Symbolic-Cognitive Proceduralism : A Model of Deliberative Legitimacy

Burkhalter et al.’s (2002) self-reinforcing model of democratic deliberation is well established, but lacks an account of legitimacy, which is a key element of most democratic-deliberative theories. We extend Burkhalter et al.’s model by proposing a new model called “symbolic-cognitive proceduralism,” which explains how democratic-deliberative processes generate legitimacy, and how such legitimacy contributes to the social reproduction of deliberation. Our proposed model accounts for perceived and normative legitimacy, at interpersonal and macro-social levels of analysis, over short and long timespans, and accords with substantial empirical evidence.

Democratic-deliberative processes are vital means of communication in legislatures (Steiner et al., 2004, pp. 138-164), self-governing bodies (Mansbridge, 1983), and civil society (Dryzek, 2000, pp. 81-114).Reformers have identified more arenas that would benefit from deliberation (Gastil, 2000, pp. 137-196;Fishkin, 1995, pp. 134-203;Gutmann & Thompson, 2004, pp. 139-187) and have introduced Citizens' Juries and other novel deliberative processes (Smith, 2009), some of which have been incorporated in large-scale systems, such as elections (Warren & Pearse, 2008). 1   Democratic-deliberative processes are group decision-making procedures in which participants-observing norms of equality and respect-gather information about an issue, prioritize values, determine policy solutions, analyze solution consequences and trade-offs, and choose a solution via a democratic decision rule (Burkhalter, Gastil, & Kelshaw, 2002, pp. 399-411).Further, legitimation is a primary purpose of democratic deliberation (Cohen, 1989, p. 22;Gutmann & Thompson, 2004, p. 10).Burkhalter et al.'s (2002) self-reinforcing model of deliberation defines the attributes, and explains the reproduction, of democratic-deliberative processes.This model warrants attention because it has been cited frequently in scholarship on democratic deliberation and related topics for its conceptual definitions and its preliminary empirical model of deliberation. 2Yet this model does not explain how deliberation yields legitimacy, which is another key concept in deliberative theory.In this essay, we extend Burkhalter et al.'s (2002) model by advancing a new theoretical model called "symbolic-cognitive proceduralism," which helps to explain the legitimacy of democratic-deliberative processes and the role of legitimacy in the social reproduction of deliberation.
In this article, we evaluate theoretical accounts of the legitimation of democraticdeliberative procedures to build the foundations of our model.We then demonstrate how our model advances that of Burkhalter et al. (2002) and has utility in contexts such as legislative and jury deliberation.We conclude by considering the limitations of our approach and the avenues it opens for further research.

Democratic-Deliberative Processes and Legitimacy
We define democratic-deliberative processes as forms of group communication in which participants analyze an issue, establish criteria for multiple solutions, and evaluate solutions while observing norms of equality and respect and using a democratic decision rule (Burkhalter et al., 2002). 3Variations on this definition abound (e.g., Chambers, 2003;Dryzek, 2000, pp. 1-7, 162-175), but it captures the deliberative and democratic dimensions of our central concept.Burkhalter et al.'s (2002) model also accounts for the sustainability of deliberative practices by applying Giddens' (1984, pp. 27-28) concept of the homeostatic loop, through which engaging in particular conduct produces cognitions and behaviors "that, in turn, are conducive to the original behaviors" (Burkhalter et al., 2002, p. 413).In this model, deliberation is predicted4 to influence citizens' thoughts and conduct such that citizens become more likely to engage in future deliberation.Deliberative participation is expected to increase citizens' deliberative communication skills, political efficacy, community-identification, and political knowledge, all of which foster habits reflecting deliberative practices (pp. 413-415, 419).These changes are predicted to increase citizens' capacity and motivation to deliberate (pp. 416-418) and attitude regarding the appropriateness of deliberation (p.414).Supporting Burkhalter et al.'s (2002) model is evidence that deliberative participation is positively associated with individuals' faith in deliberation (Gastil et al., 2010, pp. 134-146;Knobloch & Gastil, 2013, p. 239, 2015, p. 190).
