Towards a Deliberative and Associational Democracy (Research Methods for the Arts) 0748627391, 9780748627394, 9780748631483

Citation preview

Towards a Deliberative and Associational Democracy Stephen Elstub

Towards a Deliberative and Associational Democracy

Towards a Deliberative and Associational Democracy Stephen Elstub

Edinburgh University Press

For Lindsey

# Stephen Elstub, 2008 Edinburgh University Press Ltd 22 George Square, Edinburgh Typeset in Goudy Old Style by Iolaire Typesetting, Newtonmore, and printed and bound in Great Britain by Biddles Ltd, King’s Lynn, Norfolk A CIP record for this book is available from the British Library ISBN 978 0 7486 2739 4 (hardback) The right of Stephen Elstub to be identified as author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

Contents

Acknowledgements

vii

Introduction

1

1. Cultivating Autonomy: The Normative Core of Democracy

9

2. Deliberative Democracy and Autonomous Decision-Making

58

3. Institutionalising Deliberative Democracy through Secondary Associations

98

4. A Dualist Model of Deliberative and Associational Democracy

137

5. Democratising Secondary Associations

176

6. Avoiding the Mischief of Factionalism

213

Conclusion

233

Bibliography Index

237 257

Acknowledgements

I would first like to thank the various people who have kindly read previous drafts of this book, all of whom have provided invaluable advice which has enabled to the book to improve and its arguments to develop considerably. These include Matthew Festenstein, Anthony Arblaster, Michael Saward, Andrew Vincent and the Edinburgh University Press referees. Many thanks must go to all at Edinburgh University Press, especially my editor Nicola Ramsay, whose great flexibility and help made the submission process as stress-free as these things can be. I am grateful to the publishers for their permission to include extracts from the following journals: Stephen Elstub (2006), ‘Towards an Inclusive Social Policy in the UK: The Need For Democratic Deliberation in Voluntary and Community Associations’, Voluntas, 17 (1), 17–39. Stephen Elstub (2006), ‘A Double Edged Sword: The Increasing Diversity of Deliberative Democracy’, Contemporary Politics, 12 (3–4), 301–20. Stephen Elstub (2007), ‘Overcoming Complexity: Institutionalising Deliberative Democracy through Secondary Associations’, The Good Society, 16 (1) forthcoming. Stephen Elstub (2008), ‘Weber’s Dilemma and a Dualist Strategy for Deliberative and Associational Democracy’, Contemporary Political Theory, 7 (2) forthcoming May. I must also thank the reviewers of these journals for their excellent feedback and suggestions, many of which have also been incorporated into the book. Finally, I would like to thank all my friends, work colleagues and family for their support and encouragement – you know who you are! Two people deserve an extra-special acknowledgement: my mother Margaret, whose continued love and support has enabled me to reach a stage in life where I might have something interesting to write in a book; and my wife Lindsey, to whom this book is dedicated, for supporting me, and putting up with me, daily, throughout the writing of the book, and without whom the book would neither have been possible nor probable. — vii —

Introduction

This book is about changes that need to be made to improve modern democracy and specifically democratic decision-making. It sets out to answer the following questions: Why do we need democratic decisionmaking? What type of democratic decision-making do we need? What institutions do we need to make this type of decision-making feasible? Such a discussion is essential because it is thought that the structures of liberal democracy are facing a crisis of legitimacy (Habermas 1975; Hirst 1996; Fung and Wright 2001; Fung 2003a; Newman et al. 2004), and that this is reflected in the fact that citizens of liberal democracies are becoming increasingly disillusioned, dissatisfied and disenfranchised by the dominant political institutions and decision-making processes in these polities (Galbraith 1993; Stoker 2006). The general cure for these symptoms is thought to be a deepening of democracy. It is in this context – of a perceived crisis for liberal democracy and the reciprocal need for a deepening of democracy – that deliberative democracy has, over the last twenty years, become an increasingly dominant strand of democratic theory and practice, with experiments in deliberatively democratic public administration increasing throughout the UK, Europe, the USA, Australia, South East Asia and South America. Similarly, there has been an increasing focus on engaging civil society as a further solution to the problems of democracy, often termed ‘neo-tocquevillianism’ or the ‘second wave of civil society’, of which associational democracy is a dominant strand and one that seeks to transfer many functions of governance traditionally fulfilled by the state to democratic, and secondary, associations. As with deliberative democracy, this popularity of civil society is not restricted to political theory, for the world is undergoing ‘a major reappraisal of the whole state’ (Salamon and Anheier 1996: 1) and that this is being accompanied by ‘a global associational revolution’ (Salamon and Anheier 1996: 81). This book links deliberative democracy with associational democracy, arguing that when combined they complement and mutually support each other, with each —1—

Towards a Deliberative and Associational Democracy

helping to overcome the deficiencies of the other. Therefore, the argument presented here is that, in order to help solve the legitimacy problems faced by modern democracies, we should move towards a deliberative and associational democracy. In making this argument the book expands upon, and provides greater detail to, the already existing literature that connects the ideal of deliberative democracy with secondary associations (Dryzek 2000; Saward 2000; Warren 2001a; Fung and Wright 2001). To date the literature has only considered briefly some of the main arguments for and against institutionalising deliberative democracy in secondary associations, but this book discusses these, explores some of the problems in greater depth and considers some detailed, innovative and original solutions not considered elsewhere. Despite the attempt of the book to provide more detail regarding discussions on the institutionalisation of deliberative democracy in associations, than has been provided elsewhere, I acknowledge the importance of ensuring that a blueprint for institutionalisation is not put forward and hope that this has been avoided, leaving a great deal of scope for varying interpretations of the deliberative and associational model that can adapt to specific contexts. This is important because, although the recommendations for deepening democracy presented throughout the book are seen as relevant to all liberal democracies, there are vital differences and trends in culture and public administration across these countries to which the model must be able to adapt. However, the model is still thought to be relevant to all liberal democracies in so far as there remain some key similarities that transcend these differences, even if they vary in salience. There are three dimensions through which democracy can be deepened. The first is the expansion of the franchise, which means more people participate in collective decisions; the second is scope, which means more issues and topics are opened to democratic decision-making; and the third is ‘authenticity of control’, which requires ‘the effective participation of autonomous and competent actors’ (Dryzek 2000: 29). It is this third method of deepening democracy that the book focuses upon, and it is claimed here that deliberative democracy enhances authenticity. Authenticity is defined as ‘the degree to which democratic control is engaged through communication that encourages reflection upon preferences without coercion’ (Dryzek 2000: 8). At the heart of deliberative democracy is the idea that participants should communicate and exchange public reasons which should then lead them to reflect upon their preferences, and in turn make them more authentic. However, before we can start considering ways to improve contemporary democracy, we must first discuss why democracy is desirable, and note that there are many alternative justifications of it. My argument here is that personal autonomy is the normative core of —2—

Introduction

democracy and that this is the value that we should seek to enhance through democratic decision-making; the book is therefore structured around how best to cultivate autonomy and argues that deliberative democracy can best achieve this. One of the most significant criticisms levelled at deliberative democracy is that it is an irrelevant, utopian and counterfactual ideal because it is unachievable in modern, large and complex societies (Benhabib 1996: 84; Femia 1996; Warren 1996: 242; Miller 2000: 143). The book can therefore be seen as an attempt to address this problem by offering suggestions for normatively desirable, and yet practically plausible, institutional arrangements that can approximate deliberative democracy. Despite being predominantly normative, the book attempts to ensure that the normative vision can be ‘reconciled with the institutional requirements of modern society’ (Cohen and Arato 1992: 8). I therefore accept Rawls’s argument that normative theory must engage with ‘the art of the possible’, and so must be ‘feasible’ and ‘realistic’ (Rawls 1993: xv–xviii). In fact, my argument is that a deliberative and associational democracy fits more appropriately with these institutional requirements than do the current representative mechanisms of liberal democracy. In this sense, then, the book attempts to avoid the criticism of the public choice theorists that normative political theory lacks analytical reasoning and feasibility (Brennan and Lomasky 1993: 6). Therefore, the institutional discussion will be centred within the present institutional structures of liberal democracy and the capitalist economy, and will take into account both the supportive and debilitating conditions that exist therein. Even from this short discussion, the reader should now know the answers that I give to the questions with which I opened the book. We need democratic decision-making to promote autonomy. Deliberative democracy is the most suitable form of decision-making to achieve this. Secondary associations can make important contributions to ensure deliberative democracy is feasible. These are the answers, but in order to appreciate how I arrive at them, and to have an opportunity to be persuaded that they are right, you must read on, and so a more detailed overview of the structure of these arguments is provided below. Although there is already a vast literature on autonomy, and I use it as the basis of my arguments, this book maintains key elements of originality in so far as it looks specifically at cultivating autonomy in decision-making. Chapter 1 presents the case that the cultivation of the value of autonomy is the normative core of democracy. It is suggested that autonomy and democracy both aim to make their agents self-determining, therefore the relationship between them is implicit. Also, whilst democracy certainly —3—

Towards a Deliberative and Associational Democracy

promotes other important values, these are only contingently related to democracy, whereas autonomy can be cultivated only by democracy and the maximum value of other goods can only be achieved if autonomously chosen. Autonomy, then, is desirable because it is an intrinsic good for its own sake; it is constitutive of agency and people deeply value efficacy and having a causal impact on the world. Therefore, although democracy promotes many other values, it is because it promotes autonomy that these other values can be fully realised. However, the argument here is that autonomous citizens cannot be assumed, because the right conditions need to be in place to cultivate their autonomy. Therefore, the book also focuses on what features of decision-making must be present. It is argued that we should strive to reach ‘deliberative autonomy’ so that the outcomes of decisions are intended (Hurka 1993). For this to occur, free choice from a range of acceptable options based upon rationally formed preferences is required. In order to ensure that preferences are rationally formed, an adequate range of available information must be considered. Therefore it is a procedural view of autonomy that I favour, with the formation process, and not the content, of preferences determining autonomy. Chapter 1 then defends the possibility of autonomy, as defined, against several challenges that maintain the impossibility of justifying democracy upon autonomy. These are that people are socially determined, so cannot be autonomous; that autonomy inevitably leads to perfectionism, which excludes other important lifestyles, or to paternalism, which negates the possibility of autonomy; and finally that minorities cannot be autonomous in collective decisions. The chapter concludes by highlighting how collective decisions are different to individual decisions, as the former affect others. If individual decisions require private deliberation in order to be autonomous, then collective decisions require collective deliberation, including the sharing of information and reasons through public debate – which takes us on to the discussion of deliberative democracy in Chapter 2. Here, I argue that the model of deliberative democracy is the best decision-making method to cultivate the autonomy of all participants equally. A comparison of purely aggregative mechanisms of decisionmaking highlights the reasons why deliberative democracy is particularly suitable to cultivate participants’ autonomy. Deliberative democracy can increase autonomy through compromise, which leads to decisions that all accept, and which respects the agency of all. It further encourages the use of public reason as preferences must be justified to all; this will in turn encourage citizens to consider the opinions and interests of others and therefore induce people to see others as autonomous and equally important agents. Deliberative democracy cultivates both hearer and —4—

Introduction

speaker autonomy by increasing the availability of relevant information, allowing participants to express themselves freely and ensuring that the range of options to which they can vote reflects people’s rationally transformed preferences. Three challenges to these justifications for deliberative democracy are also considered: the social-choice critique that a popular will cannot be identified; a challenge from difference democracy that deliberative democracy is inevitably biased against historically disadvantaged groups; and finally that deliberative democracy requires special obligations among citizens that it cannot itself ground. Again, although Chapter 2 builds on well-established arguments, it is distinctive in the manner in which it explicitly justifies deliberative democracy via its potential to cultivate autonomy. Chapter 3 tries to expand upon, and provide greater detail of, the already existing literature that connects the ideal of deliberative democracy with secondary associations. It is suggested that it is features of social complexity, such as increased social pluralism, scale, inequality of deliberative and political skills and resources, the increasing reliance on specialists, and globalisation, that present significant barriers to deliberative democracy being meaningfully institutionalised, and also to the legitimacy and effectiveness of the state. The claim here is that, in an associational democracy, secondary associations would help overcome many of these barriers. Associations can aid in the cultivation of autonomy, assist in the institutionalisation of deliberative democracy and enable the state to be legitimate and effective. They do this by providing venues for decentralised governance, and by providing information and representation, which in turn creates informal public spheres that are based on public deliberation, are relatively voluntary and therefore provide choice over the source of representation and service delivery. They are also schools for democracy, where deliberative skills and trust are generated, in turn providing grounding for deliberative obligations. However, it is important to note that associations will not necessarily fulfil these functions. They require the right institutional framework, a high level of equality and internal democracy within each association. Furthermore, not all associations can fulfil all these functions. In fact, fulfilling one function can make it unlikely that the association can fulfil another. However, given the variability and flexibility of associations, a democratic associational ecology should mean that all functions could potentially be fulfilled. Johnson argues that the current work connecting associational democracy and deliberatively democratic decision-making is incomplete in two main ways. First, it lacks detail on the form and operation these associations would take in respect to state institutions. Second, there needs to be a —5—

Towards a Deliberative and Associational Democracy

clearer account of how deliberative procedures will operate in these associations and in ‘more formal political institutions’ (Johnson 1998: 175–6). Saward contends that further enlightenment on the issue of the appropriate siting of deliberation awaits the blending of aggregative, statist understandings of democracy and the insights of the deliberative ‘model’ within a larger realist theory of democracy that takes inevitable trade-offs and dilutions of democratic practice fully on board. (Saward 1998: 526) Chapters 4 and 5 attempt to respond to these demands, outlining the relationship of associations to the state, and providing detail of how deliberative forums and associations could operate. They will also highlight where trade-offs will need to be, and can be, made between the ideal of deliberative and associational democracy and practical exigencies, building upon the present literature that links deliberative democracy with associations, providing more detail about these aspects and outlining some innovative features unique to this deliberative and associational model. Chapter 4 specifically outlines how deliberative opinion-formation in civil society can be combined with democratic decision-making in legislative forums. It advocates a dualist strategy for associations, with the informal public sphere as the first strand. Here, the focus is on a wide variety of networks and their potential to mould public opinion and set the agenda through fostering deliberative communication between one association and another, and between all associations and the state. Mediating forums organised by quangos, with devolved powers, where representatives from associations would gather to make decisions based on the norms of deliberative democracy, provides the second strand. It accepts, however, that this is dependent upon the state devolving territorial and functional powers to these forums, and further highlights how such a system would fit in with, and yet fundamentally change, current liberal democratic institutions. The fifth chapter also reinforces claims made in Chapter 3 that in order to fulfil its potential democratic functions, associations in an associational democracy must themselves be internally democratic. The desirability of this is defended against two claims: that where costs of exit are low, the internal structure of the association is irrelevant; and that applying external standards to the internal working of secondary associations would undermine the essential liberal democratic right to freedom of association. The possibility of all associations achieving internal democracy is then defended against Michels’s ‘iron law of oligarchy’ (Michels, 1959), citizens’ reluctance to participate and the empirical restrictions of size, geographic dispersion —6—

Introduction

and time which makes associational democratisation challenging. Finally if, and how, the state and civil society could change associations to make them democratic through regulation is explored. Chapter 6 defends associational democracy against one of its most famous criticisms, Madison’s ‘mischief of factionalism’ (Madison 1966). In doing this, the deliberative and associational model of democracy is clearly distinguished from neo-pluralist and republican conceptions. The chapter suggests that a truly democratic conception of the common good can only be formed through deliberative democracy. However, it further distinguishes between the consensus and agonistic models of deliberative democracy. I argue in favour of the agonistic model because a common good cannot always be presumed; because difference is an essential resource, so should not always be eliminated; and because a focus upon the common good can allow dominant social groups to disguise their particular arguments. The book concludes by accepting that there are significant barriers to a transition to deliberative and associational democracy. Therefore the arguments in the book should be seen as a series of suggestions and proposals, rather than a prediction of how democracy might develop. However, normative ideas are still important and it is hoped that readers will find some of the suggestions for a deliberative and associational democracy compelling enough that they too might hope the changes called for here occur.

—7—

1

Cultivating Autonomy: The Normative Core of Democracy

Introduction It seems reasonable and logical to start a discussion on democracy with a discussion on why democracy is important. There are many well-established justifications of democracy, but I will present the case that a compelling justification of democratic decision-making is that it cultivates autonomy more equally than any other form of decision-making. The argument, then, is that autonomy is the normative core of democracy because they are intrinsically linked concepts, as they both aim to achieve equal selfdetermination; furthermore, autonomy has also been a central theoretical and practical aim of liberal democracy. This means that it is fair to judge liberal democracies by how well they cultivate their citizens’ autonomy. In order to completely establish autonomy as the normative core of democracy, it would be necessary to directly compare it with the alternative justifications, and demonstrate that it captures this normative core more accurately than these other justifications. However, that would entail a book in itself. Consequently, I will concentrate on putting forward, and defending, the autonomy justification of democracy. Furthermore, it will be asserted that other values, that are rightly, related to democracy, are only contingently related, where as autonomy can only be preserved by democracy. Connecting democracy and autonomy is not a new idea, but a democracy is a form of political organisation through which collective decisions are made, and as such, it is the form of decision-making that is my concern here, not the extent or scope of decisions (although this is undoubtedly a vital concern in relation to autonomy). I am concerned with the issue of how decisions should be made, in order to enhance the autonomy of the participants – an idea that has received less coverage than I believe it merits. —9—

Towards a Deliberative and Associational Democracy

The chapter starts by discussing why autonomy is an important value, arguing that it is an intrinsic value for its own sake due to its agency. Although there are other important values, those cannot be fully realised without the agency that autonomy brings. From this relationship between autonomy and agency it becomes apparent that there is an innate connection between autonomy and democracy, and autonomy and liberal democracy in particular, as ultimately they both aim to achieve selfdetermination. The next step is to define autonomy (and all its facets) and consider the implications for a decision-making process that aims at achieving equal autonomy for all. The conception of autonomy presented here is proceduralist and rationalistic, based upon revisions of Kant’s theory of autonomy. It is argued that the essential aspects of autonomy are free choice and rationality. Exactly what circumstances are required for free and rational decisions for individuals and collectives will be explored, but it is suggested that deliberation in decision-making is necessary. Following this, I consider four critiques that challenge the relationship between autonomy and democracy. The first is from communitarian critics who argue that autonomy is impossible because people are not atomistic, but embedded, therefore choice and rationality are impossible, and that means autonomy is a poor justification of democracy. The second critique argues that basing democracy on autonomy is perfectionist, as it excludes those who do not value autonomy. The third suggests that because rationality is essential to autonomy, and because some people are more rational than others, paternalism is therefore justified, which seems instinctively opposed to autonomy. The third, and final, critique argues that democracy cannot be justified on autonomy because democracy relies on majority rule and minorities cannot be autonomous. The solution to this final critique is key to my argument in this book. I suggest we must distinguish between individual and collective decisions and because our preferences in collective decisions affect others they must be formed and justified publicly, through collective deliberation. The Value of Autonomy If democracy is to be justified on the basis that it cultivates autonomy, then we must also know why autonomy is desirable. Moreover, we must also argue that it is more desirable than other values that democracy cultivates. It is argued here that autonomy is an intrinsic value because of its link to agency and its centrality to the full realisation of other values. However, it also has instrumental value in so far as autonomy increases the chances of getting what one wants; this is insufficient to justify democracy, however, — 10 —

Cultivating Autonomy: The Normative Core of Democracy

hence the appeal to its intrinsic value. Intrinsic values focus upon ‘agent evaluation’ and instrumental values upon ‘act evaluation’ (Lindley 1986: 21). It is thought that autonomy is an intrinsic good because we value our own preferences, choices, decisions and will (Hurka, 1987; Pennock 1989: 21). We also value the opportunity to act upon and pursue our own goals and form our own relations, as people want to be acknowledged as the kind of people who are capable of being self-determining. In this sense, ‘the autonomous person gives meaning to his life’ (Dworkin 1988: 31). Autonomy can be conceived as being intrinsic in two main ways. It can either be valuable ‘in itself’ or valued ‘for its own sake’. The former meaning concerns the source of value and would mean that autonomy would derive its value from itself and itself alone. The latter conception deals with the way in which it is valued and would mean autonomy is valuable for its own sake and ‘in virtue of its intrinsic properties’, not because it leads to other valuable consequences (Korsgaard 1996: 111). In this sense, autonomy is not valuable as an end but rather a condition we value (Wall 1998: 145). If autonomy was an intrinsic value ‘in itself’, then it would be valuable even in complete isolation (Moore, cited in Lindley 1986: 27–8).1 If we accepted this, however, it would mean that autonomy would be valued without an agent to value it, and this seems an untenable idea. Beardsley (1965: 11) thinks this problem can be avoided by appealing to a hypothetical and ideal observer, but even in this conception there is an agent, albeit a hypothetical agent, to value autonomy. The distinction between the two ways autonomy can have intrinsic value is important in order to distinguish the position held here from the libertarian perspective in which autonomy is valued ‘in itself’ and therefore can be experienced in isolation. The libertarian position must be rejected because autonomy requires structures: an agent needs to be autonomous in relation to something. This means that autonomy is a social value and must be exercised in a social setting. In this sense, a pre-social individual could not be autonomous; autonomy, then, requires society if it is to be achieved and valued (Clarke 1999: 43). This argument will be explored further when the communitarian critique of autonomy, as a justification of democracy, is considered later in the chapter. In order to prove that autonomy is an intrinsic good, it is necessary to demonstrate that it is better to have many good options from which to choose rather, than just the one option one would choose, regardless of the greater number of options. For Hurka, agency is central to what it means to be autonomous and provides this value: The ideal of agency is one of causal efficacy, of making a causal impact on the world and determining facts about it. And the autonomous agent, — 11 —

Towards a Deliberative and Associational Democracy

just in virtue of her autonomy, more fully realises this ideal. When she chooses among options she has two effects: realising some options and blocking others, and this gives her a larger efficacy than someone whose only effect is the first. (Hurka 1987: 366; see also Lindley 1986: 25 and Young 1986: 43, who invoke very similar arguments) However, just because autonomy is an intrinsic good, that does not mean that it is all that is required for ‘a good life’ (Wall 1998: 130). It is apparent that there are things of value other than autonomy (Hurka 1993), but, as Ackerman suggests, it is ‘not necessary for autonomy to be the only good thing; it suffices for it to be the best thing there is’ (Ackerman, cited in Sher 1997: 58). It is agency, an essential part of what it means to be autonomous, that provides an internal connection between autonomy and other values, that elevates autonomy as the most important value. This is essentially an existentialist argument and suggests that other values can only be realised if they are autonomously chosen (Sartre 1948: 32; Haworth 1986: 208; Hurka 1987: 378; Sher 1997: 58; Cohen 1998: 215–17 locates similar arguments in Locke 1955: 34 and Dworkin 1989). This is not to say that there is no good in having other values thrust upon us, only that having a causal impact releases their maximum benefit. Agents make choices based upon a potential value, but that ‘potential’ is only realised if autonomously chosen and it is therefore agency that makes autonomy an intrinsic value. Despite the quote above, Hurka (1993: 149) argues that the maximisation of all goods is not dependent on them being autonomously chosen. He argues that if Mozart had been forced into music, the music he produced outweighs this loss of autonomy. However, for Mozart the music and his life would presumably have been more enjoyable if he had chosen them.2 Moreover, his subordination is a sacrifice for the pleasure of others, so he is not being treated equally with equal respect to others. Furthermore, if other people were forced to listen to Mozart’s music throughout their life, then the value of it would be significantly less to them than if they had chosen to listen to it.3 Therefore I would suggest that the full value of goods is contingent upon them being autonomously chosen. As well as intrinsic value, autonomy has instrumental value. When one is autonomous and chooses which values and goals to pursue and how to pursue them, it is more likely that such a life will be satisfying and bring about contentment than if others (even benevolent others) make these choices and decisions for one. This is based on the idea that people are the best judge of their own interests (Dahl 1989). Consequently, people tend to resist and dislike external control, whether it is benign or not (Dworkin 1988: 111). Enabling people to participate in setting their own laws and have — 12 —

Cultivating Autonomy: The Normative Core of Democracy

control over their life allows people to avoid external control, and we can therefore see why autonomy can give value to democracy. J. S. Mill (1993) perceived autonomy as necessary for happiness, and similarly Raz suggests that ‘a person’s well-being depends to a large extent on success in socially defined and determined pursuits and activities’ (Raz 1986: 309). Although such instrumental claims about autonomy are most certainly true in most normal circumstances, there are situations in which people are manipulated to be happy, and so are happy despite being heteronomous, such as in Aldous Huxley’s Brave New World (1994).4 Nozick (1974: 42–5) makes a similar point with reference to ‘an experience machine’, which would create a virtual reality and provide people with any experience that they wanted. I think it is important, therefore, to establish autonomy as an intrinsic value. If autonomy is an intrinsic value, then we need to maximise it in the sense of maximising the number of autonomous people and maximising the cultivation of autonomy in each individual in so far as that is compatible with equal autonomy for all. This is due to all people being intrinsically equal (Dahl 1989: 86). There will often be conflicts between these two maximising principles but, due to innate equality, where there is conflict it should be the autonomy of the most people that should be maximised, rather than the autonomy of certain individuals. Wall criticises this argument by claiming that not all people will want to maximise the development of their autonomy because autonomy is not the only thing with value (Wall 1998: 184–5). It is accepted, then, that there are other important values, but autonomy is compatible with these too. In fact, it has been argued that the potential value of other things can only be fully realised if they are autonomously chosen. The maximisation of autonomy will therefore potentially achieve a greater realisation of other values. Whether or not autonomy is an important value in societies that do not value it is another matter and raises a question as to the universability of the arguments here. However, the argument here is that autonomy is certainly valued in Western liberal democratic societies, and it is to this relationship that we now turn. Autonomy and Liberal Democracy The concept of autonomy has a long political lineage. In ancient times it manifested itself in ‘autarchy’, which is the sovereignty of the city-state. In the medieval period it was used to distinguish between the ‘sovereignty’ of the church and state. However, it is in modern times, at least since the emergence of nineteenth-century imperialism, and later with the emergence of liberal democracy, that it has been applied to individuals (Young 1986: 6; — 13 —

Towards a Deliberative and Associational Democracy

Lakoff 1996: 99; Wall 1998: 131; Clarke 1999: 2). Autonomy remains a vastly contested concept. Originally ‘autos’ meant self and ‘nomos’ meant rule (Young 1986: 5; Lakoff 1996: 37). In this simple sense, autonomy means ‘to be self-governing’. When applied to the individual person, the essential idea behind the principle of autonomy is that individuals should, to a significant extent, be the author of, and therefore in control of, their own life (Raz 1986: 369; Clarke 1999: 210). If this is combined with a principle of intrinsic equality – that is, the assertion that all people have equal and independent moral value – then it seems that every individual’s autonomy should be respected equally (Dahl 1989: 86). It is then the principle of intrinsic equality that links autonomy and democracy, with both implicitly aiming for the same thing: the promotion of equal self-determination. With democracy essentially meaning ‘rule by the people’, it is apparent that it aims to ensure people can collectively control society and therefore the cultivation of their autonomy (Graham 1986: 3; Dahl 1989: Chapter 7; Warren 2001a: 60–9; Habermas 1996a; Lakoff 1996: 64). Central to the very idea of democracy is the principle that members of an association should be able to determine the conditions of that association (Held 1995: 145–6) and therefore ‘the power and appeal of democracy comes from the idea of autonomy, of choosing freely for oneself’ (Dunn, cited in Lakoff 1996: 33). Indeed, ‘we cannot make sense of the idea of democracy without appealing to both the value of intrinsic equality and the value of personal autonomy’ (O’Flynn 2006: 9). The appeal of democracy is therefore its link with autonomy, and democracy is the only political arrangement that gains its power and legitimacy from the rule of the people. To be autonomous, if you are affected by a decision, you should have the opportunity to contribute to the making of that decision: ‘Obviously, one cannot be autonomous if one’s decisions make no difference to what happens in one’s life’ (Waldron 1989: 1115). At this normative level, democracy cultivates autonomy by striving for all to have an equal opportunity to participate in the decision-making process, and therefore equal power to determine the laws of society, respecting everyone’s autonomy equally. Starting from the premise that if law is to be obeyed due to moral responsibility and not coercion then it should be viewed as legitimate by those it is to coerce, Held considers autonomy ‘the principle of legitimacy’ because it ‘seeks to articulate the basis on which public power can be justified’ (Held 1995: 153; see also Habermas 1996a: 29). In this sense, political power is legitimate to the extent that it derives from the collective autonomous choices of those it affects, or enhances citizen autonomy. For laws to be derived from the people, three criteria must be met. First, people — 14 —

Cultivating Autonomy: The Normative Core of Democracy

should participate in the process to determine laws. Second, the collective power of the people should be limited to guarantee the freedom and autonomy of all. Third, autonomy and democracy are compatible if, and only if, the autonomy of each is viewed equally (Held 1995: 147; see also Dodson 1997: 100; Warren 2001a: 62). Hence, both Bobbio and Wolff suggest that the preservation of autonomy is the primary justification of democracy (Bobbio, cited in Post 1993: 659; Wolff 1976: 21). I suggest that, more than this, autonomy is the normative core of democracy. For while democracy may cultivate other values too, it is only because it cultivates autonomy that these other values can be fully realised. As O’Flynn (2006: 43) realises, democracy may well bring other goods like ‘individual liberty, economic efficiency, and peace and stability’, but such goods are ‘only contingently related to democracy’, whereas autonomy can be secured only by democracy and democracy alone. As well as being connected to democracy in general, the concept of autonomy is central to liberal thought and to its conceptions of democracy, freedom and the individual self: ‘Autonomy is a value that has always been central to liberalism’ (Lukes 1973: 56). In normative liberal theory the liberal democratic system is often justified as the only political system that ensures autonomy. Autonomy is also embedded in Western culture and institutions; for example, the desire for autonomy has been a motivation for social movements to strive for change and liberal democracy requires its citizens to be autonomous. These relationships will be considered in turn. Galston (1995) argues that there are principally two dominant strands of liberal thought. One aims at the promotion of individual autonomy, the other is tolerance of diversity. Robert Young takes a stronger line and argues that any adequate defence of liberal democracy requires a conception of autonomy (Young 1986: 3; see also Fromm, 1955: 185 and Eckersley 1996: 222). Kymlicka similarly believes the cultivation of autonomy is the central liberal value and the normative core of liberal democracy because aspects of this form of political system, such as civil and personal rights, freedom of conscience, association, speech and constitutional government, are necessary to ensure autonomy (Kymlicka 1995: 85). Indeed liberal democracy has a tradition of being justified on providing the appropriate circumstances for individuals to select their own actions and goals. This aim has orientated itself through two processes: choosing between political alternatives, and the limitation of scope and degree of political power. We can see this in the work of Locke, who proposed that individuals could and should determine and justify their own actions, and enter into self-determined relations. He also advocated the necessity of autonomous spheres of action, which include religious, social, political and economic affairs, to ensure individual — 15 —

Towards a Deliberative and Associational Democracy

autonomy and freedom. In all these spheres, people must be free from arbitrary power in order to choose and form opinions, and regulate and coordinate their own actions. For Locke the source of political obligation lies in individual consent, express or tacit. Political authority, therefore, is to be distinguished from paternalism by being derived from the people. This is to be achieved through the rule of law, to prevent the arbitrary use of power (Locke 1988). Another classic thinker from the liberal tradition, J. S. Mill (1993), also thought it essential to base legitimate state authority upon a concept of individual autonomy (Young 1986: 44). Mill argued that individuals should be ‘sovereign’ and free from state interference when choosing and pursuing their own ends, providing they did not harm others and limit their autonomy. The individual can legitimately limit the sovereignty of the state providing it goes against their own will, for example, conscientious objectors. In turn the state must have good reason to interfere with individuals, and for Mill interference could only be permitted to prevent the individual limiting another’s autonomy. Barry accepts that it is ‘common-place’ to assume Mill thought it the role of the state to promote autonomy, but he argues that such an interpretation of Mill’s arguments ‘is clearly a travesty of Mill’s position’ (Barry 2001: 120). This is not to say that for Barry liberalism cannot be justified by reference to the value of autonomy, as he agrees that ‘autonomy-promoting liberalism’ is a ‘bona fide form of liberalism’, but that it is not the role of the state to ensure it (Barry 2001: 123). However, for Barry, liberal values and institutions can be defended upon concepts other than autonomy, and his favoured conception is equal freedom (Barry 2001: 121–2). As well as suggesting that the institutions of liberal democracy are best justified by appeal to autonomy, it is also argued that the value of autonomy is embedded within these institutions. Rawls considers autonomy a ‘political’ concept and therefore distinguishes it from the more ‘comprehensive’ and ‘metaphysical’ ideal of autonomy formulated by Mill and Kant (Rawls 1993: 10). According to Rawls, if autonomy is drawn from ‘intuitive ideas’ that are embedded in society in an ‘overlapping consensus’, which involves a range of ideas from alternative philosophical, moral and religious doctrines, it is then a political concept and latent in political culture (Rawls 1993: 14). Held argues that autonomy is indeed embedded in the public political culture of Western liberal democratic societies: ‘ ‘‘Embedded’’ . . . connotes that the principle has developed as part of, and has been constructed upon, the conceptual and institutional resources of Western democratic culture . . . in a manner that could, in principle, be understood and fully acknowledged by all citizens’ (Held — 16 —

Cultivating Autonomy: The Normative Core of Democracy

1995: 148–9). To provide evidence of the embeddedness of autonomy in Western liberal democracies, Held argues that the pursuit for equal autonomy in society has been a principal motor to change political culture. For example, feminist, race and social-class movements have won rights in liberal democracies – such as freedom of speech, expression, belief and association; freedom for women in and beyond marriage; free and equal voting in elections and universal suffrage – that contribute to the cultivation of autonomy for citizens in modern democracies by limiting the power of the state and providing citizens with equal powers of participation in the decision-making process (Held 1989: 191–2; Held 1995: 149–50; see also Cohen and Arato 1992: 554–5). In addition to arguing that state authority is morally justified if it is based on the autonomous consent of individuals, Raz argues the state has a moral duty to create the necessary conditions for autonomy (Raz 1986: 425), as autonomy will not be cultivated ‘naturally’ because the goals available for individuals to pursue are dependent upon social forms, such as ‘forms of behaviour’ and institutions that are widespread within society (Raz 1986: 308). However, Raz further suggests that in modern Western liberal societies the social forms derived from social, economic and political institutions require individuals to be autonomous. Therefore, for those who have been socialised and raised in Western liberal democracies, the value of autonomy is undeniable, and they can abandon the value only by incurring social and personal costs (Raz 1986: 394). Wall (1998: 166–9) provides further detail to Raz’s claims about modern Western societies, outlining several features of Western societies that distinguish them from others and make autonomy an essential value in them. The first factors are ‘geographic, familial and social mobility’ and ‘technological and economic innovation’. The high levels of mobility in Western societies mean that citizens must adjust to continuous economic and technological changes and therefore form and revise their aims and make regular decisions about how their nature and talents can be best employed. The second factor is ‘pluralism and secularisation’, because an extensive diversity of worldviews and religions encourages citizens to form ideas and beliefs of their own, and to choose between these competing views. The existence of this type of citizen is incompatible with authoritatively enforced religions that act as guides for action (Berger 1991: 138–9). Third, the commitment to ‘human rights’ in Western societies is based upon the idea of the individual and provides protections for the individual to be self-governing in many areas of life. In his discussion of Raz, Parekh rightly recognises that in Western societies not all people do value autonomy, but this is not Raz’s point: — 17 —

Towards a Deliberative and Associational Democracy

he only claims that the social forms of Western liberal democracies require it. A more pertinent point made by Parekh is that just because people in Western societies do value autonomy, this does not mean that they should continue to do so if it is possible for them to reject it (Parekh 2000: 92–3). However, we have already provided other reasons why they should continue to value autonomy and why it is better than a heteronomous life, suggesting autonomy is both intrinsically and instrumentally valuable. Parekh also rejects Raz’s argument that autonomy is a functional requirement for Western societies, arguing that there are many Asian immigrants in Britain who do not value autonomy but have achieved significant material success precisely because they do not value autonomy. Instead they ‘draw on the ample resources of a flourishing and tightly knit community with its readily available network of social support’ (Parekh 2000: 93). However, this view conflates autonomy with independence (which, it will be argued later, is not what autonomy is). Consequently, there is nothing necessarily inconsistent with drawing upon a network of social support and being autonomous. Moreover, Asian cultures in modern Western societies may have benefited from the presence of many social forms such as social mobility, pluralism, secularisation and human rights that not only encourage autonomy, but provide supporting conditions as well (Wall 1998: 171). A further criticism of Raz comes from Waldron (1989: 1121–2), who is against making the value of autonomy contingent upon the society we live in. In the future society could be different, and not require autonomous citizens, meaning autonomy would lose its importance. I therefore join Waldron, and other modernists, in claiming that autonomy is ‘unequivocally good’ and therefore the changes to Western liberal societies that have occurred to make autonomy important, and possible, should be applauded, preserved and enhanced (taking such a view is also the best defence against Parekh’s criticisms). Nevertheless, I present Raz’s argument here because I agree with it and because there are many who do not share a belief in the universal desirability of autonomy. In the discussion of liberal democracy, I have tried to demonstrate that autonomy has been a central concept to liberal thinkers throughout its history. By discussing the ideas of Locke and Mill, I have suggested that the concept is embedded in Western culture and liberal institutions and, furthermore, I have argued that many of the social forms of liberal democracies require autonomous citizens. We can and should therefore judge liberal democracy and its institutional arrangements on their ability to guarantee autonomy for their citizens. The fact that the cultivation of autonomy has been a principal concept behind the practice and theory of — 18 —

Cultivating Autonomy: The Normative Core of Democracy

liberal democracy does not, however, mean that present liberal democracies meet the standard of ensuring that citizens are sufficiently autonomous (Held 1995; Doyle 1990: 81). To consider what circumstances and institutions are required to cultivate autonomy equally, for all individuals in society, is certainly a liberal project and too extensive a one to cover here. Therefore my concern, in this book, is specifically how autonomy can be cultivated in collective decisions and what decision-making institutions are necessary to enable decisions to be made in this way. In order to establish this, we must now consider, in detail, the requirements of autonomous decision-making. The Requirements of Autonomous Decision-Making As it is agency that makes autonomy an intrinsic value, the first requirement of autonomy is an agent, such as a person, state, group or organisation. It is, however, the autonomy of the individual person that we will concentrate upon here. As well as an agent, it is argued that there must be rational free choice if the agent is to be autonomous. Both free choice and rationality can be interpreted in a plethora of ways, so detailed accounts of exactly what these terms mean, and require, in relation to autonomous decision-making must be provided. Free Choice The autonomous person is self-determining. This requires a degree of control over decisions, beliefs, preferences and actions: ‘I, and I alone, am ultimately responsible for the decisions I make, and am in that sense autonomous’ (Lucas, cited in Dworkin 1988: 6). Therefore people are thought responsible for their actions if they caused them. If someone has little or no choice in the matter – for example, if it was unavoidable – then this is thought of as a good reason for not blaming the person and such excuses are often used as a defence in legal and civil courts. Choice is therefore an essential ingredient of agency and therefore autonomy. From Raz we learn that an autonomous choice requires the agent to have the mental capacities to be able to exercise a choice, and for external structural requirements to be present to create a choice which encompasses the full range and extent of choice. For Raz these requirements are met if the agent is free from external constraints, there are a ‘variety of acceptable options to choose from’ and the agent has an awareness of these options (Raz 1986: 372–8; on this last aspect see also Hurka, 1987: 367). Finally, the agent must be able to substantively act upon the range of available — 19 —

Towards a Deliberative and Associational Democracy

options for them to be considered ‘acceptable’ and therefore be achievable, attainable and realistic.5 There are at least two broad interpretations of what a ‘free’ choice requires. Berlin famously outlined ‘Two Concepts of Liberty’ (1984) which he believed to be ‘in direct conflict together’ because they have contrasting aims (Berlin 1984: 23). In the negative sense of liberty, ‘I am normally said to be free to the degree to which no man or body of men interferes with my activity. Political liberty in this sense is simply the area within which a man can act unobstructed by others’ (Berlin 1984: 15–16). In contrast, the ‘positive’ sense of the word is more to do with being in control: The ‘positive’ sense of the word ‘liberty’ derives from the wish on the part of the individual to be his own master. I wish my life and decisions to depend on myself, not on external forces whatever kind. I wish to be the instrument of my own, not of other men’s acts of will. (Berlin 1984: 22–3)6 Negatively free choices are essential if choices are to be autonomous, as coercion reduces choices and therefore compromises autonomy. In fact, negative freedom is valuable for Raz and Young precisely because it contributes to autonomy (Raz 1986: 410; Young 1986: 8–9). Coercion occurs if an agent ‘prefers x over y, and continues to do so even when someone physically coerces him into doing y’ (Elster 1989: 82), thereby forcing them into taking an action without considering alternatives, and one they would otherwise not choose. Coercion therefore prevents free choices being made. Consequently, for a decision to be autonomous it is necessary for it to be based on negatively free choices, but the decision also requires these choices to be positively free, for several reasons. First, if being free is purely being able to pursue one’s own goals without being constrained, ‘a person could increase his freedom simply by scaling down his desires to match the opportunities available to him’ (Connolly 1993: 148), a view Berlin (1984) is aware of and agrees with. This is an example of Elster’s adaptive preferences (that is, changes of preferences to reduce dissonance), like Aesop’s fox who decides he doesn’t even want the grapes he cannot reach because they will be ‘sour’. Adaptive preference change is not an autonomous change, though, so the satisfaction of these preferences would not achieve freedom or autonomy. However, if I purposively change my preferences and desires so I can achieve more of them, then this is an autonomous change. Consequently, Elster defines freedom as being ‘free to do all the things that one autonomously wants to do’ (Elster 1991: 228). Second, Raz (1986: 377–8) is also aware that ‘autonomous wants’ are necessary and that people can also face constraints when formulating their — 20 —

Cultivating Autonomy: The Normative Core of Democracy

goals, not just on the pursuit and realisation of them (Christman, 1988, p. 111), meaning a person can face no constraint while being motivated by heteronomous desires, preventing them from being autonomous. For example, seduction and manipulation cause people to act on heteronomous desires and are ‘intrinsically morally objectionable’ precisely because they inhibit autonomy (Elster 1989: 82). Seduction occurs ‘when an individual initially prefers x over y, but comes to prefer y over x once he has been coerced into doing y’ (Elster 1989: 82). The important difference with coercion is that the person changes their perspective once they have been forced into it; it is therefore an example of dissonance reduction. Manipulation occurs when ‘an individual is led by a sequence of short-term improvements into preferring y over x, even if initially he preferred x over y’ (Elster 1989: 83). In these instances the same choices are available but autonomy is violated because the agent is not in control over their preferences, and is not making their own rational choices but is being constrained by an outside force. Waldron (1989: 1118) rightly notes that manipulation and seduction are very difficult to identify, as we are unlikely to know what the preferences of the person would have been without the external interference. Nevertheless, they still hinder autonomous decisionmaking because a person acting upon such desires is negatively free, but the desires have been manipulated or ‘unduly’ influenced by an outside source. This surely shows the inadequacy of negative freedom as a concept if used alone. For example, in countries with state-run media, people receive only limited perspectives and information, constraining their ability to form certain aims. For the same reasons brainwashing or other forms of behavioural control prevent autonomous choices and the formation of autonomous values. Likewise commercial advertising is seen as a method for changing people’s desires, but this does not always mean that it leads to autonomous desires, and certain advertising methods, such as subliminal clips, are banned for this very reason. Certain forms of conditioning can also manipulate an agent’s preferences, such as when a person forced into slavery for a considerable time, when offered the opportunity for freedom, chooses to remain in servitude (Sher 1997: 47). In all these cases the agent has not been restrained from doing the act, but is still not in control of their decisions. They were also unrestrained from doing other actions, but due to manipulation or seduction were unable to choose these, so choice is taken away. As already discussed, the conditions of autonomy do require choice, and conditions to ensure people can act upon, and exercise, their decisions. However, choice is not enough. The right conditions need to be in place for people to form autonomous preferences on which a choice will be made (Benn 1976: 123; Levine 1993: 160). — 21 —

Towards a Deliberative and Associational Democracy

Freedom, therefore, cannot just be an opportunity concept (Taylor 1997: 419), as an entirely negative conception of freedom assumes that people are already free from external influence. This is not the case, though, and consequently a democratic society needs to ensure that the conditions for participation in decision-making processes develop autonomous preferences within its citizens (Levine 1993: 160). If one is coerced, seduced and manipulated, then one is prevented from freely determining one’s own will or one’s goals and is therefore not autonomous. If one does not have the conditions (for example, knowledge, education and information) to rationally form and choose one’s goals in allegiance with one’s will, then one is not autonomous. Both freedoms are vital to the making of autonomous decisions (Raz 1986: 409; Levine 1993: 160). To summarise, a choice is free if there is an acceptable range of options, there is knowledge of these options, the choice is not coerced and the preferences this choice is based upon are not the result of manipulation and seduction. A free choice must therefore be free in the negative and in the positive sense to be autonomous. To achieve this, rationality is also necessary. Rationality: Between Kant and Hume To make autonomous decisions, the preferences that the choice is made on must be autonomously formed, which I suggest requires there to be reasons for the decision, so that the outcomes of the decisions are intended (Raz 1986: 300; Hurka 1993: 150). If choices were made randomly, with no intention, we would not think the person was in control of their decisions or that they were self-determining. Therefore we need to deliberate about choices when making decisions. By doing this we can gain ‘deliberative knowledge’, which gives us more intentions and therefore increases autonomy. The more knowledge we gain about the choices, the more we can have clear reasons for choosing or not choosing options, and the more likely we are to achieve our intentions and what Hurka terms ‘deliberative autonomy’, which is ‘free choice from a wide range of options that reflects practical reasoning about them’ (Hurka 1993: 151). The need for ‘deliberative autonomy’ is based upon a Kantian distinction between ‘autarchy’ and ‘autonomy’. To be autarchical, you need to make your own choices. To be autonomous, you must have reasons for these choices. Kant, whose theory suggested that to be autonomous was to live life in accordance with self-imposed maxims which have been developed through reason, argued that people could be morally autonomous precisely because they are capable of reason (Lakoff 1996: 157). Alternatively, Hume7 believed ‘that reason alone can never be a motive to any action of the will’ — 22 —

Cultivating Autonomy: The Normative Core of Democracy

and ‘can never oppose passion in the direction of the will’ (Hume 1978: 413). The central dispute here, for the relevance for autonomy, is whether preferences, desires and passions can be formed through reason and should play a part in our decision-making processes; Kant argues reason can and Hume says reason cannot. Williams (1981) clarifies this dispute by differentiating between internal and external reasons. Internal reason represents the Humean approach, where an agent has a reason for an action if they have a motive for that action, and the external approach follows Kant, where an agent has a reason for action even if they do not possess that motive. These two contrasting views and their implications for autonomy will be reviewed in turn. Kant conceives an autonomous decision as one motivated by reason, with rationality being divorced from all desire and inclination. Actions based on desire are disregarded as heteronomous, because desires are subject to causal processes, which means actions are determined and not chosen. For Clarke (1999) Kant goes too far arguing it is just whimsy and wantonness that must be avoided if autonomy is to be preserved, as acts that are motivated by desire and passion can still be autonomous. This is in accord with Williams’s critique of external reason statements like Kant’s. Williams points out that ‘no external reason statement could, by itself, offer an explanation of anyone’s action’ (Williams 1981: 106). The Kantian external approach to reason therefore appears incoherent, as it would mean that someone could gain a reason for an action from a motivation they do not have (Williams 1981: 109). It seems that divorcing reasons from desire is therefore an excessively stringent requirement for rationality and autonomous opinion formation (Lindley 1986). J. S. Mill (1993) also professed that critical rational enquiry was essential to autonomous opinion formation, but in contrast to Kant, who wanted rationality to be separated from inclinations to achieve autonomy, Mill claims that the inclinations are an inevitable part of the opinion formation process and therefore should be based upon rational critique themselves. Similarly, Lindley argues in favour of ‘active theoretical rationality’, which requires the agent deliberating over not merely preferences but also desires and beliefs (Lindley 1986: 21–7). In stark contrast to Kant, Hume perceived reason to be purely instrumental in achieving goals dictated by preferences and desires, summed up in Hume’s adage that ‘reason is, and ought only be the slave of the passions, and can never pretend to any other office than to serve and obey them’ (Hume 1978: 415). This means that it is passion that is the motivator of action, but requires rationality in order to achieve its goals, with the more intense the passion, the greater the reason for acting upon it. In this sense — 23 —

Towards a Deliberative and Associational Democracy

passion uses reason, rather than passion being subordinate to reason, as Kant perceived the relationship. Hume is right that passions, not reason, are the motivator of action. He is also right that the content of passion can be good or bad, moral or amoral and so on; in other words, passion has no objective content. From here, though, he concludes that preferences are not susceptible to rational critique unless they are based upon false supposition (Hume 1978: 416). This is a significant weakness of the Humean approach to rationality, as it suggests that an agent can be autonomous without having autonomous preferences, a position based on an inadequate conception of the role of deliberation. The Humean view of internal reason thinks that desires can only be subtracted from the agent when proved false, but not added. Williams’ approach to internal reason is more accurate when he asserts that, through deliberation, an agent can come to see that he has reason to do something which he did not see he had reason to do at all. In this way, the deliberative process can add new actions for which there are internal reasons, just as it can also add new internal reasons for given actions. (Williams 1981: 104) A related problem is Hume’s over-emphasis on passion. Being overly passionate about something can detract from autonomy by blinding people to reason (Hurka 1987: 367). Consequently, Clarke is critical of the Humean view, suggesting that ‘passion can use reason to serve its ends but passion cannot judge itself’ (Clarke 1999: 195). Kant’s categorical imperative is a way of testing passions, not of generating them; it subjects them to reason, to test the validity of these passions separate from the passion itself. Similarly, existentialists such as Sartre believe that people are responsible for their passions (Sartre 1948: 34). For existentialists, the measuring of the strength of a feeling can only be judged if it is acted upon, because if it is not acted upon then we must conclude that it is not the strongest feeling. If this is the case, we cannot therefore justify an action upon this feeling without being drawn into a vicious circle. Consequently, for Sartre, feelings are formed by action rather than being a guide to action (Sartre 1948: 37). Schumpeter (1974), Scoccia (1990) and Elster (1976) all follow in the Humean tradition and suggest that ‘economic rationality’, is the most appropriate conception of rationality. Economic rationality is a version of internal reason, where rationality is a means to bring about desired ends. This is achieved by selecting the most efficient method to achieve a given end (Scoccia 1990: 320). I think this view is mistaken, as economic rationality is simply choosing the best method to achieve one’s goal; there — 24 —

Cultivating Autonomy: The Normative Core of Democracy

is no guarantee, however, that the goals themselves will be rational and therefore the formation of these goals must be rational as well. For example, if I have the desire to cut off my ear, then through economic rationality I would be contemplating the best method to sever my ear from my head, and the best method would be dependent on my other desires, such as ‘Do I want it done quickly?’, ‘Do I want it done the least painful way?’, ‘Did I want the severed ear whole?’ There is no falsity of belief here, so it is autonomous by Humean standards, but it seems apparent there is further reason required to justify the desire and for it be considered rational. In reality it is likely to be some strange desire, possibly arising from anger or passion, which I would come to regret later (Benn and Weinstein 1971: 195). Dewey eloquently articulates the necessity of rational thought when choosing what goals to pursue and actions to take, if the chooser is to be autonomous: Impulses and desires that are not ordered by intelligence are under the control of accidental circumstances. It may be a loss rather than a gain to escape from the control of another person only to find one’s conduct dictated by immediate whim and caprice; that is at the mercy of impulses whose formation intelligent judgement has not entered. A person whose conduct is controlled in this way has at most only the illusion of freedom. (Dewey 1950: 75–6; see also Wolff 1976: 12) The full weakness of the Humean view is captured if we, once again, consider Huxley’s Brave New World (1994). The citizens here do choose and can employ economic rationality to achieve their preferences, but we would surely not want to term them autonomous, as in no way have they contributed to the rational formation of these preferences. Hume’s view of autonomy does not therefore encompass a proper conception of agency, which I have suggested is what makes autonomy an intrinsic value in the first place. In short, the rationality required for autonomy requires not just the satisfaction of whatever desires one may happen to have, but the efficient satisfaction of rationally formed desires. For goals to be minimally autonomous, they must have been ‘critically scrutinised’. This means that before a goal or value is selected, it must have had its strengths and weaknesses evaluated and considered, which requires the consideration of available information about the options and their possible consequences. Young provides a possible defence of the Humean view. He agrees that an agent’s preferences, desires and passions can be irrational, and therefore the agent may have no good reason to act upon the desire to cut off their ear, but suggests that it is not irrational to have the desire, as it is just a feeling, and so, ‘when beliefs and decisions to act are stripped away from — 25 —

Towards a Deliberative and Associational Democracy

inclinations, there is nothing left about the inclination to be irrational’ (Young 1986: 39). If this is the case, though, the Humean connection between desire, reason and action is broken. This is because as well as having reasons to do something, we can also have reasons not to do something, and there are many good reasons why someone should not cut off their ear. They could have the desire to cut off their ear and not act upon it because they have good reasons not to. In this sense, desires, preferences and passions can conflict, but the Humean conception of rationality never offers us an account of how this conflict should be resolved, even if it is adapted to accept the existence of the irrationality of acting upon certain desires. If we follow Hume then we should simply act upon the strongest desires, but that is not always possible to determine; moreover, the Humean view denies the possibility of the individual being able to distance themselves from desires and critically evaluate them (Young 1986: 42). The case being made here is that rationality provides the connection between ‘will’ and ‘choice’. There is no objective set of rational preferences, as Kant suggests, but it is the individual and their deliberation that provides the motivation of judgement behind rationality’s critical appraisal, which then determines whether someone finds a reason convincing or not. On consideration of two of the most dominant traditions in the connection between rationality and autonomy we must reject, and accept, elements of both Kant’s and Hume’s approach. From Kant we accept his argument that rationality and not inclination must be dominant to achieve autonomy. However, Kant goes too far and does not acknowledge the relevance of particular desires and interests. Hume rightly accepts the relevance of desire and preferences but does not think that these preferences can be rationally formed, therefore he thinks only economic reason is sufficient for autonomy. The view I am advocating here finds a position between Hume and Kant (although is probably more Kantian). It accepts from Hume that desires motivating actions is compatible with autonomy; and therefore sees internal reasons as important it also, however, accepts from Kant that reason must contribute to the formation of these desires, meaning that external reason also plays a significant part in the rationality required for autonomy. There is the danger, though, that this latter aspect leads to an infinite regress, and it is to this problem we now turn. Rational Preference Formation and Infinite Regress If reason is employed when forming desires and preferences, then we are in danger of an infinite regress in so far as the problem arises of how to know if the preferences used to review preferences are themselves autonomous. — 26 —

Cultivating Autonomy: The Normative Core of Democracy

Such an infinite regress is present in the work of Dworkin (1988). He calls preferences that have been rationally evaluated ‘second order preferences’ because they have been formed through the critical scrutiny and reformation of ‘first order preferences’ in accordance with one’s will. For Dworkin, lower order desires are attributable to actions while higher order desires deal with lower order desires. Dworkin expects an agent to ‘identify’ with some, but by no means all, lower order desires in an ‘important way’, and it is this identification that is the key to autonomy, for him.8 Identification occurs following a process of rational reflection, whereby the agent considers their lower order desires to be the sort of desires they want to have, and perceives them as being in accordance with their higher order beliefs and principles (Dworkin 1988). It seems apparent that this process of identification involves endorsement and not just realisation (Christman 1988: 113), which means the agent is not just appreciating the fact that they have these desires, and seeing them as part of them, but they are approving of them too. If, then, we agree with Dworkin’s suggestion that autonomy is identification of higher with lower desires, is it possible to have a desire that one disapproves of? What if a person lives a heteronomous life, but identifies with this? Moreover, people are socialised, seduced and manipulated when forming higher order, as well as lower order, desires. Hence, the higher order preferences may be as (or even more) heteronomous than the lower order ones, and therefore identification can occur while the person is still being motivated by desires that are not rationally grounded. Consequently, there is an infinite regress which any model that links rationality and autonomy must address (Christman 1991: 8; Christman and Anderson 2005: 6). This is not to say that Dworkin’s distinction between higher and lower order preferences is useless and meaningless. It is quite possible to perceive that people do have different levels of preferences, and people certainly have preferences about their preferences, and have overarching desires about their life. However, Dworkin has not said what conditions are required to ensure these higher order preferences are autonomous (Christman 1988: 113–14). Perhaps it is the process of preference formation that is key. In Christman’s case, a preference is autonomous if the person is aware of and approves of the process whereby they form or accept a preference (Christman 1991: 11). Three conditions are required for this acceptance to be adequately achieved: I. That the development of the preference was not resisted or would not be resisted. This means ‘that the agent would not resist – that is take action to counteract – the process, were she to understand it’ (Christman 1991: 13). — 27 —

Towards a Deliberative and Associational Democracy

II. That this lack of resistance was not due to factors that inhibited selfreflection. III. That the self-reflection involved in ‘I.’ was rationally motivated and did not involve self-deception (Christman 1991: 11). Christman’s theory avoids being regressive because it is not depending upon any self-appraisal based on other preferences. Instead the agent appraises the process of desire formation, and providing this appraisal is rational and devoid of self-deception, seduction and manipulation, the acceptance of the desire is autonomous (Christman 1991: 18–19). These autonomous preferences can also be used to re-evaluate already-obtained desires and beliefs. For example, if I autonomously (that is, I do not reject the process by which I) gain a new belief in the benefits to my health that can be brought about by drinking red wine, I can then use this to evaluate other desires. It may mean I start to drink red wine more regularly and other drinks, like white wine, I may desire, and therefore consume, less. However, because the desire to have more red wine was autonomously gained when it is used to re-evaluate other desires on other beverages, this has been an autonomous process. Related to this is Raz’s (1986: 389) argument that we must have current reasons for our preferences and decisions, even if they were not the reasons for its original adoption. If I already preferred red wine, but drank it because I thought other people would think me more sophisticated, but later rejected this reason, the fact that I now believe in its health benefits is the reason why I now have the preference for red wine. The key with previously formed preferences is that we acknowledge that we could abandon them (Raz 1986: 383–5). For preferences to be autonomous, the process through which they are formed must then be conscious, rational and accepted by the agent in question. This, though, still does not tell us what type of rationality is required. Two types offer different solutions: internalist rationality and externalist rationality. ‘Internalist’ and ‘Externalist’ Rationality In addition to the distinction already covered between Kant and Hume, there is a further, related, distinction that exists between ‘internalist rationalists’, who require an agent’s preferences to be consistent (that is, transitive), and ‘externalist rationalists’, who require the agent to have sufficient and objectively correct evidence (Christman 1991: 13–14). The internalist approach argues that, to be autonomous, the person must have rationally consistent means to form desires and that the resulting desires must also be consistent. Therefore an autonomous person must not be — 28 —

Cultivating Autonomy: The Normative Core of Democracy

‘guided by manifestly inconsistent desires or beliefs’ and ‘the final ends and purpose that an agent has must also be consistent with the rest of the judgements, values and beliefs to which she has committed herself’ (Christman 1991: 15). Internal rationality seems unproblematic, and essential for autonomy, but there are a few issues with external rationality that mussed be addressed. External rationalists agree that the more information a person has, the more autonomous they are. This is because ‘the effect of the information is its ability to permit the citizen to put more of himself into his deliberations, and hence to permit his preference to express himself fully’ (Lipson 1995: 2264, my emphasis). Therefore, new information can bring us to review and amend previously formed preferences, and a lack of relevant information can inhibit self-reflection. In this sense, uninformed preferences are not completely ‘authentic’ preferences, as they are not necessarily fully intended. This also means that being autonomous is ‘open ended and vague’, making it difficult to say when a person is sufficiently informed to make an autonomous decision. This causes practical problems highlighted effectively by Elster’s (1976) mushroom-gatherer. The gatherer needs to seek out a good place to pick mushrooms, but should not spend too much time comparing places, as this will reduce the time available to pick mushrooms. They do need information to make their decision more rational and autonomous, in the sense that they will increase their chances of achieving their end of picking lots of mushrooms, but it is hard to know how much information is enough. We could answer this only if we knew what effect the extra information would have on the mushroom-picker’s decision, but we do not know this until they receive this information. Information is, then, always contingent and limited, but an agent can still attain available information. Such considerations prompt Christman (1991) to reject the externalist requirement of rationality, but this has some serious implications. A person could have internally consistent preferences, but have several or even all of these preferences based on misinformation and/or manipulation if, say access to information is controlled and limited by another source. As suggested by Christman above, autonomy requires accepting the process by which one gains one’s beliefs and desires, but someone may accept the process by which they have formed their beliefs and desires and yet be unaware that such manipulation has taken place. Surely most people would not accept their preferences if they were to learn that they had been based on limited information that could have been available and would want to know what that information was in case it did have an effect upon their — 29 —

Towards a Deliberative and Associational Democracy

preferences. As we cannot rely on an ideal thinker, the only way to find out if information would affect their preferences is for them to receive it. If, having been told that their access to information has been manipulated, the person is still happy with the method of forming their desires and beliefs, then that person has failed to meet Christman’s first condition of autonomous preference formation, as they cannot fully understand the manipulation, and therefore the preference formation process, without knowing what information they have been excluded from having. If they say they have acted autonomously, then surely there is an element of self-deception which means Christman’s second condition has not been met. Moreover, the manipulation could be the very cause of making the person think that they have adequate information to form autonomous desires and beliefs. Therefore, one-sided information upsets an agent’s capacity for rationality and self-awareness and their ability to form autonomous preferences. Locke’s prisoner who is ignorant of the fact that his cell door is not locked is a pertinent example of this (Dworkin 1988: 105). Consequently, if a change in preferences is to be autonomous, then it must be the case ‘that the individual – if rational – would have done the same himself given the same knowledge about the causal process underlying the preference change’ (Elster 1976: 500). The preferences upon which these choices are made must be rationally formed, but it is the acceptance of the process – and not the content of the preference itself – that is key. Internalist rationality must be invoked to ensure that these preferences are consistent, and externalist rationality is also necessary, and involves the consideration of information and reasons for and against the preference. However, the acceptance of these factors means we must address to what extent they need to be present to achieve autonomy. To what Extent should People be Rational to be Autonomous? Autonomy, as it has been conceived here, is in danger of being indeterminate in so far as it becomes unclear what minimal standards of rationality an individual would have to display in decision-making in order to be judged minimally autonomous. There are four variables to consider: people can rationally form more or less of their preferences; they can use more or less rational consideration when forming a particular preference; they can use good or the strongest reasons; and some people may make more rational decisions than others. First, it is not clear how many of a person’s preferences must be rationally formed for them to be autonomous. For example, Haworth (1986) distin— 30 —

Cultivating Autonomy: The Normative Core of Democracy

guishes between ‘normal autonomy’ (which is necessary to be a moral and political agent) and ‘full rationality’ (a regulative ideal where all of one’s preferences, beliefs, goals and values are subject to critical appraisal) (Christman and Andersen 2005: 2). As it is autonomy in collective decision-making that is of our concern here, it is vital that preferences that enter into collective decisions are rationally formed. If autonomy does have an intrinsic good, some autonomous decisions are more important than others: ‘It is more valuable to choose goals which organise and encompass many others subordinate to them in a means-end hierarchy’ (Hurka 1987: 373). In this sense, autonomy in forming central goals such as where to live, if and who to marry and what religious beliefs to hold is more important than the achievement of a specific goal like eating a plate of spaghetti without making a mess or choosing what underpants to wear in the morning. The former dictates what other specific goals we will adopt (Hurka 1987: 373–6), while the latter are ‘too trivial to be worth seeking’ (Hurka 1993: 151). Political relations, being by their nature conflictual, involve collective action and decisions to be made which mean that we must, ‘place a premium on clarifying interests and seeking influence through argument, thus making the concept of autonomy essential to that of political self-rule’ (Warren 2001a: 62). Second, it is unclear how much rationality must be invoked when testing or forming a particular preference. Not all people will consider the options, arguments and information to equal degree. There will be the extremes of those who consider and deliberate carefully over everything, and those who act merely on impulse and do not take any time to consider reasons for actions other than what they feel like doing at that moment. However, most of us fall between these extremes and, as suggested above, will take time to deliberate on decisions such as buying a house or selecting a career, but give little consideration to choosing a flavour of ice-cream (Scoccia 1990: 321). This is sufficient for autonomy because most decisions do not require deliberation in so far as most circumstances fall into ‘fairly standard, readily recognisable categories’ – meaning that ‘living according to principle does not demand continuous ratiocination’ (Benn 1976: 127). This is in accordance with Wall’s distinction between comprehensive and peripheral choices. Comprehensive options are fundamental to the formation of an agent’s identity, whereas peripheral options are dictated by comprehensive options or are not related to comprehensive options (Wall 1998: 140). Scoccia (1990) argues that it is possible for some people to be bad at deliberating and, consequently, to make more autonomous choices by acting on impulse, as doing so reflects their ‘true self’ and will more accurately. For example, when choosing a car someone may make a more — 31 —

Towards a Deliberative and Associational Democracy

autonomous choice if they impulsively buy one because it is red rather than finding out about its mileage, fuel economy, safety credentials and so on, and deliberating over these factors. Benn accepts that someone could act autonomously on impulse if they were compelled to for the type of reason Scoccia outlines. The colour of the car may be of prime importance to that person, therefore it is a good decision. However, the person is not autonomous if ‘he acknowledges nothing as a reason for doing otherwise. Caring about nothing’ will mean ‘he sees no point controlling the inclination of the moment’ (Benn 1976: 124). For some people variety and spontaneity is the life that best suits them, but this does not mean that they have not decided this rationally. However, most people’s lives do involve long-term commitments and aims, and spontaneity is usually a barrier to achieving these (Wall 1998: 133). The third variable occurs because there are many different reasons for action; but which reasons are the important reasons for acting in order to ensure autonomy? The first requirement here is that these reasons must be relativised and contextualised, so that they are reasons of which an agent is aware in any given situation (Sher 1997: 53). However, there are limits to the extent autonomy can be relativised in this manner. We cannot, for instance, say an agent acts autonomously if, and only if, they act on what they consider to be the strongest reasons for action at a specific time because this would result in autonomy having no content at all, and any actions or reasons for action would appear autonomous, even the insane. Moreover, the connection between acting autonomously and finding one’s actual reasons for action would be lost (Sher 1997: 54). In this sense, in order to be autonomous, an agent may not need to follow their ‘strongest reasons’, but instead their actual reasons. From this premise, Sher argues that a choice or action can be reason-based even if there are stronger reasons to select an alternative option, and therefore still be autonomous. The basis of this argument is a distinction between rationality and full rationality. The conclusion is that ‘an agent may qualify as autonomous whenever he acts in response to reasons provided by what he knows about his situation that are at least strong enough’ (Sher 1997: 55). Finally, it is undeniable that some have a greater ability for rationality than others, and this could well mean that they will be more autonomous. However, it does not imply that, when making collective decisions, the autonomy of the more rational should be valued above the autonomy of the less rational. I agree with Dworkin that what matters is that all citizens have the capacity to be autonomous to a certain level, even though some will have a capacity well above this level: it does not justify a hierarchy. The fact that all are capable of this rational choice means all should have their — 32 —

Cultivating Autonomy: The Normative Core of Democracy

chosen preferences respected equally (Dworkin 1988: 31–2). It is, of course, very difficult to say what the ‘certain level’ of rationality is, which is essentially an empirical question and one which cannot be addressed here, despite the fact that it is of central importance. Nevertheless, excluding those not sufficiently autonomous from collective decisions is not legitimate because this would break the essential relationship between autonomy and democracy. Those excluded from collective decision-making would lose the opportunity to be autonomous – a point J. S. Mill (1993) seemed to be oblivious of regarding his ‘education’ benchmarks that he thought people should meet before being granted the vote. The key is to provide opportunities for all to be sufficiently autonomous, and this requires opportunities to participate in collective decisions. To summarise, the conditions to be autonomous are free choice and rationality. A choice is free if there is an acceptable range of options, there is knowledge of these options, the choice is not coerced and the preferences this choice is based upon are not the result of manipulation and seduction. The preferences upon which these choices are made must be rationally formed through deliberation, but it is the acceptance of the process, and not the content of the preference itself, which determines autonomy. Internalist rationality must be invoked to ensure that these preferences are consistent, and externalist rationality, which involves the consideration of information and reasons for and against the preference, is also necessary. Finally, there are different extents of these variables: people can rationally form more or less of their preferences, use more or less rational consideration when forming a particular preference, and use good or the strongest reasons. However, the fact that some people may make more rational choices than others does not justify inequality in collective decisions, and we should aim to create the opportunities for all to be capable of rational choices to a certain degree. Now that we have outlined the necessary features for autonomous decision-making, we can now consider various critiques of this position. Communitarianism One of the main accusations against the value of autonomy as a justification of democracy is that it is not possible because autonomy requires us to be self-determining and make choices. Communitarians, however, argue that we are completely socially determined, and therefore unable to make these choices. If people are the product of their society, they cannot make choices, as choices are based upon values and desires which have been constituted in the person by the society itself. This section will consider the — 33 —

