citizens are helpful, respectful, and trustful towards one another; and these characteristics, especially interpersonal trust, are usually considered to be forms of social capital (Green and Brock 1998; Rahn and Transue 1998; Uslaner 1998). In this chapter I explore the possible connections between the republican tradition in democratic theory and the social capital research paradigm. In a way, they are both important for the current debates about democratic theory. Social capital, in the form of associations, is often considered a solution to the ills of democracy, promoting the accountability, representation and effectiveness of democratic institutions (Putnam 1994; Boix and Posner 1996; Tyler 1998). The preceding chapters by Perczynski and Roßteutscher illustrate some of the debates that continue around these sorts of claims. The revival of republican thought influences democratic theory through its characteristic belief in deliberative democracy and the concept of the common good. In the following section I shall differentiate two strategies in the republican tradition towards the achievement of the common good. After that, I concentrate on the first of these strategies, centred in the concept of civic virtue, and discuss its potential role in the solution of collective action dilemmas. This leads the analysis to the concept of social trust and its relations to civic virtue. In the final section I shall discuss the relations between social capital in the form of associations and civic virtue. I explore the possibility that a certain form of social capital, the

Social capital, associations and civic republicanism

185

participation in networks of civic engagement, affects individual actors’ orientations towards co-operation.

Civic republicanism and the common good One of the central principles of the republican tradition in democratic theory is the belief in a notion of the ‘common good’. This republican commitment has been defined in a ‘minimalist’ way as a belief in the possibility of settling at least some normative disputes with substantively right answers (Sunstein 1988b: 1541). Nevertheless, there are important differences among republican authors concerning the means by which it could be possible to attain the common good of the community. In this section I distinguish between two different views found in republican thought about how to achieve the common good: by institutional design and through the exercise of ‘civic virtue’ by the citizenry, respectively. The basis of the set of arguments that the attainment of the common good can be enhanced through institutional design is either that (a) the civic virtue of the citizenry and the rulers is insufficient for the attainment of the common good, or (b) that civic virtue is not a necessary condition for that end. These authors defend the establishment of formal institutions of a certain kind as a means toward the fulfilment of the common good. The theory of mixed government, the separation of powers and the theory of checks and balances are probably the clearest manifestations of the institutional arguments concerning the common good. The justification of these institutional devices has historically been related to the preservation of the common good of the community against various dangers, especially the dangers of the tyranny of the poor over the rich (Richards 1994: 124–5). The distribution of functions among social orders (the one, the few and the many) or governmental bodies would compel them to put aside their sectional interests and rule with an eye to the common good. I turn now to arguments based on the individual display of civic virtue. Civic virtue is usually understood in classical republican thought as the disposition to further public over private good in action and deliberation (Dagger 1997: 14). For many republican authors, this is the sole necessary condition for the attainment of the common good of the republic. This is especially clear in the humanist writers before the Quattrocento, such as Compagni, Mussato and Latini (Skinner 1978: 43–5), but some authors of the Florentine republican tradition, such as Leonardo Bruni or even Machiavelli, can also be included on the ‘civic virtue’ side of the argument about the common good of the community (Pocock 1975: 210). Finally, in the years prior to the American Revolution, the faith of the Founders in the civic virtue of the American people was probably stronger in the aim of the creation of a virtuous republic than the necessity of building adequate institutions (Wood 1987: 100). After this brief definition of the main lines of the classical republican tradition and its relations with the concept of the common good of the community, I shall now explore the connections between republicanism and the social capital research paradigm. In the next section I discuss the relations of social capital with

186

Francisco Herreros

the concept of civic virtue. I then move on to discuss how the republican belief in the attainment of the common good can be used to question some characteristics of social capital.

Social capital and civic virtue The term ‘social capital’ was first introduced by Loury in the 1970s, to refer to the set of resources that inhere in family relations and in community social organization and are useful for the cognitive and social development of a child or young person. The most-cited definition of the concept is that of James Coleman, who defines social capital as ‘a variety of different entities having two characteristics in common: They all consist of some aspects of a social structure, and they facilitate certain actions of individuals who are within the structure’ (Coleman 1994: 302). According to Coleman (Coleman 1994: 310–11), key examples include: • • •

•

obligations and expectations; information potential (information can be acquired using social relations that are maintained for other purposes); norms and effective sanctions (for example, the norm that one should forgo self-interest and act on behalf of the collectivity, which may contribute to overcoming problems of collective action); authority relations (someone in authority has available social capital in the form of rights of control over certain activities of a person under his authority.

Some authors have emphasized the structural side of social capital: social capital as an aspect of a social structure, not an attribute of individuals (Foley and Edwards 1997; Kolankiewicz 1996: 435; Diani 1997: 133). Others have stressed the opposite, regarding social capital as a subjective phenomenon composed of a range of citizen values and attitudes, especially those related to forms of trust and reciprocity which influence or determine how they relate to each other. Regarding the two republican strategies towards the common good outlined in the previous section, the links between the second one and social capital are worth exploring. The classical republican concept of civic virtue is a promising link between republicanism and the social capital research paradigm. In this section I examine the relations between civic virtue and social capital. I shall discuss the role of civic virtue in overcoming collective action dilemmas, but, for the moment, the objective of the collective action is not going to be considered. That will be discussed in the subsequent section. As we have seen, civic virtue is usually defined in the republican tradition as the disposition to further public over private good. Nevertheless, there are differences among republican writers concerning the relation of private interests and public concerns in the republican citizen. For Plato and Aristotle, it seems there is no contradiction between private and public good: the participation of citizens in the

Social capital, associations and civic republicanism

187

public life of the polis is the only way to ground the rational nature of man, and this in turn is the necessary framework for his happiness (Domènech 1989: 78–90). Likewise, there is no conflict between private interests and public concerns for Machiavelli: citizens love their country and its laws because they perceive it as in their interest to live in that republic (Viroli 1995: 73). The relation between private interests and public concerns in the case of de Tocqueville’s ‘enlightened selfinterest’ is similar: involving himself in the affairs of the community, the citizen recognizes where his true interest lies (Oldfield 1990: 146–7). In this first interpretation of civic virtue, the concern for the common good of the republic, directly affects one’s welfare: it is part of the welfare function of the virtuous citizen. It also encompasses instrumental, and therefore rational, behaviour (it is concerned with an outcome: the attainment of a co-operative equilibrium). Other authors in the republican tradition consider the relation between private and public interests differently. The clearest example in the republican tradition of this second interpretation is probably Montesquieu. For him, the display of civic virtue is in contradiction with the pursuit of private interest, so the virtuous citizen must sacrifice his private interest on behalf of the common good (Viroli 1995: 72–3). This can hardly be understood as rational behaviour. A virtuous citizen in this sense is supposed to act according to his duty even if nobody else is behaving in that way and his action is unlikely to achieve co-operation. As we have said, the display of civic virtue can be understood as a way to solve a dilemma of collective action. In the absence of selective incentives, the usual solution to the collective-action dilemma is through repeated interactions (Axelrod 1984). The problem is that the application to the real world of this solution to the collective- action dilemma is very limited: co-operation is difficult to sustain when the game is not repeated (or when there is an end-game), when there is not full information about the other players, or when there is a large number of players. And, in the case of impersonal exchange, these conditions are rarely found (North 1990: 12–13; Elster 1985: 360–1). The presence of civic virtue can alter the outcome of the game by modifying the preferences of the players. In the Prisoner’s Dilemma there is an assumption of selfish preferences. In the case of players with civic virtue, their preferences are similar to what Amartya Sen describes as ‘sympathy’, or, more accurately, what Christopher Jencks calls ‘communitarian unselfishness’: identification with a collectivity, rather that with specific individuals (Sen 1990: 31; Jencks 1990: 54–5). These preferences can favour the achievement of a co-operative equilibrium as the solution of the game. The strategy that one player with those preferences would probably adopt is one of unconditional co-operation: in the case of a repeated game, he will choose co-operation in the first round, and he will play co-operation in all later rounds whatever the other player does. Notice, however, that the co-operative equilibrium will only be achieved if both players display civic virtue. If only one of the players is virtuous, the dominant strategy of the non-virtuous player is defection, and the outcome of the game is that the virtuous player co-operates and the non-virtuous player defects.

188

Francisco Herreros

Suppose that our virtuous citizen is placed in a world where there is uncertainty as to whether the people with whom he interacts are virtuous. What would be his strategy in these interactions? A citizen virtuous in the second sense – because it is his duty to be so, even at the expense of his private interests – would probably choose to co-operate. By contrast, a virtuous citizen in the first sense – equating his private interest with the interest of the community – can choose not to co-operate in a one-shot Prisoner’s Dilemma, because he is afraid that his goodwill will not be reciprocated. His willingness to co-operate depends on the subjective probability he assigns to the other player being virtuous, and on the ratio between the potential losses if the other player is non-virtuous and the potential gains if the other player is virtuous. In other words, it depends on the trust he is willing to place in the other player, on his beliefs about ‘social trust’. At this point we have reached one of the main subjects in the literature on social capital: the generation of social trust. As we have seen, trust is widely cited as a form of social capital in that literature. The analysis of trust in the social capital literature is focused on a certain kind of trust relation: that which arises from frequent interactions. Participation in associations is a source of this kind of trust: it is argued that trust arises in this case as a by-product of the co-operation for the appropriation of a private good (Coleman 1988; Putnam 1993: 170). The result is a kind of ‘thick’ trust, more often found, for example, in friendship. But in the case of the virtuous citizen in a world of uncertainty about the trustworthiness of others, this kind of trust is irrelevant in one-shot interactions with people that he does not know. The virtuous citizen will choose to co-operate in a one-shot Prisoner’s Dilemma if he believes that people, in general, are trustworthy. But this kind of ‘social trust’ is difficult to derive from the relations of personal trust created, according to the literature on social capital, inside associations. Trust cannot necessarily be generalized: in this sense, the empirical evidence for the creation of social trust from the participation in associations is ambiguous (Brehm and Rahn 1997; Stolle 1998). The mechanism between membership in such groups as football clubs or birdwatching societies and social trust is very unclear (Levi 1996: 47–8). This issue is not only addressed by the social capital literature. In some important ways, the search for ‘external solidarity’ in the debates on associative democracy points to the same problem (see Perczynski in this volume). But this problem is especially important for the social capital paradigm, because one of the attractive features of the concept of social capital is its influence on such factors as the effectiveness of democratic institutions, and social trust is one of the mechanisms invoked to explain this beneficial effect (Putnam 1993). It might be useful to sum up the arguments posed so far. I have argued that the display of civic virtue can be understood as a solution to a dilemma of collective action. This is possible because a virtuous citizen has co-operative preferences. Nevertheless, this solution is possible only if all the players have those same preferences, and if this structure of preferences is common knowledge. But in cases of uncertainty about the trustworthiness of other players, the virtuous citizen in the first sense will choose to cooperate in a one-shot Prisoner’s Dilemma

Social capital, associations and civic republicanism

189

only if he has social trust, that is, a belief that people in general are trustworthy. This points to one of the main problems in the social capital literature: the generation of social trust beyond the limits of associations. One possible solution to this problem is to claim there is a relation between social capital, in the form of associations, and civic virtue. The possibilities of this relation are discussed in the next section. But before that it is necessary to consider the connections between civic virtue and social trust. In order to do that, we have to determine if it is possible to sustain the view that a virtuous citizen is not only a citizen with co-operative preferences, but one who also displays social trust. For some republican authors, it is clear that widespread trust is necessary for the good of the community. Machiavelli, who often expresses a negative attitude towards trust, viewing it as a form of dependence and a weakness incompatible with autonomy, recognizes nevertheless that if suspicion becomes so widespread that there is no more trust, men are rendered incapable of citizenship and real manhood (Pitkin 1984: 21, 101). However, this general statement is not enough to claim a positive relation between social trust and civic virtue in the republican tradition. Together with arguments centred on the value of autonomy, there are other arguments against that relation. One is that the republican tradition has always been suspicious about corruption or behaviour – by governments or fellow citizens – against the common good of the community. This can be thought as contradictory to trustful behaviour. A possible strategy to connect civic virtue and social trust is to see trustful behaviour as a signal. The display of social trust by the virtuous citizen, in the form of co-operative overtures in social interactions with unknown people, can be interpreted by others as a signal of his trustworthiness. Given that a virtuous citizen (interpreted in the first sense) is committed to a certain outcome – the attainment of co-operative equilibria – he can consider bad experiences with defectors to be worthwhile if they send signals to virtuous people that he is trustworthy. If this signal works the result can be more frequent interactions between virtuous people. This group of unconditional co-operators will attain medium pay-offs higher than the group of defectors, and the result, as has been shown for a conditional strategy such as tit-for-tat, will be that the defectors transform themselves into co-operators. This is perhaps not a fully convincing argument for those who wish to defend a positive connection between civic virtue and social trust. It can be argued that when the virtuous citizen is considering his co-operative overtures as a signal of his trustworthiness he is not displaying social trust: he does not actually believe that people in general are trustworthy, but only that some people are. A possible answer to this objection is that the outcome is the same as if the virtuous citizen believes that everybody is trustworthy. According to this argument, we can defend a connection between civic virtue and social trust (or, at least, ‘virtual’ social trust). It remains for us to show the connection between social capital, in the form of associations, and civic virtue. This will be discussed in the next section.

190

Francisco Herreros

Associations and civic virtue In this section I discuss some possible connections between participation in associations – the privileged object of study in the literature on social capital because of its links with democratic governance – and change of preferences in a virtuous sense. As we have seen, social capital is usually considered to be a set of resources derived from certain social relations that are useful to attain certain ends of individual actors. In this account of social capital, it is implicit that the preferences of the actors are given. That is why the presence of social capital has sometimes been considered a neutral resource for the attainment of whatever ends. This means that the consequences of the presence of stocks of social capital are not necessarily beneficial. The classical example is the Weimar Republic: the strength of secondary groups served in this case as an organ of socialization for an authoritarian ideology (Berman 1997; Rueschemeyer, Stephens and Stephens 1992: 113–14). In the last section I advanced a possible, though not conclusive, argument that civic virtue has a positive connection to social trust. If it is argued that a certain form of social capital – participation in associations – can be a source of endogenous transformation of preferences in a virtuous sense, then it can further be argued that the presence of social capital in a given society has beneficial effects not only because it is a resource for the solution of certain social dilemmas – especially the achievement of widespread co-operation through the display of social trust – but also because it generates civic virtue. A possible way in which participation can lead to an endogenous transformation of preferences in a virtuous sense is through deliberation. The belief in the power of deliberation is opposed to an understanding of the political process as a bargaining between different groups with given preferences. The republican belief in deliberation counsels political actors to achieve a measure of critical distance from prevailing desires and practices, subjecting these desires and practices to scrutiny and review (Sunstein 1988b: 1548–9). In the republican literature we can find at least two arguments to sustain the view that deliberation can lead to the transformation of preferences towards the common good. The first of these arguments refers to the structure of the deliberative process; the second, to the actors participating in the deliberative process. This second argument is the less interesting for our discussion. It claims that deliberation can lead to the attainment of the common good because the participants have certain characteristics, those of the virtuous citizen. This strategy is not useful in the present context because it does not claim that civic virtue is an outcome of the deliberative process, but rather a prerequisite to it. The first argument contends that certain features of the process of deliberation can lead to a change of preferences. One of these features can be the rules of the process. Sometimes it is argued that if the discussion is public, there is a pressure to abstain from egoistic arguments (Elster 1995: 390). It is claimed that there is a desire in the participants not to appear selfish, because it would be embarrassing

Social capital, associations and civic republicanism

191

or shameful (Fearon 1998: 54). Another possible mechanism to explain why the participants in a deliberative process usually justify their views in terms of the common good is the psychological mechanism of reduction of cognitive dissonance: individuals tend to reconcile what they do with what they think, to reduce dissonance (Elster 1987: 113). Another feature of deliberation that can lead to a change of preferences is that deliberation can reveal private information (Fearon 1998: 46; Gambetta 1998: 22). Some participants can reconsider their preferences in the light of newly available information. A common problem with these supposed effects of deliberation is that they are open to strategic manipulation of preferences by the participants. The latter can formulate their preferences in terms of the common good, for example, solely in order to manipulate the other participants. We can borrow an illustration of this from Rousseau’s discussion of the transition from the natural state to the civil one: the proprietor convinces his neighbours (who are a threat to his property) to go through this transition invoking a general interest (protection of the weak from the strong, peace, harmony) that hides his actual selfish interests (Rousseau 1990: 179–80). Of course, the incentives for the manipulation of preferences are higher when there are conflicting interests between the partners in the discussion, as in Rousseau’s example. This is also the case for interactions between virtuous and non-virtuous citizens. There, mere discussion may not necessarily lead to the common good, because the conflict in the initial preferences of the participants can induce them to try to manipulate the other participants. Once again, we see that in order to attain the common good, the participants in the deliberation process have to display certain characteristics from the beginning: courtesy, empathy, and reasoning ability, for instance. If they possess these characteristics at least, the deliberative process can lead to real consideration of the general good of the community. Of course, conditions of equality between the participants are also required. These characteristics are a prerequisite for deliberation to take place. Nevertheless, this is not the same as claiming that the participants have to be virtuous from the beginning. Empathy and reasoning ability can be considered characteristics of the virtuous citizen, but they are not the only ones. More important are the identification of private and public interests, and the willingness to sacrifice private interests to serve public ones. Empathy can be thought of as a component of such preferences, but they obviously go beyond empathy. However, even if these prerequisites are fulfilled, it is not clear that the deliberative process would lead to a transformation of preferences in a virtuous sense. If there is an initial conflict of interest, there will always be an incentive to cheat the other participant. This is not to say that in the case of initial proximity of interest the outcome is a transformation of preferences in a virtuous sense. In that case, there would be no need for a transformation of preferences. So it is far from obvious that the deliberation process will lead to a virtuous transformation of preferences. In the case of initial conflicting preferences, the incentives to manipulate are strong. The best we can expect is probably an exclusion of the most crude expressions of selfishness, in order to avoid shame or as a manifestation of the capacity of empathy that is a prerequisite of deliberation.