Democratic-deliberative theories have been criticized on many grounds, including that citizens lack both competence (Mendelberg, 2002, p. 173) and a desire to deliberate (Hibbing & Theiss-Morse, 2002, pp. 227-228).Others have alleged that deliberation marginalizes minority voices (Young, 2000, pp. 36-50).Burkhalter et al.'s (2002) model, in particular, can be criticized on the grounds that it does not account for legitimacy.To address this deficiency, we will propose a model that helps to explain legitimation in deliberative-democratic practices within the wider scope of Burkhalter et al.'s (2002) theory.
"Perceived legitimacy" herein means the acceptability, from citizens' viewpoints, of a practice.Perceived legitimacy is illustrated by Estlund's (2008, pp. 7-9) model, in which political legitimacy stems in part from the "epistemic value" of deliberative procedures, provided such value "is publicly recognizable" (p.8).

Accounts of Deliberative Legitimacy
We preface our presentation of accounts of deliberative legitimacy with some qualifications.First, the following are theoretical accounts. 5Second, some descriptions of accounts include citations to evidence. 6Third, some of the following accounts overlap.For example, the concepts of deliberative procedures' expressiveness (Gutmann & Thompson, 2004, pp. 21-23) and of such procedures' approximating ideals (Cohen, 1989, p. 18-20) resemble one another, yet are logically distinct.Many consequences of deliberative participation predicted by the ethical account, such as enhanced concern for the common good (Cohen, 1989, p. 26), resemble (but are logically distinct from) effects predicted by the cognitive account, such as increased awareness of the common good (Mansbridge et al., 2010, p. 68).Thus, the ethical and cognitive accounts overlap.Finally, we view the following accounts as describing facets of democratic-deliberative processes.Multiple facets may simultaneously be attributes of any particular practice of deliberation.With these qualifications made clear, we now present accounts of deliberative legitimacy.
One straightforward account of deliberative legitimacy holds that deliberation constitutes the essence of democracy and therefore has intrinsic value.The intrinsic account partially explains the normative legitimacy of democraticdeliberative procedures.Thus Cohen (1989, p. 22) holds that an "ideal deliberative procedure" encompasses democratic values of justice7 and equality (pp.18-21) and a communicative value of rationality (p.21).For Habermas (1996, pp. 301, 463-490), deliberative procedures embody the democratic value of popular sovereignty.On the intrinsic account, one criterion of deliberativedemocratic processes' normative legitimacy is the extent to which they approximate an "ideal procedure" of deliberation (Habermas, 1996, p. 304;Cohen, 1989, p. 21).
Other accounts of deliberative legitimacy credit deliberation's instrumental value.All of these instrumental accounts explain how democratic-deliberative processes contribute to perceived legitimacy.Some also explain aspects of the normative legitimacy of such processes by furnishing grounds for preferring democraticdeliberative procedures to non-deliberative political processes (e.g., Martí, 2006, p. 33).Some theorists view deliberative procedures, in part, as cultural objects expressing symbolic meanings (Sunstein, 1996(Sunstein, , p. 2021)).For Gutmann and Thompson (2004), the "expressive value" (p.21) of democratic-deliberative procedures consists of "a manifestation of mutual respect among citizens … By deliberating with one another, decision-makers manifest mutual respect toward their fellow citizens … By deliberating with their fellow citizens, decision makers can … express mutual respect among free and equal citizens" (pp.21-23).Since social practices can acquire perceived legitimacy-i.e., "propriety"-through consistency with values citizens consider normative (Stryker, 1994, p. 857;Suchman, 1995, p. 579;Walker, 2004, p. 246), deliberative procedures' expression of those values-such as mutual respect-can enhance those procedures' perceived legitimacy.
On this account, citizens are predicted to apprehend the values that deliberative procedures express at the macro-social scale, via mass-media coverage of deliberative processes (Goodin & Dryzek, 2006, pp. 228-238) or, for wellestablished deliberative institutions, via public-school civics instruction (Levine, 2000, pp. 228-232).This perception is also predicted to occur during deliberative events, as participants and in-person citizen-observers witness deliberative procedures' value-expression.
Research indicates that media often cover democratic-deliberative processes (Fishkin, 2009, pp. 146-150;Rinke et al., 2013), and public-school civics instruction is common (e.g., Levine & Kawashima-Ginsberg, 2015, p. 3).Thus, media and public schools give masses of citizens opportunities to perceive valuecommunication by democratic-deliberative procedures.Some deliberative participants have testified that they perceived deliberative processes to have communicated values. 8Collectively, this evidence suggests that the symbolic account is somewhat realistic.
