Articulating the social: Expressive domination and Dewey’s epistemic argument for democracy

This paper aims at providing an epistemic defense of democracy based on John Dewey’s idea that democracies do not only find problems and provide solutions to them but they also articulate problems. According to this view, when citizens inquire about collective issues, they also partially shape them. This view contrasts with the standard account of democracy’s epistemic defense, according to which democracy’s is good at tracking and finding solutions that are independent of political will-formation and decision-making. It is also less vulnerable to the criticisms that have been raised against the standard account. To show this, the paper develops a theory of expressive domination and argues that problem-articulation works best when it is inclusive and domination-free. It also shows that democratic conflict represents a fundamental element for problem-articulation.


Introduction
The aim of this paper is to contribute to the debates on the possibility of providing an epistemic argument 1 for democracy, 2 based on John Dewey's idea that, in democratic decision-making, citizens and representatives do not merely solve but also articulate social problems. The notion of articulation points to the view that in democracy citizens not only identify problems and provide solutions to them, but they also give them their 'specific shape' (Taylor 1989, 374). That is, citizens frame and reframe their experiences of the problematic dimensions of social life in the process of identifying, defining and resolving the challenges they face. Hence, this view resists the false choice between radical constructivism, on the one hand, and realism, on the other. For the former, social problems are fully constructed by citizens in their attempts to solve them, rather than having their existence fully independent from them, as realism affirms. However, as Rahel Jaeggi (2018) puts it, for Dewey social problems are 'at once given and made', (140) neither merely objective nor subjective. 3 For Dewey, the articulation of social problems does not take place in a context free from relations of domination. Thus, the ability of some dominant groups to unequally influence how problems are articulated in democratic decision-making negatively affects the capacity of problem-articulations to address the challenges citizens face. Political institutions and practices must then confront this challenge if they are to be successful in articulating and solving collective problems.
In my view, this Deweyan insight into problems and the domination some groups exert in how we articulate them can contribute to addressing some important challenges usually directed against epistemic arguments for democracy. Some authors argue that the idea that we can appraise the value of democratic decision-making based on the quality of its decisions puts democracy in serious danger: On the one hand, some critics argue that, in order to reach better outcomesthat is, to effectively solve our collective problemswe sometimes need to exclude some groups from democratic decision-making or deny them equal political participation in those practices. A weakerand probably more compellingversion of this criticism contends that, at best, an epistemic argument for democracy makes the participation of some citizens superficial, since some group whose scope is smaller than the whole political community may be able to come to correct decisions. On the other hand, another line of criticisms argues that epistemic arguments for democracy often make social conflict and dissent as dispensable aspects of democratic decision-making. They argue that if there are correct answers to political decisions, and if the value of democracy depends on tracking them, conflict and dissent become inessential since they may prove unnecessary for tracking those correct answers. According to critics, all these worries would invalidate the idea that we can provide an argument for democracy that is based on the quality of the outcomes of democratic decision-making, and should invite us to take a different path for defending the value of democracy. 4 In my view, these worries capture the shortcomings of a group of epistemic defenses of democracy that is substantially different form the Deweyan argument I aim at developing along these lines. These accounts belong to what has been called the 'standard model' of epistemic democracy. According to this model, 'there exists, independently of the actual decision-making process, a correct decision and the legitimacy of democratic decisions depends, at least in part, on the ability of the decision-making process to generate the correct outcome' (Peter 2009, 111) Similarly, List and Goodin (2001) argue that '[t]he hallmark of the epistemic approach, in all its forms, is its fundamental premise that there exists some procedure-independent fact of the matter as to what the best or the right outcome is'. (280) 5 We will have the opportunity to discuss in more detail how the standard model falls under the two criticisms I have previously mentioned. The aim of my paper is to show that 2 Philosophy and Social Criticism 0(0) the Deweyian approach based on the ideas of articulation and expressive domination is in a better position to resist those worries. As I aim to show, this is because it challenges the view that we can have standards of correct outcomes that are independent from decisionmaking practices. Problems are, to a certain extent, given to us, but they are also made by us in the practices in which we define them. To a relevant extent then, the standards by which we measure whether problems have been effectively defined and solved are immanent to those practices. My goal is to show that, by adopting this 'immanentist' view, we can avoid the criticisms raised by opponents of the standard epistemic account, while at the same time retaining the idea that democracy can be defended based on the quality of the decisions citizens and their representatives take. By adopting such an immanentist view, however, new difficulties arise. Hence, how are we then to measure the quality of the results of democratic decision-making processes if the standards on which we rely depend on them? Are we not caught in a vicious circle? In my view, John Dewey's insight that the articulation of problems is often pervaded by relations of domination 6 can contribute to answering this question and can provide epistemic reasons for supporting democratic decision-making. As such, if we assume, as Dewey