192

Francisco Herreros

Only if the participants are virtuous from the beginning could it be possible to attain the common good. We can conclude, then, that the connection between deliberation and the generation of civic virtue is unclear, to say the least. It remains now to examine the relation between deliberation and social capital as association. According to the social capital literature, nearly all types of association can be considered forms of social capital. Nevertheless, we can also find in the literature some attempts to exclude certain types of association as forms of social capital. For example, Robert Putnam felt that vertical relations should be excluded as a form of social capital, because they are less helpful than horizontal networks in solving dilemmas of collective action; in a vertical relation opportunism is more likely on the part of both patron (exploitation) and client (shirking) (Putnam 1993: 175). Members of a vertical association may stop thinking about how to deal with each other and concentrate instead on how to cope with the shifting demands of the more powerful agents above them. In a quite different way, horizontal relations, especially networks of civic engagement, foster robust norms of reciprocity, facilitating communication and improving the flow of information about the trustworthiness of individuals (Putnam 1993: 173–4). This way of excluding a certain type of association, based on its capacity to overcome dilemmas of collective action, is not very convincing. Putnam uses it to exclude the Mafia from his analysis of social capital and institutional performance in Italy. Nevertheless, the Mafia is not an organization characterized by problems in overcoming dilemmas of collective action. It uses various enforcement procedures, including death threats, to ensure the trustworthiness of members of the organization. The result is a high internal capacity to solve dilemmas of collective action, although its effects for the wider society are the consolidation of social distrust (Gambetta 1988). A second criterion to be used in differentiating among associations distinguishes between public and private goods-producing associations (Boix and Posner 1996: 9–13). According to this criterion, civic associations dedicated to the provision of public goods will produce a stronger form of social capital than those dedicated to the provision of private goods. The reason for this is that in the case of the provision of public goods there are strong incentives to defect, so the successful maintenance over time of a public-good producing association is a signal of the creation of a robust form of social capital. This criterion is not very clear, however. The existence of public-good producing associations is not necessarily a signal of the previous presence of social capital, because social capital is not a necessary condition for the provision of public goods. Coercion or other forms of selective incentives can do it as well. Moreover, it is not clear that public-good producing associations can create a more robust form of social capital. Certainly not if we understand ‘stronger’ as ‘more enduring’. In an association created for the pursuit of a private good, social capital can be created as a by-product, for example in systems of mutual trust such as friendship. There is no reason to believe that these relations are not going to be as enduring as they would be if created inside a public-goods producing association. ‘More robust’ can also mean ‘more useful’

Social capital, associations and civic republicanism

193

for the solution of social dilemmas. It is not obvious, however, that relations of trustworthiness, for example, will be more frequent in a public-good than in a private-good producing association. I propose another criterion to differentiate between associations, based on their capacity to promote deliberation. This is related to the capacity to solve collective action dilemmas: deliberation can induce a transformation of preferences in a virtuous sense, and virtuous citizens can solve collective action dilemmas because they have co-operative preferences and display social trust. I shall draw distinctions between horizontal, vertical, civil and political associations. First, the distinction between horizontal and vertical associations. Although the criteria of differentiation are not very clear, I assume that in horizontal associations there is more equality among members, in terms of participation in the decision-making process, than in vertical associations. In vertical associations, decisions may well be adopted without the participation of most members. On the contrary, in horizontal associations, there tends to be a high rate of such participation. Deliberation is a form of decision-making incompatible with vertical relations. There are no incentives for deliberation, given that the results of the deliberation process have no influence in the working of the association. The beneficial effects of deliberation on the preferences of the participants are by-products of the participation, but the decision to participate in the deliberation process is instrumental: it serves to obtain a certain outcome (Elster 1987). Without the possibility of influencing decision-making, there are no incentives to deliberate. In the case of horizontal associations, deliberation can also be absent. If what is required is participation in the decision-making process, this can be done through voting. But at least there is room for deliberative processes. There are fewer relations of dependence between members, and so fewer conflicting interests from the beginning. This means a reduction of incentives for manipulation. Second, we have the distinction between civil and political associations. This distinction is taken from de Tocqueville and is related to the objectives of association. Deliberation as discussion can take place in both kinds of association. But, from the point of view of deliberation as a source of transformation of preferences in a virtuous way, political associations seem to be more interesting, if we consider the content of the discussion in both types of association. Some of the effects of deliberation, such as the revelation of private information or the overcoming of bounded rationality (Fearon 1998), can take place in civil as well as in political associations. But the end of deliberation, from a republican point of view, is the transformation of preferences in a virtuous way. This means that the citizen must equate his interests with those of the community, or sacrifice his private interests on behalf of the community. As we have seen, it is not clear that deliberation can lead to this transformation of preferences. But, in any case, it can be argued that if the content of discussions among the members of an association is related to the interest of the community, this transformation is more likely to occur than if the discussion is, for example, about who should play in the next match of your football club.

194

Francisco Herreros

So, according to this criterion of differentiation – the capacity to promote deliberation – horizontal political associations are those with the greatest capacity to promote deliberation, followed by horizontal civil associations. Vertical associations, political or civil, are highly unlikely to generate deliberation. Ideally, this criterion could be equated with differences in the generation of social capital, given that deliberation is related with civic virtue and this with social trust.

Conclusion My conclusion, from the point of view of the social capital research paradigm, is that a close analysis of the republican tradition in political theory can be fruitful. A conception of social capital that views it not only as a resource for the attainment of given preferences, but also as a source of the transformation of those preferences in a virtuous sense, could be important for debates on the ‘neutral’ character of social capital. I have argued that one way to defend this concept of social capital is to claim that participation in certain types of association (especially political associations with a horizontal organization) can lead, through deliberation, to a virtuous transformation of preferences; and, from that transformation to the generation of social trust. The mechanisms of this causal chain, unfortunately, are not clear enough. The connection between deliberation and civic virtue is dubious, and the argument advanced to explain the relation between civic virtue and social trust is not an argument clearly in favour of the generation of social trust, but, at best, of ‘virtual’ social trust.

Notes 1 Although not explicitly presented as such, some recent arguments regarding civic republicanism and participatory democracy, such as Barber’s proposal of a ‘strong democracy’, or Cohen and Rogers’s proposal of an ‘associative democracy’, can equally be understood as proposals for public policy aiming at the creation of stocks of social capital (Barber 1984; Cohen and Rogers 1995a).

15 Deliberative democracy versus direct democracy – plus political parties! Ian Budge

Deliberation and decision Earlier chapters have talked a lot about deliberation, without saying very much about how it will support democratic decisions. There is indeed a desire to put critical content into ‘liberal constitutionalism’, regarded as formalistic and empty (see Dryzek, Chapter 6). Perhaps as a result the rules governing deliberation and the end towards which it must tend – collective choice – have been hardly mentioned. The general assumption is that a consensus or at least a ‘workable agreement’ will emerge from informed debate. The main questions of concern to practising political scientists – the biases imparted by different electoral systems, voting cycles and political structures in general – are thus ignored as irrelevant to the fresh impetus which extended deliberation will impart to democratic processes. This sideways look at collective choice is refreshing. I shall argue below that we should all take the nature of political communication and discussion more seriously, perhaps more seriously than deliberationists do. Concentrating on deliberation to the exclusion of the procedures under which it operates, however, runs the risk of ignoring the really central questions about democracy in favour of others which are frankly more peripheral. Saward’s comments (Chapter 5 above) on the fair aggregation of preferences and their governmental representation as the central legitimating characteristics of democracy, are highly pertinent here. Put with brutal frankness, the question is whether deliberation really matters much where citizens are deprived of the right to decide, or where their decisions are ignored. The tendency among deliberationists is to assume that any set of reasonably democratic institutions will do provided that they are informed by widespread discussion and debate. This chapter argues that institutions and rules are crucial factors so far as democratic innovation is concerned. Unless citizens are given the fullest possible powers over policies affecting them, politics will remain defective in democratic terms no matter how reasoned or informed discussion is. Conversely, democracy could function with minimal public discussion even though the general hope is that fuller access to decision-making will stimulate it. This point is really the central one at issue in the dialogue between Saward (Chapter 5) and Dryzek (Chapter 6). Dryzek regards Saward as equating

196

Ian Budge

democracy with the counting of heads. I assume that Saward accepts that at bottom it is. Democracy has to be voting fairly and accepting majority decisions. There is no other way of making governments and policies conform to popular preferences – which must always be its central justification. The fact that these processes throw up problems, such as the possibility that there is no real majority for any course of action (voting cycles) or the isolation of permanently outvoted minorities, is no reason for ignoring them. On the contrary the fact that there are problems with voting procedures provides a very good reason for analysing them more intensively. Dryzek assumes, without saying so explicitly, that these problems are insurmountable, and hence that some alternative way of carrying on democracy has to be found.1 There are two arguments against this conclusion. The first is that there are many possible solutions to majority voting problems, some of which operate in actual democracies – accounting for the fact that voting cycles are rarely observed in practice (Budge 1996: 133–48). The democratic solution to permanently outvoted minorities is to allow them rights of (full or partial) secession (Budge 1996: 159–71). The other counter-argument to Dryzek’s position is that his alternative to democratic voting – ‘contestation of discourses’ – in so far as it can be pinned down, seems not very different to newspaper and media discussion of issues at the present day: ‘Examples of [the contestation of discourses] might involve industrialism, environmentalism and sustainable development; patriarchy and various kinds of feminism; realism and idealism in international affairs. In this light, public opinion can be interpreted as the provisional outcome of this contestation, as transmitted to the state through a variety of discursive means’ [bracketed words and italics mine] (Dryzek, this volume, p. 84). The point left hanging in the air, however, is how can the capitalist-dominated state be made to respond to public opinion? Surely there is no other way than voting on decisions to which the government is constitutionally bound to respond. This would point to direct democracy rather than deliberation as the way forward to any wider democratization of society. Direct democracy is thus more relevant to an extension of popular influence than deliberative democracy, though it may carry the latter in its train. To the extent that they are opposed – in the sense of institutional access being downplayed in favour of non-institutionally-anchored deliberation – direct democracy is far and away the more important instrument of democratic advancement. If trade-offs are required, deliberation should be sacrificed in favour of popular control by anyone concerned to extend democracy. To the extent that quality of public deliberation is made an argument against extensions of popular policy voting, reasons which originated in support for direct democracy (Barber 1984) are being turned against it. Political parties have also to be considered in any assessment of democracy – deliberative, associative, representative, or direct – because, if a consensus exists among empirical analysts, it is that parties are essential to it, both at a practical and theoretical level. Whether under representative or direct democracy, they largely define the nature of the policy discussion and the alternatives being voted on, as

Deliberative democracy, direct democracy and political parties

197

well as implementing (sometimes not implementing) them in government. Counterfactually, if we imagined democracy without parties and wanted to know how it could work, we have to envisage proto-parties emerging as groups of likeminded individuals getting together to advance and implement their point of view. The intimate conjunction of parties and democracies makes it curious that most analyses of deliberative democracy (with honourable exceptions, such as Cohen 1989: 31–2) do not even mention them! Can effective deliberation even proceed without party interventions (or organization by adherents or activists)? This chapter accordingly starts by considering the general role of parties in democracy, arguing that far from being a contingent means through which popular preferences may be expressed, they are in fact necessary to the shaping of such preferences, to deciding between them, and to their implementation. Adapting Michels (1915/1959: 400) we might assert ‘who says democracy says parties’. The centrality of parties has consequences for political rhetoric and the nature of deliberation which we go on to consider in Section 3. This leads naturally to the question of voting and elections, with which parties are so intimately associated (Section 4). These seem necessary mechanisms through which the conclusions reached by deliberation are transformed into action. Is there any way we can dispense with them? The answer must be ‘no’. But particular ways of organizing votes may bias results in particular ways, subverting the free expression of popular will. We need to consider these. Parties may also simplify deliberation for ordinary citizens who are normally criticized for incompetence, ignorance, fickleness, and intolerance by critics of both deliberative and direct democracy (Section 5). The chapter concludes that extended deliberation is an important adjunct to direct democracy and hopefully a consequence of it, but it is not the same thing. Effective participation is what counts: within this, deliberation may not conform to the ‘university seminar’ model assumed by most of its theorists.

Parties and democracy Is there any democracy that has ever operated successfully without political parties? Ancient Athens is the counter-example that springs to mind. But in fact there is considerable evidence that parties existed in the form of political clubs (Bonner 1967: 45, 61). Pericles could not maintain a popular majority year after year through rhetorical persuasion alone. He had to ensure that his supporters attended the popular Assembly and voted the right way – classic functions of the political machine in all ages. Mass democracy, according to the established account (Lipset and Rokkan 1967: 1–67) was created by parties organizing excluded groups and pressing for the franchise, in order to safeguard their interests. Associational groups and voluntary organizations need internal factions or parties to avoid oligarchy (Lipset et al. 1956) – although most organizational leaders manage to suppress them (Michels 1915/1959). The invariant conjunction between freely competing parties and democracy

198

Ian Budge

raises the question of whether it is more than a strong empirical correlation – is it not in some sense a necessity for any political system that aims to decide public questions by popular vote? In order to get policies or candidates voted in, is it not essential to get them on the agenda and to organize sympathizers? And what is a political party if not a group of like-minded people acting together to promote particular policies and candidates? The fact that parties are fundamentally necessary to democracy makes it odd that most discussions of democratic theory, even those sympathetic to the idea of widespread popular participation, bring them in only towards the end of their exploration of ‘citizens . . . having an equal effective input into the making of binding popular decisions’ (Saward 1998: 15; Cohen 1996: 95). Effective inputs depend on citizens being able to control both the agenda for decision-making and decisions themselves. How else can they do this but by joining together in what are in effect parties? A more relevant definition of democracy is therefore one which makes party participation as well as individual voting rights central, along the following lines: Democracy is a system in which all citizens are free to join with others, on a temporary or lasting basis, to: • advance candidates and/or polices for public adoption; • vote upon them without coercion under procedures which guarantee that the alternative with greatest popular support is accepted; • implement accepted alternatives. Such a definition recognizes both the crucial agenda-setting side of democratic debate, well-recognized in deliberative approaches (Benhabib 1996a: 70), as well as the decisional act of voting. In putting these together it follows influential procedural definitions (such as Dahl’s, 1956: 84)). Its originality however lies in stressing the collective nature of both agenda-setting and voting through parties – and one must add implementation, an oddly neglected subject but vital to popular empowerment. Parties are the sole organizations incorporating both electors and governments so they are uniquely placed to supervise implementation of accepted policy – or to be held accountable if it is not implemented. Usually a definitional stress on parties sees them in elitist terms, with leaders competing for votes, after which popular deliberation is suspended until the next general election (Schumpeter 1947: 269). It is important to note that the formulation here is just as compatible with direct democracy, characterized by permeable organizations and frequent voting on policy, as with representative democracies where both are less obvious features of politics, though not wholly absent. As I have argued in detail elsewhere (Budge 1996), agenda-setting, voting and debate need to be defined by parties even – perhaps particularly – under direct democracy. If they are not, who else can order debate sufficiently to come to definite political conclusions? The safeguard for democracy is that parties themselves can be freely formed and potentially joined by anyone in agreement with their objectives.

Deliberative democracy, direct democracy and political parties

199

If democracy is about rendering citizens equally effective in making public decisions, direct democracy must be the way to render existing democracies more democratic. The importance of parties in organizing and running direct democracy – including mass political deliberation – thus makes them central to democratic advance. In addition to the focus on associations in the previous three chapters, emphasizing the role of parties is one way of counteracting what often seems like an extreme individualistic bias in modern democratic theory – no doubt stemming from its origins in Enlightenment and liberal thinking. The stress falls entirely on the empowerment and equality of individuals. The term ‘citizen’ often appears to signify an atomized individual confronted by the collectivist power of the state, not a person rooted in family and social groups with which she is used to act, which can then form the basis of other mediating institutions. Associational theories of course take groups and organizations on board but then leave the details of how they would function to the imagination. How are they going to organize themselves internally (competing factions and groups?) or in relation to the state (political parties?). Some associational theorists such as Perczynski (Chapter 11 above) have turned to neo-corporatism to provide a model for associative democracy (see also Hirst 1993: 18). In the earliest of its protean forms (Schmitter and Lehmbruch 1979) neo-corporatism was the idea that labour and business leaders would get together with governments to hammer out an industrial policy for growth, rendering elections and party competition irrelevant to the central political decisions taken by the state. Without internal accountability of organizational leadership secured by such devices as elections and competing factions, the concept seems to sit ill with democracy. The practical relevance of neo-corporatist ideas has also been undermined by the way in which strongly ideological party governments were able to impose their own policies on both business and labour without consulting anyone outside their own ranks in the course of the eighties and nineties. Parties have therefore emerged yet again as the key mediating institutions in any form of democracy. The key to democratic advances lies in making them more open and internally democratic, and in encouraging their participation in direct policy elections (referendums and initiatives) rather than leaving them to seek an infrequent mandate every three to five years, whose ambiguity enables them to claim general authorization for everything they do with few fears of being held accountable for it. Policy elections, and party competition to get competing proposals approved, have considerable implications for the type of deliberation given to policy. We consider the effects of party rhetoric and the way it is likely to steer discussion in the next section.

Parties and deliberation The pervasive model of discussion which most deliberative theorists seem to have in mind (though it is rarely specified) is what might be called the university

200

Ian Budge

seminar. Certainly this model reasonably describes, for example, the abstract portrayal of ideal deliberation by Cohen (1989) and the specific forums of Fishkin (1991 and Chapter 1 above), though it may be less relevant to the models of theorists who focus more on diffuse deliberation in civil society (such as Dryzek). In the university seminar model a proposal is presented and detailed; this is then criticized by other participants, who may present counter-proposals. These are discussed in relation to the original. Relevant results are summarized and presented. When the discussion seems likely to bring in extraneous points or ramble off track, it is re-focused by a benevolent professor or moderator, who also encourages the shyer or more reluctant members to contribute. In the end she may get the seminar to express a collective opinion on which proposal seems most viable, though often the conclusion may be that each has advantages and disadvantages so no clear-cut decision between them seems appropriate. Of course, Dryzek’s conception of discursive democracy as set out in Chapter 6 above deliberately widens out this mode of discussion, ‘refusing confinement to reasoned, measured, gentlemanly argument’. Even here, however, the endorsement of Habermas’s conception of ‘communicative rationality’, as ‘free from domination, coercion, manipulation and strategizing’ seems to point in a different direction, where political rhetoric and debate become unrealistically sanitized. Though it manifests itself in various ways, there appears to be a common deliberative tendency to take the politics out of politics. The promise to do so has of course often been a powerful appeal of authoritarian movements which promise to stop pointless squabbling and divisiveness in favour of ‘obvious solutions’. The real lesson is that one cannot take the politics out of democratic politics, so attempts at domination and strategic manipulation will always be with us. What we can hope to do is contain them by rules and institutions, the fairness of which is a key point of debate in discussing democracy. Fishkin’s brilliant suggestion of a deliberative opinion poll (Chapter 1), where a cross-section of electors are polled before and after intensive collective discussion of a problem, guided by experts, is essentially an extension to the general public of the university seminar, or conference. The extension is a very attractive one. It has been delightful, in the two televised examples staged in Britain by Grenada (1994 and 1996), to see how ordinary people take on board informed contributions by specialists and use them open-mindedly to revise their opinions. Thus in the 1994 debate on prison sentencing majority opinion shifted from strongly endorsing longer sentences before the conference, to strongly opposing them after the weekend seminars and panels. Five years on, it is noteworthy that both Government and Opposition in the UK are still committed to tough treatment of offenders. The willingness of the public to change its opinion on the question has not been picked up by the political parties. If deliberation of this kind is to remain purely advisory, this is an effect we can anticipate. Like Royal Commission reports, think-tank recommendations, parliamentary resolutions, and other advice, governments will pick and choose according to what they think is expedient and are more likely to pay attention to the tabloid press than to deliberative polls. This, incidentally, also points up

Deliberative democracy, direct democracy and political parties

201

the limitations in Dryzek’s airy reference in Chapter 6 to public opinion being ‘transmitted to the state through a variety of discursive means’. Suppose a drastic institutional reform changed the role of deliberation of this kind from advice to control. Suppose actual powers of policy decision were given to random samples of citizens who after an extended conference were polled to decide between proposals on each issue, which were then put into effect. This would be one way of installing a deliberative democracy. Randomness of selection would ensure representativeness of panels, so every citizen would have an equal chance to participate. University-style seminars and lectures by experts would ensure due consideration was being given to all arguments. In this way deliberation would be focused and made effective, and would produce better policy – or at least the policy which would be preferred by most of the population if they gave it thought. Sampling in this way also guards against a common criticism of deliberative proposals – that they would take up too much time, that citizens are ill-informed, etc. Having decisions made by sub-sets of citizens who will be informed takes care of that. However, once such citizen samples were put into decision-making roles, how long could we anticipate the university seminar model of discussion to continue? The first reaction from those who disagreed with the decision would be to form a party – if they were not in one already – to campaign against it. If they disagreed with the whole string of proposals for reform they would campaign against the whole structure of deliberative decision-making. They would claim a voice in the selection of their experts in a new debate, warn members against dangers and consequences they had not considered, attack whatever method is used to aggregate opinions (which is in any case debatable; cf. Section 4 below). Counterparties would spring up to support the other side. Members of the panel would be investigated and interviewed to demonstrate how they were or were not unfairly pressured into taking the position they did. In other words normal politics would resume and overwhelm seminar discussions. This thought-experiment (not in any respect a proposal made by Fishkin or his associates) serves to demonstrate how hard it would be to keep deliberation on an even keel once it became a mode of decision-making. Parties or proto-parties, opposed factions and competing groups are inevitable once individuals feel something important is at stake which they need to promote or oppose. There are such political payoffs in co-operating and organizing that it is inevitable in collective decision-making. Once such groups are formed they will be interested not in deliberation in contrived forums, but in winning. This follows because parties or proto-parties do not need to discuss proposals. They already know which ones they want. Groups of this kind are bound together by an ideology in Plamenatz’s (1972) sense of assumptions about human nature and society that indicate what course of action to take in uncertain or ambiguous circumstances. Given this, their objective is generally not to develop proposals but to get them accepted. Party participation thus transforms seminar discussions into courtroom dramas. The object in forensic debate is not to make up your mind on whether or not your case is a just one. That is already given by your remit. The object is to win the case.