Towards a Deliberative and Associational Democracy

communitarian critique of autonomy, but will ultimately defend the conception of autonomy by agreeing with Caney’s view that ‘this argument fails because it foists upon liberals an implausible conception of autonomy’ (Caney 1992: 277) in so far as it makes an assumption that liberal theory’s conception of autonomy is necessarily atomistic (Levine 1993: 160). As Raz argues, ‘the completely autonomous person is an impossibility. The ideal of the perfect existentialist with no fixed biological and social nature who creates himself as he goes along is an incoherent dream’ (Raz 1986: 155). The following discussion will hopefully lend support to Raz’s argument, and show that, if properly conceived, the acceptance of the embedded individual does not mean a rejection of the liberal theory of autonomy. Sandel’s works are considered classic statements of this communitarian critique. For Sandel, ‘the notion of a subject prior to and independent of experience’ is ‘a necessary presupposition of the possibility of freedom’ (Sandel 1984: 85). It is safe to assume, I think, that Sandel would feel that, to be autonomous it is equally necessary to have an ‘unencumbered self’ prior to, and separate from, our goals and ends, and therefore autonomy is similarly impossible to achieve. Clarke succinctly sets out the core to the communitarian argument: ‘The autonomous self . . . does not exist and if the self did exist it would, in any case, not be autonomous’ (Clarke 1999: 1). Sandel is sceptical of the existence of a ‘freely choosing individual’ because individuals cannot conceive of themselves separately without consideration to their role and position within the community that has formed them, therefore we can never be separate from our ends: ‘If we are partly defined by the communities we inhabit, then we must also be implicated in the purpose and ends characteristic of those communities’ (Sandel 1984: 167). If we accept this argument, then there can be no such thing as an autonomous individual, selecting their goals and ends: I don’t choose to love my mother and father, to care about the neighbourhood in which I grew up, to have special feelings for the people of my country, and it is difficult to understand why anyone would think that I have chosen those attachments or that I ought to have done so. (Bell, cited in Barry 2001: 149) In particular, Sandel criticises Rawlsian liberals whose theories rely on an ‘unencumbered self’. To be an ‘unencumbered self’ is to be independent of one’s own values, and for them to be separate from the self in some important way is to suggest that there is ‘some subject ‘‘me’’ standing behind them’ (Sandel 1984: 86). If the self is unencumbered, then it is not the ends and desires I have that constitute my identity and define me, but something — 34 —

Cultivating Autonomy: The Normative Core of Democracy

else that is more important than any of these. For Sandel and other communitarians it is not possible to separate from one’s desires and goals because these define the person, and without them it is impossible to critically evaluate anything, as there would be nothing to motivate the evaluation or to base it on (Sandel 1984: 86). Therefore a person without any history, without any attachments, without any goals, desires and beliefs is difficult to imagine, and certainly such a person could never be autonomous (Lipson 1995: 2264). If there is no such thing as the self, if people are entirely determined by society, and its myriad of contingent factors, then there is no chooser making decisions and forming desires and goals in accordance with the sort of person they want to be. People are who they are, and it is out of their control; consequently, autonomy is a myth. Clarke defends the idea of autonomy against the communitarian critique by arguing that something rising in ‘contingently precise circumstances’ does not mean it does not exist, as everything exists in ‘contingently precise circumstances’, including autonomous choices (Clarke 1999: 136). Moreover, the way the conflict between the atomistic and situated self has traditionally been conceived invites us to see the moral universe in dualistic terms: ‘Either our identities are independent of our ends, leaving us totally free to choose our life plans, or they are constituted by community, leaving us totally encumbered by socially given ends’ (Gutmann 1985: 316–17). This is a false dualism, though, because liberals need not, and do not, deny that people are socially constituted, and accept that there is no pre-social self. They need only, and do, claim that there exists a self that can critically analyse the values and processes in the society that has socialised the self. Nevertheless, Sandel maintains that this critical reflection, because it inevitably involves distancing oneself from what one is analysing, is always ‘precarious’ and ‘provisional’ in so far as the ‘point of reflection is never finally secured outside the history itself’ (Sandel 1984: 91). There are, however, several faults which undermine Sandel’s argument. First, even if an individual’s life is not metaphysically autonomous in the sense that every aspect of it has been established and chosen by the agent, this does not rule out autonomy. Ends and attachments can still be interpreted in a unique way, even if this interpretation is not unlimited due to agents inevitably being contextualised (Clarke 1999: 111). Clarke further distinguishes between ‘the social and contingent self’, which is determined by environment, and the ‘conditions of selfhood’, which is not, and therefore enables people to ‘be able to make some admittedly limited choices within the passive presentations of experience’ (Clarke 1999: 151). Choice being limited does not mean choice is determined, or that the will is not free when making those decisions which would indicate that autonomy, — 35 —

Towards a Deliberative and Associational Democracy

although restricted, is possible, and provides important meaning to that life. If we do not accept this counter to Sandel’s case, then we must be led to conclude that only God could be autonomous, as only a God can be truly self-governing and self-determining in the sense that it forms its own nature, conditions and principles upon which to act. But this is a very strong requirement to be autonomous, as it essentially means equating autonomy with omnipotence (Clarke 1999: 42). A further criticism of Sandel comes from Kymlicka, who suggests that Sandel has misrepresented the liberal argument, and its perception of the self. To hold a belief in the autonomous individual is not to see them separate from ends per se (as, without ends, one could not even engage in rational analysis of various ends), but to see them as separate from any particular ends (Kymlicka 1988: 190). By this Kymlicka means the individual is capable of reviewing and revising their ends through rational thought and is able to change their mind and revise their goals in light of new information and different considerations. If this is the case, then ‘our self is, in this sense, perceived prior to its ends, i.e. we can always envisage our self without its present ends’ (Kymlicka 1988: 190). Raz accepts that the commitments we gain from our environment do influence our choices of whether to maintain them or not, but claims we still have choice as to whether to keep them or not (Raz 1986). Further defence is still required of autonomy, as Kymlicka’s argument does not counter the communitarian argument that ends are not chosen but discovered, as ‘constituents of our identity’. Yet this communitarian claim seems to flout our natural instincts and understandings of ourselves. We do not see ourselves as bound for life to certain ends and commitments, but believe we can (and do) make choices in life that change its direction and make new commitments, which alter our identity. It is certainly not the case that all people do make such choices. Some people seem so embedded in a particular social practice that they become blinkered and unable to review that practice in any meaningful sense. Nevertheless, to deny that anyone, or even most of us, are not capable of reviewing at least some of our attachments goes against both instinctive and empirical evidence. Sandel himself appears to admit that identity is not just a question of ‘discovery’ but includes participation from the agent in forming that identity and the ends that ‘impinge’ upon it; in this sense, ‘the bounds of the self (are) open and the identity of the subject (is) the product rather than the premise of its agency’ (Sandel 1982: 58). If Sandel does accept this, then surely he is accepting that people can reconstitute their identity and select new goals and ends at least in some form, in which case surely — 36 —

Cultivating Autonomy: The Normative Core of Democracy

Kymlicka is correct to claim that ‘at this point it’s not clear whether the whole distinction between the two views doesn’t collapse entirely’ (Kymlicka 1988: 192; see also Walzer 1989: 21; Caney 1992). Communitarians assert that the distinction is that we can interpret the social relations that we find ourselves in in different ways, but we cannot reject these relationships because we are ‘embedded’ in them. But does this mean that someone who has been brought up as a Catholic cannot reject Catholicism in later life? Surely they can, and have. Similarly some people brought up as atheists re-evaluate their convictions to commit themselves to a religion. Therefore, it does seem that people do critically evaluate goals and aims they ‘discover’ and, in light of new information, perspectives and experiences, reject or alter these goals and establish new ones. In this sense, the individual is prior to any particular aims, but not prior to aims in general. Consequently, individuals can be a ‘chooser’, and can be autonomous. Yet this is not to say that they will be unless the conditions necessary for autonomous choice are present (Raz 1986: 312). Therefore, all Sandel has proven is Nagel’s (1975) criticism that Rawls is wrong to talk of a ‘view from nowhere’, as an unencumbered self does not exist. There is no original position, there are no people not formed and framed by their environment. This does not mean that within this framework choice becomes impossible, or that people are completely passive. Neither does it mean that people cannot critically review their attachments, reflect upon their desires and then make active choices and decisions about them. Habermas accepts that people are not able to ‘choose’ everything about their life, due to the processes of cultural tradition, social integrating and socialisation. However, for him the individual still has the ability to make autonomous decisions and form autonomous preferences ‘by appropriating traditions, belonging to social groups, and taking part in socialising interactions’ (Habermas 1990: 102). In one sense, then, the agent is a ‘product’ of the traditions of their community, but also an ‘initiator’ who controls situations through their own actions for which they then must be responsible. Habermas sees the lifeworld that provides the background to each individual as a pool of resources that can be drawn upon when making interpretations of actions and forming beliefs and goals: ‘In the dialectic between nature and the socially constructed world the human organism itself is transformed. In this same dialectic, man produces reality and thereby produces himself’ (Berger and Luckmann, cited in Barber 1984: 215; see also Gould 1988: 107).9 Similarly, Swindler sees the values and ideas that are induced in people through socialisation and culture not as a straitjacket, but as a resource of — 37 —

Towards a Deliberative and Associational Democracy

‘symbols, stories, rituals and world-views, which people can draw upon to solve various problems’ (Swindler, cited in Santoro 1993: 137; see also Taylor 1985: 190–1). Without cultural resources there would be no criteria upon which to base a choice and therefore to be autonomous is ‘not to have a capacity for conjuring criteria out of nowhere’ (Benn 1976: 126). The fact that culture and community provide resources to make autonomous choice does not rule out the possibility of being autonomous, providing these resources are critically evaluated themselves: Within this conception of a socialised individual, there is room to distinguish one who simply accepts the roles society thrusts on him, uncritically internalising the received mores, from someone committed to a critical and creative conscious search for coherence. The autonomous man does not rest on the unexamined if fashionable conventions of his sub-culture when they lead to palpable inconsistencies. He will appraise one aspect of his tradition by critical canons derived from another. (Benn 1976: 126)10 Furthermore, without certain social practices an individual cannot pursue certain goals. For example, without the social recognition of ‘birdwatching’ one cannot be a ‘birdwatcher’ because otherwise this would apply to anyone who happens to see birds. Custom and tradition are therefore not incompatible with autonomy, as an agent can rationally review these connections and accept them. It is also unsurprising if they do, as customs and traditions play a significant part in forming people’s nature and rationality (Wall 1998: 138). If societies are formed and changed through the actions of individuals and collectives, then, providing some of these actions and causes are intentional, it is possible for autonomous action to take place. Therefore democracy is essential because it strives to organise society to allow for all to have an equal role in making these collective decisions, which in turn determine collective action, so that autonomy for all is equally cultivated. Democratic decision-making can mean many aspects of society that affect people have been autonomously selected (Post 1993: 673). This discussion of communitarianism should therefore show that we should not consider individuals as entirely atomistic and that the concept of autonomy, properly conceived, does not require us to do so. Accordingly, we can still provide ‘a coherent ontology in which individuality is given its full due but not at the cost of regarding individuals as isolated and abstract egos, standing in only external relations to each other’ (Gould 1988: 105). In this sense, the autonomous and rational chooser is possible, providing the — 38 —

Cultivating Autonomy: The Normative Core of Democracy

individual is not seen as a complete isolated being and that it is accepted that society and environment will form this individual, with the right environment actually making the critical reflection and choice among ends possible and not impossible. Perfectionism Perfectionism occurs when the state promotes certain ways of life (Waldron 1989: 1102). Liberal perfectionism is ‘a perfectionist theory that holds personal autonomy as a central component of human flourishing’ (Wall 1998: 127). Consequently, by seeing autonomy as an intrinsic good, and wanting to promote autonomy for all, as it is here, we ‘embrace a mild perfectionism’ (Hurka 1987: 361).11 D’Entre`ves, argues that justifications of democracy based upon autonomy, like those of Raz and Kymlicka, are guilty of perfectionism (D’Entre`ves 2002: 41–3). The central criticism of perfectionism is that ‘it will ultimately favour those individuals and groups whose conceptions of the good or wellbeing are predisposed to the value of autonomy’ (D’Entre`ves 2002: 42). Therefore, any perfectionist justification of democracy fails to be truly inclusive.12 Similarly, Gray argues that grounding liberal democracy on autonomy goes against pluralism, as he suggests that there are many good lives that are not autonomous but should still be protected by liberal democracies (Gray, cited in Wall 1998: 163). Barry argues that liberalism can provide the institutions to make autonomy possible, but cannot force people to act autonomously, and those who do not wish to be autonomous should be free to act that way (Barry 2001: 121). Autonomy certainly is a value, and if we argue (as I do here) that democratic rights are necessary to ensure autonomy, then we cannot claim, as some Kantians do, neutrality towards values. Not all people would accept the value of autonomy, for example, those guided by self-sacrifice or self-abnegation, or some religious fundamentalists. Many others may not rank autonomy as such an important value. Therefore devising a system that guarantees the autonomy of all would involve perfectionistic interference towards these people (Barry 2001: 123). The argument against perfectionist justifications of democracy is, then, that rather than promoting any particular values, democracy should be neutral because if the state is biased towards a way of life then it does not treat all people equally (Dworkin 1985: 191). The view that the state should be neutral towards conceptions of the good life has been central to contemporary liberal thought (Nozick 1974; Ackerman 1980; Dworkin 1985; Larmore 1987; Nagel 1991; Rawls 1993). However, Barry argues that — 39 —

Towards a Deliberative and Associational Democracy

liberalism is not, and never should have been, considered a justification of neutralism, as it would mean that there was ‘no existing (or possible?) worldview with which it conflicts’, and this is simply impossible (Barry 2001: 27). However, there is a distinction between strong and weak neutrality. The strong version is consequentionalist, and suggests that the state must not pass laws or policies which ‘have the effect’ of favouring any particular conception of the good life. The weaker conception of neutrality is deontological, and means the state can be neutral, providing it does not pass laws or policies ‘in order’ to favour a conception of the good life. Now Barry is right that the stronger version of neutrality is impossible, but his arguments do not rule out the weaker version, so further defence is required (Waldron 1989: 1134; Sher 1997: 4). Following this argument, Sher provides a definition of the requirement of neutrality: A law, institution, or other political arrangement is neutrally justifiable if and only if at least one possible argument for it (1) has neutral normative premises, and (2) contains no implausible premises or obvious fallacies, and (3) provides justification of reasonable strength. (Sher 1997: 26) However, what is reasonable, plausible and obvious are contentious issues themselves (Sher 1997: 26; D’Entre`ves 2002: 40). Therefore such a conception of neutrality will be unlikely to be achieved, as agreement on these factors is unlikely to be forthcoming (Caney 1995). One of the most important justifications for the neutral state is that nonneutral laws and policies restrict autonomy, and many liberals, including J. S. Mill (1993), Kant, Waldron (1989) and Rawls (1993), have made such claims. The essential premise of these theories is that to be autonomous one needs to be self-determining, and so the individual agent must make their own decisions; but if the state promotes a conception of the good, it compels individuals to act in ways that they have not chosen. In this sense, neutralism is justified on a particular conception of the good and is therefore not neutral in the strong sense. Although this position appears inconsistent, it can be argued that the state must be neutral towards the different beliefs, values and identities people can choose, but just not to autonomous choice itself. In order to defend such a position, it is essential to suggest why autonomy deserves this special treatment, which is why I have hopefully suggested that autonomy has intrinsic value that determines the full value of other things (Sher 1997: 14–15). The weak liberal neutralist position is unsatisfactory because, although it acknowledges coercion as a method of government influence over choices, it is ignorant of the fact that government can influence the options that — 40 —

Cultivating Autonomy: The Normative Core of Democracy

someone would want to select and pursue through manipulation, seduction of preferences and through restricting what options for choice are made available (Sher 1997: 36–7). It seems it is impossible for the state to be even weakly neutral in respect of individuals’ autonomy. The state will always, it seems, coerce, manipulate and seduce, even if it tries not to, as these phenomena can manifest themselves as unintended consequences that alter people’s perception of the merits of the various options available when making a decision, whether taxing, banning, legalising or subsidising (Waldron 1989: 1135; Sher 1997: 66). It does seem likely that a government’s programme will be conducted on non-neutral reasons – that is, what it thinks are the merits of particular forms of behaviour. If it does so, then the state hinders autonomy by deciding, on behalf of its citizens, the merits of the various choices rather than enabling individuals to be autonomous and make these choices for themselves (Waldron 1989). If neutrality is impossible then we must turn back to perfectionism. As with neutrality, there are two types of perfectionism: in type (1) perfectionism, governments can intentionally and actively promote the autonomy of their citizens; in type (2) perfectionism, the government can actively and intentionally promote certain pursuits over others, in addition to promoting autonomy (Wall 1998: 197).13 This is an important distinction in perfectionism. Although they are both perfectionists, Kymlicka (1990) only pragmatically accepts type (2) perfectionism, so is only a type (1) perfectionist at a normative level, as he believes type (2) perfectionism negates autonomy but is unavoidable due to the necessity of collective decisions. In contrast, Raz (1986), Hurka (1993) and Wall (1998) accept both types, believing that that type (2) perfectionism does not infringe upon autonomy: ‘The autonomy principle permits and even requires governments to create morally valuable opportunities and eliminate repugnant ones’ (Raz 1986: 417). This assertion is based upon the argument that, although autonomy is an intrinsic good, it is not the only good. However, it was argued above that, although autonomy is not the only good, the ‘good’ of other things can only be fully realised if they are autonomously chosen. Consequently, as with Kymlicka (1990) and Waldron (1989), I follow the ‘non-discrimination argument’ (Wall 1998: 198) and accept type (1) perfectionism, but reject type (2) perfectionism, as the latter can involve coercion, seduction and manipulation, which, as already discussed, goes against proper respect for autonomy because it promotes certain values above others, without the agent accepting these values. This non-discrimination argument takes us towards weak neutralism, but the belief in the prominence of autonomy for all individuals means that it is not a completely neutral position, albeit still mildly perfectionist. — 41 —

Towards a Deliberative and Associational Democracy

This leads to the pragmatic argument, accepted by Kymlicka, that type (2) perfectionism should be allowed (Kymlicka 1990: 219). If weak neutralism is unachievable, and even type (2) perfectionism inevitable, then collective restriction, which I have suggested is consistent with autonomy and therefore democracy (which can enable collective restriction), seems the only hope for the preservation of autonomy. However, Kymlicka further argues that it is not the role of the government to pursue this collective restriction, but rather should be done through the democratic participation of citizens in civil society (Kymlicka 1990: 219). The argument is based on a presumption (which will be pursued in more detail in Chapter 3), that government officials are too distant from citizens to be able to democratically decide what is of value and what should be promoted in society. Consequently, in liberal democracies, such decisions are being made in a manner incompatible with the cultivation of citizens autonomy. If the government does impose its perception of what is valuable, then this will infringe on the autonomy of the citizens, as they will not be self-determining. In contrast, secondary associations in civil society are much closer to citizens and could therefore provide a more suitable location for governance. Chapters 3 to 6 of this book, then, propose an associational democracy, grounded upon perfectionism, on the basis that collective self-compulsion is legitimate, providing this compulsion is democratic. Taylor argues that certain exclusions are inevitable, claiming that ‘liberalism can’t and shouldn’t claim complete cultural neutrality. Liberalism is also a fighting creed’ (Taylor, cited in Cooke 1997: 280). All political theories have certain doctrines that lie at their heart and cannot be transgressed. Therefore the best defence of democratic rights is not to claim them as neutral, but to say that a framework of rights enhances a conception of the good, and that people should be autonomous in selecting their own aims and goals.14 The position presented here is, then, mildly perfectionist and therefore if we do have a commitment to autonomy for all, we have a responsibility of recognising and incorporating into the decision-making process those who do not see autonomy as important (Cooke 1997: 281). Neutrality is very hard, if not impossible, to achieve, therefore some level of perfectionism is necessary. I have suggested that this is best achieved through democracy, where all have an opportunity to participate in deciding what decisions will be made. This should be based upon the intrinsic good of autonomy, as the full potential of other goods will only be realised if autonomously chosen. However, not all citizens will be interested in autonomy, so it should only be collectively sought in the political sphere.

— 42 —

Cultivating Autonomy: The Normative Core of Democracy

Paternalism Justifying democracy on autonomy, regardless of whether people want to be autonomous, is then mildly perfectionist – but is it also paternalistic? Dworkin defines paternalism as ‘interference with a person’s liberty of action justified by reasons referring exclusively to the welfare, good, happiness, needs, interests, or values of the person being coerced’ (Dworkin 1988: 121). As discussed above, not all people will be (equally) interested in being autonomous, so if it is pursued in the public sphere by providing citizens with the opportunity to participate in collective decision-making, then there could be paternalistic interference, creating a tension with the cultivation of equal autonomy for all. As with perfectionism, we can distinguish between strong and weak paternalism. Strong paternalism occurs when there is ‘intervention to protect or benefit a person, despite that person’s informed and voluntary denial of consent to the paternalistic measures proposed’. Weak paternalism ‘involves interference where there is (or believed to be) a defect in the decision-making capacities of the person interfered with, or where it is necessary to ascertain whether the person’s behaviour is fully reflective’ (Lindley 1986: 64). We must consider both types of paternalism in order to see if they are consistent with autonomy. Strong Paternalism As autonomy involves acting rationally, is it then permissible for some to force others to obey what they see as the laws and outcomes of rationality? If so, autonomy combined with rationality could actually be used to justify totalitarianism and authoritarianism, both forms of political organisation that go completely against the idea of self-government, which is the essence of autonomy. The link between autonomy and democracy would then be broken. It was argued earlier that a key requirement of autonomy is that choices are free in the negative and positive sense. Berlin, with his conception of positive liberty as being led by a rational conception of one’s ‘real’ interests, or one’s ‘higher nature’, demonstrates how autonomy constituted by rationality (as it is here) can justify coercive practices which are the antithesis to negative liberty. If rational desires designate our ‘higher selves’, then irrational impulses are our ‘lower nature’. For example, for Rawls autonomy requires being motivated by ‘higher interests’ such as ‘justice’ and ‘rightness’; to be motivated by ‘selfish interests’ and desires for money, wealth, power, food and drink would lead to heteronomy (Rawls 1993: — 43 —

Towards a Deliberative and Associational Democracy

76).15 For Berlin, such a distinction of interests, combined with the fact that all humans are not equally rational (although it may be supposed that all have the potential to be rational), justifies giving power to the most rational, who are aware of the ‘real’ interests of society, to ‘force the others to be free’. The idea is that if all were fully rational, then they would accept the laws set by the rational elite. We have seen arguments that follow this basis put forward in a variety of guises: by Rousseau, who believed conformity to a ‘general will’ was freedom and that those who did not agree were simply mistaken as to what the general will was; and by Lenin, who argued that the vanguard (supposedly rational and aware of the collective’s ‘real’ interests) led the masses (irrational and unaware of their situation as an oppressed class) to revolution and supposed higher freedom.16 This strong paternalism seems antithetical to autonomy and democracy, as it involves forcing people to conform to arguments that they are against (Berlin 1984: 24). So, are we doomed to ‘rational authoritarianism’ if we consider autonomy or positive liberty as obedience to one’s rational will? A defence against Berlin’s criticism is that all ideas can be misused (Lukes 1973: 56; Connolly 1993: 144). However, this is not a sufficient defence, as this conception of people’s ‘real’ interests or of people having ‘false consciousness’ has been the source of justification for many undemocratic processes, including totalitarianism and limitation of suffrage. The ‘misuse’, however, may be because the idea has innate faults, so a more substantive defence is clearly needed. Just such a defence comes from, Macpherson who realised that the outline of positive liberty presented by Berlin contains three different points (Macpherson 1973: 108): A. ‘The desire . . . to participate in the process by which my life is to be controlled’ (Berlin 1984: 22) B. The ability to be self-directing, to act and decide for oneself on conscious purposes and goals (Berlin 1984: 22–3), which leads to the point that C. These conscious goals should be rational and based on higher interests. This provides the key link to Berlin’s conception of positive liberty, and his related critique of autonomy – that is, that autonomy is conformity with the rational desires of all society. Macpherson argues, correctly I feel, that point ‘B’ (being self directed) is dependent on point ‘A’ (having a share of sovereignty and participating in decisions): ‘Without ‘‘A’’ the man who cannot participate in the making of political decisions is governed by rules made entirely by others i.e. is directed entirely from outside himself, which is inconsistent with ‘‘B’’ ’ — 44 —

Cultivating Autonomy: The Normative Core of Democracy

(Macpherson 1973: 109). As such, the first two parts of Berlin’s concept of positive liberty are mutually supporting and therefore make sense to come as a package. In contrast, ‘C’ (conformity to objective rational standards) is at odds with negative liberty, as it allows for coercion. It is also, for different reasons, at odds with both ‘A’ and ‘B’: the former because it allows elitist rule and the latter because it allows for some people to decide on behalf of others (Macpherson 1973: 110). In which case, ‘C’ does not follow logically from ‘B’, as many have maintained, and is, therefore, not a necessary, nor a desirable, part of the conception of positive liberty, or autonomy. Autonomy can be saved, though, if we specify, as Dworkin does, that the core notion of autonomy is ‘the ability of a person to effectuate his decisions in action’ (Dworkin 1988: 105). Although the process by which these decisions are formed must be conscious and rational, this does not legitimate others deciding for you. As already argued, a key aspect of the connection between rationality and autonomy is that the reasons supplied for motivation and justification are one’s own, have been accepted and not resisted by oneself and that this lack of resistance was based upon self-reflection that was rationally motivated and not due to self-deception. It is still possible that, on further reflection later, these values and decisions will be rejected, but this is why values, beliefs and principles are the person’s own (Kant 1964: 116; Benn 1976: 124; Cooke 1997: 274). The rationality required for autonomy is not some form of objective rationality that can be decided by others for the person, but is the formation of goals for self-direction that are consciously and rationally formed in the light of available information. For aims, goals, preferences, values and desires to be autonomous. the content is largely irrelevant: it is their process of formation which makes them autonomous and, therefore, if the motivation for the desire has come from the person, and the options and their implications have been critically scrutinised, then they will be autonomous. There is no objective content to autonomous preferences because, as Christman explains: We can imagine cases where an agent would have good reason to have such a desire. Hence, we can also imagine that the person is autonomously guided by those good reasons in formulating that desire, and so by token we can imagine it as autonomously formed. (Christman 1991: 23) It is, then, the process by which people form beliefs and make decisions that depicts whether or not they are autonomous, and not the content. There are no ‘higher’ or ‘lower’ interests, as Rawls indicates, only more or less — 45 —

Towards a Deliberative and Associational Democracy

rationally formed goals dependent on the level of consideration and amount of relevant information available. Strong paternalism is therefore not consistent with autonomy, but the concept of rational autonomy does not result in it. Justifying democracy on autonomy is therefore not a case of strong paternalism and would only be so if citizens were forced to participate when they didn’t want to. Weak Paternalism Weak paternalism can also arise from a belief in the inequality of peoples’ rationality. It seems apparent that if autonomy requires rationality, then the more rational a person is, the more autonomous a person will be. Therefore not all people can be autonomous as each other if capacities for rationality are not evenly distributed. If people are not equally autonomous, then this could mean that their rights to self-determination (which I have been arguing derives from autonomy) are not equal either. Aiming to cultivate autonomy could therefore lead to a ‘sliding scale’ which allowed for ‘differing degrees of paternalistic intervention according to the level of competence a person displays in decision-making’ (Christman 1988: 116). If this were the case, then autonomy would be a poor basis on which to justify democracy, as equal power and rights are integral to the idea of democracy. Feinberg (1971), Arneson (1980) and Van De Veer (1986) suggest that any voluntary decision cannot be interfered with, as they believe that a decision can be voluntary and autonomous without being rational, and consequently that it is a violation of autonomy to interfere with irrational choices. I do not take this position, and have argued that rationality is a requirement of autonomy. Therefore, the view considered here is that autonomy is still a constraint, but if decisions made are irrational, and therefore not autonomous, then interference is justifiable. The essential point is that an irrational decision is not an autonomous one, and therefore interfering with it is not a violation of autonomy. For example, Dworkin suggests that forcing people to wear crash helmets on motorbikes or seatbelts in cars is justifiable paternalistic intervention because those who do not wear seatbelts or crash helmets are acting irrationally (Dworkin 1988: 125).17 Similarly, we might consider the desire not to want to be autonomous, and not to want to participate in a democracy, as irrational, and therefore open to justifiable paternalistic interference. Scoccia (1990: 330–1) argues that there are three circumstances in which paternalistic interference does not violate autonomy:

— 46 —

Cultivating Autonomy: The Normative Core of Democracy

I. Failure in economic rationality – the person has autonomous goals but chooses poor means to achieve these goals, and the person would agree to the interference if she were fully rational. II. Need for high autonomy desires – the individual has ‘low autonomy desires’, lacks the capacity to form ‘high autonomous desires’ and the interference will maintain the individual’s potential to form ‘high autonomous desires’ in the future. III. Low autonomous desires – again, the individual has low autonomous desires, the interference would increase the autonomy of their desires and the person would not object to this interference if they were fully rational. According to Scoccia’s analysis, the government is justified in forcing people to wear crash helmets and seatbelts for their own good, as such interference would not detract from autonomy.18 However, this view seems to rely on objective outcomes to rational and autonomous preference formation. As with strong paternalism, weak paternalistic actions impose on the person values that they have not chosen themselves, and which may not reflect their true will thus denying them the opportunity of being fully self-determinate. It is an empirical impossibility to know what interference someone would consent to if ‘they were fully rational’. We only know what information and arguments will influence people, and how it will influence them, when it happens; we cannot presume such things, there are no ideal thinkers. Such interference inevitably involves enforcing the values of one onto another, and therefore violating autonomy because the agent in question has not accepted the value or its process. Scoccia’s justification of paternalism is based upon the assumption that those who do not want to wear seatbelts and crash helmets would change their mind when presented with the facts of the risks they undertake, but what of people who are fully aware of the risks and still choose not to wear the seatbelt and crash helmet? In this sense, this position resorts to saying the person is objectively wrong, which is a case of strong and not weak paternalism (Lindley 1986: 68–70). Weak paternalism is, however, justified, providing its motivation is to ensure that autonomy is still preserved. Just as J. S. Mill (1993) thought we should not allow people to commit themselves to slavery because it limits their future autonomy, it is legitimate to restrict autonomy at a specific time to ensure the possibility of autonomy in the future. The aim of autonomy is to be self-determining throughout one’s life, not just at any single moment (Lindley 1986: 72–3). This applies to individuals and collectives, so, just as it is legitimate to paternalistically restrict an individual from committing themselves to slavery, so it is also legitimate to prevent a collective — 47 —

Towards a Deliberative and Associational Democracy

committing themselves to an authoritarian government. Basing democracy upon autonomy is therefore justifiable paternalism for citizens who have no interest in being autonomous or participating in a democracy because it ensures that they will have the opportunity to participate in the future. Elster quite rightly points out that many things that are mistaken for paternalism are not, but are instead examples of the people (acting as a collective through democratic decisions) binding themselves to avoid certain results occurring in the future. For example, banning cigarette advertising could be an autonomous decision where a collective tries to ensure they will not ‘hear the voices of the Sirens’ and be compelled into taking an action they do not want (Elster 1976: 469–70). Therefore, people can bind themselves autonomously, as this may be done with the intention of bringing about the satisfaction of a different end (Elster 1976: 470). What is important is that the laws binding collectives are democratic, so that people are actually binding themselves and not being bound by others ‘for their own good’, with the former being compatible with autonomy and the latter being strong paternalism. There is still, however, the problem as to whether such legislation is ‘self’-binding or not if the agent has not consented to it, and it is to this issue to which we now turn. Minorities in Collective Decisions The most significant challenge to the link between autonomy and democracy is that democracy relies on majority decisions, meaning those in the minority do not consent and are therefore coerced. It is difficult to see in what sense those in the minority on decisions are autonomous if the majority, with whom they disagree, compels them. There have been many notable attempts to resolve this tension: Rousseau (1968) famously offered the general will as a solution, Wolff (1976) thinks unanimous agreement is the only resolution, and Graham’s (1982) revision of Wolff suggests we should maximise autonomy so that majority decisions are justified. However, all these positions have weaknesses and so, following Gould (1988), I shall argue that a normative conception of authority, combined with a distinction between individual and collective autonomy, provides the most promising route to a resolution. For Rousseau, autonomy could be preserved for all in collective decisions, providing everyone ‘put his person and all his power in common under the supreme direction of the general will’ (Rousseau 1968: 60–1).19 In this conception, autonomy is only ensured when all citizens will the general interest. It is then impossible for autonomous citizens to be in conflict. This is not to say that there will be complete agreement as to what the general — 48 —

Cultivating Autonomy: The Normative Core of Democracy

interest is, but Rousseau presumed that the majority would be correct. Therefore the minority can legitimately be forced to comply with this general will, and still be autonomous, in so far as they are being ruled by laws to which they would have agreed if they had interpreted the general interest correctly. Such an objective view of reason and autonomy goes against the subjective proceduralist approach offered here, though, and could lead the way to paternalism and the erosion of autonomy. Furthermore, Manin criticises the republican view of Rousseau for its lack of recognition of minority opinions because minority opinion is in no way incorporated into the will of the majority (Manin 1987: 360; see also Lessnoff 1986: 82). Consequently, we therefore have to conclude that the autonomy of the minority has not been cultivated. Such weaknesses in Rousseau’s approach led Wolff to consider unanimous direct democracy, the only form of government able to preserve the autonomy of all because only through such a system does ‘every member will freely every law which is actually passed’ (Wolff 1976: 23). As people who are ruled by laws they themselves have willed and consented to are autonomous, autonomy is reconciled with authority. Wolff is therefore against majority rule altogether, because although the autonomy of the majority is maintained, as they have willed and consented to the laws that have been passed, the autonomy of the minority is abandoned because they have not willed or consented to the law but are still coerced to obey it. Wolff’s argument shows the necessary connection between autonomy and democracy, but there are problems with it. In modern plural and diverse societies, reaching unanimous agreement seems too high a demand for authority to be legitimate, as it is unlikely ever to happen (Waldron 1993: 406). Moreover it would lead to the preservation of the status quo and would therefore tend to benefit established interests and dominant social groups (Mansbridge 1996: 48). So if we are to have a realistic conception of legitimate authority, we must abandon Wolff’s idea of unanimous agreement and form a conception that allows collective decisions to be made when not all have agreed. Without this requirement democracy could never be legitimate because it seems there will always be minorities in decisions. Furthermore, the idea that a decision made must have unanimous support from the collective gives some members of the group more influence than others because the single dissenting voter can effectively veto the decision, in that they have equal weight to all the rest of the members put together. If, therefore, the autonomy of all is to be equally respected, and if unanimity is impossible, it is hard to see why the decision should go in favour of the minority (Mansbridge 1980: 260; Gould 1988: 236–7; Barry 1991: 27). In consideration of these weaknesses of Wolff’s argument, Graham (1982) — 49 —

Towards a Deliberative and Associational Democracy

argues that we must aim to maximise autonomy and that majority rule is the best way to achieve this, as we must treat all citizens’ autonomy equally. If this is the case, then Wolff, following in the Kantian tradition, must want to universalise the maximisation of autonomy, and majority rule achieves this (Graham 1982: 133; Barry 1991: 27). In this sense, autonomy is not an absolute or an all-or-nothing affair but comes in degrees, like baldness (Young 1986: 70; Harrison 1993: 171), because one person can make autonomous decisions and choices more often than another, and a person can ‘express more autonomy in her actions’ (Van Hees 2003: 339; see also Young 1986: 51; Levine 1993: 160). We have seen further examples of variables of autonomy that come in degrees throughout this chapter, including choice, rationality and information. For Harrison, majority rule is only compatible with the equal cultivation of autonomy for all if no individuals, or groups, are more likely to be in the minority than others (Harrison 1993: 173). A democracy therefore cannot operate with permanent minorities because they would never be in the majority, and so their autonomy would not be cultivated through any decision. Perhaps, therefore, Gould is right that it is not simply the maximisation of autonomy that justifies majority rule, but equal agency. This means that majority rule should be accepted, providing that all have equal opportunity to participate in making these decisions (Gould 1988: 233). Equal agency should be preserved because ‘no one has more right to effect his or her will in joint decision than has any other and that each one is bound to respect the equal dignity of each of the others’ (Gould 1988: 237). Majority decisions do not, then, just originate from the majority, but from the collective as a whole. Gould argues that people are intentionally related, as common aims require common action, and therefore all are legitimately bound by these common purposes and procedures (Gould 1988: 221). This leads her to propose a normative conception of authority which should not be based upon hierarchical relations in which some people impose authority upon others, as this is incompatible with equal agency of all citizens (Gould 1988: 222). In this sense, individuals act as a collective exercising authority over each other, as a collective, through democratic decision-making. Autonomy is still preserved in such situations because there is a distinction between individual and collective autonomy: ‘Since the form of such common activity (making collective decisions) differs from individual activity, in which one makes decisions about one’s own actions independently of others, the nature of decision-making in common activity must also differ’ (Gould 1988: 89; see also Doyle 1990: 82; Habermas 1996a and Warren 2001a for similar distinctions). Collective autonomy is therefore — 50 —

Cultivating Autonomy: The Normative Core of Democracy

different to individual autonomy and as a result collective decisions must be viewed differently to individual ones. The latter is the formation and pursuit of private goals, the former is a collection of individuals forming and pursuing shared goals. We are not isolated individuals, but live in a society. The belief that collective decisions should be made on the same basis as individual decisions, like Wolff’s, is based on a possessive individualism in which no decision can be made without the consent of all. Collective decisions allow majority rule through a commitment to ‘the ideal that the unity of the whole must, under all circumstances, remain master over the antagonism of convictions and interests’ (Simmel, cited in Mansbridge 1980: 260; see also Gould 1988: 236–7). As Raz suggests, being autonomous is not to completely determine one’s life, as we are inevitably, and justifiably, affected by the aims and decisions of others (Raz 1986: 155–6). In collective action the autonomy of the individual can only be preserved, equally with all, through free participation with all others in the determination of these collective goals and formation of means to achieve them. This can be construed as a sacrifice of autonomy only if autonomy is viewed in the narrow sense of having total sovereignty to pursue specifically personal goals, but this would oppose the distinction between collective and individual decisions. Majority decisions are therefore justified because they ensure that the intrinsic connection between agency and autonomy is preserved because each agent’s autonomy is respected equally. If the distinction between collective and individual decisions is accepted, then the conditions of free choice and rationality required to make an autonomous individual decision will also be different in collective decisions, which are public and political. If autonomy is to be preserved in collective decisions, it is necessary for these individual opinions to be autonomous (Warren 2001a: 65). If democratic decision-making processes derive their legitimacy from maintaining autonomy of the citizens, it seems that a decision-making procedure increases autonomy by allowing more participation from participants, providing them with access to more relevant information and ensuring that the preferences that are incorporated into the decision-making processes are rationally formed. Making decisions as an individual requires deliberation, but collective decision-making is different; accordingly, in the latter case, the process of rationally forming intentions and goals should be done collectively, not privately. This means decisionmaking needs to involve a collective process of rational preference formation and, as with the individual, it is the process that is important for rational preference formation in the collective. Each individual should still form their own preferences, but because of the differences between individual and collective autonomy, and correspondingly, between individual — 51 —

Towards a Deliberative and Associational Democracy

and collective decisions, this should not be done privately. It is these considerations that lead me, in Chapter 2; to specifically argue that deliberative democracy can best cultivate autonomy. Conclusion The aim of this chapter has been to argue that autonomy is the normative core of democracy because there is an intrinsic link between them, as they are both essentially based upon the idea of self-government. This is definitely the case in Western liberal democracies, where much of the theory and practice of liberal democracy has sought the cultivation of autonomy. Autonomy is desirable because it is an intrinsic good for its own sake; it is constitutive of agency, and we deeply value efficacy and having a causal impact on the world, in terms of realising some alternatives and preventing others. Although there are other goods, their maximum value can only be achieved if autonomously chosen. I think this is a universal claim, but it is thought that to be a successful member of a liberal democracy the social norms require people to be autonomous, as the concept is embedded in Western institutions. Therefore, although democracy promotes many other values, it is because it promotes autonomy that these other values can be fully realised. Other values are, then, only contingently related to democracy, whereas autonomy can be secured by democracy alone. However, we cannot assume that citizens are autonomous, as we require the right conditions to cultivate autonomy, including an appropriate decision-making process. It is, therefore, the decision-making process, and what criteria it should meet in order to cultivate autonomy, that concerns me here. In order for autonomy to be cultivated in decision-making, there are two broad, but essential, requirements. The first is free choice, with freedom being interpreted in both the positive and negative sense. Decisions are negatively free if they are not coerced and positively free if the preferences on which decisions are based are not seduced or manipulated. This latter requirement is very difficult to secure, as, in accordance with Raz, it has been argued that people can later autonomously accept preferences they were initially seduced and manipulated into taking. Instances of manipulation and seduction are therefore very difficult to identify, but they nevertheless still detract from autonomous decisions being made. For choices to be free there must also be an acceptable range of options, and the decisionmakers must have knowledge of these options. Autonomy therefore requires that the outcomes of decisions are intended. This requires deliberation and can lead to Hurka’s ‘deliberative — 52 —

Cultivating Autonomy: The Normative Core of Democracy

autonomy’. The conception of ‘deliberative autonomy’ offered here navigates a position between Kant and Hume, in suggesting that autonomous decisions can be based upon desire, but these desires can be formed through rational deliberation. This conjures up the danger of an infinite regress, as preferences are used to review preferences, and we do not know if these first preferences were autonomous. We can, however, avoid this regress by emphasising that it is the process that is important, and providing the agent accepts the process of preference change and formation, the preferences are autonomous. Conscious acceptance involves a rationally motivated, self-reflected lack of resistance. These preferences must be consistent with each other and be based on the deliberation of relevant and available information, although this is inevitably contingent and limited. Deliberations should be focused on important decisions, which includes collective decisions, but decisions only need to be based on reasons that are strong enough. The fact that some people will have more rational preferences does not justify them a privileged position in collective decision-making; rather, the key is to make all autonomous to a certain threshold. The communitarian critique suggests that free choice, and therefore autonomy, is not possible because individuals are determined by their environment and are therefore not ‘choosing’, meaning they are not selfdetermining. However, autonomy does not require isolated individuals who determine everything about themselves, as that is a requirement that only a God can meet. Our environment may limit our choice, but it does not determine it entirely. Individuals are therefore not unencumbered individuals, separate from ends per se, but just separate from many particular ends, over which choice is possible. Cultural resources do not simply confine choice, they are also a resource that makes choice possible. In addition, as actors, we influence and change the environment, and if this action is determined by democratic decisions, then it can be intended and therefore compatible with autonomy. A further criticism is that justifying democracy on autonomy is perfectionist, as it disadvantages those whose lifestyles are incompatible with autonomy. However, it was argued that liberal democracy is a ‘fighting creed’ and cannot, and should not, be neutral to all lifestyles. The arguments here are therefore perfectionist, but only mildly so, as it is a procedural view of autonomy that I favour, with the formation process, and not the content, of preferences determining autonomy. Nevertheless, it means it is legitimate to justify democracy on autonomy and to incorporate those who do not see autonomy as important into decision-making. The procedural interpretation of autonomy offered here also means that, despite — 53 —

Towards a Deliberative and Associational Democracy

autonomy relying on rationality, it does not lead to strong paternalism, as the reasons are subjective and must be important to that individual. Weak paternalism is only justified if it aims to preserve autonomy, and therefore basing democracy upon autonomy is justifiable paternalism. Even if collective decisions are made democratically, the fact that consensus on decisions seems impossible, especially in modern and increasingly diverse and pluralistic societies, means there are minorities who are compelled to obey decisions to which they have not consented. The solution is to take a normative conception of authority and ensure equal agency. Majority collective decisions can achieve this, providing we do not confuse autonomy with sovereignty. Collective decisions are different to individual decisions, as the decisions affect others. If individual decisions require private deliberation to be autonomous, then collective decisions require collective deliberation, which includes the sharing of information and reasons through public debate: Majority rule is as foolish as its critics charge it with being. But it is never merely majority rule. The counting of heads compels prior recourse to methods of discussion, consultation and persuasion. The essential need in other words, is the improvement of the methods and conditions of debate, discussion and persuasion. This is the problem of the public. (Dewey 1988: 144) It is thought that deliberative democracy can help solve this ‘problem of the public’ and, following this argument, Chapter 2 will review two different democratic decision-making models – deliberative democracy and aggregation of preferences – from the point of view of their potential to cultivate the autonomy of all participants in decision-making equally. Notes 1. There is a further issue as to whether it is necessary to experience autonomy in order to value it. This is essentially a debate between objectivism and subjectivism which I do not think is necessary for me to resolve. However, for a discussion of these positions see Lindley (1986: 28–9), for an account of the subjectivist proposal see Von Wright (1963) and for a defence of the objectivist position see Brandt (1959). 2. As will be discussed later, even if Mozart was initially forced into music, he could still have later in life autonomously chosen to continue to dedicate his life in that direction. 3. Again, people could initially be forced to listen to Mozart’s music, but later autonomously choose to listen to it.

— 54 —

Cultivating Autonomy: The Normative Core of Democracy 4. Christman (1988: 120) points out that a sophisticated utilitarian might argue that happiness is satisfaction of autonomously formed goals, which exempts them from heteronomous criticisms, of a manipulated feeling of happiness. Just such an approach is employed by J. S. Mill (1993). 5. Clarke argues that Raz does not place sufficient emphasis on the enabling conditions of autonomy. It is certainly the case that there is a gap between having the formal rights and conditions free from interference to make autonomous decisions, and being able to actually make use of these opportunities and rights. Clarke attributes the concentration on formal autonomy within the liberal tradition to the disconnection between autonomy and selfrealisation identified in Raz. He uses an example of a person being free to choose to live in a cardboard box just as a person is free to choose to live in a palace. There are no formal constraints to either, but this ignores the lack of substance in the freedom because the homeless person who inhabits the box is substantially unable to live in a palace (Clarke 1999: 221). This point raised by Clarke is an important one, but this criticism cannot be directed at Raz, as he demands an acceptable range of options, and if one has no resources and is forced to choose to live in a cardboard box, then it seems apparent that there is not an acceptable range of options from which to choose. 6. MacCallum feels that it is misleading to talk of positive and negative freedom, because the distinction between ‘freedom from . . .’ and ‘freedom to . . .’ is not a clear or useful one. MacCallum argues that every instance of freedom consists of a triadic relationship between the agent, the absence of constraining conditions and the ability to become something (MacCallum 1967: 314). Gould realises MacCallum’s joining of the enabling conditions would be classified as ‘what one is free from the absence or lack of’, and this is ‘an oblique and strained way of referring to the positive conditions’. Furthermore, this definition of freedom makes no distinction between lack of interference from others and the provision of enabling conditions: ‘To ignore this distinction would be to conflate forbearance from action with action’ (Gould 1988: 381). We must, then, either accept Berlin’s argument that the two conceptions of liberty are in conflict or find an alternative method to MacCallum’s of linking them. 7. Hume did not explicitly refer to autonomy but rather to liberty, which for him was the ability to choose and act upon that choice. Young argues that we can substitute liberty for autonomy to have a Humean view of autonomy (Young 1986: 33). 8. In reality there are not just second order preferences: you can have a third order preference about a second order preference and so on, ad infinitum. However, we can take Young’s view that autonomy is determined by the highest order preference (Young 1986: 66). 9. This agency of individuals is often collective, with individuals acting together to achieve common aims. Gould suggests that common activity is ontologically distinct to individual action, as it cannot be explained by reference to

— 55 —

Towards a Deliberative and Associational Democracy

10.

11. 12. 13. 14.

15.

16.

17.

individual action by itself. She further warns that this activity may not be ‘common’ at all, but simply appear so because of subordination. Nevertheless, the important point remains that the collective can be a constituted entity, providing the subordination does not eliminate the capacity to change these relations through choice and action (Gould 1988: 108). It is important to note the danger of circularity in this argument. In order to break this circle, we need some more criteria about how to distinguish between those who ‘uncritically internalise the received mores’ and the autonomous person. Otherwise the autonomous person may not be autonomous at all, as ‘the critical canons derived from another tradition’ may have been uncritically internalised. However, it was argued above that this problem is avoided, providing the process of preference formation is accepted. It is only mild because of the proceduralist conception of autonomy presented here, where the reasons for decisions are subjective. However, D’Entre`ves (2002) does accept that no model of democracy can be completely inclusive. D’Entre`ves (2002) fails to distinguish between these two types of perfectionism and lumps both Kymlicka and Raz under the same perfectionist label. D’Entre`ves claims that this fact should not stop us seeking the most inclusive model of democracy possible (D’Entre` ves 2002: 43). This model is claimed to be the deliberative model of democracy, a point with which I am in complete agreement to the extent that this book is centred around this model of democracy. However, D’Entre`ves claims this is not a perfectionist model of democracy, as it contains no view of the ‘good life’ (D’Entre` ves 2002: 46). Deliberative democracy does, though, seem to favour certain aspects of the good life, such as political participation and rationality, which are surely not compatible with all conceptions of the good life. They are also key aspects of what it is to be autonomous, and in Chapter 2 I will argue that deliberative democracy is the model of democracy most likely to equally cultivate the autonomy of its citizens. In this sense, even if deliberative democracy is not justified upon the value of autonomy, it still seems to suffer from at least a mild perfectionism as D’Entre`ves has conceived it. However, I accept D’Entre`ves’ point that justifying democracy on autonomy is perfectionist, albeit a mild, liberal perfectionism. This point is undermined by Rawls’s inclusion of food and drink as selfish interests. We can only presume he meant consumption of these in the excess. Nevertheless, it is the credibility of the idea of ‘higher’ and ‘selfish’ interests that we consider here, not what higher or lower interests include. Levine (1993: 161) notes that Berlin does not advance this view on Lenin, but comes ‘perilously close’. However, this view can be found in Talmon (1952) and Popper (1962). Dworkin does not mention whether Sikhs could autonomously decide not to wear crash helmets. It seems apparent that such an action would be autonomous, as they could not wear a helmet and a turban, which is an important

— 56 —

Cultivating Autonomy: The Normative Core of Democracy part of their religious beliefs. However, Dworkin does suggest that Jehovah’s Witnesses should not be made to have a blood transfusion because this is based on evaluative differences, so Dworkin might well extend this view to Sikhs. 18. Crash helmets for motorbike riders is a better example, as it has been proven that by not wearing a seatbelt one is more likely to hurt others involved in an accident, especially fellow passengers. Therefore the government would be justified in forcing people to wear seatbelts to ensure the safety of others and protect their freedom, and so it is not necessarily an example of paternalism. 19. It is essential to appreciate the difference between the ‘general will’ and the ‘will of all’. The general will is a judgement upon the common good, while the will of all is the aggregation of private interests (Rousseau 1968: 72–5).

— 57 —

2

Deliberative Democracy and Autonomous Decision-Making

Introduction In Chapter 1 it was argued that autonomy requires democratic decisionmaking. However, it was also suggested that we could not assume citizens are autonomous and, therefore, the right conditions to ensure autonomous preferences enter collective decisions need to be provided. Furthermore, the model of decision-making used will, to a large extent, determine these conditions and there are various forms of decision-making that claim to be democratic. The claim here is that they are not all as democratic as each other, nor as effective at cultivating the autonomy of participants as each other. It is my contention that deliberative democracy, as a form of decision-making, especially promotes the aspects of democracy that cultivate autonomy, as it is the model most likely to meet the conditions of collective decision-making, required to cultivate autonomy, outlined in the previous chapter. This chapter will demonstrate how deliberative democracy will enhance the autonomy of the participants by comparing it to purely aggregative models of decision-making. In Chapter 1, having established autonomy as the normative core of democracy and then outlined the conditions for making autonomous collective decisions, we concluded with the argument that individual decisions must be conceived differently to collective decisions. In collective decisions our choices will affect others and, therefore, must be justified to others in order to respect their agency and autonomy. Individual decisions require rational deliberation to form intentions and, consequently, collective decisions require collective deliberation to form collective intentions. Unsurprisingly, this takes us towards the deliberatively democratic model of decision-making. For example, Elster (1997) distinguishes between decision-making that is suitable in the market, where it is assumed individual’s preferences will not — 58 —

Deliberative Democracy and Autonomous Decision-Making

affect others, and in the forum, where they will, and so should be publicly justified. The link between deliberative democracy and autonomy is not a new one (Cohen 1989: 25; McCarthy 1994: 48; Habermas 1996a: 118–31; Warren, 2001a; Richardson, 2002; Christman, 2005). The justification is prudential, though, as it is suggested that deliberative democracy enables preferences of the participants to become more autonomous by overcoming inequalities in information and rationality (Festenstein 2002: 103). However, these arguments tend to be implicit, whereas the relationship between deliberative democracy and autonomy needs to be expressed explicitly and in more detail. Therefore, three reasons will be considered as to why deliberative democracy can cultivate autonomy. The first is that it encourages participants to make their reasoning public, which encourages preferences to be public and also launders purely self-interested preferences; this means the autonomy of more people is respected in the preferences of participants and in the resulting collective decisions. The second and third arguments are based upon the claim that, in a deliberative situation, people communicate through language. This means that at any point there are speakers and listeners involved. It is contended that deliberative democracy enhances hearer autonomy by increasing the availability of information, which means that the rational external requirements of autonomy discussed in Chapter 1, are improved. Speaker autonomy is also enhanced because the ideal of deliberative democracy ensures all opinions are included and heard by all. When aggregation does take place, choices are determined by citizens themselves and, furthermore, are based upon reflective preferences. We must also consider three critiques that seriously challenge the connection made between deliberative democracy and autonomy. The first of these is the social choice theory critique, which argues that deliberative democracy cannot avoid the problems of cycling, ambiguity and strategic manipulation that undermine citizens’ autonomy. The next challenge comes from ‘difference democrats’, who claim deliberative democracy will fail to enhance the autonomy of subordinate groups, in so far as the norms of this model – reason, debate and impartiality – are not universalisable but are instead culturally specific and act in favour of dominant groups. The final challenge to be considered here is that deliberative democracy, and its autonomy-enhancing processes, are dependent on citizens abiding by ‘democratic obligations’, something that we cannot ensure without civic virtue. First, though, this chapter will define deliberative democracy and its two strands, briefly considering why justifying deliberative democracy on the cultivation of autonomy is superior to — 59 —

Towards a Deliberative and Associational Democracy

other dominant justifications, looking at the proceduralist and epistemic arguments, and establishing it as a model of democracy. Defining Deliberative Democracy As with democracy itself, deliberative democracy is an essentially contested concept (Elstub 2006a). However, a vague and loose core can still be identified, including the democratic strand and the deliberative strand. The democratic part is collective decision-making through the participation of all relevant actors. When interpreting the definition of democracy, a key problem is deciding what kind of participation is envisaged for the people. For deliberative democrats the answer lies in the deliberative strand in which participation should be based on the give-and-take of rational arguments (Elster 1998a: 8). In this context, a reason is ‘a consideration that counts in favour of something: in particular, a belief, or action’ (Cohen 1998: 194). The deliberative strand can therefore be described as ‘a dialogical process of exchanging reasons for the purpose of resolving problematic situations that cannot be settled without interpersonal co-ordination and co-operation’ (Bohman 1996: 27). Deliberative theorists believe that preferences will adapt to reason, conceiving preferences as being exogenous: formed during the political process, rather than prior to it. Through consideration of differing reasons, existing preferences can be transformed and new preferences formed. Elster considers preference transformation to be the defining mark of deliberative democracy: ‘The transformation of preferences through rational deliberation is the ostensible goal of arguing’ (Elster 1998a: 6). Therefore, in order for deliberation to occur, ‘reflection upon preferences in non-coercive fashion’ is required. This deliberation is democratic if these reflective preferences influence collective decisions and all have had an opportunity to deliberate equally (Dryzek 2000: 2). Such considerations have led Lindsay to argue that: Government by consent, if taken strictly, is and must be an illusion; that it is an entire mistake to suppose that there exists at any moment a ready made will of the people. The process of discovering what may be called the will of the society is a process of making it, and to that process discussion is essential. (Lindsay 1929: 430; see also Manin 1987: 352–3) This then is congruent with the conception of autonomy outlined previously, where it was the process, and an acceptance of the process, of — 60 —

Deliberative Democracy and Autonomous Decision-Making

preference and opinion formation that makes them autonomous. Therefore, deliberative democracy seems to have a general core, which includes: . . . .

the making of collective decisions involving the participation of relevant actors (the more equal this participation the more democratic) through the consideration and exchange of reasons aimed at the trans(formation) of preferences (Elstub 2006a: 303)

This ideal of deliberative democracy is perhaps best represented in Habermas’ counterfactual procedures termed ‘the ideal speech situation’ (ISS). Here communication is undistorted, as all participants are free and equal, with no power discrepancies and unconstrained from subjection, selfdelusion and strategic activity. All views are aired in an unlimited discourse, creating open participation aimed at rational consensus in which the ‘unforced force of the better argument’ is decisive (Habermas 1990: 56–8). Justifications of Deliberative Democracy There are three prominent justifications of deliberative democracy: the epistemic justification, the fair procedure justification and the prudential justification. These justifications are often cited to justify democracy in general, but the deliberative theorists make the case that deliberative democracy can promote these political values to a greater extent than other models of democracy. Consequently, the justifications focus on what deliberation contributes to democracy (Elstub 2006a: 303). In Chapter 1, autonomy was advocated as the normative core of democracy; this chapter will therefore promote the prudentialist justification, arguing that deliberative democracy is the model of democracy most suited to the cultivation of the equal autonomy of all. The superiority of the prudential justification has already been established, as it is argued that autonomy is an intrinsic value because its agency releases the full value of other goods. However, a comparison with other justifications is still worthwhile, to highlight this even further. The epistemic justification suggests deliberative democracy is good because it is the best method of producing good decisions. If another theory of democracy, or even an undemocratic method, were more reliable, then deliberative democracy would be unnecessary (Estlund 1997: 183; Elstub 2006a: 304). But the argument is that deliberative democracy is the most reliable method because, by generating public reason, it can lead to — 61 —

Towards a Deliberative and Associational Democracy

decisions that are true, well justified or commensurate with justice, needs or the common good (Bohman 1998: 403; Festenstein 2002: 99; Pellizzoni 2001; Warren 2002: 192; Elstub 2006a: 304). However, it is not clear how we know that deliberative democracy does produce decisions that promote the common good. If we could test this justification, it would mean that there is another method for identifying the common good and therefore deliberative democracy would not be required. It is also reliant upon there being a ‘real truth’ about the common good (Festenstein 2002: 99–100), something that is contested by many, including deliberative theorists (Cohen, 1989). It was also disputed here in Chapter 1, where it was argued that that there was no set ordering of values and goods that all should accept objectively, except for autonomy. A proceduralist view of autonomy was then offered, with individuals finding their own reasons for what they think has value. Those who accepted the proceduralist approach to autonomy must, therefore, reject the epistemic justification of deliberative democracy. The rejection of the epistemic justification of deliberative democracy has led many towards the proceduralist justification of deliberative democracy. This justification is based on the idea that there is no external good by which to judge decisions and therefore it is fair procedures that enable conflicts over the common good to be debated and resolved legitimately (Warren 2002: 190; Elstub 2006a: 304). The resulting decisions in deliberative democracy will be ‘just’ because they are derived from fair procedures in which all have been able to participate equally, regardless of what actual decision is reached (Cohen 1989; Elstub 2006a: 305). Such a justification contrasts with the prudential justification, and its mild perfectionism, as it favours complete pluralism in terms of values and is based on the established liberal premise (discussed in Chapter 1), that the state should be neutral towards conceptions of the good life (Nozick 1974; Dworkin 1985; Larmore 1987; Rawls 1993). It was argued that the only conceivable conception of neutrality was weak neutralism, meaning that the reason why a government may act should be neutral and therefore that reasons that are not neutral must be ruled out of democratic debate (Larmore 1987: 44). It is suggested by the proceduralists that deliberative democracy achieves this through publicity, encouraging all to offer reasons that all could potentially accept (Warren 2002: 190). It is evident that all can participate equally in purely aggregative decisionmaking methods, so the proceduralists must make the further claim that public reason increases the fairness of the procedure by encouraging participants to consider the preferences of others, and that this improves ‘the quality of preferences, opinions and reasons’. This then takes the proceduralists towards either the prudentialist justification, if the prefer— 62 —

Deliberative Democracy and Autonomous Decision-Making

ences are better informed due to these procedures, or the epistemic approach, if the claim is that these preferences are likely to be correct due to these procedures (Festenstein 2002: 102–3; Warren 2002: 193; Elstub 2006a: 305). As a consistent proceduralist should not be willing to accept that there are objectively correct decisions independent of the process of making decisions, proceduralism then must move towards the prudentialist justification to save itself from complete relativism. The proceduralist justification is excessively relativist in so far as it fails to account for why the decisions that result from the ideal procedures of deliberative democracy are ‘correct’ and based upon ‘good’ or ‘compelling’ reasons (Estlund 1997: 197). Without good reasons it is not apparent why the decision that has been produced by deliberatively democratic procedures should be selected over any of the other available options. In terms of procedure, it would be just as fair to select an option randomly by a coin toss or through a vote (Estlund 1997: 178; Festenstein 2002: 103; Elstub 2006a: 305). To avoid over-relativism, then, proceduralists only have to accept a mild perfectionism based on the value of personal autonomy, which means each individual should be able to freely choose preferences, values, goods and beliefs for themselves through rational deliberation. Perfectionism, based on the value of autonomy, therefore supports proceduralism as it allows a subjective view of the good – except for the fact this subjectivism must be grounded in autonomy and based on good reasons – and therefore does not suffer from the relativism of the purely procedural justification. This chapter will now consider, in detail, how deliberative democracy cultivates autonomy by comparing it with a purely aggregative decisionmaking mechanism. However, before we do that, we must defend the distinction between deliberative and aggregative models of democracy, in order to prove that this comparison is worthwhile. A Deliberative Model of Democracy? Saward (2000) questions whether deliberative democracy really is a distinct model of democracy, suggesting that deliberative democrats excessively elevate deliberation above voting as a form of legitimate democratic participation, as deliberation can only ever be but a part of democracy, because consensus cannot be reached. Therefore, deliberative democracy can never be a self-sufficient model because voting is still required to make a decision, a factor that will be discussed in more detail below. For Saward this means deliberative and aggregative models of democracy are not separate and distinct models of democracy, but are instead key aspects of any model of democracy (Saward 2000: 67–8; see also Squires 2002: 133–4). — 63 —

Towards a Deliberative and Associational Democracy

This argument depends, however, upon how one defines a ‘model of democracy’. Macpherson suggests that a ‘model’ should explain structural relations and have a normative element which offers a ‘model of man’ and an ‘ethically justificatory theory’ (Macpherson 1977: 2–6). With regard to explaining structural relations, public reason is absent in a purely aggregative model and, consequently, a deliberative model would produce differing structural relations and requires different forms of participation due to the differences between public and private deliberation. With respect to having a distinctive explanatory or normative approach, many other democrats and social choice theorists share the prevalent liberal conception that views the source of legitimacy as citizens’ predetermined preferences, perceiving these preferences to be endogenous and unchanging (Arrow 1963; Riker 1982). In stark contrast a deliberative model sees the formation of these preferences as the source of legitimacy, and thus perceive preferences as exogenous. These premises therefore produce differing normative and empirical claims. Furthermore, deliberation and aggregation are not elements present in all conceptions of democracy: Habermas (1996b) has suggested, in the past, that collective deliberation could lead to consensus and William Riker (1982: 5) and Rousseau (1968) have perceived democratic arrangements purely dependent on voting without any collective deliberation. Consequently, a purely aggregative model of democracy is not a mythical construct set up as a straw man by deliberative democrats. Nevertheless, Saward has further suggested that deliberative democrats have ‘overdrawn’ the distinction between deliberative and aggregative models of democracy, because citizens can deliberate in private prior to voting (Saward 2000: 68). However, the deliberative strand of deliberative democracy is different from individual deliberation. Individual deliberation is structurally different as it contains no dialogue, no give and take of reasons, and no influence between actors (Rawls 1993: 27). The aim of deliberative democracy is to achieve goals, provide solutions and resolve conflicts collectively through the expression of a plurality of opinions and interests (Bohman 1996: 55), a distinction that will be explored in more detail below. It is true that liberal democracies do not presently approximate the aggregative model of democracy, in so far as collective deliberation does occur in certain circumstances, such as through the media, in legislative assemblies and within civil society. Nevertheless, this does not mean the aggregative model does not exist as a theoretical construct. Neither does it rule out the deliberative model being a model because, as Macpherson realised, new models develop as a critique of previous models, and are suggested as a ‘corrective’ or ‘replacement’. These critiques need be upon only part of the preceding model and can therefore — 64 —

Deliberative Democracy and Autonomous Decision-Making

embody ‘substantial elements of an earlier’ model (Macpherson 1977: 8). But even if it is not accepted that deliberative democracy is a model of democracy, in as much as some form of deliberation and voting exist in many conceptions of democracy, which means they are not separate decision-making entities in the sense that deliberation cannot replace aggregation and that aggregation can be preceded by deliberation, they are still different forms of participation. Therefore it still can be useful, meaningful and enlightening to highlight the empirical and normative differences between these models in relation to democratic forms and structural relations. Although both occur in current liberal democracies, my claim in later chapters will be that there is still insufficient deliberation and that, moreover, the deliberation that does occur is not insufficiently democratic, does not lead to collective decisions and excludes many relevant voices. Deliberative Democracy and Autonomy Now we have a clearer understanding of what deliberative democracy is, and of the problems of its other justifications, we can now explore in detail the ways that deliberative democracy will cultivate the autonomy of its participants. We start by discarding the idea that deliberative democracy can or should lead to consensus, and argue instead that it can help lead to a compromise which means decisions can still cultivate the autonomy of all. It is further argued that deliberative democracy encourages the use of public reason, as preferences must be justified to all, an activity that encourages citizens to consider the opinions and interests of others. It is then claimed that deliberative democracy cultivates both hearer and speaker autonomy by increasing the availability of relevant information and allowing participants to express themselves freely. These arguments are all based upon the premise that preferences can be transformed and become more rational when reflected upon, which was earlier hailed as the defining mark of deliberative democracy (Elster 1998a: 6). Consensus In the last chapter, we learnt from Wolff that if consensus could be reached in collective decision-making, then the autonomy of all would be cultivated, as all would have consented to the laws that bind them (Wolff 1976: 23). We also learnt that, due to pluralism, consensus is unlikely to be achieved through aggregation of individuals’ privately formed preferences (Waldron 1993: 406). However, due to the potential of deliberative democracy to — 65 —

Towards a Deliberative and Associational Democracy

generate public reason, with participants trying to find reasons that are convincing to all, Cohen (1989: 23) and Habermas (1996a: 17–19) believe that a consensus would eventually be achieved in ideal deliberatively democratic decision-making. They suggest that public reason would mean people taking on board a common interest over and above their private or selfish interests, as preferences must be based on the reasons why a proposal will be good for all, which encourages people to identify with each other, and with the collective as a whole. Options based on bad, unconvincing and selfish reasons would therefore be discarded. If deliberation continued long enough, all would come to agree on the same common interest. There are, however, a number of reasons to suggest that consensus would not be achieved (Elstub 2006a: 308). Despite the fact that deliberative democracy can generate public reason, consensus may not be achieved due to the pluralism of ultimate values (Elster 1989; Christiano 1997; Weale 2000) and interests (Benhabib 1996). This factor is magnified in modern cosmopolitan and multicultural societies, where there is a mixture of cultures (Femia 1996: 378; Christiano 1997: 249; Elstub 2006a: 308). This is further exacerbated by the fact that time for deliberation is scarce and by deliberative democracies’ aim to actively engage previously excluded groups. A key democratic requirement of the ideal of deliberative democracy is that all should be included in deliberation, but more participants can lead to more opinions, making agreement harder to achieve, especially if some of these are previously unheard (Knight and Johnson 1994: 289). Debate can also increase disagreement as well as reduce it. A collective could easily have a general agreement on some issue, but a debate could generate a greater diversity of opinions on an issue as it is explored more extensively and deeply (Mansbridge 1980: 65; Knight and Johnson 1994: 289; Fearon 1998: 57; Christiano 1997: 249; Shapiro 1999: 31; Budge 2000: 203; Hibbing and TheissMorse 2002: 195; Elstub 2006a: 308). It therefore seems unlikely that moral consensus will be achieved, where ends are agreed on; however, a political consensus on means could still be achieved where there is mutual acceptance and understanding of differing opinions (Rawls 1993: Chapter 4; Dryzek 1990: 16–17, 42–3; Sunstein 1996: Chapter 2; Cohen 1998: 189–91; Richardson 2002: 173), an agreement which can be quite fruitful as ‘incommensurable moral positions’ can often find support for the same policies (Warren 2002: 188). Therefore the most important result is to have what Bohman describes as ‘plural agreement’, where citizens continue to cooperate and compromise through deliberative democratic debate despite the existence of disagreements (Bohman 1996: 34, 89). Consensus may not even be desirable. We heard in Chapter 1 how it can — 66 —

