Deliberation and Non-Deliberative Communication

The goal of this work is to clarify how certain democratic goods — notably, empowered inclusion and mutual respect — can be both antecedents to and outcomes of successful communication. When exclusion or a lack of basic mutual respect prevent deliberation from happening in the first place, where do the antecedent conditions of empowered inclusion and mutual respect come from? To answer this question, I propose distinguishing between deliberation and non-deliberative communication. More specifically, I offer a typology that distinguishes between deliberation, political communication, non-political reason-giving and non-political communication. This framework clarifies theoretical disputes and empirical mixed findings in the deliberative democracy literature and offers insight to practitioners and activists interested in using communicative practices to achieve aims related to incentivizing inclusion or promoting mutual respect.

this as the paradox of 'when water chokes'. When water is blocking one's windpipe, more water hardly seems like the solution. Similarly, if inequality or disrespect are blocking discursive influence and the uptake of reasoned arguments, how is more deliberation the solution? I propose addressing this paradox by distinguishing deliberation from non-deliberative (including non-political) communication. In the first section, I propose a general typology for distinguishing between deliberation, nondeliberative political communication, and non-political talk.
The second section considers communication that is oriented to collective issues (deliberation and political communication). In this section, I outline the problem of inequalities and exclusions that block deliberation. I describe the role that non-deliberative political communication plays in empowering inclusion when the prerequisite degree of equality and inclusion that are normally required for deliberation are absent. The third section considers communication that is not oriented to collective issues (non-political talk). In this section, I outline the problem of a lack of mutual respect that discourages interlocutors from engaging one another in deliberation. I describe how non-political, personal expression can build bonds of reciprocal interdependence when the prerequisite degree of mutual respect that is normally required for deliberation is absent. I conclude by pointing to new potential frontiers in the study of deliberation and democracy.

Antecedents and Outcomes
As Mansbridge et al. (2012: 11) explain, mutual respect is 'intrinsically a part of deliberation… to fail to grant to another the moral status of authorship is, in effect, to remove oneself from the possibility of deliberative influence'. Making a similar point, Gutmann and Thompson (1996) argue that deliberation can only address deep moral disagreements when a precondition of reciprocity and respect are met. Mutual respect is a prerequisite to deliberation because it induces both sides to make reasonable arguments the other could accept. Mansbridge et al. (2012: 11) also describe mutual respect as an outcome of deliberation: deliberation's ethical function is to 'promote mutual respect among citizens'. It is possible for deliberation to both require and cause mutual respect. However, a logical tension arises when there is very little or no mutual respect in the first place. If agents have already removed themselves from the possibility of deliberative influence by denying one another the moral status of authorship, deliberation cannot create mutual respect. This logic is formalized in the following set of statements that describe a variable (x) as an antecedent and an outcome of another variable (y): • Claim 1 (antecedent claim): Having x creates the possibility of y. • Claim 2 (outcome claim): y increases x.
Both claim 1 (the antecedent claim) and claim 2 (the outcome claim) can be true. But if claim 1 is true, then as x approaches zero the possibility of y occurring will also approach zero. This means that if both claims are true, then when the antecedent (x) is absent the outcome (y) cannot catalyze the positive feedback loop that is formalized by claim 2. To clarify what this means, consider an intuitive example: • Claim 1: Having capital creates the possibility of rents.
When a person has capital (from property or investments) they can accumulate rents which in turn will -almost by definition -increase their capital. When a person has no capital to begin with, they cannot use rents to increase their capital. The same logic applies to the relationship between mutual respect or inclusion and deliberative reason-giving. Consider: • Claim 1: Having mutual respect (the moral status of authorship) creates the possibility of deliberative influence. • Claim 2: Deliberative influence increases mutual respect.
