Deliberation and the Problems of Exclusion and Uptake: The Virtues of Actively Facilitating Equitable Deliberation and Testimonial Sensibility

In this paper, I suggest that one of the ways in which problems of exclusion from deliberation and uptake within deliberation can be ameliorated is to develop a more robust account of the deliberative virtues that socially privileged speakers/hearers ought to cultivate. Specifically, privileged speakers/hearers ought to cultivate the virtue of actively facilitating equitable and inclusive deliberative exchanges (which includes a cluster of virtues, including the practice of silence and of listening) and the deliberative virtue of training their ‘testimonial sensibility’ to correct for prejudicial judgments about other speakers.


Introduction
Deliberative democracy is roughly based on two fundamental ideas: that widespread participation in law-making processes is necessary for legal legitimacy; and that this participation should take the form of reason giving and arguments in an inclusive and equitable deliberative exchange. 1 There have been many criticisms of these features of the theory, but one of the most persistent is that deliberative democracy fails to adequately consider the structural inequalities that prevent marginalised groups and persons from speaking and being heard in the relevant way and to propose ways for making deliberation more genuinely inclusive. 2 These structural inequalities have at least two different effects in deliberative contexts: the first is that they can function to exclude marginalised persons and groups from deliberation altogether; the second is that structural inequalities can prevent hearers from recognising the communicative intent of marginalised groups and persons (that is, they inhibit 'uptake') because of various forms of epistemic injustice (Fricker 2007).
I suggest that the problems of exclusion and of uptake can be ameliorated by developing a more robust account of the deliberative virtues that socially privileged speakers/hearers ought to cultivate. Specifically, privileged speakers/hearers ought to cultivate the virtue of actively facilitating inclusive and equitable deliberative exchanges (which includes a cluster of virtues including the practice of silence and listening) and the deliberative virtue of training their 'testimonial sensibility' to correct for prejudicial judgments about other speakers (Fricker 2007).
Deliberative theory already imposes obligations on speakers to recognise one another as equals, to engage in perspective taking, to make intelligible claims, to tell the truth, and to be sincere (Habermas 1998, 322). Additional virtues include publicity, accountability, reciprocity, civility, listening, reflexivity, hope, fidelity to reason, and humility (Griffin 2011, 175). These virtues are also implicit in Brandon Morgan-Olsen's account of hermeneutical micro-climate (Morgan-Olsen 2010), Amandine Catala's account of epistemic trust (Catala 2015) and Susuan Dieleman's proposal of a 'deliberative expert' who is trained and tasked with ensuring formal/procedural inclusion and uptake within deliberation (Dieleman 2015).
In this paper, I build on these accounts of deliberative virtues in two ways: first, I suggest that while these virtues are relevant to all citizens, there may be additional virtues that socially privileged speakers ought to cultivate: the virtue of facilitating equitable deliberative exchanges, and the virtue of training one's testimonial sensibility. While we all enter deliberative exchanges with prejudicial baggage, as it were, the effects of this are not spread evenly. By this, I mean that we all enter deliberation with various unconscious prejudices, however, because we are not equally situated in relation to our social, political, or economic standing, the prejudices of the socially powerful and privileged will often silence and exclude others. If this is the case, then the onus should be on the socially privileged to address the various forms of epistemic injustice that exclude marginalized others and inhibit the uptake of their speech.
The category of privileged speaker or groups I have in mind are those speakers and groups who occupy a position of social privilege because of their race, class, or gender. While other speakers and groups may also enjoy a privileged position because of institutional support -such as scientists and other academics, politicians and policy experts -the problems of exclusion and uptake also occur in these spaces, and typically do so along gender, race, and class lines (for instance, women scientists, politicians and philosophers may be considered privileged and have institutional support, but may be excluded from or fail to achieve uptake in deliberation). For these reasons, I define socially privileged speakers and groups as those who enjoy speaking advantages because of their race, class and gender, insofar as they are able to achieve uptake or recognition for their speech because of who they are rather than what they say across a variety of informal and formal deliberative forums.
In section one, I give an account of the problems of exclusion and uptake as it arises in theories of deliberative democracy. In section two, I develop an account of deliberative virtues, which takes into consideration speakers/hearers' social position of privilege as one racy and Difference: Contesting the Boundaries of the Political (Princeton University Press, 1996) 120-35. 1 3 possible response to addressing the problems of exclusion and uptake. In section three, I respond to four possible objections.
Before proceeding, a qualification is in order. In focusing on the deliberative virtues that privileged speakers and hearers ought to cultivate to ensure the inclusion and uptake of the claims of marginalised others, I am by no means suggesting that institutions do not have a responsibility in securing inclusive deliberative environments or that their responsibilities are in some way diminished. Institutions have their own responsibilities in addressing the structural forms of oppression that exclude speakers from deliberation or prevent them from being heard in the relevant ways. They also have a part to play in drawing attention to the communicative pathologies pervading many, if not most, deliberative exchanges (perhaps through education) and in holding individual speakers to account. However, institutions (such as political, educational, or religious institutions) are constituted by people, and to varying degrees, are reliant on individuals cultivating their deliberative virtues in the appropriate ways. The account I am proposing thus envisages that institutional and personal deliberative virtues will work in tandem to address the problems of exclusion and uptake in deliberative exchanges, and I develop this relational dimension in the final section.