does, that often dominant groups have more weight in defining collective problems than subordinate groups, and that this generates difficulties in dealing with and solving those problems, we can argue that a necessary condition for political decision-making to be successful at problem-solving is that relations of expressive domination are actively addressed in democratic public life and decisionmaking. This constitutes a proceduralist epistemic argument for democracy, understood as a set of practices of institutions aiming at taking decisions under conditions of participatory equality and inclusion. This view also recognizes the essential democratic role of political conflict and dissent. Certainly, effectively addressing relations of expressive domination does not guarantee that all problems will be properly articulated and solved. Nonetheless, it sets the necessary conditions for avoiding certain fundamental ways in which our collective inquiries into public problems are usually distorted. While it remains sympathetic to existing, Dewey-based epistemic accounts of democracy, 7 the aim of this paper is to provide further insight on Dewey's epistemic groundsthat is, grounds related to the possibility of identifying, defining and solving public problemsfor supporting the democratic norms of participatory equality and inclusion, and for acknowledging the essential role conflict plays in democratic decision-making. 8 This paper is divided in four sections. First, I flesh out Dewey's original, anti-dualist perspective on problems in section (1) and draw out their implications for political power and domination in section (2). Building upon Dewey's views, the two following sections turn to its consequences for an epistemic argument for democracy. In section (3), I draw out its implications for democratic norms of political inclusion and participatory equality in the sphere of political decision-making. In section (4), I show that my Deweyan argument does not require us to consider the inherently conflictual character of politics as superfluous to problem-solving. On the contrary, it makes dissent of contentious groups a necessary condition for the successful definition and solution of social problems.

Articulating the social world
What does it exactly mean that we articulate, find and make, social problems? In this section, I will present some of Dewey's most basic claims about what social problems are and how we inquire about them. Dewey argues that the origin of any inquiry, including our democratic inquiries about public problems, emerges from the need to cope with a practical challenge in which our habitual ways of dealing with our environment are interrupted. Some new factor appears: our partners in interaction behave in unexpected ways, a new law is introduced, social customs are challenged or modified by the arrival of newcomers, economic relations are altered. At the beginning, this new situation involves a certain degree of cognitive and ontological indetermination (Dewey 1986(Dewey [1938, 108). Let me explain this claim: At this primary stage, not only we, as agents, do not know what the exact nature of the problem we are facing. Dewey claims that the very elements that led to the interruption of our habitual ways of dealing with the environment are themselves indeterminate. It follows form this that, in order to know what the problem is, we need to turn the indeterminate situation into a more determinate one. This can be achieved only if we act in the world, modifying the elements of the situation and exploring their potential outcomes.
The exploration of those potentials is a constructive process. 9 As Matthias Jung (2009) points out, according to Dewey, when we inquire into a practical challenge, we do not merely generate a one-to-one explication of the implicated (implizierte) elements of the indeterminate situation this challenge generates (210-221). On the contrary, inquirers give to the indeterminate situation a specific shape, parting from the elements constituting it. This specific shape was only potential or implicit (implizit). Thus, the explication of the implicit always involves the realization of some of the potentialities allowed by partially indeterminate situations, leaving aside other potentialities (Alexander 2013;Frega 2010). 10 This constructive dimension of inquiry has been captured in pragmatist developments of the sociology of social problems. Gusfield (1981) argues that the famous example of the construction of the drinking-driving problem as a public problem shows that 'the target character is not a given, it is not in the nature of the reality as a Ding an sich (a thing it itself) but represents a selective process from among a multiplicity of possible and potential realities which can be seen as affecting auto fatalities and injuries' (Gusfeld 1981, 3). 11 When we inquire into those indeterminate challenges and turn them into determinate problems, the realization of potentialities represents a process that is reflectively directed by the inquirers. Inquirers direct both the 'actualization,' and the 'examination', as well as the 'selection' of certain potentialities to define a problem and find its solution. 12 Certainly, the actualizations of social potentialities, as they take place in everyday social life, are not always under the control of social agents. Rather the social world is being constantly articulated in ways many of which are neither reflectively nor democratically steered. When actors inquire into public problems, however, they reflect upon those developments, consciously stimulate new ones and select some of them in line with the final goal of resolving the challenges they face.

4
Philosophy and Social Criticism 0(0) Another important feature of Dewey's view of problem-articulations is that the realization of potentialities out of an indeterminate situation involves a hermeneutical as well as a practical dimension. Firstly, it concerns the hermeneutical or interpretative potentialities linked to the categories, concepts and values through which inquirers define social problems. This is particularly important for terms and meanings that are often hermeneutically open such as what Gallie calls 'essentially contested concepts ' (1955), but also involves many other concepts, categories and values aiming at capturing our experience of the social world. Inquiry into social problems involves the use of categories whose exact meaning needs to be interpreted and which might be the object of different, even conflicting interpretations (see Bohman 1991).