202

Ian Budge

That does not eliminate deliberation and debate, but it does change its purpose. The object now is to put forward arguments which will induce third parties – judge, jury or electors – to come down on your side. Recent analyses of party election programmes show that this is done in a particular way in the public forum. Parties rarely try to convince an adverse majority of the merits of an opposed position. This takes too much time and consumes too many of their resources. Instead they try to change the frame of reference within which the issue is viewed and thus substitute for the one currently at the focus of attention another more favourable to themselves. Appropriately, Tony Blair used this tactic in the course of Granada’s televised deliberation on sentencing in 1994, with the slogan ‘Tough on crime, tough on the causes of crime’. This implied that Labour, like the Conservatives, believed in the tough, majority-supported attitude to criminals. But it also raised the question of whether a more effective approach might not be to tackle social problems such as deprivation, poverty and unemployment which drive people to crime. These are issues where the vast majority support action and where Labour is regarded as more effective than the Conservatives – which would not be the case with law and order as such. If successful in shifting the focus in this way Labour can hope to get more support either for its specific policy-proposals or for its overall programme in a general election. By putting together a whole set of ‘left-wing’ as opposed to ‘right-wing’ priorities (welfare, government intervention and peace versus order, freedom and strength) parties help to simplify choices for electors (Section 5) and avoid arbitrary majority decisions (Section 4). So their strategies serve useful democratic functions. Avoiding direct argument over the merits of competing proposals on the same issues hardly fits the standard university model of deliberation however. The choice may be between keeping deliberation out of mainstream politics and preserving these attractive features of debate, or losing them by giving ordinary citizens decisional powers. This may indicate only that deliberative theorists need to look more closely at what type of deliberation they are advocating, rather than taking it as unproblematic. Debates about overarching priorities, and attempts to shape discussions over specific issues so that they fit into the left–right dimension, do not after all signify an absence of deliberation. They may simply indicate the form that decision-oriented deliberation in a complex modern polity needs to take. While many of the participants in such debates already know what they want and are seeking to manipulate discussion rather than be persuaded by it, others will have their minds made up by them. It may even help those who do not want to deliberate, but who should not necessarily be deprived thereby of their right to participate and decide (Section 5).

Parties and procedures A major theme running through discussions of deliberative democracy is that deliberation continuously changes preferences. It is not appropriate, therefore,

Deliberative democracy, direct democracy and political parties

203

within this context, to adopt liberal assumptions about citizens having fixed preferences for proposals which then generate problems and paradoxes in voting procedures (Benhabib 1996a: 71). The preferences will change (and probably come closer?) in the course of discussion because equal participants in a public forum will be ‘forced to think from the standpoint of all involved for whose agreement one is “wooing”. Nobody can convince others in public of her point of view without being able to state why what appears good, plausible, just and expedient to her can also be considered so from the standpoint of all involved’ (1996a: 72). One can certainly agree that general arguments rather than selfish ones are necessary in public debate. The problem is, as pointed out above, that there are opposing points of view about what will benefit everyone, which parties and lawyers will try to push, depending on whose interests they represent. This serves to prevent the consensus emerging that some theorists see as the result of deliberative processes functioning properly (Dryzek 1990a). Even university seminars do not usually end with consensus, however. Face-to-face debate may consolidate rather than overcome enduring controversies, as the history of many small communities unfortunately shows. Extended public debate, with representatives of opposing views appearing on television, may also aggravate disagreements. Short of consensus, the question of how to register the outcome of discussion remains problematic. From a viewpoint sympathetic to deliberative democracy Miller (1993a: 91) points out that the probability that preferences may change in the course of discussion does not alter the fact that they will be fixed at the moment of voting. Thus the arbitrariness inherent in voting procedures remains a problem if deliberation is to lead to concrete decision. What are these voting procedures and their problems? There are three major ways of aggregating preferences: (1) By taking only first preferences and choosing the one that gets most votes. This has the disadvantages that the winning alternative is not guaranteed majority support and may indeed be strongly opposed by a majority. It also ignores preferences between the other alternatives and may lead to a tie where first preferences aggregate to equal amounts for each alternative. Ties could be broken by chance (throwing dice) or arbitrarily. A more appealing alternative might be to take the whole rank-ordering into account, in which case the other procedures listed below come into play. (2) The first of these to be suggested historically (Borda 1784) was to use the rank orders to aggregate scores. Scoring the last preference of each voter as number a (usually 0), second last a ⫹ b (b is usually 1), third last a ⫹ 2b, and so on gives scores to each alternative which reflect voters’ individual orderings. These ‘Borda counts’ can then be added over all voters and the alternative with the highest score selected. Borda counts do in a broad sense reflect the preferences of as many people as possible (Dummett 1984: 142). However, apart from their assumptions about scoring, they have two problems:

204 •

•

Ian Budge They can be manipulated. For example, by ‘strategic’ voters placing the strongest rival to their most preferred choice at the bottom of their orderings. Or a ‘stalking-horse’ alternative can run but then drop out. Borda counts may not always select the alternative that would have won in a series of votes with other alternatives.

(3) This suggests another voting procedure in which each alternative is compared in turn with each of the others, and the one which secures a majority of preferences over all these confrontations, a ‘Condorcet winner’, is then chosen. As McLean (1989: 55) observes: democracy involves each vote being weighted equally . . . this puts the concept of a Condorcet winner at [its] heart. A Condorcet winner is unquestionably the majority choice. To reject a Condorcet winner is to choose somebody who would lose a majority vote to the Condorcet winner. It is therefore to favour the minority against the majority. There is an argument that strict majoritarian decision-making is not necessarily preferable under all circumstances to getting as close to the wishes of as many people as possible (Miller 1993a: 87). However, reflecting the wishes of the majority is obviously important in a democracy. Unfortunately, these may not always be reflected by Condorcet procedures. This follows from the following considerations: •

•

Condorcet procedures are also open to manipulation. An example is given by Miller (1993a: 80) in a choice of energy options, in which nuclear power is the Condorcet winner if everyone votes sincerely. Someone strongly committed to coal might stop the nuclear option by voting insincerely for gas when the choice between nuclear power and gas comes up, thus preventing the emergence of a nuclear winner and hoping, in the resulting unstructured situation, that coal may win. A further difficulty with Condorcet procedures is that no clear majority winner may emerge. Take a situation where 30 voters prefer option A over B and B over C, 20 have the ordering B to C to A and 20 prefer C to A to B. In a series of pairwise comparisons, A wins over B (50–20): B wins over C (50–20) but C wins over A (40–30). There is thus no clear winner: everything depends on the order of voting and the specific procedures employed (for instance if the C–B vote was held first and then the B–A vote, nobody would be in much doubt that A had won, but if the sequence were the other way round, C would win).

Miller (1993a) makes the interesting suggestion that public deliberation may help resolve some of these problems by eliminating the eccentric preference orderings which generate Condorcet cycles (see also Dryzek and List 1999). This does seem a likely effect of the imposition of a public frame of reference on discussion. Miller

Deliberative democracy, direct democracy and political parties

205

also dismisses strategic voting as unlikely to be practised on any large scale owing to citizens’ inherent moral sense, perhaps buttressed by public-oriented deliberation (and one might add, difficulties of communication and co-ordination for any group wanting to organize it on a large scale). The argument for deliberation wholly overcoming collective decision problems is not conclusive however. Short of a consensus emerging there is no guarantee that deliberation will not end up with precisely the kind of opinion division between three or more alternatives that generates cycles. It is just as likely that discussion will uncover more alternative possibilities than were thought of previously, as that it will narrow them down to one or two options. Indeed, given the likelihood of party interventions of the type outlined above, deliberation may harden the divisions being voted on rather than reduce them. There are two ways that Condorcet cycles can be avoided in voting. One is to consider only two options – by focusing on one proposed change versus the status quo for example. Such a choice could be imposed by constitutional rule – proponents of change have to agree on one proposal to put up against existing policy. However, a straight two-option choice seems unlikely to emerge spontaneously either from deliberation or from party competition (very few party systems are only two-party). Cycles do not emerge in situations where options can be ordered along a single dimension. This is because those at each end of the dimension will always prefer the middle (median) position to options at the other end. Thus, whichever end the middle voter sides with will gain a true majority, at the cost of adopting her position as general policy (as she can bargain with each side to have this accepted) (Black 1958; Downs 1956). Miller argues that deliberation should help produce one-dimensional voting by (a) as mentioned above, helping to produce a single frame of reference for considering the issue or (b) allowing participants to assess the issue and decide whether it involves only one or several dimensions of choice. In the second case they could choose to avoid the risk of cycles associated with multi-dimensional decisions by agreeing to disaggregate the decision into its various components. For example, in the case of what type of power stations to build the dimensions would split into: (a) basic values of environment versus economy, and (b) technology – is coal-fired better/worse than gas-fired or nuclear in meeting the agreed basic objectives (Miller 1993a: 85–7). On each separate dimension of decision there is a middle position that is determining in the sense specified above, so votes on each aspect of the decision will be non-cyclical and conclusive. One can certainly agree with Miller that if deliberation proceeded in this way the cyclical problem would be avoided. It is however a very big if. Most people’s experience is surely that seminar-type discussion complicates the way we view the world rather than simplifies it. An obvious effect of deliberation therefore is to stress complexity, so that we cannot make rational decisions without taking all facets simultaneously into account. McLean (1989: 121–4), taking a typical academic stance, argues very strongly for non-separability and therefore for the inevitability of voting cycles. To take Miller’s example of power stations,

206

Ian Budge

arguments about the potentiality of the various technologies may change one’s values, and vice versa. The decision space is thus properly two- or threedimensional and cannot be disaggregated into separate dimensions. Deliberation is likely to strengthen this view rather than weaken it. In contrast party competition at election time, aided and abetted by the mass media, tends to assimilate all issues to a single-dimensional contrast between left and right. There is indeed no logical reason why peace, intervention and welfare should be associated with the left and contrasted with strength, order and freedom on the right. It is simply that in practice the political parties put these bundles of priorities together – aided, of course, by the arguments of generations of thinkers back to Marx on the one hand and Burke and Smith on the other. However contingent these associations are they have great force, having endured over 100 years and in many countries (Laver and Budge 1992; Klingemann, Hofferbert, Budge et al. 1994). The left–right dimension imposed on election discussion by the parties gives more of a guarantee that voting cycles will be avoided than does deliberation. It simplifies rather than elaborates political discussion. Indeed it could certainly be accused of oversimplifying it – do not religious, environmental and national issues also enter into politics, but where do they fit into left–right contrasts? If they are located in them is it not a Procrustean fit rather than a natural one? From the viewpoint of the university seminar model this is certainly true. The left–right dimension is a crude generalization which squeezes down or overlooks most of the subtleties of political issues. Nevertheless, imperfect as it is, it may be the best available way to reach non-arbitrary decisions. We shall elaborate this point further by now considering politics from the point of view of voters and electors – including those who may not (want to) deliberate.

Parties and citizens One entirely valid point made by deliberative theorists from Barber (1984) onwards is to stress the capacity of ordinary citizens to participate in political discussion and reach rational informed decisions on that basis (Marcus and Hanson 1993). Fishkin’s (1991 and Chapter 1, above) arguments and experiments have lent support to this conclusion. Their work balances the anti-majoritarian, anti-popular bias in much of political theory which sees the masses as ill-informed, intolerant, fickle, dangerous – in short, best excluded from democratic processes so far as possible. The difficulty for the traditional viewpoint is that democracy justifies itself as empowering all citizens, so seeking their exclusion leads rapidly to anti-democratic positions and arguments. Deliberative theory with its stress on popular rationality and moderation is a valuable corrective to this. However, it is surely unrealistic in assuming that everybody will have the time and interest for intensive political discussion – even if a guaranteed minimal income freed people economically from many of their usual constraints. Even elite groups such as legislators do not get to grips with the details of every policy – very few in fact. They rely on short-cuts like getting advice on

Deliberative democracy, direct democracy and political parties

207

how the issue will affect their constituency or region, relying on expert opinion, especially that of party colleagues – or above all, following the party line (D. E. Matthews 1973: 32). Even academics usually take ideological short-cuts in areas outside their own speciality – and often within it, where hard information is lacking. Ordinary folk cannot be expected to do more. Outside specially selected groups like Fishkin’s it is hard to acquire or to interpret information. Without a good educational background in many cases the effort to do so may be exhausting and discouraging. It is true that the electronic transformation of news into entertainment makes it hard not to acquire basic political information and people do spend a lot of time watching news (Budge 1996: 64–5). The stream of discrete facts and reports which hit them may not be well-enough integrated to provide a good basis for in-depth deliberative discussion, however. Deliberative theorists also have to face up to the probability that even if there is a great untapped potential among the population, which we may well believe, many may not want to deliberate. As Oscar Wilde famously remarked about socialism, deliberation takes up too many free evenings (Dahl 1970). Citizens may be perfectly rational in wanting to use most of their time in other ways than politics – indeed, one end of politics might be to help them to find alternatives. But are those who do not deliberate thereby to be excluded from decision? In that case deliberative theory might take on a distinctively elitist tone. One way out of this impasse is Lupia’s (1994) distinction between useful and merely encyclopaedic knowledge. Lupia suggests – contrary to academic and survey analysts’ concern with having comprehensive contextual information – that much of this is irrelevant to the decisions electors have to make, on Californian initiatives for example. To cast a rational vote (i.e. one in conformity with one’s preferences) under uncertainty about the effects and implications of the proposal, it does not help to know the name of the Governor or the President. But it materially aids decision to know if it is endorsed by business or consumers, Republicans or Democrats. Of course, Fishkin’s deliberative polls might also provide action-oriented information for those who have not deliberated on the issue themselves. Short of having such polls on every issue however they do not have the range of party references. Moreover, once they became so central to political debate as continual polling implies, the considerations mentioned in Section 3 will come into effect. Results will be controverted, different parties and interests will run rival polls, which will no doubt come to different conclusions, depending on who sponsored them. So we will be back to party stands as the most effective short-cut to knowing how interests line up on a vote. Parties do in fact seem the most effective intermediaries in organizing popular voting so as to get informed and rational results. Deliberation may usefully second party effort, particularly if based on interactive electronic debate which reaches large numbers of people. Parties will be major participants in such debates anyway. But they also have the potential of co-ordinating the views of participants with those not reached by them, who presumably should not be excluded from

208

Ian Budge

deciding because they have not gone so deeply into the issue. As noted in Section 4, the simplified framework which parties impose on discussion and in particular their tendency to reduce complex matters to left–right choices then help reduce problems in the way votes are aggregated for decision. Are the non-hierarchical dispersed networks praised by Dryzek in Chapter 6 alternatives to political parties in this regard? The most telling bit of evidence here is the ecological movement of the 1970s, widely regarded as a novel political phenomenon and specifically opposed to organizing themselves like traditional political parties (Poguntke 1993). By the 1990s however the Greens had turned themselves increasingly into conventional political parties because there was no other way to compete effectively. We can anticipate that networks such as the ones described by Dryzek will be under the same pressures and follow the same course of development. There is no substitute for political parties in organizing debate or influencing decisions.

Conclusions: deliberative and direct democracy – and political parties! Deliberative democracy is attractive and desirable, and surely also feasible on a wider scale than it has ever been tried – but it cannot simply be equated with direct democracy. Popular deliberation is not the same thing as popular decision – unless we want to stretch the meaning of deliberation improperly, to encompass all the short-cuts people may employ to avoid deliberation while still aiming at a rational vote. The prime agency providing such short-cuts are political parties. For this and other reasons parties are essential to direct democracy. But they may subvert the kind of reasoned academic discussion most theorists have in mind when they write about deliberation. One thrust of the argument is indeed that deliberation is not a self-evident concept. There are many forms of deliberation: academic, legal, legislative, electoral. One suspects that many theorists have in mind an academic model. However, popular political discussion will in any case assume the characteristic of forensic rather than academic debate. Political parties and interests will be arguing for policies they have already adopted. Discussion may change some minds but not those of the main protagonists. We are back with the free market place of ideas where we rely on the good sense of citizen-consumers, operating under rules of fair competition, to make sure that the best ideas get enshrined in policy. Some will certainly deliberate – perhaps on their own, as Saward suggests in Chapter 5 above. But it is even more important that citizens organize and vote, and acquire the power to decide policy that direct democracy confers. The corollary is that deliberative theorists have to consider not only what adequate deliberation itself is, but also exactly how it relates to procedures and votes and improves the quality of direct democracy. To sum up, few would dispute that deliberative forums, like university seminars, are a good thing. They feed in to discussions of democracy, provide alternatives for policy-making and improve the intellectual quality of participants. These are

Deliberative democracy, direct democracy and political parties

209

indirect effects however. We do not generally want to substitute the seminar for the parliament, for to do so would subvert the qualities of both. Why then should we think of doing so when we move from representative to direct democracy? This is a radical enough change in itself. Should we not seek to ease it by preserving rules, strategies and institutions which have made politics manageable under the old system and can do the same for the new? The real case has yet to be made for focusing instead on deliberation.