Deliberative Democracy and Autonomous Decision-Making

provide minorities with unequal power in comparison to those in the majority, but a call for consensus can also disadvantage minority social groups too, sacrificing their autonomy and hindering deliberation. The agonistic branch of deliberative democracy is not concerned by differences persisting, but rather praises differences as an essential resource for democratic deliberation, without which the deliberative process would be redundant (Young 1996: 127; Elstub 2006a: 308). Agonistics reject the idea that consensus on the common good is the sole aim of deliberation; they fear that the ‘common good’ might not be common at all, and instead be simply a perpetuation of inequality, and that consensus might be achieved due to participants feeling pressurised to conform and acquiesce to power rather than being rationally motivated (Mansbridge 1980: 32; Gould 1988: 18; Young 1996: 126; Gambetta 1998: 21). Such a consensus would not cultivate autonomy equally for all. Furthermore, it is suggested that dominant social and economic groups are at an advantage because they can put forward their preferences and opinions as ‘authoritative knowledge’, presenting their interests as neutral and in the process devaluing those with alternative beliefs, preferences and interests (Young 1997: 399; Elstub 2006a: 308), meaning the autonomy of only these dominant groups would be cultivated. If a consensus is not required or sought, then continued disagreement would not be discouraged and therefore less pressure would be exerted on subordinate groups to conform and sacrifice their autonomy. If consensus is not reached, then compromise achieved under deliberatively democratic conditions might be the best alternative (Dryzek 1990: 16–17; Festenstein 2002: 92–5; Richardson 2002: Chapter 11; Warren, 2002: 185).1 Deliberative democracy helps make compromise easier to achieve because, although it is unlikely to result in consensus, it improves understanding of the positions of others, which can in turn lead to respect and empathy for these preferences (Warren 2002: 184). Richardson distinguishes between bare compromise, where changes in means are made but not in ends, and deep compromise, where change in means and ends occurs (Richardson 2002: 146–8). Both are made possible through deliberative democracy, but deep compromise is seen as more legitimate by all, as it is supported by reasons, rather than the ‘threat or advantage of the other side’ (Richardson 2002: 152). Such a compromise is likely to occur only through public deliberation, where preferences are changed in light of the reasons and information presented by other participants towards finding ‘commonly acceptable and provisional terms to solve the problem of conflicting practical commitments’ (Festenstein 2005: 127). In Chapter 1 I argued that the autonomy of minorities in collective decisions could still be cultivated if their agency was equally respected. Deliberative democracy, by ensuring the — 67 —

Towards a Deliberative and Associational Democracy

reasons of all are heard, and through making decisions that are commonly acceptable, can still then preserve the autonomy of minorities, even when compromise is required. Deep compromise therefore helps cultivate autonomy, but is also made difficult to achieve due to the pluralism which rendered consensus impossible (Richardson 2002: 156). Furthermore, with compromise the autonomy of all is not cultivated equally, in so far as some may enter deep compromise whilst others only bare compromise (Richardson 2002: 162). It seems, then, that consensus is not possible and perhaps not even desirable. Compromise achieved under deliberatively democratic conditions is both more plausible and desirable. Nevertheless, pluralism will mean that compromise is not possible in all circumstances and therefore it is apparent that in order for decisions to be made deliberation can only ever support the aggregation of preferences and not replace it altogether (Johnson 1998: 177; Przeworski 1998: 142; Dryzek 2000: 38; Saward 2000; O’Flynn 2006: 101). Yet voting, to make a decision, makes a vital contribution to the deliberation that precedes it. It is the fact that participants will have a vote that makes deliberation necessary and desirable. The fact that other participants have a vote is a motivation to persuade and to listen to them (Goodin 1992: 134; Christiano 1997: 251; Warren 2002: 181). Furthermore, it has been argued, as in Chapter 1, that majority decisions are compatible with the equal cultivation of autonomy for all because they preserve the equal agency of all, and because individual decisions must be viewed differently to collective decisions, we should not confuse autonomy with sovereignty. It does mean, though, that we must find reasons other than consensus as to why deliberative democracy helps cultivate autonomy to a greater extent than pure aggregation. Public Reason In Chapter 1, the distinction between private and public decisions was made. It was suggested that because collective decisions are different to individual decisions, autonomy in these two contexts cannot be viewed in the same way. For the autonomy of all to be cultivated, it was argued that equal agency of all must be preserved, and that this can be done by allowing all to participate in decision-making precisely because it is collective and will affect all. It was further argued that for a preference to be autonomous, the agent must accept the process of its formation, and this acceptance must be based upon good reasons. This process also involves the consideration of alternative opinions and their justificatory reason (Mill 1993). This is even more imperative when making collective decisions because the decisions will — 68 —

Deliberative Democracy and Autonomous Decision-Making

affect all; so all should hear the reasons for the preferences of others and have the opportunity to convince others about their preferences. If the distinction between public and private decisions is accepted, it seems logical that if individual decisions need to be based on a critical reflection of the available information, then collective decisions need this critical reflection to be public. For Kant, equal autonomy means achieving reciprocity through practising ‘respectful understanding’ of the opinions of others. Participants in a collective decision must also credit moral weight to the ability of other participants to deliberate. This is what Christman terms ‘emphatic respect’ for the differences of others (Christman 2005: 342). The ‘rule of publicity’ was a cornerstone of Kant’s philosophy and ‘transcendental formula of public law’, which stated that to be right an action that relates to the rights of others must be compatible with being made public (Kant 1957: 129; Elstub 2006a: 305). By this maxim, collective decisions must be made public. If citizens are to rule themselves and make their own collective decisions, then they must respect the autonomy of others, therefore we must seriously consider the reasons of all other citizens that the decision will affect (Bohman 1996: 4; Warren 2001a: 64; Richardson 2002: 78). In addition, respect for the autonomy of others further requires us to consider them when we offer reasons, so that they are sensitive to the reasons and needs of all and that others may reasonably accept them (Richardson 2002: 80–1). Both these requirements can only be met if we reason together (Richardson 2002: 83), as public reason encourages citizens to find reasons for arrangements that will not ‘neglect the good of others’ (Cohen 1998: 197). In these senses public reason recognises citizens as being full members in the sense that they share in the sovereignty of exercising power (Cohen 1998: 222). It therefore respects all as intrinsically equal autonomous agents whose interests, preferences and beliefs are as important as one’s own (O’Flynn 2006: 51). Democratic deliberation is publicly reasoned in the sense that people offer reasons in support of their opinions and perspectives, hope they will prove convincing to others and expect these reasons, and these reasons alone, to be the motivating force of people accepting them or not. Kant’s publicity principle has had a profound influence on Rawls and Habermas and their lineage of deliberative democrats, but they offer differing interpretations of Kant’s ‘transcendental formula’ (Elstub 2006a: 305). Rawls perceived it as a hypothetical publicity test, suggesting that if a law or policy is to be right, then it must have the capacity to endure publicity. Even if this is the correct interpretation of Kant’s philosophy, Habermas seems to be right in claiming that laws and policies must actually be made and tested through rational public debate, as we have no other way — 69 —

Towards a Deliberative and Associational Democracy

of knowing if the policies have the capacity of being public (Habermas 1996a). In line with Habermas’ interpretation, democratic deliberation is generally considered to be a joint, collective activity; yet, following in the Rawlsian tradition, both Goodin (1992) and Gundersen (2000) envisage democratic deliberation as being desirable and possible outside of collective debate (Elstub 2006a: 306). The differences between collective and private deliberative theorists derive from different views on the nature of public reason. To be ‘public’ (for both groups) the reasons offered must be understandable and acceptable to all citizens or at least potentially so (Bohman 1996: 26; Elstub 2006a: 306). However, the private deliberationists see reason as ‘singular’, meaning that all will reason in the same way, negating the need for others to be present (Rawls, 1993: 227; for a critique, see Benhabib 1996: 75 and Dryzek 2000: 15). Therefore, individual citizens must consciously adopt public reason, rather than it being generated by the presence of others (Rawls 1997; for a critique, see Dryzek 2000: 15). In contrast, for the collective deliberationists, it is the very presence of other citizens that will encourage people to think ‘publicly’, the idea being that selfish reasons of the type ‘I agree with this because it will really benefit me, but disadvantage others’ will be unconvincing to others and participants in a deliberative debate will want to convince others to gain support for their proposals (Elstub 2006a: 306). Collective deliberation therefore encourages people to focus on public values in order for their arguments to persuade people of the validity of their ideas (Miller 1993: 82; Benhabib 1996: 72; Elster 1997: 12; Elstub 2006a: 306). Included in the process of collective deliberation will be those who would be disadvantaged if such selfish preferences were to prevail, which would make it very difficult to justify these prejudices to these people (Elstub 2006a: 306). However, it is not just the fact that others will have a vote that will encourage participants to offer public reasons, but also the fact that one will ‘internalise’ the norms of publicity through feeling ashamed or repulsed at the ‘inappropriateness of certain styles of argument in the public forum’ (Goodin 1992: 135). Goodin (2003: 63–4) suggests individual deliberation can generate public reason because others can be made ‘imaginatively present’ through individuals conducting ‘a wide ranging debate within their heads’. Yet this conception of deliberative democracy is difficult to distinguish from an aggregative view of democracy, as citizens can deliberate in private prior to voting (Saward 2000: 68), and Goodin therefore accepts that collective deliberation will still be necessary given that we can never know the views of others; so some will be misportrayed, others completely ignored and few put as persuasively as they would be by the agent themselves (Goodin 2003: 63–4; see also Elstub 2006a: 306). — 70 —

Deliberative Democracy and Autonomous Decision-Making

Gundersen advocates ‘dyadic’ deliberation in his ‘Socratic theory’. Groups could still assemble to make collective decisions, but communication between them would always be dyadic, with ‘serial one-to-one encounters’ (Gundersen 2000: 98). According to Gundersen, the first advantage of dyadic deliberation over collective deliberation is that it is easier to institutionalise (Gundersen 2000: 98). This may be the case, but unless it can generate the same or preferable normative consequences, it only stands if deliberative democracy is impossible to institutionalise, otherwise collective deliberation should be pursued (Elstub 2006a: 306). However, Gundersen claims dyadic deliberation is normatively superior to the collective alternative, as the relationship between participants is more interactive and therefore ‘allows each partner to more easily ascertain the other’s knowledge and interests’. This makes clarification much easier because in a group this would require the monopolisation of debate between two people (Gundersen 2000: 98–100). This seems uncertain because there may be more than one misunderstanding, sharing similarities with others. A debate about clarification could therefore take place between more than two participants and aid the understanding of many participants (Elstub 2006a: 306). Gundersen also suggests that dyadic communication will mean greater equality between participants than in collective deliberation because power in dyadic relationships is easier to challenge verbally and exit is easier, too (Gundersen, 2000: 101). This claim may be true in some cases, but certainly not in all. There are certain dyadic relationships where it is harder to challenge power verbally and exit is even harder than in collective debate; it seems to depend upon context. For example, a dyadic relationship may be dominated by one of the participants if the other holds them in high esteem, with excessive respect, or is intimidated and fearful of them. This of course can occur in collective deliberation, but other participants would be present to challenge the esteemed or feared figure with reasons. Likewise, two people may find it very hard to respect deliberative procedures because of the mutual disrespect they feel towards each other, but these feelings may be calmed by the presence of other participants in the debate (Elstub 2006a: 307). Essentially, dyadic deliberation cannot generate public reason in the same manner as collective deliberation and will therefore fail to meet the decision-making requirements that are necessary to cultivate autonomy (Elstub 2006a: 307). In accordance with the distinction between individual and collective decisions, established in Chapter 1, Elster suggests that pure aggregation of preferences (and the argument could also be applied to private deliberation) confuses types of behaviour that are apt in the marketplace and the forum. In the market the consumer can be sovereign — 71 —

Towards a Deliberative and Associational Democracy

because the different choices will only affect the consumer. This is not the case when making collective political decisions because many of the citizens’ preferences may be defective (Elster 1997: 10) and need to be justified to the rest of the polity (Brennan and Lomasky 1993: 33–4). If private, this deliberation does not open people up to the arguments of others, or force them to defend their choice. If dyadic, people only hear the reasons of one other at any one time, and do not hear the exchange of reasons between others, even when these could cause change in their own preferences (Elstub 2006a: 307). Not only is collective deliberation more likely to generate the public reason required for the cultivation of autonomy than private or dyadic deliberation, it is also similarly normatively superior to pure aggregation. It is true that using the resource of voting encourages people to privately deliberate and exercise judgement prior to making such a choice (Lieb, cited in Cronin 1989: 60–1; Saward 2000: 68), and it is this that distinguishes a vote from opinions expressed in an opinion poll, as the latter are often thought to be unconsidered opinions and treated with justifiable scepticism (Fishkin 1991). Nevertheless, this deliberation is usually private; it does not open people up to the arguments of others, or force people to defend their choice to others, therefore it does not encourage the same level of collective judgement, essential for autonomy, that debate creates: ‘The secret ballot allows the voter to express himself, but not to be influenced by others or to have to account for his private choices in a public language’ (Barber 1984: 174). Consequently, in private ballots ‘nothing stops the voter from voting on purely self-interested grounds, without any consideration for what would be a good decision for the collectivity’ (Fearon 1998: 53), which means the preferences and autonomy of others is not considered or respected (Brennan and Lomasky 1993: 217). The ability of deliberative democracy to generate public reason, which aids the cultivation of the autonomy of all in collective decision-making, leads to a well-established assumption, within deliberative democracy, that deliberators should be critically detached, and offer universal reasons that could convince everyone (Elstub 2006a: 307). For Cohen this is essential if the reasons are to be public (Cohen 1998: 186). However, this is a very demanding requirement, especially in plural societies, as, in reality, rather than offering reasons that are convincing to all, people may offer reasons that are aimed at a majority, or the largest minority. (This is of course dependent upon there being an established majority that is apparent to the participants, and as preferences will change during deliberatively democratic debate, this majority may change during the process.) If this occurs, then the reasons are not genuinely public and could even reflect the selfish — 72 —

Deliberative Democracy and Autonomous Decision-Making

interests of the majority and therefore not respect the autonomy of the minority (Elstub 2006a: 307). One suggested solution is that deliberative democracy should lead to a result ‘that enjoys the widest possible support’, not just majority support (Miller 2000: 152), as finding reasons that all can accept is just too stringent a demand, given the fact of pluralism. The danger that citizens will aim to convince the necessary numbers of people to pass a decision, rather than finding a result that gains the widest possible support, still remains and is a danger to the normative claims of deliberative democracy made here. However, the fact that deliberative democracy will encourage the use of public reason and encourage participants to consider the opinions and interests of others will help reduce (though not eliminate) this occurrence and make autonomy-cultivating decisions more likely. Furthermore, it cannot be expected that the same reasons will convince all citizens of a certain decision. Psychological research has indicated that reflective preference transformation will be limited because people are unresponsive to reasons that do not support their preconceptions of an issue. This might explain why different people will look at the same piece of evidence and use it to support their own distinct interests (Femia 1996: 378– 81). Therefore ‘the force of an argument is always relative’ (Manin 1987: 353; Dahl 1994: 31) and if rational argument is to persuade an agent of a new belief, it must start by appealing to their present beliefs (Christiano 1997: 260). Consequently, participants in debate will offer different reasons to persuade different citizens of the need for the same outcome and will therefore not be public in the way envisioned by some deliberative theorists (Gaus 1997). This does not mean deliberators can offer just any reason they think will be convincing to someone, as they must, at the very least, not contradict the reasons they have offered to others. If they did, they would be found to be insincere and willing to say anything to anyone to ensure their preferences prevail in the final decision. This advantage is specific to collective deliberation, as in ‘dyadic’ deliberation there would be nothing to ensure this civilising force of hypocrisy (Elstub 2006a: 308). Hearer Autonomy All the way back to ancient Athens and Plato, one of the major objections to giving citizens more direct powers is scepticism over their ability to make informed decisions. Deliberative democracy is seen as a solution to this, as it is suggested that through debate, providing it is open and inclusive, participants are opened up to a wide range of information, perspectives and experiences that they would be unaware of in the isolation of a voting booth, all of which enables them to make informed decisions. Hearer — 73 —

Towards a Deliberative and Associational Democracy

autonomy is therefore enhanced by public deliberation. In Chapter 1, it was argued that the more relevant information a person has, the greater their capacity for making autonomous choices and forming autonomous views, as their external rational requirements will be increased (Lipson 1995: 2269). Citizens therefore limit their autonomy if they do not receive all relevant information that is available. This is because ‘the effect of the information is its ability to permit the citizen to put more of himself ’ into his deliberations, and hence to permit his preference to express himself fully’ (Lipson 1995: 2264, my emphasis). In this sense they are not completely ‘authentic’ preferences without this information. No one participant could predict what all participants’ opinions would be or could know all the relevant information relating to a decision (Benhabib 1996: 71; Elstub 2006a: 303). However, deliberatively democratic decisionmaking helps distribute information, in so far as participants exchange opinions, experiences and information as they aim to convince others of their preferences, with the result that participants encounter perspectives and information about which that they were previously unaware, and can consequently revise their own preferences in light of this. The information provided in the discussion from the various participants may have some direct bearing on the outcomes from the various choices, which could, would or should have an effect on what decision the collective makes (Manin 1987: 349). Consequently, it is thought that such deliberation would be able to ensure that decisions are based upon reasoned analysis, as opposed to traditional and inaccurate assumptions (Sunstein 1984: 1702). Before voting, people do take account of some information, but there is other information available that would affect people’s convictions (even if just to confirm what they already believed) that they do not hear. If this information is not disseminated, then ‘it is not clear that the votes of the citizens’ represented their own best judgment. It is as though no citizen had fully expressed himself in his vote, because there was a part of him that had been inactive, but that would have been active had he received the information’ (Lipson 1995: 2252). Consequently, many voters’ preferences are not informed or considered opinions (Ryan 1999: 3). Voters are often confused as to what their preferences are, having not had the opportunity to test their views in public, or to hear a broad range of views in order to have formed firm opinions and resolve inconsistencies. The result of this is that when people come to vote, they can vote against their stated preferences (Hensler and Hensler 1979: 106). Nor are participants’ preferences full or complete. Prior to deliberation they have only tentative, unsure and incomplete ideas on certain issues (Fishkin 1991: 83). Cronin cites survey data which indicates that as many as one-third to a majority of those voting — 74 —

Deliberative Democracy and Autonomous Decision-Making

claim they needed more information to make a vote they were comfortable with (Cronin 1989: 230), while Hamilton (1970: 126) shows that this is not a recent phenomena through an investigation of citizens’ votes in referenda in the 1960s. Benhabib consequently argues that the concept of decision-making procedures based purely on aggregation ‘proceeds from a methodological fiction’, as it assumes incorrectly that a pre-political citizen will have an ‘ordered set of coherent preferences’, when in fact democratic deliberation is required to produce this, as participants will engage in rational reflection (Benhabib 1996: 71). Democratic deliberation allows people to question information and arguments put forth by partisan sources, and to form and enter into debate with their own information and arguments (Elstub 2006a: 303–4). Hence, as a decision-making process, it cultivates the collective autonomy of citizens to a greater degree than voting. Nevertheless, it is important to note that democratic deliberation will only ever increase access to available information, consequently deliberation must proceed with the understanding that information may come to light that could change the participants’ preferences (Elstub 2006a: 304). Imperfect information seems inevitable in decision-making, but exhaustive information is not required for autonomous preferences (Hyland 1995: 260), so the resulting decision must be considered ‘an act of true collective determination . . . if the concept of true collective self-determination is to have any practical force at all’ (Lipson 1995: 2253), providing that opinions or perspectives are not formally or systematically excluded. Deliberative democracy cultivates autonomy by helping to fulfil the externalist rationality requirements. It helps overcome the practical problems of gaining that information and increasing access to ‘available information’. Preferences therefore become more autonomous through the decision-making process. Speaker Autonomy Deliberative democracy enhances speaker autonomy because it enables participants’ own reasons to be heard. It also democratises the selecting of decision options, meaning the choices are more likely to meet the ‘adequate range’ criteria necessary for autonomy. A speaker’s autonomy is protected through her right to be able to say what she chooses. In a deliberative democracy participants get to express their feelings and opinions, and justify their perspectives, to the listeners by offering reasons they imagine to be convincing to them. Furthermore, a speaker’s equal agency is respected by ensuring freedom of speech in debate and in addition by enabling the intensity of preferences to be expressed (Barber 1984: 206). As Downs’ — 75 —

Towards a Deliberative and Associational Democracy

(1957) paradox of rational voting indicates, the fact that, through offering reasons, one might affect the preferences of others, and consequently gain a greater influence over final decisions than through voting alone, provides an incentive for people to critically reflect on their own preferences (Hyland 1995: 260). Deliberative democracy further helps cultivate autonomy by increasing citizens’ control over the final choices for decisions. Although deliberative democracy is unlikely to produce consensus when voting takes place, after a process of democratic deliberation people do vote for one of the available options open to them and these options will be derived from the debate that proceeded. The options available will reflect what people have felt and the opinions that have been expressed in the discussion. Deliberative democracy therefore gives the participants themselves the power to select the choices, making them more self-determining and therefore enhancing autonomy. This means choices are more likely to be of an ‘adequate range’, and knowledge of the choices will increase, which is compatible with the requirements of autonomy. In addition, choices that would not have been made available without democratic deliberation can also be generated. Collective deliberation can improve the rationality of decisions by people coming up with an idea that would not have occurred to others. More pertinently, through discussion and debate an idea can be collectively created that would not have occurred to any participants individually (Fearon 1998: 50; Smith and Wales 2000). Without this debate, the choices may not reflect how the participants feel, and this in turn limits their autonomy because it puts an external restriction on the preferences the voter can express: Talk enables us to examine rank orders, commensurable scales, and the effect of time and place; it allows us to get at what we really want as individuals and as a community. Voting freezes us into rational dilemmas . . . choices are generally more coherent and less paradoxical than the logical dilemmas extrapolated from them, especially if the choices are informed by a process of strong democratic talk. (Barber 1984: 205) However, some deliberative democrats seek to limit speaker autonomy by restricting the range of preferences that can be included in debate a priorily, by using input filters. These resemble Rawls’ (1993) belief that democratic deliberation should not adapt to pluralism as it stands, but to ‘reasonable pluralism’. Output filters reflect a Habermasian line, presuming reasonable pluralism can be achieved through the deliberative process itself and therefore that no views should be excluded prior to the deliberative process (Elstub 2006a: 310). — 76 —

Deliberative Democracy and Autonomous Decision-Making

Goodin (1986) argues that input filtering is a more efficient way of excluding irrational or misguided preferences than output filtering, as they do not enter the decision-making process and so do not influence participants at any stage. Consequently, he suggests preferences should be laundered by elites prior to debate (Elstub 2006a: 310). Gutmann and Thompson (1996) are also in favour of input filters to prevent the inclusion of preferences that challenge political equality. Despite being committed to the deliberative theory of democracy, they want to achieve the value of publicity prior to the deliberation process, obviously being sceptical of its ability to ensure public reason through its internal processes of debate (Elstub 2006a: 310). Similarly Miller (2000) and Blaug, (1999: 148–9) believe that certain reasons, such as racist arguments or violent and coercive threats, should be excluded a priori, in order to be in line with the requirements of deliberation that insist participants should be free and equal, and to further ensure the deliberative capacities of individuals are not damaged through intimidation. It seems evident that threats of violence and coercion must be ruled out, as failure to do so would mean that force itself, and not the force of the better argument, would be successful. The case for the exclusion of reasons, even if they are racist, sexist, sectarian or homophobic and so on, seems less certain (Elstub 2006a: 310). Input filtering seems undemocratic, unnecessary, and anti-deliberation (Elstub 2006a: 310). First, input filtering leads to an ever-increasing number of issues being excluded from the agenda, which can mean that ‘too many issues will form the background framework of public deliberation rather than its subject matter’ (Femia 1996: 370) and make deliberative resolution of these key conflicts impossible (Johnson 1998: 168–70). Therefore, not only is the deliberative scope reduced, but also the democratic scope of decision-making (Elstub 2006a: 310). Consequently, Cohen states that; ‘what is good is fixed by public deliberation and not prior to it’ and ‘for this reason the deliberative conception supports protection for the full range of expression, regardless of the content of that expression’ (Cohen 1989: 29–30; see also Benhabib 1996: 70). Sher agrees, arguing that it is unnecessary to a priorily restrict the agenda, as the relevant distinction is between good and bad reasons, not legitimate and illegitimate ones. Not all reasons will carry the same force. For example, racist reasons are inadequate reasons: ‘Indeed, to show that a form of discrimination is illegitimate, the best strategy is publicly to consider, and decisively to refute, the best arguments advanced on its behalf’ (Sher 1997: 5; see also Dryzek 2000: 43). However, for Goodin, the exclusion of certain opinions is necessary not because they might be successful and present in the final decision (he is confident about the role of public reason preventing this), but because they are offensive, morally — 77 —

Towards a Deliberative and Associational Democracy

wrong and therefore should not even be discussed (Goodin 1992: 138–40). Even if this is the case, there is the further problem of implementation: if input filtering is to be implemented democratically, then all relevant actors should be included. Deliberative democrats could not consistently accept that citizen’s pre-political preferences should determine the basis of censorship, which would therefore give rise to the need for a debate between all these actors in which no reasons could be a priorily excluded (Elstub 2006a: 31). However, if elites impose input filters, as Goodin suggests, they could potentially control the agenda, and therefore politics itself, (Schattsneider 1975; Barber 1984: 180), by excluding ideas that they did not like, or that challenged their power – just as anti-racist, feminist, pro-homosexual, socialist, environmentalist and libertarian views have been labelled offensive and morally wrong in the past and even today. If we want new and distinctive ideas to be incorporated into the public sphere, we cannot allow any ideas to be formally excluded, as ideas that challenge powerful groups and threaten the status quo could be excluded along with exclusive and prejudiced discourses. Therefore, if a deliberative democracy is to hold true to both its deliberative and democratic strands, no reason should be formally excluded from debate and the domain of preferences should be restricted through the process of deliberation itself (Elstub 2006a: 311). Deliberative democracy can therefore cultivate autonomy in four ways: by increasing the chances of compromise; by generating public reason so people consider the interests of others when forming preferences and making decisions; by enhancing hearer autonomy by increasing information through which preferences are correspondingly more informed; and finally, by enhancing speaker autonomy through enabling people to say what they want and increasing the likelihood that options for voting will resemble their preferences. However, three important critiques challenge these claims and must therefore be considered. The Social Choice Critique The seminal social choice theorist, Kenneth Arrow (1963), considered the key problem of how to aggregate individual preferences in order to make a collective decision or social choice. He set out five conditions that must be met to ensure a rational social choice but further argued that all these requirements cannot be satisfied at the same time through the same aggregative procedure: 1. Collective rationality requires social orderings to be transitive if individuals’ are. — 78 —

Deliberative Democracy and Autonomous Decision-Making

2. The weak Pareto condition stipulates that if every individual ranks x above y, then the collective decision should as well. 3. Non-dictatorship indicates that if the majority prefer y to x, and the minority x to y, the social ordering should not select x. 4. Universal domain: this requirement stipulates that no preferences should be excluded from collective decisions. 5. Independence of irrelevant alternatives suggests that if all individuals’ preferences are between x and y then the collective choice should be between only x and y (Arrow 1963; see also McLean 1991: 180 and Dryzek 2000: 34–35 for accessible summaries of Arrow’s social choice conditions). Arrow’s conclusion was that the only way these criteria could be met was through ‘domain restriction’ – the restriction of the range of preferences citizens could vote upon – which he argued was unacceptable, as it would violate ‘citizen authority’.2 Applying Arrow’s conclusions further, Riker (1982) has famously argued that all aggregation mechanisms are susceptible to instability and ambiguity, which allows for strategic manipulation. They are unstable in the sense that they spawn cyclical and intransitive social orderings and majorities; consequently, voting paradoxes occur.3 The problem of transitivity occurs because people’s choices do not always reflect their preferences, and decision procedures that fail the transitivity requirement are cyclical and ambiguous. Individuals’ preferences are intransitive if they prefer A to B, and B to C, but C to A; but even when individuals have transitive preferences, they may lead to collective intransitive ones. If there are three voters and voter one ranks the available options as A, B, C in preference order, voter two ranks the options B, C, A in preference order, and voter three ranks them C, A, B, then A will beat B, B will beat C and C will beat A, each winning by two votes to one. Consequently, an option can win the vote without being the first choice of any voter. Ambiguity can occur because the vote-counting process employed determines decisions. This means that, despite people’s preferences being exactly the same, different methods of counting the vote will yield very different results. Furthermore, there is no way of judging which method of aggregation most accurately aggregates people’s preferences, and so there is a sense of arbitrariness to the decisions that arise (Nurmi 1998: 334). This restricts autonomy of all citizens because the final decision in no way reflects their collective preferences, so it is hard to say in what sense the citizens are self-determining. Manipulation can occur through strategic participation and agenda — 79 —

Towards a Deliberative and Associational Democracy

control (Riker 1982: 237). According to Dryzek, ‘situations are strategic when the outcome of an actor’s choice depends on the choice(s) made by another actor or actors also pursuing goals and interests’ (Dryzek 2000: 31; see also Benoit and Kornhauser 1994: 186). Strategic participation applies equally to debate and voting, as people can strategically misrepresent their preferences in both (Cohen 1998: 198; Johnson 1998: 171), and the greater the divergence of opinion, the more likely actors are to communicate strategically (Cohen 1998: 198). Furthermore, it is very difficult to distinguish between those acting sincerely and those acting strategically (Johnson 1998: 171), or even if anyone has acted strategically – so even when no one has acted strategically, preferences still cannot be derived from aggregative mechanisms. As was argued in Chapter 1, manipulation of the decisions to be made is incompatible with a collective making autonomous decisions and being self-determining. Manipulation through agenda control is also concerning for democrats, as ‘simple majority voting can lead to any outcome in a space of issues if the agenda of voting is appropriately manipulated’ (Mackie 1998: 74; see also Miller 1993: 79). All these problems mean that popular will cannot be identified independently of the aggregative mechanism used to identify it. Therefore, Riker draws the conclusion that ‘populist’ democracy is impossible because there is no such thing as ‘the will of the people’ that should be reflected in decisions (Riker 1982: 117–19; see also Dryzek 2000: 35). Democracy must therefore be abandoned ‘not because it is morally wrong, but merely because it is empty’ (Riker 1982: 239). If the social choice theorists are right, then the link between self-government and democracy is broken and therefore the link between autonomy and democracy is broken. In fact, the autonomous decision, as I set it out in Chapter 1, is impossible. It is also a serious challenge to the claims made here for deliberative democracy. If differing decisions can arise from a deliberative process with exactly the same arguments, but different procedures, then ‘the outcome of deliberation is hostage to precisely the sort of arbitrary factors for which aggregation has repeatedly been criticised’ – that is, the problems of instability, ambiguity and manipulation (Knight and Johnson 1997: 291 and 1994: 283; see also Weale 2000: 2). If consensus could be achieved, aggregation could be avoided and therefore the social choice critique would not pertain. However, as already discussed, consensus cannot, and should not, be the aim of deliberative democracy, so we must look elsewhere if the social choice theory critique is to be avoided. I therefore consider four possible ways for deliberative democracy to avoid the social choice critique: 1) it transforms and, in turn, restricts the domain of preferences; this means that it that it can 2) change and make lines of disagreement clearer, so a more suitable choice of aggregation mechanism — 80 —

Deliberative Democracy and Autonomous Decision-Making

can be made; this latter aspect is also promoted by the fact that 3) deliberative democracy can lead to more single-peaked preferences, which reduces the dimension of disagreement and cycling; and finally, 4) it reduces strategic participation due to the public nature of debate. First, though, I cast doubt over the validity of Arrow’s ‘independence of irrelevant alternatives’ criteria for making a coherent decision. Abandonment of the ‘Independence of Irrelevant Alternatives’ Criteria The first problem with the social choice critique is that not all of Arrow’s five criteria are necessary for a coherent decision. Arrow’s fifth criterion, ‘independence of irrelevant alternatives’, demands that all preferences be ordinate, but if this is not how individuals make collective decisions, then it could be abandoned. The fact that we have preferences over our preferences, a point that was highlighted in Chapter 1 with the discussion of Dworkin (1988), means that it is possible for a second order preference to contradict first order preferences, and for second order preferences to ‘win out’ when making an individual decision.4 In short, when an individual is making a decision they may not rank their preferences in the manner social choice theorists predict. Bird (2000) therefore suggests that Arrow’s criteria of independence of irrelevant alternatives can be abandoned because people can make decisions based upon alternatives that would be considered irrelevant to the decision in hand. Social choice theorists may try to defend their position by suggesting that, higher order preference is just another preference that is attributed greater importance over one’s other preferences in this particular decision, and so is not an ‘irrelevant alternative’. Such a defence would be flawed, though, because higher order preferences are necessarily not just a preference for x over y, but a preference about the desirability of having preferences x and y to begin with. Therefore, higher order ‘volitions cannot simply be assimilated without qualification to other first order rankings of the alternatives’ (Bird 2000: 573). Just as an individual may have higher order preferences, so might a collective; for example, if it wishes to act democratically and have higher order collective preferences about wanting decisions to be fair and inclusive (Bird 2000: 575). If a collective is to operate like a democracy, then this ‘automatically entails various restrictions on the range of motivations allowed to constitute its will’ (Bird 2000: 576). The problem for society is how to ensure that this democratic identity is in place. I have argued, however, that the ideal of deliberative democracy aims at compromise, inclusion, debate and the filtering of preferences through public reason, and that this will achieve that goal. — 81 —

Towards a Deliberative and Associational Democracy

Domain Restriction For Arrow (1963) the key to making a coherent social choice is domain restriction, but deliberative democracy contributes towards this. Richardson (2002: 50) argues that social choice theory does not disprove the possibility of making collective decisions commensurate with public goods, but rather it shows that decision-making cannot operate in an ‘institutional vacuum’. However, such vacuums are not compatible with legitimate decision-making, as they are not normatively neutral and will inevitably promote some values above others (as was argued in Chapter 1 under the discussion of perfectionism and above in the discussion of proceduralism). The assertion here is that fair procedures should be justified on their ability to cultivate autonomy equally for all. The value of autonomy, as a basis for decision-making procedures, will then rule out some preference orderings and decision-making methods. For example, the possibility of dictatorship, even one based on paternalism, is ruled out tout court. The process of deliberative democracy itself further restricts the domain of preferences. As already discussed, deliberative theorists conceive preferences as being exogenous, formed during the political process rather than prior to it. Sunstein suggests preferences are produced and change in accordance with a wide range of factors, such as the context in which the preference is expressed, relevant information, existing rules, social pressures, interaction, communication and consideration of those with differing preferences, and will therefore also adapt to reason, with the latter resulting in these preferences becoming better understood (Sunstein 1985: 31–5 and 1991: 5; see also Sagoff 1998). Due to this capacity of public debate to transform preferences, deliberative democracy can reduce ‘the range of admissible preferences’ because preferences must be justified publicly and therefore preferences that cannot be publicly defended are filtered. In this way, it can ‘limit the possibilities for cycling across alternatives’ (Dryzek 2000: 73).5 Consequently, it is thought that deliberative democracy can reduce the scale of preference rankings that will need to be aggregated (Miller 1993: 81). However, many other democrats and social choice theorists share the prevalent liberal conception that perceives preferences as endogenous and unchanging. They think deliberative theorists confuse evidence of changing preferences with a change in the available ‘choice set’ on a particular decision (Miller 1993: 90; Dryzek 2000: 32). Whether preferences are exogenous or endogenous is essentially an empirical question. The results from deliberative opinion polls and citizens’ juries indicate that preferences will change when citizens participate in democratic deliberation (Fishkin — 82 —

Deliberative Democracy and Autonomous Decision-Making

1991 and 1995; Stewart et al. 1994; Renn et al. 1995; Coote and Lenaghan 1997; Coote 1997; Kuper 1997; McIver 1997; Andersen and Jagger 1999; Barnes 1999; McCombs and Reynolds 1999; England 2000; Luskin et al. 2002; Hansen 2004; Andersen and Hansen 2004; Parkinson 2006). This is certainly not conclusive empirical evidence, though, and Jordan attributes such preference changes to biased information or sees it as further proof of Fishkin’s own assertion that preferences are tentative, unsure and incomplete after, as well as prior to, deliberation (Jordan 2007: 60–1). However, the increasing volume of results from deliberative opinion polls and citizens’ juries does indicate that preferences are exogenous and that counter empirical evidence must be provided to suggest differently. The argument is, then, that deliberatively democratic decision-making ensures that individual preferences are authentic, and that through its internal process of preference laundering it also ensures that the preference domain is restricted, and consequently that clear, unambiguous results can be produced. Choosing Mechanisms of Aggregation The social choice theory critique presumes the aggregative method is selected prior to knowing what people’s preferences are, without their formation and transformation through discussion, and therefore without knowing the nature, location and extent of the disagreements that exist. Therefore, the most suitable aggregative mechanism is not known, is selected arbitrarily and therefore produces arbitrary collective decisions (Miller 1993: 85–8). Although there is no guarantee that deliberative democracy will reduce disagreement, it can nevertheless change the nature of this disagreement and make it clearer where exactly the disagreements lie and the extent of them. This increased knowledge will help reduce the misrepresentation of preferences through aggregative mechanisms because the most suitable method of aggregation, and the most relevant options for voters to choose from, can now be selected (Miller 1993: 81–6; Fearon 1998: 45). Increase in Single-Peaked Preferences The problems identified by Arrow’s theorem can be avoided if citizens’ preferences are ‘single-peaked’ (Black 1948; Miller 1993: 84). ‘Single-peakedness means that when the available options are arrayed on a continuum, the individuals preference must fall continuously on either side of the most preferred position’ (Dryzek 2000: 43; see Miller 1993: 84, for a similar definition). It is thought that if preferences are single-peaked, then there is a — 83 —

Towards a Deliberative and Associational Democracy

constraint on preference order structures, as it indicates that citizens do share a common view, even if not the same judgements: ‘When preferences are single-peaked we know not only that a majority winner exists but we also know how to find it easily’ (Barry 1991: 32). This in turn reduces the dimension of disagreement and prevents cycling (Knight and Johnson 1994: 282–3; Miller 1993: 84). Riker himself seems to admit that, to avoid cycling, decision-making does not need to achieve consensus but only ‘a common view of the political dimension’ (Riker 1982: 128; see also Knight and Johnson 1994: 283).6 Due to the central role of debate and public justification in a deliberative democracy, this common view becomes easier to achieve and in addition helps make single-peaked preferences reducible (Dryzek 2000: 43). Through debate individuals may come to realise that their preferences do contradict each other and therefore are not ‘single-peaked’. This is because debate can disamalgamate decisions that participants place different ranking orders of preference on. Once again, the dimensions of disagreement will become more apparent, allowing for the possibility that the original decision can ‘be split into components’ (Miller 1993: 85). The great advantage of this is that it can increase the chances of there being an option that can be identified as having majority support. However, Miller further recognises that it may not be in the interest of some participants to have the decision broken into components, and so they may try to strategically prevent this, but this can be avoided if the agenda is democratically formed (Miller 1993: 86). Ideally, in a deliberative democracy, if participants want to prevent decisions being broken down into components, they will have to justify this with public reasons, and whether they get their way or not would depend upon the quality of the reasons: ‘it would then be difficult to make a public argument against the disaggregation of decisions where it was clear the original choice was multidimensional’ (Miller 1993: 86). This leads us to consider the potential of deliberative democracy to reduce strategic action. Reduction of Strategic Action It is further suggested that deliberative democracy can reduce strategic activity in decision-making because it encourages the development of public reason, fosters cognitive dissonance reduction, can eliminate strategic choices from aggregation, increases the chances that the civilising force of hypocrisy will pertain and enables biased information to be challenged. As already argued, deliberative democracy encourages participants to offer public reasons for their preferences; however, many preferences can be justified publicly and participants could speak publicly on public interests — 84 —

Deliberative Democracy and Autonomous Decision-Making

but vote privately on selfish interests (Miller 1993: 76).7 By providing reasons for an option, though, one may convince oneself of it because ‘by speaking with the voice of reason, one is also exposing oneself to reason’ (Elster 1997: 12), which could lead one to internalise these norms (Goodin 1992: Chapter 7). Miller argues that human psychology requires cognitive dissonance reduction, so people would eventually become convinced of the public reasons they offered (Miller 1993: 82–3; Dryzek, 2000: 47–8). But cognitive dissonance, and its motive of inconsistency avoidance, do not lead to autonomous and reasoned preferences (Elster 1997: 12; Johnson 1998: 172). However, public debate and cognitive dissonance reduction are not the only way that deliberative democracy can reduce strategic activity. As already established, one of the advantages of deliberative democracy is that the options available to vote on will be drawn from the preceding discussion, and as a consequence, the options available to vote on may not represent one’s own selfish interests (Fearon 1998: 54). The social choice theory critique seems to rely upon debate in a deliberative democracy being a one-off, private interaction between two people, whereas it is more likely to be public, recurrent and conducted between many speakers and many listeners (Mackie 1998: 84–5). These factors produce ‘the civilising force of hypocrisy’, which sees that participants will be concerned about their reputation and will not want to be seen to be acting strategically for fear of future recriminations. Strategic participants must therefore abide by their arguments in public, even if they are really motivated by self-interest, or else risk being shown to be a hypocrite (Elster 1995 and 1998b). If people argue for one thing and vote for another (and the vote takes place in public), then they will lose credibility when they participate in future. Other participants will regard the reasons of the hypocrite as insincere and they may even face public humiliation (Brennan and Lomasky 1993: 40; Dryzek 2000: 46): ‘The moral force of the charge of hypocrisy does, then, imply the desirability of congruence between expressive and instrumental behavior’ (Brennan and Lomasky 1993: 52). Miller (1993: 83) cites evidence from Davis et al. (1988) and their research into juries, which supports this argument emprically. He further cites evidence from Orbell et al. (1988) and their research on prisoner dilemma situations, which shows how discussion can more than double instances of co-operation. Strategic behaviour is further reduced because people want to ‘think well of themselves’. In this way, people express themselves not as they are, but as they would like to be (Brennan and Lomasky 1993: 40). In addition, in social choice theory, the belief that deliberators will act strategically is based upon the presumption that the information shared is unverifiable and unproveable. This is the case when one participant has a — 85 —

Towards a Deliberative and Associational Democracy

monopoly on information and uses this information strategically. For example, tobacco companies’ own research into their product’s health risks was, for a time, the only information available. However, in most cases, one decision-making participant will not have a monopoly on information, and contrasting information, evidence and reasons will be provided from a plurality of sources (Mackie 1998: 85–6). Therefore, strategically biased or incorrect information can be disproved or at least challenged. Therefore, the worst cases of strategic behaviour in deliberative democracy can be eliminated, or at the very least made less salient. Deliberative democracy can therefore increase the chances that a coherent social choice will be made, making decisions more stable, less ambiguous and less subject to strategic manipulation. This in turn aids the cultivation of autonomy, as it increases the chances that the intentions of a collective can be identified and reflected in a collective decision. The extent that deliberative democracy can achieve these aspects will though, depend much on the institutional setting, which will be focus of argument in subsequent chapters. Difference Democrats’ Critique In the first chapter, it was argued that democracy aids the cultivation of autonomy due to its commitment to equality, and this was seen as normatively correct in so far as each citizen’s autonomy should be equally respected. It is suggested in this chapter that deliberative democracy can contribute to this required political equality through its generation of public reasoning, because this makes it more difficult for powerful groups to serve their particular interests; deliberation can, then, expose self-interest that it is in the powerful groups’ interest to hide and disguise (Elkin 2004: 61). One of the prominent criticisms against the ability of deliberative democracy to ensure this equality comes from ‘difference democrats’ such as Young (1990, 1996 and 1997) and Sanders (1997). They accept that the model of deliberative democracy is formally inclusive, in the sense that it seeks the participation of all, but claim that it is not substantially inclusive because the procedures and forms of communication privilege dominant groups over others in decision-making (Elstub 2006b: 31). This is a worrying and significant challenge because, if it is upheld, deliberative democracy would not be the most appropriate decision-making mechanism to cultivate the autonomy of all, as it would lead to the continued subordination of minority groups at the expense of dominant groups; its normative legitimacy would therefore be completely undermined. From the arguments of Young and Sanders, three criticisms of — 86 —

Deliberative Democracy and Autonomous Decision-Making

deliberative democracy can be identified. These are that it will enable certain groups to participate more and therefore dominate decision-making; that rational argument cannot challenge existing inequality; and finally, that the communication favoured by the norms of deliberative democracy are not neutral but culturally specific, and disadvantage current subordinate groups (Elstub 2006b: 32). These criticisms are now considered in turn. Some Groups Will Participate More than Others Sanders (1997: 365–6) points to evidence from juries to demonstrate that in deliberative settings it is not the quality of reasons that will persuade people but group dynamics and power structures. She argues that those who speak more gain more influence, and that those who speak the most are white males. In contrast Fishkin and Luskin (2000) cite evidence from deliberative opinion polls that suggest all social groups are able to participate fairly. All this demonstrates is that deliberative democratic decision-making requires procedures to ensure that all have an equal opportunity to participate, and effective moderation can ensure this (Parkinson 2006: 86–7). Sanders herself acknowledges her criticisms are dependent upon the verdict style of discussion rather than evidence-driven discussion being in place. Both methods are employed in juries. In verdict-driven deliberation styles, certain participants associate themselves with certain proposals early on in the process (sometimes taking an early vote). Therefore ‘verdictdriven’ style deliberation approximates more closely an aggregative model of decision-making, as it accepts the validity of pre-political preferences. In this method few preferences change and the decision usually reflects the initial views of the majority. In evidence-driven deliberation, certain options and opinions are discussed without people being categorised or formerly associated with any particular perspective (Sanders 1997: 367). ‘Evidence-driven deliberation’ appears more inclusive than ‘verdict-driven deliberation’ because it encourages all views to be expressed, more participants speak and this in turn causes more people to change their opinions. The evidencebased approach can also incorporate difference and there is a greater emphasis on all participants trying to reach an acceptable decision for all, rather than having one view winning out (Elstub 2006b: 32). Sanders argues that it is more equipped to do this as it avoids people conforming with majority opinion because of the power of majority and the force of conformity (Sanders 1997: 367), presenting the danger that conflict is suppressed, giving a false appearance of agreement. On balance, however, we can agree with Miller (2000: 146) that the evidence-driven approach more accurately embodies the characteristics of ‘good political deliberation’. — 87 —

Towards a Deliberative and Associational Democracy

Sanders’ argument therefore does not suggest that deliberative democracy necessarily favours certain social groups, but rather that when it is institutionalised the procedures used should approximate the evidencedriven model of deliberation (Miller 2000: 146–7; Elstub 2006b: 33). Sanders’ concerns are also not born out empirically, with evidence suggesting that participants from a diversity of backgrounds will demand to be heard if they feel they are being marginalised (Parkinson 2006: 142). The Inadequacy of Reason Sanders also believes that ‘insidious prejudices may incline citizens to hear some arguments and not others. Importantly, this prejudice may be unrecognised by those citizens whose views are disregarded as well as by others’ (Sanders 1997: 353). This is a serious challenge to deliberative democracy and the claim that it can lead to the equal cultivation of autonomy because this is dependent on the assertion that such prejudices can be countered by argument within the deliberative arena. However, if Sanders is right that ‘prejudices cannot possibly be challenged’ (Sanders 1997: 353), then deliberative democracy may actually reinforce inequality and exclusion. Sanders’ argument is, then, dependent on prejudice not being open to argument and critique, but being ‘invisible’ and not exposed to reason. This is really an empirical claim, and one of which we should be sceptical. What is required is more empirical research in this area to establish whether prejudice is susceptible to reason, as the implications of this are huge (Elstub 2006b: 33). Cultural Specificity Another threat to the connection between deliberative democracy and autonomy is that the capabilities required for effective participation are not culturally neutral and will disadvantage certain groups from the outset. It is suggested that it would be even more difficult for the subordinate groups to challenge the established norms of the dominant groups if the capabilities they needed to develop to be able to challenge these norms were in fact biased against them (Elstub 2006b: 33). For example, the type of language expected could be interpreted as favouring white ethnic groups, the formality of the meetings are accused of favouring the middle classes, and the rationalistic style of argument is criticised as being gender-biased against women (Bohman 1996: 116). Young believes that democratic deliberation requires neutral or universal language and premises, but that in multicultural societies characterised by — 88 —

Deliberative Democracy and Autonomous Decision-Making

pluralism, this is unobtainable. For example, Young suggests that ‘socialists, radical feminists, American Indian activists, Black activists, gay and lesbian activists’ cannot define and describe the oppression they feel in the dominant discourse of liberal individualism (Young 1990: 39). Specifically, Young claims that discourse in deliberative democracy is ‘assertive and confrontational’, ‘formal and general’, and ‘dispassionate and disembodied’ (Young 1996: 123–4). These claims will now be examined in turn. The first claim is that in a deliberative democracy ‘speech that is assertive and confrontational is here more valued than speech that is tentative, exploratory, or conciliatory. In most actual situations of discussion, this privileges male speaking styles over female’ (Young 1996: 123). It may well be the case that men employ more confrontational speech, but it is incorrect to conclude that this speech is necessarily more likely to achieve success in a deliberative situation and it could even be that the opposite is true. Young ignores the preference transformation potential of argumentation as well as the argument that speech that is ‘tentative, exploratory, or conciliatory’ could be more likely to achieve preference transformation than confrontational speech (Elstub 2006b: 34). Young (1996: 123) also seems to be under the misconception that the model of deliberative democracy is based upon competition, with participants trying to win the argument. She argues that acknowledging the force of the better argument equates to ‘conceding defeat’ because one cannot provide a counter-argument. However, this view again seems to overlook the possibility of preference transformation through debate. It is not, therefore, necessarily the case that people will always characterise preference change as ‘defeat’, as they may simply come to agree with a better argument that has been presented to them (Elstub 2006b: 34). Young acknowledges this type of preference transformation herself (Young 1996: 125). The second claim is that deliberation that is ‘formal and general’ excludes groups that need to highlight injustice in specific circumstances. It can be argued, however, that deliberation still allows this; groups can make appeals from the specific and relate it to general principles, such as justice (Miller 2000: 153–4). For example, disabled people demanding easier access to buildings may build their argument from personal experience and refer to specific buildings before broadening the argument that this lack of easy access is unjust and applies to all public and commercial buildings (Elstub 2006b: 34). Finally, Miller rightly criticises Young’s analysis of reason, which suggests that it is dispassionate and disembodied, as being based upon a false dichotomy because ‘all political speech and argument must convey the — 89 —

Towards a Deliberative and Associational Democracy

feelings and commitments of the speaker, but also give reasons either positively for some proposal, or negatively against some alternative’ (Miller 2000: 153). There is nothing wrong with demonstrating emotion in debate; in fact, it is important to show how strongly one feels about something to gain trust and commitment from others (Gambetta 1998). However, passion cannot stand alone; reasons must be supplied to convince others. People will not agree and change their preferences simply because one holds a preference strongly, but also due to the justifications for the preference. Indeed, there are certain aggregation mechanisms that allow people to place multiple votes for an option to demonstrate their passion for their preferences, but the votes do not require justification (Elstub 2006b: 34). Furthermore, we can agree with Miller that it is ‘rather insulting to disadvantaged groups to suggest that norms of argumentative rationality are loaded against them, because it implies that they cannot give coherent arguments for the changes they want to bring about’ (Miller 2000: 153). Communicative Democracy Instead of deliberative democracy, Young advocates ‘communicative democracy’ (1996), which she suggests will differ from deliberative democracy by favouring greeting, rhetoric and storytelling over rational argument. She also suggests that these techniques will make communication more compatible with pluralism because they are more amenable to the particularity of participants (Elstub 2006a: 309). ‘Greeting’ deals with how participants provide recognition amongst one another and is said to be important as it creates the right atmosphere for deliberation and can indicate a mutual respect. ‘Rhetoric’ is the use of cultural symbols and values, and can provoke and motivate participants, playing a key role in getting issues on the agenda (Gutmann and Thompson 1996: 135; Dryzek 2000: 54; Parkinson 2006: 143–4), aiding mutual understanding (Dryzek 2000: 70). ‘Storytelling’ or ‘testimony’ is the use of narratives, personal or otherwise, and is claimed to be essential because people need to share their personal experiences in order to highlight and demonstrate their specific position (Elstub 2006a: 309). Many deliberative democrats have accepted that greeting, rhetoric and storytelling could and should play a part in deliberation, and in empirical debates actually are included, suggesting that collective deliberation is compatible with ‘a range of communicative styles’ (Barnes et al. 2004; Parkinson 2006: 139–42), but have further responded by highlighting the fact that these communicative aspects are as hierarchical as the rational deliberation criticised. Just as some people are better at forming, expressing — 90 —

Deliberative Democracy and Autonomous Decision-Making

and understanding rational argument than others, so some people will have more talent for greeting, rhetoric and storytelling (Elstub 2006a: 309). Moreover, the people who have talents for these things may be those from the same dominant social groups who are talented arguers (Benhabib 1996: 82; Gutmann and Thompson 1996: 137; Dryzek 2000: 67; Miller 2000: 156–7). In addition, the inclusion of emotions can be hazardous and lead to inequality. For example, the ‘emotional culture’ of a deliberating group can lead to inequality in opportunities to participate effectively in debate and silence certain members, therefore emotion in deliberative debate should be present but contained (Thompson and Hoggett 2001). Neither can these forms of communication seem to replace the need for, or even the central role of, reason. The justification behind prioritising reason is that, if autonomy is to be cultivated, then preferences should be justified when making collective decisions. Although these other forms of communication fulfil important functions conducive to deliberation, only reason can achieve this (Elstub 2006a: 310). For example, reason is essential for answering ‘What is to be done?’ questions, which these other forms of communication cannot answer (Dryzek 2000: 71). The stories that are told will still need to highlight the situation of a broader group, not simply individuals (O’Flynn 2006: 133, 138), and will be open to contestation (Gutmann and Thompson 1996: 137). This contestation over the accuracy and validity of storytelling leads to a ‘danger that such groups will require correct storylines, and punish incorrect ones which cannot easily withstand the normalising gaze of the group’ (Dryzek 2000: 68). This is why, as Gutmann and Thompson realise, without rational argument storytelling can only bring differences to the attention of participants; it cannot resolve conflicts (Gutmann and Thompson 1996: 137). Furthermore, decisions must be based on reasons, as reasons can be shared in a manner unobtainable to storytelling which is often personal and not of common experience (O’Flynn 2006: 137). Rhetoric can actually act to conceal reasons for action (Miller 2000: 156) that could prompt preference transformation without there being sound information and reasoning behind the change, as it can involve emotional manipulation (Chambers 1996: 151–2) which conflicts with autonomous preference transformation. Fortunately, rhetorical statements can be rationally justified and scrutinised (Dryzek 2000: 52–4; O’Neill 1998), which should help counter these negative aspects. Deliberative democracy can, then, help promote the autonomy of all social groups equally, and can provide subordinate groups with the chance to challenge inequality through reason, as well as through the use of greeting, rhetoric and storytelling, but these must be additions to, and not replacements for, rational argument. However, I would warn that this is — 91 —

Towards a Deliberative and Associational Democracy

dependent upon how deliberative democracy is institutionalised, an issue that will be dealt with in Chapter 3. Deliberative Obligations The final challenge against the claim that deliberative democracy can cultivate autonomous preferences, and decisions, is concerned with the motivations of the participants and is based on the premise that participants in a deliberative democracy have ‘special obligations’ to other participants, and moreover, that deliberative democracy will not achieve the normative goods ascribed to it unless these obligations are met (Festenstein 2002: 89). However, the suggestion, from Festenstein, is that we cannot be sure that citizens will always abide by these obligations, as in certain circumstances it will not be in their interests to do so (Festenstein 2002: 89), a phenomena that may occur more in plural societies due to there experiencing less solidarity (Richardson 2002: 157). There are three key obligations, the combination of which is specific to the deliberative model of democracy, and so can be classified as ‘deliberative obligations’. The first obligation is to provide reasons that all can accept. As I have conceived deliberative democracy, this is an excessively strong demand. Instead of this requirement, I suggest that participants have an obligation to find reasons that most can accept. The second obligation is to listen and reply sincerely to all others, as none of the advantages of deliberative democracy outlined above will occur unless actors listen to each other (Gundersen 2000: 97). Information will not be increased if people will not listen to others, hearer and speaker autonomy will then not be advanced and there will be no need to make public justifications if the public ignores what one says. The third and final obligation is to try to find a proposal that is acceptable to all through the modification of proposals in accordance with the reasons of others (Festenstein 2002: 89). This means that participants should opt for the choice that best accommodates all participants’ preferences and make decisions that are not ‘tyrannical’ to a specific group. All political democracies place some obligations on their citizens but deliberative obligations are more demanding than many other models of democracy, for example, a purely aggregative model in which the only obligation is to vote. All these obligations therefore require trust and mutual understanding: if one person is to keep to these obligations, then they will expect others to do so to, and might only keep to these obligations if they trust that others do the same (O’Flynn 2006: 4). These obligations are, then, demanding and the challenge is that they will not always be kept to. However, we can invoke some of the same — 92 —

Deliberative Democracy and Autonomous Decision-Making

arguments here as we did against the social choice theorists: although a failure to abide by the obligations may be strategically motivated, it was argued above that deliberative democracy could reduce strategic behaviour. Festenstein accepts these arguments, but argues that they are insufficient to provide grounding for the deliberative obligations, as we can still imagine occasions when the costs of not abiding by them would be small. In this sense, this approach ignores ‘the distinction between a pragmatic norm and an obligation’ (Festenstein 2002: 104). Chowcat (2000) suggests that deliberative obligations will be upheld by participants simply because they have beliefs that they want to see manifested in decisions, which requires that belief to be asserted and defended. If one did not meet this obligation, then it is tantamount to accepting that your beliefs are not the best-justified, which goes against what it means to have a belief. Parkinson accepts that this might be the motivation when participants initially enter into deliberation, but feels strategic considerations could still lead to an abandonment of the obligations if they felt they were losing the debate. This is because beliefs can still be strongly held without them being defended in public, as authentication can come from a small group of the like-minded (Parkinson 2006: 38). An alternative may be that the participants will abide by the obligations because of instrumental reasons. If the same participants are involved in a number of decision-making debates over a period of time, members of the majority on one decision may find themselves in the minority on another and, hoping that this majority justify their decisions on public rather than private interests, listen to their arguments, take their interests into consideration and vote for the ‘policy outcome that enjoys the widest possible support’, accommodating minority interests where possible. If citizens do not abide by the deliberative obligations themselves when in the majority, then it would seem less likely that others will when they are in the minority (Miller 2000: 152). This is still, then, a pragmatic norm rather than an obligation and it may not always pertain. If people accept to abide by the norms of deliberative democracy for instrumental reasons, then they will abandon this commitment when it is in their interest to do so (Festenstein 2002: 97–8; Goodin 1992: 39). This point, though, ignores the educative effects of debate that have been outlined, whereby citizens accommodate the interests of others into their own. However, in certain circumstances transformation may not occur and deliberative obligations can therefore still go unfulfilled (Festenstein 2002: 98). It seems then that there are deliberative obligations that are not grounded in deliberative democracy itself. For J. S. Mill (1993), a national identity was essential to ground democratic obligations. Similarly, Miller — 93 —

Towards a Deliberative and Associational Democracy

suggests that there needs to be a shared identity that transcends all other identities to provide commitment to these deliberative obligations, a requirement which again can only be fulfilled by a national identity (Miller 2000: 158). Even if we accept the argument that national identity generates special obligations between members, there is no guarantee that these will be deliberative in character. Furthermore, if the national identity is not deliberative, but still valuable, it is not its contribution to deliberative obligations that makes national identity valuable, and so it cannot be the grounding for these obligations: ‘If the argument is that the deliberative features of a national identity make it valuable to its members, it is not clear what the specifically national aspect of those characteristics adds, if we are seeking to derive the deliberative obligations’ (Festenstein 2002: 107). Consequently, Festenstein concludes that deliberative obligations cannot be derived from the deliberative process itself, but rather from a valuable relationship such as citizenship, providing that we accept that citizenship has intrinsic value (Festenstein 2002: 104; see also Richardson 2002: 159). This does not mean the incentive to participate should be noninstrumental, but rather that there is a non-instrumental good in being part of a society that makes decisions in this manner. The point being that, if it is accepted that citizenship has an intrinsic value then citizens may well realise this value and appreciate that valuable relationships entail obligations to each other (Festenstein 2002: 105). It was already argued in the previous chapter that there was an intrinsic value in being autonomous, but this does not mean citizens will be interested in cultivating the autonomy of others, and so may not abide by the obligations. It is the case that if deliberative democracy is to cultivate autonomy, then it requires citizens to abide by deliberative obligations. However, this does not mean that there is no connection between this model of democracy and autonomy. What it does suggest is that deliberative democracy has preconditions, one of which is a certain level of civic virtue, whereby citizens gain some intrinsic value from the relationship. The challenge, then, is how to ensure that this civic virtue is present. I would suggest that this is dependent upon the institutional situation, which will be considered in the following chapters. Conclusion The intention of this chapter has been to argue that the deliberative model of democracy is the most suitable to ensure the cultivation of autonomy, as it was conceived in Chapter 1. The prudential justification was thought to be the best as the proceduralist justification of deliberative democracy, — 94 —

Deliberative Democracy and Autonomous Decision-Making

inevitably appeals to epistemic or prudentialist values. If the arguments from the first chapter were accepted then the epistemic justification cannot be accepted, as it was thought there are no objective values other than autonomy, which was conceived procedurally and, therefore, enabled individuals to find their own reasons for their own opinions. In Chapter 1, it was argued that collective autonomy differs from individual autonomy and that citizens need to form their preferences with consideration towards the opinions of other citizens, if the equal cultivation of everyone’s autonomy is to be ensured. This is effectively highlighted by a distinction between the market and the forum (Elster 1997). In the market one’s preferences do not affect others, but in the forum they do and so should be justified to these other people. Deliberative democracy encourages people to do this because it involves the making of collective decisions involving the participation of relevant actors (the more equal this participation, the more democratic) through the consideration and exchange of reasons aimed at the trans(formation) of preferences, which is the defining element of deliberative democracy. These preference changes cannot lead to consensus because of pluralism, but neither should they lead to consensus, as this can lead to the preservation of the interests of dominant groups who can disguise their partial interests as ‘authoritative knowledge’. Consensus puts undue emphasis on the need for conformity, causing preference change that is not rationally motivated and therefore not autonomous. Deliberative democracy can, though, increase autonomy through compromise, which leads to decisions that all accept and thus respects the agency of all, although again this is threatened by pluralism. Deliberative democracy also encourages the use of public reason, as preferences must be justified to all which encourages citizens to consider the opinions and interests of others and therefore induces people to see others as autonomous and equally important agents. It was further claimed that deliberative democracy cultivates both hearer and speaker autonomy by increasing the availability of relevant information, allowing participants to express themselves freely and ensuring options to vote reflect people’s rationally transformed preferences. Three challenges were considered to these arguments. The first, from social choice theory, claimed that because deliberative democracy would not achieve consensus, aggregation would be necessary, but this meant decisions would be intransitive, ambiguous and open to suspect manipulation. This was countered by arguing that deliberative democracy can aid in the selection of an appropriate aggregative mechanism, increase single-peaked preferences, reduce strategic action, restrict the domain of — 95 —

Towards a Deliberative and Associational Democracy

preferences and further cast doubt over the validity of ‘independence of irrelevancy’ criteria, all of which avoids these negative aspects. The second challenge, from difference democracy, was that deliberation would only benefit dominant groups at the expense of subordinate groups, and was countered when the arguments that subordinate groups participate less, that prejudice was not accessible to reason and that rationality is culturally specific were not accepted. It was also suggested that deliberation could make room for greeting, rhetoric and storytelling. The third critique was that deliberative democracy requires ‘deliberative obligations’ that cannot be grounded in the model of decision-making itself. This argument was accepted, but civic virtue was seen as a possible way of providing a commitment to these obligations. However, the defence of all these critiques is also dependent on how deliberative democracy is institutionalised, as this will significantly influence the extent of domain restriction, determine whether subordinate groups are equally included and affect whether or not civic virtue is present. This brings us to one of the most serious critiques of the connection between deliberative democracy and autonomy: that deliberative democracy is a utopian idea that can only act as a critique of existing arrangements because it cannot be meaningfully institutionalised. Therefore, our attention for the remainder of the book will focus on whether deliberative democracy can be institutionalised effectively, what these institutions would look like and what effects these institutional arrangements will have upon autonomy. Notes 1. Agreement, in such circumstances, could no longer be seen as a ‘truth judgment’, as some see deliberative democracy (Goodin 1992: 131), which further goes against the epistemic justification. Focusing on the ‘truth’ removes the incentive to compromise for those with incompatible conceptions (Warren 2002: 175; Festenstein 2002: 100). 2. Riker agrees, arguing that domain restriction would be unfair and undemocratic (Riker 1982: 117). 3. Voting paradoxes occur ‘whenever the relationship between the voting and the voter preferences is counter-intuitive or unreasonable in some sense’ (Nurmi 1998: 335). 4. As the discussion in Chapter 1 also indicated, this does not mean that higher order preferences are more autonomous than first order preferences, or that the higher order preferences will be decisive. For a preference to be autonomous, it must meet the conditions of the acceptance of preference formation outlined in Chapter 1.

— 96 —

Deliberative Democracy and Autonomous Decision-Making 5. Dryzek also appreciates that domain restriction can also occur through the institutional context. This argument will be reviewed in later chapters. 6. What I do not accept, though, is Knight and Johnson’s (1994) further claim that this makes the preference transformative capacities of democratic deliberation irrelevant. The arguments made above – that deliberative democracy transforms preferences in a way that makes them more compatible with autonomy – still stands. Moreover, agreement on where disagreement lies will usually require preference transformation itself, as there will be different ideas about what the source of disagreement is prior to deliberation. Furthermore, the fact that democratic deliberation can lead to preference transformation will mean that conflict and disagreement themselves are transformed. In short the preference transforming capacity of deliberative democracy still holds. 7. This phenomena could be reduced if voting occurred in public, as Mill appreciated (1993).