The symbolic account also helps to explain deliberative procedures' normative legitimacy by identifying a ground for preferring deliberative to non-deliberative procedures.That ground is that deliberative procedures emphasize mutual respect (Gutmann & Thompson, 2004, pp. 21-23), a value not emphasized as much by non-deliberative political procedures.
Cognitive accounts of deliberative legitimacy emphasize deliberative participation's potential to influence citizens' beliefs and attitudes in ways unrelated to decision quality (Chambers, 1996, pp. 202, 205, 240-241;Pincock, 2012, pp. 136-140).We use "cognitive" loosely to include some concepts involving both mental phenomena and emotions.9On the cognitive account, the perceived legitimacy of deliberative-democratic procedures-i.e., "propriety" (Stryker, 1994, p. 857)-stems in part from salutary cognitive consequences of deliberative participation.Such participation is predicted to enhance participants' issue-knowledge, awareness of their own and others' interests and values, understanding of the public good,10 and political efficacy (Benhabib, 1996, pp. 71-72;Burkhalter et al., 2002, pp. 415-417;Mansbridge et al., 2010, pp. 66-68, 70).Further, all participants, not just subsets of them, are deemed capable of experiencing these cognitive effects. 11On this account, such effects are expected to be communicated to non-participants through "ambassadorial" interactions (Pearse, 2008, p. 76), in which deliberative participants, partly due to these cognitive effects, 12 are predicted to describe their deliberative experiences to other citizens who are likely to observe these salutary cognitive effects in participants.
Evidence has also been found of deliberative participants' ambassadorial function (Gastil et al., 2010, p. 127;Pearse, 2008, pp. 76, 78).This evidence of the ambassadorial function and of salutary cognitive effects of deliberative participation suggests that the cognitive account of deliberative legitimacy is somewhat realistic.
In addition, the salutary cognitive effects of deliberative participation explain in part the normative legitimacy of deliberative procedures, by furnishing reasons for preferring deliberative to non-deliberative democratic processes.Those reasons are that the predicted desirable cognitive effects of deliberative participation are greater in number and variety than those predicted for nondeliberative democratic processes.Also, all deliberative participants are deemed capable of experiencing those salutary effects.
Theorists have also advanced an ethical account of deliberative legitimacy, holding that deliberative participation tends to "develop salutary human characters" (Sunstein, 1993, p. 141).On this account, deliberative participation is predicted to enhance perceived legitimacy-i.e., "propriety" (Stryker, 1994, p. 857)-because of its desirable ethical effects, which are expected to manifest in participants whom the public perceives as representing deliberative bodies.Further, the ethical account explains in part the normative legitimacy of democratic-deliberative procedures on the ground that deliberative participation tends to improve the moral quality of participants' personalities, as by increasing participants' concern for the public good (Pincock, 2012, pp. 139-140).Some theorists offer a communal account of deliberative legitimacy (Cohen, 1996, pp. 110-113;Cohen & Rogers, 1995, pp. 42-46;Fung, 2004, pp. 1-26;Fung & Wright, 2003, pp. 3-33) according to which deliberative procedures gain perceived legitimacy through their inclusiveness and community-building effects.These consequences are predicted to lead citizens, who observe deliberative bodies as participants or as witnesses of such consequences in their communities, to approve-i.e., grant "propriety" to-such bodies because of their consistency with values citizens consider normative (Stryker, 1994, p. 857;Walker, 2004, p. 246) and to identify with and trust those bodies (Cohen & Rogers, 1995, p. 44;Suchman, 1995, p. 575).In this view, deliberation is expected to strengthen social institutions (Cohen, 1996, p. 110-113;Cohen & Rogers, 1995, pp. 42-46;Fung, 2004, pp. 27, 70;Fung & Wright, 2003, pp. 15-17, 25) and collective identities (Felicetti et al., 2012).The communal account helps to explain democraticdeliberative processes' normative legitimacy by characterizing those processes' community-building effects as grounds for preferring those processes to nondeliberative political procedures.
