Beyond Ego and Alter: Enlarged Democratic Deliberation

According to a frequent objection coming from the tradition of political realism, deliberative democracy is impotent in the face of actors who, wielding power and money, refuse to engage in deliberation, or seek to distort deliberative processes. With the aim of disproving this objection, in this essay I proceed in three steps: first of all, I show that the realpolitik objection is based on a dyadic, two-person theoretical model of argumentative speech acts. To this model, considered limited and unsatisfactory by many sociolinguists, I counter a more complex and articulated framework. Second, I aim to demonstrate that this latter framework is capable of accounting for a temporally and spatially enlarged democratic deliberation which can be rejected or distorted barely, if at all, by agents relying on positions of power. In the third section, I highlight the many and important differences in grounding, nature and finalities between the model of enlarged democratic deliberation and forms of power politics based on strategic calculations and tactical alliances. Finally, I focus on the application of the model to societies characterized by structural injustices and distortions, with the aim of showing how it can help marginalized and victimized groups have their requests heard and discussed in the public sphere and in deliberative settings.


Introduction
This article addresses a specific objection to deliberative democracy: this objection, coming mainly from the tradition of political realism, argues that deliberative democracy is a naive project, impotent in the face of the dynamics of power prevailing in real politics and totally ineffectual in the face of powerful agents who refuse to participate in deliberative discussions, to give reasons for their actions or to listen to others. In this essay I intend to show that this objection does not necessarily hold true, and suggest how deliberative democracy can increase its effectiveness in the face of actors who, by relying on power or money, aim to reject democratic deliberation or distort its outcomes. As I argue in (1), the realpolitik objection rests on a dyadic, two-person model of public discourse, which is partial as a theoretical framework and limiting as a practical one. This model has been subjected to a great deal of criticism in the field of sociolinguistics; its descriptive scope is believed to be poor and its depiction of real discursive interactions oversimplifying. In political philosophy, however, and in particular in academic discussions concerning deliberative democracy, the dyadic model is still dominant. Therefore, with recourse to contributions from sociolinguistics, I suggest understanding deliberation on the basis of a model which goes beyond the dyadic one: I call this the enlarged theoretical model of discursive interaction. In section (2), I show that this latter model can account for a wide range of possibilities, on the part of the practical advocates of deliberative democracy, to advance deliberation even when powerful agents refuse to engage in deliberative discussions as equals. On the practical level it is possible, as I aim to show, to achieve a spatially and temporally enlarged democratic deliberation; that is, a deliberation expanded to encompass different moments, forums and actors. This account of deliberation makes it difficult, if not impossible, for the wealthy and the powerful to reject deliberation by relying on privilege: as I intend to show, 'the unforced force of the better argument' (Habermas 1998: 37) is not powerless in the face of power. In section (3) I point out the radical differences between the model of enlarged democratic deliberation and forms of power politics based on strategic calculations and tactical alliances: the account I propose has recourse to the force of big numbers, but these are obtained through and for a free, democratic deliberation, in which the force of the better argument prevails. In (4), finally, I focus on the application of the model of enlarged democratic deliberation in societies characterized by deep structural injustices: I aim to show that it can help marginalized, victimized and disadvantaged groups to make their requests heard in the public sphere and deliberative settings, contributing thereby to undermining structural inequalities and distortions.

1.
According to an objection frequently addressed to deliberative democracy by political realists, democratic deliberation is a naive pipe-dream, unable to speak to the reality of politics, understood as a site of struggle between self-interested actors competing for power (see Shapiro 1999: 29). 'Politics is about interests and power', as the subheading of Shapiro (1999) reads; 'powerful players who make it their business to shape the terms of public debate through the financial contributions they make available to politicians and political campaigns' have easy game in distorting the outcomes of deliberation (Shapiro 1999: 34; see also Shapiro 2017). Those who hold money and power can easily avoid giving reasons for their actions and confronting other people's arguments in deliberative settings, so that, in the face of wealth and privilege, the unforced force of the better argument is impotent: powerful agents often 'reject discussing an issue with their opponents, do not take them seriously, and refuse to listen to them, […] deny their opponents (equal) access to the debate and interrupt them, point to their rank or status or break off discussions arbitrarily and unilaterally' (Schimmelfennig 2003: 204). Deliberative democracy, according to its realist critics, ignores the dynamics of unequal power prevailing in real life and, in assuming from everyone 'rationality, mutual consideration, and the patient exchange of publicly justified reasons for supporting specific policies', shows 'unrealistic expectations about human nature' (Achen & Bartels 2016: 303-304).