Secondly, inquirers also articulate problems by realizing the practical potentialities of the challenges they face. We experience the world from different practical angles and take some paths against others that disclose different aspects of the initial situation. Consider, for example, the changes in how women have historically experienced sexual harassment at the workplace, the different ways they have reacted to the behaviour of their male coworkers and how these different reactions have contributed to defining the problem in different ways both at individual and collective level (see Marshall 2003). In fact, both interpretations and practical dealings with social reality are constitutive elements of the reflective articulation of social problems and are deeply entangled. Hence, even if both elements can be analytically distinguished, they are closely related, since our experiences may change according to new interpretations of categories, concepts and values, and these may, in turn, vary according to the (unexpected) consequences of the new experiences we make through our practical interventions in the social world.
Problems are gradually articulated at different stages of inquiry. 13 In the public sphere, as in the natural sciences, we often need to deal with problems that have been previously defined, and which have their own history of articulations (Chateauraynaud 2012). One only needs to consider how notions such as 'sexual abuse', or 'drinking-driving' (Gusfeld 1981) have been transformed to understand that social problems that we usually experience as given, are actually the product of a process of (re-)definition. At other times, problems can be in an initial state of emergence, corresponding to what Dewey calls 'indeterminate situations', as when a new technology has been introduced, a new norm has been implemented, or a fully unexpected event such as the recent break into the US Capitol has taken place. Importantly, even in cases where problems seem to be fully articulated and fixed, new potentials of the indeterminate situation may appear, new experiences may be made, new meanings disclosed, which motivate us to re-define our well-established problems.
Finally, problem-articulation represents an expressive process by which democratic publics articulate their own identities by forming and manifesting, that is, by 'expressing' their interests. (Dewey 1984(Dewey [1927, 327). More generally, Dewey understands inquiry as a set of expressive practices in which inquirersindividuals or communities of inquiryto the extent that they act in the social world realizing its potentialities, undergo a (trans-) formative process of self-expression (see Dewey 1987Dewey [1934, chap. 4: The Act of Expression). In democracy, problem-articulation and the formation of publics are interrelated constructive processes in which the potentialities of the social world are Serrano-Zamora explored, selected and actualized, and in which problems and collective identities are articulated. 14

Expressive domination and problem-solving
But there is a further element that deserves special attention since it will play a fundamental role for the epistemic argument for democracy that I aim at providing along these lines. According to Dewey, the articulation of social problems is not free from relations of domination. On the contrary, dominant groups in society are often better positioned to realize the hermeneutical and practical potentialities linked to indeterminate situations in ways that best fit their own perspectives and forms of life. 15 They exert what I will call 'expressive domination'. Even though Dewey does not present a systematic theory of expressive domination, he provides some examples of the power dominant groups have to define social problems in ways that are favourable to their interests and conform to their world view. For example, in his Lectures in Social and Political Philosophy (2015) Dewey shows how historically dominant groups such as the Christian Church in the Western world have succeeded in defining any reaction against their power monopoly as the problems of isolated individuals instead as the struggle between different social groups. This definition of the problem of certain forms of social conflict as the conflict between the large majority and the deviating individual has contributed to strengthen their power monopoly for years until new, more group-based definitions of social conflict have succeeded to assert themselves in public life (see Dewey 1919Dewey /1929. Dewey's idea that problems are articulated both hermeneutically and practically contributes to understanding what is involved in expressive domination and what its political and ethical consequences are. In this context, Miranda Fricker's (2017) analysis of epistemic injustice seems to capture one important dimension of it, namely, the hermeneutical. Fricker has shown that non-dominant groups often suffer from the presence of gaps in what she calls the 'hermeneutical' resource of society (Fricker 2007, 147-175). These gaps refer to the lack of symbolic means needed for accounting for individual experiences in satisfactory ways. To understand the suffering of women produced by some men at the workplace as a form of 'playful flirting' or as 'sexual harassment' substantially influences the possibilities of coping with problematic situationsthat is, of describing it in a way that properly accounts for social suffering and for providing solutions to it (Marshall 2003). According to Fricker, these gaps are to be explained by a systematically produced unequal access to the spaces of symbolic power in society such as journalism or judicial institutions. Hence, Fricker argues that if only straight, white, men have access to those spaces, the possibilities of producing concepts that capture the experience of social minorities are substantially reduced. According to Fricker, notions such as 'sexual harassment' which seem to capture best women's experiences in the workplace could only arise when feminist activist organized communicative practices such as consciousness-raising groups. In those groups, all participating women had an equal voice and could discursively share their individual experiences. Only then were they able to generate categories that worked as effective hermeneutical 6 Philosophy and Social Criticism 0(0) resources capable of properly accounting for their suffering and searching for solutions to it (Fricker 2007;Seigfried 1996, 153). But this is only a partial account of how domination pervades the articulation of social problems and of how it can be overcome. The latter also concerns the equal opportunities of some social group to freely access the realization of the practical potentialities of indeterminate situations. In On Female Body Experience (2005), Iris Marion Young shows the meaning of this latter idea by drawing on the examples of pregnancy, clothing and menstruation. According to Young, all these relevant aspects of female experience are shaped by a male-dominant culture which makes possible certain experiences and practices while others are blocked. Young shows how, following from a medicalized interpretation of pregnancy as a disease, certain practices tend to alienate women from their own experience as pregnant: the normal procedures of the American hospital birthing setting render the woman considerably more passive than she need be. Most hospitals, for example, do not allow the woman to walk around even during early stages of labor, even though there is evidence that moving around can lessen pain and speed the birthing process. (Young 2005, 58) By limiting the range of activities a woman may engage with, all these practices influence the way issues of pregnancy may be articulated (or not) as problems. By making pregnant women more passive, institutional practices of birthing setting exclude practical possibilities that would allow to re-define pregnancy as a non-pathological state.