Deliberation and Inequality
One of the essential features of deliberative democracy is that all those who are affected by an issue should be included in any discussion about it. Habermas' discourse principle, for example, states that "just those action norms are valid to which all possibly affected persons could agree as participants in rational discourses" (Habermas 1998, 138). Underwriting this principle is a system of rights that ensures equal rights of participation in the deliberative process. The discourse principle presupposes that all affected groups will be included in deliberations, and that their perspectives will be equally considered within deliberation. It also assumes that all affected parties will be able to reason and use arguments to persuade others to roughly the same extent.
However, as many have pointed out, the guarantee of formal equality and inclusiveness does not mean individuals and social groups are equal in both a procedural and substantive sense. 3 Procedural equality refers to procedural guarantees that ensure equal access to relevant deliberative forums at both the agenda-setting and decision-making stages. Meeting these guarantees is a problem of institutional design and is a necessary, although not sufficient condition to establish the sort of equality that inclusive democratic deliberation requires. 4 As Knight and Johnson point out, because deliberation requires the uncoerced give and take of reasoned argument, it also requires a substantial notion of equality of opportunity, particularly in relation to deliberative capacities (Knight & Johnson 1997, 281).

Inequality in Deliberative Capacities
Substantive equality in relation to deliberative capacities has two dimensions: first, deliberation of this sort presupposes equality of resources, such as material wealth and a level of educational attainment so that a person's assent to an argument is genuinely uncoerced. Second, deliberation requires equal capacity to advance persuasive claims. Drawing on Sen and Bohman's capacity-based conception of political equality, Knight and Johnson point out that equality of resources is an insufficient remedy for effective participation because people differ in their capacities to use available resources effectively (Knight & Johnson 1997;Bohman 1996, 128).
There are at least three kinds of capacities relevant to democratic deliberation. The first is a capacity to formulate authentic preferences. Preferences can become distorted because of asymmetries in power and resources, causing participants to adopt 'adaptive preferences,' which in turn, affects uncoerced assent in deliberation (Knight & Johnson 1997). Second is the effective use of cultural resources (Knight & Johnson 1997). As Young has pointed out, the speaking style required by deliberative democracy is culturally biased, and assumes speakers have all had the same education and training. Deliberative spaces which require persuasion through reason-giving and argumentation will tend to silence or devalue some people and groups who cannot express their ideas and needs in the language of the dominant group (Young 1996). This lack of cultural resources means that certain groups will be unable to influence others in deliberation to adopt policies that will address their needs. The third kind of capacity relates to basic cognitive abilities and skills. As Knight and Johnson put it: "unless each participant has the cognitive capacities and skills necessary to effectively articulate and defend persuasive claims, then there will be no real equality of opportunity for political influence" (Knight & Johnson 1997, 299). These concerns are confirmed in the empirical evidence which suggests that the 'well-resourced' -those with good educational qualifications and high incomes -tend to participate more in formal politics, including general elections, and in informal civil society activity (Dalton 2006). Those without these capacities and resources tend to self-exclude, not because they do not want to participate but because they feel they cannot.
Knight and Johnson suggest that to ensure substantive equality is met, there needs to be certain accommodations to remedy the asymmetrical distribution of relevant deficiencies and faculties in areas such as one's ability to reason, articulate ideas, and so on and that that these are also connected to material resources. Minimally, achieving the relevant capacities will require government support of education. However, because cognitive abilities are affected by inequalities in wealth -such as in the way material resources affect diet and lifestyle, which in turn, affects cognitive development -effective participation in democratic deliberation will require government expenditure to guarantee the social economic, and educational prerequires for effective participation (Knight & Johnson 1997, 306).

Inequalities and the Issue of 'Uptake'
However, even if these material and cognitive inequalities are addressed by way of government policy or intervention, persistent inequalities in race, class and gender will also distort deliberative spaces. This is not because marginalized groups may lack the cultural resources to make claims in ways that are recognizable by the dominant majority, but because their social position prevents them from achieving 'uptake' for their views in deliberative exchanges irrespective of how their views are expressed because of what Fricker calls 'testimonial injustice.' In her groundbreaking work, Epistemic Injustice, Fricker draws attention to the ways in which deeply ingrained, and sometimes unconscious prejudices, result in either a credibility deficit or credibility excess for certain speakers. These unconscious prejudices lead to two different kinds of injustice: testimonial and hermeneutical. The former prevents a socially marginalized speaker from being heard in the relevant way because of her social identity, while the latter prevents socially marginalized speakers from being able to give an intelligible account of their experiences (Fricker 2007, 38-9).
In the case of testimonial injustice, hearers may, for example, politely listen to what a speaker is saying; they may give her an opportunity to speak, engage in dialogue with her, and ensure that her views are represented in the communicative exchange. The speaker may have excellent communicative abilities: she may speak intelligibly, persuasively, and have many great suggestions for how to solve a social coordination problem. However, even though she meets the very stringent deliberative requirements, she nevertheless suffers from a credibility deficit so that she fails to achieve uptake for her speech. Her interlocutors may unconsciously dismiss what she has said because of some lingering unconscious bias about women. This is certainly the experience of many professional women in workplace contexts. 5 Other features of women's experience in deliberation include being interrupted by men, ignored, and spoken over. As Nancy Fraser puts it, these behaviours and practices alert us to the ways in which social inequalities can "infect deliberation, even in the absence of any formal exclusions" (Fraser 1990, 64).
