Keywords: Deliberating democracy, publicness, rationality, Habermas, Rawls, Internet

Following J.-J. Rousseau, only agreement – unanimity – may provide the complete and perfect legitimacy for a political order. It is based on individualism and human limitations (thus law) at the same time. When all individual wills unite, a true polity can secure the individual and collective good. In Rousseau, democratic rule is based upon the productive fiction of the will of all, the sovereign which is unanimous, enlightened and always already in place. It has no minority vote that is silenced by a majority and therefore may claim rights. The challenge is for the people to know what this idea stands for, what is already known. How do people reach knowledge about the common good? Rousseau’s only answer is not considered viable in a large-scale pluralist society by most observers: Decision must be made by the people itself or by their agent in the knowing of the civic religion of the community (Benhabib 1994, 30). Only in this way are individual wills obeying the common will, in fact him-, or herself. Only in this way is it possible to understand how the individual can obey a common will and still be free. A minority opinion rests then on a misunderstanding or misinterpretation about the general will. A point on which we may see more than reminiscences in communitarian theory is that people are not really asked to say their opinion about a particular proposition, but their opinion on the already determined general will on the proposition. To Rousseau, politics is about making decisions based on the general will that always already exists. (Manin 1987, 348).

Emmanuel Joseph (Abbé) Sieyés, the political theorist of the French revolution, pulled this principle closer to everyday politics, by accepting majority and representation. When unanimity cannot be reached for reasons of scarcity and necessity, it still is important to recognise it in the pluralism of society. This was the lesson of Sieyès, which made him argue for the majority principle, of which he spotted several attributes of unanimity; and therefore could step in for it (Manin 1987, 342). Majority is a practical necessity, which is justified both in its partial equivalence with unanimity and in its practical ability to make decisions. Most moral philosophy can however only justify the majority principle as a second best, due to lack of time, demography and of objective conflict of interests in the population.

This chapter briefly examines the deliberation-democratical attempt to reconcile legitimacy with rational decision-making. Deliberation harmonises the individual and the collective in, not the will of all, but the deliberation of all. I address some prominent deliberative approaches (Rawls and Habermas among others) and qualify them as strong and weak approaches. I then critically discuss the most controversial claim of deliberative theory: its claim to the possibility of impartiality in public debate.

Deliberation of all

Bernard Manin notes that deliberation, in contrast to Rousseau and partly Rawls, is the process by which everyone’s will is formed, that confers its legitimacy on the outcome, rather than the sum of already formed wills. It is necessary to contradict a long tradition, writes Manin, in affirming ‘that legitimate law is the result of general deliberation, and not the expression of the general will’ (Manin 1987, 352). He notes that when individuals approach politics, they do not know what they want. ‘We are justified in taking as a basis for legitimacy not their predetermined will but the process by which they determine their will. This is the process of deliberation. […] Without renouncing a concern for legitimacy, which in the modern world can only be based on the individual, deliberation makes it possible to avoid the exorbitant requirements of universality and unanimity’ (Manin 1987, 364).

Rousseau required that the entire people must authorise legislation. His confidence in reason was secondary to that of preferences or inclinations. Only law sanctioned by the entire people and then exercised equally, can be considered fair, and therefore legitimate. In law the impulsivity and instability of the general will is bounded and transformed into stable and specific law. It is not sufficient to rely on hypothetical criteria on what the people is likely to think, given by a representative assembly. Representation will not do because they would set themselves above the people. There can be no other sovereign than the people. The General Assembly can only consist of the people in person, or as Rousseau states elsewhere, agents who are obligated to vote and act according to the will of the people. Only in this way is it possible to guarantee that legislation reflects the will of the people, and that they are likely to obey the law. This is the principle of double generality (Kaufman 1997, 44). Or we could call them the principles of representative democracy and legitimacy. The law must be enacted by all citizens through the general will, and it must apply to all citizens in the same way. It is in the human nature to conceive of this double principle as just.

Rousseau articulated the paradoxical foundations for public opinion and democracy. Today, Benhabib argues, the answer lies in the processes of collective deliberations, the contemporary stand-in for Rousseau’s general will, and therefore what collective decision-making processes should attempt to approximate. The eloquence of the deliberative model is that it enables rationality and legitimacy in one and same process, because it generates a common moral foundation for both: Habermas argues that ‘The deliberative mode of legislative practice is not intended just to ensure the ethical validity of laws. Rather, one can understand the complex validity claim of legal norms as the claim, on the one hand, to compromise competing interests in a manner compatible with the common good, and, on the other hand, to bring universalistic principles of justice into the horizon of the specific form of life of a particular community’ (Habermas in Benhabib 1996, 25).

Public deliberative procedures

The central idea of deliberation is the idea that providing reasons acceptable to others will tend to enhance normative convergence that grounds and legitimates political, often coercive, decisions. Deliberation is viewed as reconstructive and idealised procedures of public communication where an interchange of reasonable arguments leads to agreement – on some level. Valuations and reasons are in a deliberative view dictated by the public through a public process where the public filters out what issues ought to be the object of political decision-making, and what direction such decisions should take. Decisions are in this way already endorsed in common among members of the public (Pettit 2004, 59). The overall aim of deliberation is to legitimately enable decision-making and minimise conflict.

