Jump to ContentJump to Main Navigation
Towards a Deliberative and Associational Democracy$

Stephen Elstub

Print publication date: 2008

Print ISBN-13: 9780748627394

Published to Edinburgh Scholarship Online: March 2012

DOI: 10.3366/edinburgh/9780748627394.001.0001

Show Summary Details
Page of

PRINTED FROM EDINBURGH SCHOLARSHIP ONLINE (www.edinburgh.universitypressscholarship.com). (c) Copyright Edinburgh University Press, 2022. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in ESO for personal use. Subscriber: null; date: 29 June 2022

Institutionalising Deliberative Democracy through Secondary Associations

Institutionalising Deliberative Democracy through Secondary Associations

(p.98) 3 Institutionalising Deliberative Democracy through Secondary Associations
Towards a Deliberative and Associational Democracy

Stephen Elstub

Edinburgh University Press

Abstract and Keywords

This chapter tries to expand upon, and provide greater detail of, the already-existing literature that connects the ideal of deliberative democracy with secondary associations. It suggests 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, which present significant barriers to deliberative democracy being meaningfully institutionalised. 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.

Keywords:   deliberative democracy, social pluralism, globalisation, associational democracy, autonomy cultivation, decentralised governance


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 (p.99) 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 (p.100) 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 (p.101) 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 Ok 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 (p.102) 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).

(p.103) 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 ‘post-industrial’ 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 (p.104) 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 ever-widening 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 (p.105) 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 (p.106) 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 (p.107) 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 (p.108) 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.

(p.109) 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 (p.110) 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 (p.111) 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)

(p.112) 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. 1. Reduction of Size: Smaller units, such as secondary associations, are more suited than larger units (communities/nations) to both devel oping 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 increaseing social pluralism.

  2. 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’ intervenetion, 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. 3. Reduction of Agenda: Having fewer participants, and fewer issues on the agenda, means less information is rel evant to the decision (Follesdal 1999: 15). As the availability of inform ation and knowledge of available options is a key require ment for autonomousdecisions, subsidiarity can make it easier to att ain relevant information. When the agenda is reduced, narrowed and localised, (p.113) 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 (p.114) 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 (p.115) 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, (p.116) 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, (p.117) 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 manage-rialism, 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 (p.118) 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, (p.119) 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 short-term, 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 (p.120) 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 (p.121) 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 (p.122) 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, (p.123) 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 association-alism, 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; (p.124) 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.

(p.125) 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 decision-making 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 (p.126) 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 (p.127) 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 (p.128) 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 (p.129) 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 nation-state, 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 (p.130) 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-to-face 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 (p.131) 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).

(p.132) 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 (p.133) 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 socio-economic 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 (p.134) 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 macro-economic 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.


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 (p.135) 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 (p.136) 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, delibera-tively democratic, decision-making institutions, in order to ensure both deliberation and democracy are combined, and this shall be addressed in Chapter 4.