When interlocutors grant one another the moral status of authorship they create the possibility of deliberative influence which in turn will -almost by definitionincrease mutual respect. When interlocutors deny one another the moral status of authorship, they cannot then rely on deliberative influence to increase mutual respect. When the absence of mutual respect is the problem, speakers cannot expect deliberation to be the solution. So, if more deliberation isn't the solution, what is? As others have pointed out, it is not entirely obvious where the precondition of mutual respect comes from (Dryzek 2005;O'Flynn 2007). Bohman (2000: 384) offers a thorough analysis of the paradox of ' communication that violates its own conditions of success' by considering the problem of structural inequalities that entail exclusion. Like mutual respect, equality and inclusion are also both antecedents and outcomes of deliberation. Although reciprocal justification can break down inequalities and empower inclusion, a degree of empowered inclusion is required before deliberation can occur at all. Inequalitiesasymmetrical empowerments -entail exclusion because those who are disempowered in social hierarchies are prevented from participating in or influencing public speech (Beauvais 2019a;Beauvais 2019b;Beauvais 2018). Furthermore, those who are empowered by social hierarchies can rely on asymmetrical empowerments to bypass normal communicative constraints, such as the requirement to listen and take seriously others' claims, to be open to alternatives and to treat others with respect and reciprocity (Bohman 2000;Young 2000).
As a solution to this problem, Bohman (2000: 386) points to the role of the social critic or social scientist whose job is to 'unblock communication by making distortions explicit'. However, this seems to assume that the social critic enjoys some kind of requisite empowerment, such as when the social critic enjoys a prestigious post as a university professor (and perhaps some of the correlated benefits stemming from class, gender and race). What happens when these initial empowerments are absent, such as when the social critic is a working-class Black woman in a deeply racist, sexist, and classist society? If the problem is that she is excluded from political speech and cannot influence others with her reasons, what good is more deliberation?

A Typology of Communicative Practices
This puzzle is addressed by distinguishing more clearly between different types of communicative practices that are each better suited to achieving different outcomes that are essential for democracies. I recommend distinguishing between discursive practices by asking at least two questions: 2 Are discursive practices oriented to issues of collective concern? And are discursive practices characterized by reciprocal reason-giving?
The answers to these two questions generate a typology of four sets of communicative practices (Table 1). First, deliberation, or practical discourse oriented to issues of collective concern that is characterized by reason-giving. 3 Second, non-deliberative political communication, or communication oriented to issues of collective concern that are not characterized by reason-giving. Third and fourth are non-political talk. Non-political talk is not oriented to issues of collective concern and can involve reason-giving or not. Because both types of non-political communicative practices achieve similar democratic aims, I mostly consider non-political discursive practices together, regardless of whether they involve reason-giving or not.
The very broad framework I propose also does not distinguish forms of communication based on the style of communication or level of formality. Drawing on Young (2000), I recognize that all forms of communication -whether deliberation, non-deliberative political communication, or non-political talk -can occur in a range of different styles including legalistic argumentation, rhetoric, greeting, testimony, storytelling, and negotiating. Drawing on Mansbridge (1999), I recognize that different forms of communication occur in both public arenas and formally private spaces. This broad framework also does not distinguish forms of communication based on the intentions of the speakers -for instance, whether the speakers are behaving strategically or whether communication is genuinely oriented to agreement or understanding. 4

Talking About Collective Issues
The relationship between deliberation and democracy is well-theorized. Insofar as basic conditions of justification -including equality and mutual respect -are met, offering and hearing reasons about issues of collective concern achieves a number of outcomes that are essential for democratic political systems. For instance, the process of hearing and articulating reasons for public policies helps people link their private preferences into collective opinions and agendas that underwrite collective decisions. When people perceive their preferences reflected in collective outcomes, they are more likely to perceive those outcomes as being legitimate (Chambers 1996;Habermas 1998;Manin 1987;Mansbridge et al. 2012;Warren 2017;Beauvais & Warren 2018).
The boundary between deliberation and nondeliberative political communication -talk that is oriented to matters of collective concern that is not characterized by reason-giving -should be flexible and permeable. Any time political communication is problematized -such as when a speaker raises a challenge or makes a request for clarification -political communication should switch to deliberation. When relationships are characterized by a degree of equality and mutual respect, requests for clarification prompt deliberative reason-giving (Chambers 2017). When relationships are characterized by a degree of equality and mutual respect, asking 'why?' or 'what do you mean?' should be enough to incentivize your conversation partner to switch from making statements about collective issues without justification to the practice of deliberation (and offer compelling reasons for their statements).