— 97 —

3

Institutionalising Deliberative Democracy through Secondary Associations

Introduction The book so far has presented the case that decision-making should be structured in such a way that it enhances, as much as and as equally as possible, the autonomy of all. It was further suggested that deliberatively democratic decision-making is well placed to achieve this. However, principles of autonomy, deliberation and democracy need ‘devices’ or institutions to ‘enact’ them (Saward 2003). Deliberative democracy needs institutions to ‘bring it down to earth’, to ‘give it practical import’ and to ‘make it something real’ (Blaug 1996: 49). Moreover, it was noted in Chapter 2 that the institutional context was key to decreasing the domain, and thereby avoiding transitive and ambiguous decisions, ensuring that subordinate social groups are genuinely included and to provide grounding for the essential deliberative obligations. As the book title suggests, the belief here is that an associational democracy can make important contributions to these factors, make a deliberative democracy ‘real’ and therefore contribute to the cultivation of citizens’ autonomy. The remaining four chapters of the book will deal with this argument. Overall, following the initial, mildly perfectionist argument in Chapter 1, a perfectionist justification of associational democracy is advocated. In constructing a model it is important to refrain from designing a blueprint for the institutionalisation, as the participants themselves should decide the precise details of such a model in relation to their specific context (Gutmann and Thompson 1996: 358). Consequently, the aim of the associational model outlined in the remaining chapters is to simply highlight some of the possible features that could be, and should be, included, if an associational democracy were to approximate the norms of deliberative democracy. Furthermore, I agree with Smith that ‘there is no single best design: different models will be useful — 98 —

Deliberative Democracy through Secondary Associations

in different circumstances, for different purposes at different levels and on different issues’ (Smith 2001: 90). Attention must, therefore, be paid to the context of decision-making (Saward 2003). An associational model is, then, just one of these possibilities, although, for myself at least, it is a particularly attractive one. In a deliberative democracy, however, it would have to be combined with a whole array of other institutions that also fostered key features of deliberation and democracy. In this sense, associations are far from being the whole solution, although the argument here is that they are an important and necessary part of that solution. It should be apparent that the institutionalisation of deliberative democracy is a challenging proposition. In fact, one of the most significant criticisms levelled at deliberative democracy is that it is an irrelevant, utopian and counterfactual ideal because it is unachievable in modern, large and complex societies (Benhabib 1996: 84; Femia 1996; Warren 1996: 242; Miller 2000: 143). If deliberative democracy is unachievable, then it will be unable to actually cultivate autonomy despite the normative potential it has to achieve this key aim of democracy. Perfect deliberative democracy exists only as a theoretical construct, as the ideal speech situation is a ‘methodological fiction’ (Habermas 1996a: 326) that should only be employed to steer debate on institutionalisation (Habermas 1996a: 340): ‘The idealised and demanding conditions of deliberative democracy are aspirational and therefore can only ever be approximated (rather than fully realised) in everyday politics’ (Eckersley 2000: 127; see also Cohen 1997; Leib 2004: 40). The discussion on institutionalisation starts from the premise that the present institutional framework in liberal democracies is inappropriate for the effective institutionalisation of, and close approximation of, deliberative democracy, and therefore a new institutional mix is required (Dryzek 2000: 3; Elkin 2004: 40). This is because the conception of deliberative democracy, as outlined in Chapter 2, requires a reasonable degree of active participation in debate and decision-making by affected citizens. If we look at modern liberal democracies, there is little opportunity for the public to contribute and help form policies through debate; citizens generally choose between a few parties’ already-formed manifestos at election time. Citizens therefore have few opportunities to voice new and original ideas, to expand or change the agenda of parties or government, to participate in the creative dialogical process of discussion or to contribute to decision-making. The argument, then, that decision-making in liberal democracies is not sufficiently deliberative or democratic to cultivate the autonomy of citizens. Nevertheless, the model for institutionalising deliberative and associational democracy offered here accepts a liberal democratic and capitalist framework, and the many limitations this system, and its distribution of — 99 —

Towards a Deliberative and Associational Democracy

power, presents for approximating the ideal of deliberative democracy. This is because closer approximation of deliberative democracy in practice must start from the here and now: ‘alternative institutions cannot be made out of air. Both imagining and enacting alternative institutions must begin with some elements of existing social life’ (Young 1995: 207). Nevertheless, it is still maintained that there is room within the liberal democratic system for significant institutional change that would lead to important normative developments, and the model of deliberative and associational democracy outlined in the remainder of this book, still represents a radical break from the current institutional system. This is not to say that there are no democratic or deliberative features to liberal democratic institutions and political parties, for representative legislative assemblies and the media can, and do, contribute to democracy and deliberation. Nevertheless, there is not enough open public deliberation, nor opportunities for access to these debates for all (Post 1993: 667). A location where citizens can, however, enter into debate is civil society, making it, for many, the most promising location for deliberative democracy (Benhabib 1996; Habermas 1996a; Dryzek 2000; Saward 2000; Warren 2001a; Hendriks 2006). This chapter, and those following, will explore this potential of civil society in detail. The suggestions for institutionalisation should therefore be seen as applicable to all liberal democracies. This is not to deny that there are differences in cultures, institutions, practices and trends in various countries, but rather that there still remain some key similarities that transcend these differences, even if they vary in salience (Elstub 2006b: 18). This chapter considers the specific features of social complexity that present such a challenge to the institutionalisation of deliberative democracy and argues both that deliberative democracy can be institutionalised and that civil society associations will provide a suitable location for deliberatively democratic participation. Specifically, the chapter reviews the potential associations provide to cultivate citizen autonomy. This suggestion is linked to key functions of democracy that secondary associations located in civil society are particularly apt to fulfil. It is argued here that they are suitable locations for governance, provided the principle of subsidiarity is applied; that they can provide effective information and representation; that they can increase and improve the provision of information, and that this enables them to contribute to public discourses in the public sphere; and that they can foster key political and civic skills and dispositions. The fulfilment of these functions enables associations to make indispensable contributions to enabling deliberative democracy to adapt effectively to interconnected features of social complexity that include increased social pluralism, scale, inequality of deliberative and political skills — 100 —

Deliberative Democracy through Secondary Associations

and resources, the increasing reliance on specialists, and globalisation. The fulfilment of these important democratic functions also enables them to enhance autonomy in their own right. Clarification of Terms In this chapter, and for the rest of the book, the terms ‘secondary associations’, ‘civil society’ and ‘associational democracy’ will be used regularly. It is therefore essential to explain and clarify these terms. Due to the sheer diversity of associations, the definition of an ‘association’ needs to be broad. However, I would suggest that G. D. H. Cole’s definition of ‘association’ (1920a) is too broad, although it does provide a useful starting point. He defines association as: Any group of persons pursuing a common purpose or aggregation of purposes by a course of cooperative action extending beyond a single act, and, for this purpose, agreeing together upon certain methods and procedures and laying down, in however rudimentary a form, rules for common action. At least two things are fundamental and necessary to any association: a common purpose and, to a certain extent, rules of common action. (Cole 1920a: 37) More specifically, I am referring to associations that are voluntary (to a degree), secondary and located in civil society. A consideration of each of these factors will provide a clearer conception of what is meant here by ‘association’. Civil society associations are secondary because they are not the primary association of people, which is the state itself. They are voluntary because members choose to join them (but the exact extent to which they are voluntary will be considered in detail later in this chapter). There are many competing definitions of civil society. Engaging fully with these is a useful and interesting process, but beyond the remit of this chapter and book. On a general level, civil society is differentiated from the economy and the state (Cohen and Arato 1992: 20; Dryzek 2000: 23; Young 2000: 158–60; Cerny 2006: 96). Only if this is the case, in a market society, can civil society develop a critical political power because the state incorporates the political sphere of parties, political organisations and parliament, while the organisations of the economy are those involved in production and distribution. Although state and market organisations do share similarities of organisation and communication with those in civil society, and are similarly institutionalised through rights, by contrast — 101 —

Towards a Deliberative and Associational Democracy

they seek to control and manage either the state or economy (Cohen and Arato 1992: viii–ix). Parsons highlighted that states and market-orientated economic organisations are distinct from secondary associations because they are dominated by the media of (legal and administrative) power and money. Associations, however, are mainly constituted by common purposes, identities or interests, and their prime mechanism of organisation is influence (Parsons 1971; see also Warren 2001a: 54; Young 2000: 163). Following this analysis, civil society can be defined as ‘a sphere of social interaction between economy and state, composed above all of the intimate sphere (especially the family), the sphere of associations (especially voluntary associations), social movements, and forms of public communication’ (Cohen and Arato 1992: ix), reproduced through both independent action and institutionalised laws. Marking out precise boundaries between civil society, the state and the economy is far from easy, since individuals and associations ‘are themselves embedded in economic and political-institutional contexts’ (Cerny 2006: 96), and is increasingly difficult as the interactions between these spheres becomes increasingly complex (Hirst 1996: 99; Hendriks 2006: 488–9). To clarify, though, it is not the case that anything that happens outside of the state and the economy should be classified as civil society, but rather only conscious association-building, associational life and institutionalised forms of communication. Associations that are located in civil society are complex, extensive, flexible and diverse, for example, trade unions, business and professional organisations, welfare and charity organisations, service clubs, community associations, recreational associations, environmental groups, educational organisations and cultural organisations (and this list is far from exhaustive) (Van Deth 1997: 1). All of these coagulate ‘multiple loyalties and identities’ (Cerny 2006: 86). They therefore have a huge range of goals, including individual and social, inclusive and exclusive, scarce and unscarce, material and symbolic/psychological goods (Warren 2001a: 94–126), and are ‘overlapping’ (Truman 1951) and ‘cross-cutting’ (Simmel 1955). In addition, associations have a variety of organisational structures which affect opportunities for voice and exit (Warren 2001a: 98). They come in an array of sizes, from those with a handful of members, to those with millions, and can operate at a range of levels, including local, regional, national, transnational and international. Associations are also embedded in, or orientated to, different types of media: market, bureaucracy and association (Parsons 1971). They operate within a multi-level state with ‘multiple points of access’ (Truman 1951). This complex, fluctuating, often spontaneous variety of associations in civil society can lead to what Warren terms a ‘democratic ecology of associations’ (Warren 2001a: 12). — 102 —

Deliberative Democracy through Secondary Associations

Alexis de Tocqueville, in his classic study on Democracy in America (1945), was possibly the first to offer a sustained argument that civil society associations could develop autonomy in their participants through their contribution to democracy (Warren 2001a: 70). Recently, these ideas of de Tocqueville have been rejuvenated and related explicitly to democracy, with associations justified as possible mechanisms to deepen democracy. This advocacy of secondary associations as key units for democratic participation can be termed ‘associative or associational democracy’, defined as ‘a model of participatory democracy based on self-governance of internally democratic, voluntary and functional groups’ (Perczynski 2000: 163; see also Hirst 1994: 112; Bader 2005: 323). This resurgence of the theoretical advocacy of associations as venues for democratic participation has arisen in the context of the perceived failure of the nation-state in ‘postindustrial’ societies (Hadley and Hatch 1981; Martell 1992; Hirst 1994; Cohen and Rogers 1995; Perczynski 2000; Warren 2001a; Fung 2003a). It is to the failures of the state caused by, and contributing to the development of, changes in social complexity – which also present barriers to the institutionalisation of deliberative democracy – that the chapter now turns. The Barriers of Social Complexity From the work of Zolo (1992), Bohman (1996) and Femia (1996), which explores the relationship between democracy and complexity and, in the case of the latter two, deliberative democracy and complexity specifically, it is apparent that modern democracies are plagued by the problems associated with social complexity. For Femia ‘social complexity’ is associated with the ‘number and variety of elements and interactions present’. In addition, it is increasing, primarily due to rapid changes in technology (Femia 1996: 360). This increase in social complexity has led to a decline in the relevancy, potency and ability of the nation-state to fulfill many functions, meet the needs of society and meet standards of democratic legitimacy, and this has led to a reappraisal of the state across the world (Salamon and Anheier 1996: 1; see also Cohen and Rogers 1995; Dryzek 1996; Hirst 1996: 105–106; Beck 1997; Cohen 1999: 211; Young 2000; Warren 2001a: 4). Deliberative democracy itself could help overcome many of these aspects of social complexity. Its capacity to change preferences means that options that were previously unavailable to decision-makers become possible. It further enhances governability by improving the legitimacy of decisions and consequently increasing the likelihood of compliance with these decisions (Barber 1984). Deliberative democracy also helps overcome inequalities, as — 103 —

Towards a Deliberative and Associational Democracy

these are exposed to, and must be justified in, public debate, further enabling subordinate groups to offer reasons against the privileged position of dominant groups: ‘silence serves the wealthy and powerful well, and public argument is a primary means through which poorer and weaker members of society can have influence’ (Warren 1998: 150). Deliberatively democratic decision-making also helps overcome the need for specialists, as it increases the amount of information disseminated. It also includes more people’s preferences (Dryzek 1990; Warren 2002: 194; Parkinson 2006: 1) by including more participants and requiring them to provide reasons and information to justify their preferences, and provides the opportunity for participants to question information and arguments put forth by partisan sources, as well as entering into debate with their own information. Due to this greater amount of information that is made available, democratic deliberation has the instrumental value of improving the quality of the decisions, as these will be based upon ‘better knowledge of the important facts’ (Christiano 1997: 248, 255; Warren 2002: 194) and participants’ preferences will change, becoming more knowledgeable and better informed (Manin 1987: 349) as bounded rationality is reduced (Fearon 1998; Warren 2002: 194). Moreover, these preferences are more likely to be based on good and public reasons, which is not necessarily the case with specialists operating in private committees. Experts still play an essential role in this process, but their status as ‘experts’ is no longer sufficient, and they must establish their authority on an epistemic basis (Warren 2002: 195). However, for these benefits of deliberative democracy to be achieved, it needs to be institutionalised, and the features of social complexity form significant barriers to the meaningful institutionalisation of deliberative democracy (Femia 1996; Bohman 1996). Nevertheless, the discussion below should also highlight that, in addition to presenting significant barriers, some features of social complexity also provide conducive conditions in some circumstances (Bohman 1998). There are many aspects of social complexity, but this chapter will focus on those that are perceived to be the most significant in terms of deepening democracy. The first of these is increased plurality, a result of societies becoming ever more diverse and multicultural. A growing perception of the state is that it excludes certain subordinate social groups, in so far as its universal approach is increasingly unable to take account of an everwidening diversity of needs. This ratcheting-up of complexity has in turn contributed to the corresponding retrenchment of the state as a welfare agency (Cahill 1994: 183). Similarly, increasing pluralism has compromised the effectiveness of traditional and formal representative structures of liberal democracies, as their ability to include all social groups in decision-making — 104 —

Deliberative Democracy through Secondary Associations

has correspondingly declined (Phillips 1995). Increased pluralism also makes deliberative democracy unlikely, as it decreases the chance of reaching consensus on a common good and makes the inclusion of all relevant views harder to attain. The second aspect is that modern societies are too big and contain millions of people, dispersed over large geographical areas, which has led to the centralisation and bureaucracy that characterises present nation-states and means decisions are far removed from citizens, with little potential for meaningful participation. Problems of scale also challenge the possibility of democratic deliberation, with its reliance on participation in discussion. To have all citizens meet together and deliberate together, actually or virtually, is an empirical impossibility, especially if debates are to be inclusive and have depth (Bohman 1996: 2; Walzer 1998: 68; Parkinson 2006: 6, 151; O’Flynn 2006: 98). We must also be wary of the time citizens have available to participate (Adanaby, cited in Schattsneider 1975: xiv; Parkinson 2006: 151), with participation in deliberation being potentially more time consuming than other forms of participation, such as voting. The third aspect of complexity is the inequality of resources and deliberative skills in society that are necessary to participate effectively. These democratic capacities are of two types: civic capacities, which involve civic consciousness and trust, and deliberative skills, which include listening to and analysing the assertions of others, and rationally forming and expressing one’s own preferences in light of available information in a manner that will be persuasive to others. Civic capacities are essential to the equal cultivation of autonomy for all, as they increase the chances of citizens empathising with the concerns of others and considering their arguments with an open mind. As discussed in Chapter 2, autonomy requires other citizens to be able, and inclined, to appreciate one’s situation and needs in order to ensure one’s claims have a chance of being accepted or having influence on the preferences of others and in decisions. Deliberative skills are necessary to participate in collective debate effectively, and deliberative democracy will enhance the autonomy of all its participants only if all have sufficient deliberative skills. It is, then, important to democracy that these skills be distributed widely and reasonably equally. In Chapter 1, I defended deliberative democracy against the claim that these capacities are necessarily culturally specific to dominant groups. However, the claim that all social groups do not currently have the equal opportunity to develop these capacities still stands. Liberal democracies, with their reliance on capitalist modes of production and distribution, have been seen as a key cause of these inequalities. In addition, state structures are thought to be increasingly ineffective at — 105 —

Towards a Deliberative and Associational Democracy

dealing with them, due to their inability to identify and meet citizens’ needs in a manner that will reduce inequalities, or to offer the chances of meaningful participation in centralised state structures. Such inequalities mean that deliberative democracy could therefore effectively lead to rule by elites, particularly in a deliberative democracy (Bohman 1996: 3), as its requirements of reasoned debate raise the threshold of participation compared to other forms of decision-making. These three factors of social complexity are intensified by the fourth, which is the growing need for greater levels of specialism in making decisions. New problems arise as society changes and, correspondingly, the state expands its functions. This combined with constant technological development has led to the emergence of many more problems requiring technical solutions, and therefore decisions are thought to require high levels of expertise. This, in turn, leads to a decline in informed participation, as being informed requires too much time (Femia 1996: 365). Present trends of increasing division of labour, and new technologies, has meant citizens are incapable of participating directly in making decisions, which correspondingly leads to a decline in democracy (Bohman 1996: 151–2; Femia 1996: 364). As well as making lay citizen participation in a deliberative democracy more difficult to achieve, this has led to the decreasing legitimacy of state institutions, as they increasingly rely on experts for policy decisions, which has in turn led to the proliferation of quasi-autonomous non-governmental organisations (quangos) in all aspects of the policy process. The fifth and final aspect to be considered is globalisation, which is itself incredibly complex, and has intensified many of the above elements, too (Cerny 2006: 105). Globalisation has led to, and been caused by, increasing global competition and integration, increasing technological diversity, and rapid change and increased dispersion of the labour market (Cohen and Rogers 1995: 10–11; Hirst 1995: 109; Cohen 1999: 211). This has significantly hindered the potential for democratic control, especially at the level of the nation-state, which now does not, and cannot, monopolise the functions of governance (Hirst 1996: 103). This means the state and its various institutions are ‘less suited than they once were to ensuring a reasonable fair society’ (Cohen and Rogers 1995: 10–11). In fact, the state is now seen as ‘part of the problem’ in terms of its institutions, structures and processes, its capacity to provide coherent collective action and to effectively operate in international affairs (Cerny 2006: 91). Correspondingly, national identity is waning and the power of markets, and particularly multinational corporations, is rising, as they control the most resources to achieve their aims (Cerny 2006: 92–3). These actors now benefit from a — 106 —

Deliberative Democracy through Secondary Associations

virtual ‘veto power’ in many key policy areas (Ringen, cited in Eisfeld 2006: 18), with policies overall being determined by international and transnational factors (Cerny 2006: 94). Globalisation has therefore intensified problems of scale, as it is thought that many decisions now need to be made at a transnational, or even international, level (Held 1995; Dryzek 2006; Eisfeld 2006: 19). This correspondingly increases the plurality of those affected by decisions, and therefore of those who should be included, either directly or indirectly, in decision-making, as the ‘inside/outside’ boundary of the nation state is transgressed, causing the fluctuation and repositioning of the public/private divide (Cerny 2006: 105). Consequently, identifying those who are affected also becomes more problematic. Furthermore, globalisation has contributed to inequalities and fuelled the need for expertise in decision-making. Together, these aspects of complexity provide significant barriers to the institutionalisation of deliberative democracy. However, part of the solution could be associational democracy, to which we now turn. The Functions of Secondary Associations The conclusion drawn from these changing economic, political and cultural aspects of power, attributed to social complexity, is that the nation-state cannot remain as the key focus for political participation (Kohler 1993: 609). It is argued that secondary associations could be an attractive alternative location for direct political participation, for they could enable non-statist planning, decision-making, task-fulfilment and interaction, as well as providing channels for citizens entering into public discourse (Hadley and Hatch 1981; Martell 1992: 166; Hirst 1994; Cohen and Rogers 1995; Perczynski 2000; Warren 2001a; Fung 2003a; Bader 2005; Eisfeld 2006: 15). Secondary associations would then become venues for self-governance, which would reduce the state’s burden. Furthermore, it is hoped that ‘democracy might, via its associative media, expand within and beyond its current state-centred venues’ (Warren 2001a: 9), deepening democracy to a greater degree than was ever possible through the state alone. The arguments that secondary associations in an associational democracy can alleviate many of the problems of social complexity affecting the state will be reviewed. In addition, and central to our debate here, will be the further argument that these secondary associations can also be locations of participation that will enable deliberative democracy to be meaningfully approximated, and therefore also overcome the features of social complexity that cause many commentators to see deliberative democracy as a purely utopian ideal. The contribution that secondary associations can make to the overall cultivation of autonomy, if — 107 —

Towards a Deliberative and Associational Democracy

used as key decision-making actors, will also be considered. All three arguments are linked to key functions that secondary associations are particularly apt to fulfil. Associations as Locations of Governance The exclusivity and overstretching of the roles of the principal fulcrums of liberal democratic institutions, political parties and representative legislative assemblies that were seen as the democratic solution to this problem of scale (Mill 1993) has led to a legitimation crisis for such structures and the state as a whole (Habermas 1975; Hirst 1996; Fung and Wright 2001: 5; Newman et al. 2004: 204). This is because the representative structures and bureaucratic administration that characterise the modern state ‘frequently operate in unjust, unaccountable and ineffective ways’ (Fung 2003a). Consequently, it is argued that, in modern, complex and globalising society, where public services have diversified societies and the polity becomes increasingly differentiated (Rhodes 1997: 7), a single, central, unitary elite cannot ‘exercise positive directive control’, which means that decentralisation and a plurality of organisations is required to avoid ‘governance failure’ (Hyland 1995: 262; Hirst 1996: 103; Bovens et al. 2001). For associational democrats, secondary associations are the solution because they have the potential to be ‘democratically self-governing’. Furthermore, by reducing the scale of decisions, they operate at ‘accessible decentralised levels’ which offer greater levels of inclusiveness in collective decision-making, as citizens can participate more fully and with greater knowledge of the affairs being discussed’ (Martell 1992: 166; Warren 2001a: 196). Potentially, then, if associations are devolved sufficient powers, they could implement legislation and fulfil ‘quasi-public functions’ in support, or in place, of the state, which could help overcome both the problem of size and the need for specialists (Cohen and Rogers 1995: 44; Warren 2001a: 69). They should therefore fulfil ‘as many social activities as possible’ (Bader 2005: 323; see also Hirst 1994: 112), removing much of the need for a ‘central co-ordinating mechanism’ such as the state (Bohman 1996: 156). They could also relieve and redirect some normative pressures of legitimation away from the state, making it easier for the institution to meet standards of legitimacy and freeing up the government’s time and resources to concentrate on other functions (Habermas 1975; Hirst 1994; Hirst 1996; Perczynski 2000: 164). Autonomy is cultivated if these associations are deliberatively democratic, with decentralised powers, as citizens have the opportunity to participate in the decisions and exert more control in the social activities that affect them. — 108 —

Deliberative Democracy through Secondary Associations

There is, however, nothing inherently democratic about decentralisation, as it can mean the restriction or elimination of legitimate participants in making decisions (Warren 2001a: 196), which would be an unjustifiable sacrifice of their autonomy. Decentralisation alters the nature of participation, though, which inevitably changes the nature of political conflict: the outcome of all conflict is determined by the scope of its contagion. The number of people involved in any conflict determines what happens; every change in the number of participants, every increase or reduction in the number of participants, affects the result. (Schattsneider 1975: 2) Decentralisation is, accordingly, only democratic, and only aids in the enhancement of autonomy, to the extent that it ‘socialises conflict’ by the linking of collective actions to collective justifications that include all those affected (Schattsneider 1975; see also Warren 2001a: 201–2). Decentralisation therefore needs to be based upon a sound principle in order to provide guidance on who should receive devolved powers, on what policy areas, to what extent and on how is it should be implemented. One possibility is ‘subsidiarity’, a principle which legislates for both regional and functional decentralisation (Kohler 1993: 617; Bosnich 1996: 1; Mylod 1998: 1) and whose guiding idea is that ‘decisions are taken as closely as possible to the citizen’ (Follesdal 1999: 3). Consequently, subsidiarity can bring collective actions and decisions closer to the citizens they affect, and thereby aid self-governance (Warren 2001a: 191), making deliberative participation available to more citizens (Warren 2002: 188–9). Therefore, what the principle of subsidiarity would legislate for is the removal of many functions currently fulfilled by states to secondary associations (Warren 2001a: 69). Originating in Catholicism, the idea behind subsidiarity is that there are various levels of organisation, and hence there is an apt and relevant level of organisation for each function that society wishes to fulfil. Only if the function cannot be achieved at the lower level should it then be passed up to the higher level. It is then about providing the relevant level of association with the appropriate powers and resources to exercise their rights and fulfil the necessary functions. Subsidiarity is therefore not achieved through the limitation of collective power, as in the pursuit of the free market, as libertarians have suggested (Bosnich 1996: 3). It does not always protect against centralist intervention, but can, at times, legitimise it if it is seen to be the required level to fulfil a necessary function; however, it does place the onus of justification upon the centralists and therefore tends towards decentralisation. It can though, conversely, lead to the passing up — 109 —

Towards a Deliberative and Associational Democracy

of functions to transnational and international levels of organisation, a factor that is on the increase under globalisation. Regrettably, the libertarian/market interpretation of subsidiarity has dominated. In terms of social policy, associations in the USA and the UK have mainly been used as a substitute for social welfare spending (Salamon and Anheier 1996: 2) or what Newt Gingrich calls ‘replacing the welfare state with an opportunity society’, which is essentially focusing the supply of welfare upon the market (Gingrich, cited in Cohen 1999: 229). The USA is said to have the largest associational culture, both actually and relatively, but that this has meant that social welfare provision has been severely restricted (Salamon and Anheier 1996: 98). The recent trends in the USA have gone against its tradition, started in the 1960s, where the government provided direct funding for associations to provide services. There has been a recent ‘explosion’ of associations with devolved powers to fulfil government contracts in the USA (Warren 2001a: 191, 33; Mylod 1998: 2). However, such a trend has led not to increased democratisation of services, but rather to their privatisation and reduced government spending, with little accountability in service delivery being evident (Warren 2001a: 194). In the UK, with ‘New Right’ and ‘Third Way’ ideas in the ascendancy, there has been an increasing market role for provision, which has been part of a growing trend in social policy to reduce the role of the state. In line with this policy trend, the Conservative Government introduced the ‘Citizen’s Charter’ in 1991, the Housing Association Act 1982 and the National Health Service and Community Care Act 1990. For the Conservatives the attraction of the Voluntary and Community Sector (VCS) (as the collection of secondary associations is now termed in the UK) was that it could contribute to the internal, or quasi, market, therefore increasing competition, which in turn should lead to greater efficiency and choice in welfare. Consequently, there was the rise of the ‘contract culture’ in the 1990s, with secondary associations competing for government funding (Elstub 2006b: 19). Under Labour the marketisation of the VCS has continued, with the emergence of private funding initiatives and a continuation of the contracting system, although changes have been made to this inherited system (Whitfield 2002: 129). The Labour government has attempted to establish the VCS as a ‘partner’ in decision-making through policy compacts and with policies such as the National Strategy for Neighbourhood Renewal, Better Government for Older People and Modernising Local Government (Taylor and Warburton 2003: 327–8; Elstub 2006b: 19). In addition, the VCS has been devolved powers to distribute welfare with the aims of achieving democratic renewal and overcoming social exclusion by enabling — 110 —

Deliberative Democracy through Secondary Associations

increased citizen participation. To achieve this, more promising programmes have been instigated, such as Sure Start, Supporting People, Sustainable Communities Plan, New Deal for Communities, New Deal for Young Unemployed People, the 1999 Immigration and Asylum Act and Community Legal Services, all of which are reliant upon secondary associations (Cohen 2002; Harris et al. 2003: 94; Elstub 2006b: 19) but have not embraced the full potential of self-governing associations advocated by associational democrats (Rouse and Smith 1999: 254; Elstub 2006b: 20). More power needs to be devolved to the VCS, as decision-making powers are still retained by elites and government decision-making is still too centralised despite the rhetoric of ‘partnership’ and ‘policy compacts’ (Langan and Clarke 1994: 86; Rao 1996: 161; Forbes and Sashidharan 1997: 486–90; Langan 2000; Whitfield 2002: 139–40; Taylor and Warburton 2003: 328). This feeling was expressed by associational members in a recent empirical study (Parkes et al. 2004: 320–1). Moreover, there is evidence to suggest that this phenomenon is more pronounced for subordinate groups (Clarke 1998), with government bodies and officials failing to consult associations representing excluded groups (Craig et al. 2002; Taylor and Warburton 2003: 332–4). It is therefore important to distinguish the call here for decentralisation of powers to associations from this consumerist/market approach, adopted by certain governments, which aims to reduce collective service and welfare provision, and to redirect the focus for provision to the market, family and charity. The argument made here is about ‘empowerment’ for citizens over essential services and embraces a different philosophy that is in direct conflict with the consumerist approach (Croft and Beresford, cited in Forbes and Sashidharan 1997: 485; Elstub 2006b: 20). Therefore, associations would still require public resources to be established/continued to ensure adequate levels of welfare (Young, 1990: 85). One of the key justifications of subsidiarity is that it enhances autonomy. If there are functions individuals could fulfil themselves, or through participating in a more immediate association, a higher association is denying autonomy by taking away control. Subsidiarity therefore sets the conditions for individual autonomy, but also assumes the capability and the desirability of the individual to be autonomous, given the right conditions: Subsidiarity seeks to enhance the full development of human personality by promoting conditions in organisations of every sort that give individuals the greatest possible opportunity to reflect, choose, and act for themselves, and to take responsibility for the outcomes. (Kohler 1993: 619) — 111 —

Towards a Deliberative and Associational Democracy

The principle of subsidiarity is, furthermore, clearly compatible with both aspects of deliberative democracy: deliberation, as ‘the continuous and active involvement of those directly affected in an ongoing discourse about the way their lives should be ordered is a key feature of subsidiarity’ (Kohler 1993: 622; see also 619); democracy because a normative ideal of subsidiarity is that ‘policies must be controlled by those affected, to ensure that institutions and laws reflect the interests of the individuals under conditions where all count as equals’ (Follesdal 1999: 2). Subsidiarity, then, requires democracy, and even if democracy does not necessarily require subsidiarity, the joining of the two can lead to a deepening of democracy, as it enables citizens more opportunity to directly participate in some decisions by reducing the scale of these decisions. The argument that deliberative democracy is counterfactual because it cannot be implemented on a large scale does not therefore prove the impracticality of deliberative democracy, but rather demonstrates the necessity for the units of decision-making to be reduced. Subsidiarity is the most coherent principle to achieve this, hence deliberative democracy is unlikely to be achieved without it. Follesdal (1999) identifies three strong, mutually supporting connections in relation to how subsidiarity can enable deliberative democracy to adapt to features of social complexity: 1. Reduction of Size: Smaller units, such as secondary associations, are more suited than larger units (communities/nations) to both developing shared interests through deliberation and to securing their representation (Follesdal 1999: 15). As inclusion of all affected becomes more attainable, subsidiarity can also help address increasing social pluralism. 2. Reduction of Domination: Subsidiarity would reduce the exterior domination over preferences of the members of associations as it specifically prescribes the justifiable grounds for ‘exterior’ intervention, providing the ‘institutional space’ necessary for democratic preference formation to be based upon collective deliberation (Follesdal 1999: 15). Such institutional space is essential if preferences are not to be seduced, coerced or manipulated and therefore be autonomously (trans)formed through reasons. 3. Reduction of Agenda: Having fewer participants, and fewer issues on the agenda, means less information is relevant to the decision (Follesdal 1999: 15). As the availability of information and knowledge of available options is a key requirement for autonomous decisions, subsidiarity can make it easier to attain relevant information. When the agenda is reduced, narrowed and localised, — 112 —

Deliberative Democracy through Secondary Associations

participants are better able to attain and understand the relevant information, knowledge and choices, and what these entail, on any specific issue (Fung and Wright 2001: 28). This, in turn, will make their preferences more autonomous, and helps address the need for specialists to make decisions. The reduction of the agenda can also help counter the social choice theory critique, outlined in Chapter 2, where it was suggested that agenda reduction of amalgamated issues can reduce preference domain, which can lead to a more coherent collective decision, which further aids the autonomy of those participating. Although the tendency will predominantly be towards decentralisation, the exact content of what decisions should be taken, and at what level, is not stipulated by the principle and would have to be decided through the political process, ideally a deliberatively democratic one. This presents a significant problem for a transition to a deliberative and associational democracy, as such decisions are initially likely to be made through current processes. However, it is envisioned that nearly all welfare services could be devolved to associations. In addition to enabling deliberative democracy to adapt to several features of social complexity, the application of subsidiarity, and the decentralisation of key roles of governance and service formation and delivery to secondary associations could help overcome the state’s inability to adapt to expanding diversity. As highlighted earlier, one of the principal weaknesses of the welfare state in modern diverse societies is that it can only offer universal social service provision. The state by its very nature, tends to be inflexible (owing to accountability through universal rules) and sometimes arbitrary (as when universal rules produce different results under different circumstances). Because of their distance from social actors, states often have to resort to complex systems of inducements and monitoring to achieve results. (Warren 2001a: 88) Although different social groups have many needs that will be similar, or the same, because they have different social locations, there are usually some distinct needs, too (Young 1990: 185; Cohen and Rogers 1995: 61; Forbes and Sashidharan 1997: 495). As social pluralism increases, so do the differences in need. Differences have increased in Western societies as social relationships have become more plural and diverse. For example, societies have become progressively multicultural; there is a greater proliferation of identities; the number of lone parents has grown; the need for child-care has — 113 —

Towards a Deliberative and Associational Democracy

increased in line with greater participation in the workplace by women; this, and the increases in life expectancy, has increased the need for elderly care; and families are becoming more plural – but still we see policies trying to encourage traditional family responsibility, making state welfare provision increasingly arbitrary (Elstub 2006b: 21). For example, in the UK, Williams (1989: XI) argues that the unique experiences of women and racial minorities have not been recognised, and this has led to racism and sexism in state welfare provision. There has also been a failure to address these issues and offer a ‘progressive welfare strategy which incorporates the needs and demands’ of subordinate groups (see also Langan and Clarke 1994: 75; Forbes and Sashidharan 1997: 492; Hussain et al. 2002). In the UK, specific areas of welfare delivery that fail in this differentiated needs test by not meeting the needs of women are pensions (Walker 1999) and housing and employment strategies for young mothers (Kidger 2004; Giullari and Shaw 2005). Delivery of health and social services which ignores the needs of ethnic and racial minorities seems rife (Drake 2001), for example the provision of domestic violence services (Burnam et al. 2004), psychiatry and mental health services (Langan and Clarke 1994; Forbes and Sashidharan 1997), and social service provision of care for disabled and older people (Langan and Clarke 1994; Hussain et al. 2002). Therefore, racial and ethnic minorities and women are expected to fit in with current and universal service provision, rather than having the provision adapt to their needs (Pascall 1997: 140; Hussain et al. 2002). The solution is to ensure ‘equivalent conditions differentiated by need’, if the needs of all are to be met (Gould 1996: 180). Essentially this means variable delivery to meet the varying needs of different social groups is required to achieve a level of equality in access to services (Elstub 2006b: 21). This exclusion of certain groups is often ‘unconscious’, resulting from ‘well-meaning people’, because it is structurally caused and therefore cannot be overcome without a change to the methods of policy formation and delivery. As Young puts it, these ‘oppressions are systematically reproduced in major economic, political and cultural institutions’ (Young 1990: 41). Young therefore suggests that new institutions for the provision and formation of services are required (see also Fraser 1987: 104; Elstub 2006b). Due to their immense diversity, associations appear to be the most suitable organisational framework to meet the ever-growing differentiated needs and demands of citizens (Taylor 1996: 67; Warren 2001a: 92), an opinion supported by a recent empirical study in the UK on the contribution the VCS makes to democracy from the perspective of members of associations and national and local government representatives (Parkes et al. 2004: 315). The decentralization of service formation provision to diverse — 114 —

Deliberative Democracy through Secondary Associations

associations would also enable minority cultural groups to preserve and nourish their distinct lifestyles (Shachar 2001: 121), and therefore enhance the autonomy of such groups. The processes of decentralisation of key roles to associations would also enable citizens to participate in the definition of their own needs, and in the design and implementation of methods and mechanisms best suited to meet these needs, reducing the control of specialists. Currently, state service provision perceives service users as consumers rather than citizens; accordingly, their participation is not valued or encouraged. According to Martell (1992: 159), state interpretation of people’s needs involves the imposition of factitious ‘objective’ and ‘homogenous’ needs regardless of what citizens want: ‘It disenfranchises citizens from deciding together what their interests could be and how a settlement could be reached amongst them all’, it assumes that ‘the state can somehow express, represent and execute externally and from above plural needs as one unified will’ (Martell 1992: 170). For example, through the use of ‘experts’, women have been excluded from defining their needs, ‘subverting women’s own expertise’ and consequently ‘dominating their lives’. This is not to say that there have not been policies that have been favourable to women, but only that some have resulted in ‘social control and stigmatisation’ (Pascall 1997: 27; see also Fraser 1987, 1989). Williams (1989 and 2000a) argues that it is the absence in debate of members from these social groups that leads to their interests being assumed, something which the institutionalisation of deliberative democracy aims to solve. Decentralisation of powers can help overcome this subordination, as it can ensure opportunities for social groups to form and articulate their needs inclusively, in a manner that will directly affect policy (Powell and Guerin 1997: 63; Elstub 2006b: 22). The decentralisation of roles of governance to associations is also compatible with globalisation, as one of the consequences of globalisation has been the spreading of the ability of governance actors ‘above and below’ the nation-state (Brenner 2004; Hirst 1996: 97), leading to ‘overlapping webs of governance’ (Cerny 2006: 104). For Cerny, ‘a globalising world is a pluralizing world’ (Cerny 2006: 98), as these elements are in fact ‘inextricably intertwined’ (Cerny 2006: 99). Globalisation increases spaces and opportunities for action for an increasingly large array of associations, in comparison to the relatively restrictive, nationally centred, nation-states: Globalisation pulls more and more actors outwards, downwards, and upwards, both forcing and drawing them in to operate on, and to attempt to manipulate and reshape, a complex mix of old and new international, — 115 —

Towards a Deliberative and Associational Democracy

transnational, regional, translocal, and local stages and playing fields. (Cerny 2006: 98) Therefore, an associational democracy is more compatible with the dynamics of globalisation than is the state. Although globalisation, as a feature of social complexity, presents a significant barrier to democratic control and deliberative democracy at the level of the nation-state, it does provide the pluralised environment that is ripe for deliberative democracy to be instigated through secondary associations. The Limits of Associationalism Defended In order to make the case that secondary associations present a superior alternative to the state for fulfilling many processes of governance, it is important to consider three criticisms from Stears (1999), because his arguments are indicative of those that defend the state against the criticisms that have been made above. Stears is concerned that associational provision will increase inequality, is sceptical that ‘needs’ do differ much across society, and also doubts that citizens could, should and would accurately define their needs and the best methods to fulfil them. These arguments will be considered in turn. The first criticism is that a move away from uniform welfare provision would lead to inequality of service owing to the fact that some associations would be better providers than others, due to their having more resources such as money and staff. This will occur even if associations start on a level playing field because they will vary in their provision (Stears 1999: 584; see also Burns 2000: 967). Although, I accept this argument and realise that the scenario it depicts is undesirable, I do not think it supports the conclusion that Stears reaches – namely, that the provision of welfare should remain state centred. If associations were state-funded, then it would not necessarily be the case that some associations had more money per person, although we can accept that associations with higher memberships could have a greater total income and could achieve greater economies of scale as a result (Elstub 2006b: 28). Hirst (1994) thinks that this inequality can be avoided because associations can approximate elements of free market competition. Citizens would be attracted to what they perceived to be better associations, meaning the ‘worse’ associations would have to improve or lose their members. There are, however, problems with this. First, if associations are internally democratic, and are poor at meeting members’ needs, this will be because of the decisions they themselves have made (Elstub 2006b: 28). However, — 116 —

Deliberative Democracy through Secondary Associations

this does then provide an incentive for associational members to make good decisions, as they will be subject to them. Without decentralisation, ‘the consequences of one’s decisions are statistically negligible’ (Fung and Wright 2001: 28), as they are through national and local voting mechanisms (Downs 1957). Second, if certain associations are more efficient and therefore attract more members, this will potentially reduce their capacity for internal democracy, based upon the norms of deliberative democracy, as will be discussed in Chapter 5. This might in turn make the association less efficient at meeting members’ needs. Third, it is questionable to what extent associations can approximate the free market, as different associations have different opportunities for exit, which restricts market competition. There are also many other sources of inequality besides funding, as will be highlighted later in this chapter, so Hirst’s suggestion is not an adequate response to Stears’ criticism (Elstub 2006b: 28). Government league tables suggest state agencies themselves suffer from regional inequalities in service provision, leading to what is commonly termed in the UK a ‘postcode lottery’. Stears’ proposal therefore suffers from his own criticism. However, this point is still relevant if state provision would be more even. More to the point, this argument fails to respond to the claims, discussed above, that in a diverse society ‘equivalent conditions differentiated by need’ are necessary to ensure justice of welfare (Elstub 2006b: 28). The key to this is whether the provision is uneven in order to meet relevant differing needs, meaning that ‘different’ does not equal ‘worse’, as some have suggested (Jordan 2007: 62). Part of the problem with the current Labour government’s approach to associations in the UK has been the continued desire for uniform provision, which has meant the powers devolved to associations have been clawed back through managerialism, regulation and performance measures (Hirst 1996; Parkinson 2006: 51–2). Recent research suggests that the VCS in the UK offers differentiated services because it aims to meet differing needs. However, this does not seem to be the case with state agencies, indicating that their services are arbitrarily uneven (Taylor 1996; Ware and Todd 2001; Warren 2001a; Parkes et al. 2004). Stears rejects the arguments about difference and pluralism, only accepting that individuals’ needs will differ between society and society (Stears 1999: 583). Stears therefore does not accept that needs are open to interpretation, and that there is conflict over what are ‘needs’, believing needs are defined objectively, or at least quasi-objectively, and therefore the state is the best mechanism to decide what these needs are and to provide services to fulfil them. This argument ignores not just the evidence, provided earlier, about the increase of diversity but also the argument — 117 —

Towards a Deliberative and Associational Democracy

that the state has indulged in social control when defining the needs of many social groups. The reality seems to be that neither state agencies nor associations have a consistent definition of needs (Pascall 1997: 27) and they should therefore be established through discourse in which all can participate (Fraser 1992; Gould 1996: 181). The aim of institutionalising deliberative democracy in secondary associations is to try and achieve these aims (Elstub 2006b: 28). Following on from this, Stears (1999: 577) rejects the argument that individuals will know how to fulfil their own welfare needs on the grounds that these issues are too complicated, and experts are therefore needed. He also believes people aim to achieve immediate goods over future goods, meaning the long-term interests and needs of people will not be met without paternalism. Stears (1999: 579) further distinguishes between preferences and needs, and argues that if a mistake about needs is made, this is more serious than mistakes about people’s preferences. For Stears needs are independent of individual choice and cannot change, whereas preferences can, the implication being that meeting needs is too important to leave to the citizens themselves. Here, it can be argued that Stears overestimates the distinction between needs and preferences in the welfare system. For example, Posner has indicated that welfare as ‘preference satisfaction’ is in fact logically equivalent because welfare is inevitably based upon what citizens are willing to pay (Posner, cited in Sagoff 1998: 220). Moreover, needs are essentially contested, are not self-evident and are always open to interpretation (Elstub 2006b: 29). These various interpretations are not ‘unproblematic’ and are therefore politically contested, with different social groups competing to make their interpretation dominant (Fraser 1989: 162–6). Therefore, it is not irrelevant from ‘what perspective and in light of what interests’ needs are interpreted (Fraser 1989: 164). Experts’ interpretations of needs, whether they be the state’s, associations’ or other sources’, are political and therefore should be subject to dispute. Central to the theory of deliberative democracy is the suggestion that not only preferences but also perception of needs can be improved through rational argument (Habermas 1996a: 305–6), and the increase in available information from democratic debate will assist this. Deliberative democracy can enable interpretations of all needs to be compared and contested through reasons, rather than automatically privileging any particular, and inevitably political, interpretation. It is suggested here that a deliberative and an associational democracy combined can enable challenges to the state’s ‘expert’ categorisation of needs (Elstub 2006b). Stears (1999: 579) further indicates that because the associational system allows individuals to define their own needs from a subjective point of view, — 118 —

Deliberative Democracy through Secondary Associations

people are likely to mistake preferences for needs or purposely claim much more than they actually ‘need’. Earlier, it was argued that social inclusion requires differential provision to meet relevant needs. The issue here, then, is how do we establish what are relevant differences? (Gould 1996). In response to the first point, Stears’ claim could equally be made about state representatives’ interpreting needs; it is not made clear how Stears thinks bureaucrats will be best placed to do this. Surely we can maintain that people know their own needs best, and that, although mistakes will be made, these will be less frequent than when bureaucrats and politicians define people’s needs for passive citizens (Elstub 2006b: 29–30). One of the core tenets of deliberative democracy, covered in detail in Chapter 2, is that ‘the force of the better argument’ will prevail (Habermas 1990). In reality, people will not only adapt preferences and interpretations of need because of good reasons but also due to other factors, such as the source of the information, the manner in which it is provided and pressure to conform with the majority. Despite this, ‘reasons’ remain privileged in deliberative democracy, in comparison to other forms of decision-making, and deliberative democracy does not accept pre-political and unreformed preferences and interpretations of needs (Elstub 2006b: 30). As argued in Chapter 2, this will mean preferences and conceptions of needs are more autonomous. If long-term needs can be rationally proven to be superior to short-term ones, then deliberative democracy, with its exchange of reasons, will be better situated to encourage people to focus on long-term, rather than shortterm, needs. Furthermore, in the deliberative situation, information and arguments will be available from ‘experts’ and many of the members of associations will become ‘experts’ themselves, if they are the paid officials or long-term volunteers who act as permanent members of the association or as representatives of that association. But they will have to justify their assertions with reasons, which will help alleviate the aspect of complexity that is the domination by specialists (Elstub 2006b: 30). Similarly, in terms of the second of these points by Stears, in a deliberative decision-making framework, it would not be enough for one person to claim a preference is a need to have this met, because they must convince others that it is a need as well. The decisions around defining needs and how they are to be met are, then, not individual but collective decisions, where each person will be one voice in the deliberative arena (Elstub 2006b: 30). All groups will tend to highlight common justifications for their ideas, needs, demands and so on, but it is how persuasive the reasons are that will determine whether these will be accepted or not (Young 1990: 185–6; Sunstein 1985). Different needs will have to be justified through reasons in order to be accepted as relevant by others. It is this — 119 —

Towards a Deliberative and Associational Democracy

interchange of reasons that will help in ‘judging better and worse interpretations of people’s need’ (Fraser 1989: 181). Stears thinks such an argument just undermines the normative claims of participatory associations: ‘If, therefore, associationalists accept that subjective preferences should not be the determining feature in shaping welfare provision then they are left without their claim against the structure of current arrangements’, that is that they are uniform and inflexible (Stears 1999: 581). The dispute seems to be over whether needs are socially constructed, with human agency playing a central role in the construction of competing interpretations, or are real and waiting to be discovered or, more accurately, interpreted. If one sides with the realist interpretation, then this still does not preclude the existence of differing needs, even if we deny that welfare should be based upon subjective preferences. Therefore, the arguments that the centralised state is too distant from citizens to consult them in order to find out what their needs are, and is too inflexible to be able to supply a genuinely differentiated service, are still relevant. Likewise, the suggestion that the current state mechanisms that exist are inadequate to enable citizens to participate in democratic debate about their needs and how they can be best met, are valid. And if the needs of each individual citizen are objective, unlike preferences, this does not mean that anyone’s interpretation of anybody’s needs is necessarily accurate prior to being involved in a process of collective deliberation, where new information and the perspectives and experiences of other citizens is made available (Elstub 2006b: 31). Furthermore, as Fraser argues, if one believes needs are socially constructed and interpreted, this does not lead to complete relativism, but rather highlights the importance of ‘interpretive justification’, which should be through ‘communicative processes that most closely approximate ideals of democracy, equality and fairness’ (Fraser 1989: 182), which is what deliberative democracy does. Therefore, in both objective and subjective conceptions of need, deliberative democracy should play an essential role in determining between competing interpretations. Nevertheless, Stears’ criticisms do highlight why it is so important for deliberative democracy to be the organisational norm for associations if they are to distribute services, as without this the associational arguments are open to many of the faults he attributes to them. But, just as an associational democracy aids the institutionalisation of deliberative democracy, deliberative democracy aids associational democracy (Elstub 2006b: 31). A final criticism, not given by Stears, but still relevant to the role of associations providing key services, is that important needs will not be provided for because associational provision will be patchy. This would — 120 —

Deliberative Democracy through Secondary Associations

certainly be the case at first, but the greater number and extent of opportunities to deliver services, the stronger the incentive to form associations to deliver them (Offe 1995: 127). Therefore, if the state devolves power and duties to associations, and provides funding, as advocated here, then this will encourage more associations to form, particularly associations to meet the needs of subordinate groups, who perceived that under a state relatively unresponsive to associations, or only responsive to the main dominant associations, they would have little or no chance of affecting policy and service delivery (Elstub 2006b: 31). If associations become an increasingly dominant method for delivering services, then more people would seek to become members of them and overall their public profile would be increased, enabling secondary associations to come ‘out of the shadows’ (Salamon and Anheier 1996: 116). Associations and Representation Even in an associational democracy, where the principle of subsidiarity is prevalent, and where secondary associations fulfil many functions currently fulfilled by the state, not all functions of governance will be devolved to associations. Broader policies at local, national, regional, transnational and international level will be made, and citizens need to be represented in these processes if they are to be democratic. Representation has been seen as essential to democracies in order to address the problem of scale (Mill 1993), as it enables all to have their preferences included in debate without all having to directly participate, and also helps overcome inequalities in participation because representatives are thought to represent the interests of their constituents more effectively than the constituents themselves. In addition, it relieves demands on excessive participation, helping to meet the challenge of ‘time’ and the ‘work – democratic life’ balance, whilst still ensuring citizens’ views and opinions are incorporated into decision-making processes. However, the mechanism of representation is key to the achievement of these aims, and the claim here is that secondary associations are particularly apt at providing the relevant representation required for deliberative democracy. This is because they enable a diversity of citizens with similar beliefs, preferences and needs to combine their voice and therefore increase the chance that they will be heard by others citizens, associations and the relevant state agency, in a detailed manner. The plurality and flexibility of associations means that collections of people from social groups, and with an array of beliefs and interests, can form associations in order to form and represent their interests and — 121 —

Towards a Deliberative and Associational Democracy

preferences, giving voice to groups excluded by present institutional mechanisms and their media of power and money. They can help equalise representation because citizens without adequate resources for political mobilisation can combine their resources and so increase their potential political influence, for which the commitment of the members is key, and this is more evenly distributed than money, which can be accumulated (Warren 1998: 19). Furthermore, they represent interests that are not territorially-based, which would go un(der)-represented through party politics, and therefore can overcome the restrictions and limitations of territorial representation (Herring 1929; Mansbridge 1992: 41; Cohen and Rogers 1995: 42–4; Warren 2001a: 83–4; Fung 2003a). They have a ‘variable geometry’, as they are flexible enough to represent their members at local, regional, national, transnational and international level. Indeed, there has been a growth of NGOs operating at these broader levels in response to globalisation (Cerny 2006: 94), leading to a ‘global civil society’ (Edwards 2004) in which associations are unevenly, and fluctuatingly, ‘strategically located’ to provide effective representation throughout the multiple and multiplying decision-making arenas (Cerny 2006: 96). There is a need for functional representation because social and cultural groups, and many distinct interests and beliefs, are not territorially-based, but should be represented and included in the decision-making process. In a democracy it is unfair for dominant groups to monopolise representation, as this is not consistent with the cultivation of autonomy for all. Consequently, subordinate social groups have looked towards associations to provide representation throughout history, making it the ‘critical resource for those who lack influence based on economic resources, cultural hegemony, prestige, and so on’ (Rosenblum 1998: 208). Associations therefore add to speaker autonomy for citizens, increasing their ability to have their views heard by others, which in turn enhances hearer autonomy. Associations and the Provision of Information Much of the content that secondary associations represent is information that they have formed, collected and organised. They are particularly useful at this because they specialise in certain areas which are of particular relevance to their members, and therefore help overcome, or at least counter or alleviate, the need for specialists in decision-making. Such information helps associations hold government officials and institutions accountable (Mansbridge, 1992; Cohen and Rogers 1995: 65; Christiano 1996), and is likely to be more local and practical, more detailed, refined and abundant, than information from others forms of representation (O’Neill, — 122 —

Deliberative Democracy through Secondary Associations

2003; Fung, 2003a). Furthermore, secondary associations create a division of labour in the collection and organisation of information, achieving economies of scale that enable citizens to acquire levels of information that they would be unable to obtain by themselves and therefore contribute to overcoming inequality in information (Hirst 1994: 34–40; Cohen and Rogers 1995: 42–3; Warren 2001a: 71–2). In addition, due to their close involvement with their members, associations can provide information that would otherwise be unavailable to the state, such as experiential knowledge (Cohen and Rogers 1995: 43; Davies 2007: 56), which, as discussed in Chapter 2, is vital to ensuring inclusion in the deliberative process (Young 1996; Sanders 1997). Through these capacities associations can make important contributions to external rationality necessary for autonomous preference formation and decision-making. In addition, this representation of information enables a level of coordination in a disparate and plural society that is unachievable by the state and market. Environmental policy, provides a good example of this, as it is currently limited due to the problems that the state has with ‘command, control and co-operation’ in establishing environmental public standards in the face of a diversity of sites, enforcing compliance to the standards and gaining co-operation in setting standards. Greater co-operation from a plurality of associations could lead to more relevant and detailed specialist information about environmental damage and costs of environmental protection. They can provide co-operation from members to agreed environmental legislation and in the implementation of environmental protection methods. Associations can further help in the process of dissemination of knowledge and information about the new measures to other groups, such as consumers, all in a manner unobtainable to state agencies (Cohen and Rogers 1995: 58). Hayek concurs with this failing of the state, famously arguing that it is not possible to communicate all information to a centralised planner (Hayek 1937). Associational democracy, by pluralising social activities and their coordination through a huge array of associations, removes the need for one, centralised co-ordinator, and aids in the retrieval, dispersal and co-ordination of this disparate array of information and knowledge (although these same problems still occur in terms of co-ordinating information within each association, although to lesser and varying degrees). However, Hodgson (1998) maintains that these co-ordination problems still affect associationalism, as central to Hayek’s thesis is the suggestion that much of this information acquired through specialisation is partial, fragmented, practical and habitual in nature, and only tacitly held by individuals, which means it cannot be the subject of rational deliberation (Hayek 1937; Hodgson 1998; — 123 —

Towards a Deliberative and Associational Democracy

Pennington 2003). This is a serious challenge to the viability and desirability of deliberative democracy per se (Pennington 2003), as well as to its possible institutionalisation. Consequently, for Hayek markets are the only mechanism that can co-ordinate and disperse tacit information and knowledge (Hodgson 1998). Drawing on Neurath’s associationalism, O’Neill turns to scientific knowledge to demonstrate the co-ordinating failings of the market, and the superiority of secondary associations in this department. Taking Hodgson’s (1998: 409) premise that tacit knowledge provides much of the foundations of science, O’Neill highlights the fact that scientific communities are a beacon for co-ordination of this information and knowledge, but achieve this largely without the market (O’Neill 2003: 200). The market can disseminate some types of information, but not all e.g. scientific, merit, health and love. Furthermore, the market can be directly responsible for the destruction of knowledge that is local and practical, as we have seen with the growth and dominance of global markets, as many actors and their information and knowledge become marginalised due to their lack of relevant resources to compete effectively in the market, or because it has no market value at all, or because it cannot be transferred across cultures. Again this can be highlighted through environmental examples, with local knowledge and information of soil conditions and crop varieties having little market value and in danger of being lost in market-dominated societies (O’Neill 2003: 201–2). Secondary associations, in many instances, will therefore be the superior mechanism to the market at pluralising, and therefore increasing, co-ordination and dissemination of information and knowledge, as they reduce the scale of decisions. This makes the inclusion of all affected, and all relevant knowledge and information, accessible and consequently makes deliberatively democratic decision-making achievable. However, for Pennington, Hayek’s analysis here definitively means that tacit knowledge and information cannot be communicated linguistically, rendering deliberative democracy redundant with respect to this ‘large body’ of social knowledge (Pennington 2003: 731). In contrast, as was discussed at length in Chapter 2, it has been theoretically asserted that such knowledge can be linguistically communicated, and that this is an essential aspect of deliberative democracy, if all speakers in multicultural and diverse societies are to be included (Young 1996; Sanders 1997; Dryzek 2000; Miller 2000). Empirical evidence further suggests that such communication does occur in deliberative events (Parkinson 2006: 139–42). Pennington’s mistake is, therefore, to assume that deliberative democracy is solely reliant on reason as a form of communication.

— 124 —

Deliberative Democracy through Secondary Associations

Associations and Informal Public Spheres In exchanging, representing and communicating ideas, information, beliefs and preferences, secondary associations generate deliberation and form a generalised debate in the informal public sphere. Public spheres can be characterised as spaces ‘in which citizens deliberate about their common affairs, and hence an institutionalised arena of discursive interaction’ (Fraser 1992 110; see also Warren 2001a: 34). By participating in the informal public sphere, associations bring new speakers into public debate and change its parameters, too. The role of associations as communicators in the public sphere is an intrinsic one. This is because associations are established through communication between individuals themselves and because many try to influence the preferences of the general public, and members of other associations, by representing and voicing the views and interests of their members, trying to convince these other actors in the informal public sphere of their validity (Bohman 1996: 138; Habermas 1996a: 369; Warren, 2001a: 78–80). To achieve this, associations must be able to ‘employ and appeal to norms of publicity’; limiting their potential as strategic actors (Habermas 1996a; see also Mansbridge 1992). Through contributing to the development of informal public spheres, associations therefore increase the prevalence of public reason in civil society, and between civil society and the state, which was established as a vital requirement to autonomous collective decisionmaking in Chapter 2. The public sphere then mediates between these associations (Cohen 1999: 215). These informal public spheres can appear at local, national, transnational, international or functional level (Germain 2001), making them vital to democracy in the global era, as the state is ‘hollowing out’ and institutions and modes of governance and debate increasingly occur at multiple levels (Bache and Flinders 2002; Rhodes 1997). They also help overcome the scale problem, as the public sphere transcends elements of time and space, potentially enabling all to participate in an anonymous discourse. Increased Choice for Political Participation and Representation In addition to providing opportunities for governance and representation, associations also increase the choices available for governance, political participation and representation, due to their relative levels of voluntariness in comparison to more rigid, territorially-fixed structures of participation and representation in liberal democracies. This freedom of association is ensured through legal rights and is central to the promotion of autonomy of — 125 —

Towards a Deliberative and Associational Democracy

citizens; it is therefore essential to the promotion of the key values of liberal democracy, too (Dahl 1989: 233). In addition to adapting to social pluralism more adeptly, and generating opportunities for deliberation, this function makes an important contribution to the cultivation of autonomy itself. One of the key requirements for autonomous decisions, established in Chapter 1, was that they are voluntary and therefore the fact that secondary associations are, to a degree at least, voluntary, means that they are compatible with autonomy. Moreover, if they are voluntary, if they become key locations for decision-making, as associational democrats advocate, then citizens will get to choose the mechanism through which they contribute to collective decisions, which would lead to the further cultivation of autonomy. As was discussed in Chapter 1, for a choice to be genuinely voluntary it must be free in the positive and negative sense, which requires there to be an acceptable range of options (in this case, associations to join) and knowledge of these options (again, in this case, associations). Furthermore, the choice must not be coerced and the preferences the choice is based upon should not be the result of manipulation and seduction. A choice to join an association can therefore be more or less voluntary, depending on the extent to which it meets these criteria. Associational membership is negatively free because associations are voluntary in the sense that they rely on active consent to join and therefore the members get to choose which association would perform a function for them (Hirst 1994: 24). In comparison to communities, associations also contain relatively high opportunities for exit. Therefore ‘voluntariness’ of association is a principle of social provision which contrasts starkly with state collectivism, where there is little choice and little negative freedom (Hirst 1994: 4). As Hirst claims, associations are ‘communities of choice’ and not ‘communities of fate’ (Hirst 1994: 49–56). Associations allow for citizens to be social and political without being dominated by an all-encompassing shared way of life or community norm (Warren 2001a: 45). Such encompassing identities, that characterise community relations, can hinder opportunities for deliberative democracy: ‘A strong community is constituted in such a way that its practices and traditions are securely interconnected with its social functions, which are in turn closely related and integrated with its members’ identities’ (Warren 2001a: 46). Communities are less capable of dealing with the increasing plurality of preferences through democratic debate than are associations, as force is invoked more often. Exit is much easier in associations, meaning force cannot be used so easily and this means debate is often more likely to be used to resolve conflict. Furthermore, associations usually have a narrower — 126 —

Deliberative Democracy through Secondary Associations

focus than communities, allowing participants to put aside differences such as ‘religion, nationality, occupation, ethnicity, and so on’ and agree on single purposes: ‘In contrast, in any complex and pluralistic society the (encompassing) communitarian impulse to connect every issue and identity tends to stop collective action in its tracks’ (Warren 2001a: 46). Due to the ‘democratic ecology of associations’, like-minded people can form an organisation around any shared theme, for example, occupation, hobby, identity, political interest, economic interest, ideology, religion and so on, but communities generally involve a variety of these themes (Young 1990: 184). Associations represent many identity-based, cultural, class and religious social groups, which is considered to be essential to achieving inclusive multicultural democracies. The protection of these identities is similarly thought essential to achieving individual autonomy (Kymlicka 1995). Membership of such social and cultural groups is not voluntary to the same degree as interest-based associations, as membership is less contingent and usually derives from birth, with highly restrictive opportunities for exit. This has prompted Eisenberg to talk of the ‘necessary illusion of the voluntary association’ (Eisenberg 2006: 71). Nevertheless, given the huge variety of associations, there is still choice over which associations to join, have represent and receive services from, for people from these social and cultural groups, making them, to a degree not recreated in geographical communities, voluntary. Not only are the types of associations diverse but so too are the motivations for joining them, as are the bases of perceived membership (Rosenblum 1998: 5). Rosenblum similarly warns of making the libertarian error of classifying all associations as voluntary and further suggests that ‘there are always alternative understandings of an association’s nature and purpose, and competing classifications’. Voluntariness is one of these (Rosenblum 1998: 6) and Roßteutscher (2000) questions the voluntariness of associations in an associational democracy. He argues that if secondary associations are to be key providers of essential services, be locations for governance and provide representation (as was argued above), then people have little choice but to become members of some secondary associations. There is a challenge, then, to the negative freedom involved in the choice to be a member. However, once again, citizens will still have a myriad of associations to choose from. Therefore, it will not be compulsory to be a member of any particular association, providing there is an acceptable range of choices. In terms of providing an acceptable range of choices, secondary associations, by their very nature, are diverse, not only embracing but also providing the opportunities for a plural society due to the ‘dense social — 127 —

Towards a Deliberative and Associational Democracy

infrastructure’ that they form. Consequently, they offer a wide range of options for citizens to choose from. Moreover, in an associational democracy, associations would become a location of central importance for political participation and service provision, and the likelihood is that more associations would be mobilised, increasing the acceptability of this range of options, especially if more funding and resources were made available to achieve this (Offe 1995: 127; Elstub 2006b: 31). In terms of current awareness of these options, there may well not be sufficient knowledge of the array of associations amongst citizens; however, in an associational democracy, it is likely that citizens will make more of an effort to acquaint themselves with the available relevant associations, and that associations will make more of an effort to ensure citizens are aware of their existence and aims. This could be facilitated by a charter of all associations, for example an expansion of the UK’s present National Centre for Voluntary Organisations. In addition, the media is likely to cover associations in greater scope and detail as they become primary political actors, which will in turn increase public knowledge of these associations. However, it will be important for associations to publicise themselves and actively endeavour to recruit potential members (Rosenblum 1998: 189). We should therefore not overestimate the level of choice for political participation and representation that associations present, and should not conceive of them as completely ‘voluntary’, as this can ‘implicitly marginalize groups that cannot meet this standard’ (Eisenberg 2006: 77). Nevertheless, in contrast to purely geographically-based and rigid structures, secondary associations do offer greater choice, aiding the cultivation of autonomy, adapting better to high levels of social pluralism and being more conducive to deliberation. Schools of Democracy For deliberative democracy to effectively cultivate the autonomy of all, all need relatively equal skills and chances to engage in debate (Behabib 1996; Cohen 1998). However, liberal democracies are rife with inequalities of resources to form and participate in associations, many of which both derive from and cause inequality in the distribution of the democratic capacities required for effective participation. Overcoming these inequalities is essential if we are to ensure that deliberative democracy is not elitist, enabling the educated, with the most advanced rhetorical skills, to dominate. However, it has been suggested that active and equal participation in secondary associations can provide the circumstances necessary for democratic capacities to be developed (Warren 2001a: 61). In addition, it is thought that — 128 —

Deliberative Democracy through Secondary Associations

associations can provide ‘the free spaces’ where ‘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’ (Evans and Boyte 1992: 17–18; see also Mill 1993; Putnam 1993 and 2000; Walzer 1994: 189; Cohen and Rogers 1995; Verba et al. 1995; Galston 2000a) that make citizens more likely to participate, co-operate and consider the interests of others. These are vital values to deliberative democracy, and the cultivation of autonomy, but importantly could also mean that participation in secondary associations may provide the sense of citizenship that Festenstein (2002) has suggested is necessary to ground deliberative obligations, and that was highlighted as a key problem for deliberative democracy in Chapter 2. Both are strains of thought which date back to de Tocqueville (de Tocqueville 1945 vol. 2: 117) and J. S. Mill (1993). De Tocqueville (1945) claimed that associations can develop trust because people are encouraged to form bonds away from the primary associations of family and friends, which in turn enables people to become aware of the consequences of their actions on others, and therefore their interdependency. Mill (1993) argued that it was through participation at local and decentralised levels that citizens learnt key political skills and broadened their political outlook, but the most recent and dominant articulation of the argument comes from Putnam (1993 and 2000), who argues that participating in associations creates ‘social capital’, an important aspect of which is ‘generalised reciprocity’, amongst members, whereby interests are broadened and become more public in orientation. If this is the case, participating in associations can provide people with a sense of responsibility, and ‘enlightened self-interest’, with members becoming aware of their mutual dependency with members of other associations and appreciating the relevance of their interests, needs and preferences, thereby fostering the civic consciousness and trust that is necessary for collective action (see also Olson 1972; Barber 1984; Mansbridge 1995; Habermas 1996a; Warren 2001a: 73). In comparison with market and state relationships based on inequality, hierarchy and compulsion, associational relationships are more voluntary and equal, consequently they are based upon consent, which deepens these civic capacities and autonomy, too (Warren 2001a: 42). Therefore, trust and concern for the public good cannot be generated at the level of the nationstate, but only in arenas such as secondary associations, with powers decentralised to them (Elkin 2004: 55–7). This phenomenon is likely to be enhanced if the relationships between associations and their members are based upon deliberatively democratic communication. This is because collective deliberation encourages people to offer public justifications for — 129 —

Towards a Deliberative and Associational Democracy

their preferences, and to listen to the opinions of others. Furthermore, Putnam has argued that once these capacities of trust and civic virtue have developed, citizens can then co-operate to solve collective problems, which in turn helps develop trust and civic virtue even further, so a ‘virtuous cycle’ is developed (Putnam 1993; see also Verba and Nie 1972: 186; Elsdon et al. 1995; Van Deth 1997: 14 for further empirical evidence to support Putnam’s claims). Olsen claims all secondary associations can contribute to the development of democratic skills and civic trust (Olsen 1972: 319). However, although trust and civic virtue could be generated within an association, it seems unlikely that they will be generated across society and between associations: How does intragroup trust become trust of strangers outside the group? Why does the willingness to act together for mutual benefit in a small group such as a choral society translate into willingness to act for the common good or to become politically engaged at all? (Cohen 1999: 219–20) Cohen is further sceptical that ‘the interpersonal trust generated in face-toface interactions [is] the same thing as ‘‘generalised trust’’ ’ (Cohen 1999: 220). Interpersonal trust is, by its very nature, specific to its context, and needs reciprocation to be directly experienced, so cannot ‘simply be transferred to others or to other contexts’ (Cohen 1999: 221). In addition, associations are, by their nature, exclusionary and competitive, at best providing the location for ‘shifting involvements’ of individuals (Rosenblum 1998). In Putnam’s (2000) own terms, it is more likely that participation in associations will generate ‘bonding social capital’ in homogenous networks rather than ‘bridging social capital’ in heterogeneous relationships, due to the social homogeneity of associations that results from their voluntary membership (Mutz 2006: 35–6). Cohen’s and Rogers’ more minimal claims that associations can promote a ‘civic consciousness’, defined as a recognition and commitment to democratic procedures and norms as the basis for social co-operation and trust in the commitment of others to do the same (Cohen and Rogers 1995: 43–4; see also Rosenblum 1998: 59; Warren 2001a: 7), therefore seem more accurate than Putnam’s. Nevertheless, ‘civic consciousness’ should still be sufficient to ground deliberative obligations. Problems for Associational Democracy Despite the potential of secondary associations to fulfil these key democratic functions, which help deliberative democracy in practice to adapt to many — 130 —

Deliberative Democracy through Secondary Associations

of the features of social complexity, there are still a number of significant problems that associational democracy would face. Those considered here are the inability, for a variety of reasons, of all associations to be able to fulfil all these functions, and the high levels of socio-economic inequalities that are present in Western capitalist liberal democracies. The Variability of Associations Due to the variability of associations that has been highlighted, not all associations will be able to fulfil all the functions that are attributed to them in this chapter. The fact that they are apt to fulfil one function may well mean they are unsuitable to fulfil another. Moreover, some types of secondary association hinder democracy, rather than promoting it (Fung 2003a: 515). For example, the principle of subsidiarity only outlines the idea that functions must be fulfilled at the lowest possible level of association; it does not mean this is the case whatever the type or nature of that association. Associations will be more suitable for subsidiarity if their aims are not contested, and this is enhanced by a low cost of exit whereby members are more likely to leave if there is substantial dispute than to stay and contest. Obviously associations that do not aim to perform collective functions will be unsuitable for subsidiarity (Warren 2001a: 191–3). Neither are all associations equally apt at providing representation or at contributing to the informal public sphere. Vested associations will not be good at representing differences, as this involves providing opposition which involves the association sacrificing some of its established interests and relationships to other associations or the state. Moreover, if they are attaining benefits for their members, they do not want this to be highlighted and become a public issue in the public sphere. In fact, they usually operate to keep such issues off the public agenda because they have something to lose if that ‘difference’ becomes an issue. Groups that generally benefit from the status quo like this might be the CBI and BMA in the UK. Consequently, vested associations highlight commonalities rather than differences, which has some positive and negative dimensions for deliberation: positive in the sense it will help groups move towards consensus upon a common good and reach deliberative compromises, but negative in the sense that it excludes certain groups from this deliberation (Warren 2001a: 173). Effective representation is further hindered if the association is not united, and offers conflicting claims, preferences, goals and reasons. This is more likely in associations with low opportunities for exit (Warren 2001a: 185). — 131 —

Towards a Deliberative and Associational Democracy

There are a number of factors which affect an association’s ability to develop civic and deliberative skills. First, they will be enhanced by an association that deals with collective action (Rosenblum 1998: 206; Warren 2001a: 72), as associations involved in conflicts, regardless of whether this conflict be internal or external, will provide increased opportunity for such skills to be used. Associations that are politically orientated will also present more opportunities for participants to develop these aspects. Furthermore, the fewer opportunities for easy and low-cost exit from the association, such as neighbourhood associations and housing associations, the greater the chance of developing political skills, as this encourages members to internalise political conflict and again will provide opportunities to develop political skills. Groups with high opportunities for exit can still develop members’ political skills, providing their focus is the development of public material goods, and inclusive social goods, as these can only be achieved through co-operation. These factors are strengthened if the association is embedded in social media, as this focuses the association on commonalities rather than conflict, which in turn is concentrated if opportunity for exit is high (Warren 2001a: 152). This variability of associations is not essentially a problem, as the great strength of associational life is its plurality, which means all the functions can still be fulfilled, providing there is a diversity of specialised associations to form a ‘democratic ecology of associations’ (Warren 2001a: 12). As discussed earlier, this ecology is likely to be increased in number, and diversity, in an associational democracy. Socio-Economic Inequalities The most significant problem for the associational system, advocated here, are the current levels of socio-economic inequalities that plague liberal democracies, and the associational system itself. The current political processes of interest group competition are analogous with the market, with interest groups competing for the loyalty of citizens and money, which is then used in competition to lobby government, contributing to distributive unfairness. Those with greater resources – for example, associations pursuing business interests – are best able to represent their interests and so policies continue to be biased towards these already dominant groups (Manin 1987: 355; Young 1990: 72; Cohen and Rogers 1995: 41; Achterberg 1996: 169–70). This is especially so when many policies are zero-sum games, meaning that if some associations win, others lose (Cerny 2006: 87). Inequalities in power and money are perpetuated in associational mem— 132 —

Deliberative Democracy through Secondary Associations

bership (Schattsneider 1975; Verba et al. 1995: Chapter 12; Salamon and Anheier 1996; Van Deth 1997: 9; Skocpol 1999: 66–73). Barriers to participation include the lack of income and education, and the presence of discrimination, which prevent equal participation in the associational system (Verba et al. 1995). Individuals ‘do not simply ‘‘join’’ associations; they are recruited’, and ‘dispositions’ to join associations are affected by factors like ‘ghettoization’ and ‘chronic unemployment’, which results in people lacking the necessary resources to form their own associations and/ or the opportunities to be recruited into existing ones (Rosenblum 1998: 189). Consequently, socio-economic inequalities mean some people have much more choice over which associations they will join, if any, meaning they are not voluntary to all in the positive sense of freedom. Therefore, the greater the socio-economic level, the greater the level of associational participation, and this reduces the potential of associations to instil civic virtues and political skills throughout the citizenry (Gutmann 1998: 3–31; Skocpol 1999). Similarly, informal public spheres are plagued by inequality of access, which affects its potential to fulfil deliberative roles in a democratic manner, enabling the discourses of the powerful to dominate and ‘crowd out everyone else’ (Mansbridge 1992: 48). Moreover, Fraser argues that socio-economic inequalities cause the cultural ethos developed by socioeconomic groups to be unequally valued. She further suggests that in everyday life, and within the public sphere, such powers are magnified because inequality in the political economy affects opportunities for access to participation, therefore public spheres are not, and cannot be, neutral and equally ‘expressive of any and every cultural ethos’ (Fraser 1992: 120). In a deliberative democracy, or any democracy, it is essential that ‘the political sphere must be protected from being determined by spillover effects from social or economic inequalities in the society’ (Fishkin 1991: 31). These inequalities seriously threaten the ability of an associational and deliberative democracy to cultivate the autonomy of all, and such a system could in fact be quite elitist if these socio-economic inequalities are not alleviated. The undemocratic effects of inequality will be softened by pluralism. If citizens have multiple and fluid membership in associations, then the inequalities from each sphere should be contained, to a certain extent, as well as ensuring that democratic power is not determined by any single ascriptive characteristic (Warren 2001a: 215). Therefore, associations can prevent socio-economic injustices being translated into the association to which one is a member. For example, those with low-status occupations can still receive high levels of respect and prestige in an association if they have certain attributes and skills that are useful to the association. In a local — 133 —

Towards a Deliberative and Associational Democracy

football club a lawyer could be playing with a bin collector, or in an association providing support for those who have suffered from mental illness a cleaner and bank manager could be sharing their feelings and experiences. However, this equalising effect of associations decreases, or can be completely eliminated, in hierarchical associations (Rosenblum 1998), which further indicates the need for associations to be democratised. Cohen and Rogers suggest limiting individual financial contributions to political groups, lowering barriers of entry to political processes and macroeconomic measures such as ‘inheritance taxes, income redistribution and subsidies for the organisation and representation of under-represented interests’ to prevent excessive inequalities being generated in the first place (Cohen and Rogers 1995: 37). Young suggests there is a need for differentiated resource allocation to associations by the state, to address inequalities between social groups that have arisen from historical processes of disadvantage and oppression (Young 1995: 212). Schmitter suggests state funding could be distributed to associations through citizens allocating with vouchers to their favoured associations. All citizens would receive some vouchers, and poorer citizens could be given more. Citizens would then give their vouchers to the association(s) they felt would best represent their interests and fulfil their needs (Schmitter 1995: 171–80; see also Fishkin 1991: 99–100 for a similar proposal). Such a proposal is compatible with individual autonomy, as each citizen would control the distribution of some resources. These measures would also make an important contribution to alleviating the worst socio-economic inequalities of the associational system, but they are far from sufficient. Socio-economic inequality is too vast and complex an issue to fully deal with here in this chapter, but over the next three chapters other various methods that should be included in a deliberative and associational democracy, in order to reduce the intensity of the negative effects of socio-economic inequalities, will be advocated. Conclusion Features of social complexity present significant barriers to deliberative democracy being meaningfully institutionalised, and to the legitimacy and effectiveness of the state. The claim here has been that in an associational democracy secondary associations would help overcome many of these barriers. Secondary associations provide routes to accommodating social pluralism, as they offer suitable locations for decentralisation, which makes the inclusion of all affected easier to accommodate. This factor, combined with the increased levels and greater dissemination of information that — 134 —

Deliberative Democracy through Secondary Associations

associations provide, aids co-ordination in a plural society. Membership of associations is relatively voluntary, and with their diversity, citizens have high levels of choice over sources of provision of key services, participation and representation. They are also able to represent a greater diversity of people than are territorially-based mechanisms. Associations can also generate civic consciousness in their members, which is desperately needed in plural societies. Secondary associations provide a passage for overcoming the barrier of scale as they are relatively smaller scale units that can enable citizens to directly participate in debate through which preferences, needs and information are formed. This is consistent with the principle of subsidiarity, which legislates for decisions to be taken at the lowest appropriate level to allow for those affected to participate and fulfil the task or provide the service they need and want. Decentralisation reduces the domain, making clear and unambiguous decisions more likely and therefore further reducing the problems caused by scale. Associations represent their members, enabling more people to be represented and included in decision-making and public debates, which can lead to the development of informal public spheres that can transcend elements of time and space. Furthermore, associations aid in the dissemination of information to large numbers of geographically dispersed people. Associations contribute to the more equal generation of key democratic capacities by providing the free spaces for citizens to learn, and develop, these capacities by participating on a small scale. Associations aid equality in democracy more broadly by providing representation, and access to public spheres for a diversity of groups that are not territorially based and that currently tend to be un(der)-represented. Again, economies of scale in the collection, and dissemination, of information are also achieved by associations. This last aspect also aids in the reduction of the reliance on specialists. Furthermore, if associations are devolved powers there is less relevant information, and greater knowledge of the issues, which in turn aids coordination. Associations provide specialist information themselves, including experiential information that might otherwise not be available, therefore countering the dominance of specialists in current policy processes. Finally, globalisation is accommodated because associations are so flexible that they can operate at international and transnational level, developing public spheres there and, if structured appropriately, even contributing to the decision-making processes of international and transnational formal institutions. However, an associational system is far from perfect and faces many — 135 —

Towards a Deliberative and Associational Democracy

significant problems. Not all secondary associations could fulfil all these functions. The extent of opportunities for exit, the media the association was orientated to, whether it was vested in this media, and the type of goals the association aimed to achieve, all affected the potential of any particular association to fulfil the democratic function. However, providing there is ‘a democratic ecology’ of associations, which there is likely to be in an associational democracy, these functions will still be fulfilled. Of greater concern are the current levels of socio-economic inequalities, which mean some associations have far greater resources to achieve their aims, and have greater access to, and are more likely to have their views considered in, the public sphere. Furthermore, people from dominant social groups will find it easier to join and participate in associations. They are more likely to join, too. Differential resource allocation to disadvantaged social groups is necessary, then, but not sufficient and this problem will be addressed in the following chapters, as will be other significant problems that threaten the democratic and autonomy-cultivating credentials of a deliberative and associational democracy. The approximation of deliberative democracy within the associations is essential to this system, but also difficult to achieve due to the same barriers of social complexity that affect its institutionalisation more broadly, which will be addressed in Chapter 5. There is also the likelihood that the mischief of factionalism will prevent the common good being promoted, and this will be confronted in Chapter 6. First, though, deliberations within associations and those occurring in the informal public sphere need to be linked to a variety of formal, deliberatively democratic, decision-making institutions, in order to ensure both deliberation and democracy are combined, and this shall be addressed in Chapter 4.