Deliberation is reconstructive in that it might take place in current democracies, but often in highly imperfect ways and often not at all, due to exogenous causes. Therefore, the principle might be used normatively as a model on which actual democratic procedures might be assessed. Its procedural aspect infers that conditions of communication need to be considered fair, and include free individuals protected by legal rights.

Most importantly, it is public for many reasons; transparency inhibits manipulation and control. Lobbying, for instance, is from a deliberative approach seen as a problem, because of its intransparent nature. Reasonable arguments need to be public in order to be considered attractive and persuasive to others. It acknowledges the necessity for a modern society to make a series of decisions involving the preferences and interests of its citizens, which all the more underline the importance of legitimacy. Public deliberation takes ‘hostages’: it ties individuals to opinions in the early phases of an opinion-formation process that they feel are hard to abandon later on.

Legitimacy in a Rawlsian conception of liberal democracy is a matter of public justification in accordance with the constitution, which all citizens are expected to endorse. To Habermas, legitimacy emerges in several layers of the public sphere and lifeworld-grounded communicative action, and through the principles of discourse ethics, modified from Apel’s universal ethics. In both, public (as opposed to private), reasonable communication is essential, which subsequently lead to reasonable and acceptable decisions. Habermas however, is far more detailed and sociologically inclined in his normative, but also occasionally empirical, interest in the public sphere as the testing bed and sounding board of deliberative democracy. Rawls’ example of a site practising reasonable deliberation – the US Supreme Court – impresses no sociologist.

John Rawls’s concept of ‘public reason’ points to the central principle of legitimacy in a liberal democracy. Justification of principles must take place in a concrete, social process that is political in a wide sense, in addressing politically relevant questions, and yet connected to moral norms. The reasonable way to proceed should be seen as the result of an overlapping consensus in the political culture. Rawls points towards the principle of autonomous public and reasonable communication as the procedure where decisions are legitimated and grounded. The aim is binding agreement, preferably on the basis of the substantial output, or at least on what is at stake, which makes any acceptable output from what is seen as a democratic process legitimate. A less ambitious (but I would say just as important) aim of public deliberation is to avoid or minimise conflict, particularly violence. A common source of conflict is a feeling of exclusion in questions that the group thinks should be heard. Public debate prevents this by opening the floor for all groups and viewpoints.

The principle of impartiality is key in the public deliberative approach to ensure that everyone involved experiences fair and acceptable procedures, and from this, legitimacy. In the deliberation approach, legitimacy is viewed as endogenous consent or acceptance resulting from public reasoning among free and equal citizens. If public justification is allowed to legitimate decisions, the distance between the decision and one’s own preferences cannot be too long. If the end result tends to be suboptimal for many or all in the communication process, Przeworski (in Elster 1998) has argued, legitimation in the form of justification will probably not take place. Ideally, impartiality requires considerable time and space; deliberation and representation in the whole society. Hasty decisions and incomplete representation represent the main obstacles to impartiality (Gargarella, in Elster 1998, 260).

Deliberation is a key to impartiality because it informs all those involved about the essential knowledge and viewpoints in question, and forces all participants to filter out the most self-interested arguments. What Jon Elster calls the ‘civilizing force of hypocrisy’ (Elster 1998, 12) implies that even self-interested speakers will tend to argue in terms of the public interest because of the normative presuppositions in public discourse about necessity of reasoning, if one wants to influence the debate. Adherents of deliberative theory argue that a winning strategy is to emphasise the arguments that presumably will persuade others, and to suppress self-interested claims. Conformism and mainstreaming (the spiral of silence) will also tend to suppress extreme views. Another civilising force relates not to the taming of egocentrism but to the problem of extremism: Public deliberation may bring participants to see their own latent prejudice in a wider perspective, and either opportunistically modify them or become convinced that they are ethically unacceptable.

These arguments of civilising forces that handle psychological opportunism and manipulation are of a less noble character than Habermasian theory about implicit norms of normative rightness in discourse. The vital difference is that Elster’s and other views can without problem be combined with views that tend to stress that public debate rarely enhances consensus. Deliberation is an education process by which exchange of opinions assists the reproduction of a rational and moral base for politics. Lastly, it clarifies reasoning by testing reasons in debate. Particularly this last point makes the deliberative approach argue that contrary to other approaches, deliberation generates not only legitimate output, but also qualitatively better solutions. In this sense, the outcome of deliberation may be seen as a collective good, produced by the procedure that enabled it.

Approaches to deliberation

Today the deliberative approach to democracy represents the dominant paradigm in liberal political theory, and has done so since the mid-1990s. Since then, it has moved from a strong to a weaker (or from a thick to thin) version. In spite of the influential contributions of Rawls and Habermas, a moderate version, which includes empirical considerations on the public sphere, seems to prevail.

Strong deliberative theory emphasised discursive and cognitive processes of justification at various public levels in society that change preferences of those involved, and enhance preferential convergence and consensus – before decisions are finally made by the representative assembly. Habermas thus insisted that there is more to deliberation than fairness and impartiality. However, with the new emphasis on law in Between Facts and Norms, Habermas modified the weight on rational responsibility of the public sphere. In a rational division of labour, the public sphere provides themes and positions, whereas the decisions are left to formal, representative institutions.