However, as I started to explain, asymmetrical power relations can block reason-giving and thus prevent misunderstanding or disagreement from being addressed through the forceless force of compelling reasons. The importance of distinguishing between political communication and deliberation is illustrated precisely in the moments when political communication is problematized by misunderstanding or disagreement but fails to become deliberation. Consider the power asymmetries between employers and their employees. When an employer expresses a political opinion, an employee who disagrees might not be secure enough to challenge the employer at all. Even if the employee does challenge their employer, the employer might take advantage of their position of authority by ignoring or scoffing at the employee's challenge. While not all idiosyncratic power asymmetries undermine democratic health, 5 democratic systems typically involve competing interests and experiences, and justice requires that people can raise and challenge the validity of claims on topics that affect them. The problem I am centrally concerned with happens when structural inequalities -inequalities between salient social groups in a society -systematically prevent disempowered social group members' discursive challenges from receiving uptake. For instance, in capitalist societies characterized by racialized and gendered classhierarchies, men and White people are more likely to be employers (who give orders) and women and people of colour are more likely to be employees (who receive orders) (Davis 2011;Young 2011). When structural inequalities prevent misunderstandings or disagreements about collective issues from being addressed through reason-giving, communication is far less likely to achieve the outcomes which are essential for well-functioning democracies. Being misunderstood and not being able to explain oneself, misunderstanding others and not receiving clarification, or experiencing disagreement and not explaining/hearing justifications for different sides of an argument will have measurably different effects on communication partners than the process of being misunderstood and being empowered to explain oneself, misunderstanding others and receiving clarification, or experiencing disagreement and explaining/hearing justifications for different sides of an argument. Distinguishing between deliberation and political communication -recognizing that reason-giving matters -has implications for the empirical study of communication. Many empirical studies do recognize the importance of reason-giving. The Discourse Quality Index, the earliest attempt to measure deliberation empirically, includes justification as a central feature of deliberative quality (Steenbergen et al. 2003). However, reason-giving is an important missing variable in many well-cited studies of deliberation.
For instance, Mutz (2006) uses an index of intense, sustained political disagreement in communication networks to measure ' deliberation', but does not identify whether sustained political disagreement is characterized by reason-giving. 6 One of Mutz's key claims is that deliberation suppresses voting. Mutz's central finding contradicts other research showing that deliberation mobilizes voter turnout and other acts of political engagement (Gastil, Deess, and Weiser 2002;Knobloch and Gastil 2015). Recognizing the importance of reasongiving points to an explanation for these mixed findings. It seems plausible that Mutz is actually measuring the effect of experiencing disagreement in political communication that fails to be addressed through reason-giving, perhaps because power asymmetries constrain speakers' abilities to effectively raise reciprocal validity claims. This intuition is supported by the fact that the kind of intense, sustained disagreement in communication networks Mutz identifies is more common among the disempowered. In the United States, people of colour the poor, and the least educated perceive the most sustained disagreement in conversation networks (Mutz 2006).
This contrasts with other studies that operationalize deliberation as communication organized into highquality discursive forums (juries and deliberative mini publics) that are organized to address matters of collective concern (Gastil, Deess and Weiser 2002;Knobloch and Gastil 2015). It is possible that in societies marked by racialized and class-based social hierarchies, organizing disagreement into democratic innovations helps neutralize inequalities in these institutional settings. Future empirical research could more explicitly compare the consequences of leaving disagreements about collective issues unaddressed in the public and addressing disagreements about collective issues through deliberative reason-giving by organizing talk into democratic innovations that neutralize inequalities. I hypothesize that the former (leaving political disagreement unaddressed) reinforces political exclusion by suppressing other political practices, including voting. By contrast, I suspect that the latter (neutralizing inequalities and exclusions to address disagreement through deliberation) reinforces political inclusion by mobilizing other political practices, including voting.