These behaviours run quite deep, in the sense that as Friker argues, various forms of testimonial and hermeneutical injustices happen all the time, and they may, in fact, be the normal social baseline. They are so normalized that we do not have the moral language for understanding the wrongs that are done to a person when she is treated in this way (Fricker 2007, 39). Similarly, Amy Allen has argued most of our communicative interactions are distorted. A full examination of the ways in which communication is distorted, especially in the public sphere, potentially undermines the theory of communicative action: 'it becomes difficult to make sense of systematically distorted communication at all, inasmuch as this notion relies implicitly on the possibility of undistorted communication …" (Allen 2008, 106). If distorted communication is a symptom of asymmetries in power, and these asymmetries are not only deeply embedded in all our institutions, but form and distort our very subjectivities then it is difficult, according to Allen, to see how one "can achieve the kind of reflexive distance from one's beliefs, practices and norms, and life projects … that supplies that notion critical bite" (Allen 2008, 106). According to Bohman, the consequences of this failure to achieve uptake are profound: politically marginalized groups are publicly excluded because they cannot successfully initiate the joint activity of public deliberation; conversely, these groups cannot at the same time, avoid political inclusion because they are the addressees of laws and policies over which they had no real control or influence (Bohman 1997, 333).

Epistemic Injustice and Deliberative Theory
More recently, there have been attempts to bring this work on epistemic injustice in dialogue with the literature on deliberative democracy, both to demonstrate other kinds of communi-1 3 cative pathologies operative in deliberative contexts, and to propose ways of addressing the problem of uptake in deliberation. Brandon Morgan-Olsen builds on Fricker's concept of hermeneutic injustice to identify another type of exclusion, which he refers to as 'conceptual exclusion' (Morgan-Olsen 2010). This functions to politically exclude citizens both by inhibiting their ability to make certain political claims, and by reducing the likelihood that those claims will be accessible to the public at large. He proposes that one way of addressing this kind of exclusion is to create 'hermeneutical micro-climates,' both within excluded groups to help them make sense of their claims and between different micro-climates to help bridge gaps in understanding and bring the claims of the excluded group to the wider political culture (Morgan-Olsen 2010, 238-239).
Also drawing on Fricker's account of epistemic injustice, Amandine Catala identifies 'hermeneutical domination' as another way in which marginalized groups are excluded from deliberation. This refers to the majority groups' wrongful dismissal of the testimony of the minority because of perceptions that the minority groups are epistemically untrustworthy. As a consequence, the minority group is denied the opportunity to contribute to collective hermeneutical resources and becomes subjected to a public discourse that is primarily formulated by the majority (Catala 2015, 427-8). Catala argues that testimonial injustice can be undone through epistemic trust: that the majority ought to and can recognize the minority's expertise in relation to their own lived experience of oppressive practices and institutions (Catala 2015, 432). To facilitate this trust, she suggests implementing strict rules of deliberation, including listening carefully, speaking respectfully, being responsive to others' contributions, and being self-critical (Catala 2015, 436).
Susan Dieleman draws attention to and develops Fricker's idea of 'virtuous hearing,' which can ameliorate epistemic injustice by requiring individuals to train their testimonial sensibility so that they can pick up and correct the identity prejudices they are making. Acting in an epistemically virtuous way requires that the hearer not dismiss a speaker because she is struggling to find the right words or because her style of communication is not suitably rational for the context in which she speaks (Dieleman 2015, 804). Implicit in all these accounts is that the majority has various responsibilities in addressing the problem of uptake and that hearers ought to cultivate various virtues, including training their epistemic sensibility. But as Dieleman notes, who are these virtuous hearers? Who is responsible of taking on this kind of training? Is it the case that all people, "in all domains of their lives, are required to stretch the bounds of what they are willing to consider epistemically worthwhile?" (Dieleman 2015, 795). And in what context should this occur? In philosophical discussion, where precise and clear articulation of arguments is crucial, or only in political contexts? (Dieleman 2015, 795).
The argument developed in the next section responds to this question, by suggesting that while we all have various responsibilities across all deliberative contexts (including in philosophical deliberation), privileged speakers have more onerous responsibilities. Specifically, they are responsible for the virtue of actively facilitating inclusion in deliberative spaces, including the associated cluster of virtues attached to this, such as knowing when to remain silent and listening, and the virtue of training one's testimonial sensibility. While training one's testimonial sensibility has been addressed by both Fricker and Dieleman, I develop this account in the context of different public deliberative contexts, including formal and informal ones and as it relates to socially privileged speakers.

3 3 Privileged Speakers/Hearers and Deliberative Virtues
A virtue can be defined as a practice, a disposition, or a set of habits that is held in esteem by a community or a society. The practice or habit is thought to be beneficial in some way, and its acquisition must be both voluntary and involve some degree of difficulty in mastering (Beatty 1999, 291). That is, a virtue is a habit that requires some training because it otherwise does not come easily. In acquiring the virtue, a person often has to resist temptations pulling her in other directions, such as emotions, actions and inclinations. However, once the virtue is appropriately trained, it becomes easier to perform (Beatty 1999, 291).