Proponents of fairness accounts of deliberative legitimacy (Tyler, 2006, pp. 379, 382-384) argue that the perceived legitimacy of democratic-deliberative procedures stems in part from those procedures' justness as viewed by citizens.On this account, such perceived legitimacy involves "propriety" (Stryker, 1994, p. 857), meaning citizens' approval of such procedures for consistency with the value of justice which citizens esteem, and citizens' enhanced trust (Suchman, 1995, p. 575) in such procedures resulting from the procedures' just operation.

Building a Symbolic-Cognitive Model
From the accounts just described, we select two-the symbolic and the cognitive-and combine them in a single model, which we call "symboliccognitive proceduralism," after Estlund's (2008, pp. 98-116) "epistemic proceduralism." 17 In this section, we justify including particular elements in our model, then describe our model in general terms.We then explain our model in detail and show how it builds on Burkhalter et al.'s (2002) model of deliberation.Afterward, we illustrate the application of our model to jury and legislative deliberation.

Preferring Symbolic and Cognitive Accounts of Deliberative Legitimacy
Our model is intended to satisfy criteria for evaluating theoretical models, including scope, explanatory power, parsimony, and falsifiability (Shoemaker et al., 2004, pp. 171-176), 18 and criteria used to evaluate deliberative theories: realism (Galston, 2010) and theoretical modesty (Bohman, 2004, pp. 34, 39). 19arsimony (Shoemaker et al., 2004, p. 172) requires including the fewest components necessary to explain the focal phenomena (here, perceived and normative legitimacy) within the intended scope (here, interpersonal and macrosocial levels of analysis and short-term and long-term time-spans).Realism (Galston, 2010) and theoretical modesty (Bohman, 2004) restrict the model's elements to those that, respectively, are comparatively more consistent with social reality and make comparatively narrow claims.
Of these accounts, the symbolic and cognitive seem most compelling, for several reasons.First, each of the symbolic and the cognitive accounts has greater explanatory power (Shoemaker et al., 2004, p. 172-173) than either the intrinsic or the fairness account.This is so because both the symbolic and the cognitive accounts help to explain both perceived and normative legitimacy, whereas each of the intrinsic and fairness accounts explains only one type of legitimacy.
Second, the symbolic and cognitive accounts are somewhat more realistic (Galston, 2010) than the intrinsic, communal, and epistemic accounts.We prefer the symbolic and cognitive accounts, as instrumental explanations of deliberative legitimacy, to the intrinsic because few citizens are likely to value deliberation for its own sake.In our view most citizens are likely to value deliberation because of its consequences.Further, accounts focusing on deliberative consequences offer plausible explanations for associations between deliberative procedures and perceived and normative legitimacy.For many citizens, some effects of deliberative procedures are likely to contribute to deliberation's perceived legitimacy.For some normative theorists, some of these effects are likely to contribute to deliberation's normative legitimacy. 20oreover, we prefer the symbolic and cognitive accounts to the communal because the latter seems to assume broad public support for communitarian values (Etzioni, 1993, pp. 247-267), yet research indicates such values have weak or mixed support in most democratic societies (Inglehart et al., 2004, pp. 219, 231).
We prefer the symbolic and cognitive accounts to the epistemic because most versions of the latter-those requiring procedure-independent standards of deliberative-decision quality-seem less realistic than the symbolic and cognitive accounts, whose realism we discuss above.The reason is that in diverse modern societies, cultural heterogeneity fosters frequent value conflicts, especially over policy (Benhabib, 1996, p. 73;Dryzek & Niemeyer, 2006;Gutmann & Thompson, 1996, pp. 1-51;2004, pp. 11, 20-29, 73-89).Barber (1984, pp. 129-131) argued that under such conditions of value pluralism there is no "independent ground"-a phrase we interpret to mean a procedure-independent standard-for evaluating political decisions or for reaching political consensus.In our view, Barber's (1984) critique entails that citizens are unlikely to agree on a procedure-independent standard for evaluating deliberative decisions, not in every instance, but in many instances.Such irreconcilable conflicts over procedureindependent standards may concern general values-for example, between economic equality versus economic efficiency as a standard-or specific valuecontent-such as, within the general value of economic equality, particular values of income equality versus equality of opportunity-that must be included as specific criteria in such standards in order for those standards to differentiate effectively between deliberative decisions.Accordingly, we believe citizens are unlikely to agree on procedure-independent standards for evaluating deliberative-decision quality, not in all instances, but in many instances.Since most versions of the epistemic account require procedureindependent standards for evaluating deliberative decisions, most versions of the epistemic account seem less realistic than the symbolic and cognitive accounts. 21  We therefore conclude that the symbolic and cognitive accounts have somewhat greater realism than do the intrinsic, communal, and epistemic accounts of deliberative legitimacy.