In mass democracies there will be many conversations characterized by disagreement about collective issues, and there are many instances where structural inequality and exclusion will block reason-giving. Not all disagreements under conditions of structural inequality can be channelled into institutions such as mini-publics that neutralize inequalities; democratic innovations are rare (Chambers 2009). The solution to inequalities that block reasongiving in the unstructured public is often more disruptive political communication. When inequalities systematically prevent those who are affected by collective outcomes from speaking or influencing others with reasons, it is time to start making accusations, issuing declarations, shouting polemics and making other assertions without justification (Walzer 1999).
Disruptive communication about collective issues that is not characterized by reason-giving is an effective tool for calling people to arms, 'to capture their interest, focus their energies, draw them tightly together' (Walzer 1999: 60). While there should be good reasons for shouting slogans or making statements, the acts themselves do not have to be characterized by reason-giving -slogans and statements are often shouted and presented as assertions without accompanying justifications. 7 Non-deliberative political practices -including political communication and other prefigurative, non-discursive acts (Rollo 2017;Williams and Warren 2014) -are essential for allowing otherwise silenced voices to be heard. Disruptive political communication is essential for incentivizing political inclusion under conditions of structural inequality. Scholars interested in the relationship between inequality and political activism might try to clarify the conditions under which inequalities and exclusions block deliberation and the disempowered become more politically disengaged, versus the conditions under which inequalities and exclusion block deliberation and the disempowered become mobilized to engage in more disruptive political communication to break down exclusion.

Talking About Personal Experiences
Not all communication is oriented to collective issues. Often, talk is expressly framed in terms of personal experiences. Communication that is not oriented to collective issues, or non-political talk, can be about matters the public should care about (see Mansbridge 1999) but it differs from political talk because it is not framed as such.
Instead, non-political talk is framed as a personal or private experience (a feature of personal biography), rather than as a public or collective issue (a feature of society and collective history). Non-political talk might be oriented to establishing interpersonal connections, developing selfor collective identities, or might be purely expressive.
The key to understanding the distinction between nonpolitical and political talk is understanding the distinction between framing something as a personal experience ' of the milieu' versus framing something as 'public issues of social structure' (Mills 2000: 8). A person may complain about being poor and how humiliating they find the experience of seeking personal assistance. But so long as they frame this as a personal experience without relating the experience of their milieu to structural features of the economy, they are engaging in non-political talk. When the speaker makes the connection between their personal experiences with poverty and the machinations of history or structure of society by drawing their interlocutors' attention to the unemployment rate, or to the historical exclusion of members of their social group from wellpaid jobs and education, they are participating in political communication or deliberation.
What people need to make the leap from non-political communication (oriented to personal experiences) to political communication (oriented to issues of collective concern) is a quality of mind we might call the political imagination (see Mills 2000). 8 When feminists say 'the personal is political' they are asking people to use their political imaginations. 'The personal is political' asks people to look at women's experiences as public issues, to make the link between biography and history, to shift from non-political communication to political communication (see Mansbridge 1999). Ideally, communication shifts easily from non-political talk to political communication (or deliberation) by engaging the political imagination and shifting the framing from personal experiences to collective issues (and in the case of shifting to deliberation, offering rational justifications).
Ideally, all forms of communication -whether oriented to collective issues or not -help develop deontic commitments (webs of reciprocal obligation) (Brandom 1998). This is because talking should involve reciprocally raising and responding to utterances, which requires making and responding to requests or bids for connection (Driver and Gottman 2004;Gottman and Driver 2005). When I respond to a speaker's utterancewhen I affirm my interlocutors bid for connection -this creates an expectation that my interlocutor will respond to my utterance in return. The basic reciprocal obligations created from raising and responding to utterances form the basis of the precondition of a commitment to reciprocity that underwrites social cooperation. These basic deontic commitments help develop mutual respect.