The free reasoning among equals that a deliberative interpretation of democracy assumes is, according to Rawls, connected to a model procedure for the individuals involved, and for the society, which would encourage such reasoning procedures. Free reasoning is an ideal setting, which participants and institutions of society hold as a normatively right way to base decisions. Therefore, participants will appeal to considerations at a distance from one’s own desires, that are generally recognised as having weight, and will prioritise arguments which they think others will find acceptable. About why this is the case, the approaches differ. Still, there are few substantial differences between Rawls and Habermas’ more recent conception of the relationship between opinion-formation and will-formation, or between informal and formal political discourse. To be sure, Rawls has been more explicit than Habermas on the concern for stability, and less oriented towards the potential in the public sphere for instituting processes of social change. His reconstruction has been more limited in that it, in spite of its many improbable presuppositions, describes a reality not so far from the current status in what he calls well-ordered societies.

The idea of deliberative democracy, following Rawls, is based on three elements: one is the idea of public reason, the second is constitutional democratic institutions that specify the setting for deliberative legislative bodies, and the third is the knowledge and desire on the part of citizens to follow public reason ‘and to realize its ideal in their political conduct’ (Rawls 1997, 772). Rawls makes clear that Habermas’ concept of the public sphere and his own concept of public reason are not the same (Rawls 1995, 140 note). The public reason of political liberalism is the reasoning of legislators, executives and judges, candidates for public office and party leaders, and others who work in their campaigns, ‘and the reasoning of citizens when they vote on constitutional essentials and matters of basic justice. The ideal of public reason does not have the same requirements in all these cases.’ Habermas’ public sphere is, according to Rawls, much the same as what he called ‘the background culture’ in Political Liberalism. To Rawls, the background culture consists of the civil sphere of churches, associations, the university, and other scientific societies, and all kinds of political meetings. In other words, Habermas’ broader concept of a public sphere includes parts of Rawls’ public reason, most importantly the legislating assemblies. Habermas’ concept also includes weak informal publics, Rawls’ background culture, where notions about the good life thrive, but where public reason cannot be expected.

While Habermas includes the notion of a rational public sphere involving the communication of civil society, journalism and citizens who gather in political dialogue, Rawls, in the interest of securing a rational discussion on what is reasonable, distinguishes between public and non-public uses of reason. Public uses are will-formation in parliamentary, courtly and governmental, and party-related assemblies, and in political campaign debates (Rawls 1993 215-16). Non-public (social) reason is connected to more or less formal networks and associations in the civil sphere. In the domains of non-public uses of reason, reasons refer to particular comprehensive doctrines, whether they are moral, religious, or ideological worldviews. Rawls has no idea here of a rational or even reasonable public discourse among or across such movements or voluntary associations, which could approach one another discursively.

A difference here from Habermas is that Rawls’ domain of public reason is concerned with locating reasonable norms for society as a whole, whereas non-public use of reason in a society of a plurality of doctrines, is concerned about their respective norms and ideals. It is necessary for Rawls to restrict the public uses of reason in order to achieve some plausibility for reaching acceptable norms in an overlapping consensus. Democratic deliberation is reserved for formal political procedures due to the need for stability in a pluralist society. Here the chances are present that reasons are offered that could be endorsed or at least accepted by the other parties, because of their shared conception of justice for the pluralistic society as a whole. These are seen by Rawls as constitutional concerns that must be reserved for delimited deliberating for operating with a certain discipline. They can operate under the limits of how to reason what themes should be addressed, and how to give evidence for various positions. They are however to operate publicly and through discourse with reasons and facts that are seen as widely accepted, in order for the norms to reach a status as public, and that need to be justified as such. While norms are developed in restricted representative fora, responsible citizens accept the limits of the public use of reason.

Rawls’ citizen engages in non-public, social uses of reason, that generate topics and positions for the deliberations that take place in fora of the public uses of reason, and the responsible citizen knows her civil duties and accepts the substantial decisions. In the citizen’s moral duty of civility lies the responsibility to respect a decision that does not necessarily accord with one’s own beliefs and values.

Also, in Rawls, the burden of filtering topics from the non-public uses of reason into the public uses of reason seem to lie to the public uses of reason itself. This seems to him to be the most realistic and feasible solution in order to avoid political paralysis or destabilisation. Whereas Habermas places considerable emphasis on the normative crystallisations of apt political topics and rational arguments in the public sphere, Rawls leaves the transformation or translation from the diversity in society at large into the art of the possible, to politics itself. In practice, both projections see formal politics as exposed to pressures from below; from civil society and the public sphere on issues that require decisions; unjust conditions, signs of crisis will direct themselves to politics; the difference being that this power of the public and civil spheres is fully acknowledged in Habermas’ theory. In Habermas, political decision-making is validated by public and private autonomy in pragmatic, ethical and moral discourses secured by law. To what extent this may actually take place is an empirical question. The normative and reconstructive point in Habermas is that this is how it can and should emerge. Therefore, this is a reconstruction around which we ought to organise society.