In sum, expressive domination concerns the capacity of citizens to participate in the process of the configuration of social problems under conditions of equality, both hermeneutically and practically. A central aspect of Dewey's view is that expressive domination has negative consequences for the capacity to articulate into problems the practical challenges citizens need to face in ways that are conducive to effective solutions. Hence, while problem-articulation is to be seen as a constructive process, this does not need to be read as if there were no better and worse ways of articulating problems.
We should thus distinguish three ways in which expressive domination hinders a proper articulation of problems. First, indeterminate situations cannot be solved effectively if some dominant social groups have been able to articulate them one-sidedly, generating problem-distortions (1). Second, dominant groups may block the very possibility of articulating indeterminate situations. On the one hand, situations might remain too vague to be properly handled by social actors; on the other hand, situations may be kept too determined for new, alternative definitions to be articulated (2). Third, domination may also entail articulating problems such that the articulation of other problems can be negatively affected (3). Surely, these three possibilities are not mutually exclusive and may take place simultaneously and reinforce each other. Let us briefly look at each of them: (1) As the cases I have presented show, dominant groups often manage to articulate indeterminate situations into problems excluding other, often subaltern groups. In this way, the specific definition of those problems only fits the interests and Serrano-Zamora worldviews of their members. 16 The question, therefore, is why this should be a problem for addressing the practical challenges we need to face in situations that are indeterminate. Have we not argued that the potentials of indeterminate situations can be developed in many ways? How can we then distinguish between good and bad problem-articulations? A way to answer this question is by pointing to Dewey's idea that the practical challenges we face are composed by many elements and factorspractices, meanings, facts, objects, valuesrelating to each other complex ways. As in the case of 'sexual harassment' or 'birthing', problemarticulations carried out in a context of expressive domination by some groups only realize the hermeneutical and practical potentialities of certain factors, leaving other elements of the situation latent or repressed. In this case, unilateral problem-articulations can be expected to fail to properly address the objective elements of the practical challenge at the source of the indeterminate situation. They generate problem-distortions. A male-dominated culture contributes to articulating problems that hardly relate to the situations of practical challenge experienced by many women. The dominant group often provides a description of a situation that makes invisible certain details that are relevant for women's experience. It also often imposes the interpretation of certain categories or values in ways that do not capture relevant aspects of women's experience. In these cases, any honest inquiry into work relations and pregnancy, for example, must take the possibility into consideration that what appears as a proper articulation of a practical challenge represents, in fact, a false understanding of the objective features of the indeterminate situation. 17 But does this understanding of problem-distortion not amount to accepting the main premise of the 'standard account' of the epistemic defense of democracy? Are we then not assuming that there is something like an independent standard of correct problemarticulations which democratic decision-making should be able to track? The answer to these questions is both yes and no. On the one hand, Dewey considers that expressive domination distorts problem-articulations because it makes the latter unable to capture the practical challenges that are involved in the interruption of our habitual dealings with the world. On the other hand, without actively inquiring into the world, those practical challenges are so indeterminate that they cannot provide by themselves a standard of correct problem-articulations. On the contrary, this can only be done once practical and hermeneutical potentialities of a situation have been sufficiently developed into problems we can deal with. And this can only be accomplished by regular citizens, scientists, representatives, or any other group or institution inquiring into collective problems. 18 To this extent, we cannot talk, as defenders of the standard account of epistemic democracy do, of a standard of correctness that is independent of the practices by which we articulate problems. Certainly, there are better and worse articulations of indeterminate situations into problems, since the latter may correctly address the objective features of the practical challenges we face. Contrary to what the standard epistemic argument for democracy would assume, however, indeterminate situations themselves cannot offer an objective point of view from which problems and their solutions can be assessed. Rather, this point 8 Philosophy and Social Criticism 0(0) of view must be actively produced, and it is precisely the way in which this happens that determines the possibilities for the successful articulation of practical challenges into problems so that they can guide us towards their solution.