Notes 1 Though Dryzek and List’s (1999) contribution to the ECPR workshop which stimulated this book suggested that deliberation may help to reduce the number of dimensions involved in collective choices, and thus to help induce single-peakedness in a set of preferences on a given issue (see the discussion of Miller’s (1993a) work on this issue in Section 4).

Conclusion

Variation, innovation and democratic renewal Michael Saward

The permutations of democratic innovation are many and complex. What really matters of course is which innovations are compelling as ideas and resonate with real political needs (though the shape of the latter is rarely agreed) – and which can find the political space to develop. Since the work of democracy is never done, any work like this one will necessarily be work in progress. Any attempt to summarize the key points of the fifteen chapters here in an effort to chart where and how they may take democratic theory forward will necessarily lack subtlety and completeness. Nevertheless I shall make a brief foray into that territory.

Dimensions of variation To help us, I offer here a modest illustrative taxonomy of persistent major (and mostly institutional) themes within democratic theory, together with some selective comments on how and why the arguments in this book pick up some more than others, and how they relate to the project of democratic theory as a whole. Broadly speaking, we can distinguish a number of key lines of variation with respect to democratic institutions, which together define a general menu of democratic possibilities. I put to one side more clearly non-institutional areas such as the justification of democracy, and more clearly empirical topics such as the economic or cultural contexts in which democracy is more likely to prosper. Though not limitless, the range of problems and issues, both theoretical and empirical, thrown up within and across the six dimensions I put forward is huge. I do not present it here to suggest that this volume deals with all, or even many, dimensions; rather, to help the reader to locate, within the broad sweep of democratic theory and innovation, the topics we cover. •

•

First, how ought the political unit and political community in which democracy is to be practised be understood, in terms of geography (whether for example the unit is to occupy a continuous slice of territory, or should be larger or smaller), population size, the inclusiveness of effective membership and degree of cultural homogeneity? Second, what constitutional constraints should democratic majorities face?

214

•

•

•

•

Michael Saward How far should constitutions entrench rights that are beyond the reach of majorities? Third, what are to be the formal structures determining the relationships between distinct sub-groups within the unit? Is the state (if it is a state) to be unitary or federal? If federal, then functionally or territorially? How far should formally separate sub-groups have autonomy from the ‘master’ unit to determine their affairs? Fourth, what is to be the balance between different forms of popular participation in the making of collective decisions, in terms of both (a) the balance between direct and representative forms or institutions, and (b) the balance of variations within each of these two basic forms? Fifth, how are relations of accountability to be structured, how are ‘accounts’ to be given, and to whom (by representatives, for example)? At how many points of ‘removal’ should effective decision-makers operate from their constituencies, or from those affected by their decisions? And sixth, how are the respective roles of the public and the private spheres, and formal and informal modes of political activity, to be understood, and which is taken to provide what in terms of the requisites of a healthy democratic structure?

I do not know how to construct such a list without building in biases of one sort or another. Nonetheless, accepting these six variations as a baseline for the moment, to what extent have the chapters in the book helped us to see ways in which the boundaries of democratic ideas are being or could be pushed back, or redefined? (I take on board the fact that some chapters comment on innovations, while others press for their own).

Political unit and political community One area where it might be said that political theory lags behind rapid real-world change is in rethinking democracy for the (it seems) global era. Held (1995) has mapped out a ‘cosmopolitan model of democracy’, but in terms of systematic efforts his is a lonely voice. In the introduction I mentioned that ‘global democracy’ was not a theme covered by more than passing comments in this book, and thinking about the taxonomy here reminds us that (re-)definitions of political community – whether national ones remain the baseline ones, and whether territorial ones will remain more important than cultural ones, for example – lie at the centre of democratic innovation. The contributions here work mostly (and implicitly) with the model of the political community as the nation-state. But the exceptions are illuminating. Eriksen shows how the most-developed experiment in supra-national political units, the EU, may require quite different ways of thinking by democratic theorists. What may normally be seen as a lack of democracy – unelected officials working through committee structures inured from popular pressures – may be legitimized as embodiments of forms of deliberation which reinforce inclusive and consensual

Variation, innovation and democratic renewal

215

decision-making. The heightened emphasis on deliberation, and the reduced emphasis on majoritarianism – themes present in one way or another in most chapters – renders Eriksen’s argument possible and makes it more plausible to contemporary eyes. Further, the idea of democracy-across-boundaries is glimpsed in provocative ways in certain discussions. Political ecologists or green political theorists are understandably keen on transgressing boundaries, since environmental problems and crises are hardly respectful of them. It is clear that we are seeing a new marriage of (a) deliberative and (b) ecological ideas that in turn enables creative forms of geographical boundary-crossing; for Dryzek, informal ecological networks of non-state actors across borders represent an instance of discursive democracy, no less than the vital ingredient in the forward progress of democratization in an age when liberal and constitutionalist ideas dominate national and global agendas. If we blend concerns such as the importance of local perspectives and empowerment (Bang and Dyrberg), the potential for local face-to-face participative forums (Smith), the local and cross-border openness of civil society to new linkages and ways of looking at problems (Rättilä, Dryzek), and new ways of regarding larger units as democratic (Eriksen), we begin to see that democratic innovations in the coming years will likely (and should) include a continuing loosening up of assumptions about the essentially national character of democratic political units and communities.

Constraining majorities It is interesting that in terms of formal decision-making no contributions here could be described satisfactorily as ‘majoritarian’. Most authors describe or seek new ways to constrain, educate, mould, build or even bypass conventional democratic majorities. The reasons behind this emphasis vary; some are distrustful of the incentives and knowledge confronting ordinary citizens (Fishkin and Luskin), others concerned more with the dangers of excluding the marginalized and/or the need for mechanisms to incorporate them (Rättilä, Squires, Meier, Eckersley). Associationalists and their critics debate whether traditional majoritarianism might be bypassed entirely, in favour of a decentralized, community and groupcentred system of regulation and decision-making (linking with the point on localism, above) – reminding us in the process that it is the individualistic basis of democratic theory that makes majoritarianism a central category, and that such a basis is in the end optional. And while Mills and King would bolster green constitutional constraints, Budge would institute more direct democracy but in a way that seeks to balance the impact of popular passion and reason. Where majoritarianism remains part of the picture of democratic theory, the main theme is the role that (structured or unstructured) deliberation may play in building majorities of a special sort – well-informed and focused on issues of common concern. Saward and Budge express concern about the potential elitism of deliberative emphases of certain types of discussion; whatever the merits of

216

Michael Saward

their arguments, we might conjecture that the apparent triumph of democracy over its great rivals has enabled democratic thinkers to ‘think the unthinkable’, concentrating on new forms of constraint on majoritarianism to the point that majoritarianism may no longer be regarded as a (if not the) core ingredient of any worthwhile democratic theory. It is evident too that taking other-than-majoritarian perspectives is being allied to a creative proceduralism that is moving beyond more traditional contrasts between procedural and substantive principles in democratic systems. Deliberative theorists are primarily proceduralists, though procedural rights are not underestimated; procedural innovations rather than substantive constraints were seen as central to inclusion of groups and concerns that stand in danger of marginalization in the chapters of Meier, Squires, Eckersley and Perczynski in a way that a few years ago might have been surprising. But this is, to borrow Beitz’s (1989) words, a ‘complex proceduralism’ which, as I have indicated, can by no means be taken as a return to strict majoritarianism; indeed, the opposite appears to be the case. The new proceduralism worries about inclusion and presence more than formal equality, voices and reasons more than votes and counting. Dissenting voices are plenty – represented here by Budge, Saward, perhaps Roßteutscher – but this trend may represent a steady, decisive sea-change in basic ways in which democracy is understood.

Sub-group relationships Traditionally in democratic theory sub-groups within a larger political unit or political community have been understood in primarily territorial terms. Cultural factors, such as the distinctiveness of a group in terms of language or dialect, religion or ethnicity, may have been taken into account but were regarded as parasitic upon occupation of a continuous ‘slice’ of territory. This way of viewing sub-groups – or of failing to view them because they did not fit the mould – meant that territorial solutions were the primary way of addressing issues of sub-group demands. Federalism, in a variety of strong or weak territorial forms, could be the solution, the sub-group-sensitive way of holding a political community (essentially a national one) together by granting a measure of separateness or autonomy. Today’s democratic innovation in this area appears to be the downgrading of the territorial dimension and the upgrading of issues such as ethnicity, gender, age and lifestyle. Distinct sub-groups are more often seen in terms of degrees of exclusion or the suffering of past injustices, than in territorial terms. To put the point another way, ‘difference’ is being deterritorialized as it is being problematized for democratic theorists, to the point where to talk (for example) of ‘cultural federalism’ is still to talk something of an outmoded language. At the same time, if democratic theory is tending to divide us up in new ways – not as territorial units or as electoral majorities and minorities, but as identity groups defined variously by self-conscious identification or experience of past and present injustice and exclusion – it worries about how it might at least minimally re-unite us in those areas that require common action and common consent across

Variation, innovation and democratic renewal

217

gulfs of difference and mutual misrecognition. For example, Hirst’s associationalism is decentralist and localist in a manner that leads him in the end to reinforce in certain ways the overarching, formal and egalitarian representative framework (see Perczynski, and Roßteutscher); deliberative democrats, from those happy to propose modest supplements to traditional liberal democratic practices (Fishkin and Luskin) to more radical ‘communicative democrats’ (see Squires’s discussion of the ideas of Young), seek ways in which important lines of communicative and informative commonality of concern might be forged across widely accepted gulfs of ignorance, group defensiveness and distrust. Through these debates, in general terms, we see a wide acceptance of quite radical pluralism in group identification, moral relativism of sorts, and the problematization of inclusion – well beyond the forms of interest-group pluralism of the 1950s and 1960s and its darker neopluralist successors.

The making of decisions Textbooks dealing with the idea of democracy will more often than not repeat the mantra that representative and direct democracy respectively form the two basic models on offer. The debates and innovations represented in these pages transcend the separation and opposition between these two models by (depending on your perspective!) ignoring or transcending the distinction. Budge would lead us to accept that the representative–direct distinction hides what is really a continuum, that more emphasis on the direct side is both feasible and desirable, and that deliberative democracy (of particular sorts at least) cannot reasonably substitute for more direct democracy. This deliberative–direct divide is reinforced perhaps by Fishkin and Luskin’s distrust of plebiscitary egalitarianism and proposal of new deliberative forums to shore up representative institutions (and the importance of representation as a democratic concept generally). Certainly Saward sees the deliberative model as veering away from the more direct forms that he too favours. Other innovations discussed here blur the issue in further, provocative ways. Is associative democracy direct or representative, for example? If the regulation of many aspects of people’s welfare locally is in the hands of genuinely local associations, is that not direct democracy despite the national representative framework? Are proposals and institutions for ‘inserting’ traditionally marginalized voices into the heart of decision-making bodies (Squires, Meier, Bang and Dyrberg) – and also the ‘voices’ even of those interests who literally have no voice of their own (Eckersley) – subversive or reinforcing of received wisdom about the structure of representative democracy? If ‘everyday makers’ really are able to shape the terms of their lives (including their political lives) as they see fit, and their experience is generalizable, perhaps formal mechanisms such as direct and representative ones perform together a marginal democratic role rather than the long-assumed central one? And what of Dryzek’s reclaimed category of ‘discursive democracy’, which on one reading at least is sufficiently anti-statist as to bypass the direct– representative divide as a significant issue in democratization. On this issue and on the evidence before us, few generalizations are possible but

218

Michael Saward

the permutations are many and fascinating. One could conjecture that the direct– representative distinction will indeed be, and come to be seen to be, less important in democratic theory and practice in developed countries and other contexts in the future. This in part reflects the fact that direct democracy requires indirect institutions for its realistic functioning (Budge 1996; Saward 1998), but also because the emphasis on talk, discussion and deliberation will continue to bear fruit in new deliberative forums that may increasingly take centre stage in deciding issues in democratic systems. For better or worse, real inclusiveness has come to mean more than direct or representative inclusiveness – it is also deliberative inclusion.

Accountability Accountability is not a term that has been used frequently in this book. That in itself may be a sign of shifts in democratic theory debates about what the real innovations are. Traditionally, democratic accountability concerns the ultimate constitutional responsibility of an office (or more precisely its occupant) for a range of activities or outcomes. There is some evidence in the attention of the authors here that there is a shift in theoretical concern from accountability to (a) the giving of reasons (b) on a more constant basis. The giving of reasons is critical to deliberative models, especially that of Gutmann and Thompson (1996), though its significance does vary in the very different formulations of (for example) Gutmann and Thompson, Fishkin and Dryzek. However they view it exactly, deliberative and discursive democrats do conceive of the democratic conversation/debate/argument as being more or less ongoing, however punctuated and halting it may be. In this way reason-giving, too, is constant, and a continuing obligation upon those whose role it is to act in the name of a broader public (such as elected officials in particular). If Eriksen is right, in supra-national settings it may be the case that even reason-giving pressures within small, elite groups amounts to a democratically satisfactory form of accountability – and perhaps a reasonable substitute for traditional understandings of that concept. On the other hand, there are reassertions of the importance of the familiar forms of accountability here too (from Budge). As we might expect in a field so contested as democratic theory, there are no claims without counter-claims.

Formal and informal, public and private At different levels of debate as evidenced in this book, the tyrannies of both structure and structurelessness are of concern to democratic theorists.1 Rättilä and Dryzek in their different ways show concern that state structures are suspect for fundamental democratic reasons, and that attempts by states to formalize/ colonize political action in the private sphere of civil society can represent frontal challenges to democracy. Others are in various ways happier to bring the dynamism of civil society to bear on complacent and insufficiently inclusive or responsive formal institutions, for example by creative new forms of representation for nature or for marginalized groups. Further, the partial formalization of

Variation, innovation and democratic renewal

219

local civil groups, such as religious organizations, local businesses and voluntary organizations, is envisaged in associationalist visions of future democracy. If there is a dominant thread – and I suggest this with some caution – then the importance to democracy of the spontaneous, the civil and the informal is being emphasized in innovative thinking about democracy – hardly surprising, perhaps, in a time when the end of the era of the state as we know it is being widely predicted.

A future for democratic theory? Conclusions to books by their nature grant authors an unusual freedom. In my comments above I have made various suggestions while simply bracketing important themes, such as the extent to which (and ways in which) real democratic systems may incorporate theoretical innovations; how theoretical questions about democratic institutions can be separated from, for example, more philosophical concerns in democratic theory; the extent to which the authors in this volume are variously critics, commentators or proponents (indeed originators) of democratic innovations they discuss; and even the differential relevance of most of the innovations discussed to the countries of the rich North rather than the developing South. Having confessed, however, let me attempt a summary. Democratic theory is in an innovative, dynamic phase – the apparent triumph of democracy has been accompanied by debate about the meaning, value and realization of democracy as fluid, pluralistic and intense as probably any other recent period of scholarly debate around political ideas. Continuing debate, in some cases social-scientific experiment and in others real-world embodiment or adoption, will test innovations prominent in this wave of theorizing. In the middle of this phase, we cannot see at all clearly its destination – the model of democracy, perhaps radically different from today’s variants, that may result. But we might conjecture at this point that many of its core components may be drawn from the following elements: •

• •

•

•

the counter-intuitive products of thinking in political ecology, which may make for example the political representation of non-human interests perfectly thinkable in time; the recognized importance of an open, fluid, free civil society not just as an adjunct of the state but as a realm of self-governance in its own right; the deterritorialization of democratic politics and the reassertion of particularistic identities based on cultural and experiential attachments rather than proximity in itself; the importance of talk, deliberation and reason-giving to shaping the activities of formal governmental structures, and to the meaning and practice of accountability, through both structured microcosmic forums and lessstructured civic forums; a renewed emphasis on the importance of power being exercised from the bottom up by more educated and self-aware citizens in a context of widelyaccepted value pluralism, through the use of direct democratic mechanisms

220

•

•

Michael Saward to supplement representative ones, and more generally the willingness and capacity of ordinary people to take control of their own lives and thereby take decisions out of the sphere of external governance structures; a decline in national political units and a concomitant reassertion of locality and region, giving rise to a context in which ‘shared sovereignty’ and equivalents become the norm rather than the exception; and a decline in the significance of, or emphasis given to, the democratic institutions of voting and majoritarianism, and a rise in the same of democratic talk and informed consensus (at least as an aspiration).

I do not claim that any of the present authors accepts or endorses this interpretative summary – indeed I do not myself endorse it in the sense of wanting it (all) to come about and believing it to be unambiguously more democratic than more conventional visions. I have merely sought to be faithful to what I have perceived in different arguments in the book. Be all this as it may, the meaning and status of deliberation, representation and association will continue to evolve – at times in tandem, at others in tension with each other. We hope our collective efforts may have added something positive in the larger effort to understand, and in modest ways to direct, this complex evolution.

Notes 1 On this point I have drawn upon helpful remarks by Robyn Eckersley which summarized a range of themes covered in the ECPR Workshop ‘Innovation in Democratic Theory’.

Bibliography

Abowitz, D. A. (1990) ‘Socio-political participation and the significance of social context’, Social Science Quarterly 71, 3: 543–66. Abramson, J. (1994) We, The Jury, New York: Basic Books. Ackerman, B. (1991) We the People I: Foundations, Cambridge, MA: Harvard University Press. Alexy, R. (1989) A Theory of Legal Argumentation, Oxford: Clarendon Press. Almond, G. A. and Verba, S. (1963) The Civic Culture, Princeton, NJ: Princeton University Press. Althaus, S. L. (1998) ‘Information effects in collective preferences’, American Political Science Review 92, 3: 545–58. Andersen, S. S. and Burns, T. (1996) ‘The European Union and the erosion of parliamentary democracy: a study of post-parliamentary governance’, in Andersen, S. S. and Eliassen, K. A. (eds) The European Union: How Democratic Is It?, London: Sage. Anderson, C. J. (1996) ‘Political action and social integration’, American Politics Quarterly 24, 1: 105–24. Andeweg, R. B. (1985) ‘De Burger in de Nederlandse Politiek’, in Andeweg, R. B., Hoogerwerf, A. and Thomassen, J. J. A. (eds) Politiek in Nederland, Samson, H. D: Tjeenk Willink. Apter, D. (1964) Ideology and Discontent, New York: Free Press. Arendt, H. (1958) The Human Condition, Chicago: University of Chicago Press. Arendt, H. (1961) ‘The crisis in culture’, in Between Past and Future, New York: Meridan. Arrow, K. J. (1951) Social Choice and Individual Values, New York: Wiley. Axelrod, R. (1984) The Evolution of Cooperation, New York: Basic Books. Bader, V. (1998a) ‘Associative democracy and the incorporation of ethnic and national minorities’, University of Amsterdam (unpublished paper). Bader, V. (1998b) ‘Problems and prospects of associative democracy, revisited’, University of Amsterdam (unpublished paper). Balibar, É. (1996) ‘Is European citizenship possible?’, Public Culture 8: 355–74. Bang, H. P. and Dyrberg, T. B. (1997) ‘Det lokale som slagmark og eksperimentarium for demokratisk styring’, in Gundelach, P. et al. (eds) Det lokale, Ålborg: Ålborg Universitetsforlag. Bang, H. P. and Sørensen, E. (1997) ‘Fra græsrødder til hverdagsmagere’, The Democracy Project, Democracy from Below, Copenhagen: University of Copenhagen. Bang, H. P. and Sørensen, E. (1998a) ‘The everyday maker: a new challenge to democratic governance’, COS-Rapport, No. 3, University of Copenhagen. Bang, H. P. and Sørensen, E. (1998b) ‘Hverdagsmageri på Indre Nørrebro – Institutionel udvikling og identitetsdannelse’, in Hansen, B. and Jensen, C. (eds) Statskundskab: Grundbogen i Statskundskab, Copenhagen: Akademisk Forlag.