Disagreement and struggle, as we know too well, takes place all the time; in Habermas, this fact points towards procedural and institutional imperfections, including social inequalities. But Habermas too would have to recognise that disagreements may be reasonable in that they bring in new ways to see things and reflect real conflict of values or of interest; he has also come to realise that some disagreements among ethical worldviews, particularly religious, may not be resolved discursively. These aspects put Habermas’ position closer to Rawls’ view on political institutions as arenas of reasonable overlapping consensus or compromise of comprehensive doctrines.

Rawls’ version is liberalist and contractual in that it in view of pluralism can settle for only a minimalist yet moral common denominator of institutional ways to distribute resources fairly. Under conditions of groups and citizens with reasonable worldviews, society is able to construct an overlapping consensus on how conflicts and disputes are to be settled in order for all to live together. Liberal theory claims to be neutral with respect to the plural conceptions of a good life, it merely postulates the realistic possibility for such a society to settle for some political principles for legitimate politics. If all parties can agree on some fair terms on how to settle conflicts, the results will be seen as fairly just by the involved reasonable, yet incompatible doctrines in the sense that all will respect the end result. Legitimacyis secured by the fact that all relevant groups, through the overlapping consensus, acknowledge the procedures.

Rawls views the principle of overlapping consensus as more than a Hobbesian modus vivendi: the conception of such consensus as justice or fairness is presented as based on some basic and moral principles of justice, which would have to be accepted by all groups that want to be heard and accepted in society. It is stronger than a modus vivendi, and weaker than a common political culture: Rawls’ regulative idea of justice as fairness is presented as being appropriate to secure both pluralism and stability.

Weaker approaches

A set of views of what can be seen as public deliberation, as grounding for legitimate politics leads to a variety of proposals that concentrate more on broad public deliberation in the public sphere than proposed by Rawls, and other ways to justify the rationality of deliberation than in Habermas’ discursive ethics.

Such a modified or reformed deliberative approach expands the concept of communication to include cultural and ‘disruptive’ forms of expression on the grounds that their underlying message might be of an argumentative or reasonable kind. Such a moderate view on deliberation states that changing of preferences might take place in favourable circumstances, but that what deliberation always does, is to generate a common understanding on the dimension of conflict, thus enabling consensus on what the disagreement is about. In this latter view, deliberation is a sorting and clarifying process, where some positions are transferred to the market or the private sphere, some are postponed, others are possibly agreed on, and finally some remain as a common understanding of the remaining conflict. On this ground a decision may be made that is less susceptible to instability, and more likely to stand as legitimate. Knight and Johnson (1994, 284) argue for a modest deliberative position: ‘Rather a political outcome is legitimate, if at all, because it survives the deliberative process, because it is produced by the sort of reasoned argumentation under fair procedures that defines deliberation as a critic ideal.’ In Jean Cohen’s version, the end result will not necessarily be consensus, but the public reasoning will tend to reduce the diversity of preferences, since they will be shaped underway by the public reasoning (Cohen in Elster 1998, 199). When an argument is considered to appeal to wide-spread considerations, decisions on the basis of it will have a good chance to be considered legitimate by the great majority.

Allen Buchanan states that ‘an entity has political legitimacy if and only if it is morally justified in wielding political power, where to wield political power is to attempt to exercise a monopoly, within a jurisdiction, in the making, application, enforcement of laws.’ (Buchanan 2002, 659–60) And yet, Buchanan’s understanding of moral justification is weak. He argues that the concept of consent is unsuited to the political world. It is extraordinarily demanding and in practice impossible to meet. Any political power that uses its capacity will fall short of the demand to consent simply because of the dominance embedded in any use of power. Consent complements political power; in practice, it would be impossible to combine or integrate the two: ‘If consent is really necessary for political authority then there are not and are never likely to be any entities that possess political authority’ (Buchanan 2002, 699). Few if any political powers such as governments have ever enjoyed consent from the vast majority of the citizens. That is not a serious problem: In political life, consent is neither necessary nor possible to legitimate political authority. Rather, politics is about how to conduct legitimate politics in a context of dissent. To make people obey the law, much is done if citizens feel that the law provides the protection of individuals and human rights, and that individuals have equal access to just institutions. But are we then still talking about moral justification?

The ‘republican’ Philip Pettit argues that democracy is a system for empowering the public reasons and common valuations recognised among people. It can, he argues, be projected as a two-dimensional ideal. The first dimension involves general elections to ensure that initiatives supported by people are identified and implemented. The second dimension is to ensure that only public valuation matters, ‘that whatever initiatives are adopted are justifiable by reasons that are commonly recognized as relevant in the public arena’ (Pettit 2004, 60). If the first dimension is guarded by general elections, the second must be ensured through what Pettit calls institutions that are largely contestatory in character. Individuals and institutions must be in the position to challenge government decisions, arguing that they are not well anchored in the public reasons recognised in the political community, and therefore should be buried. Pettit applies the metaphor of editors to illustrate the role of such individuals and institutions; the need to authorise the policies that the government and representatives have authored, because ultimately they are, together with the authors, the ultimately responsible. The notion of editors seems to be more than a metaphor here, since the mass media, along with the civil sphere organisations, are the chief agenda-setters of the political public sphere, and in the position to undertake investigative reporting and alarm the public sphere by blowing the whistle. What he calls ‘largely contestatory in character’ could in this respect refer to critical journalism that spurs and disciplines critical public debate.