(2) The second epistemic deficit that may arise from relations of expressive domination is the blocking of problem-articulations. This may take place in two opposite ways. On the one hand, situations leading to problems may be kept so indeterminate that they cannot be properly grasped as concrete problems. In other words, they may be kept too vague to be properly handled. Domination then concerns the inability of certain groups to have access to certain experiences and practices, as well as to discursive spaces of interpretation, so that the potentialities of indeterminate situations cannot be sufficiently realized to become an object of inquiry. On the other hand, relations of expressive domination may contribute to experiencing prevalent problem-articulations as fundamentally fixed, as if they represented the only possible way to articulate a situation of practical challenge. (3) Finally, the third epistemic deficit follows from the multiplicity of problems democratic societies need to deal with. In public inquiries, we are not only dealing with one problem but with many problems at the same time (or at different times). The way we define and solve a problem, we fix social realitythat is, we fix the meaning of its categories, identities, values and possible usesin ways that affect the possibilities of defining and solving other problems and of prioritizing problems that need to be addressed urgently versus problems that are framed as 'not so urgent'. Here expressive domination affects the ability of some groups to articulate social reality in ways that block the possibility for other groups to articulate the practical challenges affecting them.

Political inclusion and equality
In this section, my aim is to show that the view developed above provides epistemic groundsthat is, based on the quality of the outcomes of democratic decision-making processes 19for defending maximal political inclusion and equality in participation, two norms guiding democratic institutions and practices. 20 As will be argued, this view is less vulnerable to critique than what I have called the 'standard account' of the epistemic justification of democracy, according to which there is a standard of correctness for political decisions that is independent from the practices by which we take those decisions. First, I show why the standard account fails at providing a convincing epistemic underpinning of both democratic norms. In a second step, I develop an epistemic proceduralist argument based on the idea that reducing expressive domination is a necessary (though not sufficient) condition for producing good problem-articulations. In a third step, I argue that this reading contributes to (re-)interpreting the meaning of those democratic norms. Firstly, I follow Cristina Lafont in arguing that, if we accept that there are standards of correctness that are independent from decision-making practices, and that democracy can be defended on its capacity to approaching or 'tracking' those standards, the risk of falling Serrano-Zamora into some form of exclusionary epistocracya government of the 'knowers' that does not overlap with that of the members of the political communitybecomes acute. This is because, under these premises, we cannot exclude the possibility that a group of individuals that does not overlap with the political community may know what the correct decision is. In the literature, we can find a strong and a weak formulation of this worry. Defenders of the stronger version argue that including some (groups of) citizens in democratic decision-making can take us far from the correct decision. 21 In this regard, authors like Helène Landemore (2013), who provides one of the most sophisticated epistemic defenses of democracy based on the standard account, have convincingly argued that problems can be best solved if a broad diversity of perspectives is included in decision-making. Landemore also argues that this diversity can be guaranteed only if we include the largest amount of citizens possible. However, one has good reasons to wonder if Landemore's view can resist a weaker version of this criticism, as it has been formulated by Lafont. Lafont's critique draws Landemore's assumption that, an implication of this [epistemic, JSZ] argument is that, to the extent that including everyone is not feasible, an alternative solution is to restrict the group of problem solvers to a representative sample of the larger cognitive diversity: a group of representatives chosen by lottery. (Landemore 2014, 188, quoted from Lafont 2020 According to Lafont, this lottocratic solution to the problem of feasibility is problematic to the extent that it 'excludes the majority of citizens from political decisionmaking by leaving it up to the few, randomly selected citizens who become members of the assembly' (Lafont 2020, 97). Thus, 'those not selected via lottery would simply lose the ability to influence or shape political decisionsbe it through voting for those who represent their political views, joining political campaigns, or running as representative' (97). Thus, 'a representative sample may suffice for engaging in the deliberative search for the best answers, whereas the rest of the citizenry becomes dispensable'. According to Lafont, while this lottocratic solution can be justified by an epistemic argument for democracy that is based on the substantive quality of its outcomes, it 'has nothing to do with enabling citizens to participate in co-shaping the outcome of the decision-making process' (97).