Many deliberative theorists recognize some truth in these objections: for Carolyn Hendriks, actors with vested interests only support processes of citizen deliberation when they know they can gain an instrumental advantage from them (see Hendriks 2006). For Iris Marion Young, every activist for justice knows only too well that 'the powerful officials have no motive to sit down with him, and even if they did agree to deliberate, they would have the power unfairly to steer the course of the discussion' (Young 2001: 673). Archon Fung maintains that, more often than not, 'more powerful actors are unwilling to deliberate' and 'seldom engage weak ones with a willingness to constrain themselves according to the norms of deliberation' (Fung 2005: 408). Furthermore, ' economic inequalities […] enable wealthier parties to improperly displace communicative power by mounting threats, purchasing compliance, drowning out other perspectives, mobilizing many forms of support, or simply privatizing some area of concern out of the domain of public deliberation' (Fung 2005: 413). From these objections, realist theorists claim the futility of the project of deliberative democracy, while radical proponents of deliberative democracy conclude that, in the face of naked power and privilege, deliberation should give way to non-deliberative forms of activism (this, at least, is what Iris Marion Young makes her activist character say in Young 2001: 673). In any case, it seems that deliberative democracy cannot but surrender in the face of powerful parties who intend to make strategic use of it or do not want to listen to others.
Both realistic objections to deliberative democracy and theories of deliberation are generally based on a two-person, or dyadic, model: even when deliberation takes place between several participants, it is usually comprehended according to the dyad of a speaker and an addressee who alternate in their roles. The traditional twoperson model is well described by Erving Goffman: Traditional analysis of saying and what gets said seems tacitly committed to the following paradigm: Two and only two individuals are engaged together in it.
[…] The full concern of the person speaking is given over to speaking and to its reception, the concern of the person listening to what is being said. The discourse, then, would be the main involvement of both of them.
[…] Over the course of the interaction, the roles of speaker and hearer will be interchanged in support of a statement-reply format, the acknowledged current-speaking right -the floor -passing back and forth.
[…] It is felt that without requiring a basic change in the terms of the analysis, any modification of conditions can be handled: additional participants can be added, the ensemble can be situated in the immediate presence of nonparticipants, and so forth (Goffman 1981: 129).
In a number of studies in the field of sociolinguistics, as well as in accounts by many theorists of deliberative democracy and in the objections by their opponents, it is often Ego and Alter, or A and B, or the two partners of discussion, who confront each other, even when they are not the only subjects engaged in the interaction (see Clark & Carlson 1982;Levinson 1988;Plot 2009). In other words, deliberation is often understood as an interaction taking place only between those, among the participants, who interchangeably assume in the discussion the role of speaker and addressee. This, of course, does not mean that the deliberation only includes two people: a democratic deliberation can bring together hundreds of participants, who can address their arguments and counter-arguments to individual or collective subjects (i.e. groups or categories of people). However, only two roles are relevant for the dyadic model and, sometimes, also for the active participants in deliberations: the interchangeable roles of speaker and addressee. Whoever decides to intervene in a deliberation having been confronted with objections or demands to provide reasons, usually responds to these objections by assuming, as his/her only interlocutor and listener, the individual or group that addressed him/her; he/she, in other words, acts as a speaker directing his/ her utterances towards a specific individual or collective addressee, on which he/she focuses all his/her attention. In turn, those who had previously played the role of speakers had in mind a specific individual or group as a recipient for their utterances, on which, and on no-one else, they were focusing their attention. Any other people who are present in the discussion, but do not participate directly, tend to slip out of the picture. Hearers are not quite considered as actual participants in the deliberation, neither in the theoretical model, nor, all too often, by those who actively intervene in a public discussion. This two-person framework is reflected also in the previously considered objection to deliberative democracy: for the realist critics of deliberative democracy, the fact that powerful or wealthy actors are often unwilling to engage seriously in democratic discussion or to listen to the arguments of others marks the impracticability of deliberation in real politics. The common saying according to which 'it takes two to talk' seems to condemn deliberation to impotence in all those cases in which the most powerful agents are unwilling to listen or participate, or simply reiterate their position without taking into account the reasons of others.