In 1994, Seyla Benhabib defined democratic legitimacy as ‘the belief on the part of the citizens that the major institutions of society and the decisions reached by them on behalf of the public are worthy of being obeyed’ (Benhabib 1994, 27). The definition is close to Weber’s. Legitimacy is people’s belief in the democratic institutions that rule, and nothing is said here about why and how the legitimacy comes about. More surprisingly, coming from a political philosopher, there are no validating criteria for what can be seen as legitimate. Neither does Benhabib’s definition privilege a deliberative understanding over other considerations. However, Benhabib adds that the recognition that the institutions truly are worth obeying can only be fulfilled ‘if such decisions are in principle open to appropriate public processes of deliberation’ (Benhabib 1994, 31).

Benhabib sympathises with the idea of deliberative democracy, and argues that a practical reason can be and actually is articulated in discussion. The deliberative model, following Benhabib, is proceduralist in that it emphasises institutional procedures and practices for attaining decisions on matters that would be binding for all. ‘Proceduralism is a rational answer to persisting value conflicts at the substantive level’ (Benhabib 1994, 34). Further, it recognises conflict of objective interest: ‘Procedures can be regarded as methods for articulating, sifting through and weighting conflicting interests.’ Procedures allow for a multitude of modes of political organisation and modes of articulation, including contestation. ‘It is through the interlocking net of these multiple forms of associations, networks and organizations that an anonymous “public conversation” results.’

Participatory parity

Another deliberalist approach of a weaker kind is presented by Nancy Fraser. In discussions with Axel Honneth, Fraser (2001) has argued that his post-Hegelian way of treating recognition misses the point, and that recognition is a question of justice. Fraser laments the divide between the two positions, arguing that redistribution and recognition – currently seen in the debate during the 1990s about identity politics versus welfare politics – had drifted further away from one another, and the two corresponding positions in moral philosophy, morality and ethics (in Habermas’ terms) consider themselves mutually exclusive. Nevertheless, argues Fraser, there is no real contradiction between them; indeed, they are dependent on each other. Rather than belonging to two different moral categories (morality and ethics), they address different dimensions of society (economy and culture): ‘Justice today requires both redistribution and recognition; neither alone is sufficient’ (Fraser 2001, 22).

But how are they to be reconciled? Nancy Fraser argues that the two positions of redistribution and recognition, morality and ethics, can be integrated by viewing claims for recognition as justice claims, and so make them reachable for moral justification. Contrary to Charles Taylor, Etzioni, Honneth and others, Fraser argues that recognition can be seen as an issue of justice and morality when it is considered as a question of status. Misrecognition is viewed as simply wrong because it denies certain groups the possibility of participation equal to other groups. As a universal norm, what Fraser calls ‘participatory parity’ can be seen as a binding norm if justifying under fair terms interaction under conditions of value pluralism (Fraser 2001, 27). Misrecognition means that certain groups cannot participate fully in society, which clearly is unjust. Fraser notes that the standard of participatory parity can only be applied dialogically and discursively through democratic processes of public debate. ‘In such debates, participants argue about whether existing institutionalized patterns of cultural value impede parity of participation and about whether proposed alternatives would foster it – without exacerbating other disparities. For the status model, then, participatory parity serves as an idiom of public contestation and deliberation about questions of justice’ (Fraser 2001, 41).

More strongly, Fraser adds, participatory parity represents the idiom of public reason, the preferred language for conducting democratic political argumentation on issues of both distribution and recognition. In Fraser’s attempt at treating recognition deontologically as a matter of justice (rather than as a matter of the good life), which can be justified or not in public debate according to standards of participatory parity, she adheres to a deliberation approach that negotiates between different ‘ethical’ positions without the ambition of normative consensus.

Fraser represents a ‘weak’ deliberative approach, which emphasises discussion and contestation without relying on discourse ethics or principles of cognitively generated consensus. Fraser is backed by Zygmunt Bauman and others in this (Bauman 2001, 145). In this model, struggles for recognition and procedural distributive justice form coalitions and campaigns of more temporary and issue-oriented character. Bauman concludes with his own terms that fusing the tasks of distributive justice and the policy of recognition is the meaning of social justice in the present ‘liquid-modernity’ era, while campaign politics compounding the two is its prime and perhaps its sole, available strategy.

The problem of impartiality

Returning to Habermas and Rawls, we have seen that it is assumed that citizens must accept that when they take part in collective will-formation, they must consider the interests of all parties involved. This fact requires all who participate to consider all arguments about public interests, while ignoring arguments about private interests. This application of public reason works impartially, in that only general interests or particular interests that concern a wider collective (women, ethnic minorities, people with physical disabilities, etc.), can be taken into consideration as special cases.

This view sets certain standards and principles for the public sphere. The norm of impartiality is not considered viable by scholars who advocate the politics of contestation and identity recognition. To Iris Young (1990) and others, this detachment from values and desires under the banner of universality, distances politics from what it really is about: to engage in struggles for marginalised groups and particular interests. The deliberative norm of impartiality, Young argues, is really the norm of the dominant groups and interests, and is used against particular groups as informal mechanisms of exclusion. From the norm of neutral public reason, marginalised interests are seen as deviant, and attempts to speak out for them will tend to be ignored. An inclusive public sphere cannot operate with the norm of impartiality but will allow individuals and groups to argue from their particular interests and needs, as they experience them within their status as citizens.