222

Bibliography

Bang, H. P. and Sørensen, E. (1998c) ‘Hverdagsmagerne – en udfordring til demokratiet og til samfundsforskningen’, in Andersen, G. et al. (eds) Festschrift on the Occasion of the Danish Constitution’s 150th Birthday, Copenhagen: Reitzels Forlag. Barber, B. (1984) Strong Democracy, Berkeley: University of California Press. Barns, I. (1995) ‘Environment, democracy and community’, Environmental Politics 4, 4: 101–33. Barrett, M. and Phillips, A. (eds) (1992) Destabilizing Theory, Cambridge, UK: Polity Press. Barry, J. (1996) ‘Sustainability, political judgment and citizenship: connecting green politics and democracy’, in Doherty, B. and de Geus, M. (eds) Democracy and Green Political Thought, London: Routledge. Bartels, L. (1996) ‘Uninformed votes: information effects in presidential elections’, American Journal of Political Science 40, 1: 194–230. Beck, U. (1992) Risk Society, London: Sage. Beck, U. (1995) Ecological Enlightenment, New Jersey: Humanities Press. Beck, U. (1998) ‘Politics of risk society’, in Franklin, J. (ed.) The Politics of the Risk Society, Cambridge, UK: Polity Press. Beetham, D. (1992a) The Legitimation of Power, London and Atlantic Heights, NJ: Macmillan and Humanities Press International. Beetham, D. (1992b) ‘Liberal democracy and the limits of democratization’, Political Studies, special issue, 40: 40–53. Beetham, D. and Lord, C. (1998a) Legitimacy and the EU, London: Longman. Beetham, D. and Lord, C. (1998b) ‘Legitimacy and the European Union’, in Nentwich, M. and Weale, A. (eds) Political Theory and the European Union, London: Routledge. Beitz, C. (1989) Political Equality, Princeton, NJ: Princeton University Press. Bellamy, R. and Castiglione, D. (1997) ‘Building the Union: the nature of sovereignty in the political architecture of Europe’, unpublished paper, University of Reading/University of Exeter. Bellamy, R. and Castiglione, D. (1998) ‘Between cosmopolis and community: three models of rights and democracy within the European Union’, in Archibugi, D., Held, D. and Köhler, M. (eds) Re-imagining Political Community, Cambridge, UK: Polity Press. Benhabib, S. (1992a) Situating the Self, New York: Routledge. Benhabib, S. (1992b) ‘Models of public space: Hannah Arendt, the liberal tradition, and Jurgen Habermas’, in Calhoun, C. (ed.) Habermas and the Public Sphere, Cambridge, MA: MIT Press. Benhabib, S. (1994) ‘Deliberative rationality and models of democratic legitimacy’, Constellations 1, 1: 26–52. Benhabib, S. (1996a) ‘Toward a deliberative model of democratic legitimacy’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. Benhabib, S. (ed.) (1996b) Democracy and Difference, Princeton, NJ: Princeton University Press. Bennett, W. L. (1998) ‘The uncivic culture: communication, identity, and the rise of lifestyle politics’, Political Science and Politics 31, 4: 741–61. Berman, S. (1997) ‘Civil society and political institutionalization’, American Behavioural Scientist 40, 5: 562–74. Bessette, J. M. (1994) The Mild Voice of Reason, Chicago: University of Chicago Press. Beyers, J. and Dierickx, G. (1997) ‘Nationality and European negotiations: the working groups of the Council of Ministers’, European Journal of International Relations 3, 4: 435–71. Beyers, J. and Dierickx, G. (1998) ‘The working groups of the council of the European Union: supranational or intergovernmental negotiations?’, The Journal of Common Market Studies 36, 3: 289–317. Birch, A. (1971) Theories and Concepts of Modern Democracy, London: Routledge. Black, D. (1958) The Theory of Committees and Elections, Cambridge, UK: Cambridge University Press.

Bibliography 223 Blaug, R. (1996) ‘New theories of discursive democracy: a user’s guide’, Philosophy and Social Criticism 22, 1: 49–80. Bloom, W. (1990) Personal Identity, National Identity and International Relations, Cambridge, UK: Cambridge University Press. Bohman, J. F. (1990) ‘Communication, ideology, and democratic theory’, American Political Science Review 84: 93–109. Bohman, J. (1996) Public Deliberation, Cambridge, MA: MIT Press. Bohman, J. F. (1998) ‘The coming of age of deliberative democracy’, Journal of Political Philosophy 6, 4: 399–423. Bohman, J. F. and Rehg, W. (1997a) ‘Introduction’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA and London: MIT Press. Bohman, J. F. and Rehg, W. (eds) (1997b) Deliberative Democracy, Cambridge, MA and London: MIT Press. Boix, C. and Posner, D. (1996) ‘Making social capital work: a review of Robert Putnam’s Making Democracy Work: Civic Traditions in Modern Italy’, Harvard University Center for International Affairs Working Paper Series 96 (4). Bonner, R. J. (1967) Aspects of Athenian Democracy, New York: Russell and Russell. Borda, J. D. de (1784) ‘Memoire sur les elections au scrutin’ in Memoires de L’Academie Royale des Sciences Annee 1781, Paris: Academie des Sciences. Bousetta, H. and Swyngedouw, M. (1999) La Citoyenneté de l’Union Européenne et l’enjeu de Bruxelles, Bruxelles: CRISP. Brehm, J. and Rahn, W. (1997) ‘Individual-level evidence for the causes and consequences of social capital’, American Journal of Political Science 41, 3: 999–1023. Broere, M. (1998) ‘Tweedeling Ugandese samenleving wordt groter en groter’, Internationale Samenwerking 13, 6: 14–17. Budge, I. (1993) ‘Direct democracy: setting appropriate terms of debate’, in Held, D. (ed.) Prospects for Democracy, Cambridge, UK: Polity Press. Budge, I. (1996) The New Challenge of Direct Democracy, Cambridge, UK: Polity Press. Bühler, W., Kanitz, H. and Siewert, H. J. (1978) Lokale Freizeitvereine, St. Augustin: Institut für Kommunikationswissenschaften der Konrad-Adenauer-Stiftung. Burnheim, J. (1985) Is Democracy Possible?, Cambridge, UK: Polity Press. Burt, J. (1994) ‘John Rawls and the moral vocation of liberalism’, Raritan 14, 1: 133–53. Campbell, B. L. (1985) ‘Uncertainty as symbolic action in disputes among experts’, Social Studies of Science 15: 429–53. Campbell, D. T. and Stanley, J. C. (1963) Experimental and Quasi-Experimental Designs for Research, Chicago: Rand-McNally. Carton, A. (1995) ‘Paritaire democratie: een definitieve stap vooruit’, in Nederlandstalige Vrouwenraad (ed.) Jaarboek van de vrouw 1994, Leuven: Acco. Chambers, S. (1995) ‘Discourse and democratic practice’, in White, S. (ed.) The Cambridge Companion to Habermas, Cambridge, UK: Cambridge University Press. Chambers, S. (1996) Reasonable Democracy, Ithaca, NY: Cornell University Press. Charney, E. (1998) ‘Political liberalism, deliberative democracy, and the public sphere’, American Political Science Review 92: 97–110. Cohen, J. (1989) ‘Deliberation and democratic legitimacy’, in Hamlin, A. and Pettit, P. (eds) The Good Polity, Oxford: Basil Blackwell. Cohen, J. (1996) ‘Procedure and substance in deliberative democracy’, in Benhabib, S. (ed.) Democracy and Difference, Princeton NJ: Princeton University Press. Cohen, J. (1997a) ‘Deliberation and democratic legitimacy’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA: MIT Press. Cohen, J. (1997b) ‘Procedure and substance in deliberative democracy’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA: MIT Press.

224

Bibliography

Cohen, J. (1998) ‘Democracy and liberty’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Cohen, J. L. and Arato, A. (1992) Civil Society and Political Theory, Cambridge, MA: MIT Press. Cohen, J. and Rogers, J. (1992) ‘Secondary associations and democratic governance’, Politics and Society 20, 4: 393–472. Cohen, J. and Rogers, J. (1994) ‘Solidarity, democracy, association’, in Streeck, W. (ed.) Staat und Verbände, Opladen: Westdeutscher Verlag. Cohen, J. and Rogers, J. (1995a) ‘Secondary associations and democratic governance’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Cohen, J. and Rogers, J. (1995b) ‘Solidarity, democracy, association’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Coleman, J. S. (1988) ‘Social capital in the creation of human capital’, American Journal of Sociology 94: 95–120. Coleman, J. S. (1994) Foundations of Social Theory, Cambridge, MA: Harvard University Press. Converse, P. E. (1964) ‘The nature of belief systems in mass publics’, in Apter, D. (ed.) Ideology and Discontent, New York: Free Press. Conway, M. M. (1991) Political Participation in the United States, Washington, DC: Congressional Quarterly. Coote, A. and Lenaghan, J. (1997) Citizens’ Juries, London: IPPR. Coote, A. and Mattinson, D. (1997) Twelve Good Neighbours, London: Fabian Society. Crenson, M. A. (1983) Neighborhood Politics, Cambridge, MA: Harvard University Press. Crosby, N. (1995) ‘Citizen juries: one solution for difficult environmental questions’, in Renn, O., Webler, T. and Wiedemann, P. (eds) Fairness and Competence in Citizen Participation, Dordecht: Kluwer. Dagger, R. (1997) Civic Virtues, Oxford: Oxford University Press. Dahl, R. A. (1970) After the Revolution, New Haven: Yale University Press. Dahl, R. A. (1947) ‘Workers’ control of industry and the British Labour Party’, The American Political Science Review 41: 875–900. Dahl, R. A. (1956) A Preface to Democratic Theory, Chicago: University of Chicago Press. Dahl, R. A. (1989) Democracy and its Critics, New Haven: Yale University Press. Dahlerup, D. (1998) ‘Using quotas to increase women’s political representation’, in Karam, A. (ed.) Women in Parliament, Stockholm: IDEA. Davies, S. and Sang, B. (1997) ‘Perfect strangers’ (chapter of manuscript), London: King’s Fund. De Tocqueville, A. (1995) La democracia en América, Madrid: Alianza Editorial. Dechamps, A. (1993) ‘Volunteering in Germany’, in Smith, J. D. (ed.) Volunteering in Europe, London: The Volunteer Centre UK. Dekker, H. (1994) ‘Socialization and education of young people for democratic citizenship’, in Holly, R. (ed.) Political Consciousness and Civic Education during the Transformation of the System, Warszawa: Instytut Studiow Politycznych PAN. Dekker, P., Koopmans, R. and van den Broek A. (1997) ‘Voluntary associations, social movements and individual political behaviour in Western Europe’, in van Deth, J. W. (ed.) Social Participation, Voluntary Associations and Political Involvement in Representative Democracies, London and New York: Routledge. Delli Carpini, M. X. and Keeter, S. (1996) What Americans Know About Politics and Why It Matters, New Haven: Yale University Press. Dewey, J. (1927) The Public and Its Problems, London: George Allen and Unwin. Diamond, I. and Hartsock, N. (1998) ‘Beyond interests in politics: a comment on Virginia Sapiro’s “When are women’s interests interesting?” ’, in Phillips, A. (ed.) Feminism and Politics, Oxford: Oxford University Press.

Bibliography 225 Diani, M. (1997) ‘Social movements and social capital: a network perspective on movement outcomes’, Mobilization 2, 2: 129–47. Dienel, P. (1989) ‘Contributing to social decision methodology: citizen reports on technology projects’, in Vlek, C. and Cvetkovich, G. (eds) Social Decision Methodology for Technical Projects, Dordecht: Kluwer. Dienel, P. (1996) ‘Das “burgergutachten” und seine nebenwirkungen’, Forum fur Interdisziplinare Forschung 17: 113–35. Dienel, P. and Renn, O. (1995) ‘Planning cells: a gate to “Fractal” mediation’, in Renn, O., Webler, T. and Wiedemann, P. (eds) Fairness and Competence in Citizen Participation, Boston: Kluwer. Di Stefano, C. (1990) ‘Dilemmas of difference’, in Nicholson, L. (ed.) Feminism/Postmodernism, New York: Routledge. Dobson, A. (1993) ‘Critical theory and green politics’, in Dobson, A. and Lucardie, P. (eds) The Politics of Nature, London: Routledge. Dobson, A. (1996a) ‘Democratising green theory: preconditions and principles’, in Doherty, B. and de Geus, M. (eds) Democracy and Green Political Thought, London: Routledge. Dobson, A. (1996b) ‘Representative democracy and the environment’, in Lafferty, W. M. and Meadowcroft, J. (eds) Democracy and the Environment, Cheltenham: Edward Elgar. Doherty, B. and de Geus, M. (1996) Democracy and Green Political Thought, London: Routledge. Domènech, A. (1989) De la Ética a la Política, Barcelona: Crítica. Douglas, M. (1985) Risk Acceptability According to the Social Sciences, London, Routledge. Downs, A. (1956) An Economic Theory of Democracy, New York: Harper and Row. Dummett, M. (1984) Voting Procedures, Oxford: Clarendon Press. Dunn, J. (1992) Democracy, Oxford: Oxford University Press. Dryzek, J. S. (1987a) ‘Discursive designs: critical theory and political institutions’, American Journal of Political Science 31: 656–79. Dryzek, J. S. (1987b) Rational Ecology, Oxford: Basil Blackwell. Dryzek, J. S. (1990a) Discursive Democracy, Cambridge, UK: Cambridge University Press. Dryzek, J. S. (1990b) ‘Green reason: communicative ethics and the biosphere’, Environmental Ethics, 12: 195–210. Dryzek, J. S. (1992) ‘Ecology and discursive democracy: beyond liberal capitalism and the administrative state’, Capitalism, Nature, Socialism 3, 2: 18–42. Dryzek, J. S. (1995) ‘Political and ecological communication’, Environmental Politics 4, 4: 13–31. Dryzek, J. S. (1996) Democracy in Capitalist Times, New York: Oxford University Press. Dryzek, J. S. (1999) ‘Transnational Democracy,’ The Journal of Political Philosophy 7: 30–51. Dryzek, J. S. (2000) Deliberative Democracy and Beyond, Oxford: Oxford University Press. Dryzek, J. S. and List, C. (1999) ‘Social choice theory and deliberative democracy: a reconciliation’, paper presented at the Joint Sessions of the European Consortium for Political Research, Mannheim. Easton, D. (1953) The Political System, Chicago and London: University of Chicago Press. Easton, D. (1965a) A Framework for Political Analysis, Englewood Cliffs, NJ: Prentice-Hall. Easton, D. (1965b) A Systems Analysis of Political Life, New York: John Wiley and Sons. Eckersley, R. (1990) ‘Habermas and green political theory: two roads diverging’, Theory and Society 19: 739–76. Eckersley, R. (1996) ‘Greening liberal democracy: the rights discourse revisited’, in Doherty, B. and de Geus, M. (eds) Democracy and Green Political Thought, London: Routledge. Eckersley, R. (1999) ‘The discourse ethic and the problem of representing nature’, Environmental Politics 8: 24–49. Egeberg, M. (1999) ‘Transcending intergovernmentalism? Identity and role perceptions of national officials in the EU decision-making’, Journal of European Public Policy 6, 3: 456–74. Elshtain, J. B. (1995) Democracy on Trial, New York: Basic Books.

226

Bibliography

Elster, J. (1985) Making Sense of Marx, Cambridge, UK: Cambridge University Press. Elster, J. (1987) ‘The market and the forum: three varieties of political theory’, in Elster, J. and Hylland, A. (eds) Foundations of Social Choice Theory, Cambridge, UK: Cambridge University Press. Elster, J. (1988) ‘Introduction’, in Elster, J. and Slagstad, R. (eds) Constitutionalism and Democracy, Cambridge, UK: Cambridge University Press. Elster, J. (1992) ‘Arguing and bargaining in the federal convention and the assemblée constituante’, in Malnes, R. and Underdal, A. (eds) Rationality and Institutions, Oslo: Universitetsforlaget. Elster, J. (1995) ‘Forces and mechanisms in the constitution-making process’, Duke Law Review 45: 364–96. Elster, J. (1998a), ‘Introduction’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Elster, J. (1998b) ‘Deliberation and constitution making’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Elster, J. (ed.) (1998c) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Eriksen, E. O. (1999) Kommunikativ Ledelse, Bergen: Fagbokforlaget. Eriksen, E. O. and Fossum, J. E. (eds) (2000) Democracy in the European Union, London: Routledge. Eriksen, E. O. and Weigård, J. (1997) ‘Conceptualizing politics: strategic or communicative action’, Scandinavian Political Studies 20, 3: 219–41. Esquith, S. L. and Peterson, R. T. (1988) ‘The original position as social practice’, in Political Theory 16, 2: 300–34. Estlund, D. (1993) ‘Who’s afraid of deliberative democracy? On the strategic/deliberative dichotomy in recent constitutional jurisprudence’, Texas Law Review, 71: 1437–77. Estlund, D. (1997) ‘Beyond fairness and deliberation: the epistemic dimension of democratic authority’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA: MIT Press. Etzioni, A. (1993, 1995) The Spirit of Community, London: Fontana Press. Etzioni-Halevy, E. (1993) The Elite Connection, Cambridge, UK: Polity Press. Evans, S. M. and Boyte, H. C. (1992) Free Spaces, Chicago: University of Chicago Press. Fearon, J. D. (1998) ‘Deliberation as Discussion’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Ferguson, K. (1993) The Man Question, Berkeley: University of California Press. Filipowicz, S. (1997) Pochwala rozumu i cnoty, Krakow: Znak. Fischer, F. (1993) ‘Citizen participation and the democratization of policy expertise: from political theory to practical cases’, Policy Sciences 26: 165–87. Fishkin, J. S. (1991) Democracy and Deliberation, New Haven: Yale University Press. Fishkin, J. S. (1995) The Voice of the People, New Haven: Yale University Press. Fishkin, J. S. (1997) The Voice of the People (expanded edition), New Haven: Yale University Press. Fishkin, J. S. and Luskin, R. C. (1999) ‘Bringing deliberation to the democratic dialogue’, in McCombs, M. and Reynolds, A. (eds) The Poll with a Human Face: The National Issues Convention Experiment in Political Communication, Mahwah, NJ: Lawrence Erlbaum. Fleming, M. (1995) ‘Women and the “public use of reason” ’, in Meehan, J. (ed.) Feminists Read Habermas, New York: Routledge. Fleras, A. (1985) ‘From social control towards political self-determination? Maori seats and the politics of separate Maori representation in New Zealand’, Canadian Journal of Political Science 18, 3: 551–76 Foley, M. W. and Edwards, R. (1997) ‘Escape from politics? Social theory and the social capital debate’, American Behavioural Scientist 40, 5: 550–61.