Lafont's criticism is that, if we base our epistemic argument for democracy on the quality of the outcomes of decision-making processes, we can come to justify forms of decision-making that fail to 'enable citizens to participate in co-shaping the outcome of the decision-making process'. Importantly, my claim is that this is not necessarily the case, since Lafont's worry applies only to a view that the quality of the outcomes of decisionmaking can be measured based on an independent standard of correct outcomes. The Deweyan account I have previously outlined contends that, since social problems are not merely given but also made, and since expressive domination negatively affects the possibility of providing good problem-articulations, democratic systems must ensure that dominant groups cannot develop the hermeneutical and practical potentialities of social problems unilaterally, that is, in conditions of expressive privilege. Hence, a necessary condition to avoid expressive domination is that every citizen potentially concerned by a

10
Philosophy and Social Criticism 0(0) practical challenge can equally participate in the development of the potentialities of the indeterminate situation. This account thus challenges the worry that a limited group of citizens may have some form of privileged access to the correct outcome of a political decision-making processes. This is so because, contrary to other strategies based on the standard account, the idea that citizens' main epistemic task consists in realizing the potentialities of the problematic situations they face, excludes the possibility of an epistemic privilege of some reduced group of citizens. Since the only way to figure out a problem-articulation that properly addresses the practical challenges we need to face is by developing its hermeneutical and practical potentialities, we cannot systematically exclude anybody from equally participating in that process under the risk that some important aspect of the situation remains systematically obscured. In my view, the epistemic argument for democratic inclusion and equal participation strengthens Dewey's idea that democracy cannot be limited to its strictly political form. 22 The reason is that, for democratic systems to be able to articulate and resolve citizen's problems, democratic societies must fight expressive domination not only regarding political decision-making but also the rest of spheres of social interaction of a democratic community. Hence, the practical challenges and the potentials of social reality are not only articulated in the political sphere but extend to all our social relations. Dewey's idea that democracy is a way of life finds then its epistemic counterpart in the insight that a democratic society, in order to solve its problems, needs to address relations of expressive domination. 23

Articulation and dissent
A second challenge to the standard account can also be avoided by the present articulative view, namely, one related to the essential role of dissent and democratic conflict play in political decision-making. According to Urbinati's criticism of epistemic accounts of democracy based on the quality of outcomes (2014), democratic procedures 'presume [...] that dissent (which diversity can engender) is good as an injection of vitality and reviewability into the democratic process, not yet necessarily a means to truer outcomes'. (98) In fact, 'while truth tends to overcome dissent, democratic procedures presume dissent always'. (98) Moreover, "when the political arena is inhabited by conflicting interpretations of what a true idea means, compromise between them looks difficult to achieve and logically impossible. Since the opposite of truth is error, it makes no sense to tolerate an error unless those who hold it see it as a temporary error to be overcome". (99) In her defense of the standard account, Landemore has provided a convincing response to Urbinati's (and Rawls') concerns (Landemore 2017). This consists in arguing that we do not need to operate on the basis of an intolerant and sectarian notion of truth. Rather, we can 'put forward a pluralized concept of truth by which truth would mean different things depending on the domain of application'. (Landemore 2017, 285) This would give space for conflict and dissent as part of our attempts at tracking correct political outcomes. Nonetheless, Landemore's view seems to be more vulnerable to Lafont's weaker critique to the standard account of epistemic democracy, namely, that it 'ignore[s] and stipulate[s] away the need for such political struggles to actually take place and succeed'. (Lafont Serrano-Zamora 2020, 99, author's emphasis). According to Lafont, the problem is not so much that assuming an independent standard of correctness or truth makes any democratic dissent impossible, but that it makes it unnecessary because it can be substituted by other means: The main problem [of the epistemic argument for democracy based on the quality of outcomes, JSZ] concerns its systematic implications for democratic theory. For this assumption stipulates away the democratic significance of political disagreement. It is one thing to contendagainst deep pluraliststhat political disagreements can be overcome. It is quite another to stipulate political disagreement away in assuming that, once decision makers hit on the right political answers, agreement by decision takers will simply follow. This assumption eliminates a task (and an epistemic dimension) of political deliberation that is quintessential to democracy, namely, the need to reach agreement with others by justifying political decisions to them with reasons that they can reasonably accept so that they can identify with them and endorse them as their own. This step, of course, would be superfluous if one assumes that the best solution to a political question 'must be obvious to all of them when they are made to think of it'. (Lafont 2020, 99) As in the previous section, I contend that the Deweyan approach outlined here is not vulnerable to this charge. Rather, it considers conflict and dissent as necessary elements of any democratic decision-making process that aims at effectively articulating and solving citizens' problems. The reason for this is twofold. First, dissenting groups often bring together individuals whose social experiences challenge the articulations of social problems that are hegemonic in the political debate. This function is essential since hegemonic problem-articulations may be distorted even when democratic norms of inclusion and participatory equality are respected and when participants in public discussion are willing to be reasonable. As James Bohman (1996) notes, '[...] many prejudices, ideologies, and biases may be too widely shared to be eliminated, even if all citizens intend their reasons to be public, unless new publics emerge and change the context of deliberation'. (208) Against this background, contending minorities often question what they see as the result of epistemic deficits such as problem-distortions and promote what they take to be more correct