The dyadic model, however, cannot account for the complexity of discursive interactions. This point is diffusedly maintained in the field of sociolinguistics. According to Levinson, 'the assumption of dyadic verbal interchanges as the basic (or sole) form of human communication' leads to 'serious mistakes, or at least oversimplifications' (Levinson 1988: 163-164); Dell Hymes goes as far as to claim that ' even if such a scheme is intended to be a model, for descriptive work it cannot be' (Hymes 1972: 58). For Erving Goffman, the dyadic framework is 'too gross', as 'it takes global folk categories (like speaker and hearer) for granted, instead of decomposing them into […] analytically coherent elements' (Goffman 1981: 129), while Lewiński insists on the oversimplifying character of the model (Lewiński 2013: 161). 'Speakers perform illocutionary acts not only toward addressees, but also toward certain other hearers' (Clark & Carlson 1982: 333): what we say to someone, we often intend to be heard also by other people present in the same room. Similarly, in a deliberative context, enumerating in front of an interlocutor his or her unfulfilled responsibilities may have the purpose of recalling them to him or her, as well as communicating them to other hearers.
In reality, deliberation never takes place only between speaker(s) and addressee(s): when the participants in a deliberation respond to each other's statements, they are not alone; there are hearers listening to them. Hearers do not intervene directly in the discussion, but keep track of what others say, relate what they hear to possible past deliberations and draw assessments about the coherence and accountability of those participating in the front line. On the basis of these reflections, they can at some point take sides or develop their own positions. In a deliberation, therefore, one never speaks only in front of the addressee(s) and for the addressee(s); reasoning in these terms, understanding one's own actions or those of others according to the two-person model, is limiting and self-defeating. The point, within sociolinguistic studies, is well expressed by Michael Warner: it is remarkable how little work in even the most sophisticated forms of theory has been able to disentangle public discourse from its self-understanding as conversation: in addressing a public, however, even texts of the most rigorously argumentative and dialogic genres also address onlookers, not just parties to argument. […] The agonistic interlocutor is coupled with passive interlocutors, known enemies with indifferent strangers, parties present to a dialogue situation with parties whose textual location might be in other genres or scenes of circulation entirely. The meaning of any utterance depends on what is known and anticipated from all these different quarters.
[…] The interactive relation postulated in public discourse, in other words, goes far beyond the scale of conversation or discussion to encompass a multigeneric lifeworld organized not just by a relational axis of utterance and response but by potentially infinite axes of citation and characterization (Warner 2002: 420-421).
In a deliberation, therefore, every position must be reflexive: participants must understand themselves as nodes within a field of interactions extended in time and space, in which every utterance is not only heard by the addressee, but also by every other actual or potential hearer. Deliberation can never be assimilated to a twoperson exchange which can proceed and bring results only if both partners (be they individuals or groups) continue their conversation with dedication while aiming at mutual understanding: in the absence of a reciprocal willingness to engage in an exchange of reasons, there are, at any time, other current or potential interlocutors to which the practical advocates of democratic deliberation can turn. The actors who refuse to engage in deliberation never do so only in front of those who called them out on their responsibilities (and who, according to the twoperson model, would at this point remain powerless, as it is impossible to discuss with someone who will not listen or even show up). They also do so in front of all the people actually or potentially interested in the problem at issue or in the advancement of democratic deliberation at large. The proponents of deliberation can therefore turn to all these other actors in order to make their normative demands known, to expose the conduct of the original addressees, and, thereby, to raise the costs of rejecting reason and democracy: 'parties who are initially outsiders to a two-party dispute can display alignment to particular positions within the dispute' (Goodwin & Goodwin 1990: 101), and ' can dispute a particular position for different reasons and by different means' (Maynard 1986: 281).

2.