Bibliography 227 Foucault, M. (1980) ‘Power and strategies’, in Foucault, M. (ed.) Power/Knowledge: Selected Interviews and Other Writings 1972–1977, New York: Pantheon Books. Foucault, M. (1981) The History of Sexuality: An Introduction, London: Penguin Books. Foucault, M. (1988) ‘The ethic of care for the self as a practice of freedom’, interview with Michel Foucault, in Bernauer, J. and Rasmussen, D. (eds) The Final Foucault, Cambridge, MA: MIT Press. Fraser, N. (1992) ‘Rethinking the public sphere: a contribution to the critique of actually existing democracy’, in Calhoun, C. (ed.) Habermas and the Public Sphere, Cambridge, MA: MIT Press. Fraser, N. (1995) ‘Recognition or redistribution? A critical reading of Iris Young’s Justice and the Politics of Difference’, Journal of Political Philosophy, 3, 2: 166–80. Fraser, N. (1997) Justice Interruptus: Critical Reflections on the ‘Postsocialist’ Condition, New York: Routledge. Gambetta, D. (1988) ‘Mafia: the price of distrust’, in Gambetta, D. (ed.), Trust, Oxford: Basil Blackwell. Gargarella, R. (1998) ‘Full representation, deliberation, and impartiality’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Gaus, G. F. (1997) ‘Reason, justification and consensus: why democracy can’t have it all’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA and London: MIT Press. Giddens, A. (1994) Beyond Left and Right, Cambridge, UK: Polity Press. Goodin, R. (1983) ‘Ethical principles for environmental protection’, in Elliot, R. and Gare, A. (eds) Environmental Philosophy, Milton Keynes: Open University Press. Goodin, R. (1992) Green Political Theory, Cambridge, UK: Polity Press. Goodin, R. (1996), ‘Enfranchising the earth, and its alternatives’, Political Studies, 44: 835–49. Goul Andersen, J. (1996) Membership and Participation in Voluntary Associations in Scandinavia in a Comparative Perspective, Aalborg University: Politics and Public Administration. Gould, C. (1996) ‘Diversity and democracy: representing differences’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. Green, M. C. and Brock T. C. (1998) ‘Trust, mood, and outcomes of friendship determine preferences for real versus ersatz social capital’, Political Psychology 19, 3: 527–44. Grundahl, J. (1995) ‘The Danish consensus conference model’, in Joss, S. and Durant, J. (eds) Public Participation in Science, London: Science Museum. Guadagnini, M. (1998) ‘The debate on women’s quotas in Italian electoral legislation’, Swiss Political Science Review 4, 1: 97–102. Gunnell, J. D. (1998) The Orders of Discourse, Lanham: Rowman and Littlefield Publishers. Gutmann, A. (1996) ‘Democracy, philosophy, and justification’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. Gutmann, A. and Thompson, D. (1996) Democracy and Disagreement, Cambridge, MA and London: The Belknap Press. Habermas, J. (1971) Toward a Rational Society, London: Heinemann Educational Books. Habermas, J. (1984) The Theory of Communicative Action I: Reason and the Rationalization of Society, Boston: Beacon Press. Habermas, J. (1987) The Theory of Communicative Action II: Lifeworld and System, Boston: Beacon Press. Habermas, J. (1989) Structural Transformation of the Public Sphere, Cambridge, MA: MIT Press. Habermas, J. (1990) Moral Consciousness and Communicative Action, Cambridge, UK: Polity Press. Habermas, J. (1992) ‘Further reflections on the public sphere’, in Calhoun, C. (ed.) Habermas and the Public Sphere, Cambridge, MA: MIT Press. Habermas, J. (1994) ‘Three normative models of democracy’, Constellations, 1: 1–10.

228

Bibliography

Habermas, J. (1995) Moral Consciousness and Communicative Action, Cambridge, MA: MIT Press. Habermas, J. (1996a) ‘Three normative models of democracy’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. Habermas, J. (1996b) Die Einbeziehung des Anderen. Studien zur Politischen Theorie, Frankfurt: Suhrkamp. Habermas, J. (1996c) Between Facts and Norms, Cambridge, UK: Polity Press. Hansen, M. H. (1991) The Athenian Democracy in the Age of Demosthenes, Oxford: Blackwell. Hamlin, A. and Pettit, P. (eds) (1989) The Good Polity, Oxford: Blackwell. Harrison, R. (1993) Democracy, London: Routledge. Hayward, B. M. (1995) ‘The greening of participatory democracy: reconsideration of themes’, Environmental Politics 4, 4: 215–37. Hayward, T. (1997) Political Theory and Ecological Values, Cambridge, UK: Polity Press. Hayward, T. and O’Neil, J. (eds) (1997) Justice, Property and the Environment, Aldershot, UK: Avebury. Heater, D. (1990) Citizenship, London: Longman. Held, D. (1993) ‘Democracy: from city-states to a cosmopolitan order?’, in Held, D. (ed.) Prospects for Democracy, Cambridge, UK: Polity Press. Held, D. (1995) Democracy and the Global Order, Cambridge, UK: Polity Press. Hirst, P. (1989) ‘Introduction’, in Hirst, P. (ed.) The Pluralist Theory of the State, London: Routledge. Hirst, P. (1992) ‘Comments on “Secondary Association and Democratic Governance” ’, Politics and Society 20, 4: 473–80. Hirst, P. (1993) ‘Associational democracy’, in Held, D. (ed.) Prospects for Democracy, Cambridge, UK: Polity Press. Hirst, P. (1994) Associative Democracy, Cambridge, UK and Amherst, MA: Polity Press and University of Massachusetts Press. Hirst, P. (1995) ‘Can secondary associations enhance democratic governance?’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Hirst, P. (1997) From Statism to Pluralism, London: UCL Press. Holden, B. (ed.) (2000) Global Democracy, London: Routledge. Honig, B. (1992) ‘Toward an agonistic feminism: Hannah Arendt and the politics of identity’, in Butler, J. and Scott, J. W. (eds) Feminists Theorize the Political, New York: Routledge. Hooghe, M. (1998) ‘Selectieve uitsluiting in het Belgisch politiek systeem. Innovatie en protest door nieuwe sociale bewegingen’, Res Publica 40, 1: 3–21. Immergut, E. M. (1995) ‘An institutional critique of associative democracy’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. IPU (1992) ‘Women and political power’, survey carried out among the 150 national parliaments existing as of 31 October 1991, Geneva: IPU. IPU (1997) ‘Democracy still in the making’, Men and Women in Politics: A World Comparative Study, Geneva: IPU. Janicke, M. (1996) ‘Democracy as a condition for environmental success: the importance of non-institutional factors’, in Lafferty, W. M. and Meadowcroft, J. (eds) Democracy and the Environment, Cheltenham: Edward Elgar. Jencks, C. (1990) ‘Varieties of altruism’, in Mansbridge, J. J. (ed.) Beyond Self-Interest, Chicago and London: University of Chicago Press. Joerges, C. and Everson, M. (2000) ‘Challenging the bureaucratic challenge’, in Eriksen, E. O. and Fossum, J. E. (eds) Democracy in the European Union, London: Routledge. Joerges, C. and Neyer, J. (1997) ‘From intergovernmental bargaining to deliberative political processes: the constitutionalisation of comitology’, European Law Journal 3, 3: 274–300. Johnson, J. (1993) ‘Is talk really cheap? Prompting conversation between critical theory and rational choice’, American Political Science Review 87: 74–86.

Bibliography 229 Johnson, J. (1998) ‘Arguing for deliberation: some skeptical considerations’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Jones, M. (1996) ‘Increasing women’s representation via gender quotas: the Argentine Ley de Cupos’, Women and Politics 16, 4: 75–98. Jones, M. (1998) ‘Gender quotas, electoral laws, and the election of women: lessons from the Argentine provinces’, Comparative Political Studies 31, 1: 3–21. Joss, S. and Durant, J. (eds) (1995) Public Participation in Science, London: Science Museum. Kaldor, M. (1996) ‘Nation-states, European institutions and citizenship’, in Einhorn, B., Kaldor, M. and Kavan, Z. (eds) Citizenship and Democratic Control in Contemporary Europe, Cheltenham: Edward Elgar. Katz, D. (1985) ‘Nationalismus als sozialpsychologisches problem’, in Winkler, H. A. (ed.) Nationalismus, Königstein/Ts.: Athenäum. Kerremans, B. (1996) ‘Do institutions make a difference? Non-institutionalism, neoinstitutionalism and the logic of common decision-making in the European Union’, Governance 2: 217–40. Klingemann, H. D., Hofferbert, R., Budge, I. et al. (1994) Parties, Policies and Democracy, Boulder, CO: Westview. Klüver, L. (1995) ‘Consensus conferences at the Danish Board of Technology’, in Joss, S. and Durant, J. (eds) Public Participation in Science, London: Science Museum. Kolankiewicz, G. (1996) ‘Social capital and social change’, The British Journal of Sociology 47, 3: 427–41. Kommission für Zukunftsfragen der Freistaaten Bayern und Sachsen (1997) ‘Erwerbstätigkeit und Arbeitslosigkeit in Deutschland’, Vol. III, Bonn. Kornhauser, W. (1959) The Politics of Mass Society, Glencoe, IL: The Free Press. Krebs, A. (1997) ‘Discourse ethics and nature’, Environmental Values 6: 269–79. Kulynych, J. (1997) ‘Performing politics: Foucault, Habermas and postmodern participation’, Polity 30: 315–46. Kymlicka, W. (1995) ‘Multicultural citizenship’, in A Liberal Theory of Minority Rights, Oxford: Clarendon Press. Kymlicka, W. and Norman, W. (1994) ‘Return of the citizen: a survey of recent citizenship theory’, Ethics 104: 352–81. Laffan, B. (1996) Constitution-Building in the European Union, Dublin: Institute for European Affairs. Lafferty, W. M. and Meadowcroft, J. (1996a) ‘Democracy and the environment: congruence and conflict – preliminary reflections’, in Lafferty, W. M. and Meadowcroft, J. (eds) Democracy and the Environment, Cheltenham: Edward Elgar. Lafferty, W. J. and Meadowcroft, J. (eds) (1996b) Democracy and the Environment, Cheltenham: Edward Elgar. Laver, M. J. and Budge, I. (eds) (1992) Party Policy and Government Coalitions, London: Macmillan. Lazarsfeld, P. F., Berelson, B. and Gaudet, H. (1948/1968) The People’s Choice, New York: Columbia University Press. Leiss, W. (1990) Under Technology’s Thumb, Montreal: McGill-Queen’s University Press. Levi, M. (1996) ‘Social and unsocial capital: a review essay of Robert Putnam’s Making Democracy Work’, Politics and Society 24, 1: 45–55. Lijphart, A. (1986) ‘Proportionality by non-PR methods: ethnic representation in Belgium, Cyprus, Lebanon, New Zealand, West Germany, and Zimbabwe’, in Grofman, B. and Lijphart, A. (eds) Electoral Laws and Their Political Consequences, New York: Agathon Press. Lipset, S. M. and Rokkan, S. (1967) Party Systems and Voter Alignments, New York: The Free Press. Lipset, S. M., Trow, M. and Coleman, J. S. (1956) Union Democracy, Glencoe, IL: The Free Press. Lodge, J. (ed.) (1989) The European Community and the Challenge of the Future, London: Pinter.

230

Bibliography

Lord, C. (1998) Democracy in the European Union, Sheffield: Sheffield Academic Press. Lowi, T. J. (1969) The End of Liberalism, New York: W. W. Norton. Luban, D. (1996) ‘The publicity principle’, in Goodin, R. (ed.) The Theory of Institutional Design, Cambridge, UK: Cambridge University Press. Lupia, A. (1994) ‘Shortcuts vs. encyclopedias: information and voting behaviour in California insurance reform elections’, American Political Science Review 88: 643–76. Lupia, A. and McCubbins, M. D. (1998) The Democratic Dilemma, Cambridge, UK: Cambridge University Press. Luskin, R. C. (2000) ‘From denial to extenuation (and finally beyond): political sophistication and citizen performance’, in Kuklinski, J. H. (ed.) Thinking about Political Psychology, New York: Cambridge University Press. Luskin, R. C. and Fishkin, J. S. (1998) ‘Deliberative polling, public opinion and democracy: the case of the National Issues Convention’, presented at the annual meeting of the American Association for Public Opinion Research, Saint Louis, MO, May 14–17. Luskin, R. C., Fishkin, J. S. and Jowell, R. (1996) ‘Considered opinions: deliberative polling in the UK’, presented at the International Society of Political Psychology Annual Meeting, Vancouver, Canada, 30 June–3 July . Luskin, R. C., Fishkin, J. S., Jowell, R. and Park, A. (1999) ‘Learning and voting in Britain: insights from the deliberative poll’, presented at the annual meeting of the American Political Science Association, Atlanta, Georgia, 2–5 September 1999. Luskin, R. C., Fishkin, J. S. and Plane, D. L. (1999) ‘Deliberative polling and policy outcomes: electric utility issues in Texas’, presented at the annual meeting of the Association for Public Policy Analysis and Management, Washington, DC, 4–7 November 1999. Luskin, R. C. and Globetti, S. (1997) ‘Candidate versus policy considerations in the voting decision: the role of political sophistication’, unpublished manuscript, Department of Government, University of Texas at Austin. Macedo, S. (ed.) (1999) Deliberative Politics, Oxford: Oxford University Press. McIver, S. (1997) ‘An evaluation of the King’s Fund citizens’, Juries Programme, Birmingham: Health Services Management Centre. Mackie, G. (1998) ‘All men are liars: is democracy meaningless?’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. McLean, I. (1987) Public Choice, Oxford: Blackwell. McLean, I. (1989) Democracy and New Technology, Cambridge, UK: Polity Press. MacPherson, C. B. (1977) The Life and Times of Liberal Democracy, Oxford: Oxford University Press. Majone, G. (1993) ‘The European Community between social policy and social regulation’, Journal of Common Market Studies 31, 2: 153–70. Majone, G. (1996) Regulating Europe, London: Routledge. Majone, G. (1998) ‘Europe’s “Democratic Deficit”: the question of standards.’ European Law Journal 4, 1: 5–28. Manglapus, R. S. (1987) Will of the People, New York and London: Greenwood Press. Manin, B. (1987) ‘On legitimacy and political deliberation’, Political Theory 15, 3: 338–68. Mansbridge, J. (1983) Beyond Adversarial Democracy, Chicago: University of Chicago Press. Mansbridge, J. (1995) ‘A deliberative perspective on neocorporatism’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Mansbridge, J. (1996) ‘Using power/fighting power: the polity’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. March, J. G. and Olsen, J. P. (1989) Rediscovering Institutions, New York: The Free Press. March, J. G. and Olsen, J. P. (1995) Democratic Governance, New York: The Free Press. Marcus, G. E. and Hanson, R. L. (eds) (1993) Reconsidering the Democratic Public, University Park, PA: University of Pennsylvania Press.

Bibliography 231 Mares, A. (1999) ‘Kiezen voor Brussel: de Brusselse Vlamingen en hun politieke vertegenwoordiging’, in Witte, E. and Mares, A. (eds) Twintig jaar onderzoek over Brussel, Brusselse Thema’s 6, Brussels: VUB Press. Marques-Pereira, Bérengère (1998) ‘La citoyenneté politique des femmes’, Courrier Hebdomadaire 1597, Bruxelles: Crisp. Marshall, T. H. (1950) Citizenship and Social Class, Cambridge, UK: Cambridge University Press. Mathews, F. (ed.) (1996) Ecology and Democracy, London: Frank Cass. Mathews, J. (1989) Age of Democracy, Melbourne: Oxford University Press. Matthews, D. E. (1973) US Senators and Their World, New York: Norton. May, J. D. (1978) ‘Defining democracy: a bid for coherence and consensus’, in Political Studies 26, 1: 1–14. Mayer, I., de Vries, J. and Geurts, J. (1995) ‘An evaluation of the effects of participation in a consensus conference’, in Joss, S. and Durant, J. (eds) Public Participation in Science, London: Science Museum. Mead, G. H. (1962) Mind, Self and Society from the Standpoint of a Social Behaviorist, Chicago: University of Chicago Press. Meehan, E. (1993) Citizenship and the European Community, London: Sage. Michelman, F. I. (1997) ‘How can the people ever make the law? A critique of deliberative democracy’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA: MIT Press. Michels R. (1915/1959) Political Parties, New York: Dover. Middlemas, K. (1995) Orchestrating Europe: the Informal Politics of the European Union, 1973–95, London: Harper Collins. Mill, J. S. (1975) ‘Considerations on representative government’, in Mill, J. S. (ed.) Three Essays, Oxford: Oxford University Press. Mill, J. S. (1991) Considerations on Representative Government, New York: Prometheus Books. Miller, D. (1993a) ‘Deliberative democracy and social choice’, in Held, D. (ed.) Prospects for Democracy, Cambridge, UK, and Stanford, CA: Polity Press and Stanford University Press. Miller, D. (1993b) ‘Three conceptions of citizenship’, paper presented to the Workshop on Citizenship and Plurality at the ECPR Joint Sessions, Leiden, 2–7 April 1993. Mills, M. (1996) ‘Green democracy: the search for an ethical solution’, in Doherty, B. and de Geus, M. (eds) Democracy and Green Political Thought, London: Routledge. Milofsky, C. (1988a) ‘Scarcity and community’, in Milofsky, C. (ed.) Community Organizations, New York and Oxford: Oxford University Press. Milofsky, C. (1988b) ‘Structure and process in community self-help organizations’, in Milofsky, C. (ed.) Community Organizations, New York and Oxford: Oxford University Press. Montin, S. (1998) Lokale demokratiexperiment – eksempel och analyser, Stockholm: Fritzes kundtjänst. Moravcsik, A. (1998) The Choice for Europe, Ithaca, NY: Cornell University Press. Mossuz-Lavau, J. (1998) Femmes–Hommes, pour la Parité, Paris: Presses de Sciences Po. Mouffe, C. (ed.) (1992) Dimensions of Radical Democracy, London: Verso. Muller, E. N. and Seligsen, M. A. (1994) ‘Civic culture and democracy’, American Political Science Review 88, 3: 635–52. Nentwich, M. and Weale, A. (1998) ‘Introduction’, in Nentwich, M. and Weale, A. (eds) Political Theory and the European Union, London: Routledge. Newton, K. (1976) Second City Politics, Oxford: Clarendon Press. Newton, K. (1997) ‘Social and political trust’, paper to the conference on Democratic Institutions: America in comparative perspective, Washington, DC, 25–27 August 1997. North, D. C. (1990) Institutions, Institutional Change and Economic Performance, Cambridge, UK: Cambridge University Press.