— 136 —

4

A Dualist Model of Deliberative and Associational Democracy

Introduction In the previous chapter, self-governing secondary associations were advocated as a mechanism with the ability to help institutionalise deliberative democracy. Providing they are devolved sufficient powers and are based on the norms of deliberative democracy, it was argued that associations could represent their members effectively and contribute to deliberations within the informal public sphere. However, these deliberations within associations and those occurring in the informal public sphere need to be linked to formal, deliberatively democratic, decision-making institutions, in order to ensure both deliberation and democracy are combined. As already highlighted, this book is essentially about decision-making, and if decisions are to cultivate autonomy, then deliberatively democratic decision-making is required. For this to occur, deliberative democracy must be effectively institutionalised, which requires that deliberation and decision-making must be linked: ‘Unless a direct link can be established and maintained between informal deliberation and formal decision-making the decisions made cannot realistically benefit from the legitimacy generated by the deliberation alone’ (Squires 2002: 142; see also Bohman 1996: 177; Dryzek 2000: 2; Leib 2004: 5–6, 39). If collective deliberation is not linked to decision-making, then the fact that participants’ preferences are more prudent seems irrelevant, as autonomy will not be enhanced. If only those in established and elitist representative assemblies, such as parliament, make decisions, then democracy would not be deepened. If deliberation is located only in the informal public sphere, then we must be sceptical as to whether the resulting decisions could be actualised. Therefore, in order to cultivate political autonomy, and closely approximate the ideal of deliberative democracy, public reasoning — 137 —

Towards a Deliberative and Associational Democracy

must be translated into collective decisions (Warren 2001a: 61): ‘We may say that while decision without deliberation is blind, deliberation without decision is empty’ (Gould 1996: 176). If deliberation and democratic decision-making cannot be joined, then popular sovereignty and autonomy is lost. Unfortunately, attempts at combining deliberation and decision-making with citizen participation lead to a Weberian dilemma (1978): ‘Either decision-making institutions gain effectiveness at the cost of democratic deliberation or they retain democracy at the cost of effective decisionmaking. In either case, citizenship, deliberation, and decision-making fail to be linked together’ (Bohman 1996: 178). However, a dualistic democracy is seen as a possible method for overcoming this Weberian dilemma and connecting the deliberations of citizens with decision-making institutions. Hendriks suggests that there are two broad types of strategy for institutionalising deliberative democracy. First, there is micro deliberative democracy, which focuses on ideal deliberative procedures within structured decision-making arenas within the state. Second, there is macro deliberative democracy, which favours informal and unstructured deliberative communication aimed at opinion formation within civil society outside, and often against, the formal decision-making institutions of the state. Micro deliberation tends to be too elitist, excluding too many participants, and while macro deliberation is more open, there is a failure to sufficiently empower citizens and make their participation effective unless this deliberative communication is linked to decision-making and micro venues. Hendriks argues that it is therefore essential for micro and macro deliberative democracy to be integrated (Hendriks 2006). This chapter offers a dualist model of deliberative and associational democracy, arguing that it provides a reasonable balance to the Weberian dilemma by connecting the macro deliberations of citizens with micro democratic decision-making institutions. It involves two dimensions for secondary associations: the first is their participation in the communicative processes of the informal public sphere (as has already been highlighted in Chapter 3); the second is their incorporation into the institutionalised and decision-making processes of the formal public sphere. Another attempt at a dualist model for institutionalising deliberative democracy is Habermas’ (1996a) ‘two track’ model in which deliberation is not restricted to representative assemblies, or to the informal public sphere, but is present in a combination of both informal citizen settings and formal representative institutions, enabling both micro and macro deliberation. In the first track, parliament (or comparable representative institutions) would remain the central focus for decision-making, but would be supported by — 138 —

A Dualist Model of Deliberative and Associational Democracy

decentred deliberation in the second track, the informal public sphere. In this model, public opinion generated in the public sphere can provide influence only through mechanisms like elections, the media, protests, demonstrations, petitions, commissions and consultation processes. In this model there is a cavernous separation between the power to decide and the power to influence deliberation because the public are separated from key decision-making arenas. The results of their deliberations will not be made into decisions and the model therefore fails to sufficiently overcome the Weberian dilemma (Bohman 1996: 186), or to cultivate citizens’ autonomy. Furthermore, Habermas’ model is too reliant on the judiciary to provide the link between popular and elite deliberation (Leib 2004: 32–3). In fact, it is hard to see what the differences are between Habermas’ two-track model and the present arrangements dominant in most liberal capitalist-welfare democracies, except for the fact that the legislative assemblies would need to be more deliberative than they currently are (Uhr 1998).1 Nevertheless, in common with Hendricks (2006), this chapter argues that a dualist strategy remains a promising approach for institutionalising deliberative democracy. In contrast to Habermas and Hendriks, however, I envisage the same secondary associations communicating in the informal public sphere and participating in making legislation in the formal public sphere. The ‘formal deliberative structures’ would not function ‘in a separate realm from active public spheres’, which is Hendriks’ main concern regarding associational democracy as a method of institutionalising deliberative democracy (Hendriks 2006: 497). The model here, then, significantly differs from Hendriks’ ‘integrated system’, as she is against associational democracy altogether due to scepticism that associations will represent a sufficient diversity of citizens, that associational representatives would be good deliberators due to inevitable partiality, and that they could avoid being co-opted by the state (Hendriks 2006: 497). The first of these significant concerns, over representation, was dealt with in Chapter 3 where it was argued that associations are in fact particularly apt at representing diversity, especially in an associational democracy where the ‘democratic ecology’ of associations will be expanded. The remaining concerns will be responded to in the course of this chapter. In the dualist model advocated here, as with Habermas’ model, associations would deliberate in the informal, unregulated discourse of the informal public sphere, which would act as a democratic agenda setter and opinion transformer. It is suggested that ‘discursive structures’ are required to interpret this agenda, that the media needs to be pluralised and that multiple spheres are necessary to reduce the problems of inequality and pluralism in the public sphere. Second, the model includes decentralisation — 139 —

Towards a Deliberative and Associational Democracy

of decision-making to quasi-autonomous non-governmental organisations (quangos) to hold forums where representatives from relevant secondary associations participate in deliberatively democratic decision-making, to create binding legislation. The preferences the representatives will express in these forums derive from deliberatively democratic decision-making of the members of the secondary association, meaning these associations are themselves internally democratic to some degree. The representatives from the associations who participate in the forums will therefore be principal agents and held accountable to their membership. Issues being considered here include whether the problems of representation, partisanship and transmission can be overcome. The chapter then addresses the criticism of dualism, which suggests that such a system will lead to the co-option of civil society by the state and would therefore fail to deepen democracy. Overall it is suggested that this dualist method offers a reasonable trade-off to the Weberian dilemma, effectively combining deliberation of citizens with decision-making, and enabling deliberative democracy to be approximated, thereby achieving the cultivation of autonomy that has been ascribed to it. The Informal Public Sphere Despite currently not approximating deliberative democracy very closely, civil society in liberal democracies generates deliberation through communication between its organisations, which forms a generalised debate in the informal public sphere. Following Habermas (1996a), it is suggested that the informal public sphere is capable of fulfilling two key elements of the dualist approach: the creation of ‘public opinion’ and ‘agenda setting’. As discussed in Chapter 3, public spheres are dependent upon flows of communication between secondary associations and other organisations. These flows of communication can then influence the opinions of the public. In doing so, they will be more likely to be based upon reason and be publicly orientated, as in order to convince the ‘general public’ of the validity of their concerns and preferences, associations must be able to ‘employ and appeal to norms of publicity’ (Habermas 1996a). Furthermore, in the informal public sphere discourse ‘is always likely to be less constrained than within the state’, meaning that insurgent discourses and identities can be established (Dryzek 2000: 79). Not only is public opinion generated through discourse in the public sphere, but these opinions are also more likely to be based upon reason and be publicly orientated. It is, then, associations that provide the ‘social infrastructure’ that enables opinion formation, that can question and hold accountable present sources of authority, and generate new issues, new beliefs, solutions, perspectives and — 140 —

A Dualist Model of Deliberative and Associational Democracy

ideas through public discourse (Cohen 1999: 215). It is these factors that make informal public spheres such promising locations for deliberative democracy. A dualist model of associational democracy is likely to enhance these processes because, as secondary associations become more important political actors, their discourses will too. Outside the informal public sphere, many organisations avoid public processes so they can assert private interests and maneuver themselves to gain vested state powers via nonpublic processes. They do this through funding political parties, lobbying and private consultation that leads to the subversion of formal representative institutions. Associations that are vested will try to avoid public debate and employ money and power to achieve their goals, only entering into public debate when they are forced to justify their actions, privileges and preferences (Warren 2001a: 165). When bargaining with government officials, sanctions and rewards are used to apply pressure on the government. However, inside the public sphere, the effect of interest groups is limited as these techniques are ineffectual, with the dominant media in the public sphere being communication, as opposed to money and coercion (Habermas 1996a: 369; Warren 2001a: 80). Consequently, convincing reasons become increasingly influential at transforming preferences and mobilising public opinion: ‘Public opinion can be manipulated, but neither publicly bought nor publicly blackmailed’ (Habermas 1996a: 364). The fact that associations often pursue narrow interests and assert unequal influence over decisions through private processes is, though, a huge concern and will be addressed fully in Chapter 6. There are various ways in which the informal public sphere and political system can influence each other to establish an agenda. These models are the inside access model, mobilisation model and outside initiative model. If the decision-making agenda is to reflect the public discourses of the informal public sphere, it should be set through the ‘outside access model’, as it is the only model of agenda setting that endorses communication in the informal public sphere. In the other two models the initiative to put an issue on the agenda comes from office-holders and political leaders. The pressure upon the formal political system to consider the issue, from the informal public sphere, in the ‘outside access model’, can be produced in three broad ways, all of which can be either democratic or undemocratic. The first method is for a group to articulate a grievance, communicate with other groups (with the aim that they take on board the issue) and pressurise decision-makers to deal with the issue and put it on the formal agenda (Cobb et al. 1976: 132). More specifically, the informal public sphere can change political discourse, which in turn affects how ‘terms are defined’ and ‘issues are framed’, and — 141 —

Towards a Deliberative and Associational Democracy

thereby ‘influence political culture’. Examples of this would be the racial equality and feminist movements. The second method is through the establishment of its own policy forums, whereby pressure can be placed on legislative bodies, holding them accountable.2 Protest located in the public sphere can in this way pose the threat of political instability, causing the government to react (Dryzek 2000: 101–3). Habermas correctly observes that it is usually through the inside access or mobilisation models that issues are placed on the agenda. He portions the blame on the mass media’s techniques of drawing information from powerful and organised elites and pursuing strategies that lower the discursive level of public communication (Habermas 1996a: 380). Despite deserving this blame, the media still plays a significant role in communicating ideas, needs, preferences and issues within the informal public sphere, and in setting the political agenda. In fact, the mass media constitute publics themselves, as one of their primary roles is to induce public debate on issues that they raise, and to justify the raising of such issues in the first place, although often in a manner that does not reflect the ‘complexity’ and ‘impersonal features’ of the issues concerned (Parkinson 2006: 164). However, the media in liberal democracies currently reflects, and consequently reinforces, the vast disparities of economic and political power, and is not accessible to all actors in civil society. Associations outside the political system, or outside large organisations, will have a much reduced chance of influencing media output due to its market structure. This factor is accentuated if the views of the association fall outside ‘centrist’ or ‘established opinions’ that dominate the media (Habermas 1996a: 377; Bohman 1996: 132, 140–1; Warren 2001a: 168). Therefore, the transformation of the media is probably one of the most essential requirements for the meaningful approximation of deliberative democracy (Habermas 1996a: 378). Exactly what framework the media would need to adopt, and how these changes would occur, is a study in itself, and therefore outside the remit of this book, despite its obvious importance to the institutionalisation of deliberative democracy. Nevertheless, normatively, a democratic and pluralised media that does not tacitly reflect inequalities is necessary. The above points notwithstanding, civil society still plays a ‘surprisingly active and momentous role’ in agenda setting (Habermas 1996a: 381). For example, opposition to nuclear arms and power, genetic engineering, ecological threats, Third World debt, the encumbrance of risk, racial discrimination and gender inequality have all arisen from the informal public sphere (Habermas 1996a: 359; Beck 1999). Therefore, the informal public sphere can undoubtedly change institutions, forcing them to adapt to new publics offering new visions, interpretations, issues and beliefs. — 142 —

A Dualist Model of Deliberative and Associational Democracy

Institutions must somehow interact with these new publics, even if they are simply trying to suppress them rather than democratically communicate with them: ‘In the process, institutions are changed in a variety of ways: in their concerns, in their ongoing interpretation of rules and procedures, in their predominant problem-solving strategies and so on’ (Bohman 1996: 201). Even if they do not then get to change actual decisions, they force government organisations, corporations and powerful associations to justify their position on the issue through public reasons. A deliberative and associational dualist model increases the chances that the ‘outside initiative model’ of agenda setting is employed by making secondary associations a key locus of political participation and representation. Consequently, the media and state will inevitably be encouraged to give the public sphere more attention, just as political parties receive much attention now. If they did not, then the media’s credibility and legitimacy would seriously be challenged, as secondary associations would also become central legislative participants through the decentralised forums (as will be discussed in more detail in the following section). In turn, secondary associations would become more likely to enter the discourse in the informal public sphere ‘because they could no longer afford simply to attempt to buy votes from elites . . . or pay marketing firms’ (Leib 2004: 119).3 Micro deliberative sites, such as the forums, require a clear and often narrowly focused agenda to be effective and to enable rational decisionmaking and good deliberation (Thompson and Hoggert 2001: 358). Unfortunately, there is no mechanism for establishing a deliberative majority opinion in the informal public sphere. It is therefore not clear which public opinion should be ‘taken up’ by the representative institutions (Bohman 1996: 179–80). There is, then, still a need for what Habermas terms ‘discursive structures’ that link the informal public spheres and the formal decision-making forums that will allow for influence and channel communication between the deliberative opinion formed in the public sphere and the decisions taken in the institutions. However, Habermas leaves these ‘discursive structures’ virtually unspecified, in an attempt not to lay down a blueprint for a deliberative democracy (Habermas 1996a: 226–8). Without such structures, the agenda will be set by political elites, a situation that brings a significantly increased danger of an overly narrow agenda that, by definition, excludes the views of affected groups and therefore frames decisions. As already suggested, the greater focus on informal public spheres provided in a dualist model, advocated here, helps overcome this, enabling macro deliberative processes to provide an agenda which generates open debate over both the agenda itself, and its definition. Nevertheless, this is unlikely to be specific enough for the forums, and the agendas arising from — 143 —

Towards a Deliberative and Associational Democracy

the macro public sphere will still need to be interpreted. Inevitably, this seems to be a role that would still involve current political elites (Parkinson 2006: 128–33). However, the process of interpreting the competing agendas of the informal public sphere could still be combined with more democratically deliberative methods. An innovative approach to these ‘discursive structures’, from Parkinson, is for local and national governments to have committees whose function is to gather submissions from civil society groups. This could be combined with processes like an ‘electronic town hall’, where thousands of citizens would assemble to debate and vote on the agenda before it is formalised (Parkinson 2006: 170), which would make agenda setting a more equal and deliberative process. There are, however, other problems relating to inequality of access to the informal public sphere. We heard in Chapter 3 that in current liberal democratic societies, the informal public sphere is plagued by inequality of access, which affects its potential to fulfil deliberative roles in a democratic manner and enables the discourses of the powerful to dominate and ‘crowd out everyone else’ (Mansbridge 1992: 48). Further, socio-economic inequalities cause the cultural ethos developed by socio-economic groups to be unequally valued, meaning public spheres are not, and cannot be, neutral and equally ‘expressive of any and every cultural ethos’ (Fraser 1992: 120). Associations that are not recognised as deliberators are unable to influence debate through dialogue and therefore resort to other methods, such as demonstrations, protests and civil disobedience, to achieve recognition, gain influence and encourage an issue on to the agenda (Fraser 1992: 123). This creates the danger that rhetoric can supersede reason in these informal public spheres (Parkinson 2006: 164). This is a serious challenge to the informal public sphere’s agenda setting role, as inequalities can enable the issues and concerns of some to be ‘heard’ louder and more often, but more importantly enable them to limit the agenda to protect the status quo from which these groups benefit (Bachrach and Baratz 1962). In fact, by ‘promoting one political agenda item, civic activists may succeed in driving other issues away’ (Crenson, cited in Lukes 1974: 44). Consequently, this inequality severely limits a public sphere’s potential to fulfil both deliberative and democratic roles. If this is the case, then autonomy of all members of all social groups is not cultivated equally. Habermas (1996a) is right, however, that ‘influence’ cannot entirely escape democratic connotations, despite these inequalities, as unless the public finds the assertions of associations convincing in some way, they will not be influenced by them. The assertions must therefore be relevant in some way for the people to be persuaded and have their preferences transformed. Habermas envisages a single public sphere, suggesting that socio— 144 —

A Dualist Model of Deliberative and Associational Democracy

economic inequalities can be ‘bracketed’ so that participants deliberate together as peers. This view seems mistaken, though, as Fraser is aware, because all discursive arenas are situated in a broader socio-economic environment which forms many aspects of the individual participants, making it impossible for them to bracket these inequalities. ‘Bracketing’ would also be biased towards the dominant social groups, as it is ‘tantamount to filtering diverse rhetorical and stylistic norms through a single, overarching lens’ (Fraser 1992: 120–6). It seems apparent that Habermas also fails to deal with the plurality of the arguments, identities and interests that will exist in informal public spheres, and the problems these create. Fraser suggests multiple public spheres are the solution, as they provide subordinate groups with the arenas in which to deliberate and form collective preferences, goals, strategies and identities away from the unequal influence of dominant groups. This is important because participation in a public sphere is not just about asserting neutral preferences, but about forming one’s own identity, preferences and needs. This is why subordinate social groups, for example women, workers, racial minorities, homosexuals and disabled people, have been motivated to form alternative publics, or what Fraser terms ‘subaltern counter publics’, described as ‘parallel discursive arenas where members of subordinated social groups invent and circulate counter discourses to formulate oppositional interpretations of their identities, interests and needs’ (Fraser 1992: 123). These are factions or enclaves that are often excluded and become factionalised by the political process. Nevertheless, they aim to disseminate their beliefs, and communicate these to as broad a public as possible, through networks with other associations and between public spheres (Fraser 1992: 124). The dualist model outlined here has many features that should aid the multiplication of existing informal public spheres, such as functional and territorial devolution, the creation of a multiplicity of legislative arenas (the mediating forums) and the promotion of secondary associations to a prominent role in governance, and it is essential that each forum develops extensive informal public spheres around it (Bohman 1996: 188). Such features are in stark contrast to Habermas’ ‘two track model’, where the public sphere is centred on one dominant legislative arena. Although they help combat inequality of access, multiple public spheres also cause problems for deliberative democracy. For Squires, because they are multiple and difficult to demarcate, it is hard to establish whether inclusivity has been achieved or not, meaning the attempt to legitimise the informal public sphere on the basis of providing inclusive deliberation ‘will necessarily fail’ (Squires 2002: 139). Despite the faith that many deliberative theorists have placed in them, multiple public spheres consequently remain — 145 —

Towards a Deliberative and Associational Democracy

far removed from an approximation of the deliberative ideal in which all affected participants get to have their views heard by all, and hear the views of all others. With a myriad of sub-publics come incomplete and disjointed discourses, meaning that, at best, ‘a partial insight into issues through discussion’ can be achieved (Bohman 1996: 182). Furthermore, if these multiple publics compete, rather than communicate, with each other, then the informal public sphere will not fulfil its function of ‘filtering out nonpublic reasons’ (Bohman 1996: 180), which will compromise the autonomy of public opinion. Although these multiple public spheres are often excluded and become factionalised by the political process, they themselves aim to disseminate their beliefs and communicate these to as broad a public as possible. This is one of the key meanings of what it is to be in a public, whether subaltern or not. Even if a public is empirically small, with only a few participants, they are always part of a ‘potentially’ larger public, the general public (Fraser 1992: 124). Although not all networks are public, they do enable secondary associations to spread their message to other public spheres that would otherwise not hear, or address, such issues. In addition, networks enable the pooling of resources and information between, as well as within, secondary associations and therefore create economies of scale that can address some of the socio-economic inequalities that exist when a group is trying to be heard in the informal public sphere (Bohman 1996: 136). The challenge is to ensure that these relationships are co-operative, with information being shared, resources pooled and joint programmes and processes initiated. Co-operation is important, as the trust established through networks of communication between associations can ‘provide assurances to members that their own willingness to co-operate will not be exploited by others’ (Cohen and Rogers 1995: 44). Such trust is imperative in a deliberative democracy because it requires participants to commit to its specific norms and procedures. In Chapter 2, following Festenstein, I argued that deliberative democracy required grounding for its necessary deliberative obligations. One possibility is that grounding could occur through the civilising force of hypocrisy, also discussed in Chapter 2. Through regular interaction within the forums, members from various associations will form co-operative relationships, and will not want to sour their image through selfish actions outside of the forum. This will affect how their arguments are perceived within the forum and how honest the information is seen as being. However, as Festenstein appreciates, if it is due to instrumental reasons that an association will be civic, then when it is in their interests to abandon civility, then abandoned it will be (Festenstein 2002: 97–8), and the civilising force of hypocrisy, although certainly a help, might not be — 146 —

A Dualist Model of Deliberative and Associational Democracy

sufficient to reign in this self-interest. In Chapter 3, the case was presented that associations could provide a ‘civic consciousness’, involving a recognition and commitment to democratic procedures and norms sufficient to ground these deliberative obligations and ensure co-operative relationships (Cohen and Rogers 1995: 43–4; Rosenblum 1998; Warren 2001a: 7). Some types of association have a greater capacity to co-operate and coordinate within associational networks. Associations that aim to secure public material goods will inevitably feel an incentive towards co-operation in order to achieve these, while those aiming for inclusive social goods will want to co-operate with other associations out of principle. Associations pursuing individual material goods have strategic incentives for cooperation and co-ordination; however, this might not be pursued through the informal public sphere because the incentive for secrecy might dominate in such situations. In contrast, associations that pursue exclusive identity goods will be unsuited to contributing to co-operation, as they will have little principled or strategic incentive to do so. In fact, members of the association may think co-operation with other associations is actually a betrayal of their principles (Warren 2001a: 197–8). Moreover, the variety of groups involved in these networks and movements means there is still conflict over aims and beliefs (Young 1990: 83) and they can excessively ‘oversimplify, personalise, and polarise debates’, too (Parkinson 2006: 167). Due to these issues raised by the existence of multiple publics, Fraser classifies interaction in the public sphere as a contestation of publics, as opposed to a contestation of discourses. In Fraser’s analysis, contestation is distinct from deliberation; consequently, she sees no reason why deliberation will predominate in the informal public sphere (Fraser 1992: 125). However, this analysis rests on an overly narrow definition of deliberation. It was asserted in Chapter 2 that, as long as preference reflection is generated by communication, then deliberation has occurred, and contestation can generate this reflection (Dryzek 2000: 76; see also Achterberg 1996). Hunold (2001) provides a persuasive account of how associational representation is changing across most nations to provide a more supportive environment for deliberative democracy in public spheres, away from the neo-pluralist model of associational competition (see also Firorino 1995; Gundersen 1995). The neo-pluralist model had previously dominated because associations could not trust others to collaborate and to abide by resulting decisions. What has emerged in the place of neo-pluralism is state-sponsored ‘pluralism by the rules’, through which associational actors gain ‘assurance in the form of rules which structure how the game of pluralism will be played’ (Weber, cited in Hunold 2001: 162). Through the — 147 —

Towards a Deliberative and Associational Democracy

proliferation of policy and issue networks, co-operation amongst a huge array of associations and public officials is increasing, and actively promotes deliberation. Such processes are increasing as actors believe their interests will be promoted more successfully through such deliberative and collaborative networks than through the competition and aggregation associated with traditional pluralism, in which most parties end up being disillusioned and frustrated with the process and resulting decisions. Consequently, pluralist arrangements are becoming more deliberative and corporatist in their nature (Hunold 2001: 162). Instances of macro deliberation, like the informal public spheres discussed here, are suitable for generating deliberative influence and opinion formation, but are what Fraser (1992) terms ‘weak’ publics, as they are usually peripheral to decision-making arenas. In order for deliberative democracy to be approximated, the Weberian dilemma to be successfully resolved and the normative goals attributed to it generated, arenas for collective decisions are also required. Decision-making arenas are ‘strong’ publics, and examples of micro deliberation. Micro deliberative forums are further required to help counter the inequalities that might exist in the informal public spheres discussed above (Hendriks 2006: 496). It is these formal arenas to which this chapter now turns its attention. The Formal Public Sphere – Mediating Forums If the Weberian dilemma is to be traversed, and deliberation and democracy effectively combined, the second part of the dualist strategy must ensure that secondary associations participating in the discourse of the informal public sphere have access to deliberative legislative arenas. The suggestion here is that mediating forums, with territorially and functionally devolved powers, could be legislative arenas in which representatives from relevant secondary associations would assemble and therefore fulfil the second branch of the dualist strategy. Traditionally, mediation is not followed by legislation, and decisions made are not imposed; however, it is essential that these forums are legislative arenas that produce binding decisions in the form of policy that would be implemented and enforced by the relevant level of government or quango. This is essential if the Weberian dilemma is to be evaded, for without power over policy the arrangements would be open to the same criticisms leveled at Habermas’ two-track model. Similarly, the main flaw of Hendriks’ (2006: 501) ‘integrated deliberative system’, which also combines micro and macro deliberative sites, is that it only results in ‘recommendations for decision-makers’, rather than decision-making itself, and as Dahl — 148 —

A Dualist Model of Deliberative and Associational Democracy

says: ‘Democracy involves debate and discussion, but these are not enough if they remain inconclusive and ineffective in determining actual policies’ (Dahl, cited in Gastill 1993: 16). Given the right institutional design, all those associations that represent people affected by a certain issue can be included into democratic decisionmaking forums. As well as ensuring that those previously excluded from decision-making processes are now included, this also forces previously vested associations to give public reasons to justify their preferences and interests. If these forums do not have the power to pass binding legislation, then empirical evidence suggests that it is more likely that some associations, especially those representing commercial interests, will choose not to participate or withdraw if they feel the forums are going against their interests, especially if they have ‘a lot to lose’ (Hendriks 2002: 65; Hendriks 2006; see also Cohen and Rogers 2003: 252). In contrast, if the forums are binding, empirical evidence indicates associations are more likely to want to participate in collective deliberation. precisely because they have a lot to lose or gain, as the most effective way to influence outcomes will be through participation (Fung and Wright 2001: 24; Newman et al. 2004: 213; Hendriks 2006). These associational forums therefore bare a strong resemblance to European corporatism (Warren 2001a: 119), with the exceptions that corporatism has dramatically fewer groups integrated into the decisionmaking process and the groups that are included are stable, changing little.4 Corporatism, then, tends to promote the interests of certain privileged groups, and their inclusion into the decision-making process at the expense of other overlooked/ignored/marginalised/excluded groups (Mansbridge 1992: 41; Offe 1995: 120), which leads to ‘capture’, ‘patronage’ and ‘clientelism’ (Streek and Schmitter 1986): In summary tripartite bargaining arrangements tend further to diminish the role of territorial representation without necessarily putting a socially inclusive and public accountable system of functional representation in its place. (Hunold 2001: 161) Such a practice goes against democratic equality, and only enhances the autonomy of the members of associations who are represented. Moreover, corporatist discussions tend to be private, rather than public, affairs (Hunold 2001: 161). Associational democracy, particularly when it aims to approximate the ideal of deliberative democracy, is quite the opposite. It aims at the inclusion of all groups into public decision-making processes, and seeks the formation of new groups to represent un(der)- represented — 149 —

Towards a Deliberative and Associational Democracy

voices, and corporatist arrangements seems to be increasingly moving towards such a model (Hunold 2001). We cannot, therefore, look to traditional tried and tested corporatist arrangements to gain insight to how the mediating forums might actually operate, but must turn instead to other types of deliberative forums. In Chapter 2, empirical evidence from legal juries indicated that the ‘evidencedriven’ style of deliberation, as opposed to the ‘verdict-style’ of deliberation, was essential to ensure inclusion of all deliberators in the debate (Sanders 1997: 367), and in the UK the evidence-driven style of deliberation does seem to be on the increase in partisan forums, especially in natural resource management (Connelly and Richardson 2004). There is extensive empirical evidence available from non-partisan deliberative forums such as citizens’ juries and deliberative opinion polls that indicates citizens have the competence to address complicated issues, that participants will change their preferences in light of reasons and information, and that they can arrive at compromise decisions (Fishkin 1991 and 1995; Stewart et al. 1994; Renn et al. 1995; Coote and Lenaghan 1997; Coote 1997; Kuper 1997; McIver 1997; Andersen and Jagger 1999; Barnes 1999; McCombs and Reynolds 1999; England 2000; Luskin et al. 2002; Hansen 2004; Andersen and Hansen 2004; Parkinson 2006). Perhaps more revealing research on citizens’ juries, for the purposes here, from Thompson and Hoggett (2001) is concerned with the development of factions within deliberative arenas, that could offset the benefits of the deliberative process. They advocate combining deliberative plenary sessions with sub-divided deliberative forums, as factions and ‘internal psychological divisions’ are less likely to develop in small groups. Moreover, these subgroups do not need to have ‘rigidly defined boundaries’, so long as they have revolving membership, as with German planning cells.5 This ensures all get to hear the views of all, and get to express their views to all (Thompson and Hoggert 2001: 358). Furthermore, this helps overcome the problem of scale within the forums, as, despite the use of decentralisation and representation, which both help alleviate this aspect of complexity, many decisions, and therefore forums, will involve a large number of associational representatives, which is a barrier to effective, equal and inclusive deliberation. This is, however, obviously a trade-off between the ideal of deliberative democracy, in which all participants are involved in the same debate, and the practical necessities of real-life decision-making. Research based on non-partisan deliberations fails to fully indicate how the associational mediating forums might operate in reality. Partisan deliberative forums are on the increase, and although they do not replicate the associational forums outlined here, especially in the aspect of decen— 150 —

A Dualist Model of Deliberative and Associational Democracy

tralising final and binding decision-making power to the forums, which is obviously a crucial dynamic, they do provide insight and lessons about how deliberation in such arenas is likely to proceed. Evidence from Barnes et al. (2004), in their study of a range of forums in the UK, indicates that deliberation in forums is not always generated even though citizens enter into dialogue with officials. This occurs due to a lack of awareness of, and inclusion of, varying types and sources of knowledge, discourses and forms of expression, and also differing levels of respect for participants. It is consequently suggested that deliberation can be more successful in identitysharing groups (which bodes well for deliberation within the associations discussed in Chapter 5) rather than ‘forums established by officials’ (as the forums advocated here are) (Barnes et al. 2004: 106). Here we see the Weberian dilemma present again, as deliberation might well operate most successfully away from locations where it cannot directly affect decisions. However, this is largely overcome if the participants have prior experience of participation (which the associational members should quickly gain in this model) and if, ‘prior to engagement with officials’, they are able to ‘construct their own definitions’ of the issues to be debated (Barnes et al. 2004: 107). In the dualist model the debates within associations themselves should ensure this occurs. The forums will also vary in their nature to quite an extent, as they will be organised by different quangos, address different issues and include vastly different sets of participants. The problem with this, though, is that each participant has their own ‘institutional assumptions’ about how the debate should proceed and these issues need to be resolved before more substantial and issue-orientated dialogue can occur (Barnes et al. 2004: 107). Consequently, Barnes and her collaborators conclude that, if deliberative democracy is to be effectively approximated, then practices ‘need to break out of many of the institutional constraints within which they operate’ (Barnes et al. 2004: 107). Accordingly, the dualist model of associational and deliberative democracy requires extensive institutional change. It is apparent, then, that given this research a key criteria democratic procedures and mediators must ensure is equality within the forums, but this is something that is very hard to achieve. Once again, the democratic potential is threatened by embedded socio-economic inequalities, which produce inequalities between associations and mean that the decisionmaking processes operate in favour of dominant groups. Dryzek warns that institutional designs similar to the forums outlined here may therefore lead to manipulation and capture by associations representing such dominant groups. Manipulation can occur through their greater control of resources, allowing them to ‘cloak private interests in a rhetoric of public concern’ as — 151 —

Towards a Deliberative and Associational Democracy

well as ‘making superficial concessions to opponents and thereby secure passive acquiescence on the part of potential troublemakers’ (Dryzek 1990: 81; see also Fung and Wright 2001: 36). If consensus decisions are not reached, and a certain decision goes against the interests of one group, unequal distribution of power can mean that groups still have effective veto over decisions, as their co-operation may be essential for it to be implemented. For example, polluting industries must co-operate if pollution is to be reduced. Furthermore, associations representing corporate interests have an organisational and financial advantage over other associations (Smith 2001: 78). On the other hand, if the forums are successful in avoiding manipulation by the more powerful associations, and actually challenge the dominance of these groups, then there is a further danger that these dominant associations will resort to private mechanisms outside of the forms, such as lobbying, to re-assert their influence (Fung and Wright 2001: 34). As suggested earlier, a dualist model reduces the strategic benefits of such non-deliberative actions, but cannot eliminate them. The forums must therefore balance the power between the associations, so that the threat and sanction of force is removed, and reason and influence will predominate, which increases the focus on public goods, rather than private interest. Corporatist arrangements such as the forums are particularly apt at achieving this (Mansbridge 1992: 48). Empirical research on a variety of devolved, partisan and deliberative forums from across the world, co-ordinated by Fung and Wright (2001), suggests that, despite these threats, the dualist model of deliberative and associational democracy would provide several routes to the balancing of power in order to enable deliberation to occur in the forums. First, experts and citizens in the forums would both have to justify their views to each other through reasons, which places them on a more equal plane (Fung and Wright 2001: 22). Furthermore, the presence of other partisan associational representatives would assist in the checking and regulation of powerful groups who might attempt to capture the deliberative process to pursue narrow selfinterests (Fung and Wright 2001: 22). The forums will include the subordinate groups who are usually excluded from decision-making arenas, and because they are participating, the resulting decisions are likely to be fairer (Fung and Wright 2001: 26). Relative equality will also be aided through the decisions the forums make by ‘delivering effective public action to those who do not generally enjoy this good’ (Fung and Wright 2001: 26). Moreover, because the forums will be deliberative, decisions are more likely to be based on reason, as opposed to money, power, numbers or status, and therefore lead to more equitable policies (Habermas 1996a; Fung and Wright 2001: 26; Warren 2001a). In reality, though, people will not — 152 —

A Dualist Model of Deliberative and Associational Democracy

only adapt preferences in accordance with reasons, but also due to other factors such as the source of the information, the manner in which it is provided, the psychological dynamic of the group and pressure to conform with the majority. Despite this, however, ‘reasons’ remain privileged in deliberative democracy, in comparison to other forms of decision-making. One of the principal functional advantages of associations being involved in decision-making processes is that once the decision has been made, it generally becomes easier to introduce and enforce (Barber 1984; Fung and Wright 2001: 26), which contributes significantly to navigating the Weberian dilemma. The forums are also likely to result in fewer legal challenges, something which plagues much environmental legislation (Fiorino 1995), for example, meaning a deliberative and associational democracy could lead to less costly legislation and more expeditious policy enforcement (Hunold 2001: 154).6 As the forums increase the likelihood that all relevant views will be included in decision-making, legislation is more likely to survive legal challenges if they do occur (Hunold 2001: 154–5). This is partly to do with the fact that members of the associations who have participated in making the decision are more likely to see the process as a legitimate one, and therefore accept the consequent decision, even if it is not what they hoped for, than they would if it was imposed by an external authority. Moreover, if the members have been engaged in democratic debate about these issues themselves, and seen how an associative member in a forum has represented their ideas, then they can see how their own views may have influenced that debate, again making the resulting decision even more legitimate. The associations can then help in the implementation of the legislation, either through the carrying out of the services/activities set out in it, or in dissemination of information. Due to the fact that legislation would now be easier to enforce, more options become open to political debate, rather than being ruled out tout court (Fung and Wright 2001: 18). It also means powerful organisations will have less ability to veto any legislation that they dislike because their co-operation will become less important due to the increased co-operation of other associations (Cohen and Rogers 1995: 65–6; Smith 2001: 78). There are many other issues and problems that must be addressed if these mediating forums are going to be seen as a credible alternative to the current legislative bodies’ dominant in liberal democracies. Many of these will now be addressed in a bid to provide more detail to this sketch of these institutions. First, the nature and manner of decentralisation to the forums will be considered; the chapter will then set out how quangos could take responsibility for organising these forums, specify what role existing institutions of governance would fulfil under this dualist system, and finally address the nature and form of representation in the forums. — 153 —

Towards a Deliberative and Associational Democracy

Decentralisation In Chapter 3, decentralisation of many social activities directly to secondary associations, based upon the principle of subsidiarity, was advocated. It was further established that broader collective decisions would be required, and the suggestion here is that decentralised forums could make these in a dualist model of deliberative and associational democracy. This would require a considerable increase in decentralisation than is currently present in most liberal democracies, but once again, would help adapt deliberative democracy to the barriers of social complexity, and help accommodate the state to these features, too. As with decentralisation direct to associations, decentralisation to the forums must be implemented cautiously in order to ensure that those who are affected are not excluded from participating in, or being represented in, the making of the decision. The decentralisation should, therefore, again be based on the principle of subsidiarity. As discussed in Chapter 3, although the tendency will predominantly be towards decentralisation, the exact content of what decisions should be taken and at what level is not stipulated by the principle and would have to be decided through the political process. Examples can still be envisaged and proposed, though, and types of health and social care delivery, local transport, policing, education and planning could all be determined, to an important extent, locally, with the interpretation of ‘local’ varying from context to context and allowing for the ‘delineated ecosystem habitats’ (Fung and Wright 2001: 21) thought essential for democratic environmental control (Eckersley 2000: 120). Issues requiring functional forums include telecommunications and biotechnology, whereas issues relating to foreign policy and national defence are likely be made nationally. This associational model of democracy offers decentralisation on two fronts, functional and territorial. Functional decentralisation is achieved through the devolution of powers to quangos, who would hold the forums for the relevant associations to participate in the decision-making process on functional issues that cut across geographical areas, and take account of associations’ specialist knowledge in these areas. Territorial decentralisation is achieved by devolving powers to local areas, where quangos will again hold the forums for relevant associations to participate in deliberatively democratic decision-making processes, through which local and regional policy will be formed (Martell 1992: 166), and take advantage of the local knowledge and information citizens here will have of the issues in question (Fung and Wright 2001: 26). As Hyland argues, there is no good reason why the agency concerned with regulating nuclear power should be the same — 154 —

A Dualist Model of Deliberative and Associational Democracy

agency that deals with food additives, the education system and road networks (Hyland 1995: 263–4). In order to ensure effective control and accountability of forum representatives by those being represented, there should be only two layers of deliberation and decision-making in the dualist model: the first, direct participation in the associations, and the second, representatives from these associations participating in the forums. There should be functional forums organised across territorial boundaries that might deal with issues such as inter-regional travel, bio-technology and fraud. As well as forums territorially organised at local and regional level that might deal with local transport, social care, tourism and planning, there must not be several levels of forums for the formation of any one policy in these or any other policy areas. Consequently, a decision cannot be passed up to a further forum to be discussed by representatives who have come from a ‘lower’ forum. This means that a series of forums on a particular issue, for example, bio-techology, cannot be held in such a way that every locality holds its own forum, and then these forum representatives move on to a regional forum, and representatives from the regional forum then move up to a national forum, with the aim being a national biotechnology policy addressing the issue or even passing it to a transnational forum. In such circumstances, the likelihood is that the association’s representatives participating in the local forum would elect the representatives to participate in the regional forum and so on. This would mean that at the regional, and especially the national, forums, representatives would have little credibility over the claim that they represent their constituents, and are accountable, as they would be too far removed from the citizenry by too many layers of forums and representation. If decisions need to be made that cut across the local and regional territories, then they must be made through functionally decentralised forums. If a national policy is required over an issue, then this must be made by national government; if a transnational decision is necessary, then this should be made in transnational institutions. This demonstrates, then, how essential it is that the principle of subsidiarity be applied coherently and consistently when decentralising powers because, in reality, this will not always be possible due to the compromises that will be made when interpreting subsidiarity through the political system. Decentralisation to forums helps reduce the preference domain, enabling clear, unambiguous decisions to emerge from the forums. It also helps alleviate the social choice theory critique of democratic practices. Decentralisation is, furthermore, compatible with the forums’ need for a specific focus, which is why a combination of elites and ‘discursive structures’ are needed to interpret and ‘take up’ the discourses emerging from the informal — 155 —

Towards a Deliberative and Associational Democracy

public sphere: The forums ‘could not decide anything if they were chaotic’ (Leib 2004: 120). Due to the serial deliberative structures, and the territorially and functionally devolved forums, there is potentially a problem of transmission and co-ordination of the decisions (Goodin 2003: 56). Associations, as locations for governance, providing services to their members under powers devolved by the state, will make decisions that will primarily affect the members of that association (Perczynski 2000: 169). However, when associations are also to participate in the formation of legislation in the forums, then the decisions they make most certainly will affect people outside of the association, and even the forum, so the problems of transmission therefore still pertain. Decentralisation can encourage participants to see the issue under debate as ‘unique, isolated phenomena abstracted from social relations’. Furthermore, participants, forum organisers and mediators could be compelled to see disputes in this manner, as it would make them ‘easier to mediate and resolve’ (Smith 2001: 78). There is, then, inevitably a discursive dilemma, similar to Weber’s, that again relates to the problem of combining democracy and deliberation. Decisions could be responsive to the reflective preferences of the representatives assembled in the forum, regardless of whether they are rationally compatible with decisions made previously in other forums. Alternatively, decisions could be rationally consistent but be unresponsive to representatives’ preferences. The former is more democratic, but at a sacrifice to deliberation, while the latter is more deliberative, but at a loss to democracy (Pettit 2003: 138). In the mediating forums advocated here, the decisions will inevitably be responsive to participants’ preferences, but for Pettit this means the decisions will be arbitrary and capricious (Pettit 2003: 155). He argues it is more important that decisions meet deliberative requirements and are rationally compatible than be democratically responsive. One solution is to ensure that all decisions remain contestable, especially as participants will change over time (Pettit 2003: 156). This seems appropriate and the forums should continue until the legislation has been reviewed and, if necessary, reformed. Once this process is completed, and the forum has become defunct, a new forum could be generated to readdress the issue and provide further reform, if required. In general, Pettit’s discursive dilemma is a dilemma for decision-making and could be alleviated by the flexibility of the mediation process, which is a feature of this dualist model, as the forums can operate at a variety of levels and across an array of functions (Smith 2001: 80), which are also strengths of secondary associations. Moreover, due to the mediation being based upon the norms of democratic deliberation, no arguments or reasons are — 156 —

A Dualist Model of Deliberative and Associational Democracy

formally excluded, so there is nothing to prevent any of the participants from trying to put the issues into a broader agenda. Whether these reasons have then any motivational force to the other participants depends on their convincingness. What might be required, then, is not decentralisation to the forums per se, but co-ordinated decentralisation, whereby decisions from successful forums are disseminated and diffused (Fung and Wright 2001: 22–3), which could be achieved through networks among secondary associations, local and national government and quangos. Therefore, multiple policies will, with varying techniques and strategies, be pursued simultaneously, with both the successes and failures being passed on, resulting in ‘the learning capacity of the system as a whole’ being enriched (Fung and Wright 2001: 26). The dualist model of deliberative and associational democracy advocated here also sees an even more important role for quangos, to which we now turn. Quangos and Forum Organisation The forums will need to be organised and facilitated, and here it is suggested that quangos are suitable organisations to fulfil this role. This may seem like a curious choice, as quangos have been severely criticised for their lack of democratic credentials and labelled as arenas void of democratic arrangements and processes of accountability. Indeed, they currently privilege bureaucrats, professionals and technical experts with many roles of governance, enabling them to make key policy decisions, provide advice and take action away from public scrutiny (Weir 1996: 20). They are often complex and inaccessible to the public, which exacerbates collective-action problems and therefore discourages popular participation (Weir 1996: 29). In addition, they are formed through appointment rather than election (Harden and Marquand 1997: 13; Flinders 1999) and there is a whole raft of evidence in the UK which suggests a dubious correlation between the party in government and appointments to quangos of sympathisers of the governing party (Flinders 1999). Those quangos not appointed by government ministers tend to be self-appointing, which raises further questions about undue influence and accountability. Despite their lack of legitimacy, in the last twenty years liberal democracies have seen continual quango growth, in terms of number, scope and the functional area that they cover (Kickert 2001; Van Thiel 2001; Flinders 2004). It has become an assumption of central government that quangos can more effectively implement certain policy areas, as they are at a ‘distance’ from the relevant, but inevitably bureaucratic, government departments and local authorities (Harden and Marquand 1997: 10–11). It is also claimed — 157 —

Towards a Deliberative and Associational Democracy

that they can help overcome the complexity of decisions by providing functional expertise (Weir 1996: 21). Consequently, contemporary legislation tends to provide a framework that leaves much scope for quangos to formulate rules and regulations (Hunold 2001: 151). If quangos are essential to modern governance, then the greater the legitimacy quangos have, the greater their potential for service delivery and effective governance (Harden and Marquand 1997: 19; Hunold 2001: 156): ‘The problem with bureaucratic discretion is not necessarily the existence of administrative power, but its undemocratic exercise’ (Hunold 2001: 156; see also Richardson 2002). According to Harden and Marquand, this legitimacy should be enhanced through the extension of openness, participation and increased transparency in the decision-making process, while the decisions should be based upon reason and publicly available information (Harden and Marquand 1997: 20–4). Consequently, in the UK at least, quangos are required to have some mechanisms in place for citizen participation (Davies 2007: 50), although in practice these mechanisms have questionable democratic credentials and often lead to co-option of participants in order to achieve a veil of legitimacy (Clarke 2002; Milewa 2004). The dualist model of associational democracy would help enhance the legitimacy of quangos by further employing them to organise, host and mediate the forums. First, quangos would be facilitating increased participation in decision-making, with the associations providing a vehicle for popular participation. They would therefore be ‘extending participation’ and ‘providing new forums for active citizenship’, a role envisaged for them by Harden and Marquand (1997: 19), Weir (1996: 35) and Flinders (1999). As discussed in Chapter 3, the inclusion of associational representatives would also increase the public availability of relevant information, and knowledge, which is of particular relevance to their members, and is of the type to which quangos do not necessarily have access (Cohen and Arato 1992: 20; Cohen and Rogers 1995: 42–3; Hirst 1994: 34–40; Warren 2001a: 71–2), including experiential knowledge (Cohen and Rogers 1995: 43; Davies 2007: 56), which is vital to ensuring inclusion in the deliberative process (Sanders 1997; Young 1996), and all of which is essential to solving fluid public problems (Fung and Wright 2001: 18). In the UK we have seen some quangos move towards this role, with the development of ‘dialogic intermediary organisations’ where quangos engage in dialogue with a range of stakeholders, such as the National Institute for Clinical Excellence (www.nice.org.uk 27/07/07; Davies 2007) and the Peak District Park Authority (www.peakdistrict.org 27/07/07; Elstub 2003: Chapter 7). Because decision-making in the forums will be based upon the norms of deliberative democracy, the decisions are also more likely to be based upon — 158 —

A Dualist Model of Deliberative and Associational Democracy

public reasons, which is a principal justification of deliberative democracy (Miller 1993: 82; Benhabib 1996: 72; Elster 1997: 12). In addition, the system would still gain from the expertise of quango members, but would remove their current dominance, as well as harnessing the expertise of secondary associations. In order to maximise openness, and the availability of relevant information, the quangos would also have to ensure that no relevant association is excluded from the forums. To achieve this, it is essential that participants in the forums are self-selecting. If the relevant government agency, mediators or quango members decide selection, and have the power to exclude interested agents, then vital interests and views will inevitably be excluded and the decision can be framed (Rippe and Schaber 1999: 82). This both undermines democracy, as there are those who could be affected by the collective decision that are excluded, and undermines deliberation, as relevant views, information and reasons are not heard. Moreover, it is more likely that the groups excluded from the forums would be those who are currently excluded by present modes of political party and territorial representation. Quangos are a more suitable vehicle to avoid this than governments because they are removed from the adversarial party political process and the potential political bias that comes with it. In addition, in comparison to state legislatures, quangos tend to have much more (often daily) contact with relevant secondary associations, contact that is often based on dialogue (Selden et al. 1988; Vinzant and Crothers 1998, both cited in Hunold 2001: 157–8; Davies 2007). Currently, this is a very unequal relationship, with the bureaucrats dominating and also rarely changing their views (Aronson 1993; Timney 1995; Gould et al. 1996, all cited in Hunold 2001: 158; Davies 2007: 56–7). However, the forums would change and equalise this relationship, as the bureaucrats would no longer be making the decisions themselves, but in conjunction with the associational representatives. What this does indicate, though, is the suitability of quangos to identify key stakeholders, as they have a history of constant engagement with them, although this contact tends to be with the dominant associations (Hunold 2001: 159). In order to avoid this continuing, the forums must be well advertised across a diversity of media, so that relevant associations are aware of the forums. It will also be necessary for the forum organisers to identify and contact key stakeholders, and hopefully, with the establishment of networks between associations, recruitment will also be aided. Inevitably, with quangos identifying the stakeholders, there will always be an element of ‘subjective assessment’ (Hendriks et al. 2007: 366–7), therefore the ‘mobilisation of bias is at its highest’ prior to the commencement of the forum (Smith 2001: 84; see also Rippe and Schaber 1999: 82). — 159 —

Towards a Deliberative and Associational Democracy

A related issue with recruitment arises once the forum has begun. In order to deepen democracy, prevent exclusion of affected groups and maximise the availability of public information and openness, it would be necessary to allow new associations to enter even if doing so creates some difficulties for deliberation. New participants will not have heard all the arguments offered by the incumbent participants, and therefore their preferences will not have been transformed by these arguments when they are finally aggregated, which is a fundamental aspect of deliberative democracy (Elster 1998a: 6). However, the representatives of the new associations can be brought up to speed if minutes have been kept, or meetings filmed or recorded and made available on the quango website, which would also increase the transparency and publicity of the whole process. Yet these representatives will not have had the opportunity to express their reasons, which is a further fundamental aspect of deliberative democracy. To an extent they can do this when they do join. Inevitably, though, this will be limited, as some decisions may have been made prior to their joining the forum, and although their reasons offered could well be relevant to the decision made, it would mean re-making decisions, which affects time and efficiency. Once again we see Weber’s dilemma manifesting itself. Therefore, new associations should not be allowed to join at just any stage, and it might be appropriate to have a cut-off point to prevent an association joining when it is too late for them to engage in effective deliberation and when they would be voting on pre-deliberative preferences. This seems permissible, providing the association had previously been given opportunities to participate in the forum. A further strength of quangos in relation to forum organisation is their flexibility. The mediating forums aim to include associational representatives for all those affected by a decision. This raises some significant practical problems over who is affected, and to what extent. This variability of affectedness has increased as society becomes more complex, with ‘rapid change and fluid boundaries’ (Parkinson 2006: 5). Saward suggests we need a new political unit for each political decision (Saward, cited in Smith 2001: 75–6). This is a significant advantage of mediation, as it ‘tends to be a oneoff conflict resolution or problem solving process’ (Smith 2001: 81), meaning it is more flexible for institutionalising decision-making.7 Quangos are also very flexible because they are not territorially fixed, regularly morph in shape and size, multiply and even break-up and reform (Weir 1996: 21). If they are employed to set up and mediate the associational forums under government guidelines, the institutional flexibility could be met. The quangos, and the forums that they organise, would only be temporary and formed to address a specific issue that had reached the — 160 —

A Dualist Model of Deliberative and Associational Democracy

agenda, ideally through the outside initiative model, and then narrowed by the discursive structures outlined above. The forum would last for as long as it took to make the collective decision, implement the policy (which would be carried out by the relevant elected parliament or council at local, regional or national level, or by a quango, if it is a functionally based issue, who would also ensure that responsible and applicable agents abide by the legislation), review the policy and make any necessary amendments to the policy. Such continuous participatory control throughout the whole process of public administration could also ‘increase the accountability of public power and the public’s capacity to learn from past successes and failures’ (Fung and Wright 2001: 31). This is in stark contrast to mechanisms that allow for popular participation only in the making of decisions, such as referendums, citizens’ juries and deliberative opinion polls. This review process would also be considerably more speedy because decentralisation would mean ‘the distance and time between decisions, action, effect, observation, and reconsideration’ is vastly reduced, so if poor decisions have been made, which is always inevitable even in a deliberative democracy, they can be amended expediently (Fung and Wright 2001: 26). This, then, helps to reach the balance between legitimacy and effectiveness that is central to resolving the Weberian dilemma. Following the completion of this process, both the forum and the quango could then be dissolved. Once this had occurred, in order for the policy to be changed again, it would have to go through the same process again, starting with making it onto the agenda, which again ideally would arise through the outside initiative model and interpretive discursive structures. Again we see the suitability of quangos for the role of forum organiser: due to their flexibility and malleability, they fluctuate in and out of existence, becoming defunct when the goal for their creation is accomplished, or when circumstances change in such a way that makes them redundant, but they can also be reincarnated when circumstances change again (Flinders 1997: 33). If quangos were to organise the forums, it would be essential to ensure that they do not become tools of the government. However, the likelihood of this is reduced by the fact that the forum participants will be partisans, emerging from secondary associations with stakes in the decisions, which provides both the incentive and the ability to scrutinise the quango. Nevertheless, quango appointment must be removed from government control and be subject to new laws which ensure a ‘balanced composition’ (Weir 1996: 36). Again, this is possible due to the malleability of quangos, which makes it possible to ensure that their members include a good socioeconomic mix (Skelcher 1998: 179). It seems that quangos are apt for — 161 —

Towards a Deliberative and Associational Democracy

organising functionally- and territorially-devolved forums and, in this dualist system, enable the expertise of their members to be harnessed by strong measures of democracy, citizen participation, openness, transparency, rationality and publicity. A New Role for Government and Political Parties New roles for secondary associations and quangos in organising and forming legislation inevitably mean changes for current legislative arenas. American activist Brian O’Connel has argued that civil society should be seen as a supplement to, and not a replacement of, current representative institutions, and the arguments here for a deliberative and associational democracy should be taken in this context. Nevertheless, it should be apparent that, in an associational democracy, the legal and political relationships between associations and traditional elected legislative arenas and councils would be altered dramatically in comparison with how they presently stand, which in turn would lead to a changing role for political parties, as they are the main participants in these arenas. Local, regional and national government would have a much-reduced role. They would act as an intermediary and interpreter of the competing discourses emerging from the informal public spheres and set the agenda for the forums. In this sense, the elected parliaments and councils will still retain much power (Schattsneider 1975), but this should still be exercised deliberatively and involve public participation through mechanisms such as the electronic town hall (Parkinson 2006: 170). Following this, it would then be the role of the relevant level of government to form a quango to organise the forum, although, as previously mentioned, quango composition must be regulated by stringent laws. Once a decision in the forum has been reached, it would then be the role of the relevant level of government, or quango, to implement and enforce the decision. Local and regional councils would be almost entirely relieved of their legislating roles, despite the extra powers that would be devolved from central government, as local legislation will be formed in the forums. However, national government would retain some legislative powers over decisions that the principle of subsidiary dictates need to be made at a national level – foreign policy, for example – as problems of complexity, such as size and number of participants, make it unrealistic to be able to hold a national mediating forum. It will also be the role of the various levels of government to ensure mediators are supplied and trained for the forums, possibly in the form of quangos, whose role will be to ensure the forums approximate the norms of deliberative democracy. To clarify the lines of — 162 —

A Dualist Model of Deliberative and Associational Democracy

accountability: national government would be accountable to the general electorate for the policy decisions it makes and all levels of government would be accountable to the electorate on the agendas they set for the forums, not just through the traditional mechanisms of elections and the media, but also through what should be revitalised informal public spheres. Forums are held accountable to the public by the memberships of the secondary associations and, again, the multiple public spheres that will develop around the forums. A combination of the forums, the informal public spheres, the media and the electoral system will hold the relevant level of government or quango accountable over implementation. As the role of traditional elected parliaments and councils is reduced, so too is the role and dominance of political parties, as they will inevitably have to concede many of the roles they presently fulfil to other types of secondary association. Macpherson suggested that, if liberal democracy is to include more direct democracy, with fuller participation, then mechanisms other than the present party system are required (Macpherson 1977: 8). In an associational democracy, secondary associations will increasingly become the primary location for political participation and, through the forums, become the dominant legislators. The justification is that a ‘democratic ecology of associations’ will be less exclusionary to minorities than present political parties, offering more complete representation. Furthermore, associations will not be as dominated as political parties are by political media, which will therefore be more conducive to the institutionalisation of the norms of deliberative democracy. However, as Macpherson argues, if a democracy were to approximate some form of direct/indirect system of participation, its existence in line with political parties is unavoidable, as they must be ‘assumed to be in existence’, but also desirable (Macpherson 1977: 112–13). The same is true for the dualist model: political parties will still operate and have important, but diminished, contributions to make to democracy. For example, parliament, government and local councils would still be elected on a party political basis, with the winning party/parties fulfilling the governmental roles outlined above. Parties would still need to offer policy proposals, as national decisions would still be made in parliament and the agenda for the forums interpreted by government, although with a heavy reliance on citizen-based discursive structures. It is therefore likely that political parties will not offer as wide-reaching manifestos as they presently do, but rather policy proposals for national government and a list of key issues that they feel must be addressed by the forums. As suggested earlier, these agenda issues could still be produced through consultation with secondary associations and verified by public deliberative events. There is also empirical evidence from the — 163 —

Towards a Deliberative and Associational Democracy

USA which suggests that many ‘candidates would have a propensity to operate in this manner under the conditions of a direct deliberative democracy’ (Leib 2004: 130). Representation in the Forums In order for autonomy to be cultivated to its maximum, citizen participation in decision-making was thought to be essential in order to ensure the agency of citizens is respected, and it is central to the conception of deliberative democracy presented here in this book. Consequently, many see representation as a move away from the principle of self-government, as it means all citizens do not actively make the laws that bind them (Kelsen 1949: 285; Lakoff 1996: 327; Dodson 1997: 98). Rousseau (1968) made many, now famous, assertions on the necessity of participatory democracy: ‘A people since it is subject to laws, ought to be the author of them’; ‘Legislative power belongs and can only belong to the people’; ‘Sovereignty cannot be represented’ (Rousseau 1968: 83, 101, 141). Unfortunately, achieving this democratic ideal in practice is almost impossible and, as Post appreciates, ‘what it means for laws to be ‘‘made’’ by the ‘‘same people to whom they apply’’ is not easy to understand’ (Post 1993: 659). Although the mediating forums proposed here would not allow for direct participation of all affected citizens in the decision-making processes, the combination of this institutional method with internally democratic associations does. In large-scale complex societies, representation seems inevitable, and has been accepted as essential to the functioning of modern democracies (Lakoff 1996: 176). Moreover, the discussion of the equalising force that associational representation provides, from the discussion in Chapter 3, indicates that representation also has normative value (Bohman 1998; Young 1997). Therefore, not all citizens will be able to participate in the forums, but the aim to ensure ‘that every citizen’s deliberation is represented’ in decision-making remains (Leib 2004: 37; see also Michelman 1997). Bobbio suggests that there is no absolute as to what direct democracy is, and no ‘watershed’ separating it from representative democracy, as both are linear values of which you can have more or less, with neither being able to survive without the other (Bobbio 1987: 10–11). Striking a balance between the two is perhaps the essence of the Weberian dilemma, and there are numerous alternatives as to how the balance should be made. Such analysis seems to suggest that we should discard the idea that deliberative democracy has to be, purely, a ‘direct democracy’, which in turn seems to reduce the potency of the criticism that it is irrelevant to modern liberal democracies (Cohen 1989: 30). Representation offers a solution to the — 164 —

A Dualist Model of Deliberative and Associational Democracy

problem of including all in deliberative debates, with those not participating directly still feeling as though their reasons have been aired by their representatives (Parkinson 2006: 29). For come commentators this raises a problem for the preservation of autonomy (Gould 1988: 215), as there is potentially a tension between the agency of the represented and the authority of the representatives, as well as issues around how this authority can be made compatible with equal autonomy. Dobson argues that direct participation and representation still share the same commitment in principle to autonomous decision-making. It is more ‘nuanced’ in representative democracy, but citizens can still autonomously select their representative, and hold them to account (Dobson 1996: 127). Gutmann and Thompson suggest that the aim of a deliberative democracy is to approximate a ‘system of representation that will enhance deliberation for all citizens’ (Gutmann and Thompson 1996: 154). It is suggested here that the dualist model of deliberative and associational democracy achieves just that. Certainly, it would alter the mixture of representative and direct democracy from the current parliamentary system dominant in most liberal democratic states. In addition, the dualist model still provides most citizens with the opportunity to be involved in deliberative debates. Deliberation occurs first in associations and then in forums, making it an institutional method similar to what Goodin terms ‘serial’ or ‘disjointed deliberation’ (Goodin 2003: 56). This dualist model has direct democracy and participation from citizens within the secondary associations themselves, followed by representative democracy, with representatives from the associations participating in the forums to make decisions based upon the norms of deliberative democracy. The emerging preferences, needs and identities, formed in the associations, would also be ‘voiced’ in the various informal public spheres that would be formed through informal associational communication, again generally through representatives of the association, but not necessarily the same representatives as those participating in the forums, as the division of labour for this type of representation is more likely to be dispersed throughout the associations’ membership. For example, it might be the role of some members to consult with the media, whilst others voice the preferences and communicate with a certain ‘type’ of association within networks, and still others communicate with government officials of a certain branch. Similarly, some members will participate in forum X, others in forum Y, with still other members participating in forum Z. Ideally, then, each association would diversify the privileges and burdens of the variety of representational roles. It is this combination between direct and representative democracy that — 165 —