In the previous paragraph, I have considered a well-known objection directed to deliberative democracy by the tradition of political realism. As an alternative to the dyadic model of discursive interaction on which this objection is based, I have proposed a framework, based on sociolinguistics, that takes into account other participants in democratic deliberation beyond speaker and addressee. I call this the enlarged theoretical model of democratic deliberation. In the present paragraph I develop the practical implications that, for deliberative democracy, derive from assuming the enlarged model instead of the dyadic one.
When one reasons in terms of the enlarged model, it becomes clear that deliberative democracy is not forced to declare defeat in front of powerful agents unwilling to listen to others or to give reasons for their actions.
Democratic deliberators are not the hostages of the (unlikely) willingness of the privileged to engage in deliberation as equals. Whenever the original addressee rejects deliberation, or tries to distort its outcomes, deliberation can continue with other interlocutors, i.e. with all those who, up to that moment, have listened, kept track of what was happening, and in front of whose eyes has taken place the refusal, by the original addressee, to make good arguments count instead of power. The potential interlocutors are more numerous still: they encompass everyone who is (or can become) interested in the problem being debated, who can ask questions and demand reasons for the different positions on the table, and then perhaps decide, on the basis of the arguments that will be presented to him/her, to give or deny his/her support. Deliberation should be comprehended, therefore, not in a dyadic, but rather in a spatially and temporally enlarged sense.
The model of enlarged democratic deliberation I propose shares some elements with the theory of democratic systems (on the latter, see Mansbridge et al. 2012): both see deliberative democracy as made up of several interconnected fora, as a complex unit whose parts may have relationships of complementarity or displacement encompassing different agents, moments, spaces and modalities (Mansbridge et al. 2012: 3). The model of enlarged democratic deliberation applies to a specific case that may arise, and indeed often does, within a deliberative system: the instance where deliberation is blocked by the refusal of one or more powerful actors to deliberate. Based on theoretical reflections in the field of sociolinguistics, the model of enlarged democratic deliberation constitutes a practical-normative account devised for a specific case and compatible with the theory of deliberative systems (seen as ' an overarching approach to deliberation' and not as a 'free-standing theory of deliberative democracy': see Mansbridge et al. 2012: 4) as well as with other accounts of deliberative democracy.
According to the model of enlarged democratic deliberation, the practical advocates of deliberation should maintain a reflexive awareness of the non-dual character of interaction: it is also the hearers, present or potential, who should be taken into account during a deliberation and who should be considered, from the start, as co-addressees. Even though little attention is usually paid to it, listening is a practice integral to deliberative democracy, as deliberating presupposes that all points of view made with respect to an issue have been carefully listened to. Listening, in the context of deliberation, should be understood according to what Leonard Waks calls ' apophatic listening' (Waks 2010: 2749, quoted in Dobson 2014: the practice of trying to hear and understand our interlocutors in their own terms, without immediately labelling their experiences and perspectives under our pre-existing categories, as doing this (which Waks calls ' cataphatic listening', see Waks 2010: 2743, quoted in Dobson 2014 would amount to imposing our perspective on the others and would prevent us from learning something genuinely new. At the same time, however, the suspension of our own categories is only temporary: listening in and for deliberation does not mean obliterating our own judgments and feelings to the point of making our perspective collapse into the other's (Dobson 2014: 64): in deliberation, listening means neither to reduce others to us, nor to reduce us to others, but to try and understand others while maintaining the distance necessary for an actual dialogue.
It is therefore to the hearers, and not only to the addressees, that whoever intervenes in a deliberation should, from the very first moment, try to convey his/her arguments and demands, in forms that can be comprehended by people positioned very differently on the social spectrum. When someone rejects deliberation by relying on positions of power and consolidated privilege, deliberative democrats will have to address themselves, more than ever, to these potential supporters, highlighting how the original addressee has backed out of all demands for reason-giving and how this implies an imposition of power over justified arguments. The proponents of deliberation should aim to seek support and allies through deliberative means: for example, by organizing new meetings and deliberations to present their position, summarizing relevant past events in order to inform new participants, answering objections and explaining their reasons to different publics. Deliberation can therefore continue beyond the original forum, the original addressee and the original temporal occasion, in order for it, and for justified normative demands, to gain public support through the force of good arguments. In contrast to the objections of political realists, when powerful actors reject deliberation, deliberation itself has only just begun.