In more ideal conditions, even interlocutors who perceive one another as representatives of social groups with competing collective interests will still recognize each other as being worthy of respect. In such cases, all communicative practices -whether about collective issues or not -will create constellations of commitments and entitlements that develop mutual respect. However, in some deeply divided societies, interlocutors who perceive one another as representatives of social groups with competing collective interests do not recognize each other as being worthy of respect. As I explained in the last section, when disagreement over collective issues overlaps with a lack of mutual respect, talking about collective issues will not be deliberative. When people do not grant others 'the moral status of authorship' they remove themselves 'from the possibility of deliberative influence' (Mansbridge et al. 2012: 11). When disagreement over collective issues overlaps with a lack of mutual respect, talking about collective issues can contribute to attitude polarization and worsen intergroup animosities (Dryzek 2005;Mendelberg and Oleske 2000;O'Flynn 2007;Sunstein 2002).
It is precisely because non-political talk is not oriented to collective issues that non-political communication is particularly well-suited to encouraging agents to engage as speakers rather than as group members with opposing collective interests. Because of this, nonpolitical communication across intergroup boundaries is especially effective for promoting reciprocal attitudes such as mutual respect and greater tolerance for outgroup members (Allport 1954;Christ et al. 2014;Schmid, Al Ramiah and Hewstone 2014), particularly when these conversations entail self-disclosure and perspective taking (Broockman and Kalla 2016;Davies et al. 2011). When Dryzek (2005 describes the benefits of 'rituals and indirect communication (as opposed to confrontation)' that precede deliberation, the author is describing the benefits of non-political communication. Non-political communication is a tool that connects people without priming group-based threats, thus increasing the likelihood that reciprocal obligations and mutual respect develop across social divides.
Recognizing the role that everyday, non-political communication plays in developing mutual respect across social divides -and the relationship between mutual respect, inclusion, and deliberation -helps explain some of the mixed findings in deliberative democracy research. Specifically, recognizing the importance of non-political communication for developing mutual respect across social divides speaks to the debate about whether intergroup deliberation promotes positive outcomes such as more tolerant attitudes, or whether intergroup deliberation deepens attitude polarization and intolerance. For instance, Mendelberg and Oleske (2000) present evidence that intergroup deliberation is ineffective or may worsen intergroup tensions, while Grönlund, Herne and Setälä (2015) show that deliberation can increase tolerance for outgroups. As in my first example, this apparent contradiction probably highlights the distinction that should be made between disagreement in political communication networks (that fails to be addressed through deliberation) and disagreement that is addressed through the kind of reciprocal reason-giving that occurs when the conditions of justification are sufficiently met.
As I explained in the first section, the absence of crosscutting ties and mutual respect means that empowered social group members are less likely to hear and take seriously the utterances of disempowered social group members. What Mendelberg and Oleske (2000) show is that White residents from a wealthier, homogeneously White neighborhood who used coded racial language to resist school integration were impervious to Black speakers' complaints that their arguments were racist. What Grönlund, Herne and Setälä (2015) show is thateven among White people who may not have developed feelings of moral obligation toward outgroup members (in their study, toward immigrants) -disagreement over policies impacting immigrants can be organized into highquality, deliberative institutions where the commitment to reciprocity is fostered by facilitators priming democratic norms.
Although democratic innovations can motivate mutual respect, democratic innovations are rare (Chambers 2009) and mutual respect must also be developed in the unstructured public. The solution to a lack of mutual respect that blocks reason-giving in the unstructured public is more non-political, everyday communication.
In the unstructured public, deontic commitments are fostered through cross-cutting everyday communication and social ties ('intergroup contact') between members of different social groups (see findings on intergroup contact, e.g., Allport 1954;Brown and Hewstone 2005;Davies et al. 2011;Pettigrew 1998). Supporting the intuition that everyday communication helps build the bonds of reciprocal interdependence and mutual respect required to support the strain of political disagreement, Mendelberg and Oleske (2000) also show that White residents from a separate, integrated neighbourhood heard and respected Black speakers' complaints that arguments against school busing rested on coded racial appeals.