Towards a Deliberative and Associational Democracy

enables this dualist model to elude the Weberian dilemma. Moreover, as the associational members select those who are to represent them in the various forums, there is a principal-agent connection which provides a strong and direct bond of accountability between represented and representative. This is essential to the legitimacy of the forums, as they are to be used for decision-making (Parkinson 2006: 74–84) and ensures that representatives are not a ‘self selected minority’, something that Jordan fears is an inevitable consequence of institutionalising deliberative democracy (Jordan 2007: 62). In fact, there would be a shorter chain of representation and agency than is presently found in liberal democracies with political parties, parliaments and bureaucratic agencies (Fung and Wright 2001: 18). It is, then, essential that each association included in the forums has at least a minimal democratic structure, otherwise the legitimacy of their representatives participating in decision-making is completely undermined. Secondary associations will form around a myriad of interests and identities, and in an associational democracy people are likely to be members of a number of associations, as they become key political actors. It is therefore likely that someone could be represented by a number of representatives in a given forum. Each association should have the same number of representatives in the forum, regardless of the size of their membership. This is appropriate in a deliberative democracy, as it is the inclusion of all relevant reasons, rather than an equal representation of all interests and identities, which is key, as Parkinson appreciates: ‘So long as group representatives are present in proportion to their numerical strength, identities and views which command the allegiance of many will always dominate those of the few, regardless of the reasonableness of those views’ (Parkinson 2006: 33–4; see also Williams 2000b: 125). This is a key difference between aggregative decision-making methods, in which sheer numbers are decisive, and deliberative democracy, where reasons should have more sway. Representation is therefore necesssary, but we must establish the type of representation that is required in the forums. Two central issues that must be addressed in any discussion of representation are the form of representation and what should be represented (Bobbio 1987: 5). In terms of the latter, associational representatives will mainly be representing identities and specific, rather than general interests, as this is the nature of secondary associations, which is an important form of representation in a deliberative democracy (Phillips 1995). If it is a specific interest or identity that is to be represented, then it is usually necessary that the representative share that interest or identity (Bobbio 1987: 5–8). Associations are, again, particularly apt for this type of representation, as representatives from associations will — 166 —

A Dualist Model of Deliberative and Associational Democracy

tend to share key descriptive characteristics with those they represent, for example, gender, race, class or interests. This is because representatives will be selected from the membership of the association, and to be a member of the association in the first place one would expect them to share an interest or identity; this makes these representatives particularly ‘reliable custodians’ of members’ interests and identities (Festenstein 2005: 140). Bevan lauded such representation as authentic: A representative person . . . must be of their kind. It becomes full representation only if the elected person speaks with the authentic accents of those who elected him . . . he should share their values; that is, be in touch with their realities. (Bevan, cited in Arblaster 1994: 82) If none or few of the political representatives who make decisions are from certain groups, such as women and ethnic groups, then their interests will be under-represented, no matter how paternalistic and well-meaning the representatives may be. That is not to say that there are not well-meaning politicians and people acting in public life, but that disadvantaged and minority groups need ‘more aggressive advocates’. We need only to consult history to see that when women, ethnic groups and the lower classes were not allowed political representation by a member from these groups, their interests were not effectively protected (Phillips 1995). It was only when these groups won proper rights that their conditions started to improve. Furthermore, as Phillips claims, ‘there is something odd about a democracy that accepts a responsibility for redressing disadvantage, but never sees the disadvantaged as the appropriate people to carry this through’ (Phillips 1995: 43–4). In terms of the second question, regarding how associational members should be represented in the forums, representatives can be either a ‘delegate’ or a ‘fiduciary’. If they are delegate, then they are bound completely by the wishes of those they represent. In essence, the representative is spokesperson without the authority to make decisions. In contrast, if the representative is a fiduciary, then they have more powers of authority to act on their constituents’ behalf (Bobbio 1987: 5). It is apparent the representatives must be bound to a certain extent by the interests and preferences of the associations from which they derive, otherwise the social and cultural groups who are currently under-represented will still not have their interests represented, and will remain excluded from the decision-making processes. The inclusion of a diversity of views, essential to deliberative democracy, would also be compromised. However, if they were bound too tightly by previously agreed ideas and interests, then many of the benefits that arise — 167 —

Towards a Deliberative and Associational Democracy

from discussion would be prevented from occurring, as the representatives would not change their preferences in light of new information and perspectives, and debate would cease to be an exploratory process in which new possibilities are created (Parkinson 2006: 31–2; Hendriks et al. 2007: 366). A balance between the two is therefore required: representatives must be held accountable, be bound to some degree by the preferences of their associational members and be open to dismissal if it is felt they have represented their people poorly, but they must also be free to participate fully in a discussion, and that means changing the preferences and goals with which they started. This raises a problem, as the representative will have been engaged in a democratic debate (in which their preferences were likely to be adapted) while the other members were not. Inevitably it will be the representative’s constituents, in this case the members of the association, who will ‘act as the ultimate safeguard against selling out’ (Amy, cited in Smith 2001: 80). Young argues we need ‘representation as relationship’ (Young 2000: 125), where the representatives must explain and justify the resulting decision to the members and provide the information that caused them to change their preferences; if they cannot do this, then ‘perhaps this can be traced to the co-option of the representative by other parties’ in the forum (Smith 2001: 80). The very fact that representatives will provide reasons for their preference changes, and the compromises they have reached, means the represented are more likely to trust those representing them and come to appreciate the need for pragmatic compromise (Festenstein 2005: 143). In this sense, the associational members are still involved in the discussions of the forum. It also highlights the importance of agentprincipal bonds that the associational representatives in the forums will have, as without this ‘representation as relationship’ and genuine accountability are very difficult to attain (Parkinson 2006: 32–3). This process of accountability will be aided by the publication of the forum’s minutes and the forum’s meetings could also be filmed and made available as a podcast, both being posted on the quango’s website. Furthermore, the mass media is likely to scrutinise and publicise forum debates to some degree, given their legislative function. Forum debates are unlikely to be a one-off event, so continuous debate between associational representatives and members should occur before, during and after the forum process. With the partisan role of representation outlined in this dualist associational model, there is a significant danger that representatives from the secondary associations will not be open to the transformation of preferences that is essential to the legitimacy of deliberative democracy (Smith 2000), with associational representatives’ preferences being too inflexible to make them competent deliberators (Hendriks 2006: 497). Urbinati is adamant — 168 —

A Dualist Model of Deliberative and Associational Democracy

that passionate commitment, likely to be channelled through associational representatives, will not undermine the possibility of preference transformation (Urbinati 2000: 775). In fact, Immergut thinks that partisan associational representatives are more likely to change their preferences than unpartisan citizens, as they will think of ‘policy packages’ that require compromise (Immergut 1995: 205). The empirical evidence on this is mixed, and limited, with some research (Pelletier et al. 1999; Hendriks 2002: 70; Hendriks et al. 2007) suggesting that partisan representatives will not significantly alter their preferences in deliberative situations, and other research indicating they will (Elstub 2003: Chapter 7; Fung and Wright 2003; Parkinson 2006: 136). Partisanship, though, does not always prevent deliberators from making genuine attempts to reach agreement, at least on some issues, which suggests compromise, if not preference change (Hendriks et al. 2007: 370). This was also highlighted as more realistic, and yet still normatively desirable in terms of cultivating autonomy, in Chapter 2. Nevertheless, the norm in partisan forums seems to be for agreement to be ‘elusive’ (Hendriks et al. 2007), unless the decisions are addressing local and specific issues, which has already been advocated as essential for the associational forums outlined here (Fung and Wright 2003). Despite this mixed range of empirical evidence, it is a positive benefit that participants in a deliberative democracy should be partisans in order to ensure that the information, needs and beliefs expressed are authentic and genuine: ‘Far from transcending the specific situation of citizens, deliberative reasoning rests on the premise that specificity needs to be known and acknowledged’ (Urbinati 2000: 776). Empirical evidence indicates partisanship provides greater motivation to participate (Fung 2003b: 345; Parkinson 2006: 134), increases the sustainability of the forums, and is necessary for preference change to be reflective and for decisions to be supported and implemented (Fung 2003b: 345). Representatives in the forums must, then, be given some element of freedom to operate, but this does not involve the complete abandonment of sectional interests. The Danger of Co-Option An important and established criticism of corporatism, mediation and dualist strategies is the iron law of oligarchy (Michels: 1959). The theory suggests that legislative inclusion and institutionalisation of civil society will necessarily result in ‘cooptation, deradicalisation, professionalisation, bureaucratisation and centralisation’ and eventually the dilution of aims (Cohen and Arato 1992: 557; see also Dryzek 2000: 107; Smith 2001: 81). Co-option can be defined as ‘the process of absorbing new elements into — 169 —

Towards a Deliberative and Associational Democracy

the leadership or policy-determining structure of an organisation as a means of averting threats to its stability or existence’ (Selznick, cited in Dryzek 2000: 88). Such trends go against the aims this chapter has argued the dualist model can achieve, as they would not deepen democracy and therefore are unsuitable methods by which to institutionalise deliberative democracy and could not lead to the cultivation of autonomy attributed to it. Such considerations have prompted deliberative theorists, such as Dryzek and Hendriks, to be against the inclusion of associations in the state or legislation altogether, seeing such a system as exclusive, unable to challenge inequalities, and predicting that associations would only ever be symbolically included and, moreover, co-opted (Dryzek 2000: 85; Hendriks 2006: 497). Empirical research does suggest that certain groups do exclude themselves from potential state inclusion for fear of co-option (Szasz 1995: 150; Sagoff 1999; Hendriks 2002; Elstub 2003: Chapter 7; Thomas 2003; Parkinson 2006). Specifically, Dryzek and Hendriks suggest that inclusion reduces opposition, with no real power transference, and that public policy is already determined due to state imperatives. Dryzek argues that entry of the association will only occur when the state recognises the interest of the group as a challenge to its legitimacy, not because it recognises the interest as legitimate in its own right. The group is incorporated if, and only if, the state is pursuing a certain public policy that overlaps with the claims of the group. In this sense, there is no real transfer of power to the groups, nor does the group remain as a challenge to the state’s legitimacy in the informal public sphere. Associations in these circumstances further lose the potential to be radical because they have to behave responsibility in order to ensure inclusion (Fung and Wright 2001: 34). Overall, then, the public sphere is sanitised, with little democratic transfer of power away from the state being achieved. There are many examples of such co-option occurring. Dryzek provides the example of the Salinas administration of 1988–94 in Mexico, who introduced PRONASOL (National Solidarity Campaign), which combined grassroots participatory democracy and centralised guidance, but resulted in co-option (Dryzek 2000: 92–3). Young points to the Black Liberation Movement operating in the USA in the 1930s, the neighbourhood movements of the Mission District in San Francisco towards the end of the 1970s and the ‘New Populism’ movement, all of which were widely supported but ultimately were unable to bring about the institutional change they initially sought, due to co-option (Young 1990: 89–90). Warren suggests the German Greens and the Equal Rights Amendment in the USA also gained strategic victories at the expense of core principles that motivated their members (Warren 2001a: 121). — 170 —

A Dualist Model of Deliberative and Associational Democracy

However, the point of the dualistic approach considered here is that there is no loss of a vigorous civil society because the associations still remain there, whilst also gaining a legislative role in the forums, and when agents are given decision-making power the danger of co-option is significantly alleviated (Leib 2004: 121; Hunold 2001: 158). Implicit in Dryzek’s argument is that the same association cannot achieve both elements of a dualistic strategy, but this surely depends on the institutional framework that incorporates associations into the state. This chapter suggests a devolved forum system could ensure this, and there are examples of associations that have successfully combined oppositional and co-operative strategies, such as the Texas Industrial Areas Foundation (Warren 2001b) and the Pacific Institute for Community Organizing (www.piconetwork.org 20/07/07; Wood 2001). It is apparent that the associations that participated in the forum that produced the policy, but disagreed with the policy, could still oppose it through contributing to critical discourse in the informal public sphere. Associations that agreed to the policy, or most of its elements, would have their oppositional edge blunted to a degree, but could still oppose other policies, as well as criticising the interpretation of the agenda. As participation in the forums is, to a large extent, self-selecting, such oppositional groups could not be excluded from the forums in the future, which reduces the capability of the state to co-opt them. Moreover, once associations are given more power, they will ‘play important roles in stabilizing and defending participatory institutions against counter reforms’ (Fung 2003a). Dryzek’s second claim is that inclusion of all relevant secondary associations is unnecessary because policy is never completely undetermined. All states must fulfil the imperatives of accumulation and legitimation, which means that groups in opposition to the state will be incorporated impotently, and only when their interest is directly related to a state imperative. The only elements of public policy to be decided are how best to meet the state imperative and how to achieve a balance between incompatible imperatives, for which the group can help with ideas, information and enforcement, but there is no real transfer of power to these associations. Granted, those groups whose interests do not coincide with the state’s imperatives will find it very difficult to affect public policy in a meaningful way. However, there does seem to be more scope for secondary associations being included into the policy process than Dryzek gives credit for. The state imperatives of legitimation and accumulation, although restricting, are very broad and leave plenty of scope and plenty of alternatives for public policy, particularly as these imperatives can be in conflict, meaning tradeoffs need to be made. Consequently, civil society could play a relevant role — 171 —

Towards a Deliberative and Associational Democracy

in deciding where the trade-offs between these imperatives should be made, even if it is inevitably constrained from abandoning one or other of the imperatives altogether. Moreover, there are many areas of public policy that have little relevance to either of the state imperatives, so the role of associations here seems even less constrained, a point that Dryzek (2000) accepts but then dismisses as peripheral zones of public policy which must still not transgress state imperatives. Inclusion of associations is currently not a state imperative and, for Dryzek, is unlikely ever to be (Dryzek 2000: 93). Nevertheless, state imperatives do change over time, and therefore it is possible that inclusion could become an imperative. Legitimation is already a state imperative, and the state would be more democratically legitimate if it did include all associations, rather than excluding some. It is possible to imagine that if these associations were to put significant pressure onto the government to be included, it could become part of the state’s imperative of legitimation to include these associations. However, if, due to present exclusion, these groups are in a subordinate position to the extent that they do not have the resources to gain effective mobilisation, then it is unlikely that they will successfully pressurise the government, and if it is not going to be excluded groups who are the agency for change, then it is not clear who it will be. I presume that it is this type of analysis that has led Dryzek to claim that he cannot think of a scenario under which inclusion would become a state imperative (Dryzek 2000: 93). However, my argument here is supported by Warren, whose analysis suggests that in very conflictual policy areas the relevant state agencies’ imperatives of increasing legitimacy may provoke an interest ‘in the democratic mix of associations’ (Warren 2001a: 217). Dryzek does acknowledge that democracy does require that ‘reflective preferences influence collective outcomes, and so both an orientation to the state and discursive mechanisms for transmission of public opinion to the state are required’ (Dryzek 2000: 162). It is precisely because of this that a dualistic model is absolutely necessary. If power is transferred from the state to associations in a democratic forum, then the associations will not be powerless even if they will be restricted by the state imperatives. As mentioned in Chapter 3, initial institutional change must accept liberal democratic structures and the capitalist economy as being in place. In relation to the issue of democratisation, these forces inevitably bring with them both limitations and supportive conditions. State imperatives are, then, characteristic of these opposing forces.

— 172 —

A Dualist Model of Deliberative and Associational Democracy

Conclusion On a normative level, it has been suggested that deliberative democracy can make citizens’ preferences more prudent, making them more autonomous. However, this cannot be achieved unless deliberative democracy can be meaningfully approximated. In doing this, the ideal of deliberative democracy faces a Weberian dilemma: how to ensure that deliberation and democracy are effectively combined so that citizens actively engage in deliberation while ensuring the results of the deliberations are actualised into binding decisions. Weber’s dilemma can never be completely avoided, and is always present in practical considerations of democracy. Indeed, in the course of this chapter we have seen several tensions between democracy and deliberation. Overall, there is the tension between achieving efficiency in decisionmaking and deliberation versus increasing participation, and the problem of linking macro deliberation in the informal public sphere with micro deliberation and decision-making. Dualism itself was seen as the solution, with the same associations participating in each. More specific tensions included agenda setting. It was suggested the informal public sphere can, and should, set the agenda, as it provides a relatively open arena for public communication in which associations in civil society could raise a variety of concerns. However, micro deliberative arenas, where decisions are made, require clearly defined and specific agendas for efficient decision-making, which the informal public sphere cannot provide. Governments will inevitably play a key role in this process, but should invoke mechanisms for popular deliberation to achieve this. Second, there was inequality of access to the informal public sphere, resulting in the voice of subordinate groups being excluded or marginalised. Multiple and fluid associational membership; multiple public spheres generated by a multiplicity of legislative forums; a role fulfilled here through the devolved deliberative mediating forums; close relations with micro-publics, once again through the forums – all of these were thought to alleviate inequalities. These same inequalities will also affect the deliberative forums, but their inclusiveness, legislative power and invocation of deliberative democracy helps balance power, to a degree, in these sites. Given deliberative democracy’s need for preference transformation, we also considered whether decisions should be responsive to reflective preferences regardless of whether these decisions are rationally compatible, and how associational members should be represented by fiduciaries, who are held accountable by ‘representation by relationship’. In this model, territorially and functionally devolved forums guided by — 173 —

Towards a Deliberative and Associational Democracy

the principle of subsidiarity are employed to ensure citizens’ deliberations affect decisions, with quangos providing the flexible organisation to host these forums. Within the forums, representatives from interested secondary associations assemble to make decisions based upon the norms of deliberative democracy. Further connection between citizens’ deliberations and democratic decision-making is ensured by the requirement that these representatives must come from secondary associations that comply with the norms of deliberative democracy at a basic level, an issue that will be explored in full in Chapter 5. Co-option of civil society by the state is a significant danger to such a system, but with the same associations participating in the informal and formal public sphere, there is still plenty of opportunity for critiques of the state. The current liberal democratic and capitalist system inevitably limits the ability of the dualist model of deliberative and associational democracy to respond effectively to these dilemmas in terms of eliminating private influences of pressure groups, achieving equal levels of power among all socio-economic groups and in terms of avoiding co-option, but we must start from the here and now, and that means liberal democracy. Nevertheless, this model offers genuine and radical alternatives to the current institutional make-up that will much more closely approximate deliberative democracy. This is not to say that the dualist model is the only possible method by which to achieve an effective link between citizens’ deliberations and decision-making and macro and micro deliberation, but it is one possible institutional mix that emphasises the importance of democracy as well as deliberation, and could enable autonomy to be cultivated. Notes 1. There are further problems with this model. For a good discussion of these, see Bohman (1996), Leib (2004) and Hendriks (2006). 2. In the dualist system here, this method would be reduced due to the devolved legislative forums that secondary associations could participate in. 3. This is obviously a circular argument. However, the point is that once an associational democracy has been achieved, the ‘outside initiative model’ of agenda setting will be much more predominant than it is now. The problem remains how to achieve the associational model in the first place, so that this phenomenon can occur. 4. In Europe it has generally been capital and labour that have been included in a tripartite system (Mansbridge 1995: 136), which was the case when practised by ‘radical left-wing local authorities in Britain in the 1980’s’ (Martell 1992: 168). 5. German planning cells are similar to, although usually larger than, the more widespread citizens’ juries, where a small group of citizens are brought together

— 174 —

A Dualist Model of Deliberative and Associational Democracy to discuss an important issue and produce policy advice. See Renn et al. (1995) for a detailed discussion. 6. Although there is some evidence to the contrary (Jacobsen 1998; Fung and Wright 2001: 35), this is from forums which were excessively controlled by government officials. 7. This is a disadvantage in terms of deliberative democracy because if it is not an ongoing process, then factors such as the ‘civilising force of hypocrisy’ are less likely to pertain (Mackie 1998: 84–5; Dryzek 2000: 46).

— 175 —

5

Democratising Secondary Associations

Introduction The central argument of this chapter is that in the deliberative and associational model of democracy, secondary associations should be democratised in order to ensure that they have an internal democratic structure whereby all the members of the association participate in the decisions made in that association: ‘The democratic associational model of voluntary organisations assumes that members should not only be expected, but actively encouraged to participate in the running of the organisations’ (Lansley, cited in Powell and Guerin 1997: 166; see also Perczynski 2000; Fung 2003a). As already discussed, there are many forms of participation, and several ways decisions can be made democratically. The case was made in Chapter 2, that deliberative democracy was the model of decisionmaking most suitable for cultivating autonomy, which was established as the normative core of democracy in Chapter 1. Consequently, this chapter makes the stronger claim that associations should engage their members in deliberatively democratic decision-making (Young 1990: 91; Elstub 2006b). Chapter 3 established that associations can fulfil many functions that contribute to the institutionalisation of deliberative democracy and to the cultivation of citizens’ autonomy. However, if associations are to fulfil these functions effectively, then the internal structure of the associations must be democratic and based upon the norms of deliberative democracy. If associations are to be venues for subsidiarity and offer scope for more small-scale participation, then they must allow participation and therefore democratise their decision-making structures. If the principle of subsidiarity is to be introduced, then those associations that will be devolved powers to fulfil key public activities and deliver important services must be accountable to those they serve, which suggests the need for an internal democratic structure within the association (Hadley and Hatch 1981: 147; Putnam 1993: 147; Warren 2001a: 36; Elstub 2006b). If associations are to provide — 176 —

Democratising Secondary Associations

authentic information and representation, then these must be linked to the participation of all those who are said to be represented: ‘deliberative reason should not be divided so that representatives give reasons while citizens merely receive them’ (Gutmann and Thompson 1996: 358–9; see also Mansbridge 1995: 143). Participation also needs to be deliberative, as preferences are not set but must be formed through deliberation in order to be autonomous. If associations are to be schools of democracy and develop citizens’ capacities to participate deliberatively, then this can only be achieved if members get to participate in deliberative democracy. It will be more likely that associations will produce public-spirited, tolerant, knowledgeable citizens, and thus create civic virtue, social trust and improve co-operation and civic attachment, if they are horizontally and democratically organised (Mansbridge 1992: 51; Putnam 1993: 173–5; Warren 1996: 241). In Chapter 4 it was further suggested that there must be a principal-agent link between the associational representatives, participating in the decision-making forums and the associational members they represent if the forums are to be legitimate decision-making arenas (Parkinson 2006: 89, 96–7). Furthermore, as argued in Chapter 1, citizens need opportunities to assert influence over decision-making if their agency, and hence autonomy, is to be preserved. Some degree of participation from those affected by decisions is therefore a fundamental democratic principle. This is not to say that secondary associations should be the only mechanism through which citizens participate, but that in an associational democracy they need to participate in these associations to some degree. However, there are five key arguments against this claim, which must be considered. The first two challenge the normative claim for participation in secondary associations. The first of these two is the argument that in an association with high opportunities for exit, internal democracy is not necessary for autonomy. The second of these two considers the legitimacy of legislating associations in this manner, as to do so goes against freedom of association because forcing legislation on associations and regulating their internal structure – or, alternatively, allowing associations complete control over their membership and proposals – can infringe upon the autonomy of excluded citizens. The remaining three arguments challenge the possibility of associations being internally democratic on empirical grounds. The chapter considers Michels’ famous ‘iron law of oligarchy’ argument, where it is claimed that democratising any organisation is impossible because representatives cannot be effectively held accountable to the membership. We then move to discuss whether most citizens actually want to participate, and can participate in associations, with apathy, the logic of collective action, the desire to maintain social harmony and empirical restrictions of — 177 —

Towards a Deliberative and Associational Democracy

size, time and disparate membership all restricting the possibility of democratising associations. The final argument concerns the need for state regulation to ensure associations meet democratic standards, whether this will eliminate the advantages associations offer and whether associations can in fact be reformed in this manner. Exit and Voice Following Hirschman’s (1970) famous distinction between exit and voice, Warren argues that an internally democratic structure for all associations would not necessarily contribute to democracy overall and could even detract from it: ‘the greater the chances for exit from an association, the lesser the chances that voice will have an impact within the association’ (Warren 2001a: 96). This means that if an association has high costs of exit, then it is more important for it to have a democratic structure that gives members voice, but correspondingly means that associations that have low exit costs will not have the incentive, or need, to incorporate the voice of dissenting members. Opportunities, and need, for voice therefore increase as the opportunities for exit decrease (Hirschman 1970: 34). In Chapter 3, the fact that associations are relatively voluntary, to a significant degree, in the sense that they involve choice over entry and exit, was lauded as a strength that aids the cultivation of autonomy and fosters deliberation. However, it also means that the membership will be fairly homogenous in terms of the members’ purposes, which, according to Warren, will lead to a general consensus over goals and means of the association. Consequently, voice will not be encouraged, and in fact dissenters seen as challengers to the consensus will be encouraged to remain silent or leave the association rather than threaten ‘the solidarity, mission or purpose of the group’ (Warren 2001a: 104). It is due to this homogeneity of associations that Femia finds the avocation of secondary associations as a location for deliberative democracy ‘mystifying’, as people join such associations precisely because they ‘pander to a well-defined set of preconceptions or passionately held convictions’ (Femia 1996: 3). It is also true that the greater the diversity and proliferation of associations, the more choice there will be and so the less costly exit will be, as there will be similar alternatives to choose from. In this sense, there is undoubtedly a connection between ease of exit and opportunities for voice in an association. In an associational democracy there is likely to be a proliferation and diversity of associations, as they will become a key avenue for political participation and governance and will increase in number and variety if public finances are provided for their formation and running, and if new types of association are sought (Offe 1995: 127). — 178 —

Democratising Secondary Associations

Most importantly, associations with high exit levels may have a ‘purer’ message to represent and can have greater legitimacy in claiming accurate representation of their members’ beliefs and preferences. In this sense, associational democracy externalises conflict and creates ‘silence within an association’ (Warren 2001a: 97). Associations will then encourage high exit rates for members who disagree with the messages of the association, rather than deal with the dissent in a democratic way, to ensure their public voice is not weakened by uncertainty and remains clear and unanimous (Warren 2001a: 36). Undemocratic associations are still important to democracy, despite their undemocratic structure, as they still make important contributions to discourse and opposition in informal public spheres. For example, there are groups that are conservative in outlook and demands but relatively democratic in structure, such as US anti-abortion group Operation Rescue (www.operationrescue.org 27/07/07), and those that are very hierarchical, such as Greenpeace (Dryzek 2000: 100; Barry 2001: 165), and Berry’s evidence from a host of associations in the USA seeking increased environmental protection, consumer rights, racial and gender equality suggests that undemocratic associations are more effective at achieving their aims (Berry 1999). Therefore, it is less important to have a democratic structure, and principal-agents bonds between members and the associational representative when associations are participating in informal public spheres. This is because having such requirements would raise the costs of participating to too high a level, reduce the more informal types of communication often prevalent here (Parkinson 2006: 154) and could compromise the diversity of public reasons included in the informal public sphere, and therefore diminish the quality of public deliberation. In terms of associations participating in the informal public sphere, then, exit is sufficient for accountability and legitimacy. Nevertheless, if associations involved in delivering services are participating in the decision-making forums, I maintain that an internal democratic structure that provides voice for the members is still essential, as exit will not be sufficient to achieve legitimacy and accountability here by itself. Hirst is in favour of a combination of exit and voice, but in contrast to the argument here suggests associational members should only participate to elect the association’s governing body, due to the reluctance of most citizens to participate and to assert more voice in ways more substantial than a periodic vote. He argues that exit is sufficient to achieve further accountability because members will simply move to a competing association if the service or representation they receive from their current association is unsatisfactory (Hirst 1996: 113). — 179 —

Towards a Deliberative and Associational Democracy

However, I still maintain that more voice than this is needed. Associations can reduce exit, even if other exit costs are low, by instigating loyalty in the membership. However, Warren maintains that exit levels will still be high because of the ‘costs and uncertainties of the internal political process’ (Warren 2001a: 97). But exiting to join another association presents significant costs and is certainly uncertain, so retaining current membership might be the more desirable option. As Hirschman noted, low exit cost is not necessarily a good thing for the association, as those who leave may well be the most ‘quality sensitive’ and an association can lose members who would be a valuable resource (Hirschman 1970: 33). The presence of an internal democratic structure may prevent their exit, though, if they feel they can voice their concerns and influence the running of the association. Furthermore, I think Warren and Femia overstate the homogeneity that will exist in associations. It is true to say that there will be a shared interest, belief, preference or identity that will motivate people to join or form an association, but this may be as general as being a single mother, wanting to help protect the environment or being a train driver. There is nothing to suggest that because one shares these common factors, there will be exact agreement on what the purposes or methods of the association should be (Young 1990: 48): Group representation avoids most of the pitfalls in appealing to shared experiences as an automatic guarantee. It makes no claims to essential unities or characteristics; it recognises 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. (Young 1990: 54) Furthermore, the association will be unable to determine if a consensus exists or not, without having a democratic debate where they discuss the issue and realise whether they do in fact agree on purposes. Even then, it would not be proof that there was a prior consensus, as the agreement could have been generated by the debate. Empirical evidence highlights the need for arenas where citizens can deliberate amongst the like-minded, or those with similar interests and identities, prior to deliberation in more combative forums (Campbell and Oliver 1996; Barnes and Bowl 2001, both cited in Barnes et al. 2004: 107), and democratised associations can provide this opportunity. It might be the case that, despite the presence of dissenters in an association, the majority of the members are in agreement with the — 180 —

Democratising Secondary Associations

associational elites. However, to know this, an association would have to have a minimally democratic structure in which a majority could be established through aggregation. Either that or the fact that associational members remained members of the association is a sign of tacit consent; But even in associations with relatively low exit costs, there are always costs of some kind. Evidence from trade unions in the USA suggests voice is more effective than exit at gaining the attention of the association’s elites, even though these unions are generally not very democratic (Freeman and Medoff, cited in Mansbridge 1992: 44). Other empirical evidence indicates that associations do not passively represent the interests of their members, as these interests and preferences are revised and transformed through interaction of the membership of the association (Streek and Schmitter 1986). This interaction should therefore be at least partially deliberative and democratic, rather than purely aggregative. Just as it was argued in Chapter 3 that the state will not have a consistent definition of people’s needs, neither will associational elites (Pascall 1997), and these therefore need to be established through democratic debate (Fraser 1992; Gould 1996; Elstub 2006b). Undemocratic associations that provide important services can be agents of social control, even with opportunities for exit (Elstub 2006b: 28–9). One of the great strengths of associations is their proximity to the ‘lifeworlds’ (Habermas 1990) of citizens, making them useful arenas for the identification of new problems, enabling the representation of new perspectives and information, and delivering services that meet the needs of the citizens. However, if associations presume a consensus already exists amongst their members, then they will not be sensitive to new problems and information, differing types of need or interpretations of needs, and the effectiveness of the association in delivering services and participating in decision-making in the forum will be compromised. Without internal democracy, elites will not be fully aware of the interests and preferences of their membership, and consequently these will be distorted (Mansbridge 1992: 51). In contrast to informal public spheres, the representation in decisionmaking forums of preferences that are not agreed to by an association’s members, or that are asserted by representatives of it who are not accountable, is not democratic. Neither is the delivery to people of services solely devised by associational elites. Therefore, legitimacy, and accountability of associational service delivery and representation in macro deliberative and formal public spheres should not be achieved by either exit or voice, but through a combination of both.

— 181 —

Towards a Deliberative and Associational Democracy

Freedom of Association Opportunities for associational exit help the cultivation of citizens’ autonomy by increasing choice. However, if associations are to meet standards of internal democracy before they are devolved powers and prior to being included in the forums, choice is decreased. Here we see a paradox of an associational system: A government that is constitutionally dedicated to liberal democratic principles has a strong interest in supporting a vast assortment of associational activities among its citizens. But it also has a strong interest in regulating associations so that they support a liberal democratic form of government and public policies that are consistent with liberal democratic principles. (Gutmann 1998: 18; see also Barry 2001: 123) Therefore, the requirement for associations to have an internally deliberatively democratic structure raises two important questions for the freedom of association: their freedom of voice and their freedom of membership. These two arguments against the democratisation of associations originate from opposing ideas. The first is the suggestion that it is illegitimate not to grant an association political recognition because it does not have an internal democratic structure, as this causes excessive external control and infringes the right to freedom of association and association voice. The second argument is that associations should not be free to select their membership and regulate themselves internally because this can mean ‘dangerous opinions’ are expressed and allows exclusive membership, both of which can lead to discrimination. Both these arguments are now considered in turn. Restriction of Associational Voice The question of whether it would be legitimate to exclude associations that do not meet internal standards of democracy from collective decisionmaking and from supplying services is a contentious one, as it would confine their political participation to the informal public sphere. The criticism, presented effectively by Gutmann, is that by failing to recognise associations that are internally undemocratic, we also fail to treat citizens as autonomous agents who can choose the type of association that would suit their needs and identity best (Gutmann 1998: 23). The suggestion is that individuals can choose to join an undemocratic association autonomously therefore this decision should be respected and the autonomy of those — 182 —

Democratising Secondary Associations

making such a choice should not be sacrificed by excluding undemocratic associations. Gutmann accepts that autonomous associational membership requires informed ongoing consent (Gutmann 1998: 23), but she also thinks that this consent can be tacit: ‘to be consistent with living life as a free person, continued membership in an association must be a sign of ongoing consent to the association’s purposes’ (Gutmann 1998: 23). The idea, then, is that if individuals consent to the undemocratic structure of their association, then this is permissible providing there are opportunities for exit. It is certainly possible, and even likely, that certain members of associations will not want to participate in their associations (Walzer 2004: 85), however, there may be many who do, and denying them this opportunity can ‘preclude some citizens from shaping their own relation to the polity’ (O’Flynn 2006: 149). However, Kateb warns that ‘the web of relations housed in an association can take on tremendous value, greater than the goals of the association’ (Kateb 1998: 37). In this classical liberal view, this would mean that the freedom of association should take priority, no matter how undemocratic it is. Weak paternalism is, however, justified, providing its motivation is to ensure that autonomy is still preserved. If we cast our minds back to Chapter 1, weak paternalism was claimed to be legitimate, and therefore it is legitimate to restrict autonomy at a specific time to ensure the possibility of autonomy in the future. Therefore, it is also legitimate to prevent people having undemocratic associations supply them services and represent them in collective decision-making forums. This is justifiable, even for citizens who have no interest in being autonomous or participating in a democracy because it ensures that they will have the opportunity to participate in the future. In Chapter 1 it was also argued that we cannot assume citizens are autonomous, only ensure the right conditions are present. The requirement that associations should have an internally democratic structure is aiming to meet a key requirement for autonomy, if associations are to become primary political vehicles. I do not think this leads us to a Rousseauian paradox of ‘forcing people to be autonomous’, as without these democratic conditions people will not have sufficient control over their lives and the acceptability of choice will be diminished, not enhanced. In Chapter 1, a mild liberal perfectionism was also justified. This perfectionism favoured democracy in order to cultivate autonomy. Here we see the consequences of this, which is the exclusion of undemocratic associations that could not ensure the autonomy of their members. I return, then, to Taylor’s point, that liberalism is a ‘fighting creed’ (Taylor, cited in Cooke 1997: 280). Autonomy is the key value of modern liberalism and I am suggesting that excluding undemocratic associations — 183 —

Towards a Deliberative and Associational Democracy

from participating in formal decision-making arenas is one of the best methods to cultivate it. Barry disputes this type of argument, maintaining that it is not part of liberalism to ‘insist that every group must conform to liberal principles in its internal structure’ and that ‘liberal principles themselves demand that groups should have the utmost freedom to handle their affairs in accordance with the wishes of their members’ (Barry 2001: 147–8). As discussed in Chapter 1, Barry rejects a conception of liberal democracy based upon the value of autonomy, so would be against the line of argumentation in this book; however, he acknowledges that justification of liberal democracy on this value has significant consequences for the freedom of association, which is defended upon the value of autonomy, the opportunity for members of associations to make informed and reflective choices from a range of realistic options (Barry 2001: 147). My argument is that without engaging in deliberatively democratic discussion, citizens are unable to make these informed and reflective choices, and consequently associations should be organised internally around its norms. Apart from our views on autonomy, the key difference between my position and Barry’s is that he suggests that associations should not have to conform to the same liberal democratic principles that are used to regulate public bodies. However, in an associational democracy, associations become a primary tool for political participation, representation, policy formation and service delivery, so do at least become quasi-public bodies. I therefore maintain that they should be regulated by the same principles, if they wish to gain this level of political recognition. If they do not, then, as Barry suggests, they should be free to have whatever internal structure they choose, providing exit it is still possible, and can still participate in the discourses of informal public spheres. This means that, in an associational democracy, undemocratic associations such as the Catholic Church would still be permitted to exist, but they would not be permitted to run schools or participate in collective decision-making. It was argued in Chapter 2 that there should be no a priori censorship of views, as the deliberative process itself would launder prejudiced, irrational and undefendable preferences. The demand for internal democracy in associations could be seen as a contradiction to this, as it inevitably restricts the voice of some associations and could also been seen as a sinister challenge to the validity of ethnic groups (O’Flynn 2006: 148). However, it in fact only restricts associations that are elitist and hierarchical, for we can imagine associations formed around racist and sexist opinions forming these opinions democratically. The freedom of the association, and its voice, is therefore only restricted to the extent that it cannot be undemo— 184 —

Democratising Secondary Associations

cratic, and this restriction is entirely compatible with the mild perfectionism justified. The laundering of the opinions of such associations will occur through macro deliberations in the informal public sphere and micro deliberations in the devolved forums, as they attempt to couch their views in public terms. In an associational democracy, where associations become a primary location for political participation, freedom of association is definitely compromised because people will not have much choice about whether to join one or not. Neither will associations be left with much choice about their internal structure if following anything other than the democratic norms means exclusion from service delivery and policy-making. Nevertheless, although freedom of association is an important right, and aids autonomy, it should not be elevated to the most essential liberal right so that everything else is secondary: ‘seeing the right of association as fundamental . . . gives considerable power to the group, denying others the right to intervene in its practices whether in the name of liberalism or any other moral ideal’ (Kukathas, cited in Barry 2001: 131–2). The conditions necessary for autonomy should be privileged due to its intrinsic value established in Chapter 1. Freedom of association is essential to this, but democracy is more so, therefore such associations should be infused with democratic processes. As associations are not purely voluntary, they cannot be purely equated with individual freedom and therefore democratic decision-making in public and private institutions should take priority (Hirst 1996: 99). Freedom of Associational Membership This criticism, in contrast to that covered above, suggests that internal democracy of associations would give too much control, and power, to the associational members, especially over membership, which would allow associations to exclude people. There are several high-profile examples that are symptomatic of this, such as the Supreme Court decision of Roberts v United States Jaycees (1984), where it was decided that the Jaycees (the junior chamber of commerce) could not legally discriminate, and so should be forced to accept women members. In contrast, the Supreme Court’s ruling on Boy Scouts v Dale concerned a man who was dismissed from the scouts for being gay. The court ruled that this was a justifiable decision for the Boy Scouts to take, based upon the First Amendment right of ‘expressive association’, or, as Galston refers to it, the right to ‘organise around the articulation of its preferred core values and to select members and leaders consistent with those values’ (Galston 2000b: 929). Recently in the UK — 185 —

Towards a Deliberative and Associational Democracy

legislation was passed that prevents such discrimination by organisations and was opposed by the Catholic Church, as they did not want to let homosexuals adopt children through their organisation. Kateb (1998) suggests freedom of association is essential to freedom because it enables freedom of choice, and therefore it is essential that members should be allowed to choose the goal of the association and the criteria for membership (see also Barry 2001: 127–8). This, then, suggests that associations should be free from state interference and regulation, based upon the classical liberal value of individual negative freedom, which, as discussed in Chapter 1, is necessary for autonomy. However, all actors, both individuals and associations, are bound by the harm principle in liberal theory, so there may be many justifications for placing limits and regulations on associations: ‘Associations exist within fields of power relations, and absolute claims for freedom of association can produce a society within which there are very few freedoms’ (Warren 2001a: 26). The argument in the previous section should indicate that I am not against state regulation of associations, therefore my argument against not restricting associations in membership choice is not to do with freedom of association, but to do with the freedom of speech that is essential to deliberative democracy: ‘Without access to an association that is willing and able to speak up for our views and values, we have a very limited ability to be heard by many other people or to influence the political process’ (Gutmann 1998: 3). As Rosenblum suggests, ‘voluntary association typically precedes expression’, so interfering with membership will change the preferences of that association (Rosenblum 1998). Therefore, just as associations should be allowed to select their own ‘message’, they should also be allowed to select their own membership requirements: the Catholic Church, for example, would not remain the Catholic Church if it could not exclude those with conflicting beliefs (Barry 2001: 151). Similarly, if the Sierra Club or Greenpeace had to admit people from the mining industry as members, then the voice of the association would be ‘muddied’ and this would reduce the effectiveness of their representation of their other members, restricting the autonomy of the association excessively, and making them less effective participants in public discourse (Warren 2001a: 36). Young argues that subordinate social and cultural groups require associations that exclude those with alternative identities, particularly those from dominant groups. Without separate associations, subordinate groups will struggle to form a shared and democratic identity, needs, preferences and interests (Young 1990: 167).1 Gutmann concludes that if the primary purpose of the association is expressive, religious or intimate, then it should then be exempt from legal — 186 —

Democratising Secondary Associations

restraints on membership, and therefore be allowed to be discriminatory (Gutmann 1998: 31). The central justification of this is that if such associations cannot choose their membership, and exclude when they want to, then free speech is compromised, and that is ‘tantamount to outlawing the expressive association’ (Gutmann 1998: 11–12). I would, however, extend this to include all associations, not ‘just primarily’ expressive ones. The problem with this, though, is who is to decide if an association is primarily expressive? Even if expression is not an association’s primary focus, that does not mean that it is a role that it does not wish to play on certain occasions, and restricting freedom to choose membership requirements will alter expression. However, Gutmann also warns against groups like the Ku-Klux Klan, whose discourses are of ‘hatred, degradation, and denigration of fellow citizens and fellow human beings’. Consequently, she asks whether a liberal democratic government should distinguish between progressive and regressive associational discourses (Gutmann 1998: 3). In Chapter 2, it was argued that such censorship goes against the core of deliberative democracy, as it would take legitimate issues off the agenda, and one of the central justifications of deliberative democracy was that it can induce reflection and transform preferences, and that those with little justification could not be publicly defended. Input filtering and agenda restriction was not, therefore, seen as necessary in a deliberative democracy. This same argument can be applied equally well to secondary associations. As was discussed in the Chapters 3 and 4, the informal public sphere, in which the associations are located, is a contestation of discourses; therefore, all discourses, no matter how exclusive or abominable to some, should be included (Dryzek 2000: 75–6). It should not just be progressive discourses that are included, as who would be the judge of which are the progressive discourses? Whoever had this power could potentially control the public sphere by excluding ideas that they did not like or that challenged their power. If we want new and distinctive ideas to be incorporated into the public sphere, then we cannot allow any ideas to be formally excluded. This was the justification, given above, for allowing undemocratic associations to participate in the informal public sphere. If we want progress to continue, then all ideas must be considered, new ones sought. New ideas that challenge powerful groups, that threaten the status quo, will be excluded along with exclusive discourses if we allow the formal exclusion of any discourse. Mansbridge concurs, arguing for the necessity of ‘organisational and deliberative enclaves’, regardless of content, ‘where oppositional thought can grow’ (Mansbridge 1996: 59). It is an infringement of autonomy to say to people, ‘you cannot hold and express this view.’ This is the great strength of deliberative democracy it launders preferences through its — 187 —

Towards a Deliberative and Associational Democracy

process, not prior to it, and therefore does not infringe autonomy. Therefore, I maintain that associations must have the autonomy to decide their own discourses free from interference from the state and, impose their own membership criteria as well, in order to be consistent with the norms of deliberative democracy. The Iron Law of Oligarchy Achieving internal democracy within secondary associations requires accountability, which is very difficult to achieve because of the complicated nature of the relationships within associations, especially in comparison to state agencies and the market. State agencies have clear guidelines and laws, and are organised hierarchically, internally and externally. Market organisations are accountable to their shareholders and are regulated by the market. Secondary associations, in contrast, have variable forms of regulation in order to achieve accountability and are accountable to a number of different agents, for example, governors, members, staff, government and taxpayers (Rochester, cited in Powell and Guerin 1997: 163; Leat 1996). If associations are to become more internally democratic, then it is essential that these lines of accountability are clearly defined. Associations must be made accountable to their members, which should include staff, and all members should have a chance to participate in the decision-making structure. The issue, though, is whether such accountability is possible to establish. Michels’ (1959) study of the German Social Democratic Party (SDP) suggested that there is an ‘iron law of oligarchy’ that made popular control of representatives impossible in all organisations. The SDP had sought an internal democratic structure but, according to Michels’ study, due to their longevity of position, the leaders dominated and were not held accountable to the party members. Michels not only extended his conclusions beyond the SDP to include all political parties, but thought this pattern was inevitable for all organisations, hence the ‘iron law of oligarchy’, which meant that democracy was impossible because representatives would always dominate the represented: ‘Who says organisation, says oligarchy’ (Michels 1959). Michels’ argument is based upon the assumption that democracy is impossible without organisation and representation. In order to provide people with a location for participation, organisations were essential, and some level of representation was necessary to provide economy of time and scale, as was argued in Chapters 3 and 4. However, for Michels representation causes a divide between the representatives and the represented which makes the represented subordinate. Through being leaders, leaders gained — 188 —

Democratising Secondary Associations

specialised knowledge, and this, combined with what Michels perceived was the incompetence of the masses, led to oligarchy in all organisations. ‘The masses’ – or, in our brand of organisation, the associational members – are passive in the knowledge they receive from the leadership, allowing for the inevitable domination and manipulation by the leaders, who become a professional elite with much more detailed knowledge than the ordinary member. In addition, in secondary associations, in order for the association to be successfully managed and to compete with other associations, much strategic and financial control is centralised to the association’s elites. This means that the interests of these elites dominate the actions of the association, and that they are incentivised to maintain this dominant status (Cerny 2006: 87). For Michels, this is a universal law that would occur at any time, in any culture, no matter what democratic procedures were employed in the organisation. If Michels is correct about the ‘iron law of oligarchy’ affecting all organisation and rendering democracy impossible, then this has serious consequences for associational democracy. If associations cannot be democratised, then they cannot be the units to allow citizens to participate in deliberatively democratic decision-making and public administration. Moreover, if Michels is right about the incompetence and passivity of the rank and file members of an organisation, then deliberative democracy in practice would be impossible, whether it was secondary associations or another type of organisation, or combination of organisations, that were to be employed as the location of participation. If Michels’ theory on oligarchy is correct, and it is a ‘law’, then it must be universal. However, it is suggested that his analysis is specific to the SDP of the time of his research, which means that universal generalisations cannot be drawn from it. According to Wainwright (1994), three features in particular seem specific to the SDP at this time: a collective will for a single centre of power, a specialised nature of knowledge and incompetence among the masses. Moreover, as Beetham (1977) realised, he confused the achievement of revolutionary change with the deepening of democracy, an important distinction to which we will now turn. Revolution or Democracy? Beetham argues that Michels’ argument shifted from the problem of socialists achieving revolutionary change, to the possibility of democracy – a shift that created ambiguity and undermined Michels’ findings and meant that all evidence inevitably pointed to his elitist conclusion. Beetham explains this clearly, thus: — 189 —

Towards a Deliberative and Associational Democracy

If the working class support their leaders in the pursuit of reformist policies, then, on the original revolutionary perspective, this can only be because they have been misled in their conception of their interests, and is therefore an example of elitist deviation. If, on the other hand, the leaders were to prove in advance of their followers in revolutionary zeal, then, on a more conventional understanding of democracy, this must be oligarchical also. (Beetham 1977: 17) From this confusion identified in Michels’ study, we can conclude that his ‘iron law’ may well not be an iron law at all, and certainly not one that rules out democracy within all organisations. It is more likely that it rules out revolutionary activity. A Collective Will for a Single Centre of Power Michels assumed that all political organisations would aim to develop a collective will, and aim to take over a single centre of power, such as the state. This would mean these associations becoming appropriated by the state, and co-opted. However, unlike political parties, secondary associations in an associational democracy do not aim to be, and are not required to be, the single centre of power. Secondary associations would seek access to decisions-making forums, where they would participate in a decisionmaking process with other associations. This would require devolving power to that forum, but not to any single or specific coalition of associations. The forums and quangos used to facilitate them would be fairly fluid, with a changing set of associational representatives depending on the agenda and the forum. Therefore, they will not necessarily become appropriated by the bureaucracy of the state in the manner envisaged by Michels (Wainwright 1994: 215). Moreover, as the discussion of co-option in the last chapter suggested, the associations will also remain in civil society and contribute to discourses in the informal public sphere, which can challenge the state and forum decisions. The Specialised Nature of Knowledge and the Incompetence of the Masses The iron law of oligarchy states that the interests of the leaders and members of an association will, at times, be separate. Associations may then just lead to the increased representation of the interests of the elite, and not of the deliberatively formed opinions of the members. Michels’ assumption here is that the knowledge relevant to political organisation is of a technical and specialised nature that ordinary members cannot — 190 —

Democratising Secondary Associations

understand, meaning the elites are again not subject to effective control by the mass membership (Michels 1959). However, the premise at the heart of deliberative democracy is that people are able to understand information of a technical nature, and make judgements on it, as the empirical evidence on deliberative opinion polls and citizens’ juries cited in Chapter 2 indicates (Fishkin 1991 and 1995; Stewart et al. 1994; Renn et al. 1995; Coote and Lenaghan 1997; Coote 1997; Kuper 1997; McIver 1997; Andersen and Jagger 1999; Barnes 1999; McCombs and Reynolds 1999; England 2000; Luskin et al. 2002; Hansen 2004; Andersen and Hansen 2004; Parkinson 2006). Moreover, as argued in Chapter 3, participants will improve these skills the more they participate and will become better, and better, at evaluating technical information. This is even more likely to be the case if the associational members have decision-making power within their association, as the incentive to become informed, and for others to inform oneself, increases. Therefore, the relative incompetence of the general associational member may well be due to the absence of opportunities for participation in their organisation (Hyland 1995: 256–7). Second, not all relevant information will be technical. Personal information and experiences will have significant relevance to the debates within an association. The great advantage of deliberative democracy is that it democratises the collective accumulation of knowledge and information, with all inputs being judged by each individual on the basis of their rational potency, at least to some extent. No one person, or collection of people, is thought to have a monopoly of relevant knowledge, so all opinions need to be included by deliberative relationship (Young 2000). The divide between represented and representatives will therefore be reduced. If associations attempt to have an internally democratic structure based on the norms of deliberative democracy, then, although information and knowledge will remain disaggregated, there will still be a greater pooling of this knowledge and information, and each participant will improve their levels of understanding. Empirical evidence of the Dutch Green Left supports the possibility of this. According to Wainwright, their policies are formed through a network of organisations that enable the party membership to participate and contribute to the formation of a strategy for action which is based upon the principle that party members have valuable experience and knowledge to give, and that policy-making is itself a political issue that involves interests and values, rather than a purely technical matter that is the domain of ‘neutral’ experts (Wainwright 1994: 220). The fact that elites might dominate an association excessively seems an incentive, and justification, for representatives to be bound by internal democratic mechanisms to ensure responsiveness to the membership: ‘The — 191 —

Towards a Deliberative and Associational Democracy

natural response to the problem of disjunction is to require greater use of such mechanisms of responsiveness among groups that are granted quasipublic status’ (Cohen and Rogers 1995: 71). Associational elites, in democratic associations where they are elected, will be held accountable to their members’ preferences by the forthcoming election (Hyland 1995: 253). Within secondary associations, elites are much closer to their members than in governmental and political party structures, as there are far fewer decisions made within an association than there are across local or national government. In addition, as only some associational decisions will be relevant to some members, and other decisions relevant to other members, there will be a division of labour in regulating and holding elites accountable (Hyland 1995: 262–3). From this discussion of Michels we can conclude that his ‘iron law of oligarchy’, which is invoked to rule out the possibility of democratising any organisation such as secondary associations, is far from being a law. At best we can say Michels accurately described the situation in the SDP in 1911, but the conclusions he drew cannot be taken out of this context. At worst we can say Michels’ argument is incoherent because it confuses revolutionary change with democracy. Participation in Associations Empirical studies suggest that most associations in the world do not have a democratic structure. This is true whether in the UK (Taylor 1996: 61, 66), Ireland (Powell and Guerin 1997: 167), the USA (Fukuyama 1995) or the Scandinavian democracies, where membership is quite extensive (Roßteutscher 2000: 178). This was a problem noted by associational members themselves in a recent study in the UK (Parkes et al. 2004: 315), although other research in the UK suggests that many members of associations believe their organisations are very democratic and inclusive (Taylor and Warburton 2003: 329). However, those who do participate are not representative of their populations. The lack of internally democratic associations has led Roßteutscher to conclude that an associational democracy ‘is a theoretical construct, having little in common with empirical reality’ (Roßteutscher 2000: 178). There are associations that constantly strive for a participatory, internally democratic structure, such as Sikkuy in Israel, which is an association located in the Western Galilee Misgav region and is based on the grassroots participation of local Jewish citizens (www.sikkuy.org.il/english/home.html 12/07/2007). Another example would be the Massachusetts Climate Action Network, an association dedicated to halting the threat of global — 192 —

Democratising Secondary Associations

climate change through reducing emissions of greenhouse gases in our communities and the state, which seeks active grassroots participation (www.massclimateaction.org 12/07/2007), or Brighton and Hove Community Sector Forum, which is an umbrella association that holds forums for all voluntary associations in the Brighton and Hove area (www.cvsectorforum.org.uk 12/07/2007). Further examples in the UK come from the disability movement and include associations such as the British Council of Disabled People (www.bcodp.org.uk 20/07/2007), the Long-Term Medical Conditions Alliance (www.lmca.org.uk 20/07/2007), Choices and Rights (www.choicesandrights.org.uk 20/07/2007) and the Disabled People’s Direct Action Network, which all have relatively horizontal and democratic decision-making structures (Parkinson 2006: 16). These are just a small number of examples from the many relatively democratic associations that exist, but there are a host of others, especially environmental, women’s and anti-globalisation associations, that have democratic structures, too (Catt 1999: 43–4). In the UK, several associations have an aspiration to achieve, and are attempting measures to introduce, specific deliberatively democratic practices (Roberts 1996). However, it is certainly the case that these are the exceptions, and not the norm, and while it does demonstrate the possibility of democratising associations, it might also indicate that people do not want to participate in associations. According to Warren the prevalent individualist culture provokes apathy (Warren 1996: 263). People ‘free ride’ rather than making the effort to participate themselves, which leads the associations to be even more hierarchical (Olson 1965). Recent empirical evidence in the UK, produced by the Office of the Deputy Prime Minister (ODPM), investigated levels of desire for political participation and found that most citizens did not want to participate more than they currently do (ODPM 2005: 73) (although this study did not take into account the extent to which the participation included an opportunity to have an impact on decisions). Meadowcroft’s study into participatory mechanisms employed by UK local governments indicated a similar lack of desire for more participation (Meadowcroft 2001: 40). In contrast, empirical evidence from the Power Inquiry suggests that citizen involvement can be maintained, and in fact concludes that ‘participatory approaches to decision-making are now coming of age’ (Power Inquiry 2006: 228). Jordan (2007: 58) is sceptical of methods to increase participation, though, arguing that citizens will not want to participate if they cannot affect decisions, and most participatory mechanisms still retain final decision-making power to political elites. This clearly shows that, in order to increase participation, power must be taken from these elites and given to citizens. Parkinson tells us about members of citizens’ juries and — 193 —

Towards a Deliberative and Associational Democracy

deliberative opinion polls who demanded more decision-making power (Parkinson 2006: 157). The argument, then, is that if people are given real opportunities to participate in decision-making that affects them, and know that participation can actually affect those decisions, then they will participate. Downs’ rational choice theory demonstrated that the more chance participation can affect outcomes, the more likely people are to participate, and vice versa (Downs 1957). In an associational democracy, participation can make a real difference to lives because citizens can affect their children’s schools, their workplace, and key services such as healthcare or job training (Fung 2003a). Participatory methods of decision-making make it easier for people to have their voice heard as money, status and expertise become less influential (Fung, 2003a). Decentralisation, in the manner advocated in Chapters 3 and 4, could then help increase participation, as ‘the less populous the autonomous unit, the greater will be the effectiveness of an equal share of power’ (Hyland 1995: 261). Decentralisation can then lead to more influence on decisions, which can lead to greater efficacy and, consequently, more participation. Efficacy is the belief that one’s participation would influence collective decisions; ‘the feeling that individual political action does have, or can have, an impact upon the political process, i.e., that it is worth while to perform one’s civic duties’ (Campbell et al. 1954: 187). Efficacy has two components, internal and external. The internal aspect is the self-perception of individuals as to the degree that they have the necessary skills to influence and impact decisions. The external component is the feeling that institutions are structured in a way that will allow them to have an impact. As a concept, it is not objective but subjective, as it is dependent upon whether someone perceives that their participation would have an impact (Lane 1959). Despite it being subjective, however, the resources available to an individual, and the extent and manner of institutional opportunities for participation, influence feelings of internal and external efficacy, as will ‘success’ in participation: In developing efficacy nothing succeeds like success . . . But one can also be trained otherwise: a history of discouragement and failure produces passive and fatalistic individuals; they come to lack the psychological resources to act even when circumstances permit. (Warren 2001a: 71) The lack of current levels of participation may well, then, be due to the scarcity of opportunities for participation to have a real impact on decisions. The problem with participating in deliberatively democratic decisionmaking is that preferences are expected to change, at least to some extent, — 194 —

Democratising Secondary Associations

through debate. Consequently, it is hard to interpret whether an individual has been ‘successful’, especially if success is perceived as being able to recognise one’s original preferences in the final decision. However, Warren (2001a: 71) points out that new social movements have bred feelings of efficacy in terms of ‘consciousness raising’, and it is perhaps in this manner that deliberative democracy might enhance efficacy. Through their audit of citizen participation in local decision-making processes in the UK, Lowndes et al. (2006) provide a number of elements that need to be in place for citizens to participate, summed up by the C.L.E.A.R acronym. The first requirement is that citizens ‘can’ participate, which requires the presence of necessary resources and knowledge. As suggested in Chapter 3, socio-economic inequalities comprehensively determine the presence of these skills, and it was suggested that resources must be distributed more broadly to amend this. However, once participation in associations has commenced, these skills are more likely to be developed. The second criteria is that citizens must ‘like to’ participate, which is more likely if they feel attached to their community. Again, this can be generated through participation in associations, and the resulting development of social capital and civic identity. Next, citizens must be ‘enabled to’ participate, and this is facilitated through secondary associations, especially if they have the ability to affect decision-making through powers being devolved to the associations directly, or through their representatives participating in the decision-making forums. Participation is further enhanced if citizens are ‘asked to’ participate. If secondary associations are required to meet standards of internal democracy, then the members will be asked to participate. Finally, if participation is to be sustained, decisionmaking must ‘respond to’ the participants, which means the services delivered to associational members must be determined by the recipients and, further, that associational representatives participating in the informal and formal public spheres must, in the first instance of debate at least, articulate the preferences formed through debates within the association. In addition, if associational members feel their representatives have influenced decisions in the forums, then participation is more likely to continue within the associations themselves. However, citizens must also realise that they will not always ‘get their way’ in these decisions and, moreover, that representatives will transform their preferences and seek compromises with the forums (Lowndes et al. 2006). The associational model therefore seems to include the key criteria necessary to maximise citizen participation. The issue still remains, though, whether this participation is likely to be deliberative. For Mutz, deliberation and participation undermine each other, as citizen participation is possible, — 195 —

Towards a Deliberative and Associational Democracy

but deliberative participation is not, and deliberation is possible but not through citizen participation. Therefore, the conclusion drawn from her own research into three representative surveys of Americans and their political networks is that ‘there are fundamental incompatibilities between theories of participatory democracy and theories of deliberative democracy’ (Mutz 2006: 2). However, this is because deliberative participation is more likely to occur in arenas where people are ‘surrounded by those who agree with them’, as a shared purpose can ‘promote the kind of passion and enthusiasm that are central to motivating political participation’ (Mutz 2006: 3). This, then, does suggest that deliberative participation in secondary associations may well be possible. Other empirical evidence suggests that participation in public deliberation is more likely in arenas where participants share certain aspects of identity (Barnes et al. 2004: 106), such as they do in secondary associations. Nevertheless, Mutz’s research also indicates that, at present, deliberation is not occurring in secondary associations, as they are only accounting for 5 per cent of political discussions (Mutz 2006: 28). Other research also suggests that participation in a deliberative democracy might be less attractive than other forms of participation because of its public nature, with people not wanting to reveal their preferences to strangers (Conover et al. 2002). Much of this is due to the fact that many people are so desperate to maintain social harmony that they refrain from engaging in argument or making a contribution that will destroy solidarity (Mutz 2006: 84). These problems are intensified in an associational democracy because citizens will need to be members of many cross-cutting associations, and exposure to cross-cutting social influences tends to lead to lower levels of participation as people aim to maintain their social relationships, or because this cross-cutting exposure produces ambivalence, which in turn reduces activism (Mutz 2006: 123). However, other evidence, cited by Mutz herself, indicates that levels of participation are at their highest in extremely heterogeneous or extremely homogenous social contexts. The suggestion here is that in heterogeneous contexts, people participate to protect their interests, whereas in homogenous contexts people participate because civic norms are prevalent and supportive (Campbell, cited in Mutz 2006: 97). Consequently, this could mean that there is the safe type of supportive environment to create high levels of deliberative participation in a single association (Mutz 2006: 115), as when discussion does occur in secondary associations, it tends to be between similar people, for that is the nature of associations (Mutz 2006: 35). Furthermore, deliberative participation might well be sustainable in the forums, as these will be — 196 —

Democratising Secondary Associations

heterogeneous locations in which participants will have much to gain or lose due to the fact that they will result in decisions. In Chapter 3 there was a detailed discussion of the features of social complexity and the barriers these present to the institutionalisation of deliberative democracy. The same features of complexity are present within associations, and present similar barriers to achieving deliberative democracy within secondary associations, such as time, number and disparity of members. Although they may present to a lesser extent than in nationstates, these problems are increased exponentially the larger the association. There is, then, a paradox for associations: the larger the association, the more power it might enjoy in the informal public sphere, due to its having more people and resources; however, the larger the association, the less power an individual member can have on the decision-making of that association, producing a diminished feeling of efficacy within it.2 Furthermore, the larger the association, the less opportunity people will have to contribute to collective deliberations (providing the meeting length does not increase or meetings do not increase in number in proportion to membership size, which affects time required to participate, and in turn affects participation levels). There is, then, a trade-off in size between the overall power of the association and the ability of the association to be democratised. One of the inevitabilities of a very large association is that it will have to develop structures of representation, especially if the association’s membership is geographically dispersed, either with representatives from each region/locality gathering for meetings after being informed by their local membership or through splitting into autonomous committees with distinct areas of control. Geographically dispersed groups are, then, less compatible with deliberatively democratic decision-making. If the association is geographically dispersed, this can lead to ever greater inequality, as decision-making power is concentrated into the hands of a few members who will have authority to make more decisions, and have greater influence over other members due to their ‘central location in the communication network’ (Gastill 1993: 131). Much relies on having democratic procedures, such as elections, to constrain the power of the representatives and hold the associational elites accountable to the members, but the ability of the members to do this is hindered by the fact that the association’s elites are geographically removed and due to the reduced opportunity for ordinary members to engage in face-to face collective deliberation, except perhaps at an annual meeting. If groups are to be geographically dispersed, there need to be enough members locally for people to meet and discuss the issues with, and then they can elect a member to represent the results of these deliberations at another level within the association. In addition, — 197 —

Towards a Deliberative and Associational Democracy

continuous developments in communication technology has made communication across geographically dispersed locations much easier. Therefore, deliberative democracy can still take place through conference calls or virtual forums. Such mechanisms make participation easier, could even increase participation levels and are facilities that all associations could make use of to greater or lesser extents. The face-to-face element of deliberative democracy is lost, though, which does have downsides. Deliberators tend to act more morally when communicating in person (Kagan, cited in Goodin 1992: 130), which may well dissuade smaller, and local, associations from being overly reliant on such mechanisms. Furthermore, the structure of the devolved system which I have outlined for associational democracy will discourage the development of a completely dispersed membership. If decision-making authority is to be devolved to local and regional forums, then there will not be much advantage to a dispersed membership. Perhaps it is a fault of the system, then, that it will discourage such associations, as it will reduce the autonomy of those who would wish to form such an association. However, many issues will require functional forums, which will not discriminate against geographically dispersed associations. Most associations, though, are unlikely to want to only participate in purely functional forums, meaning that they are likely to adopt a coalition structure with fairly autonomous sub-units operating at a local level. This will mean that the principle-agent bond between representatives and associational members will be weaker in the purely functional forums, as there will be less opportunity to form the deliberative relationship that is essential to hold associational elites accountable. Consequently, it seems inevitable that larger, geographically dispersed associations will have to combine top-down and bottom-up styles of internal democratisation, as does the USA’s Christian Coalition (www.cc.org 23/07/07), for example. As was suggested with the forums, within the association it is possible to combine deliberative plenary sessions with sub-divided deliberative forums, and there is no need for these sub-groups to have ‘rigidly defined boundaries’, as long as they have revolving membership. This further ensures that all get to hear the views of all, and get to express their views to all (Thompson and Hoggert 2001: 358). The problem of scale within associations can therefore be accommodated through such innovative procedures. Once again, there is a trade-off between the ideal of deliberative democracy, in which all participants are involved in the same debate, and the practical necessities of real-life decision-making. If levels of participation are to be maintained in an associative democracy, then participatory demands must not be too excessive, in both the number and the duration of meetings. If meetings are too frequent, or too long, then — 198 —

Democratising Secondary Associations

this will put some people off participating regularly, or even at all. The more time required for participation, the less equal participation is and, as Blaug notes, the motivation of participants is democracy’s most ‘significant resource’. It is important, therefore, not to make too excessive demands upon participants, as this could result in ‘reducing the energy available to the group’ (Blaug 1999: 145). The length of meetings will also affect equality within the meeting: ‘since members do not grow weary at the same rate, their participation levels also begin to diverge’ (Gastill 1993: 104). People who are tired lose interest and participate less, which will effect the collective deliberations. Political participation is a scarce good because it requires time, and participation in a deliberative democracy is more costly in comparison to other forms of participation, such as voting. This further affects equality, as often those from subordinate groups have less time to participate. Participation will, then, always have to trade-off with other goods (Warren 2001a: 126). This is why a deliberative and associational democracy can only expect, and therefore only requires, associations to be minimally democratic. This includes electing the various representatives for the various forums, after a debate on the potential representatives; participating in debate periodically (annually or bi-annually, perhaps) to decide the overall aims and means of the association; and participating in debates over what ideas, preferences, beliefs and interests should be articulated in the forums, as and when forums relevant to the association are set up. Associations that are directly devolved powers to deliver services for their membership would need similar meetings to decide what needs should be met and how best to meet them. This might sound like a lot of timeconsuming participation, especially as people will be members of several associations, and it is certainly more demanding than the current liberal democratic system requires, and more than most citizens currently do participate. However, not all associational members will be interested in every issue that makes it to a forum that is relevant to their association or to every aspect of the service an association might be delivering. Therefore, I maintain that such democratisation would not be excessively demanding or time-consuming. Moreover, this is simply a suggestion as to the level of participation that might be introduced. These are issues that will be unique to each individual association and must be addressed by the association. No blueprint can be applied to such problems, as different measures will suit different associations. What this discussion of empirical restrictions should demonstrate, though, is that there are significant obstacles to be overcome if associations are to be internally democratic, but these in themselves do not demonstrate the impossibility of this democratisation. They do highlight the fact that — 199 —

Towards a Deliberative and Associational Democracy

not all associations will be able to democratise their internal structure to the same degree, but all should be able to meet minimal standards of internal deliberative democracy. Overcoming Inequalities in Participation Throughout this book we have been reminded of the threat that socioeconomic inequalities pose to democratic deliberation, to participation in associations, to the potential of associations to effectively fulfil their potential functions, and therefore to the equal enhancement of autonomy for all citizens. Present political, social and economic relationships are not conducive to significant and equal citizen participation (Young 1990; Smith 2001: 90), and economic inequality in many liberal democracies, such as Britain and the USA, is increasing. In Chapter 3 we heard how inequalities in power and money are perpetuated in associational membership (Schattsneider 1975; Verba et al. 1995: Chapter 12; Salamon and Anheier 1996; Van Deth 1997: 9; Skocpol 1999: 66–73), with barriers to participation including the lack of income and education, and the presence of discrimination preventing equal participation in the associational system. Therefore, the more egalitarian the distribution of resources, the more associational participation we can expect (Verba et al.: 1995). Similarly, Putnam has argued that the lower the socio-economic inequalities, the greater the chance the association will be democratic and foster horizontal and voluntary relations (Putnam 1993: 147; see also Warren 2001a: 36). If participation in all associations that will contribute to the development of participation skills, or allow for influence in the public sphere and therefore in some way help atone for the socio-economic inequalities, is unequally distributed, then the consequences of an associational democracy would be the reinforcement of elite rule, so democracy would not be deepened and autonomy not cultivated throughout society. State methods to redistribute resources more equally, and publicly fund associations, were therefore seen as essential, and outlined in Chapter 3. Macpherson recognised that there was a class disparity to political participation, but that this was perpetuated in a spiral of cause and effect – that is, the fact that the lower strata participate less means that they are less able to organise, form demands and then articulate and effectuate them. This in turn leads to the domination of decision-making by higher-strata groups who do not protect the interests of the lower strata, which reduces their opportunities to participate. Due to the fact of this cycle of low participation and socio-economic inequality, Macpherson argues that ‘a more equitable and humane society requires a more participatory political — 200 —