The efforts of the proponents of democratic deliberation must be targeted at advancing and resuming deliberation, as this is the context in which, more than in any other, good reasons can count instead of the imposition of power. The practical advocates of deliberative democracy have, therefore, to mobilize -through deliberative means -enough public support to persuade agents wielding power or money to (re-)engage in practices of deliberation (Fung 2005;Hendriks 2006). Ideally, this process should aim to change their antideliberative convictions, although powerful actors will probably be more sensitive to strategic factors, such as the foreseeable worsening of their public image if they continue to reject deliberation (Fung 2005: 399). To assert the unforced force of the best argument in the face of naked power, in short, the force of big numbers may sometimes be necessary, but even the public support needed to convince powerful parties to resume deliberation can derive from deliberation itself and from the force of good reasons: it is through deliberative modalities that, in the account proposed here, the force of good reasons can lead to the power of big numbers.
In order for a spatially and temporally enlarged deliberation to be possible, and for it to counter the imposition of power and privilege over the exchange of reasons, in democratic deliberations hearers should be taken into account from the very beginning: one never speaks only in front of his/her addressee and only for him/her. On the basis of this reflexive awareness of the non-dyadic character of discursive interactions in deliberative settings, every speaker should try to explain his/her positions in a way that all participants can understand, and try to make his/her actions and statements justifiable to everyone. The proponents of deliberation, moreover, should make sure that all interested parties, both in practical deliberations and in society at large, have the possibility to be informed about the matters of discussion, and should explain the steps that from their perspective have led to the current situation. They should also clarify for all, in easily understandable language, every possible technical or complex issue concerning the matter in question, which, if not made comprehensible, could drive potential interested parties away. Discussions and thematic meetings should be organized by the proponents of deliberation to explain their reasons and answer doubts and objections from the public. In doing so, it is essential that deliberators always maintain an open-minded attitude to alternative ideas and perspectives, and remain willing to possibly change their mind. As claimed by Robert B. Talisse, deliberation is problematic for anyone who 'takes himself to know what justice is and what justice requires' (Talisse 2005: 428); the epistemic and transformative value of deliberation, in order to unfold, requires open-mindedness on the part of all participants.
Different channels, including the internet, could also be used by the proponents of deliberation to circulate their claims, to discuss the issues on the table and to distribute information (in this regard, see the experience of the EJ Working Group narrated in Dodge 2009). Faced with the need to gain public support in the face of the unwillingness of powerful actors to deliberate, the deliberators could also try to mobilize media attention (Fung 2005: 403): in this, they should aim to explain their positions and to make the general public aware that a certain issue is being discussed, highlighting its implications for justice and its possible consequences for other groups. They should explain why even people not directly involved in the problem at issue -but having a sense of justice -should be interested in it, and highlight how the matter of discussion may have wider implications, or be of interest to more people or groups than perceivable at first sight. In the process of gaining public support, the choices regarding the actions to be taken might best be decided through deliberation: democratic deliberation has the power not only to bring about solutions acceptable to all those involved, but also to facilitate the ideation of actions not yet undertaken. As an example of this, Jane Mansbridge narrates how teachins were born in a democratic deliberation among forty-six members of the University of Michigan, who had gathered to decide how to protest against the war in Vietnam (see Mansbridge 2010: 74).
One problem which, on the whole, affects deliberative democracy, and therefore also the model of extended democratic deliberation, is the reluctance of some citizens to participate in deliberative meetings. The main reasons for their refusal to be involved are, according to recent studies, their concentration on the private sphere, a perceived lack of competence in political matters, public meeting avoidance, conflicts of schedule, political alienation and a perceived lack of impact of the deliberative processes on the political system (Jacquet 2017: 9). However, the model of enlarged democratic deliberation, due to its very nature, presents a vantage point for mitigating this difficulty. When an issue begins to be widely debated in a community, collective attention arises around it and a large public turns directly to decisionmakers to be heard, a snowball effect is set in motion: the interest and the willingness to participate, even on the part of quite indifferent and reluctant citizens, also increases (Oliver, Marwell & Teixeir 1985;Curato & Niemeyer 2013;Pulkkinen 2013). More and more citizens realize that what is happening could well bring about concrete changes for the future and, in consequence, decide to get informed and, possibly, involved. The model of enlarged democratic deliberation implies precisely the creation of such public engagement as can lead even initially reluctant citizens to want to learn more about the topic at issue and join the discussion in deliberative contexts. Of course, there is no guarantee that this will always happen; success in this endeavour depends on many factors, such as the context, the issue being discussed, and also the ability of the proponents of deliberation to foster people's commitment and leverage their sense of justice.