Deliberative theory already requires speakers and hearers to cultivate various virtues. For example, the deliberative virtue of reciprocity is most commonly invoked in the literature, as this virtue is thought to increase common ground among citizens and increase the chances of a fair, just and mutually agreeable decision being reached (Griffin 2011, 178). Gutmann and Thompson define reciprocity as the capacity to "seek reasons that can be justified to all parties who are motivated to find fair terms of social cooperation" (Gutmann & Thompson 1996, 52). It functions to regulate public reason by limiting the kinds of reasons that citizens can give in defense of various claims that they can make of each other (Griffin 2011, 178).
Other deliberative virtues include civility and tolerance. These encourage citizens to talk and listen to one another with respect. According to Levine, civility can play an important role in encouraging citizens "to challenge ideas strenuously without attacking people as individuals or as a member of a group" (Levine 2010, 16). Virtues like civility and tolerance can enable a diversity of perspectives within deliberative exchanges. However, as Griffin points out, these virtues do not show us how citizens develop or change their values and beliefs so that they genuinely are less self-interested and more sensitive to the truth (Griffin 2011, 179). Moreover, while these deliberative virtues may give the appearance of fostering inclusive and equitable deliberative exchanges, marginalized speakers may nevertheless not be heard because of testimonial and hermeneutical injustice.
In response to this issue, Griffin suggests the cultivation of deliberative virtues such as reflexivity, where citizens question their own positions on various issues and their value commitments; the virtue of hope, which provokes the feeling that all is not lost and that citizens can make an impact; fidelity to reason, which ensures that citizens deliberate internally through complex deliberative processes rather than on the basis of inaccurate stereotypes, or dogma; and humility, which refers to a capacity to admit when one is wrong and to correct one's views accordingly (Griffin 2011, 180-3).
To this list, I suggest an additional two further deliberative virtues that are relevant to socially privileged speakers/hearers: the virtue of facilitating inclusive deliberative environments (which includes a cluster of virtues including the virtue of knowing when to remain silent and the virtue of listening) and the deliberative virtue of training one's testimonial sensibility. A focus on these under-examined deliberative virtues as they relate to socially privileged hearers/speakers offers one way of addressing the issues of exclusion from deliberation, and the problem of uptake within deliberation.

The Deliberative Virtue of Facilitating Inclusive and Equitable Deliberative Exchanges
While all speakers should cultivate the deliberative virtues of reciprocity, civility, and tolerance, privileged speakers may have more onerous or demanding responsibilities to cultivate additional deliberative virtues, such as that of facilitating or actively fostering inclusive deliberative exchanges. This would require them to be highly attuned to deliberative pathologies and adopt the practice of calling them out when they see them occurring, in order to reveal or expose the communicative pathologies at play in the exchange. Such behaviors can include those identified by Fraser, such as privileged speakers dominating deliberation, speaking over the top of others or interrupting (Fraser 1990) and those identified by Fricker, such as discrediting the claims of others on no other grounds than the social identity of the speaker, patronizing other speakers, or professing to know something of another's social experience (in colloquial terms, many of these behaviors are referred to as 'mansplaining').
Given that many of these pathologies and/or prejudices have been habituated over time, it may well be that other privileged speakers/hearers are unaware they are occurring (i.e., that they are dominating the discussion, speaking over others, or dismissing what other speakers say simply because of the speaker's social identity), and so drawing attention to these subtle and unconscious dynamics might function to bring the issue into the open and to raise awareness on the part of other privileged participants. This could, in turn, encourage other deliberative participants of the same privileged group to exercise due diligence with respect to their own speech, thus facilitating more equitable deliberative exchanges. For example, these responsibilities could involve a privileged male participant pointing out that a woman has been cut off or that her views have not been given full consideration; it could mean silencing other men in the exchange and asking for other views to be considered. While these are typically the functions of chairpersons or facilitators, they need not be, especially in highly unregulated deliberative environments.
To highlight with a more specific example: a recent philosophy conference I attended had only one female speaker and three women in the audience out of a total of 30 audience members. During question time, several hands were raised, including the hand of one of the women. The (male) chair called on the five men to ask their questions, then cut the discussion off because of time restrictions. The women in the room were visibly annoyed, but also clearly uncomfortable about saying anything explicitly. Instead, one of the male speakers pointed out what had (unintentionally) occurred: that the only woman who wanted to speak was not given an opportunity to do so, and that it would not matter if the schedule was pushed back a little to re-dress this inadvertent silencing. Pointing this out revealed something about the ways in which women were excluded from discussion, even though they were physically at the conference and how this exclusion was operating at a subtle and perhaps unconscious level. The intervention successfully altered the communicative dynamics for the rest of the day. Male participants became more aware of the ways in which they were dominating the discussion, to the exclusion of other voices and perspectives, and took responsibility for their own communicative behavior.