Democratising Secondary Associations

system’ (Macpherson 1977: 94). Macpherson’s analysis here seems accurate; however, it is necessary to extend his conclusions about low-strata classes to all subordinate groups (Young 1990). We still remain in the dark as to which should come first: above, it seems clear that more opportunities for participation would instigate greater socio-economic inequalities; but at the same time, as long as socio-economic inequalities are present, the current levels of low participation, particularly in the most marginalised socio-economic groups, will be perpetuated. Macpherson acknowledges that we are caught in a vicious circle because ‘it is unlikely that either of these prerequisite changes could be effected without a great deal more democratic participation than there is now’ (Macpherson 1977: 100). If equal democratic participation in institutions, including associations, requires greater socio-economic equality, and greater socio-economic equality is only likely to be achieved through increased democratic participation of subordinate groups, then we are left with a ‘chicken and egg’ question: ‘Which is to come first?’ Gutmann (1980) argues that there should be no further increase in institutionalised opportunities for democratic participation until greater distributive justice is achieved. This is due to the fact that those with the greater resources – that is, the dominant groups – will be able to promote their interests at the expense of those from subordinate groups (Gutmann 1980: 191–7). However, I side with Young, who claims that waiting for distributive fairness before aiming for greater participative opportunities would not only postpone deepening democracy to an ‘indefinite utopian future’, but also make this achievement exceptionally unlikely, as ‘weakening relations of domination so that persons have greater institutionalised opportunity to participate in discussion about and the making of decisions that affect them itself is a condition for achieving greater distributive fairness’ (Young 1990: 94). Without changing the parameters for distributive decisions that have been stabilised in welfare capitalist states for some time, significantly greater socio-economic equality cannot be achieved. The parameters will be changed by including new participants (Schattsneider 1975) in distributive decision-making processes, which will allow for greater socio-economic equality to enable equal participation amongst all citizens: ‘Economic equalisation and democratisation . . . foster one another and should occur together to promote social justice’ (Young 1990: 94). Macpherson’s analysis of J. S. Mill and Marx seems to lend support to Young’s argument. He showed that they both thought an increase in equality would be reciprocated by an increase in participation, and vice versa. So we should not expect either of the changes to be completed before an increase in the other will begin. Consequently, Macpherson suggests — 201 —

Towards a Deliberative and Associational Democracy

that we can find loopholes in the circle where there are either cumulative increases in equality or in democratic participation (Macpherson 1977: 101). It seems, then, that an increase in participation could lead to a decrease in inequality, which could lead to a further increase in participation. If inequalities need to be removed prior to the commencement of deliberation, then Jordan asks whether we would need deliberative democracy at all after this (Jordan 2007: 67). However, even with reduced inequalities, conflicts and disputes would still exist and there would still be a need to make collective decisions. As argued in Chapter 2, deliberative democracy would still be the decision-making method that is most compatible with the cultivation of autonomy for all. Greater socio-economic equality would improve the functioning not just of deliberative democracy, but any form of democracy. We should not expect to achieve complete equality, but neither does legitimacy nor effective deliberation require it (Mansbridge 1996: 54; Fung and Wright 2001: 24). No democracy has, or ever will, achieve complete equality, so this cannot be a requirement for justification (Mansbridge 1996: 54) and must instead remain a guiding ideal. It is a ‘rough approximation of political equality’ that we are aiming for (Mansbridge 1996: 55), sufficient to ensure that deliberators are on a ‘par sufficient for deliberative cooperation to be attractive’ (Fung and Wright 2001: 24). The degree of equality that Rousseau prescribed, where ‘no citizen shall ever be wealthy enough to buy another, and none poor enough to be forced to sell himself’ (Rousseau 1968: Book II, Chapter 11), is perhaps what we should aim for. However, ‘most policy outcomes in today’s democracies do not derive from procedures that even approach that standard of fairness’ (Mansbridge 1996: 55). As suggested above, an increase in participation in associations, within a deliberative and associational democracy, could be the necessary first step. But is their any evidence to suggest that this increase will ever arrive? Evidence of Conducive Dispositions I would agree with Macpherson that reaching a participatory democracy is the key problem, and this is the case for deliberative and associational democracy as well: The main problem about participatory democracy is not how to run it, but how to reach it. For it seems likely that if we can reach it, or reach any substantial instalment of it, our way along the road to reaching it will have made us capable of running it, or at least less incapable than we are now. (Macpherson 1977: 98) — 202 —

Democratising Secondary Associations

The dualistic model outlined here certainly could not operate with present levels of participation, but ‘such a system could not have been reached except by a people who had thrown off their apathy’ (Macpherson 1977: 111). For the norms of deliberative democracy to become established within associations, there must be ‘supportive beliefs’ and ‘deliberative obligations’ present. Even if these supportive beliefs are generated by the implementation of deliberatively democratic rules and procedures, and civic virtue generated through participation in associations, ‘they are likely to emerge only after a time-consuming process in which people ‘‘get used to’’ and begin to feel ‘‘at home’’ in the new institutional framework’ (Offe 1995: 117). The big test, then, is whether the associations can endure the transitional period, while the complementary attitudes and behaviour become the dominant norm. Unfortunately, aspects of liberalism, such as possessive individualism and market competition, have had such an encompassing effect in Western capitalist societies that such patterns and norms of behaviour will be very difficult to replace, preventing the required change in dispositions that deliberative associations will need to cement. This barrier to institutional change is enhanced by the inevitable fact that institutional change in liberal democracies occurs in a situation where the institutional framework and practices that we seek to replace are dominant. As Femia appreciates, if a transition to a deliberative democracy is to occur, then ‘it is incumbent upon deliberative democrats to identify social trends or factors what would favour such a transformation’ (Femia 1996: 390). Before any steps towards a deliberative and associational democracy can be made, then, there must be ‘active popular support’, which means the promise of an associational democracy must ‘connect with the deeper aspirations to democratic order’ (Cohen and Rogers 1995: 90). However, Cohen and Rogers do not explain what the deeper aspirations are likely to be (Offe 1995: 125). I have argued here that these ‘deeper aspirations’ are the desire to be as autonomous as possible in a collective, but another necessary change is that people need to appreciate that in order to cultivate their autonomy further, they must develop and exert their own capacities, and that this can only be achieved through association with other citizens, and through a change to the present institutionalised decision-making framework (Macpherson 1977: 99). Decline in political participation throughout liberal democracies is already starting to raise serious questions over the legitimacy of present institutions, and there is evidence of a growing apathy towards, and dissatisfaction with, political institutions (Galbraith 1993; Stoker 2006). One hypothesis is that people are becoming disenfranchised from the system, as they come to appreciate that the institutions of liberal democracy — 203 —

Towards a Deliberative and Associational Democracy

mean that their participation makes no difference to the decisions that will be made. If this is the case, then the next process that needs to begin is for people to actively seek, and campaign for, new opportunities and methods of participation. Inglehart, in his important work, presents the case that a ‘post-material ethos’ is growing in developed liberal democracies, generation by generation. Key aspects of this post-material ethos include the centrality of selfgovernment, expectations of ‘competent government’ and a growing disposition to question authority. According to Inglehart, the principal reason why this post-material ethos is growing, and becomimg more prevalent, is that material needs are becoming increasingly satisfied, and consequently preferences are shifting towards public and social goods (Inglehart 1990). Weale has suggested that the rise of deliberative democracy has occurred due to this post-material ethos (Weale 2000: 1) and Warren cites empirical evidence from associational studies in the USA that support Inglehart’s prognosis. Studies by Ladd (1999) and Bennett (1998) both suggest membership in and numbers of informal associations, recreational associations, lifestyle associations and functional associations – that is, those related to care, school, work and neighbourhood – is on the increase. Berry‘s (1999: Chapter 3) evidence shows associations that form new social movements, and represent subordinate groups, are on the increase in the USA, and that they are becoming increasingly successful in influencing the agenda. The study by Verba et al. indicates that decline in voter turn-out does not coincide with a decline in associational membership, and reports increases in participation in community associations (Verba et al. 1995: 68–91). However, studies by Putnam (2000) and Skocpol (1999) suggest that more traditional forms of representative association are in decline. The difference between these empirical studies may be due to the likes of Putnam and Skocpol ‘equating civil society with traditional forms of voluntary association’ (Cohen 1999: 212). In contrast, present participation takes place in a myriad of small-scale, face-to-face groups (Cohen 1999: 226, 241). Nevertheless, the evidence that conducive dispositions for a more participatory and deliberative system are growing are indeed limited, and not sufficient to convince any sceptic. Much more research into this area is therefore required. Regulation of Associations If secondary associations are to be required to have a minimally democratic structure in order to be devolved powers to distribute services and before they can participate in the forums, then their organisational and decision— 204 —

Democratising Secondary Associations

making structures will have to be regulated, and it is suggested that state regulation can have undesirable consequences for associations. The absence of regulation is in the classical liberal tradition, where the state acknowledges key social, economic and personal areas where negative freedom will be enhanced if the state ensures they are protected from interference through the enforcement of negative rights by the legal system. However, these approaches are not neutral and are based on philosophical and ideological principles enacted by the decision not to regulate. Therefore, by not interfering with the internal structure of an association, the state is not being neutral towards associations, but is pursuing a value. In fact, it is impossible for the state to take a neutral stance towards associations, just as the state cannot be neutral towards values (see Cohen 1999 for a more detailed discussion of this). State actors must therefore decide which values in society they wish to promote and, similarly, they must also decide what aspects of associations they wish to enhance in civil society. What we have, then, are a variety of relationships that can exist between the state and associations, with alternative possibilities for levels of, and types of, intervention (Warren 2001a: 222). For Mansbridge, pursuing a laissez-faire approach to associations is illegitimate because it means different interests will have vastly unequal power within both the formal and informal public spheres. This is why neo-corporatist institutional arrangements, such as the mediating forums advocated in Chapter 4, redress the imbalance of laissezfaire tactics by ensuring decision-making takes place in fair conditions that can help redress these inequalities (Mansbridge 1995: 135), and by enabling all democratic associations to have equal access to decision-making forums, which also leads to multiple informal public spheres and, in turn, enables access to macro deliberatively democratic debate for subordinate groups. If organisations want to receive public funding, they already need to meet certain criteria for internal organisation (Young 1990: 85). In fact, state regulation of civil society has exploded over the last two decades at local, national and EU level (Rhodes 1997; Laughlin and Scott 1997). Associations, in particular charities, are regulated in the UK with regards to recordkeeping, auditing, accounting, health and safety, the use of profits, marketing and advertising, funding, investments, political activities and management structures. This regulation is justified because it is seen as essential that high standards of probity are ensured in the services these associations deliver. However, this regulation has also been seen as oppressive, enabling the state to co-opt civil society. This amount of regulation can also compromise associations’ ability to deliver services as they see fit (Taylor 1996), and to become increasingly homogenous (Powell and Guerin 1997: 157). Consequently, the advantages of associations as service providers are — 205 —

Towards a Deliberative and Associational Democracy

lost, as the justification of their increased political role is that they can fulfil certain key democratic functions more effectively than the state, by offering greater flexibility and diversity of provision. However, associations can also interpret this regulation, especially as it is constituted and imposed through an array of organisational sources (Newman 2005). Regulation does not all derive from the state, though. Associations regulate each other and associational members make important contributions to the regulation of their own associations, and providing regulation to ensure internal democracy would further enable members of an association to regulate the quality of services themselves. Therefore, regulation to ensure the internal structures of associations are democratic may well in fact reduce the current, and more controlling, forms of state regulation of associations. Because associations are in closer proximity to service users than are the state agencies, they have the advantage that trust and the commitment to democratic processes can be cultivated amongst their members, meaning stringent state regulation and monitoring is not necessary (Hirst 1994: 169). Despite reducing other forms of state regulation, though, it seems inevitable that the state must ultimately be the safeguard of internal democracy in an association, even if it devolves this power to another agency. However, associational members will also help regulate this, and the more established associational democracy becomes, the more opportunities for internal regulation there will be. It must be considered, then, whether it is possible for undemocratic associations to be transformed into democratic associations through the state, and other sources of regulation. Cohen and Rogers (1995: 46–7) believe that associations are artifactual because their dispositions and structures are not based upon the natural character of citizens who enter, and form, the associations. Artifactuality of associations is an essential principle to accept if the internal and external relations of associations is to be purposefully changed and altered. Both the nature of the associations, and the nature of the members, are influenced and affected by their environment, which includes many factors, such as the structure of political institutions, the economy and prevailing cultural norms. Different associations and members will be affected differently by these features, depending upon their location in society due to cleavages of power, economics, gender, race, ethnicity, disability and so on. The structure of political institutions, and the economy, may seem quite static and established in relation to these cleavages of power, but they can be, and have been, changed through public policy. If these structures can be changed by public policy, then it is fair to say that the disposition and structure of associations can be changed, as can citizens’ attitudes and approaches to them: — 206 —

Democratising Secondary Associations

In claiming that associations are artifactual, we do not mean to suggest that they are simply political creations or that they ought to be treated as such. But it is both an empirical and normative mistake to treat the extent and forms of group organisation as a scheme of private ordering to which politics must simply adapt. In part reflecting political choice, the incidence and structure of groups and the patterns of group representation can be changed through political choice. (Cohen and Rogers 1995: 47) Warren argues that Cohen and Rogers are overstating the malleability of associations by calling them ‘artifactual’. However, in comparison with other units of political organisation, such as communities, they are, he accepts, ‘intrinsically more fluid in the sense that they can be brought into existence in response to a problem or opportunity without requiring social and psychological integration with every other social attachment’ (Warren 2001a: 45). This is indeed the case because, even though associations foster and require social bonds, an overarching community or lifestyle does not determine them, due to their voluntary nature. Associations can be constructed and disbanded, can change purpose and structure, whereas communities are not malleable in this sense. Hirst accepts the artifactuality of associations as presented by Cohen and Rogers (Hirst 1995: 102–3), but he rejects the claim that, because associations are artifactual, they can be ‘re-artifacted’ by the state, and that through public policy their roles, and distributions of power within and between them can be changed. He further argues that associations could prove resistant to state legislation partially due to the state being less powerful than Cohen and Rogers assume (Hirst 1995: 103–4). This is a view supported by Offe, who also accepts that associations are artifactual, but rejects the idea that they are contingent, that is, tractable, elastic and alterable. He cites the examples of British industrial relations, British union reform and the abolition of German industry subsidies as evidence to support this (Offe 1995: 123). Hirst goes on to argue that, even if we accept that the state can ‘re-artifact’ associations, then it must be neutral, based upon consent and acting in the common good in order to be legitimate: ‘How can state agencies acquire the competence, neutrality and legitimacy to perform this function of crafting? They must be autonomous enough to act on society and yet must possess sufficient public support that those actions can be sustained’ (Hirst 1995: 106). Furthermore, the majority opinion upon which the state would act in such circumstances ‘may be regarded as itself an artifact of the very associational structure and culture which is at default’ (Hirst 1995: 106). In short, public opinion will be affected by the deficient and unequal nature — 207 —

Towards a Deliberative and Associational Democracy

of present associations. If the state follows that opinion, the state may not be able to achieve the aim of democratising associations. And if the state acts against the present preferences of the majority, where will be the legitimacy, and how will the state be prevented from acting in its own interest? The problem is that this process of crafting an alternative throws an excessive weight either on the capacity for reform of the state or the possibility of a consensus about the virtues of reform on the part of existing parties and associations. (Hirst 1995: 108) Hirst’s view of the possibility of generating more democratic relationships between associations consequently seems more realistic than Cohen and Rogers’. He acknowledges that, due to the present inequality of power and influence between associations, reform to the associational system is made harder because the powerful associations could ‘derail’ the process if they felt reforms went against their interests. The fact that a more equal and democratic relationship between associations could lead to less power for the dominant associations makes it unlikely that they will accept it occurring (Hirst 1995: 108). Cohen and Rogers (1995) criticise the state, arguing it is too distant from the people, and too inflexible to be legitimate, but then think it possible that the state could be the main agent of reform for associations. Rosenblum warns that the benefits of associations to their members tend not to be intended and, consequently, state intervention to mould associations tends to fail to produce the desired results (Rosenblum 1998). Hirst therefore ‘advocates a process of rebuilding associations from below, by political campaigning and voluntary action in civil society’ (Hirst 1995: 111). Associations themselves would be the agents for associational reform. Such a process, over a long period, could help to alleviate socio-economic and other forms of political inequality, which would improve associations’ ability to gain reform in the future. Having the associations themselves as the main agents of change is also much more in the associational tradition than Cohen and Rogers’ statist approach. Figgis (1913), G. D. H Cole (1920b) and Laski (1925) were also against the centralising power of the state, questioning its democratic legitimacy and opportunities for participation; consequently ‘they believed that associations are most effective when they are constructed by citizens rather than by the state’ (Hirst 1995: 112; see also Schmitter 1995: 171). There are problems with leaving reform to associations themselves, though. Tamir argues that if associations are left alone by the state, then only a few will fulfil the democratic functions ascribed to them, with many — 208 —

Democratising Secondary Associations

developing and fostering undemocratic capacities (Tamir, cited in Warren 2001a: 20), although she also agrees that if there is state intervention, then the diversity of democratic effects that arise from the diversity of associations would be eliminated. This is because the diversity, freedom and spontaneity of associations are their greatest assets, and excessive state intervention takes this away. Such dilemmas have prompted Roßteutscher to conclude that associational democracy is doomed to failure: 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. (Roßteutscher 2000: 177) However, as already argued, the state cannot be neutral towards associations; there are, in fact, just different types of interference and degrees of intervention. As Lukes’ (1974) analysis of power demonstrates, a decision not to do anything and maintain the status quo is a decision itself, so state intervention at some level seems inevitable. Although we may not know the full effects of each state intervention, we know that state procedures and policies do have an affect (Warren 2001a: 20–1). Tamir does conclude that welfare state programs can help produce an active civil society, and believes that this is a liberal good. It does seem strange, then, for her to conclude that the state cannot be used to promote a democratic good (Warren 2001a: 21). The simple fact of the state requiring associations to meet internal democratic standards before they can distribute services and participate in legislative forums will itself provide a huge incentive for associations to democratise. Perhaps ‘what is needed is a skilful combination of the state approach and the societal approach towards associations’ (Bader, cited in Perczynski 2000: 169). Perczynski thinks that in the initial stage, of forming a public of associations, the state should play a greater role in creating suitable circumstances for the formation of democratic associations. However, as democratic associations start to establish themselves, the state should play a much reduced role of interference, because ‘otherwise groups might lose their natural character, which . . . is their biggest asset and the basis of their robustness’ (Perczynski 2000: 169). It seems apparent that the state can influence the nature and relationships of associations, at least to some extent. However, if this is to occur, then it must be a finely balanced process, as excessive state interference will eliminate the advantages of civil society, rather than enhance them. There are also serious problems over the legitimacy of the state undertaking such — 209 —

Towards a Deliberative and Associational Democracy

an operation, as state legitimacy is itself in question, hence the need for an associational democracy in the first place. If the state follows opinion generated currently in civil society, this may simply lead to the enhancement of present inequalities, rather than their eradication. Nevertheless, the state cannot take a neutral approach to associations and should therefore make decisions to pursue what it sees as the fairest and most democratic approach. However, the limitations of state interference mean that civil society must provide much of the impetus for change and democratisation, and social movements seem to offer the most promising source for this impetus. In many cases, it has been social movements that have created new associations and new publics, applying pressure for the democratisation of existing institutions, enhancing public discourse and providing locations for political participation (Cohen and Arato 1992: 548). Examples of social movements from the USA creating opportunities for direct deliberative participation include the Student Non-Violent Coordinating Committee (Polletta 2002), the Texas Industrial Areas Foundation (Warren 2001b) and the Pacific Institute for Community Organizing (www.piconetwork.org 27/07/07; Wood 2001). Extending Mutz’s analysis from earlier, social movements may increase participation precisely because they bring together like-minded people (Mutz 2006: 99–100). It is, then, perhaps most likely that any pressure for further institutional changes along the lines of the model of deliberatively democratic associations, as well as the source for the change in attitudes necessary for such a framework to be both demanded and then implemented, is to come from associations themselves. Conclusion This chapter has established the normative necessity for associations to have an internal democratic structure if they are to play a central role in a deliberative and associational democracy. Even where there are good opportunities for exit, associations must be democratic, as both voice and exit are necessary to ensure legitimacy and accountability for associations distributing services and for associational representatives participating in formal decision-making public spheres, although exit is sufficient for those associations wishing to participate only in informal public spheres. It was further suggested that it is legitimate to regulate associations in this manner, as state neutrality towards associations is impossible, and internal democracy will enhance autonomy, which is a more important political value in liberal democracies than is freedom of association. In making this demand, we do though take both a weak paternalist and mild perfectionist — 210 —

Democratising Secondary Associations

stance. However, if associations are to contribute effectively to deliberation, then they must have control over membership criteria, otherwise they cannot control their views, and consequently many opinions will be lost prior to the commencement of public deliberation. The chapter also tried to establish the empirical possibility of democratising associations. Michels’ ‘iron law of oligarchy’ is not a law, but was based upon the specific context of the German SDP; however, mechanisms are necessary to ensure the accountability of representatives to the membership, if oligarchy is to be avoided. Although associations are currently not very democratic, in an associational and deliberative democracy associations would meet the C.L.E.A.R. requirements that are most conducive to encouraging participation, primarily because they will enable citizens to affect legislation that affects them, providing them with a feeling of efficacy, and because associations tend to bring together similar types of people, which provides a supportive environment for participation. However, sustainable levels of participation are inevitably limited and we should not expect too much participation from associational members. Associations should therefore only be required to be minimally democratic. Empirical restrictions of size, time and disparity of membership all influence the potential of an association to democratise internally and mean that not all associations will be able to meet the same level of democratisation, without ruling out democratisation itself. Instead, innovative procedures are required to overcome these practical exigencies and these innovative procedures should be devised internally by the membership itself. The state cannot take a neutral stance towards associations, and it should therefore try to ensure associations fulfil their potential democratic functions, and limit the undemocratic consequences of associations. This is a difficult balance to reach, though, if associations are not to be completely co-opted by the state, and associations themselves must take much of the responsibility. The state also needs to try and address present socioeconomic inequalities, which affect equality of participation, but this needs to go hand in hand with an opening up of more opportunities for participation, which will secure more equality in the long-run. It is difficult to say whether citizens do want to participate more, but there is certainly a widespread, and deeply felt, dissatisfaction with current elitist and liberal democratic institutions. Although the prevalent individualistic culture is not conducive to increased participation, there might be a postmaterial ethos emerging which heightens interest in political issues. The initial impetus for change though, is likely to come from civil society itself, and particularly from new social movements.

— 211 —

Towards a Deliberative and Associational Democracy

Notes 1. I accept this argument, but Young does not extend this special privilege of marginalised groups to all associations. In contrast, she argues that gentlemen’s clubs should not be permitted because they are just supporting ‘networks of privilege’ that dominant groups already possess and could do without (Young 1990: 197). I would suggest that all associations, even all-male ones, require this same opportunity, though, as they too need to democratically form shared needs and identities in order to effectively contribute to public discourse. 2. This is in fact only true in one sense of power, that is, voting for the final decision. However, in deliberative democracy, an individual’s argument can still influence the whole membership.

— 212 —

6

Avoiding the Mischief of Factionalism

Introduction A dualist model of deliberative and associational democracy has been advocated and outlined. This involved secondary associations actively participating in the making of collective decisions. Many issues and problems with this model were addressed in the previous chapters, but a key one still remains: as with any model of democracy, the question arises, ‘how is it possible to ‘‘reconcile each individual’s free pursuit of his own objectives with the common good?’’ ’ (Manin 1987: 351). This is of particular importance to a deliberative democracy, which has been justified on its ability to promote the common good (Cohen 1989; Bohman 1996; Habermas 1996a). Institutionalising deliberative democracy through secondary associations has been seen by many as unworkable because secondary associations tend to pursue narrow interests, promoting factions rather than the common good. If the common good is sacrificed, and narrow interests dominate in collective decision-making, then it is clear that only those whose narrow interests are ensured will have their autonomy secured. Madison provided one of the most famous, and longstanding, republican conceptions of factions. He warned that if deliberative assembles are captured by factions, then instability, increased conflict, disregard of the common good, disregard for justice and rights and coercive rule of majorities would follow (Madison 1966: 16). He defines a faction as: A number of citizens, whether amounting to a majority or minority of the whole, who are united and actuated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community. (Madison 1966: 17) Such considerations have led even the most avid fans of associational democracy to suggest that ‘any comprehensive and plausible solution to the — 213 —

Towards a Deliberative and Associational Democracy

problem of faction must include efforts to insulate a politics of the common good from more particularistic aspirations of association’ (Cohen and Rogers 1995: 24). For Madison, the only way to solve the mischief of faction is to remove its causes, or control the negative effects. The causes are perceived to be liberty, secured by freedom of association, the removal of which would be worse than factional mischief itself, as this ignores the democratic functions that associations fulfil, outlined in Chapter 3, which make significant contributions to the institutionalisation of deliberative democracy and the cultivation of autonomy. Secondary associations are an inevitable, and necessary, aspect of liberal democracy and are protected by the important right of freedom of association, which is vital to the cultivation of citizens’ autonomy. Secondary associations are, then, ‘unavoidable political facts’ (Cohen and Rogers 1995: 26; see also Mansbridge 1992, Hendriks 2002). Furthermore, it is an ‘implausible assumption to think that the state can resist the demands and supplications of organised business interests in an environment densely populated by those interests’ (Cohen and Rogers 1995: 25). The other solution would be to ensure all citizens have the same preferences and interests, which Madison rightly concludes is impossible, even in the most homogeneous societies (Madison 1966: 19). This latter solution becomes even less attractive, and impossible, given the increasing plurality of preferences, interests and identities present in modern, multicultural, liberal democracies that were outlined in greater detail in Chapter 3. Therefore, for Madison the mischief of faction should, and can, only be alleviated through controlling the effects. The consequent challenge that must be addressed, if liberal democracies are to be more democratic and cultivate autonomy, is how to integrate secondary associations into the political system in a democratic format that is compatible with the norms of public deliberation. The previous three chapters offered a dualist model as a method of achieving just this. This chapter will now outline how this model can avoid the problem of faction so strongly connected to associational democracy and ‘maximize deliberative benefits’ while ‘minimiz[ing] rent seeking costs’ of secondary associations (Mansbridge 1992: 47). If this is to be achieved, then it is certain that this dualist model of deliberative and associational democracy must be starkly different to the neo-pluralist model of interest group competition. In contrast to Madison, for neo-pluralists, factions are ‘a structural source of stability and the central expression of democracy’ because they ensure the fair competition of interests (Held 1996: 201). However, neo-pluralists reject the idea that a common good exists (Fraser 1992: 141). For these pluralists, it is the diversity of interests in society that protects a polity from the ‘tyranny — 214 —

Avoiding the Mischief of Factionalism

of the majority’ precisely because it would be separated into competing factions. In the post-war era, the dominance of these pluralist ideas led public administrators to assume that there was no longer a unified common good that could be successfully identified and realised. It was thought that the public interest should therefore be decided through a continuous competition for power among various groups and associations (Reich 1988: 128). The role of government was then to reconcile and minister these competing, and often conflicting, demands, with the public interest being achieved if, and only if, a solution was devised that would accommodate the majority of the competing groups, or at least the most powerful ones (Reich 1988: 135). The problems of faction are intensified further, in pluralist politics, by the fact that certain groups are over-represented in the bargaining process due to inequalities in opportunities to organise and gain access to the relevant bargaining arenas. As has been noted throughout the discussions on secondary associations in this book, the pressure group system is vastly unequal in terms of membership and resources. Consequently, the system currently has many undemocratic aspects, and does not equally represent the diverse interests of a whole society. Those associations that represent the interests of business lobbies, unions and professional groups, situated closely to economic media, have large financial resources to support their political aims through buying access to the mass media, funding political parties, lobbying and funding legal action. Furthermore, associations representing businesses have the opportunity to threaten capital flight and business relocation (Warren 2001a: 183), a threat with ever increasing credibility in the global market (Dryzek 2000: 143). The neo-pluralist interest group system ‘depoliticises public life’ because all interests are allowed to compete, whether they are selfish or normative in nature, therefore ‘one does not win by persuading a public that one’s claim is just’ (Young 1990: 72–3). The neo-pluralist model also means that associations espousing ideological beliefs, representing identities or advocacy associations promoting other-regarding claims are excluded, as there is no apparent way that their claims could be judged, reconciled and compensated (Reich 1988: 131). Furthermore, interest groups’ lack of internal democracy suppresses internal conflict, making a false appearance of unanimity and representing positions that have not been endorsed by the whole membership. Moreover, the neo-pluralist processes of interest group bargaining and competition are hidden from public scrutiny, as they are conducted in private and prevent active participation from citizens (Young 1990: 72–3). In addition to distinguishing the deliberative and associational from the — 215 —

Towards a Deliberative and Associational Democracy

neo-pluralist model, it is further essential for this chapter to address republican criticisms of associational democracy. In general, republicanism is strongly associated with the belief that public policy should aim at advancing the common good by seeking to safeguard policy processes from private interests and increase the power of public authorities that are in a position to consider, and act upon, the common good, ultimately enabling collective decisions to ‘transcend the mere sum of individual preferences’ (Fraser 1992: 129–130; see also Schattsneider 1975: 23; Offe 1984: 173; Mansbridge 1996: 49). Republicans are therefore against the assertion of private, narrow and selfish interests associated with factions and interest groups, as they are perceived to detract from the common good. Such considerations led Arendt to claim that citizens who participate in associations based upon private interests are ‘blackmailers’ and not ‘citizens’ (Arendt, cited in Mansbridge 1996: 49). In sum, for most republicans secondary associations threaten opportunities for democratic deliberation, rather than contributing towards it. However, the republican conception of the common good is contested here. Specifically, this chapter will distinguish the deliberative and associational model of democracy from neo-pluralism and republicanism through addressing many key disputes that arise from the differing features and concepts of neo-pluralism, republicanism, deliberative democracy and associational democracy in relation to the mischief of factionalism and the common good. These include highlighting the advantages of public, over private, decision-making processes; asserting the necessity of including private interests in deliberations on the common good; a discussion of how a universal will, thought necessary to realise the common good, excludes minorities; the proposal that requiring agreement on the common good can suppress conflict; the contention that a collective cultural or national identity cannot achieve the reciprocal trust necessary to ground deliberative obligations, but that associational affiliations can; and an argument for the superiority of decentralisation over centralisation in alleviating the mischief of factionalism. Public versus Private Processes The benefits to autonomy of collective and public deliberation were discussed at length in Chapter 2. However, further consideration with regard to the common good is required here, as it is the predominance of public deliberation in civil society that provides the clearest, and most important, distinction between the associational model advocated here, the neo-pluralist model and Rousseau’s republicanism. — 216 —

Avoiding the Mischief of Factionalism

The neo-pluralist model operates precisely to forestall the emergence of public discussion and decision-making. Each interest group promotes its own specific interest as thoroughly and forcefully as it can, and need not consider the other interests competing in the political marketplace except strategically, as potential allies or adversaries in its own pursuit. (Young 1990: 190). Despite the fact that many associations do represent narrow conceptions of interest (Schattsneider 1975: 26), this neo-pluralist analysis of secondary associations as purely self-interest promoters seems mistaken, as interest groups try and justify their own ‘special interests’ through public reasons when participating in public spheres (Schattsneider 1975: 26; Habermas 1996a; Femia 1996: 381). Even the dominant and vested associations, when encouraged or forced to go public, make similar claims to the common good and justify the status quo upon principles such as justice, freedom, autonomy, merit and efficiency. This is not to say their interests actually resemble common goods, though, as ‘an association’s appeals to commonality may mask conditions and interests that are not, in fact, common’ (Warren 2001a: 176; Schattsneider 1975: 26–7). For example, American tobacco companies tried to keep health issues relating to smoking off the agenda, but when these issues reached the public agenda they could not argue their profits were more important than people’s health because that is not a publicly convincing reason. Consequently, they justified their actions by appealing to principles of the common good, such as freedom of choice, freedom from government intervention in the market and distributive justice in terms of the taxation levied from tobacco sales (Warren 2001a: 177). The key, then, is for the public themselves to decide whether the reasons are convincing enough and whether the preferences would promote common or private goods. An associational democracy need not necessarily lead to associations’ being narrowly focused and factionalised, if their interaction is regulated by the norms of deliberative democracy. The forums, and surrounding multiple public spheres, will bring together representatives from associations with common concerns, and the norms of deliberative democracy will ensure that they view and treat each other as ‘equal partners in addressing those shared concerns . . . precisely because discussion in these arenas requires fashioning arguments acceptable to those others’ (Cohen 1997: 430). Such a situation is far removed from the neo-pluralist interest group system because, rather than encouraging associations to pressure the state with private and selfish claims, it demands that the associations formulate proposals and offers reasons, which the other associations could accept — 217 —

Towards a Deliberative and Associational Democracy

in macro and micro deliberative arenas (Hunold 2001: p. 155). This would ‘plausibly drive argument and proposed action in directions that respect and advance more general interests’. This, in turn, would mean that associational democracy did not lead to the mischief of faction and increased bargaining, but to the possibility of the advancement of the common good (Cohen 1997: 430). For Rousseau, the common good can be realised without public deliberation, as it will be ‘evident, simple and luminous’, and citizens can thus identify it in private. In fact, though, ‘long debate, dissension, and tumult betoken the ascendance of private interests and the decline of the state’ (Rousseau 1968). There seems, then, to be a contradiction in Rousseau’s analysis here, as he argues that the common good is only attainable if democracy is insulated from private interests, but that this can only be achieved if people contemplate the common good in private. The suggestion is that minority groups only need to fear the tyranny of the majority when the factions are motivated by personal interests (as Madison assumed all would be), as they cannot be tyrannised by a majority motivated by the common good. Manin criticises Rousseau, however, arguing that ‘a legitimate decision does not represent the will of all, but is one that results from the deliberation of all. It is the process by which everyone’s will is formed that confers its legitimacy on the outcome, rather than the sum of already formed wills’ (Manin 1987: 352). This is consistent with the conception of autonomy outlined in Chapter 1, where the process of preference formation was seen as key to autonomy. As discussed in Chapter 2, the idea implicit in many conceptions of deliberative democracy is that public debate, in which all are included, will encourage participants to offer ‘public’ arguments, as naked self-interest will be unjustifiable to the collective. Therefore, public debate will not lead to the ascendance of private interest, but to a focus on public interests. Rousseau, with the absence of public deliberation in his model, in contrast, seems to be unable to offer a convincing account of why citizens’ private deliberations will lead them to focus on the common good (Manin 1987: 352). For example, when an issue first reaches the agenda there is no guarantee the majority will be right about what decision would reflect the common good. However, following a democratic debate, there is a much greater chance that the majority decision will now be ‘correct’ (Waldron 1993: 413). Common Good versus Particular Interests Not all republicans are against public deliberation. In fact, deliberative democracy has great relevance to republicanism, as both schools of thought — 218 —

Avoiding the Mischief of Factionalism

share a mutual commitment to participation in debate to resolve political and moral disputes aimed at advancing the common good (Warren 2002: 174; Elstub 2006a: 301). If we cast our minds back to Chapter 2, two contrasting models of deliberative democracy were outlined: the consensus and agonistic models. Much of this dispute between these two models regarded whether consensus should be the aim of deliberative democracy. There it was decided that consensus was not always possible due to plurality of views, or desirable because it could be derived from dominance rather than being rationally motivated. A similar dispute exists between these two deliberative models over the common good. The consensus model suggests the principal justification of deliberative democracy is that it ensures people will focus upon the common good, not assert private interests and thereby reduce the differences in perception on the common good. This is achieved through the sharing of information, which ensures ‘distorted’ conceptions of the common good are reduced, and through the need to offer reasons that others find acceptable (Cohen 1997: 420–1; Elstub 2006b: 25). In contrast, the agonistic model rejects the idea that consensus on the common good is the sole aim of deliberation, fearing that the ‘common good’ might not be common at all, and instead simply a perpetuation of inequality, and consequently the agnostic model favours the inclusion of partial interests into deliberative arenas (Mansbridge 1980: 32; Gambetta 1998: 21; Gould 1988: 18; Young 1996: 126; Elstub 2006b: 26): When discussion participants aim at unity, the appeal to a common good in which they are all supposed to leave behind their particular experience and interests, the perspectives of the privileged are likely to dominate the definition of that common good. The less privileged are asked to put aside the expression of their experience, which may require a different idiom, or their claims of entitlement or interest must be put aside for the sake of a common good whose definition is biased against them. (Young 1996: 126) As Mansbridge has realised, it is minorities and subordinate groups that will suffer most from this: ‘The less powerful may not find ways to discover that the prevailing sense of ‘‘we’’ does not adequately include them’ (Mansbridge, cited in Fraser 1992: 130). This occurs because dominant groups are able to present their particular interests and norms as ‘authoritative knowledge’, and therefore less likely to be disputed (Young 1997: 399). Republicanism’s ‘commitment to a unified practice’ can consequently lead to ‘cultural imperialism’, the universalisation of the dominant group’s ‘experience and culture’ achieved through the monopolisation of channels of ‘interpretation, which enables dominant social and cultural groups to — 219 —

Towards a Deliberative and Associational Democracy

establish the norms of society and correspondingly the perspectives and norms of subordinate groups become invisible’ (Young 1990: 183–4, 58–9; see also Reich 1988: 138; Urbinati 2000: 774). Subordinate and marginalised social and cultural groups therefore have to assimilate with these norms, such as the ‘language and symbols of public deliberation’ of the dominate group, or exclude themselves from public debate (Festenstein 2005: 128).1 This does not necessarily mean, however, that dominant groups are acting strategically when doing this. Christiano talks, for example, of cognitive bias, which prompts individuals to be more responsive to their own interests than to others, a factor that is intensified in modern diverse societies. This means that ‘if many advance conceptions of justice in public discussion that reflect their interests, those who lack opportunities to advance their own will lose out’ (Christiano 1997: 259). If this is the case, then in order to ensure equality and fairness in decision-making, we must allow the assertion of the interests of all, as dominant groups will have a greater advantage in disguising their interests as the common good. It then follows that the deliberative and associational model, with its aims to include an array of associations representing a diversity of interests and identities in opinion-formation and decision-making, is vital to equality in deliberation, and to approximating the agonistic model of deliberative democracy (Elstub 2006b). According to Young, the republican criticism of secondary associations is based upon the assumption that group differences necessarily mean that there are conflicts of interest; but interests are not always in conflict, however, and when they are compatible, deliberation between interest groups can lead to understanding of where interests do converge, which then encourages associations to take on board the interests of others, and the public good as a whole, in their considerations and preferences. In fact, the source of conflict is often due to the exclusion of groups in the first place. The increased representation of such groups, achieved by secondary associations, can change these ‘structured relations of oppression’ and therefore change the nature of the conflict, reducing it in many instances. If ‘the alternative to stalled decision-making is a unified public that makes decisions ostensibly embodying the general interest which systematically ignore, suppress, or conflict with the interests of particular groups, then stalled decision-making may sometimes be just’ (Young 1990: 189). Consequently, the ‘mischief on behalf of hidden injuries, is an effect that democracy could do without only in a just society with convergent interests’ (Warren 2001a: 35), which, we have highlighted, do not exist. Although the common good does not always pertain, justifiable decisions with regard to these issues still need to be made (Young 1990: 189; Warren — 220 —

Avoiding the Mischief of Factionalism

2002: 175). Without their consideration in public deliberation, private interests may not be transformed so that a consensus or compromise can be reached, and interests will thus continue to conflict (Warren 2001a: 24). Consequently, the public agenda should never be restricted or predetermined, as deliberative democracy ‘encourages discourse about the lines separating the public and the private’ (Benhabib 1996: 76; see also Fraser 1992: 129). Many issues central to public policy that were previously considered private concerns, such as domestic, environmental and economic, were previously thought to be issues for the individual, market or private ownership. Without allowing associations to challenge the current agenda, issues of public concern would forever be set in stone (Young 1990: 66). Moreover, associations provide essential ‘protected enclaves’, in which the members can deliberate upon their shared (private) interests as well as the common good: ‘Members of these groups may legitimately take particularist as well as universalist stands, as they may legitimately challenge the underlying assumptions of these forms of universalism around them’ (Mansbridge 1996: 57). Therefore, there is no need to deny the ‘partiality of affiliation, of social or group perspectives, that constitutes concrete subjects’ (Young 1990: 100). In contrast, Squires (2000) argues that synthesising group representation and democratic deliberation, as the deliberative and associational model aims to, leads to instability in decision-making. This is because group representation is necessarily partial, and therefore cannot be reconciled with deliberative democracy, which, for her, is essentially an attempt to achieve rational impartiality. However, deliberative democracy should not expect participants to be ‘impartial rationalisers’, if it is to be realistic (Johnson 1998: 174; Festenstein 2002: 90). As Elkin contends, ‘citizens are unlikely to accept that something is in the public interest unless it is connected to their private interest’ (Elkin 2004: 56–7). Consequently, participation in macro and micro deliberation is unlikely to occur without the strategic motivation associated with associations (Hendriks 2006). Empirical research suggests that associations, including interest groups, contribute extensively to public deliberation, but out of instrumental motivations rather than to enhance the common good. Such strategic motivations include improving the public image, to distribute information, to improve their understanding of public opinion, to stimulate policy reform and to avoid non-participation costs (Hendriks 2006: 70–2). Partisanship in politics is therefore inevitable – we can not dismiss and eliminate these natural human motives – but neither is partisanship necessarily bad. Instead, we must ‘think about how [it] can be harnessed in the effort to constitute a good political regime’ (Elkin 2004: 75). The model of delib— 221 —

Towards a Deliberative and Associational Democracy

erative and associational democracy attempts to do just this. It accepts partisanship and self-interest as facts of the political process, but tries to incorporate these interests into a deliberatively democratic opinion formation and decision-making process that will encourage consideration of the interests and autonomy of others. The associational members deliberate within the association, in the forums and informal public spheres, precisely because they are particularly situated. Associations are collections of people who share some situation; they are then there to assert beliefs, interests and information that are connected to the association and themselves, not to transcend or bracket their identities and specificity. Although associational representation is necessarily partial, this does not mean that group representatives cannot come to appreciate the claims and interests of others, or that these claims will not affect their preferences. In entering public deliberation, associations still need to justify their aims through public reasons and through this ‘these interests can be stretched’ in a manner that leads to consideration of the common good (Elkin 2004: 57). However, this does not mean rational impartiality in the sense that all should accept a common good independent of their own interests: 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) Eckersley makes a similar point, claiming that, rather than ‘impartiality’, deliberative democracy should aim instead at ‘enlarged thinking’, which is an ‘other-regarding orientation’ in the formation and justification of preferences (Eckersley 2000: 121). Squires argues that this is itself an appeal to impartiality (Squires 2000: 103), but I would suggest it is not an appeal to impartiality, but rather an attempt to make preferences ‘less partial’, the idea being that impartiality is in fact undesirable and impossible to achieve. However, such analysis does lend support to Festenstein’s (2002) claim that deliberative obligations will be abandoned when it is in the interests of actors to do so, which means there is a distinct danger that public spheres, both informal and formal, could descend to bargaining between narrow and fixed interests, if they are not appropriately mediated. Universal Common Will versus Minority Opinions A classic articulation of the republican view of the common good was provided by Rousseau, in his vision of a general will embodying the — 222 —

Avoiding the Mischief of Factionalism

common good (Rousseau 1968). Rousseau thought it essential for all to participate in the realisation of the common good, so that it represents the will of all, and not the private interests of all, but he also believed that private interests, such as those represented by secondary associations, should be eliminated from this process because the common good is seen as something distinct from the private interests of individuals, which detract from the development and maintenance of a common will (Manin 1987: 341–2). Associations are perceived to detract from this unanimity because they pursue private interests: When special interests begin to make themselves felt, and when smaller societies influence the larger one, the common interest changes, and finds opponents. Unanimity no longer reigns, the general will is no longer the will of all, contradictions and debates arise, and the best point of view is no longer accepted without disputes. (Rousseau, cited in Manin 1987: 346) For Rousseau, the associational democracy advocated to institutionalise deliberative democracy would, therefore, prevent this unanimity and collective will, which is essential for the realisation of the common good, forming. However, as suggested in Chapter 1 Rousseau’s thought on the common good contains a lack of recognition of minority opinions because majority opinion is interpreted as the general will. Rousseau suggests that the minority in a decision are still autonomous because they are obeying laws that are ‘what they really wanted’ and that they were simply ‘mistaken’ about what they wanted. Rousseau believes that because the sovereign is composed of every citizen, it will not have an interest contrary to that of any member. This further highlights the necessity that the members of the community have a universal shared set of interests and ethical values, or what Rousseau describes as ‘universal and compelling power’ (Rousseau 1968, Book 2, Chapter iv: 74). If universal interests and values exist, and the community will not intentionally make a decision that goes against the whole community, then decisions cannot damage any individual in that community. The idea of a general will dates to pre-modern times, before nation-states were established, when homogenous city-states would claim to have unified common interests (Martell 1992: 164–5).2 However, now that nation-states are the principal unit of political organisation (which is still the case even in the global era), and are characterised by a cacophony of social divisions and differences, the idea of a common will cannot be sustained. Social divisions are further increasing as the nation-state is eroded through globalisation, with even more plural societies and multicultural identities emerging. If a — 223 —

Towards a Deliberative and Associational Democracy

common will is established at all in contemporary societies, in reality it is the will of the majority: ‘The general will is more often than not a mythical construct, an imaginary community, which is insensitive to, and suppresses, real diversity and pluralism’ (Martell 1992: 165). As Manin argues, minority opinion is in no way incorporated into the will of the majority in Rousseau’s theory. Consequently, the autonomy of permanent minorities, or any individual who did not conform to the ethical views of the majority in the community, will not be cultivated by these decisions, but sacrificed to those of the majority. Again, this is something that is even more pronounced in contemporary, diverse and multicultural societies. Therefore, an alternative institutional design to Rousseau’s is necessary to ensure minorities can still register their opinions and arguments, even if the majority does not accept them. Manin goes on to argue that secondary associations can best fulfil this aim, and in fact suggests that this is the ultimate justification for pluralism and associations because they can represent minority and under-represented interests, ensuring that the majority must consider minority opinions, or meet with opposition and resistance which provides important ‘counterforces, checks and balances’ to majority will (Manin 1987: 360–1). The associational mediating forums, and the informal public spheres, are an attempt to ensure that minorities will have their views heard and can still continue to challenge decisions over which they disagree and represent alternative opinions to the majority. Therefore, for Manin, the goal of pluralism should be democratic deliberation, as it can help overcome some of the inequalities of the associational system. Actively seeking the views of minority or subordinate groups adds to the diversity of the debate, ensures representation of presently un(der)represented groups and ensures that groups that are in the dominant culture hear the arguments these associations provide, increasing available information and therefore enhancing the autonomy of all participants in the debate. It also forces the associations of the dominant groups to respond to these arguments and to provide reasons defending their position, as without the presence of minority opinion there is no need for the majority to offer public reasons to justify itself. Common Identity versus Group Affiliations Some deliberative democrats take too strong a republican stance, feeling the need for unity and consensus to be in existence prior to the commencement of deliberation, if the common good is to be reliably identitified. As we learned from the discussion of deliberative obligations from in Chapter 2, some form of trust that could ensure commitment to these obligations is — 224 —

Avoiding the Mischief of Factionalism

upheld is required, and Walzer (1994) is therefore right to suggest that there needs to be a ‘shared understanding’ prior to democratic discourse. For communitarians, this ‘shared understanding’ is to be produced though a common culture, which can generate reciprocal trust because it allows for a shared ‘understanding of norms, values and roles’ (Festenstein 2005: 144). However, as Festenstein argues, this is not the same as trust and can even be a source of distrust (Festenstein 2005: 145). Miller suggests that for trust to be prevalent there needs to be a shared identity that transcends all other identities. For him, this can be fulfilled only by a national identity, which generates trust and shared commitments to democratic government throughout society (Miller 2000: 158): ‘if we share a common nationality, we will possess the motivations to overcome other sorts of social difference in order to co-operate on a deliberative basis’ (Festenstein 2005: 148). However, given the diversity of identity that exist in modern multicultural societies, we cannot assume shared understandings of a national identity exist, on top of which such a suggestion further eliminates the preference transformation role of deliberation. Festenstein appreciates that this leaves us with a dilemma. If national identity is essential to generate the trust required for deliberative democracy, then challenges to the perceived shared cultural norms and values can deteriorate the trust necessary for public deliberation. This seems to occur whether we take a realist or constructivist view of national identity. If national identities are real and waiting to be discovered, or more accurately interpreted, then public deliberation in which all perspectives are included is still required to improve the ‘discovery’ process, as not all will identify with the same common values of the national identity in the same way. If national identities are socially constructed, and interpreted, with human agency playing a central role in the competing interpretations, then public deliberation in which all cultural groups are included is similarly required. If we exclude issues of the content of national identity from collective deliberation, then minority cultural groups are excluded from the prevailing identity, their autonomy is sacrificed, leading them to choose, once again, between a rock (assimilation) and a hard place (exclusion). If these issues are included on the agenda, then division increases and the national identity thought necessary for deliberative democracy is lost (Festenstein 2005: 153). Unity and consensus upon a shared national or cultural identity therefore cannot be a prerequisite for deliberative democracy. This is not to say that national identity does not have an important role, but only that this needs to be constructed through democratic deliberation itself, and so cannot ground this deliberation in the first place (Young 1996: 127). Deliberative models that assume the common identity is in place prior — 225 —

Towards a Deliberative and Associational Democracy

to the commencement of deliberation fail to meet the condition of Rawls’ reasonable pluralism, which requires that ‘no comprehensive moral or religious view provides a defining condition of membership of the foundation of the authorization to exercise political power’ (Cohen 1997: 408). The common good can therefore not be derived from shared cultural and national identities, as this will simply reflect the common good of dominant social groups, excluding many others, especially in increasingly plural, multicultural societies. I have suggested a more civic trust can be generated through participation in secondary associations, which only requires a common set of general political values through which all other conflicts should be resolved. This is less prescriptive and is open to a plurality of identities, while still providing the foundation of trust necessary for the commencement of deliberative democracy. Many secondary associations will not be interest groups but identity groups, based around factors such as religion, ethnicity, race, gender, culture and sexual preference, and the internal commonality within such associations is pre-given. For Rosenblum, this means that these associations will be unable to find shared interests with other associations, making them poor contributors to deliberation and rendering them unable to generate the reciprocal trust that public deliberation requires. The inclusion of such associations in both formal and informal public spheres would therefore increase, rather than decrease, factionalism (Rosenblum 1998: 131). It is true to say that the primary focus of such associations is to represent differences as opposed to commonalities; however, commonalities on a broader abstract level can still be established (Warren 2001a: 176). Such associations will be able to make public arguments based on principles such as justice, equality, freedom and autonomy, which will hold resonance with other associations who are also making appeals to the same principles. For example, in the UK, the Countryside Alliance has appealed for the continuing right to hunt foxes on the basis of individual and group freedom. Having made these arguments, it would then be harder for them not to publicly accept similar claims for toleration of diversity of sexual preference and religious freedom. Furthermore, associations from all these various identity cultures could form coalitions and networks of communications amongst themselves, as they are similarly situated in terms of being excluded from the dominant cultural norms. Moreover, such associations still contribute to the contestation of discourses in public spheres, and can still stimulate preference reflection, which is the core of democratic deliberation, as was highlighted in Chapters 2 and 4. In an associational democracy, citizens will join a variety of associations because people have multiple and cross-cutting identities, interests and — 226 —

Avoiding the Mischief of Factionalism

beliefs. This will alleviate factionalism because ‘people are more likely to have some basis for understanding and empathising with others in societies where they inhabit crosscutting and overlapping roles’ (Warren 2001a: 16; see also Hirst 1994: 49–55; Achterberg 1996: 169), as the discussion of civic virtue in Chapter 3 indicated. Due to this overlapping membership, and because associations create space for the participation of citizens over a greater number of functions, the participants will have ‘a growing sense of their own effectiveness’, which will be ‘the best protection against the parochialism of the groups in which they participate’ (Walzer 1994: 189–90). The deliberative and associational model of democracy could, therefore, help achieve what Martell terms ‘pluralist social negotiation’, which would allow each association, and their members, to preserve their distinct identities but still deliberate with other associations and demonstrate trust and respect for their needs and demands (Martell 1992: 171). Centralisation versus Decentralisation In contrast to Rousseau, Madison thought the best way to control the effects of faction was representative democracy, as he thought representatives would be better situated to determine the common good. He was also aware of the danger that the representative body could itself form a powerful faction, and a large state was seen as the cure for this danger because it ensured there would be a high number of representatives, a greater number of suitable citizens to be representatives and an increased chance of unsuitable representatives being identified by the citizenry. A large polity would further ensure that there were a great variety of interests, which would make it more difficult for any faction to form and mean that a majority would only be identifiable on the common good (Madison 1966). Citizens would be involved in the political process through voting in elections, and this would ensure citizens could protect their own interests without the need for the ‘violence of factions’ (Madison 1966: 19). Voting for representatives would then insulate the political process from factions, as it ‘enables the majority to defeat their sinister views by a regular vote’ (Madison 1966: 19). The associational model outlined in Chapters 3, 4 and 5 is at odds with the solutions to factionalism advocated by Madison, and by republicans following in his tradition. In fact, sympathisers of Madison would argue that such an associational model is completely open to the ‘mischief of factionalism’ that Madison identified, with the decentralisation of decisions enabling narrow, facticious interests to form and the associational mediating forums being captured by the associational representatives pursuing — 227 —

Towards a Deliberative and Associational Democracy

their narrow interests and destroying public deliberation focused on the common good. In a similar vein, Cohen would also be against the mediating forums, as they would be organised around local, sectional and issue-specific areas. For Cohen, such an institutional framework would be unable to develop sufficient opportunities for open-ended deliberation and only incorporate a narrow range of interests which would be unable to produce a ‘comprehensive conception of the common good’ (Cohen 1989: 31). However, this argument seems misplaced. First, Cohen is a strong advocate of secondary associations as a location for active participation and, as he rightly accepts, if these associations are internally organised around the norms of deliberative democracy, as I have advocated, then the sectional interests will be more coherent and autonomous (Cohen 1989: 31). Furthermore, the idea of the forums is that they will bring together representatives from associations with a diverse selection of interests, but ones that are relevant to the decision area in question. Not all people, and not all associations, will be interested in every issue, especially if decisions are to be devolved territorially and functionally. Consequently, there is no reason why the forums should only ‘bring together a narrow range of interests’ (Cohen 1989: 31). Manin rightly recognises that in modern complex societies it is impossible to have collective deliberation exploring every possible outcome (Manin 1987: 357). The contention here, though, is that decentralisation could in fact enable decision-makers to appreciate the common good, as it becomes clearer who is affected by the decisions and what the effects of a given decision would be. In Chapter 3, we learnt that Cohen himself believed that participating in associations can lead to a ‘civic consciousness’ that included a recognition and commitment to democratic procedures and norms as the basis for social co-operation and trust in the commitment of others to do the same (Cohen and Rogers 1995: 43–4). If this is the case, then some of the most detestable features of factionalism could be alleviated. For Elkin, decentralisation of decision-making to, and active participation from, citizens, enables them to see how their actions and interests affect others; it hones participants’ skills for considering particular interests in light of the common good and improves their ability to make judgements upon which assertions are in the public interest, as these aspects cannot be achieved if politics is removed from the citizens in large polities in which they are simply observers (Elkin 2004: 55–6). This echoes Rousseau, who argued that when decisions are made remotely, it becomes ‘difficult to realise the advantages [a person] might hope to draw from the continual privations good laws impose’ (Rousseau 1968, Book II, Chapter VII). If Elkin and Rousseau are right, — 228 —

Avoiding the Mischief of Factionalism

then secondary associations and mediating forums can improve the focus on the common good by offering venues for decentralisation and active participation in decision-making. With associations participating in public deliberative and legislative forums, the influence pressure groups could gain through private processes will be lessened, which would reduce the mischief of faction, even though processes of lobbying are likely to never be eliminated. Hayek argues that undue factional influences are ‘the inescapable result of a system in which government has unlimited powers to take whatever measures are required to satisfy the wishes of those on whose support it relies’ (Hayek 1979: 13–15). This is a set of circumstances that would be prevalent in Madison’s large, centralised and representative republic. However, in the dualist system envisaged here, governmental powers would be significantly reduced, as the devolved forums would become the primary legislative bodies. Despite this, there will still be plenty to gain for interest groups by accessing the government through private processes, as the government will still play a significant role, especially in relation to agenda setting. This, then, seems an inevitable consequence of upholding liberal democratic rights, as it is not possible to outlaw lobbying and private bargaining relationships with state representatives. Therefore, it is perhaps the best we can hope for, and a significant achievement, to reduce these private and unequal methods of factional influence and is an important difference between the deliberative and associational, and the neo-pluralist, models. Conclusion There are mischiefs of faction in the neo-pluralist model of associations, as it approximates a model of the market not suitable for making public policy fairly and democratically. It also embodies the norms of bargaining, which is private and based on power, and therefore goes against deliberative democracy. The neo-liberal model does not engage the associations into deliberatively democratic decision-making, therefore they do not have to provide public reasons to justify their interests and aims, nor do they necessarily hear the arguments, experiences and interests of other associations, which could impact upon how they perceive their own interests. Second, the neo-liberal model is vastly unequal and makes no attempt to readdress this inequality by equalising resources, guaranteeing representation of excluded groups and protecting the democratic process from economic media, where possible. Third, the neo-pluralist model does not seek to engage the members of each association in internal democratic debate. Consequently, in the neo-pluralist model, the interests represented — 229 —

Towards a Deliberative and Associational Democracy

have not been deliberatively formed by the members and therefore are not as autonomously formed as those in the dualist model. All these factors mean the common good would be impossible to realise if the neo-liberal model was approximated. However, the deliberative and associational model of democracy set out in this book differs markedly from the neo-pluralist system, and promotes the public sphere to the forefront of politics, politicises public life, and in turn cures the mischief of faction. This is achieved because the requirement that decision-making forums must approximate the norms of deliberative democracy ensures that representatives from the associations will have to justify their interests to the public. Second, in the dualist model the associations are to be internally democratic, again approximating the norms of deliberative democracy, so the resulting positions, although not necessarily endorsed by the whole membership, will result from democratic decision-making which does not accept the pre-political preferences of its members. Third, the forums are not to be private but public decisionmaking arenas, open and accessible to representatives from all affected associations, creating a genuinely public process that incorporates ideological, normative, advocacy and identity groups, in addition to those pursuing their own interests. Fourth, the informal public sphere will become more prominent because associations involved in making legislation will be involved in public communication here, too. This, combined with the fact that, in an associational democracy, associations become more prominent political actors, leads to the informal public sphere receiving more attention. It is suggested that these measures will transform the associational system, reducing the prevalence of narrow interests associated with secondary associations, and significantly reducing the inequalities of influence that infect the present pressure group system. Therefore, decision-making through bargaining is replaced with democratic deliberation. This means that representatives from associations must justify their claims with public arguments, if they are to be accepted, and that decisions will not be made purely on the basis of self-interest, as such decisions could not be justified. Participants are also presented with a range of new information and experiences from other associations, which will be a resource for interest transformation. It will be more likely that it will be the force of these reasons that will either be convincing, or not, to the other associations, rather than levels of membership or money. Furthermore, by making associations a key and central locus for political participation, citizens will be encouraged to be members of several associations, and become more aware of the interests of other associations. It is, then, combining the associational model with the norms of deliberative democ— 230 —

Avoiding the Mischief of Factionalism

racy that enables the mischief of factionalism, that is associated with associational democracy, to be avoided, or at least dampened. The agonistic model of deliberative democracy best addresses the problems of achieving a common good under conditions of pluralism and mulitculturalism. It allows subordinate groups to express their particular interests and therefore challenge the particular interests of dominant groups, which can be disguised as the common good. It is the process of debate and information-sharing that is key, as this can lead to more otherregarding preferences, which are in line with the common good. Moreover, the agonistic model accepts that, although there is a common good in many situations, in others there may not be, and it cannot be presumed a priori, as this is itself an area for democratic debate. Where there is no common good, and interests conflict, a decision still needs to be made, and this should still be made under the same conditions of deliberative democracy. Unlike the neo-pluralist model, the republican model believes in, and aims to promote, the common good through collective decision-making. It does not, however, offer an attractive or plausible method of forming the common good, as it requires the existence of a pre-political consensus in the form of a shared cultural or national identity, neither of which can be presumed but must be formed during the political process itself, if it is to be genuinely inclusive of all cultural groups. Therefore, by relying on shared cultural and national identities, republicanism is not sensitive to diversity and minority opinion. If it includes these opinions in debate, then the identity that grounds the trust in deliberation is lost. I, instead, suggest a civic trust generated through participation in the associations themselves could ground the trust necessary to ensure the commitment to deliberative obligations by providing a common set of general political values. Similarly, the republican conception of the ‘common good’ and ‘general will’ cannot respond to the challenges of pluralism and multiculturalism that are increasing rapidly in modern day societies, without excluding minorities, sacrificing their autonomy for the benefit of the majority and thus ensuring such minorities become established subordinate groups. A common or general will cannot rest upon universal interests or values, as these do not exist, and the presumption that they do excludes those who disagree with the majority. All values and interests must be considered in any formation of the common good, and secondary associations, with their great diversity and flexibility, are the best mechanism to achieve this. The dualist model of associational democracy would not, and should not, prevent associations with narrow interests from participating in collective decision-making, as republicans have advocated, as such an exclusion would eliminate the motivation to enter deliberation. In addition, and — 231 —

Towards a Deliberative and Associational Democracy

as the agonistic model of deliberative democracy appreciates, there cannot be impartial participants in deliberative democracy, and that a belief that there can be allows dominant groups to disguise their private interests as appeals to the common good. Furthermore, when private interests conflict, if they are left unaddressed and unresolved then the status quo will be maintained and this will usually benefit already dominant groups, thereby undermining the autonomy of those in subordinate groups. Finally, decentralisation (as opposed to centralisation) is more likely to reduce factionalism because decentralisation of decision-making to, and active participation from, citizens enables them to see how their actions and interests affect others and improves skills and judgement for considering the common good, whereas centralised decision-making methods leave citizens purely as observers. In addition, through decentralisation, governmental powers will be significantly reduced, making it more difficult for associations pursuing narrow interests to capture decision-making assemblies that determine all-important decisions. Notes 1. Certain cultural groups may decide to take the route of exclusion from public deliberation, as by entering it their identities and lifestyles are opened to examination by public deliberation. Ultimately, however, exclusion seems more costly than inclusion, even for such marginalised groups (Festenstein 2005: 130–1). 2. As Martell (1992) notes, it is unlikely these city-states were as homogenous as they thought, and that the ‘general will’ was in fact formulated upon the exclusion of large sectors of society.

— 232 —

Conclusion

The book has presented a normative argument for deliberative and associational democracy based upon the justification that such a democracy can cultivate the autonomy of all citizens relatively equally. To get to this conclusion, it was argued that autonomy is the normative core of democracy, that deliberative democracy is the decision-making model most likely to cultivate the autonomy of citizens, and that secondary associations enable deliberative democracy to overcome features of social complexity and be meaningfully institutionalised. If associations are to be mechanisms for deliberative democracy, then a dualist strategy must be employed in order to ensure that deliberation and decision-making are linked. This involves networks of communication within the informal public sphere, which can both transform preferences and set the agenda. The other requirement of the dualist model is access to the legislative arenas, which must be devolved, both territorially and functionally, consistent with the principle of subsidiarity. One possibility is mediating forums, organised by quangos, which bring together representatives from secondary associations to make decisions based upon the norms of deliberative democracy. Such a model differs considerably from neopluralism, as all relevant associations are included and must justify their preferences publicly, which encourages them to consider the interests and opinions of other associations, and in turn reduces the mischief of faction connected to secondary associations. The combination of both deliberative and associational democracy therefore brings out mutual strengths and helps overcome many weaknesses of each model that would be present if they were not combined. The book offers, then, a mild perfectionist argument for participation in democratic associations, for although not everyone will want to be autonomous, even in liberal democracies, in order to be able to make this decision, citizens need to be guaranteed the conditions to be autonomous and to change their mind and seek autonomy in the future. Moreover, — 233 —

Towards a Deliberative and Associational Democracy

liberalism is a fighting creed, not a neutral framework devoid of values and principles. As autonomy is conceived procedurally, the argument here is still compatible with the huge diversity of views and identities that are present in contemporary diverse societies. Whether such levels of citizen participation can be sustained is a significant problem; however, by combining participation and representation, in sequential deliberation, high levels of participation are not required, in addition the system would enable citizens’ participation to affect collective decisions, and therefore their lives, which is essential if people are to participate. Inglehart claims that there are conducive dispositions to participatory democracy emerging, but the empirical evidence is very mixed and there needs to be much more empirical study on the prospect of such phenomena occurring. All this is also dependent on associations being artifactual, which I suggest they are, to a certain extent. Although the state must play some role towards democratising and funding associations, the fact that it cannot be neutral towards civil society means that much of the impetus for change will have to come from radical discourses within civil society itself, such as new social movements. Even if Inglehart is right, though, current socio-economic inequalities present a huge threat to the achievement and effective functioning of both deliberative and associational democracy. Although they also both help alleviate certain inequalities, a precondition of both is also greater equality. We are then caught in a vicious circle: we cannot achieve greater equality without participation in associations and deliberatively democratic decision-making, but we cannot have either of these without greater equality. I agree with both Young and Macpherson that, to break the cycle, we must try and increase participation and decrease inequality simultaneously. More research is therefore required to produce innovative ideas about how this could be achieved, and to measure the success of reducing inequalities on participation levels, and vice versa. The book has included many suggestions on how this could be achieved. Another reason not to be optimistic about a transition to a deliberative and associational democracy is the danger posed by co-option. The state will only allow the entry of associations into legislative arenas such as the mediating forums if the aims of these associations match state imperatives. However, as one of the state imperatives is legitimacy, inclusion of associations in decision-making could still be achieved but it would require a significant development of Inglehart’s post-material ethos and an increased demand for opportunities for meaningful public participation, in order for such strong demands of legitimacy to be placed upon the state. However, the dualist model of associational democracy overcomes some of — 234 —

Conclusion

the problems of co-option because it allows the same associations access to legislative arenas that are also participating in state critique in the informal public sphere, so the loss to a vital and critical civil society should not occur. Nevertheless, the key problem remains that any restructuring of the present institutional framework towards an approximation of deliberative democracy will be resisted by powerful groups who benefit from current arrangements and who will lose their present advantages if a deliberative and associational democracy occurred, so this model may well remain purely aspirational (Eckersley 2000: 131). It should, therefore, be apparent that I am not claiming that people will unite behind the idea of a system of deliberative and associational democracy, but I hope I have argued sufficiently that such a system would better promote the equal autonomy of all than does the present system. If people were to unite behind this idea, then it could provide the required ‘supportive beliefs’ that institutional reform obviously requires. I am not suggesting this will happen, only that it is possible and that it should. I have identified changes that are occurring in liberal democracies that are conducive to such a shift, but similarly there are many changes and factors present that hinder such a transition. However, starting from the premise that current democracies are flawed and can be significantly improved, I believe the ideas in the book make an important contribution to the debate about how democracy should be changed. As Young appreciates, social change arises from politics, and not philosophy, but ideas can provide an essential role in that process of politics because ‘they dislodge our assumption that what is given is necessary. They offer standpoints from which to criticise the given, and inspiration for imagining alternatives’ (Young 1990: 256). Hopefully this book achieves all these: points out the failings of current democracy and offers the tempting alternative that is deliberative and associational democracy. Many will object to the claim that autonomy is the normative core of democracy, and will favour democracy for other reasons. Others will not be convinced that participation in deliberation is most important, but might nonetheless see secondary associations as a solution to many of the complexities and challenges of contemporary public administration. Still others will favour deliberative democracy, but remain unconvinced that secondary associations can make an important contribution to this. This book does not suggest that autonomy is the only important value, that deliberation should be the only form of participation, or that secondary associations are the only institutions we need – far from it, in fact. However, I still maintain that all of these arguments are at their most persuasive when they are connected; that autonomy is a very important value, that — 235 —

Towards a Deliberative and Associational Democracy

deliberative democracy is a desirable method of making decisions because it enhances autonomy, and that secondary associations are unavoidable political facts that can, if structured correctly, help institutionalise deliberative democracy. Therefore, to deepen democracy we should move towards a deliberative and associational democracy.