3.
As we have seen in (1) and then, on a practical level, in (2), finding an alternative framework to the dyadic one allows room for a temporally and spatially enlarged model of deliberation. Through the proposed framework, it becomes possible to confront, by deliberative means, those who could otherwise easily disengage themselves from communicative demands for reason-giving. In this section, we specify the many important differences between the model of enlarged democratic deliberation and forms of strategic political action purely based on power (including the power of big numbers). Realist theorists, on the whole, see politics as a site of struggle between self-interested actors competing for power, in which what counts are strategic power games, the ability to form alliances for one's own interest and to prevail over an ' other' who is seen as an adversary or an enemy (see Schmidt, 2005). If necessary or convenient, the victory over one's adversary can take place thanks to the power of big numbers, i.e. by mobilizing, for strategic purposes, significant public pressure. Such power of big numbers is extremely different, in its foundations, nature and objectives, from the power of big numbers to which the model of enlarged democratic deliberation aims at making recourse. In the latter: 1. the force of big numbers is obtained through good reasons and deliberation: all those who give their support to the deliberators in order to pressure the original anti-deliberative actor, do this on the basis of arguments of justice that they have found convincing; they have discussed these arguments with the proponents of deliberation, have perhaps also objected to them and received satisfactory answers, and have finally deemed the deliberators' cause worthy of practical support. People are motivated to pressure the powerful who refuse to deliberate, not by strategic interest, but by arguments of justice discussed through deliberative means. 2. the power of big numbers is aimed at the prevailing of good reasons and deliberation, both with regard to the specific case at issue and over the long term, in society at large: the pressure on the anti-deliberative actor is not exerted in order to conquer him in a power struggle, but to the purpose of resuming a deliberation in which the better argument (and not force) will prevail. The ultimate goal of the deliberators and those who support them, moreover, is an overall change in society. In the long run, they aim to transform politics into an arena in which decisions are taken in a democratic, deliberative and participatory fashion, rather than behind closed doors by actors wielding power and money. 3. to the proponents of deliberation, anti-deliberative actors are not adversaries or enemies who must be defeated and must succumb (either by force or strategic manipulation) to an interest other than their own; they are considered, from the outset, as partners in a future deliberation within which their arguments will be considered with fairness and attention.
These characteristics of the model of enlarged democratic deliberation rule out (even against possible deliberators who might misunderstand it) the recourse to the support of a wider public that might aim to take the side of the deliberators without sharing their deliberative approach and finalities. The long-term objective of the proponents of deliberation must be a society in which deliberation is the decision-making norm. Therefore, their first commitment must be to deliberative democracy itself, and they must welcome only the support of those who share with them both this long-term commitment and the short-term objective of bringing the original antideliberative actor to resume deliberation on the particular problem at issue. Similarly, the model of enlarged democratic deliberation also rules out the possibility that deliberators seeking support for a specific issue may enter into purely strategic alliances with associations or groups; it does, however, encourage the building and maintaining, through deliberation and for deliberation, of relationships of collaboration between groups and associations faced with a common problem. When confronted by anti-deliberative actors, it often happens that the proponents of the deliberation may at first feel isolated; sometimes they cannot find enough support among the public they initially manage to reach out to, especially when this public is manipulated by the media or holds prejudices or strategic interests against the group the deliberators belong to. Where this is the case, the proponents of deliberation might benefit from extending their range of action, particularly when they know, or suppose, that other communities elsewhere are dealing with the same or similar issues. For example, a group that has unsuccessfully asked the local government for certain relevant and controversial decisions to be taken through deliberation, if unable to elicit enough support for its cause locally, can get in touch with other communities in which the same or similar decisions are being taken undemocratically. Groups from different communities can meet (face to face or through conference call services) to discuss and deliberate on how to get wider visibility and public support with respect to the problem they have in common; they can compare their different contexts and hypothesize solutions, share experiences, give each other material and moral support and, in general, broaden their reach and the proposals they can count on. Subsequently, on the strength of the new ideas and proposals that emerged from these discussions among proponents of deliberation, they will probably be more capable of finding the right key to attract the public's attention to the problem in question and to increase attendance at the meetings they will organize, each in their own community, in order to discuss it. In this way, an issue that local anti-deliberative actors had silenced in a single community, can achieve a much wider resonance across different regions, to the point that, on the part of the powerful, refusing to deliberate becomes difficult or impossible. These forms of bottom-up self-organization between local groups, just like the big numbers they can gather together, are achieved through deliberation and for deliberation: everyone shares a normative and practical commitment to deliberation and all are motivated by the goal of making good reasons prevail. While the model of enlarged democratic deliberation does not shy away from recourse to big numbers, it does so only in order to resume or begin a deliberation founded on the force of the better argument. These forms of cooperation are radically different from alliances based on self-interest and on strategic calculations, where partners are instruments for one's own ends and opponents are adversaries that must be defeated with the aim of increasing one's own power.