Conclusion
The main contribution of this work is to outline a typology of communicative practices that clarifies how certain democratic goods -notably, empowered inclusion and mutual respect -can be both antecedents to and outcomes of successful communicative processes. As I explained in the first section, it makes sense to say that inclusion and mutual respect are required for deliberation to occur at all and that deliberation empowers inclusion and increases mutual respect. However, if both claims are true then when inclusion and mutual respect are absent, deliberation cannot empower inclusion or increase mutual respect. When inequalities entail exclusion or a lack of basic mutual respect prevents deliberation from happening in the first place, where do the antecedent conditions of empowered inclusion and mutual respect come from?
I have shown that this question can be addressed by distinguishing between different types of communication. I propose two general criteria for distinguishing between communicative practices: first, is communication oriented to collective issues or private experiences? Second, is communication characterized by reciprocal reasongiving? These criteria produce a typology of four sets of communicative practices: deliberation, non-deliberative political communication, and non-political talk (nonpolitical reason-giving and non-political communication, which I have considered together). I have discussed how, in political systems where asymmetrical power relations systematically block deliberation, non-deliberative political communication -such as shouting slogans and making statements -is essential for incentivizing political inclusion. In societies marked by deep divisions, non-political communication is essential for developing a degree of interpersonal mutual respect before speakers try to talk through their political disagreements.
My discussion has implications for democratic practitioners -for instance, local decision-makers, activists, and practitioners in the field of democratic innovation -interested in engaging members of the public in discursive practices to promote more informed public opinion, expand political inclusion, or promote reciprocal attitudes and moral inclusion. Democratic practitioners should be attentive to the problems of democracy that need to be solved in order for a given engagement process to be successful. For instance, polarized public opinions in segregated societies with little intergroup contact might be organized into carefully designed institutional forums to encourage participants to reach more fully informed opinions through deliberation. Even within these institutional forums, practitioners should begin engagement processes with exercises aimed at promoting low-conflict, interpersonal interaction (intergroup ' contact') before attempting deliberation about matters of collective concern. For instance, opening participatory practices with ice-breaking exercises that entail self-disclosure and perspective taking, and engaging participants in group tasks that are solved through cooperation.
Since deliberative democracy's 'systemic turn' it has become commonplace for scholars to attend to the distinct aims that different deliberative or communicative moments achieve in broader political systems. This is an important development that has greatly progressed the discipline. My purpose here has not been to introduce a new ' approach' to the study of deliberative democracy. Rather, by proposing a typology for distinguishing between deliberative and non-deliberative communicative practices I am hoping to help address 'the twin dangers of "concept-stretching" and "criteria weakening"' that arise in some applications of deliberative systems approaches that judge all political practices by the degree to which they are deliberative or contribute to deliberation' (Owen and Smith 2015). Ignoring non-deliberative discursive practices runs the risk of over-extending the concept of deliberation and carries the danger that ' almost every communicative act may qualify as "deliberative" (at least in function), leading to the problem of concept stretching' (Bächtiger et al. 2010: 48). Concept stretching reduces conceptual clarity and worsens the problem of scholars talking past one another (Neblo 2007). By clarifying conceptual confusion, the typology I have outlined also helps explain mixed findings in studies of deliberation and points to future pathways for empirical research. For instance, recognizing that addressing political disagreement through reason-giving is a central feature of deliberation may help clarify whether deliberation mobilizes or suppresses other political practices such as voting. The leading study that suggests deliberation suppresses voting does not tap into political reason-giving and instead uses a measure of fairly intense, sustained political disagreement as a proxy for deliberation (Mutz 2006). It seems likely that disagreement that goes unaddressed by rational discourse -perhaps because structural inequalities prevent the disempowered from speaking out or being heard -can demobilize political engagement. Future empirical studies might compare the effects of political disagreement under different antecedent conditions to see how variation in equality and inclusion shape how different people respond to political disagreement (with reason-giving or without) and how this impacts subsequent political behavior. It would be interesting to identify the features that motivate the disempowered to turn to disruptive political communication when reason-giving fails (and the features that demobilize any political engagement). Future theoretical work might offer a normative framework that more carefully details when non-deliberative, disruptive political communication -as opposed to deliberation or another form of communication -is normatively justifiable in a democracy.