232

Bibliography

Nursey-Bray, P. (1983) ‘Consensus and community: the theory of African one-party democracy’, in Duncan. G. (ed.) Democratic Theory and Practice, Cambridge, UK: Cambridge University Press. O’Riordan, T. and Cameron, J. (eds) (1994) Interpreting the Precautionary Principle, London: Earthscan Publications. Offe, C. (1997) ‘Micro-aspects of democratic theory: what makes for the deliberative competence of citizens’, in Hadenius, A. (ed.) Democracy’s Victory and Crisis, Cambridge, UK: Cambridge University Press. Offe, C. (1998) ‘Demokratie und Wohlfarthstat: eine europeische Regimeform unter dem Stress der europeischen Integration’, in Streeck, W. (ed.) Internationale Wirtschaft und nationale Demokratie, Frankfurt am Main: Herausforderungen für die Demokratitheorie. Oldfield, A. (1990) Citizenship and Community, Routledge: London. Paehlke, R. (1980) ‘Democracy, bureaucracy, and environmentalism’, Environmental Ethics 10: 291–308. Paehlke, R. (1995) ‘Environmental values for a sustainable society: the democratic challenge’, in Dryzek, J. S. and Schlosberg, D. (eds) Debating the Earth: The Environmental Politics Reader, Oxford: Oxford University Press. Paehlke, R. (1996) ‘Environmental challenges to democratic practices’, in Lafferty, W. M. and Meadowcroft, J. (eds) Democracy and the Environment, Cheltenham: Edward Elgar. Pag, S. (1987) ‘The relation between the commission and national bureaucracies’, in Cassesse, S. (ed.) The European Administration, IISA. Parry, G., Moyser, G. and Day, N. (1992) Political Participation and Democracy in Britain, Cambridge, UK: Cambridge University Press. Pateman, C. (1970) Participation and Democratic Theory, Cambridge, UK: Cambridge University Press. Paye, O. (1997) ‘Féminiser le politique: recitoyennisation ou tribalisation? Une réponse du monde politique Belge’, Sextant 7, 1: 139–61. Pearce, D., Markandya, A. and Barbier, E. (1989) Blueprint for a Green Economy, Earthscan: London. Perczynski, P. (1995) ‘Political pluralism in English guild thought’, in Ryszka, F. (ed.) Historia, Idee, Polityka, Warszawa: Scholar. Perczynski, P. (1998) ‘Liberal influence on British socialist thought – the concept of combined freedom’, in Kujawinska-Courtney, K. and Machnikowski, R. M. (eds) Liberalism Yesterday and Today, Lodz: Omega-Praksis. Phillips, A. (1993) Democracy and Difference, Oxford: Polity Press. Phillips, A. (1995) The Politics of Presence, Oxford: Oxford University Press. Phillips, A. (1996) ‘Dealing with difference: a politics of ideas or a politics of presence?’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. Phillips, A. (ed.) (1998) Feminism and Politics, Oxford: Oxford University Press. Pierson P. and Leibfried, S. (eds) (1995) European Social Policy: Between Fragmentation and Integration, Washington, DC: Brookings Institution. Pitkin, H. F. (1984) ‘Fortune is a woman’, in Gender and Politics in the Thought of Niccolò Machiavelli, Berkeley and Los Angeles: University of California Press. Pitkin, H. F. (1967) The Concept of Representation, Berkeley: University of California Press. Plamenatz, J. (1963) Man and Society (two vols.), London and New York: Longman. Pocock, J. G. A. (1975) The Machiavellian Moment, Princeton: Princeton University Press. Poguntke, T. (1993) Alternative Politics, Edinburgh: Edinburgh University Press. Popkin, S. L. (1991) The Reasoning Voter, Chicago: University of Chicago Press. Portes, A. and Landolt, P. (1996) ‘The downside of social capital’, The American Prospect 94: 18–21.

Bibliography 233 Price, V. (1999), ‘Political information’, in Robinson, J. P., Shaver, P. R. and Wrightsman, L. S. (eds.) Measures of Political Attitudes, New York: Academic Press. Przeworski, A. (1998), ‘Deliberation and ideological domination’, in Elster, J. (ed.) Deliberative Democracy, Cambridge, UK: Cambridge University Press. Putnam, R. (1993) Making Democracy Work, Princeton, NJ: Princeton University Press. Putnam, R. D. (1995a) ‘Bowling alone’, Journal of Democracy 6, 1: 65–78. Putnam, R. D. (1995b) ‘Tuning in, tuning out’, Political Science and Politics 28, 4: 664–83. Putte, M. Van de (1997) ‘Politieke partijen en allochtonen’, in Hubeau, B. and Foblets, M. C. (eds) Politieke participatie van allochtonen, Leuven: Acco. Rahn, W. and Transue, J. E. (1998) ‘Social trust and value change: the decline of social capital in American youth’, Political Psychology 19, 3: 545–65. Rawls, J. (1972) A Theory of Justice, Oxford: Oxford University Press. Rawls, J. (1993) Political Liberalism, New York: Columbia University Press. Rawls, J. (1997) ‘The idea of public reason revisited’, The University of Chicago Law Review 64, 3: 765–807. Renn, O., Webler, T. and Wiedemann, P. (eds) (1995) Fairness and Competence in Citizen Participation, Dordecht: Kluwer. Rhodes, R. (1997) Understanding Governance, Buckingham, UK: Open University Press. Richards, C. J. (1994) The Founders and the Classics, Cambridge, MA: Harvard University Press. Riesenberg, P. (1992) Citizenship in the Western Tradition, Chapel Hill, NC: The University of North Carolina Press. Riker, W. H. (1982) Liberalism Against Populism, San Francisco: Freeman. Rose, N. (1996) ‘Governing advanced liberal democracies’, in Barry, A. et al. (eds) Foucault and Political Reason, London: UCL Press. Rosenblum, N. (1998) Membership and Morals, Princeton, NJ: Princeton University Press. Rousseau, J. J. (1990) Discurso Sobre el Orígen y los fundamentos de la Desigualdad entre los Hombres, Madrid: Tecnos. Rueschemeyer, D., Stephens, E. H. and Stephens, J. D. (1992) Capitalist Development and Democracy, Chicago: University of Chicago Press. Sabel, C. F. (1995), ‘Bootstrapping reform: rebuilding firms, the welfare state, and unions’, Politics & Society 23: 5–48. Sagoff, M. (1988) The Economy of the Earth, Cambridge, UK: Cambridge University Press. Sagoff, M. (1992) ‘Technological risk: a budget of distinctions’, in Cooper, D. E. and Palmer, J. A. (eds), The Environment in Question, London: Routledge. Said, E. (1989) ‘Representing the colonized: anthropology’s interlocutors’, Critical Inquiry 15: 205–25. Salamon, L. M. and Anheier, H. K. (1994) The Emerging Sector, Baltimore: Johns Hopkins University Institute for Policy Studies. Sanders, L. M. (1997) ‘Against deliberation’, Political Theory 25: 347–76. Sapiro, V. (1998) ‘When are interests interesting?’, in Phillips, A. (ed.) Feminism and Politics, Oxford: Oxford University Press. Sartori, G. (1987) The Theory of Democracy Revisited. Volume One, The Contemporary Debate, Chatham, NJ: Chatham House. Sasse, C. E. (et al.) (1977) Decision Making in the European Community, London: Praeger. Saward, M. (1993) ‘Green democracy?’, in Dobson, A. and Lucardie, P. (eds) The Politics of Nature, London: Routledge. Saward, M. (1996) ‘Must democrats be environmentalists?’, in Doherty, B. and de Geus, M. (eds) Democracy and Green Political Thought, London: Routledge. Saward, M. (1998a) The Terms of Democracy, Cambridge, UK: Polity Press. Saward, M. (1998b) ‘Green state/democratic state’, Contemporary Politics 4, 4: 345–56. Sawicki, J. (1991) Disciplining Foucault, New York: Routledge.

234

Bibliography

Scharpf, F. W. (1994) ‘Community and autonomy: multi-level policy-making in the European Union’, Journal of European Public Policy 1: 219–42. Scharpf, F. W. (1999) Governing in Europe, New York: Oxford University Press. Schlesinger, P. H. and Kevin, D. (2000) ‘Can the European Union become a sphere of publics?’, in Eriksen, E. O. and Fossum, J. E. (eds) Democracy in the European Union, London: Routledge. Schlosberg, D. (1999) Environmental Justice and the New Pluralism, Oxford: Oxford University Press. Schmitter, P. C. (1994) ‘Interests, associations and intermediation in a reformed post-liberal society’, in Streeck, W. (ed.) Staat und Verbände, Opladen: Westdeutscher Verlag. Schmitter, P. C. (1995) ‘The irony of modern democracy and the viability of efforts to reform its practice’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Schmitter, P. C. (1996) ‘Is it really possible to democratize the Euro-Polity?’, in Føllesdal, A. and Koslowski, P. (eds) Democracy and the European Union, Heidelberg: Springer. Schmitter, P. and Lehmbruch, G. (eds) (1979) Trends Towards Corporatist Intermediation, London and Beverly Hills: Sage. Schudson, M. (1997) ‘Why conversation is not the soul of democracy’, Critical Studies in Mass Communication 14: 297–309. Schumpeter J. A. (1947) Capitalism, Socialism and Democracy, New York: Harper. Schumpeter, J. A. (1976) Capitalism, Socialism and Democracy (fifth edition), London: Allen and Unwin. Schuppert, G. F. (1997) ‘Assoziative demokratie’, in Klein, A. and Schmalz-Bruns, R. (eds) Politische Beteiligung und Bürgerengagement in Deutschland, Bonn: Bundeszentrale für politische Bildung. Schwarz, N. (1996) ‘Ehrenamtliches engagement in Deutschland’, Wirtschaft und Statistik 4: 259–66. Sen, Amartya (1990) ‘Rational fools: a critique of the behavioural foundations of economic theory’, in Mansbridge, J. J. (ed.) Beyond Self-Interest, Chicago and London: University of Chicago Press. Shapiro, I. (1996) Democracy’s Place, New York: Cornell University Press. Skinner, Q. (1978) The Foundations of Modern Political Thought. Volume One: The Renaissance, Cambridge, UK: Cambridge University Press. Smith, D. H. (1977) ‘Voluntary action and voluntary groups’, in Inkeles, A. (ed.) Annual Review of Sociology 1: 247–70. Smith, G. (1999) ‘Pluralism, democratic deliberation and environmental values’, paper to the Deliberative Democracy and Green Politics Panel, PSA Conference, University of Keele, April 1999. Smith, G. and Wales, C. (1999) ‘The theory and practice of citizens’ juries’, Policy and Politics 27: 295–308. Smith, G. and Wales, C. (2000, forthcoming) ‘Citizens’ juries and deliberative democracy’, Political Studies. Stewart, J., Kendall, E. and Coote, A. (1994) Citizens’ Juries, London: Institute of Public Policy Research. Stolle, D. (1998) ‘Bowling together, bowling alone: the development of generalized trust in voluntary associations’, Political Psychology 19, 3: 497–525. Streeck, W. (1995) ‘Inclusion and secession: questions on the boundaries of associative democracy ‘, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Strong, T. B. (1994) Jean-Jacques Rousseau, London: Sage. Sundberg, J. (1995) ‘Women in Scandinavian party organizations’, in Karvonen, L. and Selle, P. (eds) Women in Nordic Politics, Aldershot: Dartmouth. Sunstein, C. R. (1988a) ‘Constitutions and democracies: an epilogue’, in Elster, J. and

Bibliography 235 Slagstad, R. (eds) Constitutionalism and Democracy, Cambridge, UK and Oslo: Cambridge University Press and Universitetsforlaget. Sunstein, C. R. (1988b) ‘Beyond the republican revival’, Yale Law Journal 97: 1539–90. Sunstein, C. (1993) ‘Democracy and shifting preferences’, in Copp, D., Hampton, J. and Roemer, J. E. (eds) The Idea of Democracy, Cambridge, UK: Cambridge University Press. Sunstein, C. (1997) ‘Deliberation, democracy and disagreement’, in Bontekoe, R. and Stepaniants, M. (eds) Justice and Democracy: Cross-Cultural Perspectives, Honolulu: University of Hawaii Press. Taylor C. (1989) ‘Cross-purposes: the liberal-communitarian debate’, in Rosenblum, N. (ed.) Liberalism and the Moral Life, Cambridge, MA: Harvard University Press. Taylor, C. (1991) ‘Die Beschwörung der Civil Society’, in Krzysztof, M. (ed.) Europa und die Civil Society, Stuttgart: Klett-Cotta. Thompson, D. (1999) ‘Democratic theory and global society’, The Journal of Political Philosophy 7: 111–25. Tilly, C. (1975) The Formation of National States in Western Europe, Princeton, NJ: Princeton University Press. Torgerson, D. (1997) ‘Policy professionalism and the voices of dissent: the case of environmentalism’ Polity 29: 345–74. Torgerson, D. (1999) The Promise of Green Politics: Environmentalism and the Public Sphere, Durham, NC: Duke University Press. Traxler, F. and Schmitter, P. C. (1995) ‘The emerging Euro-polity and organized interests’, European Journal of International Relations, 1, 2: 191–218. Traugott, M. C. (1992) ‘The deliberative opinion poll is a well-intentioned but flawed concept’, Public Perspective 3, 2: 27. Tyler, T. R. (1998) ‘Trust and democratic governance’, in Baithwaite, V. and Levi, M. (eds) Trust and Governance, New York: Sage. Uhr, J. (1998) Deliberative Democracy in Australia, Melbourne: Cambridge University Press. Uslaner, E. M. (1998) ‘Social capital, television and the “mean world”: trust, optimism, and civic participation’, Political Psychology 19, 3: 441–67. Uyttendaele, M. and Sohier, J. (1995) ‘Les quotas féminins en droit électoral ou les paradoxes de l’égalité’, Journal des Tribunaux 114: 249–56. Van Deth, J. W. (1992) ‘De politieke betekenis van Maatschappelijke participatie’, Acta Politica 27, 4: 425–44. Van Deth, J. W. (1996) ‘Voluntary associations and political participation’, in Gabriel, O. W. and Falter, J. W. (eds) Wahlen und politische Einstellungen in Westlichen Demokratien, Frankfurt am Main: Peter Lang. Van Deth, J. W. and Kreuter, F. (1998) ‘Membership of voluntary associations’, in van Deth, J. W. (ed.) Comparative Politics, London: Routledge. Van Gunsteren, H. (1998) A Theory of Citizenship, Boulder, CO: Westview Press. Van Molle, L. and Gubin, E. (1998) Vrouw en politiek in België, Tielt: Lannoo. Verba, S., Schlozman, K. L. and Brady, H. E. (1995) Voice and Equality, Cambridge, MA: Harvard University Press. Viroli, M. (1995) For Love of Country, Oxford: Oxford University Press. Walzer, M. (1981), ‘Philosophy and democracy’, Political Theory 9, 3: 379–99. Walzer, M. (1991) ‘Constitutional rights and the shape of civil society’, in Calvert, R. E. (ed.) The Constitution of the People, Lawrence, KS: University Press of Kansas. Walzer, M. (1994) Thick and Thin, Notre Dame, IN: University of Notre Dame Press. Warren, M. (1992) ‘Democratic theory and self-transformation’, American Political Science Review 86: 8–23. Warren, M. (1995) ‘The self in discursive democracy’, in White, S. (ed.) The Cambridge Companion to Habermas, Cambridge, UK: Cambridge University Press.

236

Bibliography

Warren, M. (1996) ‘What should we expect from more democracy?’, Political Theory 24: 241–70. Warren, M. (1998) ‘What should democrats expect of associations?’, paper presented at the Annual Meeting of the American Political Science Association, Boston, 3–6 September 1998. Weber, M. (1924) ‘Rede auf dem ersten Deutschen Soziologentag in Frankfurt 1910’, in Weber, M. (ed.) Gesammelte Aufsätze zur Soziologie und Sozialpolitik, Tübingen: J. C. B. Mohr. Weigård, J. (1995) ‘Representasjon og rasjonalitet. Deliberasjon og aggregering som politiske koordineringsprinsipper’, in Eriksen, E.O. (ed.) Deliberativ Politikk, Bergen: TANO. Weiler, J. H. H. (1991) ‘The transformation of Europe’, Yale Law Journal 100: 243–83. Weiler, J. H. H. (1995) ‘Does Europe need a constitution? Demos, telos and the German Maastricht decision’, European Law Journal 1, 3: 219–58. Weiler, J. H. H. (1999) The Constitution of Europe, Cambridge, UK: Cambridge University Press. Wessels, W. (1991) ‘The EC Council: the Community´s decision-making center’, in Keohane, R. O. and Hoffmann, S. (eds) The New European Community, Boulder, CO: Westview Press. Westlake, M. (1995) The Council of the European Union, London: Cartermill. Williams, M. S. (1998) Voice, Trust, and Memory: Marginalized Groups and the Failings of Liberal Representation, Princeton: Princeton University Press. Winter, L. De (1988) ‘Belgium: democracy or oligarchy?’, in Gallagher, M. and Marsh, M. (eds) Candidate Selection in Comparative Perspective, London: Sage. Wood, G. S. (1987) The Creation of the American Republic: 1776–1787, Williamsburg, VA: The University of North Carolina Press. Wright, E. O. (1995) ‘Introduction’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Wynne, B. (1996) ‘May the sheep safely graze?: a reflexive view of the expert–lay knowledge divide’, in Lash, S., Szerszynski, B. and Wynne, B. (eds) Risk, Environment and Modernity, London: Sage. Young, I. M. (1990) Justice and the Politics of Difference, Princeton, NJ: Princeton University Press. Young, I. M. (1995) ‘Social groups in associative democracy’, in Wright, E. O. (ed.) Associations and Democracy, London: Verso. Young, I. M. (1996) ‘Communication and the other: beyond deliberative democracy’, in Benhabib, S. (ed.) Democracy and Difference, Princeton, NJ: Princeton University Press. Young, I. M. (1997) ‘Difference as a resource for democratic communication’, in Bohman, J. F. and Rehg, W. (eds) Deliberative Democracy, Cambridge, MA: MIT Press. Young, I. M. (1998) ‘Polity and group difference: a critique of the ideal of universal citizenship’, in Phillips, A. (ed.) Feminism and Politics, Oxford: Oxford University Press. Zimmer, A. (1996) Vereine – Basiselemente der Demokratie, Opladen: Leske and Buderich.