views on the problems they articulate. This takes place in two steps: challenging the 'taken-for-granted' status of hegemonic problemarticulations and proposing alternative articulations that better address the challenges they face. Regarding the first step, Bohman argues that cultural codes entrench taken-for-granted meanings, making them given 'facts' rather than social constructions; they 'appear as transparent descriptions of reality, not as interpretations, and are apparently devoid of political content.' Such transparent descriptions and accepted meanings guide definitions of problematic situations, and in so doing they limit the possibilities that are available to deliberators. collective actors in social movements have to challenge this taken-for-granted character and show these meanings to be only some of many possibilities, as the women's movement has done with gender identity (Bohman 1996, 209) But this disclosive function is certainly not the only contribution to problemarticulation social movements make. Contending groups also argue that their articulations are more adequate than the hegemonic ones, which they see as distorted. In other words, contending groups are not only concerned with showing that other articulative possibilities of practical challenges exist, but also with showing that their articulation responds in a more adequate manner to the partially indeterminate situation. The previous example of the dispute around the proper articulation of women's experiences of sexual harassment at the workplace is paradigmatic for this kind of conflict. Feminist attempts at bringing the notion of 'sexual harassment' to the public and legal spheres represents the attempt to present new, better articulations of social problems. 24 The second reason is linked to the fact that contentious politics entails a certain level of collective organization under conditions of interaction that differ from hegemonic ones. This often allows for the generation of 'free spaces' (Croch 2001) where non-hegemonic ways of experiencing and interpreting social reality are made possible. 25 As such, they often represent spaces where some of the hermeneutical and practical potentials of practical challenges that often remain obscured by existing relations of domination can be collectively experienced and articulated. This allows dissenting individuals both to gain confidence about their own social experience of problematic situations and to develop new views that contribute to disclosing the meaning of the problematic situation. At the local level, this makes possible the correction of epistemic deficits such as problemdistortions and problem-indeterminations. As in the case of consciousness-raising groups, these free spaces for hermeneutical and practical articulation are often necessary if an effective struggle in the broader public sphere is to be pursued. 26 These arguments not only show that dissent is compatible with an epistemic approach to democracy based on the substantive quality of democratic decisions. Considering that epistemic deficits will take place even in conditions of inclusion and equalitywhich represent, as I have argued in the previous section, necessary conditions for avoiding those deficitsthey also show that dissent and contentious politics are also necessary conditions for promoting good problem-articulations. Furthermore, as unique spaces for world-disclosure and experience, they represent a non-substitutable dimension of democracy's epistemic dimension.

Conclusion
This paper has outlined a weak epistemic argument for democracy that draws on Dewey's idea that in decision-making citizens articulate (discover and make) social problems, and that problem-articulation is pervaded by relations of domination. I have presented some central elements of this expressivist view to explore some of its implications for an epistemic approach to democracy. To put it succinctly, the paper provides a proceduralist argument for democracy grounded on the claim that the articulation of problems under conditions of expressive domination generates problem-distortions, as well as other kinds of obstacles to addressing the practical challenges citizens must face. I have argued that this view is better prepared to resist the criticisms that have been addressed to the 'standard accounts' of the epistemic argument for democracy. First, it resists the challenge Serrano-Zamora that some limited group of citizens may have access to the independent standard of correctness that democratic practices and institutions would be meant to track. While one may draw this conclusion from a view that there is a pre-existing standard of correct decisions which needs to be 'discovered', the ideas of problem-articulation and expressive domination exclude this possibility. Importantly, I have argued that this argument contributes to rethinking the meaning of the democratic norms of maximal inclusion and equal participation and to underpinning the Deweyan thesis according to which democracy cannot be reduced to its strictly political forms.
Secondly, contentious politics represents an essential condition for avoiding the epistemic deficits of collective articulation given the pervasiveness of biases and distorted worldviews. Firstly, in situations where the political debate is pervaded by biases, contentious politics contributes to rearticulating problems in the public sphere. Secondly, political conflicts are themselves modes of realizing social possibilities that would otherwise remain latent. As 'free spaces', they are the condition for the articulation of problems in ways that challenge their hegemonic articulation.
The present approach differs from the standard account of democracy to the extent that the latter assumes that an independent standard of correct decisions or right solutions to problems exists. I have shown that by denying the independence of a standard from the processes of inquiry, my approach can better resist Urbinati's and Lafont's criticisms. However, dispensing with an independent standard of correctness does not entail the denial of the existence of (weak) standards for the articulation and solution of problems. To this extent, my epistemic argument can avoid the deep pluralist view according to which we need to move away from any idea of correct outcomes. This anti-epistemic move would consist in the rejection of the idea that democratic decision-making has as its essential aim the effective resolution of collective problems (Urbinati 2014, Mouffe 2005). My Deweyan claim is that, at the beginning, practical challenges represent indeterminate situations that must be transformed into concrete problems, and that epistemic deficits are likely to arise when this is not done in ways that reduce expressive domination. There cannot be a privileged group of knowers with exclusive access to the right outcome. 27 Nor can there be a group of agents who would be systematically excluded from the possibility of realizing some of its ontological and hermeneutical possibilities without harming democracy's problem-solving function.