— 236 —

Bibliography

Achterberg, W. (1996), ‘Sustainability and Associative Democracy’, in W. M. Lafferty and J. Meadowcroft (eds), Democracy and the Environment, Cheltenham: Edward Elgar, 157–74. Ackerman, B. (1980), Social Justice in the Liberal State, New Haven: Yale University Press. Andersen, N. V. and Hansen, K. (2004), ‘Deliberative Democracy and the Deliberative Poll on the Euro’, Scandinavian Political Studies, 27 (2): 261–86. Andersen, I. and Jagger, B. (1999), ‘Danish Participatory Models, Scenario Workshop and Consensus Conferences: Towards More Democratic Decision-Making’, Science and Public Policy, 26 (5), 331–40. Arblaster, A. (1994), Democracy, Buckingham: Open University Press. Arneson, R. J. (1980), ‘Mills vs. Paternalism’, Ethics, 90, 470–80. Arrow, K. (1963), Social Choice and Individual Values, New Haven: Yale University Press. Bache, I. and Flinders, M. (2002), Multi-Level Governance: Interdisciplinary Perspectives, Oxford: Oxford University Press. Bachrach, P. and Baratz, M. (1962), ‘Two Faces of Power’, American Political Science Review, 56, 947–52. Bader, V. (1995), ‘Citizenship and Exclusion: Radical Democracy, Community and Justice. Or, What is Wrong with Communitarianism?’, Political Theory, 31 (3), 211–46. Bader, V. (2005), ‘Associative Democracy and Minorities Within Minorities’, in A. Eisenberg, and J. Spinner-Halev, (eds), Minorities Within Minorities, Cambridge: Cambridge University Press, 319–39. Barber, B. (1984), Strong Democracy, Berkeley: University of California Press. Barnes, M. (1999), Building a Deliberative Democracy: An Evaluation of Two Citizens’ Juries, London: IPPR. Barnes, M., Knops, A., Newman, J. and Sullivan, H. (2004), ‘The Micro-Politics of Deliberation: Case Studies in Public Participation, Contemporary Politics, 10 (2), 93–110. Barry, B. (1991), ‘Is Democracy Special?’, in B. Barry (ed.), Democracy and Power: Essays in Political Theory 1, Oxford: Clarendon Press, 24–60.

— 237 —

Bibliography Barry, B. (2001), Culture and Equality, Cambridge: Polity Press. Beardsley, M.C. (1965), ‘Intrinsic Value’, Philosophy and Phenomenological Research, 26, 1–17. Beck, U. (1997), The Reinvention of Politics: Rethinking Modernity in the Global Social Order, Cambridge: Polity Press. Beck, U. (1999), World Risk Society, Cambridge: Polity Press. Beetham, D. (1977), ‘From Socialism to Fascism: The Reflection Between Theory and Practice in the Work of Robert Michels’, Political Studies, 25 (1), 3–24. Benhabib, S. (1996), ‘Toward a Deliberative Model of Democratic Legitimacy’, in S. Benhabib (ed.), Democracy and Difference: Contesting the Boundaries of the Political, Princeton, NJ: Princeton University Press, 67–94. Benn, S. I. and Weinstein, W. L. (1971), ‘Being Free to Act, and Being a Free Man’, Mind, 80, 194–211. Benn, S. I. (1976), ‘Freedom, Autonomy and the Concept of a Person’, Proceedings of the Aristotelian Society, 66, 109–30. Bennett, L. (1998), ‘The Uncivic Culture: Communication, Identity and the Rise of Lifestyle Politics’, Political Science and Politics, 31, 741–61. Benoit, J. P. and Kornhauser, L. (1994), ‘Social Choice in a Representative Democracy’, American Political Science Review, 88 (1), 185–92. Berger, P. (1991), The Sacred Canopy, New York: Anchor Books Berlin, I. (1984), ‘Two Concepts of Liberty’, in M. Sandel (ed.), Liberalism and its Critics, Oxford: Blackwell, 15–36. Berry, J. M. (1999), The New Liberalism and the Rising Power of Citizen Groups, Washington, DC: Brookings Institution Press. Bird, C. (2000), ‘The Possibility of Self-Government’, American Political Science Review, 94 (3), 563–77. Black, D. (1948), ‘On the Rationale of Group Decision-making’, Journal of Political Economy, 56, 23–34. Blaug, R. (1996), ‘New Theories of Discursive Democracy: A User’s Guide’, Philosophy & Social Criticism, 22 (1), 49–80. Blaug, R. (1999), Democracy: Real and Ideal: Discourse Ethics and Radical Politics, New York: SUNY Press. Bobbio, N. (1987), The Future of Democracy, Minneapolis: University of Minnesota Press. Bohman, J. (1996), Public Deliberation: Pluralism, Complexity and Democracy, Cambridge, MA: MIT Press. Bohman, J. (1997), ‘Deliberative Democracy and Effective Social Freedom: Capabilities, Resources, and Opportunities’, in J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 321–48. Bohman, J. (1998), ‘The Coming of Age of Deliberative Democracy’, Journal of Political Philosophy, 6 (4), 400–25. Bosnich, D. A. (1996), ‘The Principle of Subsidiarity’, Religion & Liberty, 6 (4), 1–4, available at www.acton.org/publicat/randl/96jul_aug/bosnich.html. Accessed 29 May 2002.

— 238 —

Bibliography Bovens, M., t’Hart, P. and Peters, B. G. (2001), Success and Failure in Government Policy, Cheltenham: Edward Elgar. Brandt, R (1959), Ethical Theory: The Problems of Normative and Critical Ethics, Englewood Cliffs, NJ: Prentice Hall. Brennan, G. and Lomasky, L. (1993), Democracy and Decision: The Pure Theory of Electoral Preference, Cambridge: Cambridge University Press. Brenner, N. (2004), New State Spaces: Urban Governance and the Rescaling of Statehood, Oxford: Oxford University Press. Brighton and Hove Community Sector Forum (2007), available at www.cvsectorforum.org.uk. Accessed 12 July 2007. British Council of Disabled People (2007), available at www.bcodp.org.uk. Accessed 20 July 2007. Budge, I. (2000), ‘Deliberative Democracy Versus Direct Democracy–Plus Political Parties!’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 195–209. Burnam, E., Smailes, S. and Chantler, K. (2004), ‘ ‘‘Culture’’ As a Barrier to Service Provision and Delivery: Domestic Violence Services for Minoritised Women’, Critical Social Policy, 24 (3), 332–57. Burns, D. (2000), ‘Can Local Democracy Survive Governance’, Urban Studies, 37 (5–6), 963–73. Cahill, M. (1994), The New Social Policy, Oxford: Blackwell. Campbell, A., Gurin, G. and Miller, W. (1954), The Voter Decides, Evanston, IL: Row, Peterson and Company. Caney, S. (1992), ‘Liberalism and Communitarianism: A Misconceived Debate’, Political Studies, 40, 273–89. Caney, S. (1995), ‘Anti-Perfectionism and Rawlsian Liberalism’, Political Studies, 43 (2), 248–64. Catt, H. (1999), Democracy in Practice, London: Routledge. Cerny, P. (2006), ‘Plurality, Pluralism, and Power: Elements of Pluralist Analysis in an Age of Globalisation’, in R. Eisfield (ed.), Pluralism: Developments in the Theory and Practice of Democracy, Opalden, Germany: Barbara Budrich, 81–110. Chambers, S. (1996), Reasonable Democracy: Jurgen Habermas and the Politics of Discourse, London: Cornell University Press. Choices and Rights (2007), available at www.choicesandrights.org.uk. Accessed 20 July 2007. Chowcat, I. (2000), ‘Moral Pluralism, Political Justification and Deliberative Democracy’, Political Studies, 48, 745–58. Christian Coalition (2007), available at www.cc.org. Accessed 23 July 2007. Christiano, T. (1990), ‘Freedom, Consensus, and Equality in Collective DecisionMaking’, Ethics, 101, 151–81. Christiano, T. (1996), ‘Deliberative Equality and the Democratic Order’, in I. Shapiro and R. Hardin (eds), Political Order: Nomos 38, New York: New York University Press, 251–87.

— 239 —

Bibliography Christiano, T. (1997), ‘The Significance of Public Deliberation’, in J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 243–77. Christman, J. (1988), ‘Constructing the Inner Citadel: Recent Work on the Concept of Autonomy’, Ethics, 99, 109–24. Christman, J. (1991), ‘Autonomy and Personal History’, Canadian Journal of Philosophy, 21 (1), 1–24. Christman, J. (2005), ‘Autonomy, Self-Knowledge, and Liberal Legitimacy’, in J. Christman and J. Andersen (eds), Autonomy and the Challenges to Liberalism, New York: Cambridge University Press, 330–57. Christman, J. and Andersen, J. (2005), ‘Introduction’, in J. Christman and J. Andersen (eds), Autonomy and the Challenges to Liberalism, New York: Cambridge University Press, 1–23. Clarke, C. (1998), ‘Self-Determination and Paternalism in Community Care: Practice and Prospects’, British Journal of Social Work, 28, 387–402. Clarke, P. (1999), Autonomy Unbound, Aldershot: Ashgate. Clarke, R. (2002), New Democratic Processes, London: IPPR. Cobb, R., Ross, J. K. and Ross M. H. (1976), ‘Agenda Building as a Comparative Political Process’, American Political Science Review, 70, 126–38. Cohen, J. and Arato, A. (1992), Civil Society and Political Theory, Cambridge, MA: MIT Press. Cohen, J. (1999), ‘Trust, Voluntary Association and Workable Democracy: The Contemporary American Discourse of Civil Society’, in M. Warren (ed.), Democracy and Trust, Cambridge: Cambridge University Press, 208–48. Cohen, J. (1989), ‘Deliberation and Democratic Legitimacy’, in A. Hamlin and P. Pettit (eds), The Good Polity: Normative Analysis of the State, Oxford: Blackwell, 17–35. Cohen, J. (1997), ‘Procedure and Substance in Deliberative Democracy’, in J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 407–37. Cohen, J. (1998), ‘Democracy and Liberty’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 185–231. Cohen, J. and Rogers, J. (1995), ‘Secondary Associations and Democratic Governance’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 7–98. Cohen, J. and Rogers, J. (2003), ‘Power and Reason’, in A. Fung and E. O. Wright (eds), Deepening Democracy: Institutional Innovation in Empowered Participatory Governance, London: Verso, 237–55. Cohen, S. (2002), ‘Dining with the Devil: The 1999 Immigration and Asylum Act and the Voluntary Sector’, in S. Cohen, B. Humphries and E. Mynott (eds), From Immigration Controls to Welfare Controls, London: Routledge, 141–56. Cole, G. D. H. (1920a), Social Theory, New York: Frederick A. Stokes. Cole, G. D. H. (1920b), Guild Socialism Re-stated, London: Leonard Parsons. Connelly, S. and Richardson, T. (2004), ‘Exclusion: The Necessary Difference between Ideal and Practical Consensus’, Journal of Environmental Planning and Management, 47 (1), 3–17.

— 240 —

Bibliography Connolly, W. (1993), The Terms of Political Discourse, Oxford: Blackwell. Connover, P., Searing, D. and Crewe, I. (2002), ‘The Deliberative Potential of Political Discussion’, British Journal for the Philosophy of Science, 32 (1), 21–62. Cooke, M. (1997), ‘Authenticity and Autonomy’, Political Theory, 25 (2), 258–88. Coote, A. (1997), Twelve Good Neighbours, London: Fabian Society. Coote, A. and Lenaghan, J. (1997), Citizens’ Juries: Theory into Practice, London: IPPR. Craig, G., Taylor, M., Monro, S., Wilkinson, M., Bloor, K. and Syed, A. (2002), Contract or Trust?, Bristol: Policy Press. Cronin, T. (1989), Direct Democracy: The Politics of Initiative, Referendum and Recall, London: Harvard University Press. Dahl, R. (1989), Democracy and Its Critics, New Haven: Yale University Press. Dahl, R. (1994), ‘A Democratic Dilemma: System Effectiveness versus Citizen Participation’, Political Science Quarterly, 109 (1), 21–34. Davies, C. (2007), ‘Grounding Governance in Dialogue? Discourse, Practice and the Potential for a New Public Sector Organizational Form in Britain’, Public Administration, 85 (1), 47–66. Davis, J., Stasson, M., Ono, K. and Zimmerman, S. (1988), ‘Effects of Straw Polls on Group Decision-Making: Sequential Voting Pattern, Timing and Local Majorities’, Journal of Personality and Social Psychology, 55, 918–26. D’Entre`ves, M. P. (2002), ‘Political Legitimacy and Democratic Deliberation’, in D’Entre`ves (ed.), Democracy as Public Deliberation: New Perspectives, Manchester: Manchester University Press, 39–52. Dewey, J. (1950), Experience and Education, New York: Macmillan. Dewey, J. (1988), ‘The Public and its Problems’, in J. O. Boydston (ed.), John Dewey: The Later Works, 1925–1953, Volume 2, 1925–1927, Carbondale: Southern Illinois University Press. Dobson, A. (1996), ‘Representative Democracy and the Environment’, in W. M. Lafferty and J. Meadowcroft (eds), Democracy and the Environment, Cheltenham: Edward Elgar, 124–39. Dodson, K. (1997), ‘Autonomy and Authority in Kant’s Rechtslehre’, Political Theory, 25 (1), 93–109. Downs, A. (1957), An Economic Theory of Democracy, New York: Harper and Brothers. Doyle, J. (1990), ‘Democratic Autonomy: Fulfilling Revolutionary Promises’, Revolution and Human Rights, 80–5. Drake, R. (2001), The Principles of Social Policy, London: Palgrave. Dryzek, J. (1990), Discursive Democracy: Politics, Policy and Political Science, Cambridge: Cambridge University Press. Dryzek, J. (1996), Democracy in Capitalist Times: Ideals, Limits and Struggles, Oxford: Oxford University Press. Dryzek, J. (2000), Deliberative Democracy and Beyond: Liberals, Critics, Contestations, Oxford: Oxford University Press. Dryzek, J. (2006), Deliberative Global Politics, Cambridge: Polity Press.

— 241 —

Bibliography Dworkin, G. (1988), The Theory and Practice of Autonomy, Cambridge: Cambridge University Press. Dworkin, R. (1985), A Matter of Principle, Cambridge, MA: Harvard University Press. Dworkin, R. (1989), ‘Liberal Community’, California Law Review, 77 (3), 479–504. Eckersley, R. (1996), ‘Greening Liberal Democracy: The Rights Discourse Revisited’, in B. Doherty and M. de Geus (eds), Democracy and Green Political Thought, London: Routledge, 212–36. Eckersley, R. (2000), ‘Deliberative Democracy, Ecological Representation and Risk: Towards a Democracy of the Affected’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 117–32. Edwards, M. (2004), Civil Society, Cambridge: Polity Press. Eisenberg, A. (2006), ‘Pluralism and the Politics of Diversity’, in R. Eisfield (ed.), Pluralism: Developments in the Theory and Practice of Democracy, Opalden, Germany: Barbara Budrich, 59–79. Eisfeld, R. (2006), ‘Introduction’, in R. Eisfield (ed.), Pluralism: Developments in the Theory and Practice of Democracy, Opalden, Germany: Barbara Budrich, 11–20. Eldson, K. T., Reynolds, J. and Stewart S. (1995), Voluntary Organisations: Citizenship, Learning and Change, Leicester: NIACE. Elkin, S. L. (2004), ‘Thinking Constitutionally: The Problem of Deliberative Democracy’, Social Philosophy and Policy, 21 (1), 29–75. Elster, J. (1976), ‘Ulysses and the Sirens: A Theory of Imperfect Rationality’, Social Science Information, 16 (5), 469–526. Elster, J. (1989), Solomonic Judgments: Studies in the Limitations of Rationality, New York: Cambridge University Press. Elster, J. (1991), ‘Sour Grapes: Utilitarianism and the Genesis of Wants’, in A. Sen and B. Williams (eds), Utilitarianism and Beyond, Cambridge: Cambridge University Press, 219–38. Elster, J. (1995), ‘Strategic Uses of Argument’, in K. Arrow, R. Mnookin, L. Ross, A. Tversky and R. Wilson (eds), Barriers to Conflict Resolution, New York: Norton, 237–57. Elster, J. (1997) ‘The Market and the Forum: Three Varieties of Political Theory’, in J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 3–33. Elster, J. (1998a), ‘Introduction’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 1–19. Elster, J. (1998b), ‘Deliberation and Constitution Making’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 97–122. Elstub, S. (2003), Cultivating Autonomy: A Case for Associational and Deliberative Democracy, PhD thesis, Sheffield: University of Sheffield. Elstub, S. (2006a), ‘A Double Edged Sword: The Increasing Diversity of Deliberative Democracy’, Contemporary Politics, 12 (3–4), 301–20. Elstub, S. (2006b), ‘Towards an Inclusive Social Policy in the UK: The Need for Democratic Deliberation in Voluntary and Community Associations’, Voluntas, 17 (1), 17–39.

— 242 —

Bibliography England, T. (2000), ‘Rethinking Democracy and Education: Towards an Education of Deliberative Citizens’, Journal of Curriculum Studies, 32 (2), 305–13. Estlund, D. (1997), ‘Beyond Epistemic Fairness and Deliberation: The Epistemic Dimension of Democratic Authority’, in J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 173–204. Evans, S. M. and Boyte, H. C. (1992), Free Spaces: The Sources of Democratic Change in America, Chicago: University of Chicago Press. Fearon, J. (1998), ‘Deliberation as Discussion’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 44–68. Feinberg, J. (1971), ‘Legal Paternalism’, Canadian Journal of Philosophy, 1, 105–24. Femia, J. (1996), ‘Complexity and Deliberative Democracy’, Inquiry, 39, 359–97. Festenstein, M. (2002), ‘Deliberation, Citizenship and Identity’, in D’Entre`ves (ed.), Democracy as Public Deliberation: New Perspectives, Manchester: Manchester University Press, 88–111. Festenstein, M. (2005), Negotiating Diversity: Culture, Deliberation, Trust. Cambridge: Polity Press. Figgis, J. (1913), Churches in the Modern State, London: Longmans, Green and Co. Fiorino, D. J. (1995), Making Environmental Policy, Berkeley: University of California Press. Fishkin, J. (1991), Democracy and Deliberation: New Directions for Democratic Reform, London: Yale University Press. Fishkin, J. (1995), The Voice of the People: Public Opinion and Democracy, New Haven: Yale University Press. Fishkin, J. and Luskin, R. (2000), ‘The Quest for Deliberative Democracy’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 17–27. Flinders, M. (1997), ‘Quangos: Why Do Governments Love Them?’, in M. Flinders, I. Harden and D. Marquand (eds), How to Make Quangos Democratic, London: Charter 88. Flinders, M. (1999), ‘British Quasi-Government: History, Diversity and Debate’, Talking Politics, 11 (3), 176–81. Flinders, M. (2004), ‘Distributed Public Governance in Britain’, Public Administration, 82 (4), 883–909. Follesdal, A. (1999), ‘Subsidiarity and Democratic Deliberation’, Arena Working Papers, 21, available at www.arena.uio.no/publications/wp_21.htm. Accessed 29 May 2002. Forbes, J. and Sashidharan, S. P. (1997), ‘User Involvement in Services–Incorporation or Challenge?’, British Journal of Social Work, 27, 481–98. Forester, J. (1999), The Deliberative Practitioner, Cambridge, MA: MIT Press. Fraser, N. (1987), ‘Women, Welfare and the Politics of Need Interpretation’, Hypatia, 2 (1), 103–21. Fraser, N. (1989), ‘Struggle over Needs: Outline of a Socialist-Feminist Critical Theory of Late Capitalist Political Culture’, in N. Fraser (ed.), Unruly Practices:

— 243 —

Bibliography Power Discourse and Gender in Contemporary Social Theory, Cambridge: Polity Press, 161–87. Fraser, N. (1992), ‘Rethinking The Public Sphere: A Contribution To The Critique Of Actually Existing Democracy’, in C. Calhoun (ed.), Habermas and The Public Sphere, Cambridge, MA: MIT Press, 107–42. Fromm, E. (1955), The Sane Society, London: Routledge. Fukuyama, F. (1995), Trust: The Social Virtues and the Creation of Prosperity, London: Penguin. Fung, A. (2003a), ‘Associations and Democracy: Between Theories, Hopes and, Realities’, Annual Review of Sociology, 29, 515–40. Fung, A. (2003b), ‘Survey Article: Recipes for Public Spheres: Eight Institutional Design Choices and their Consequences’, Journal of Political Philosophy, 11 (3), 338–67. Fung, A. and Wright, E. O. (2001), ‘Deepening Democracy: Innovations in Empowered Participatory Governance’, Politics and Society, 29 (1), 5–41. Fung, A. and Wright, E. O. (2003), ‘Countervailing Power in Empowered Participatory Governance’, in A. Fung and E. O. Wright (eds), Deepening Democracy: Institutional Innovation in Empowered Participatory Governance, London: Verso, 259–89. Galbraith, J. K. (1993), The Culture of Contentment, London: Penguin. Galston, W. (1995), ‘Two Concepts of Liberalism’, Ethics, 105, 516–34. Galston, W. (2000a), ‘Civil Society and the ‘‘Art of Association’’ ’, Journal of Democracy, 11 (1), 64–70. Galston, W. (2000b), ‘Freedom of Association: Review Essay’, American Political Science Review, 94 (4), 929. Gambetta, D. (1998), ‘ ‘‘Claro!’’: An Essay on Discursive Machismo’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 19–43. Gastill, J. (1993), Democracy in Small Groups: Participation, Decision-Making and Communication, Philadelphia: New Society Publishers. Gaus, G. (1997), ‘Reason, Justification, and Consensus: Why Democracy Can’t Have It All’, in J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 205–42. Germain, R. (2001), ‘Global Financial Governance and the Problem of Inclusion’, Global Governance, 7, 411–26. Gilligan, C. (1982), In a Different Voice, Cambridge, MA: Harvard University Press. Giullari, S. and Shaw, M. (2005), ‘Supporting or Controlling? New Labour’s Housing Strategy for Teenage Parents’, Critical Social Policy, 25 (3), 402–17. Goodin, R. (1986), ‘Laundering Preferences’, in J. Elster and A. Hylland (eds), Foundations of Social Choice Theory, Cambridge: Cambridge University Press. Goodin, R. (1992), Motivating Political Morality, Cambridge, MA: Blackwell. Goodin, R. (2003), ‘Democratic Deliberation Within’, in J. Fishkin and P. Laslett (eds), Debating Deliberative Democracy, Oxford: Blackwell, 54–79. Gould, C. (1988), Rethinking Democracy: Freedom and Social Co-operation in Politics, Economics and Society, New York: Cambridge University Press.

— 244 —

Bibliography Gould, C. (1996), ‘Diversity and Democracy: Representing Differences’, in S. Benhabib (ed.), Democracy and Difference: Contesting the Boundaries of the Political, Princeton, NJ: Princeton University Press, 171–86. Graham, K. (1982), ‘Democracy and the Autonomous Moral Agent’, in K. Graham (ed.), Contemporary Political Philosophy: Radical Studies, Cambridge, MA: Harvard University Press, 113–37. Graham, K. (1986), The Battle of Democracy, Totowa, NJ: Barnes and Noble Books. Gundersen, A. (1995), The Environmental Promise of Democratic Deliberation, Madison: University of Wisconsin Press. Gundersen, A. (2000), The Socratic Citizen: A Theory of Deliberative Democracy, Oxford: Lexington Books. Gutmann, A. (1980), Liberal Equality, Cambridge: Cambridge University Press. Gutmann, A. (1985), ‘Communitarian Critics of Liberalism’, Philosophy and Public Affairs, 14 (3), 308–22. Gutmann, A. (1998), ‘Freedom of Association: An Introductory Essay’, in A. Gutmann (ed.), Freedom of Association, Princeton, NJ: Princeton University Press, 3–31. Gutmann, A. (2003), Democracy and Identity, Princeton, NJ: Princeton University Press. Gutmann, A. and Thompson, D. (1996), Democracy and Disagreement, London: Harvard University Press. Habermas, J. (1975), Legitimation Crisis, Boston, MA: Beacon Press. Habermas, J. (1990), Moral Consciousness and Communicative Action, Cambridge: Polity Press. Habermas, J. (1996a), Between Facts and Norms: Contributions to a Discourse Theory on Law and Democracy, Cambridge: Polity Press. Habermas, J. (1996b), ‘Three Normative Models of Democracy’, in S. Benhabib (ed.), Democracy and Difference: Contesting the Boundaries of the Political, Princeton, NJ: Princeton University Press, 21–30. Hadley, R. and Hatch, S. (1981), Social Welfare and the Future of the State, London: Allen and Unwin. Hamilton, H. D. (1970), ‘Direct Legislation: Some Implications of Open Housing Referenda’, American Political Science Review, 64, 124–37. Hansen, K. (2004), Deliberative Democracy and Opinion Formation, Odense: University Press of Southern Denmark. Harden, I. and Marquand, D. (1997), ‘How to Make Quangos Democratic’, in M. Flinders, I. Harden and D. Marquand (eds), How to Make Quangos Democratic, London: Charter 88, 8–30. Harris, M., Halfpenny, P. and Rochester, C. (2003), ‘A Social Policy Role for FaithBased Organisations? Lessons from the UK Jewish Voluntary Sector’, Journal of Social Policy, 32 (1), 93–112. Harrison, R. (1993), Democracy, London: Routledge. Haworth, L. (1986), Haworth: An Essay in Philosophical Psychology and Ethics, New Haven: Yale University Press.

— 245 —

Bibliography Hayek, F. A. (1937), Individualism and Economic Order, London: Routledge and Kegan Paul. Hayek, F. A. (1979), Law, Legislation and Liberty, Vol. 3: The Political Order of the Free People, Chicago: University of Chicago Press. Held, D. (1989), ‘Citizenship and Autonomy’, in D. Held (ed.), Political Theory and the Modern State, Cambridge: Polity Press, 189–213. Held, D. (1995), Democracy and The Global Order: From the Modern State to Cosmopolitan Governance, Cambridge: Polity Press. Held, D. (1996), Models of Democracy, 2nd ed, Cambridge: Polity Press. Hendriks, C. (2002), ‘Institutions of Deliberative Democratic Processes and Interest Groups: Roles, Tensions and Incentives’, Australian Journal of Public Administration, 61 (1), 64–75. Hendriks, C. (2006), ‘Integrated Deliberation: Reconciling Civil Society’s Dual Role in Deliberative Democracy’, Political Studies, 54 (3), 486–508. Hendriks, C., Dryzek, J. and Hunold, C. (2007), ‘Turning Up the Heat: Partisanship in Deliberative Innovation’, Political Studies, 55 (2), 362–83. Hensler, D. and Hensler, C. (1979), Evaluating Nuclear Power: Voter Choice on the California Nuclear Energy Initiative, Santa Monica, CA: Rand Corporation. Herring, E. (1929), Group Representation before Congress, Baltimore, MD: Johns Hopkins University Press. Hibbing, J. and Theiss-Moore, E. (2002), Stealth Democracy, Cambridge: Cambridge University Press. Hirschman, A. (1970), Exit, Voice and Loyalty: Responses to Decline in Firms, Organisations and States, Cambridge, MA: Harvard University Press. Hirst, P. (1994), Associative Democracy: New Forms of Economic and Social Governance, Cambridge: Polity Press. Hirst, P. (1995), ‘Can Secondary Associations Enhance Democratic Governance?’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 101–13. Hirst, Paul (1996) ‘Democracy and Civil Society’, The Political Quarterly, 67, 97–115. Hodgson, G. (1998), ‘Socialism Against Markets? A Critique of Two Recent Proposals’, Economy and Society, 27, 407–33. Hume, D. [1739] (1978), A Treatise of Human Nature, Oxford: Clarendon Press. Hunold, C. (2001), ‘Corporatism, Pluralism and Democracy: Towards a Deliberative Theory of Bureaucratic Accountability’, Governance, 14 (2), 151–67. Hurka, T. (1987), ‘Why Value Autonomy?’, Social Choice Theory and Practice, 13, 361–82. Hurka, T. (1993), Perfectionism, Oxford: Oxford University Press. Hussain, Y., Atkin, K. and Ahmad, W. (2002), South Asian Disabled Young People and their Families, Bristol: Policy Press. Huxley, A. [1932] (1994), Brave New World, London: Flamingo. Hyland, J. (1995), Democratic Theory: The Philosophical Foundations, Manchester: Manchester University Press. Immergut, E. (1995), ‘An Institutional Critique of Associative Democracy’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 201–6.

— 246 —

Bibliography Inglehart, R. (1990), Culture Shift in Advanced Industrial Society, Princeton, NJ: Princeton University Press. Jacobson, L. (1998), ‘Local Timber Collaboration Stalls in National Arena’, Planning, 61 (11), 22–3. Johnson, J. (1998), ‘Arguing for Deliberation: Some Skeptical Considerations’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 161–84. Jordan, G. (2007), ‘Policy without Learning: Double Devolution and the Abuse of the Deliberative Idea’, Public Policy and Administration, 22 (1), 48–73. Kant, I. [1795] (1957), ‘Perpetual Peace’, in L. W. Beck (ed.), Kant: On History, Indianapolis, IN: Bobbs-Merrill. Kant, I. [1785] (1964), Groundwork of the Metaphysics of Morals, New York: Harper and Row. Kateb, G. (1998), ‘The Value of Association’, in A. Gutmann (ed.), Freedom of Association, Princeton, NJ: Princeton University Press, 35–63. Kelsen, H. (1949), General Theory of Law and State, Cambridge; MA: Harvard University Press. Kickert, W. (2001) ‘Public Management of Hybrid Organisations: Governance of Quasi-Autonomous Executive Agencies’, International Public Management Journal, 4 (2), 135–50. Kidger, J. (2004), ‘Including Young Mothers: Limitations for Supporting Teenage Parents’, Critical Social Policy, 24 (3), 291–311. Knight, J. and Johnson, J. (1994), ‘Aggregation and Deliberation: On the Possibility of Democratic Legitimacy’, Political Theory, 22 (2), 277–96. Knight, J. and Johnson, J. (1997), ‘What Sort of Political Equality Does Deliberative Democracy Require?’, in J. Bohman and W. Rehg (eds) Deliberative Democracy, Cambridge, MA: MIT Press, 279–319. Kohler, T. C. (1993), ‘Lessons from the Social Charter: State Corporation, and the Meaning of Subsidiarity’, University of Toronto Law Journal, 43, 607–28. Korsgaard, C. (1996), The Sources of Normativity, Cambridge: Cambridge University Press. Kuper, R. (1997), ‘Deliberating Waste: The Hertfordshore Citizens’ Jury’, Local Environment, 2 (2), 139–53. Kymlicka, W. (1990), Contemporary Political Philosophy, Oxford: Clarendon Press. Kymlicka, W. (1995), Multicultural Citizenship: A Liberal Theory of Minority Rights, Oxford: Clarendon Press. Kymlicka, W. (1988), ‘Liberalism and Communitarianism’, Canadian Journal of Philosophy, 18 (2), 181–204. Ladd, E. C. (1999), The Ladd Report, New York: Free Press. Lakoff, S. (1996), Democracy: History, Theory, Practice, Oxford: Westview Press. Lane, R. (1959), Political Life: Why People Get Involved in Politics, Glencoe, IL: The Free Press of Glencoe. Langan, M. and Clarke, J. (1994), ‘Managing in the Mixed Economy of Care’, in J. Clarke, A. Cochrane and E. McLaughlin (eds), Managing Social Policy, London: Sage, 73–92.

— 247 —

Bibliography Langan, M. (2000), ‘Social Services: Managing the Third Way’, in J. Clarke and S. Gewirtz (eds), New Managerialism, New Welfare?, London: Sage, 152–68. Larmore, C. (1987), Patterns of Moral Complexity, Cambridge: Cambridge University Press. Laski, H. (1925), A Grammar of Politics, London: Allen and Unwin. Laughlin, M. and Scott C. (1997), ‘The Regulatory State’, in P. Dunleavy, A. Gamble, I. Holliday and G. Peele (eds), Developments in British Politics 5, London: Macmillan, 205–19. Leat, D. (1996), ‘Are Voluntary Organisations Accountable?’, in D. Billis and M. Harris (eds), Voluntary Agencies, London: Macmillan, 61–79. Leib, E. J. (2004), Deliberative Democracy in America: A Proposal for A Popular Branch of Government, University Park, PA: Pennsylvania State University Press. Lessnoff, M. H. (1986), Social Contract, London: Macmillan. Levine, A. (1993), The General Will: Rousseau, Marx, Communism, Cambridge: Cambridge University Press. Lindley, R. (1986), Autonomy, London: Macmillan. Lindsay, A. D. (1929), Essentials of Democracy, London: Oxford University Press. Lipson, M. (1995), ‘Autonomy and Democracy’, Yale Law Journal, 104 (8), 2249– 76. Locke, J. [1689] (1955), Letter Concerning Toleration, Indianapolis, IN: BobbsMerrill. Locke, J. [1689] (1988), Two Treatises of Government, London: Everyman. Long-Term Medical Conditions Alliance (2007), available www.lmca.org.uk. Accessed 20 July 2007. Lowndes, V., Pratchett, L. and Stoker, G. (2006), ‘Diagnosing and Remedying the Failings of Official Participation Schemes: the CLEAR Framework’, Social Policy and Society, 5 (2), 281–91. Lukes, S. (1973), Individualism, Oxford: Blackwell. Lukes, S. (1974), Power: A Radical View, London: Macmillan. Luskin, R., Fishkin, J. and Jowell, R. (2002), ‘Considered Opinions: Deliberative Polling in Britain’, British Journal of Political Science, 32, 455–87. MacCallum, G. (1967), ‘Negative and Positive Freedom’, Philosophical Review, 76, 312–34. Mackie, G. (1998), ‘All Men Are Liars: Is Democracy Meaningless?’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 69–96. Macpherson, C. B. (1973), Democratic Theory: Essays in Retrieval, Oxford: Oxford University Press. Macpherson, C. B. (1977), The Life and Times of Liberal Democracy, Oxford: Oxford University Press. Madison, J. (1966), The Federalist Papers, New York: Doubleday. Manin, B. (1987), ‘On legitimacy and Political Deliberation’, Political Theory, 5 (3), 338–68. Mansbridge, J (1980), Beyond Adversary Democracy, London: University of Chicago Press.

— 248 —

Bibliography Mansbridge, J. (1992), ‘A Deliberative Theory of Interest Representation’, in M. Petracca (ed.), The Politics of Interests: Interest Groups Transformed, Oxford: Westview Press, 32–57. Mansbridge, J. (1995), ‘A Deliberative Perspective on Neocorporatism’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 133–47. Mansbridge, J. (1996), ‘Using Power? Fighting Power: The Polity’, in S. Benhabib (ed.), Democracy and Difference: Contesting the Boundaries of the Political, Princeton, NJ: Princeton University Press, 46–66. Martell, L. (1992), ‘New Ideas of Socialism’, Economy and Society, 21 (2), 152–72. Massachusetts Climate Action Network (2007), available at www.massclimateaction.org. Accessed 12 July 2007. McCarthy, T. (1994), ‘Kantian Constructivism and Reconstructivism: Rawls and Habermas in Dialogue’, Ethics, 105, 44–63. McCombs, M. and Reynolds, A. (eds) (1999), The Poll with a Human Face: The National Issues Convention Experiment in Political Communication, Mahwah, NJ: Lawrence Erlbaum. McIver, S. (1997), An Evaluation of the King’s Fund Citizens’ Juries Programme, Birmingham: Health Services Management Centre. McLean, I. (1991), ‘Forms of Representation and Systems of Voting’, D. Held (ed.), Political Theory Today, Cambridge: Polity Press, 172–96. Meadowcroft, J. (2001) ‘Community Politics, Representation and the Limits of Deliberative Democracy’, Local Government Studies, 27 (3), 25–42. Michelman, F. (1997), ‘How Can the People Ever Make the Laws? A Critique of Deliberative Democracy’, J. Bohman and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 145–71. Michels, R. [1911] (1959), Political Parties, trans. E. and C. Paul, New York: Dover Publications. Milewa, T. (2004), ‘Local Participatory Democracy in Britain’s Health Service: Innovation or Fragmentation of a Universal Citizenship?’, Social Policy and Administration, 38 (3), 240–52. Mill, J. S. (1993), Utilitarianism, On Liberty, Considerations on Representative Government, G. Williams (ed.), London: Everyman. Miller, D. (1993), ‘Deliberative Democracy and Social Choice’, in D. Held (ed.), Prospects for Democracy, Cambridge: Polity Press, 74–92. Miller, D. (2000), ‘Is Deliberative Democracy Unfair to Disadvantaged Groups?’, in D. Miller, Citizenship and National Identity, Cambridge: Polity Press, 142–60. Mutz, D. (2006), Hearing the Other Side: Deliberative Versus Participatory Democracy, New York: Cambridge University Press. Mylod, R. J. (1998), ‘Subsidiarity and Health Care Reform’, Religion & Liberty, 8 (1), available at www.acton.org/publicat/randl/98jan_feb/mylod.html. Accessed 29 May 2002. Nagel, T. (1975), ‘Rawls on Justice’, in N. Daniels (ed.), Reading Rawls: Critical Studies on Rawls’ ‘‘Theory of Justice’’, Stanford, CA: Stanford University Press, 1–15.

— 249 —

Bibliography Nagel, T. (1991), Equality and Partiality, Oxford: Oxford University Press. National Institute for Clinical Excellence (2007), available at www.nice.org.uk. Accessed 27 July 2007. Newman, J., Barnes, M., Sullivan, H. and Knops, A. (2004), ‘Public Participation and Collaborative Governance’, Journal of Social Policy, 33 (2), 203–23. Newman, J. (2005), ‘Participative Governance and the Remaking of the Public Sphere’, in J. Newman (ed.), Remaking Governance: Peoples, Politics and the Public Sphere, Bristol: Policy Press, 119–38. Nozick, R. (1974), Anarchy, State and Utopia, New York: Basic Books. Nurmi, H. (1998), ‘Voting Paradoxes and Referenda’, Social Choice and Welfare, 15, 333–50. Offe, C. (1984), Contradictions of the Welfare State, Cambridge, MA: MIT Press. Offe, C. (1995), ‘Some Skeptical Considerations on the Malleability of Representative Institutions’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 114–32. Office of the Deputy Prime Minister (2005), New Localism: Citizen Engagement, Neighbourhoods and Public Services, London: ODPM. O’Flynn, I. (2006), Deliberative Democracy and Divided Societies, Edinburgh: Edinburgh University Press. Olson, M. (1965), The Logic of Collective Action, Cambridge, MA: Harvard University Press. Olson, M. (1972), ‘Social Participation and Voting Turnout: A Multivariate Analysis’, American Sociological Review, 37, 317–33. O’Neill, J. (1998), ‘Rhetoric Science and Philosophy’, Philosophy of the Social Sciences, 28 (2), 205–25. O’Neill, J. (2003), ‘Socialism, Associations and the Market’, Economy and Society, 32 (2), 184–206. Operation Rescue (2007), available at www.operationrescue.org. Accessed 27 July 2007. Orbell, J. M., van der Kragt, A. and Dawes, R. (1988), ‘Explaining DiscussionInduced Co-operation’, Journal of Personality and Social Psychology, 54, 811–19. Pacific Institute for Community Organizing (2007), available at www.piconetwork.org. Accessed 27 July 2007. Parekh, B. (2000), Rethinking Multiculturalism: Cultural Diversity and Political Theory, Basingstoke: Palgrave. Parkes, T., Taylor, M. and Wilkinson, M. (2004), ‘From Protest To Partnership? Voluntary and Community Organisations in the Democratic Process’, in M. J. Todd and G. Taylor (eds), Democracy and Participation: Popular Protest and New Social Movements, London: Merlin, 307–25. Parkinson, J. (2006), Deliberating in the Real World: Problems of Legitimacy in Deliberative Democracy, Oxford: Oxford University Press. Parsons, T. (1971), The System of Modern Societies, Englewood Cliffs, NJ: Prentice Hall. Pascall, G. (1997), Social Policy: A New Feminist Analysis, London: Routledge.

— 250 —

Bibliography Peak District Park Authority (2007), available at www.peakdistrict.org. Accessed 27 July 2007. Pelletier, D., Kraak, V., McCullum, C., Uusitalo, U. and Rich, R. (1999), ‘The Shaping of Collective Values through Deliberative Democracy: An Empirical Study from New York’s North Country’, Policy Sciences, 32, 103–31. Pellizzoni, L. (2001), ‘The Myth of the Best Argument: Power, Deliberation and Reason’, British Journal of Sociology, 52 (1), 59–86. Pennington, M. (2003), ‘Hayekian Political Economy and the Limits of Deliberative Democracy’, Political Studies, 51, 722–39. Pennock, R. J. (1989), ‘The Justification of Democracy’, in G. Brennan and L. E. Lomasky (eds), Politics and Process: New Essays in Democratic Thought, Cambridge: Cambridge University Press. Perczynski, P. (2000), ‘Active Citizenship and Associative Democracy’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 161–71. Pettit, P. (2003), ‘Deliberative Democracy, the Discursive Dilemma, and Republican Theory’, in J. Fishkin and P. Laslett (eds), Debating Deliberative Democracy, Oxford: Blackwell, 138–62. Phillips, A. (1995), The Politics of Presence, Oxford: Clarendon Press. Polletta, F. (2002), Freedom is an Endless Meeting: Democracy in American Social Movements, Chicago: Chicago University Press. Popper, K. (1962), The Open Society and Its Enemies, Princeton, NJ: Princeton University Press. Post, R. (1993), ‘Managing Deliberation: The Quandary of Democratic Dialogue’, Ethics, 103, 654–78. Powell, F. and Guerin, D. (1997), Civil Society and Social Policy: Voluntarism in Ireland, Dublin: A & A Farmar. Power Inquiry (2006), Power to the People, London: Power Inquiry. Przeworski, A. (1998), ‘Deliberation and Ideological Domination’, in J. Elster (ed.), Deliberative Democracy, Cambridge: Cambridge University Press, 140–60. Putnam, R. D. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton NJ: Princeton University Press. Putnam, R. D. (2000), Bowling Alone. The Collapse and Revival of American Community, New York: Simon and Schuster. Rao, N. (1996), Towards Welfare Pluralism: Public Services in a Time of Change, Aldershot: Dartmouth. Rawls, J. (1971), Theory of Justice, Cambridge, MA: Harvard University Press. Rawls, J. (1993), Political Liberalism, New York: Columbia University Press. Rawls, J. (1997), ‘The Idea of Public Reason Revisited’, University of Chicago Law Review, 94 (3), 765–807. Raz, J. (1986), The Morality of Freedom, Oxford: Clarendon Press. Reich, R. B. (1988), ‘Policy Making Democracy’, in R. Reich (ed.), The Power of Public Ideas, Cambridge, MA: Harvard University Press.

— 251 —

Bibliography Renn, O., Webler, T. and Wiedmann, P. (1995), Novel Approaches to Public Participation in Environmental Decision-Making, Amsterdam: Kluwer. Rhodes, R. (1997), Understanding Governance: Policy Networks, Governance, Reflexivity and Accountability, Buckingham: Open University Press. Richardson, H. (2002), Democratic Autonomy: Public Reasoning About the Ends of Policy, Oxford: Oxford University Press. Riker, W. (1982), Liberalism Against Populism: A Confrontation Between the Theory of Democracy and the Theory of Social Choice, San Francisco: W. H. Freeman. Rippe, K. P. and Schaber, P. (1999), ‘Democracy and Environmental DecisionMaking’, Environmental Values, 8, 75–88. Roberts, J. (1996), ‘From Discipline to Dialogue’, in R. Munro and J. Mouritsen (eds), Accountability: Power, Ethos and the Technologies of Managing, International London: Thomson Business Press. Roßteutscher, S. (2000), ‘Associative Democracy: Fashionable Slogan or Constructive Innovation’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 172–183. Rosenblum, N. (1998), Membership and Morals: The Personal uses of Pluralism in America, Princeton, NJ: Princeton University Press. Rouse, J. and Smith, G. (1999), ‘Accountability’, in M. Powell (ed.), New Labour, New Welfare State: The Third Way in British Social Policy, Bristol: Policy Press. Rousseau, J. J. [1762] (1968), The Social Contract, Harmondsworth: Penguin Books. Ryan, P. (1999), ‘Australia Deliberates: A Republic: Yes or No? The First Australian Deliberative Poll’, available at www.i-d-a.com.au/presentations.htm. Accessed 5 May 2000. Sagoff, M. (1988), The Economy of the Earth, Cambridge: Cambridge University Press. Sagoff, M. (1998), ‘Aggregation and Deliberation in Valuing Environmental Public Goods: A Look Beyond Contingent Pricing’, Ecological Economics, 24, 213–30. Sagoff, M. (1999), ‘The View from Quincy Library: Civic Engagement and Environmental Problem Solving’, in R. K. Fullinwider (ed.), Civil Society, Democracy and Civic Renewal, New York: Rowman and Littlefield, 151–83. Salamon, L. and Anheier, H. (1996), The Emerging Nonprofit Sector, Manchester: Manchester University Press. Sandel, M. J. (1982), Liberalism and the Limits of Justice, Cambridge: Cambridge University Press. Sandel, M. J. (1984), ‘Morality and the Liberal Ideal: Must Individual Rights Betray the Common Good?’, The New Republic, 190, 7 May, 1915–17. Sanders, L. (1997), ‘Against Deliberation’, Political Theory, 25 (3), 347–76. Santoro, E. (1993), ‘Democratic Theory and Individual Autonomy’, European Journal of Political Research, 23, 121–43. Sartre, J. P. (1948), Existentialism and Humanism, London: Methuen. Saward, M. (1998), ‘Deliberation, Difference and Democratic Institutions’, Government and Opposition, 33 (4), 519–26.

— 252 —

Bibliography Saward, M. (2000), ‘Less Than Meets the Eye: Democratic Legitimacy and Deliberative Theory’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 66–77. Saward, M. (2003), ‘Enacting Democracy’, Political Studies, 51, 161–79. Schachar, A. (2001), Multicultural Jurisdictions: Cultural Differences and Women’s Rights, Cambridge: Cambridge University Press. Schattsneider, E. E. (1975), The Semi-Sovereign People: A Realist’s View of Democracy in America, Orlando: Holt, Rinehart and Winston. Schmitter, P. (1995), ‘The Irony of Modern Democracy and the Viability of Efforts to Reform its Practice’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 167–83. Schumpeter, J. (1974), Capitalism, Socialism and Democracy, London: Unwin University Press. Scoccia, D. (1990), ‘Paternalism and Respect for Autonomy’, Ethics, 100 (2), 318–34. Shapiro, I. (1999), ‘Enough Deliberation’, in S. Macedo (ed.), Deliberative Politics, New York: Oxford University Press, 28–38. Sher, G. (1997), Beyond Neutrality: Perfectionism and Politics, Cambridge: Cambridge University Press. Sikkuy (2007), available at www.sikkuy.org.il/english/home.html. Accessed 12 July 2007. Simmel, G. (1955), Conflict and the Web of Group Affiliations, trans. K. H. Wolff, and R. Bendix, New York: Free Press. Skelcher, C. (1998), The Appointed State: Quasi-Governmental Organisations and Democracy, Buckingham: Open University Press. Skocpol, T. (1999), ‘Associations without Members’, American Prospect, 45, 66–73. Smith, G. and Wales, C. (2000), ‘Citizens’ Juries and Deliberative Democracy’, Political Studies, 48 (1), 51–65. Smith, G. (2000), ‘Toward Deliberative Institutions’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 29–39. Smith, G. (2001), ‘Taking Deliberation Seriously: Institutional Design and Green Politics’, Environmental Politics, 10 (3), 72–93. Squires, J. (2002), ‘Deliberation and Decision-Making: Discontinuity in the TwoTrack Model’, in D’Entre`ves (ed.), Democracy as Public Deliberation: New Perspectives, Manchester: Manchester University Press, 133–56. Squires, J. (2000), ‘Group Representation, Deliberation and the Displacement of Dichotomies’, in M. Saward (ed.), Democratic Innovation: Deliberation, Representation and Association, London: Routledge, 93–105. Stears, M. (1999), ‘Needs, Welfare and the Limits of Associationalism’, Economy and Society, 28 (4), 570–89. Stewart, J., Kendell, E. and Coote, A. (1994), Citizens’ Juries, London: IPPR. Stoker, G. (2006), Why Politics Matters: Making Democracy Work, Basingstoke: Palgrave. Streek, W. and Schmitter, P. C. (eds) (1986), Private Interest Government: Beyond Market and State, London: Sage.

— 253 —

Bibliography Sunstein, C. (1984), ‘Naked Preferences and the Constitution’, Columbia Law Review, 84, 1689–732. Sunstein, C. (1985), ‘Interest Groups in American Public Law’, Stanford Law Review, 38 (1), 29–87. Sunstein, C. (1991), ‘Preferences and Politics’, Philosophy and Public Affairs, 20 (1), 5–26. Sunstein, C. (1996), Legal Reasoning and Political Conflict, New York: Oxford University Press. Szasz, A. (1995), ‘Progress through Mischief: The Social Movement Alternative to Secondary Associations’, in E. O. Wright (ed.), Associations and Democracy, New York: Verso, 148–56. Talmon, J. (1952), The Rise of Totalitarian Democracy, Boston, MA: Beacon Press. Taylor, C. (1985), ‘Atomism’, Philosophical Papers: Volume Two, Cambridge: Cambridge University Press. Taylor, C. (1997), ‘What’s Wrong With Negative Liberty?’, in R. Goodin and P. Pettit (eds), Contemporary Political Philosophy: An Anthology, Oxford: Blackwell, 418–28. Taylor, M. (1996), ‘What are the Key Influences on the Work of Voluntary Agencies?’, in D. Billis and M. Harris (eds), Voluntary Agencies, London: Macmillan, 13–28. Taylor, M. and Warburton, D. (2003), ‘Legitimacy and the Role of the UK Third Sector Organisations in the Policy Process’, Voluntas, 14 (3), 321–38. Thomas, C. W. (2003), ‘Habitat Conservation Planning’, in A. Fung and E. O. Wright (eds), Deepening Democracy Institutional Innovations in Empowered Participatory Governance, London: Verso, 144–72. Thompson, S. and Hoggett, P. (2001), ‘The Emotional Dynamics of Deliberative Democracy’, Policy and Politics, 29 (3), 351–64. Tocqueville, A. de [1835–1840] (1945), Democracy in America, New York: Vintage. Truman, D. (1951), The Governmental Process, New York: Alfred A. Knopf. Urbinati, N. (2000), ‘Representation as Advocacy: A Study of Democratic Deliberation’, Political Theory, 28 (6), 758–86. Uhr, J. (1998), Deliberative Democracy in Australia: The Changing Place of Parliament, Cambridge: Cambridge University Press. Van Deth, J. (1997), ‘Social Involvement and Democratic Practice’, in J. Van Deth (ed.), Private Groups and Public Life: Social Participation, Voluntary Associations and Political Involvement in Representative Democracies, London: Routledge, 1–21. Van De Veer, D. (1986), Paternalistic Intervention, Princeton, NJ: Princeton University Press. Van Hees, M. (2003), ‘Acting Autonomously Versus Not Acting Heteronomously’, Theory and Decision, 54 (4), 337–55. Van Thiel, S. (2001), Quangos: Trends, Causes and Consequences, Aldershot: Ashgate. Verba, S. and Nie, N. (1972), Participation in America: Political Democracy and Social Equality, New York: Harper and Row.

— 254 —

Bibliography Verba, S., Schlozman, K. and Brady, H. (1995), Voice and Equality: Civic Voluntarism in American Politics, Cambridge, MA: Harvard University Press. Von Wright, G. (1963), The Value of Goodness, London: Routledge and Kegan Paul. Wainwright, H. (1994), Arguments for a New Left: Answering the Free-Market Right, Oxford: Blackwell. Waldron, J. (1989), ‘Autonomy and Perfectionism in Raz’s Morality of Freedom’, Southern California Law Review, 62 (3–4), 1097–152. Waldron, J. (1993), Liberal Rights: Collected Papers 1981–1991, Cambridge: Cambridge University Press. Walker, A. (1999), ‘The Third Way for Pensions (by way of Thatcherism and avoiding today’s pensioners)’, Critical Social Policy, 19 (4), 511–27. Wall, S. (1998), Liberalism, Perfectionism and Restraint, Cambridge: Cambridge University Press. Walzer, M. (1984), ‘Liberalism and the Art of Separation’, Political Theory, 12, 315– 30. Walzer, M. (1989), ‘The Communitarian Critique of Liberalism’, Political Theory, 18, (1), 6–23. Walzer, M. (1994), ‘Multiculturalism and Individualism’, Dissent, 41, 185–91. Walzer, M. (1998), ‘On Involuntary Association’, in A. Gutmann (ed.), Freedom of Association, Princeton, NJ: Princeton University Press, 64–74. Walzer, M. (2002), ‘Equality and Civil Society’, in S. Chambers and W. Kymlicka (eds), Alternative Conceptions of Civil Society, Princeton, NJ: Princeton University Press, 34–49. Walzer, M. (2004), Politics and Passion: Towards a More Egalitarian Liberalism, New Haven: Yale University Press. Ware, P. and Todd, M. J. (2001), The Voluntary Sector, Sheffield: Sheffield Hallam University Press. Warren, M. (1996), ‘What Should We Expect from More Democracy?’, Political Theory, 24 (2), 241–70. Warren, M. (1998), ‘What Should Democrats Expect of Associations? The Ambiguities and Complexities of their Democratic Functions’. Paper presented for the American Political Science Association, Boston, 3–6 September. Warren, M. (2001a), Democracy and Association, Princeton, NJ: Princeton University Press. Warren, M. (2002), ‘Deliberative Democracy’, in A. Carter and G. Stokes (eds), Democratic Theory Today, Cambridge: Polity Press, 173–202. Warren, M. R. (2001b), ‘Power and Conflict in Social Capital: Community Organizing and Urban Policy’, in B. Edwards, M. W. Foley and M. Diani (eds), Beyond Tocqueville: Civil Society and the Social Capital Debate in Comparative Perspective, London: University Press of New England, 169–82. Weale, A. (2000), ‘Conversations and Democracy’, Centre for the Study of Democracy: Bulletin, 7 (2), 1–2. Weber, M. (1978), Economy and Society, trans. G. Roth and C. Wittich, Berkeley: University of California Press.

— 255 —

Bibliography Weir, S. (1996), ‘From Strong Government and Quasi-Government to Strong Democracy’, Political Quarterly, 67, 20–38. Whitfield, D. (2002), ‘Afterword’, in M. Todd, P. Ware and G. Taylor (eds), Markets & the Welfare State, Sheffield: Sheffield Hallam University Press, 121–50. Williams, B. (1981), ‘Internal and External Reasons’, in B. Williams, Moral Luck, Cambridge: Cambridge University Press, 101–13. Williams, F. (1989), Social Policy: A Critical Introduction, Cambridge: Polity Press. Williams, F. (2000), ‘Principles of Recognition and Respect in Welfare’, in G. Lewis, S. Gewirtz and J. Clarke (eds), Rethinking Social Policy, London, Sage. Williams, M. (2000), ‘The Uneasy Alliance of Group Representation and Deliberative Democracy’, in W. Kymlicka and W. Norman (eds), Citizenship in Diverse Societies, Oxford: Oxford University Press, 124–52. Wolff, R. P. (1976), In Defence of Anarchism, London: Harper Torchbooks. Wood, R. L. (2001), ‘Political Culture Reconsidered: Insights on Social Capital from an Ethnography of Faith-Based Community Organizing’, in B. Edwards, M. W. Foley and M. Diani (eds), Beyond Tocqueville: Civil Society and the Social Capital Debate in Comparative Perspective, London: University Press of New England, 254–65. 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 E. O. Wright (ed.), Associations and Democracy, New York: Verso, pp. 207–13. Young, I. M. (1996), ‘Communication and the Other: Beyond Deliberative Democracy’, in S. Benhabib (ed.), Democracy and Difference: Contesting the Boundaries of the Political, Princeton, NJ: Princeton University Press, 120–35. Young, I. M. (1997), ‘Difference as a Resource for Democratic Communication’, in J. Bohman, and W. Rehg (eds), Deliberative Democracy, Cambridge, MA: MIT Press, 383–406. Young, I. M. (2000), Inclusion and Democracy, Oxford: Oxford University Press. Young, R. (1986), Personal Autonomy: Beyond Negative and Positive Liberty, London: Croom Helm. Zolo, D. (1992), Democracy and Complexity: A Realist Approach, University Park, PA: Pennsylvannia State University Press.

— 256 —

Index

accountability, 188, 211 agency, 25, 50, 54 agenda setting, 80, 141–3, 144, 173, 174n aggregation, 58, 59, 63–5, 75, 79–80, 83 apathy, 193, 203 Arrow, K., 78–9, 81 associational competition, 147 associational democracy, 1, 5, 6, 7, 103, 130–4, 216–18, 230 dualist model, 137–74, 231, 233 associational discourses, 187 associationalism, limits of, 116–21 associations neo-liberal model, 229, 230 neo-pluralist model, 229, 230 undemocratic, 182–3, 184, 187 vested, 131, 141, 144, 217 see also secondary associations autarchy, 13, 22 authority, normative conception, 50, 54 autonomy, 9–54, 55n, 58, 183–4, 185, 235 and agency, 10, 11–12, 19, 25, 50, 52, 54 vs autarchy, 13, 22 and communitarianism, 33–9, 53 criticisms of, 10, 33–52 deliberative, 4, 22, 52, 53 and democracy, 2–4, 59, 65–78 and free choice, 15, 19–22, 52, 53 hearer, 59, 65, 73–5, 78, 95, 122 individual vs collective, 15, 30–3, 50–1, 95 instrumental value, 12–13

intrinsic value, 11–12, 13, 52 and liberal democracy, 13–19 maximisation of, 13, 48, 50 and minorities in collective decisions, 48–52 and paternalism, 43–8, 54 and perfectionism, 39–42, 53 and preference formation, 218 procedural interpretation, 53, 62 and rationality, 30–3 speaker, 59, 65, 75–8, 95, 122 Barry, B., 16, 39–40, 184 Beetham, D., 189–90 Berlin, I., 20, 43–4, 44 Bevan, A., 167 choices, comprehensive vs peripheral, 31 Christiano, T., 220 Christman, J., 27–8, 29–30, 45 citizens’ juries, 150, 191 citizenship, 94, 129 civic consciousness, 130, 147, 228 civic virtue, 94, 130 civil society, 100, 101–2 co-option by state, 140, 169–72, 174, 190, 234–5 Clarke, P., 24, 34, 35–6, 55n co-option, 140, 169–72, 174, 190, 234–5 cognitive dissonance reduction, 85 Cohen, J., 130, 206–7, 208, 228 collective action, 132 collective will for single power centre, 190

— 257 —

Index common good, 67, 213, 215, 216, 219, 220, 231 communication, dyadic, 71 communicative democracy, 90–2 communitarianism, 33–9, 53, 225 compromise, 67–8, 78, 96n, 221 consensus, 65–8, 80, 95, 180, 219, 221 Cronin, T., 74–5 cultural imperialism, 219 decentralisation, 139, 154, 194 and factionalism, 227–9, 232 and forums, 154–7 and secondary associations, 108, 109, 111, 113, 115 decisions collective vs individual, 54, 58, 68–9, 70, 72 instability in making, 221 majority, 51, 68 and minorities, 48–52 requirements for autonomous, 19–33 transmission of, 140, 156 unanimous, 48, 49 deliberation dyadic, 71–2 evidence-driven vs verdict-driven, 87–8 deliberative democracy, 1–7, 202, 230, 235–6 and aggregation, 58, 59, 63–5, 75, 79–80, 83 agonistic model, 219, 220, 231, 232 and autonomy, 59, 65–78 and compromise, 67–8, 78 and consensus, 65–8, 219 criticisms of, 59–60, 87–90, 99 and cultural specificity, 88–90 definition, 60–5 and deliberative obligations, 92–4, 95 and difference democracy, 86–92, 95 and domain restriction, 82–3, 96n, 97n dualist model, 137–74 and hearer autonomy, 73–5

institutionalisation of, 96 justification, 61–3, 94–5 and language, 88–9 macro vs micro, 138, 148, 173 and prejudice, 88 and public reason, 68–73, 78 and social choice, 78–86, 95 and speaker autonomy, 75–8 and supportive beliefs, 203 ‘two track’ model (Habermas), 138–9, 145, 148 and varied group participation, 87–8 deliberative obligations, 92–4, 98, 129, 130, 146–7, 203, 216, 222, 224, 239 democracy, 1–3 internal, 184 schools of, 128–30 vs revolution, 189–90, 192 see also associational democracy; communicative democracy; deliberative democracy; difference democracy; dualist democracy; liberal democracy D’Entre`ves, M. P., 39, 56n Dewey, J., 25 difference democracy, 86–92, 95 discourses, contestation of, 147, 187, 226 discrimination, 182, 185–6 discursive dilemma (Pettit), 156 diversity, tolerance of, 15 Dryzek, J., 170–2 Dworkin, G., 27, 43, 45, 46 efficacy, 194–5, 211; see also Weberian dilemma elites, 77, 78, 142, 143–4, 181, 189, 191–2, 197–8 Elkin, S. L., 228 Elster, J., 20, 48, 71–2 environmental policy, 123 equality see inequalities exit, 126, 131–2, 136, 178–81, 210 vs voice, 178–81, 210

— 258 —

Index pressure groups, 229 subordinate, 219, 220 Gundersen, A., 71 Gutmann, A., 182–3, 186–7

factionalism, 7, 136, 213–32 and decentralisation, 227–9, 232 and neo-pluralism, 214–15 and representative democracy, 227 Femia, J., 178, 180, 203 Festenstein, M., 92–3, 94, 146, 225 forums and cooperation, 146–7 and corporatism, 149–50 and decentralisation, 154–7 and devolution, 173 and equality, 151, 152 and legal challenges, 153 and lobbying, 152 and manipulation, 151–2 mediating, 148–69, 228, 229, 233 non-partisan deliberative, 150 and quangos, 157–62 and representation, 164–9 and subsidiarity, 174 temporary nature, 161 vs markets, 71, 95 Fraser, N., 120, 145 freedom of association, 125, 177, 182–8 of choice, 10, 15, 19–22, 52, 53, 121 negative, 20, 21, 22, 55n, 126, 186 positive, 20, 22, 43–5, 55n, 126 of voice, 182–5, 186–7

Habermas, J., 37, 61, 69–70, 138–9, 142–3, 144–5 happiness, 55n harm principle, 186 Harrison, R., 50 Hayek, F. A., 123–4, 229 hearer autonomy see autonomy Held, D., 16–17 Hendriks, C., 138, 139 Hirst, P., 116–17, 179, 207–8 Hodgson, G., 123–4 Hume, D., 22–4, 25–6, 55n Hunold, C., 147–8 Hurka, T., 11–12, 52 hypocrisy, civilising force of, 85, 146, 175n

German Social Democratic Party (SDP), 188–9, 211 globalisation, 106–7, 115–16, 135 Goodin, R., 70, 77–8 Gould, C., 50–1, 55–6n government, new roles for, 162–4 Graham, K., 49 Gray, J., 39 greeting, 90–1 groups dominant, 219, 220, 232 excluded, 212n, 220, 232n and identity, 226 interest group competition, 132, 214–15

ideal speech situation 61 identity common vs group affiliations, 224–7 cultural, 225–6 national, 93–4, 225, 226 shared, 94 impartiality, 222 incompetence of masses, 191 inequalities, 132–4, 199, 200–2, 211, 234 of access, 144, 173 in participation, 200–2 and pluralism, 133 of service, 116–17 and social complexity, 105–6 infinite regress, 27–8 information, 74–5, 85–6, 122–4, 191 Inglehart, R., 204, 234 interests conflict of, 220, 221 particular, 218–22 and preferences, 214 private, 221, 232

— 259 —

Index Johnson, J., 5–6 juries, 87 Kant, I., 22–3, 26, 69 Kateb, G., 186 knowledge specialised nature, 190–1 tacit, 124 Kymlicka, W., 36, 37, 41 law, legitimacy of, 14–15 liberal democracy, 13–19 liberty see freedom Lindley, R., 23 Locke, J., 15–16 Lowndes, V., 195 MacCallum, G., 55n Macpherson, C. B., 44–5, 64–5, 163, 200–3, 234 Madison, J., 7, 213–14, 227, 229 majority decisions of, 51, 68 opinion of, 223–4 tyranny of, 215 Manin, B., 224 manipulation, 21, 41, 79–80, 151–2 Mansbridge, J., 187, 205 markets, 71, 95, 124 mass media, 142 Michels, R., 188–91, 192 Mill, J. S., 13, 16, 23, 33, 93, 129 Miller, D., 85, 89–90, 93–4, 225 minorities 67–8, 219, 231 and collective decisions, 48–52 opinions of, 222–4 multiculturalism, 127, 223–4, 231 Mutz, D., 195–6 needs, 116–20 neo-pluralism, 147, 214–15, 217, 229, 230 neutrality, 39–41, 42, 62 of state, 62, 205, 209, 210, 211

objectivism vs subjectivism, 54n oligarchy, iron law of (Michels), 169, 177, 188–92, 211 opinion polls, 150, 191 Parekh, B., 17–18 Parkinson, J., 93 participation, 177–8 C.L.E.A.R. requirements, 195, 211 level of, 87–8, 199, 203, 211 overcoming inequalities in, 200–2 and secondary associations, 177, 192–6, 198–9 strategic, 80, 81, 93 see also Weberian dilemma partisanship, 140, 169, 221, 222 passions, 24 paternalism strong, 43–6, 47, 54 weak, 43, 46–8, 54, 183 Percyzynski, P., 209 perfectionism, 39–42, 53, 56n, 98 mild, 63, 183, 185, 233 type 1, 41 type 2, 41, 42 Pettit, P., 156 pluralism, 104–5, 133, 148–9, 224, 226, 231 political parties, new roles for, 163–4 post-material ethos, 204 preferences, 59, 81, 82–3 formation, 26–8, 30–1, 51, 52, 96n, 97n, 218 input/output filtering, 76–8, 187 and interests, 214 and needs, 118–19 single peaked, 81, 83–4 transformation, 60, 65, 73, 89, 95, 168–9, 173 processes, public vs private, 216–18 public good see common good public policy, 221 public reason, 65, 68–73, 84, 95 public sphere

— 260 —

Index formal, 138, 139, 148–69; see also forums; quangos informal, 125, 133, 138, 139, 140–8, 187, 230; and agenda setting, 141–3, 144, 173, 174n; inequality of access, 144, 173; and opinion formation, 140–1 multiple, 145–6, 173 single, 144 publicity principle (Kant), 69 Putnam, R. D., 129, 130 quangos, 140, 157–62, 174 advantages, 159–61 recruitment to, 159–60 and stakeholders, 158, 159 rational choice theory (Downs), 194 rationality, 10, 45 economic, 24–5, 26 ‘internalist’ and ‘externalist’, 28–30, 33 and Kant and Hume, 22–6 and personal autonomy, 30–3 Rawls, J., 16, 43, 45, 69 Raz, J., 13, 17–18, 19, 20–1, 28, 34, 36, 51, 52, 55n reasons, types of, 23, 32 reciprocity, 69 representation, 121–2, 131, 139, 140, 164–9, 173, 181 republicanism, 216, 218, 219–20, 222–3, 231 rhetoric, 90–1 Riker, W., 8, 79 Rogers, J., 206–7, 208 Rosenblum, N., 226 Roßtuetscher, S., 209 Rousseau, J.-J., 44, 48–9, 202, 218, 222–4, 228 Sandel, M. J., 34–5, 36–7 Sanders, L., 86, 87–8 Saward, M., 63, 64 scale, problem of, 105, 121, 125, 197, 198

Scoccia, D., 31–2, 46–7 secondary associations, 2, 5–6, 213, 214, 220, 226–7, 228, 235–6 character, 101, 102, 125–8, 131–2, 180, 197, 206–7 and coalitions/networks, 226 and decentralisation, 108, 109, 111, 113, 115 definition, 101, 102 and democratisation, 176–212 exit from, 126, 131–2, 136, 178–81, 210 and freedom of association, 177, 182–8 functions, 107–16 and funding, 134 geographically dispersed, 197–8 and increased choice, 125–8 inequalities in, 132–4, 136, 197, 199 and informal public sphere, 125, 133 as locations of governance, 108–16 vs markets, 124 and neutrality of state, 205, 209, 210, 211 participation in, 177, 192–6, 198–9 and problem of scale, 197, 198 and provision of information, 122–4 reform, 208–9 regulation, 178, 204–10 and representation, 121–2 and strategic motivation, 221 and subsidiarity, 109–10, 111–13 seduction, 21, 41 self-interest, 222 service provision, 205–6 Sher, G., 32, 40 social complexity, 100, 103–7, 112–13, 134 and globalisation, 106–7 and inequalities, 105–6 and need for specialism, 106, 135 and pluralism, 104–5 and problems of scale, 105, 121, 125 social movements, 210 speaker autonomy see autonomy

— 261 —

Index state and accumulation, 171 and co-option of civil society, 140, 169– 72, 174, 190, 234–5 legitimacy of, 172, 209–10 neutrality of, 62, 205, 209, 210, 211 see also welfare state state imperatives, 171–2 state regulation, 178, 204–10 Stears, M., 116, 117–20 storytelling, 90–1 subsidiarity, 109–13, 131, 155, 162, 174, 176, 233 Sunstein, C., 82 Swindler, A., 37–8

Voluntary and Community Sector (VCS), United Kingdom, 110–11, 114, 117 voting and information, 74–5 votes vs opinions, 72 voting paradoxes, 96n

Tamir, Y., 208–9 thinking, enlarged, 222 Tocqueville, A. de, 103, 129 transitivity, 79 trust, 129, 130, 224–5, 226, 227, 231

Waldron, J., 18 Wall, S., 17, 31 Warren, M., 178–9, 180, 193, 207 Weberian dilemma (democratic participation vs efficiency), 138, 139, 140, 148, 151, 153, 160, 161, 164, 173 welfare state, and universal provision, 113–14, 116, 117, 209 Western society, nature of, 17 will, general, 223–4, 231 vs minority opinions, 222–4 vs will of all, 48, 57n Williams, B., 23 Wolff, R. P., 48, 49, 50, 65

voice vs exit, 178–81, 210 freedom of, 182–5, 186–7

Young, I. M., 86, 88–9, 90, 186, 220, 234, 235 Young, R., 25–6

— 262 —