Recognizing the distinct value of non-political talk for developing interpersonal ties and bonds of mutual respect also helps clarify the mixed findings on intergroup deliberation. The question of whether intergroup deliberation promotes positive outcomes (such as promoting more mutual respect and tolerance) or worsens intergroup conflict probably depends on the extent to which people engage in everyday, non-political talk because they engage one another in conversations about matters of collective concern. In societies marked by deep divisions, non-political communication is essential for building the kinds of deontic commitments that enable members of different social groups to navigate the strain of collective rule by supporting the illocutionary aspects of discursive conflict resolution.
The second goal of my work has been to highlight the essential role that non-political, private expression can play in democracies. Unlike existing typologies that distinguish between discursive processes (Conover and Searing 2005;Landwehr 2010), the typology that I have presented here is not singularly concerned with communication oriented to collective issues or public speech. A promising new frontier in the study of deliberative democracy is the link between everyday social interaction and non-political communication -the realm of private expression that is not oriented to collective issues -and the desire/capacity to talk about collective issues. As I explained, it would be beneficial if people could shift easily between private expression and communication oriented to collective issues by engaging the political imagination and making the link between their biography and our collective history. But are all social group members equally able to engage the political imagination? What features motivate or disincentivize different people from engaging their political imaginations and shift the frame of conversations from private experiences to collective issues? Exploring the role that non-political communication plays in democracies -and exploring the relationship between non-political talk and communication oriented toward collective issues (including deliberation) -are promising avenues for future research.

Notes
1 Landwehr (2010: 102) considers the 'requirements for decision-making to be successful' and discusses the role that four ideal-type modes of communication play in decision making. Mansbridge (1999: 215) considers informal talk in the formally private sphere that is explicitly political (topics that are drawn to the attention of the public, as something the public should discuss as a collectivity, with a view to possible change'). Conover and Searing (2005) offer an empirical analysis of the kind of everyday political communication that Mansbridge (1999) describes. Elster (1997: 26) is concerned with political debates defined as 'public in nature and instrumental in purpose'. 2 Depending on the analyst's goals, even finer distinctions might be appropriate. For instance, it might be productive for scholars studying communicative practices leading up to a collective decision to make more fine-grained distinctions between communicative practices oriented to collective decisions (e.g., see Landwehr 2010). My central contribution with this broad typology is to highlight the democratic importance of non-deliberative communication (which others have done as well) and expressly non-political communication (which has received less attention from political scientists). 3 The typology of four discursive practices I have proposed preserves Jurgen Habermas's original definition of deliberation (practical discourse) as rational discourses about matters of collective concern (Chambers 1996, Habermas 1984, 1990. 4 For a discussion of this distinction see Habermas (1984). There is a debate around the relevance of arguing/deliberation vs. negotiation/bargaining in politics (Austen-Smith 1992;Elster 1997Elster , 2017Risse 2000). My typology does not speak to this debate, although it could be used fruitfully with existing distinctions between speech that is oriented to understanding/strategic speech. A comparative analysis of, for instance, understanding-oriented/ strategic communication about collective issues versus understanding-oriented/strategic non-political communication might be a fruitful avenue for future research. 5 There are instances where inequalities might be acceptable, such as when interests between those leading and those following are identical (see Mansbridge 1977). 6 See Klofstad, Sokhey and McClurg (2013) for a discussion of measuring disagreement in conversation networks. 7 Of course, disruptive speech can involve reason-giving.
But if inequalities and exclusion are blocking the uptake of reasons then what the speakers are relying on for effect is not the forceless force of good reasons. The speakers are relying on the degree of disruption: the loudness of the speech, the discomfort of the performative acts, etc. 8 Mills (2000: 19, see endnote 2) famously refers to this quality of mind as the 'sociological imagination'. However, as Mills admits, the term 'political imagination' can be used just as well to refer to the same idea. I use political imagination because it makes more intuitive sense in a framework analyzing the democratic potential of discursive practices. Aside from using the term political imagination instead of sociological imagination, I use Mills's original distinction between personal experiences (what Mills calls personal 'troubles') and issues, as well as Mills's original description of the state of mind required for linking biography to history, more or less faithfully.