While cultivating the virtue of facilitating inclusive deliberative exchanges might risk producing a paternalistic response: say, men speaking on behalf of women or racial minorities in ways that reproduce communicative pathologies, it also does not place the onus on marginalized groups to fight to be heard, or to educate others. The risks of paternalism might also be further mitigated by developing greater sensibility to communicative dynamics, and learning when to stay silent, or when to appropriately silence others who may be engaging in deliberatively pathological behavior. While many philosophers and political theorists have drawn attention to the how privileged persons can silence marginalized speakers, there has been less attention to how silence can open up communicative exchanges.
Silence is practiced in many other deliberative contexts, including in the classroom or in professional psychotherapeutic relations. A therapist's practiced silence relative to a client functions to promote or provoke disclosure; a teacher's use of silence in the classroom can draw out a class. In each of these cases, as Ferguson puts it, "silence functions as a demand, not for silence in return, but for narrative participation. Silence thus evokes non-silence: it incites interaction without demanding it" (Ferguson 2003, 57). Dauenhauer has demonstrated the ways in which silence is an active human performance, one which involves a 'yielding' before another and which is outside one's control. As he puts it: "in performing silence one acknowledges some center of significance of which is not the source, a center to be wondered at … the agent is aware that the doing of silence opens him to meet that which lies beyond his control" (Dauenhauer 1980, 24-5).
Silence is also a necessary condition for the virtue of listening. As Fiumara argues, listening authentically does not so much require perfecting linguistic capabilities as much as it requires silence or a "listening silence" (Fiumara 1990, 96-7). The importance of listening is often invoked as necessary for countering the problem of uptake, but what does this actually involve? Good listening, as Beatty has argued, is a virtue, one that is beneficial in most communicative situations. It is a virtue that requires deliberate training to overcome states such as laziness, inattention, egoism, narcissism, dogmatism, and resistance to change and selftransformation (Beatty 1999, 291). The good listener permits individuals to be themselves, to express meanings and judgments that are significantly different from the listeners, to tell their story in their own way, even if the listener can formulate the story more coherently or compellingly. While a good listener can intervene at times to seek clarification, these interventions must be sensitive to allowing the speaker to speak in her own way. The good listener has the courage to listen to things they might find threatening and is open to views that are foreign to her own (Beatty 1999, 292-3). Good listening opens the way to not only greater understanding of others, but to self-transformation. As Lipari puts it: "by stepping of the cliff of ego-bound self-certainty, the communicative acrobatics of listening to others as other makes ethics and transformation (of selves, others, and even worlds) possible" (Lipari 2013, 157). Cultivating the deliberative virtue of facilitating or promoting inclusive deliberative exchanges and achieving uptake may thus require a sensitivity to silence; an understanding of when to remain silent so as to create a space for others to speak in the first place (to ensure inclusively) and to really listen to what the other is saying (to ensure uptake of what they are saying).
Of course, this virtue is not going to be possible to do in every deliberative exchange. It might work better in more structured deliberative contexts, such as town hall or council meetings, academic conferences, and workshops, in classrooms, government cabinet meetings, and televised discussion panels, to name a few. But it could also be possible in unregulated exchanges, such as in online forums. All it would take is one or two privileged deliberative participants developing the virtue of facilitating equitable exchanges, by exposing certain communicative dynamics, critically engaging with other members of the social group (say, critical and respectful confrontations with other men), to perhaps shift the direc-1 3 tion of a debate, or cause others to re-think their speaking practices and judgments, thereby promoting more equitable deliberative exchanges.
Suggesting that individual deliberative participants who occupy positions of social privilege have more onerous responsibilities in cultivating these deliberative virtues in communicative exchanges can also facilitate a form of group responsibility. By bringing communicative pathologies into the open, individuals can engage in critical confrontations with members of their own group, encouraging the group to think critically about its bias, thereby improving individual and collective awareness. Individual responsibility might then, improve deliberative exchanges within both regulated and unregulated spheres of deliberation.
There are two main advantages to including this to the list of deliberative virtues. The first is that a focus on the deliberative virtues that ought to be cultivated by privileged speakers is better able to address the kind of deliberation that occurs in unregulated deliberative exchanges, where formal deliberative rules may not be so easily enforced. As Aikin and Clanton put it, in the 'real world' we often face multiple and differing debates, some formal, but most informal, time constraints and other less than ideal settings for deliberation. It makes sense, then, to think about what could be done to mitigate these practical problems that we face in actual informal and unregulated deliberations, with a view to improving the deliberation that occurs there (Aikin & Clanton 2010, 412). Aikin and Clanton propose a form of group responsibility that can only be fulfilled in and through the deliberative interactions and cooperation of many different actors.
To this end, they propose a list of deliberative virtues that speakers ought to cultivate, including deliberative wit, friendliness, temperance, courage, sincerity, and humility. The cultivation of these virtues will improve the quality of deliberation in groups (Aikin & Clanton 2010, 414-8). While I think these virtues are important, they may also inadvertently place more communicative burdens on some speakers. For example, requiring speakers to cultivate wit, described as the ability to think creatively, critically, and constructively (Aikin & Clanton 2010, 414) may place additional burdens on speakers suffering from hermeneutical injustice, who may not have the resources to even articulate their experience, let alone, do so with wit. Nor does this account directly address the issue under examination in this paper: that differently situated speakers and hearers may have different and more onerous responsibilities than others. The account I propose requires privileged individuals to cultivate the deliberative virtue of facilitating inclusive exchanges with a view to not only including marginalized persons and groups, but to also facilitate greater awareness and confrontations within their own privileged group.