4.
The societies in which we live are characterized by widespread structural injustices and inequalities; racism, sexism, homophobia and other forms of prejudice and negative discrimination are common in them; the media are often used strategically by the most powerful actors to orient people's opinions in the desired direction. In this paragraph, we shall see how the model of enlarged democratic deliberation relates to these structural injustices and distortions.
When a powerful actor refuses to deliberate, as we have seen, the proponents of deliberation must aim to win the attention of a wider public and the media, in order to pressure him/her to resume deliberation. However, deliberators may face many other obstacles besides one or more anti-deliberative actors. They may belong to a marginalized or stigmatized ethnic or social group within a society largely characterized by prejudice; they may be faced with a public driven by dogmatic ideologies or strategic interests and, therefore, unwilling to listen to their arguments and discuss them; or they may be confronted with a society in which mainstream media are strongly oriented by the political and economic interests of the most powerful actors and capable of influencing large parts of the public opinion. In all these cases, structural distortions and inequalities constitute a further obstacle to deliberation and to the deliberators' possibilities for action. Here, usually, the first challenge for the proponents of deliberation, especially if they belong to groups subject to marginalization, prejudices and structural injustices, is to gain visibility in the public sphere and make their claims heard through their own voice. In doing this, they will have to face major structural barriers, and, even when they succeed, the task of breaking through preconceptions, dogmatic stances and mediainduced convictions will have only just begun.
The key, nonetheless, is to think outside the dyadic model of communication and to act on several fronts. That is, the proponents of deliberation should at the same time try to get through to the wider public (even if initially largely indifferent or hostile) and seek allies among sympathetic sub-publics or other marginalized and discriminated groups, through forms of self-organization like the ones described in the previous paragraph. Independent newspapers and alternative TV and radio stations can also be relevant instruments to inform a larger audience about one's situation and to make one's requests and arguments heard. By establishing collaborative relationships with sympathetic groups, the proponents of deliberation will be able to get moral and practical support; they will also gain access to other publics or sub-publics, while at the same time continuing to strive to bring at least sections of the originally hostile public to develop forms of enlarged thought (Young 1997). To this end, information and discussion meetings, open to all, can possibly be organized by the deliberators on the specific problem that the antideliberative actors refuse to discuss. Becoming involved in the discussion of a certain issue presupposes, however, curiosity and interest in the matter at issue on the part of the public itself, as well as, often, a favourable disposition towards the proponents of discussion; in many cases, and especially when the group proposing the deliberation is marginalized or subject to prejudice, this interest and favourable disposition must first be generated.
In this case, there remains the possibility of nondeliberative (but not anti-deliberative) demonstrations aimed at drawing attention to the group and its demands. The group in question can implement traditional forms of demonstration such as rallies, banner drops, sit-ins, etc., or experiment with means of expression specifically aimed at raising attention among the public and the media: flash mobs, acts of creative protest such as spontaneous or choreographed dances, silent protests, countercelebrations, performance protests, etc. (see Hanna et al. 2015). All these modalities, at first glance, are nondeliberative; however, they are not anti-deliberative, since, in the model here proposed, they are not considered as alternatives to deliberation, but rather aim to initiate and encourage a deliberative discussion by generating attention and curiosity in the public. These activities, therefore, need to be combined with the setting-up of contexts (after the demonstration itself, or in other spaces and moments) in which the members of the proponent group can explain in depth the reasons for the demonstration, inform the public about their situation, present their perspective and discuss it with all those interested. It is essential that non-deliberative demonstrations go hand in hand with discursive moments of explanation, information and discussion, appealing both to the empathy and the rationality of the people involved; only in this way will it be possible to bring a good number of people to put aside prejudices or false beliefs induced by the mainstream media, and to gain their support.