Aknowledgments
I would like to thank for their helpful comments Lisa Herzog, Matteo Santarelli, Tullio Viola, Emmanuel Renault, Arvi Särkelä as well as the audiences of the ESPF Grundlegung at the University of Groningen, the workshop "Struggle, Conflict, and Critique" at the Universitat de Barcelona, and the journée d'études "Naturalisme, Philosophie Procesuelle et Democratie Participative chez John Dewey" at the Université Paris Nanterre.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/ or publication of this article. in a map of the globe. This insertion, moreover, was not merely additive, but transformative of a prior picture of the world as to its surfaces and arrangements. It may be replied that it was not the world which was changed but only the map.
[…] The map of the world is something more than a piece of linen hung on a wall. […] A potential object for further exploration and discoveries now existed in Europe itself; a source of gold,; an opportunity for adventure; an outlet for crowded and depressed populations […]; in short: an agency of new events and fruitions, at home as well as abroad ' (124-125). 10. This is what is involved in Dewey's characterization of inquiry: 'Organic interaction becomes inquiry when existential consequences are anticipated; when environing conditions are examined with reference to their potentialities; and when responsive activities are selected and ordered with reference to actualization of some of the potentialities, rather than others, in a final existential situation'. (Dewey 1986(Dewey [1938, 111) 11. Also drawing on the pragmatist tradition, Daniel Cefaï (1996) has developed a sociological approach to public problems such as the 'new poverty', 'malaise of the banlieues', or that of 'voluntary abortion' that also denies that problems are purely objective, as functionalists would argue, or, on the contrary, purely the construction of social actors. According to Cefaï, we need to go beyond this dichotomy, understanding the emergence of public problems as processes of narrative 'configuration' (Gestaltung), where objective and subjective elements are simultaneously at play. 12. See note 10. 13. Hendriks and Dzur (2021) show this difference in comparing how citizens governance spaces deal with indeterminate problems that will be determined during their problem-solving activities while Ostrom's (1990Ostrom's ( , 1996 examples deal with more determined problems such providing access to clean water. 14. On the possibility of linking Dewey's epistemic approach to the tradition of expressivism, which goes back to the work of Johan Gottfried Herder up to Charles Taylor (1985Taylor ( , 1989Taylor ( , 2015. For a more detailed discussion, see Jung 2009, Serrano Zamora, 2017, 2021, Viola 2019. See also Anderson andPildes 2000. 15. As Christopher Ansell (2011) has put it 'politics and power are refracted through the experiential basis of concepts. Power is exercised by controlling the experiential associations that people develop with concepts and by controlling the associations between concepts'. (34) Knight and Johnson (2011) have also discussed the idea of power in the context of Dewey's theory of democracy. 16. Note that, for Dewey, interests are also articulated (see Santarelli 2019). 17. Note that this does not mean that there is only one right articulation of a problem. My claim is just that there are wrong interpretations. 18. As we will see later, here the 'articulatory' activity of dissenting groups such as feminist, environmentalist, or workers movements also plays an essential role, since they generate nonhegemonic spaces for problem-articulations that are essential to redress existing problemdistortions. 19. Lafont argues that there are two ways of understanding the epistemic function of democracy: an outcome-based view and a justification-based view. 20. Note that the aim of this paper is not to argue that there may be alternativeepistemic and nonepistemicarguments for democracy. In my view, an epistemic argument for democracy of the kind I am defending here is compatible with non-epistemic defenses of democracy based on values such as autonomy and equality. See Landemore 2017. 21. see Urbinati 2014. 22. See Dewey 1984[1927. For a discussion, see Frega 2019. 23. This paper adds an epistemic argument to the Dewyean ideal of democracy as non-domination (see Rogers 2009). 24. In her empirical survey, Ana-Maria Marshall (2003) has reconstructed how understandings of 'sexual harassment' have developed in women's claiming for their rights in US courts. 25. Mansbridge and Morris (2001) provide an overview of different empirical examples. 26. As for the disability rights movements, Sharonn Croch (2001) has shown how central the creation of 'free spaces' is for the building of certain frames by which disabled people articulated their specific problems. 27. This does not mean that some social groupfor example, scientists, victims of oppression and locals affected by some environmental problemmight have better access to problematic social situations and need to be more seriously listened to than members of other groups.