A second advantage of focusing on the responsibilities of socially privileged participants to cultivate these deliberative virtues is that it does not place the onus on excluded or marginalized groups to speak in the 'right' sorts of ways to achieve recognition for what they say (and in any case, as we have seen, this might not make a significant difference, since these speakers are often excluded or not heard because of who they are rather than what they say). Nor does it require them to call out this kind of behavior themselves (which can be quite stressful or intimidating in some cases). Rather, it places the onus on socially privileged speakers to take responsibility for addressing the issues of exclusion from and inequality within deliberation by being conscious of and modifying their own behavior. It also places the onus on them for facilitating equitable deliberative exchanges, ensuring that marginalized speakers achieve uptake for their speech, creating space for marginalized 1 3 speakers to make intelligible their experiences, and raising awareness of these issues within their own social (privileged) group.
As Medina puts it, it should not be the responsibility of oppressed groups to educate dominant or privileged others about their situation. Oppressed subjects are not, as Medina puts it, obligated "to facilitate the communicative and epistemic agency of more privileged subjects" especially if "that can worsen their precarious situation and deepen their oppression" (Medina 2012, 116). Nor should they bear the deliberative labour of educating privileged others, which as Nora Berenstain argues, is a type of 'epistemic exploitation.' It occurs when privileged persons compel marginalised persons and groups to produce knowledge or explanation about the nature of the oppression they face (Berenstain 2016, 570). This is a form of epistemic oppression because it requires a significant amount of unrecognised, uncompensated, and emotionally taxing labour on the part of the marginalised to provide information, resources and evidence of oppression to the privileged, which, is then often dismissed (Berenstain 2016, 570). Placing the onus on privileged speakers to educate themselves and their social group by facilitating equitable deliberative exchanges (for example, by calling out certain behaviors perpetrated by members of their own group that function to exclude already marginalized speakers) can potentially circumvent this problem.

The Deliberative Virtue of Training One's Testimonial Sensibility
Given that the problems of exclusion and uptake are produced and compounded by the prejudicial judgments that socially privileged speakers/hearers make of marginalized persons and groups, the onus should again be on them to train their testimonial sensibility in an appropriate way. Training one's testimonial sensibility requires, as Fricker argues, the cultivation of various skills and virtues in ways analogous to ethical training. According to Fricker, ethical sensibility is acquired initially by our being inculcated and socialized into the attitudes of the day. But we are soon able to criticize those attitudes, and so we may, over time, come to distance ourselves from any given commitment. Ethical training thus comprises two types of input: social and individual: "one develops an ethical sensibility by becoming inculcated into a historically and culturally specific way of life … where this is to be construed as a matter of ongoing ethical socialization" (Fricker 2007, 82). The social dimension is passively acquired from the ethical community, but developing an appropriate ethical sensibility requires an individual to develop a suitably critical distance from her primary ethical socialization in light of the experiences that life offers her, experiences which may be in tension or at odds with her socialization (Fricker 2007, 82).
Fricker argues that virtuous hearers acquire their epistemic socialization in analogous ways: the hearer is initially passively socialized by way of a background 'theory' of socially situated trustworthiness, which is internalized over time (Fricker 2007, 83). In the same way a person develops an ethical sensibility through critical distance and reflection, a hearer will become aware that her experiences in testimonial exchanges are in tension with the sensibility she has passively internalized. She thus becomes responsible for adjusting her sensibility to accommodate this new experience. How might this ethical training play out in deliberative exchanges?
Privileged participants should train themselves to be on high alert about the credibility judgments they are making and be aware of the social position they occupy. This training, I suggest, requires that hearers develop the deliberative virtue of humility, identified by 1 3 Griffin. Humility enables citizens to change their minds about the beliefs and preferences they hold, to recognize the error of their judgments, to correct their beliefs and preferences accordingly, and to base their decisions on good evidence, rather than dogmatism, stereotypes, or rigid belief systems (Griffin 2011, 184). While Griffins suggests all citizens should cultivate this deliberative virtue, and that this will lead to better deliberation and decision making, I suggest that there may be additional obligations on privileged groups to cultivate humility because it will better enable them to train their testimonial sensibility, thereby ameliorating the problem of uptake.

Objections
There are at least four possible objections to this argument that differently situated speakers have more onerous obligations to cultivate deliberative skills to address the problems of exclusion and uptake in deliberative exchanges. The first is given how deeply insidious communicative pathologies of this kind are, how might we gain critical distance from prejudicial judgements about marginalized speakers, especially since in some cases, we may not even be aware that it is occurring? Second, why would the privileged want to give up their social positions of power and influence, especially in political contexts? (Laverty 2014, 36). Third, why shouldn't they dominate deliberation, if they are better at giving reasons and understanding the issues because of their formal education? (Laverty 2014, 43-44). Finally, and perhaps most importantly, in building on Fricker's account of developing individual deliberative virtues, this account is also vulnerable to the same criticisms made by Rae Langton (Langton 2010) Elizabeth Anderson (Anderson 2012) and Kristie Dotson (Dotson 2012) namely that the focus on individual virtues to respond to structural injustice is likely to fail.