The image of what we may call the public sphere of recognition looks different from the Habermasian public sphere in that it emphasises the aesthetic, affective and expressionist. It is characterised by individuals and groups that individually and collectively seek attention and visibility through various unorthodox means in order to gather support and to confront power in order to effect change for the particular group. Rather than arguing for improvements against the backdrop of the universal interests of society, improvements are claimed on the basis of the living conditions of the actual group. For this reason, the forms of presenting their interests and claims are not necessarily argumentative, but often symbolic, rhetorical, confrontational and demagogic. A politics of recognition is more concerned with symbolism in order to create support for policy initiatives. It also tends to focus on political leaders as symbols of struggles, rather as representatives of the people. Who they are is considered more important than in the deliberative approach, which is more narrowly concerned with how they argue. The focus on the identity of representatives in politics of recognition makes them less concerned with the distinction between public and private, which is considered as a means of domination.

The means of struggle may be drastic: civil disobedience, hunger strikes, violent confrontations. The police typically appear as a symbol of hegemonic power and to confront them occasionally is a point in itself. The primary aim is not to convince, but to attract attention, win support and challenge power. News reports and statements from a political administration that consider themselves impartial is rejected as such, because impartiality is seen as a particular form of power rhetoric. The problem of the public sphere seen from this perspective is not that public debate is biased vis-à-vis marginalised groups. The politics of recognition cannot argue along this way, as can deliberative perspectives. Rather, the problem according to the politics of recognition is that the interests of one’s own group are marginalised, treated badly or considered unworthy by society, or that the interests of the group (in abortion, wearing a turban or head scarf, praying at work, designed working spaces for physically disabled, etc.) are not met by the prevailing political power. In this view, impartiality is part of the problem. The hegemonic emphasis on reason and argumentation tends to leave out interests which are outside the political and cultural mainstream and in possession of little or no formal power to back their views.

A related problem is the point that impartiality is a means of ideology on the part of power that is reserved for the mainstream of society. Impartiality must be respected because it suppresses the tendency of giving our own interests unreasonable focus when weighted against other interests. Does it satisfactorily detect particularities of society? Or does this impartiality blind society from seeing groups and interests at the margin of the ordered, mainstream society? Does it create a bias toward majority, normal, middle class, male, interests? The multitude of incidents in Europe during years with stable social democratic or liberal Governments, more than suggests that large groups feel left out and that the legal/deliberative approach to democracy ignores them. Modern rule can attain its legitimacy without the consent of such groups. Modern societies are increasingly culturally and religiously pluralist, which poses particular challenges for a post-Enlightenment deliberative approach to politics. In other words, what is the deliberative reply to those who do not care to, or are not able to, deliberate? To those who feel that the main sites of the public sphere are closed for them? To those who walk the streets and squat buildings again and again, knowing that the political elite does not listen, but do it all the same.

The deliberalist-oriented reply is that without impartiality, the public sphere loses its democratic function, as it becomes reduced to a battlefield where all parties will lose, particularly minority interests. The question is rather: ‘How can the public sphere be configured so as to ensure that the claims of the marginalised will be heard and receive impartial consideration?’ (McBride 2005, 499). An answer to this would be not to abandon the key principles of equality but to enhance them in order to include greater segments of the citizenry in deliberations, in order to enable the democratic discourses themselves to single out reasonable ways of resolving them. Additionally, the question is how decision-making procedures may be organised in more transparent and accountable ways through a number of democratic reforms like public hearings. In other words, the challenge from the politics of recognition is met with further democratisation of deliberation. The way is not to give special protection to groups, through institutionalising special group representation, group vetoes, etc. suggested by Young. This would only, according to McBride, strengthen counter-deliberative tendencies. In other words, not only deliberation, but also recognition can serve ideological purposes.

However, in many countries strong norms to ensure relatively equal gender representation exist in political institutions (in addition to technical election procedures, which ensure that less-populated areas in the periphery are de facto over-represented in the national assembly.) With regard to the public sphere ‘below and before’ decision-making procedures, McBride argues that if incentives for group representatives are to be the key mechanism for rendering the public sphere inclusive, ‘then the “differences” presented to the public must inevitably undergo significant pre-deliberative filtering and packaging prior to public consumption and may drastically inhibit the expression of dissent from within the groups concerned over and above its impact on deliberation more generally’ (McBride 2005, 500). However this problem can be, and actually is bypassed in some countries by designing a media subsidy structure that favours the opinionated quality press with a relatively small readership. Such a structure, decided through opinion-formation and will-formation in the public sphere and political fora, broadens and opens the public sphere for a more diverse set of interests and claims. Also the special privileges which benefit strong public service broadcasting companies in Europe, is justified on the basis of their constructive contribution to a public sphere. Rather than formal group representation, a middle way is state interventions in the public sphere on the structural level to allow for marginal interests, ‘lifeworlds’ and viewpoints to be heard and discussed.

Towards affirmative pluralism?