At this stage, as well as when deliberations with the original anti-deliberative actors resume, the role of facilitators is crucial: often overlooked and under-theorized (Moore 2012: 147, 149), their contribution is particularly needed in deliberations where the participants, in order to understand each other, have to overcome prejudices, disparities in power and, in some cases, ancient grudges and old suspicions. In such contexts, the facilitator must of course 'balance participation, create a respectful climate, and stimulate, clarify, and summarize discussions' (Trénel, 2009: 253). However, his/her role also extends to mitigating the effects of adversarial framing (Kadlec, Sprain & Carcasson 2012: 28-29), issue polarization (Fung 2005: 414) and speaker domination (Crosby & Nethercutt 2005), as well as to setting the discussion tone in a way that empowers disadvantaged participants and makes them feel free to express themselves, so as to minimize forms of 'internal exclusion' (Young 2000: 53). In order to perform these tasks, a good facilitator will probably assume in these circumstances an attitude of moderate involvement (according to the categorization proposed by Dillard 2013: 220), while keeping a neutral stance on the matter at issue, so as not to exacerbate polarizations and group contrapositions: his/her purpose should not be to try to frame the topic in a certain way or orient the positions of the participants, but to foster on their part the adoption of reflexive attitudes and the consideration of the perspectives and reasons of others.
Even in societies characterized by structural distortions and injustices, oppressed and marginalized groups have, therefore, some opportunities to gain visibility and support for their demands. Unfortunately, their efforts are not always successful, and the very circumstances that make those efforts necessary are deeply wrong and unfair. Remedying structural injustices is a task that requires commitment from many different actors on many different fronts, and, as such, goes far beyond the limited scope of the model proposed here. The model of enlarged democratic deliberation, however, offers significant theoretical and practical advantages: first of all, it allows consideration of a much larger number of actual and potential participants in deliberative procedures, staying clear of the oversimplifications of the dyadic model on which the realpolitik objection is based (1). The model here proposed is also able to account, on the practical level, for a spatially and temporally enlarged deliberation, which cannot easily be rejected or distorted by those who rely on their power or wealth to impose their will (2). It allows deliberators to rely on the power of big numbers obtained through deliberation and for deliberation (3), and, finally, it proposes a path for marginalized groups to follow in order to try and make themselves heard, even in societies characterized by deep structural inequalities (4). The more frequently anti-deliberative actors are forced by public pressure to resume deliberation, the more widespread deliberative modalities will become in society; and the more often political decisions are taken through free and inclusive processes of deliberation, the more sharply structural injustices will be undermined.
Every occasion on which anti-deliberative actors are forced by public pressure to go back on themselves constitutes a victory not only with regard to the specific matter at issue, that will be decided democratically, through the force of the better argument, but also with regard to the general diffusion of deliberative modalities and the struggle against structural inequalities. The very results of making deliberation as inclusive as possible will promote a more just society, which, in turn, will foster the further diffusion of inclusive deliberative processes capable of bringing about just results, realizing thereby the virtuous circle delineated by Iris Marion Young (2000: 27-36).
Political realists claim that politics know no other reason than force. However, even if lacking the force of power and money, those possessing the force of good reasons can obtain, through deliberation and for deliberation, the power of big numbers. The accounts developed by political realists neglect the normative element and emphasize power; but deliberative democracy, while favouring the first, cannot abstract from the latter. And it does not have to: its power is the power of good arguments, as well as the power of the many people who good arguments can bring together for justice. Through this power, deliberative democracy can oppose those who would rather impose their own privilege over reason and justice. It is in this way that, in the non-ideal conditions characterizing the societies we live in, even more disadvantaged citizens can have the chance to counter the inequalities affecting their position.