With respect to the first issue, just because critical distance from unconscious prejudicial judgements is difficult, it does not make it impossible. Socially and politically, there has been a great deal of awareness raising about 'white privilege,' 'male privilege' and 'class privilege.' This should prompt people to think about the social advantages they enjoy and whether this causes them to make prejudicial identity-based judgments in deliberative contexts.
With respect to the second issue, it may well be that some speakers are quite comfortable with their position of privilege and the advantages it confers; they may have no desire to give this up or for more inclusive deliberative forums. Or more worryingly, they may think that they have already given up a lot, and resist any further obligations for ameliorating inequality. There is already considerable resistance or backlash to the consciousness raising about white privilege and male privilege, and a general fear about loss of power. 6 It would seem then, that those who are most in need of greater testimonial sensitivity are also those who have been most resistant to the awareness that is required to train it. 7 There are several responses to this problem. First, it is important to note that this account is intended to operate at both a normative and descriptive level. At the normative level, I 1 3 suggest that all socially privileged participants in deliberation ought to develop these virtues for the reasons I have given. At the descriptive level, I concede that many will resist these responsibilities perhaps because they think that that women and other minorities have already become more privileged or because they remain unaware of their own position of privilege. The fact that this resistance exists does not affect the normative account: irrespective of what some socially privileged participants think about these issues, the fact is that structural inequalities persist in our societies, and that they distort communicative exchanges in quite profound ways. Consequently, privileged speakers/hearers have different and more onerous responsibilities for training these deliberative virtues at the normative level. Second, at the descriptive level, it is unlikely that all privileged speakers/hearers will train these deliberative virtues, but there are many others who are capable of recognising their privileged position when it is pointed out to them or who are aware of the position they occupy. At a practical level of implementation, it is these persons who, I suggest, should engage with recalcitrant privileged people, including by making oppression and privilege visible to them, making the case for why inclusion and uptake are in their interests (such as by demonstrating how exclusion limits the available solutions and ideas to problems, which over the long-term, disadvantages everyone), and by making it personal for them. 8 At a descriptive level, the account does not require all privileged persons to train these virtues (although they ought to morally). But it does require that members of privileged groups take responsibility for the issues of exclusion and uptake in deliberation, and educate members of their own group who are responsible for perpetuating these deliberative pathologies. It should not be the responsibility of marginalised groups and persons to educate others (either normatively or descriptively) but that this should fall to the socially privileged.
The third objection is that privileged speakers perhaps should dominate deliberations, perhaps because they have more expertise about the issue, and are better able to give reasons because of their education and training. However, as I have suggested, perhaps these speakers are not in fact, better placed to make decisions and we (and they) only think they are better placed because of 'credibility excesses.' Irrespective of their formal education, they will nevertheless not have knowledge or understanding of the experiences and lives of others which are different from their own, and for these reasons, they should, as Catala argues, trust the epistemic privilege of others. Moreover, the issue is not necessarily one of whether excluded and marginalized speakers can deliberate about an issue, but whether there is a space for them to contribute at all in the deliberative exchange.
The final objection concerns the focus on individual solutions to address an ostensibly structural problem. For example, in her review of Fricker's work, Langton notes that Fricker offers an individual remedy in response to a structural problem, whereas it may be that a structural remedy will be more effective. Langton uses the example of how 'testimonial injustice' towards girls in school exams has been solved: not by training the virtues of examiners and expecting them to exercise responsibility, but by the structural remedy of making exams anonymous (Langton 2010, 463).
Similarly, Anderson has argued that: In the face of massive structural injustice, individual epistemic virtue plays a comparable role to the practice of individual charity in the context of massive structural 8 See Smith, n 47.
poverty. Just as it would be better and more effective to redesign economic institutions so as to prevent mass poverty in the first place, it would be better to reconfigure epistemic institutions so as to prevent epistemic injustice from arising. Structural injustices call for structural remedies. (Anderson 2012, 171).
For example, in employment contexts, structural remedies to prevent discrimination include requirements that institutions make employment decisions such as hiring, firing and promotion on explicit objective measures rather than subjective assessments; managers are given sufficient time to make careful decisions against the criteria; that the evaluation context avoids priming stereotypes; and managers are held accountable for discrimination when it occurs (Anderson 2012, 168). Kristie Dotson, drawing on Langton, argues for a "conceptual revolution," a fundamental change in our schematas or frameworks for generating knowledge (Dotson 2012, 30). She argues that a focus on individual virtues is a reform that may make our credibility judgments more accurate, but does not in itself challenge the value of credibility: "the value of credibility remains the same; it is how we pursue it that alters" (Dotson 2012, 28).
Structural change is of course, fundamental, and institutions have an important role in addressing the problems of exclusion and uptake as they occur in communicative exchanges. Institutions should be designed in such a way that enables all groups and person to deliberate and to also hold the privileged to account to ensure that they cultivate the virtues I have been proposing. The account of privileged speaker responsibility is intended to operate in tandem with widespread structural changes, but to also fill in some gaps or deficiencies in relation to the capacities of an institutional or structural response to these issues.