For the stronger deliberative approaches, the concept of justice is key, however more as a goal than as a condition for consensus. How can a society reach a common understanding of justice as a basis for just politics? The emphasis is on the procedure protected by law: If everyone accepts the process of deliberation as acceptable, the outcome must be considered legitimate. Political power is guided by the ideals invested in the procedure, and transformed there into generalised opinions. The emphasis is on how regimes of political principles or specific policies are constituted and formed through rational discourses at various levels of society. Politics is considered legitimate through a moral understanding backed in the concept of justice.

The deliberative approach argues that there is a strong link of legitimacy between political decision-making and the constitution of ‘popular will’ in deliberation. Political decisions, it is argued, are seen as legitimate because deliberation simulates a general will in a normatively stronger manner than elections. The deliberative approach tends to locate itself in a paradoxical position by stating its closeness to the Kantian and Rousseauian tradition on the general will; to recognise that legitimacy from a general will, if admirable, is unachievable in a large-scale democracy. The deliberative approach is simply, in Habermas’ words, an intermediate position between the republican-communitarian approach, and the liberalist-aggregative approach.

It is essential in the deliberative approach that not only elites but all citizens (in principle) must be included in deliberation at various levels, irrespective of their position and social status, and also that all available relevant facts and reasons are included in the deliberations. The ideal is broad inclusion of reasonable topics and viewpoints, which deliberation filters into agendas and proposals for political decision-making. Consequently, the challenge is to remove obstacles to inclusion, without turning to formal, positive discrimination mechanisms on a non-deliberative basis, such as instructing political parties to have a certain ratio of gender representation on the lists for general elections, reserving seats for particular minorities in the parliament, or giving particular groups particular veto rights.

The deliberative approach is concerned about the problems of unfair non-recognition, as a consequence of only recognising universal features (‘people’, citizens, etc.), which causes personal and social problems of many sorts. It has however, problems with the politics of recognition position, which argue that groups may demand extended right from arguments of difference rather than from arguments of equality. Only arguing from what defines oneself as individuals and groups, which may be essential to our identity, is met with scepticism on the grounds that the celebration of particularity and ‘authenticity’ is a turning away from society as a rationally governed collective. One should not be concerned about one’s particularities as such, but about one’s rights as a citizen despite particularities. Contrary to the politics of recognition (in contestation-oriented and communitarian guises), the deliberative approach argues that there is no contradiction between respect for particularities and universal reason. McBride argues that the politics of recognition based on recognition as respect as particularly developed by Axel Honneth, can be harmonised with the deliberation approach. Confronted with a politics of egalitarian recognition aiming for respect, political power can work for removing ‘inequalities that unfairly restrict the freedom of persons to seek social endorsement of their identities and projects’ (McBride 2005, 502). And this, I may add, is a reconstructive view, because this is precisely what took place during the previous century: hierarchies were flattened, the notion of citizen was dramatically expanded, and the public sphere included new segments of the population. Within the frames of capitalism European society conceived of itself as a more open, tolerant and egalitarian, pulling in new groups and visualising discrimination, poverty and abuse. The effect of visualisation, facts and debate has together yielded growing tolerance of minority needs. The argument has not been that groups were to be endorsed because of their particularity (as ‘travelling people’, gays or mentally ill) but because everyone is entitled to live according to their rights as citizens.

It is worth noting the mutual criticisms of totalising tendencies that have been raised over the fence between varieties of politics of recognition and the deliberative approach. Charles Taylor has argued that the prevalent Enlightenment approach of democracy potentially underlines totalitarian tendencies in their emphasis on the universal and undifferentiated nature of democracy, because it may serve to suppress particularities like minorities. Conversely, McBride suggests that what she calls the ‘politics of authentic recognition’ (Taylor) risks pursuing a totalising agenda by putting so much emphasis on securing others’ agreement of one’s own values, ‘in the guise of promoting pluralism’ (McBride 2005, 503). This mutual suspicion only reflects the inherent tension in democracy between minority and majority interests, that cannot be dissolved or resolved in either of the two approaches. More important however, is that both perspectives are actually operative in current societies and particularly in the public sphere, as complements and correctives, which no doubt seem to add a further argument in favour of the deliberative approach, since the struggle for recognition as respect can only present claims within the discursive frame of deliberation and decision-making.

The weight of the argument derives not only from the argumentative power but more realistically from resources available and how much support the argument – carried by an organisation or a movement – can mobilise in civil society and in the public sphere. Reason is backed by strategic power. In so far as such power is translated into votes or forms of numerical support in the public sphere (demonstrations, petitions), they are highly legitimate, Jane Mansbridge has argued in many contexts, for a normative understanding of democracy. No society and no politics exist without coercion. The ideal of coercion-free communication works as a dream to be pursued, perhaps even unavoidably because ideals are inherent in commonality. Even if it is impossible to reach, it could be used as a moral guiding light. That is however Habermas’ argument; to him the mission of political theory is hardly to produce illusions.