Index

Abramson, J. 34 accountability 97–8, 218–19 Ackerman, B. 80 advocacy issues 94–5 agenda-setting 149, 198 ‘aggregative’ model of democracy 67–9, 122, 139–45 passim, 203–4 Almond, G. A. 179 ambit claim for ecological democracy 119–20, 124, 131 Amsterdam Treaty 60 animal species, proxy representation of 9, 122, 129–30, 138, 219 Arendt, Hannah 42, 45, 57, 121, 178 Argentina 109 argument, processes of 56 Aristotle 186 Arrow, K. J. 54 associations 10–11, 70, 73, 75, 104; active or passive involvement in 179; civil or political 193–4; definition of 182; factions within 197; functions of 164, 172–3; granted ‘semipublic’ status 165–6, 176; horizontal or vertical 193–4; impact on democracy 179–81; producing public or private goods 192–3 associative democracy 10–12, 35, 42, 162–70, 172–83, 199, 217–19; alternative designs for 163–6; and citizenship 166–8; criticisms of 175–6; definition of 163; empirical research on 178–80, 182; functions of 181–2 Athenian democracy 19, 197 Australia 23–4, 86 Bader, V. 169 Bang, H. P. 151 Barber, B. 146–8, 174, 178, 182, 206 bargaining 61–4 Barry, J. 123 Beck, Ulrich 117–18, 141–2 Beetham, D. 65, 68 Beitz, C. 216 Belgium 9, 107–15 Benhabib, Seyla 7, 41–50, 67, 71–2, 87

Bennett, W. L. 115 Bessette, J. M. 80 Bill of Rights (US) 80 biopiracy 88 Birch, A. 138 Blair, Tony 202 Bohman, J. 5–6, 66–7, 79, 82, 124–6 Borda counts 203–4 ‘bottom-up’ processes 9, 88, 144, 146–7, 157, 169, 219 Boyte, H. C. 174, 182 Bruni, Leonardo 185 Bühler, W. 182 bureaucracy 149 Burke, Edmund 8, 96 Burnheim, J. 38 Campbell, B. L. 142 Campbell, D. T. 21 Chambers, S. 38 citizens’ juries 19–21, 29–37, 59, 70, 82–3, 104, 126, 135, 168–9 citizenship: active conception of 32, 139, 161–3, 170; and associative democracy 166–8; as distinct from nationhood 56; scepticism about notion of 166; ‘secondary’ 176; universal 101; see also education, civic; European citizenship civic consciousness 162, 167 civic engagement 9–10, 146, 150–1, 192 civic republicanism 120, 184–7 civic virtues 11, 161–2, 179–80, 185–94 civil society 10, 41, 47–50, 70, 127, 149–50, 172–4, 218–19 co-decision procedure 60 cognitive dissonance 191 Cohen, Joshua 7, 11, 35, 67–75, 80, 84, 139–40, 164–9, 174–7, 182, 197, 200 Coleman, James 186 collective identity 58 collective-action dilemma 187–8, 192–3 comitology 63–4 Committee of Permanent Representatives (COREPER) 64

238

Index

Committee of the Regions 59 common concerns 150, 153–4, 157 ‘common good’ 73, 139, 156, 184–7, 190–1 common interests or values 57, 123; of European states 61–4 communicative action and interaction 48, 61–5, 81 communicative competence 119, 129, 140 communicative democracy 8, 128–9 communicative ethics 86, 102–3 communicative rationality 55, 60–2, 81, 140, 200 communicative styles 46–7 ‘communitarian unselfishness’ 187 communitarianism 148, 170, 172–4, 178, 182 ‘communities-of-fate’ 119–20, 127–8, 162, 166 community activities 180–1 Condorcet procedures 204–5 conflict, suppression of 37 conflict-prone neighbourhoods 180 consensus 37, 67, 84, 125, 139–41, 157, 195, 203, 220; as a feature of the EU 62; rational 62 consensus conferences 19–20, 30–7 passim, 59, 126 consent 68–9 consociationalism 113–14 constitutional states 56 constitutionalism 133–4, 137–8, 144–5 conversation, democratic 46–7, 218 co-operative equilibrium 187, 189 Copenhagen 150 corporatism 175–6; see also neo-corporatism corruption 189 cosmopolitan model of democracy 12, 128, 214 cost–benefit analysis 123 Crenson, M.A. 180 crime 26 cultural differences 142–3 Dahl, R. A. 77 decision-making 37–8, 201–3, 217–18; democratization of 142; as distinct from informal participation 50–1; environmental 118–19 deconstruction 97 deliberation 31–4; boundaries of 143–4; citizen participation in 206–8; civilizing effect of 44; and decision-making 38, 201–3; definition of 85–6; democratic value of 17–18; and direct democracy 197, 208; distinctiveness of 78–9; on environmental matters 133–41, 145, 215; forms of 205, 208; and the generation of civic virtue 192, 194; and group representation 100–5; incompatibility with vertical relationships 193; informal 50; as a means of getting political results 55; as only one part of the political process 68; and the politics of

presence 108; and risk-taking 142; role in democracy 104–5; tension with representation 34–6; and the transformation of preferences 190–1, 193–4, 202–3; see also ‘university seminar’ model deliberative capacity 80 deliberative democracy 5–12, 54, 56, 118, 163, 184, 206, 217; achieved by stages 18–19; aims of 44; and associationalism 168–70; and the concept of the public 40; criticisms of 66, 76, 83–5, 93, 102, 123–7, 209; and ecology 120, 122, 130–1; educational aspects of 122, 126; exclusive implications of 76; ideal type of 73–4; liberal constitutionalist form of 78–81; and long-range decisions 123; ‘otherregarding’ orientation of 122; problems with 133–4, 143–5; public acceptance of 43; as a regulatory ideal 124–5, 127, 131; typology of 71 deliberative (opinion) polls 19–27, 29–38 passim, 59, 70, 77, 82–3, 104, 126, 168–9, 200, 207; analysis of 24 ‘demarchy’ 38 demo-elites 152–5 democracy: definition of 198; elements of 4–5; -of-the-affected or for-the-affected 119; representative or participatory 162, 168 democratic deficits 7, 30, 54, 59, 63, 82, 118, 128, 167 democratic theory, themes of 213–14 demos 154 Denmark 19, 30, 33, 150 descriptive representation 106, 108 dialogue, constrained and unconstrained 121, 126 Diamond, Irene 95–6 difference: politics of 98–9; recognition and acceptance of 150, 153–4, 157, 216 difference democrats 81, 85–6, 128 direct democracy 4–5, 12, 163, 169–70, 180, 196–9, 208–9, 217–18 disagreement, expectation of 125–6, 155 discourse ethic 10, 101, 119–20, 125 discourse model of the public sphere 43–6 discourse theory 54–9, 64 discursive democracy 7, 78, 81–3, 86, 89, 135, 200, 215, 217 discursive rationality 41, 43, 48 displacement, strategies of 93–104, 129 diversity democrats 85; see also difference democrats Dobson, A. 136–8, 140 Douglas, M. 142–3 Downs, Anthony 18 Dryzek, J. S. 31, 33, 51, 72, 135–6, 140, 195–6, 200–1, 218 Easton, D. 155–6 Eckersley, R. 136

Index ecological costs, externalization of 122 ecological democracy 4, 9, 117–19, 122, 127, 130–1, 215, 219; ambit claim for 119–20, 124, 131; critique of 136–7 ecological values 133–45 ‘economy of moral disagreement’ 126 education, civic 77, 122, 126, 164–7, 173–4, 180 ‘egalitarian reciprocity’ 41, 43–5, 48 elections: equality achieved by 18–19; role in democracy 84, 220; as a source of legitimacy 67; see also voting elite influence 22, 41, 75, 84, 146–55, 173, 198, 207, 215 Elster, J. 36, 134, 168 empathy 191 empowerment 99, 146, 174, 199 endogenous view of democratic politics 155 ‘enlarged thinking’ 45, 121, 126, 128 Enlightenment thought 101, 199 environmental advocacy and monitoring 129–30 environmental impact assessments 129 environmental justice movement 117 environmentalism and democracy 136–7 equality, political 17–19 ‘equality of voice’ 135 Eriksen, E. O. 84 Esquith, S. L. 75 Estlund, D. 80 Etzioni, Amitai 170 Etzioni-Halevy, E. 151 European citizenship 161, 166–7 European Commission 59–60, 63, 167 European Convention on Human Rights 60 European Council of Ministers 6, 59–60, 167 European Court of Justice 53, 59–60 European Parliament 6, 53, 59–60, 167 European Union 4, 6, 12, 53, 57–65, 82, 167, 214 Evans, S. M. 174, 182 everyday makers 151–7, 217 experimentation in democracy 3 expertise vis-à-vis lay involvement 146–57 external solidarity see ‘other-regard’ externalization of costs 122; see also risk face-to-face dialogue 36, 203 factions 197 Fearon, J. D. 68 federalism 216 feminist theory 45, 47, 93–5, 98, 103–4 Fishkin, J. S. 72, 146–8, 200–1, 218 focus groups 23 Foucault, M. 9, 87 France 109 Fraser, N. 46, 95 freedom of expression 47, 80 French Revolution 177

239

‘full-information’ preferences 18–19, 22 future generations, proxy representation of 138 gender quotas 107–15 gendered identity 94 Germany 19, 29, 31, 33, 177 Giddens, A. 117–18 globalization 214 Goodin, Robert 122, 136 Gore, Al 23 governance as distinct from government 147–9, 157 Granada Television 200, 202 Green parties 113, 129–30, 208 greeting 102 group differentiation 98–9 group representation 8–9, 35, 93–105, 115, 129 guarantees of political representation 109–10, 114–15 guild socialism 163 Gutmann, A. 69, 80, 125, 218 Habermas, Jürgen 7, 10, 41, 43, 46–52, 57–8, 62, 81, 87, 101–2, 119–20, 122, 139–40, 200 Hartsock, Nancy 95–6 Hayward, Bronwyn 135, 140, 143 Hayward, Tim 143 Held, David 128, 214 heuristics 22 ‘high politics’ 51, 155 Hirst, Paul 11, 163–9, 174–7, 217 Hooghe, M. 114 ideas, politics of 107–8, 115, 128 identity, sense of 114–15; see also collective identity; gendered identity identity groups 99–100, 216 ideology 201 impartiality 100–4, 121 inclusion, strategy of 94–6 inclusiveness of democratic institutions 31, 41, 121, 162–3, 218 incompleteness problem 80 informal processes 48–51 instrumentalism 136–7 integration of communities 180–1 interaction, modes of 61–4 interest-group pluralism 99–103, 217 interest groups 163–5, 175–6, 182 intergovernmentalism 60 International Monetary Fund 130 Interparliamentary Union 109 Italy 111, 184 Janicke, M. 142 Jencks, Christopher 187 Joerges, C. 63–4 Johnson, J. 124 juridical democracy 82

240

Index

justice 100, 102, 156 justification of claims 55 Kaldor, M. 161 Kant, Immanuel 45, 119–20 knowledge: ‘useful’ or ‘encyclopaedic’ 207 Kulynych, J. 46–7, 51 language, representation based on 110, 112, 114 larger assemblies 36 lay involvement in decision-making see expertise Lebanon 109 left–right dimension in politics 206 legitimacy 33, 38, 44, 52, 66–72, 76, 84, 87; differential degrees of 69–70; of the EU 64–5; ‘output-oriented’ and ‘input-oriented’ 54 legitimation, processes of 55, 57–62 Lehrer, Jim 23 Leibfreid, S. 65 liberal constitutionalism 78–87, 195 liberal democracy 107, 120 Liberal Democrats 24 liberal feminism 94 liberalism 41–2, 46–7, 136, 147–8, 153, 199, 203 Lord, C. 65 lot, selection by 19 ‘low politics’ 155; see also everyday makers ‘low-information rationality’ 22 Lowi, T. J. 82 Lupia, A. 207 Machiavelli, Niccolò 185, 187, 189 McLean, I. 204–5 MacPherson, C. B. 12 Madison, James 59 Mafia, the 192 majority decisions 54, 67, 196, 204, 215–16, 220; constraints on 134; in the EU 62 Manin, B. 38, 66 Mansbridge, Jane 35–6, 70, 162 marginalized groups 98–100, 106–8, 217–18 Marques-Pereira, B. 112 Marshall, T. H. 161 ‘mass society’ 173 media attention to democratic innovation 21, 33, 36, 200–1, 207 ‘mediated’ approaches to democracy 9, 174–5, 199 meritocratic rule 22 Michels, R. 197 Mill, John Stuart 5, 22, 25, 69 Miller, D. 203–5 Milofsky, C. 178 minimalist model of democracy 75 modernization, political 142 Montesquieu, Baron de 187

National (UK) Health Service 26 National (US) Issues Convention 23, 25 national unity 97 neo-corporatism 199 neo-republican theory 162 Netherlands, the 30 network form of organization 88–9 networks of civic engagement 149–54, 192 neutrality, principle of (for political action) 42–3, 143 New Zealand 109 Newton, K. 179 Neyer, J. 63–4 norms 55, 123, 167 objective judgement 100–4 Offe, Claus 76 ontological issues 94–5 opportunism 192 ordinary, the, politics of 147–8, 151–6 passim ‘original position’ 72–5 ‘other-regard’ 162–3, 167–70 parliamentary assemblies, representativeness of 108–9 participation 41, 43, 135, 180–1, 206; compulsory 177; as distinct from formal decision-making 50–1; extension of 50–1; purpose of 47; social or political 179–80; ‘virtual’ 52 participatory democracy 79, 98, 150, 161–8 passim, 172, 174 passivity of citizens 178 Pateman, C. 12 Paye, O. 112–14 perfectionism and anti-perfectionism 153, 157 Pericles 197 Peterson, R. T. 75 Phillips, Anne 9, 34–5, 95–8, 102, 107–8, 128 Pierson, P. 65 ‘pillarization’ 177 Pitkin, Hannah 103, 106 Plamenatz, J. 201 planning cells 19–21, 29, 31 Plato 22, 186 plural voting 22, 69 pluralist theory 173–8; ‘old’ and ‘new’ 174–6 policy, influence on 87 political capital, building of 156–7 political community: concept of 155–6; definition of 214 political parties 11, 71–2, 75, 106, 176, 196–202, 206–8 political rights 80 politicization of administration 149–50 ‘politics of presence’ 9, 97–8, 107–8, 114–15, 128–9 popular sovereignty 57–8, 167 postmodernism 47, 94, 121, 124

Index power 42, 47, 124–7; legitimation of 68 precautionary principle 129–30, 134, 137, 144–5 preferences: co-operative 193; information on 164; manipulation of 191; ‘pre-political’ 76; subjective 140–3; transformation of 139–41, 144, 190–1, 193–4, 202–3; see also ‘fullinformation’ preferences Prisoner’s Dilemma 187–9 proceduralism 46, 216 professionalization 150–1 protective democracy 153 proxy representation 122, 129–30, 133, 137–8 Przeworksi, A. 67 public, the, concept of 40 public consultation 17 public goods 192–3 public interest 57, 59 public openness 36, 40, 48, 167, 170 public opinion 58; formation of 82, 84, 87, 196; ill-informed and well-informed 19; informal 48 public reason 45, 70, 85, 87, 153–6 public sphere 7, 70, 80–4, 101, 120, 151; autonomy of 41, 47; discourse model of 43; European 58–9; infrastructure of 48; multiple forms of 58; role of 51–2, 57, 86–7; as a social space 49–50 Public Utility Commission 23–5 Putnam, Robert 156, 177, 180–2, 184, 192 qualified majority rules 60 quotas 111; see also gender quotas radical feminism 94 random sampling 10, 19–21, 26, 30–1, 38, 106 ‘rational ignorance’ 18, 21, 147 rationality: of decision-making processes 44–7; of deliberative processes 50–2; ‘lowinformation’ form of 22; modern forms of 48; political 59; see also discursive rationality Rättilä, T. 87 Rawls, John 9, 42, 45, 70–5, 80, 84–5, 104, 147–8, 153–7 reason-giving 55, 218–19 reciprocity 192 referendums 56 regulatory functions 164 Rehg, W. 66–7 religion, freedom of 80 representation 5, 8–12, 59, 64, 106–7; based on language 110, 112, 114; diversity of 127–9; guarantees of 109–10, 114–15; in parliamentary assemblies 108–9; pre- and post-deliberation 20; in risk-taking decisions 118–19; tension with deliberation 34–6; of women 96–8; see also proxy representation representative democracy 162–5, 168–9, 196, 198, 209, 217

241

‘representative thinking’ 121 republicanism 41–2, 148, 161–2, 178, 189–90; see also civic republicanism res publica 57 reserved seats 109–10 rhetoric 102 Riker, W. H. 54 Rio Declaration 129 risk: assessment of 118, 127; aversion to 118, 122; ecological 133–5, 141–5; externalization and displacement of 117–18, 122, 128–9 Rogers, Joel 11, 35, 164–9, 174–7, 182 Rousseau, Jean-Jacques 174, 177, 191 Sagoff, M. 136, 143 Said, E. 125 Sanders, L. M. 52 Sapiro, Virginia 95–6 Saward, M. 85–6, 136, 141, 195–6 Scharpf, F. W. 57 Schlosberg, D. 88 Schmitter, Philippe 165, 167–8, 174–7 Schudson, M. 46 Schumpeter, J. A. 75 Schuppert, G. F. 178 scientization of politics 122 secession, rights of 196 secret ballots (in deliberative polling) 20 Sen, Amartya 187 sentencing policy 200, 202 shared sovereignty 220 small-group discussions 25–6, 31–2, 35–6 Smith, G. 135, 140 Social and Community Planning Research 23 social capital 11, 170, 172–4, 181, 184–94; association as a form of 192; definition of 186; trust as a form of 188–9 social choice theory 81, 84 social group differentiation 98–9, 102–3 social inequality 176 social learning 121–2, 126 social rankings 54 solar power 24 Sørensen, E. 151 South Africa 70 Stanley, J. C. 21 state structures 218 state-of-the-environment reporting 129 storytelling 102 strategic voting 204–5 sub-group relationships 216–17 subjective preferences 140–3 subjectivity 94–5 subsidiarity 60 Sunstein, Cass 84, 126 supra-nationalism 63–4, 214, 218 sustainable development 123 Sweden 151 Switzerland 18, 19

242

Index

Taylor, C. 94–5, 178 territorial divisions 216 Texas 23, 29 Thompson, D. 69, 80, 125, 218 de Tocqueville, Alexis 165, 172, 174, 187, 193 ‘top–down’ processes 169 town meetings 36 Traugott, M. C. 21 trust 181, 186–9, 193–4 Uganda 109 Uhr, J. 80 United States 17–19, 23, 29, 69, 109, 185; Supreme Court 80 ‘university seminar’ model of deliberation 197, 199–209 utilitarianism 136 Van Gunsteren, H. 162 veil of ignorance 157 Verba, S. 179 veto power 60, 99 ‘virtual’ participation 52 ‘virtual’ social trust 189, 194 voluntarism 173, 176–7 voluntary organizations 150

voting: manipulation of systems 204; one-dimensional 205–6; organization by political parties 199, 207–8; see also plural voting voting cycles 204–6 voucher systems 165, 176 Walzer, M. 75, 80 Warren, M. 31–2 Weber, Max 149, 178, 181 Weiler, Joseph 60 Weimar Republic 190 welfare states, dismantling of 183 Wilde, Oscar 207 will-formation 58 women, inclusion and representation of 94–8 women’s movements 46 ‘workable agreements’ 126, 195 World Bank 130 World Trade Organization 130 Young, Iris Marion 8, 34–5, 85–6, 93–105, 124, 127–8, 140 Zimmer, A. 174 zipper lists 111, 113