First, the sorts of structural changes in employment, education, and political contexts, to name a few, depends on the relevant institution initiating and implementing structural reform, and holding individuals to account. But institutions are constituted by people, who collectively, must initiate these structural changes and ensure they are being adhered to. In these contexts, those in positions of power and privilege have onerous responsibilities to ensure these changes are made and enforced, so an account of individual deliberative virtues of the kind I propose is relevant here.
Moreover, informal deliberation about injustice is quite often a crucial precursor to any structural reform or any kind of conceptual revolution. Structural reform is thus contingent on individuals coming together to discuss an issue and collectively agitate for change. Therefore, we cannot lose sight of informal spheres of deliberation, and of suggesting ways of making them more equitable. These informal spaces are better able to include the perspectives of marginalized and oppressed groups, and they are often the places where issues are brought to public awareness and put on a legislative agenda, leading ultimately to structural reform. And changes to our concepts and values, including concepts like 'credibility' and the value we place on it, are also fundamentally altered through these informal discursive processes. While a focus on privileged speaker and hearer responsibility for cultivating deliberative virtues will not, on its own, achieve these complex objectives, it may go some way to making informal deliberative spaces more inclusive and facilitating uptake.
Second, as I have argued, institutions may not be up to the task of securing inclusivity and equality on their own because of the pervasive nature of testimonial injustice (a point also conceded by Anderson (2012, 169) and because many of these structural reforms rely on people in positions of power addressing their own prejudices. Consider, for instance, 1 3 the case of a university enacting widespread structural reform in the employment context. These reforms are intended to ensure equal representation of all groups on selection panels for hiring and promotion, to ensure that women are not penalized for career interruptions to have children, and to implement flexible work arrangements so that employees can manage work and care responsibilities. It may offer training for managers in cognitive bias and how to overcome it and have robust accountability measures in place. These measures might be very effective at overt forms of discrimination, such as sexual harassment in the workplace, however, they may be less effective at the more insidious forms of exclusion and uptake that may not be as easy to identify.
A male manager may still silence women and other marginalized groups in meetings, or dismiss what they say, or harbor biased views that affect his decision-making, and so on. It is not always easy to appeal to the structural measures in place to address this kind of exclusion and failure of uptake. In these situations, it may be up to other privileged speakers and hearers to facilitate more equitable deliberative exchanges, by perhaps endorsing the view that has been dismissed to give it a fair hearing, or by pointing out what is occurring (which may or may not be intentional on the part of the manager). Privileged speakers have less to lose in developing these virtues and putting them to good use, and so the burden should fall to them.
Third, institutions may have limited reach in relation to informal deliberative spaces, especially in online contexts. For example, while social media platforms can attempt to foster equitable deliberative exchanges at an institutional level by prohibiting certain kinds of discriminatory speech that may exclude speakers in online deliberations, this is also difficult to do on account of the sheer volume of online deliberation. It may be up to socially privileged individuals to step in and fill this accountability gap. The same could be said in institutional contexts. While the structures in place may work better at enforcing deliberative responsibilities in formal deliberative spaces -such as on a selection panel -they may be less effective at the level of everyday, informal communication that occurs in small meetings, post-conference deliberations, and so on.
A focus on the responsibilities privileged speakers have in cultivating additional deliberative virtues is not, therefore, intended to replace structural or institutional measures at addressing the problems of exclusion and inequality but to work in tandem with them. While individual charity will not solve poverty, charity inculcates a mindset about the importance of income distribution, that may act as a precursor to collective action and more structural change. Similarly, arguing that privileged speakers have more onerous responsibilities to facilitate more inclusive deliberative exchanges and to train their judgements accordingly is intended to inculcate a mindset about overcoming prejudicial judgments, that may in turn, act as a precursor to more structural reform.
I have argued that the problems of exclusion and uptake in deliberative exchanges arise, in part, because we are not similarly situated as speakers. Power imbalances in deliberative environments adversely affect marginalized groups and persons in various ways, making it difficult for them to achieve uptake in the relevant ways. While we are all guilty, to varying degrees of making prejudicial judgements that may function to exclude others, the effects of this are not evenly distributed. Often it is those who occupy social, political, economic positions of power and privilege whose behavior and prejudicial judgements about people's social identity function to exclude others from deliberation and within deliberation. As such, I have argued that socially privileged speakers ought to cultivate different virtues: the virtue of facilitating equitable deliberative exchanges (and the cluster of virtues that go with it), and the virtue of training one's testimonial sensibility. The account I have proposed is a modest response to the issues of exclusion and uptake in deliberative spaces. Identifying the onerous responsibilities that privileged speakers and hearers have is intended to work in tandem with structural reform: institutions may need to step in to enforce these responsibilities when they can, so it is not left entirely up to the good will of privileged speakers/hearers, but speakers/hearers may also need to step up to bolster institutional norms or fill in the gaps in institutional coverage.

Declaration of Interests
No funding was received to assist with the preparation of this manuscript and the author does not have any relevant financial or non-financial interests to disclose or competing interests to declare that are relevant to the content of this article.

Open Access funding enabled and organized by CAUL and its Member Institutions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.