The sociological fact remains: no society has ever produced equality to the extent that differences of interest have been rendered irrelevant for the development of society. Social relations, particularly of some scale, are power relationships, and politics is, at best, about legitimate power, Herrschaft or rule. More than ever this is so in modern capitalist crisis-ridden societies. Legitimacy cannot be derived from a situation free from conflict, but neither can it be built upon conflict alone, since something must prevent the conflict from escalating. Jane Mansbridge is probably correct in concluding that, ‘In practice, actual democracies can produce at best a “rough” or “good enough” legitimacy, based on citizens’ generalisable interests in creating conditions of relatively willing cooperation (Mansbridge in Benhabib 1996, 54). They will always create only institutions ‘that are reasonably just in view of the circumstances’ or, that do not ‘exceed the limits of tolerable justice’.

Additionally, the willingness to accept certain forms of coercion probably tends to derive from a tacit and conventional consensus embedded in cultural values. We should also add another most important point: In contemporary Western Europe, people are aware of the brutal political conditions in regions such as northern Africa and the Middle East, and they know that political stability and a ‘liberalism of fear’ are important values in themselves and function as a legitimate condition. This also relates to the legitimacy of legality.

However, for Mansbridge and others close to an ‘agonistic’ approach, the procedurally fair and substantially just will remain as normative convictions for participants in the struggles, as will the almost unavoidable disappointments and frustrations that lead to new episodes in the never-ending story of social struggle. The perceived or experienced imperfections in all democratic procedures occasionally produce protests, strikes and sometimes civil disobedience, instantiated by formal but normally legitimate mechanisms. Nevertheless, legal and political settlements and compromises will tend to produce conflict. As the sociology of law tells us: Legal processes may reduce and enlarge conflicts.

Rather than simply emphasizing the importance of the public sphere, conflict-oriented or ‘contestative’ views point to the importance of a wide variety of social institutions that organise the disadvantaged, the different and the weak, like collectives, churches, interest groups, and cultural communities, where members see their interests and values protected or articulated in a deliberative manner. Some of these ‘counterpublics’ may take universalistic, for instance environmental, standpoints, while others may take local and delimited positions to specific problems. In some instances, as was the case in Istanbul in summer of 2013, a local modernisation issue evolved into a universalistic demand for a liberal democracy. The discourses may be of various cultural kinds, with little similarities to the public reason presented in quality newspapers. The aim may realistically not even be to win the battle, but to forge bonds of solidarity, reorient themselves, create new strategies and plan ahead (Mansbridge in Benhabib 1996, 58). She argues: ‘in the polity we can best preserve these irritants by encouraging ‘affirmative pluralism’ in democratic talk. In our democracies, we need to find ways of removing coercion as much as possible from the arenas in which we struggle to understand what is just and unjust. This means a public discourse not completely overwhelmed with the massive resources of existing forms of domination’ (Mansbridge in Benhabib 1996, 59).

Conclusion

What has been called ‘foundationalist’ theory, theory with the ambition to systematise and reduce morality to some principles and ideas, cannot adequately describe and sociologically appreciate substantial, personal and moral sentiments. A moral theory of the public sphere has the unrealistic ambition to formulate moral ideas for all domains or sectors of social life, which seems unrealistic and runs counter to at least some central social theories of differentiation. One could perhaps even argue that it runs counter to Habermas’ own use of systems theory. It makes more sense to address politics as historical products of a particular modern constellation, and to focus on consistency, institutional practices and their effects, the political development and use of morality, ways to formulate and see things, the application of power, responsibility, etc. We are concerned with the actual society. The sociological non-normative question is when and how citizens, often in one or another aggregated or synthesised version, in fact consider political power as legitimate in relation to a particular issue, a particular regime, or in relation to a political system. The guiding question is, What is going on? How has what is considered reasonable or natural come to be seen like that? As sociology often does, we may construct our categories and types from actual practices we see unfolding. A critique of this is perfectly possible without normative ideals as a beacon or standard. I do not deny that basic political ideas are discussed in principal and moral terms, but the open and broad public sphere cannot be seen as impartial and procedural. The emphasis on a rational public sphere may have led to a displacement of politics in studies in the formation of public opinion. Since it tends to downplay the political mechanisms that seek to produce legitimacy from other norms, for instance from what is necessary or inevitable, it looks at people as citizens and as the sovereign body, which in practical politics it is not.

Rawls defines reasonable as those who accept liberal procedures, in other words political liberalism. Those who do oppose the liberal principles are in this way excluded from what is considered political in the liberal view, or relegated to the private sphere, because they unacceptably jeopardise the dominant liberal principles in the public sphere. Still political liberalism insists on its neutrality with respect to substance. That a certain minimal acceptance of political principles must be in place in order to take part in democratic discourse is evident, but this is also a political point of view, an argument for a distinct political perspective: political liberalism. What is unreasonable, viewed from Rawls’ perspective of political liberalism, may be fairly reasonable from another perspective. Rawls’ political liberalism cannot come across as anything but a particular model of political democracy, as the main features of a particular regime. Rawls gets caught, following Chantal Mouffe, in a circular argument: ‘political liberalism can provide a consensus among reasonable persons who by definition are persons who accept the principles of political liberalism (Mouffe in Benhabib 1996, 250). Mouffe argues that this is a logical conclusion in political liberalism because ‘antagonistic principles of legitimacy cannot coexist within the same political association without putting in question the political reality of the state. However, to be properly formulated, such a thesis calls for a theoretical framework that asserts that the political is always constitutive – which is precisely what liberalism denies’ (Mouffe in Benhabib 1996, 250).