Our third set of reasons, regarding our preference concerning the ethical account, involves the criteria of scope, explanatory power, parsimony (Shoemaker et al., 2004, p. 172-173), and theoretical modesty (Bohman, 2004, pp. 34, 39).The symbolic account addresses perceived legitimation primarily at the macro-social level, whereas the cognitive and ethical accounts do so at the interpersonal level. 21Our evaluation of the epistemic account does not reach versions of that account that do not posit a procedure-independent ground for evaluating deliberative decisions (Bohman, 2006, pp. 185-188;Gutmann & Thompson, 2004, p. 42;Peter, 2009, p. 128).We address this in the conclusion.
Therefore, the scope and explanatory-power criteria (Shoemaker, 2004, p. 172-173) suggest that our model should include the symbolic and either the cognitive or the ethical account, in order to address both the interpersonal and macro-social levels of analysis.Further, most effects predicted by the cognitive and ethical accounts overlap, and both accounts address both normative and perceived legitimacy.Accordingly, the criteria of scope, explanatory power, and parsimony suggest our model should include either the cognitive or the ethical account, but not both.To choose, we employ the criterion of theoretical modesty (Bohman, 2004, pp. 34, 39), which favors comparatively narrow claims.The ethical account makes broader claims-concerning the "transformation" of participants' characters (Pincock, 2012, pp. 138-142)-than does the cognitive account, which predicts only changes to participants' beliefs, attitudes, intentions, and political efficacy.Thus parsimony, explanatory power, scope, and theoretical modesty lead us to prefer the cognitive over the ethical account of deliberative legitimacy.

Symbolic-Cognitive Proceduralism
Having justified our model's elements, we now introduce it.We call our model of deliberative legitimacy "symbolic-cognitive proceduralism," because the model derives from Estlund's (2008, p. 98) "epistemic proceduralism."We base our model on Estlund's (2008) for three reasons.Estlund's (2008, pp. 98-116) model is parsimonious, as it relies only on the epistemic account; it explains perceived and normative legitimacy; and it applies to deliberation by juries and legislatures (pp. 156-158, 201-202).We wish our model to be parsimonious-and reasoned that employing a parsimonious foundation would increase the likelihood that our new model would be parsimonious-to explain perceived and normative legitimacy, and to apply to deliberation by juries and legislatures as well as to newer deliberative designs that feature elements of jury and legislative deliberation (Carson et al., 2013;Crosby & Nethercut, 2005, pp. 111-118;Gastil et al., 2014;Warren & Pearse, 2008).
According to our symbolic-cognitive model, 22 several processes contribute to the perceived legitimacy, i.e., "propriety" (Stryker, 1994, p. 857), of democraticdeliberative procedures.Through the symbolic function, such procedures are predicted to express, primarily at the macro-social level through media and public 22 In our model, the symbolic and cognitive functions are considered facets of democraticdeliberative processes.These functions are predicted to operate simultaneously, in different ways, and at different levels of analysis.That the symbolic and cognitive functions inhere in a particular democratic-deliberative practice does not preclude other functions, such as the ethical or communal, from also being attributes of that practice.education, widely accepted democratic values.Via the cognitive function, deliberative procedures are expected to produce, at the interpersonal level, desirable cognitive effects in deliberative participants.After deliberative events, such participants are predicted, partly because of these cognitive effects, to engage in "ambassadorial" interactions (Pearse, 2008, pp. 76) in which the participants recount their deliberative experience to other citizens.Those other citizens-who during these interactions are expected to observe in participants those desirable cognitive effects and therefore to perceive favorably those participants as representatives of deliberative processes-are predicted to approve of those processes.Over time, democratic-deliberative procedures having acquired perceived legitimacy in the form of "propriety" are predicted to become established as social facts (Stryker, 1994, p. 857) to which the public is expected to orient itself, resulting in increased perceived legitimacy in the form of "validity" (Walker, 2004, pp. 246-248).This accumulated perceived legitimacy is predicted over time to foster citizens' expectations (Beetham, 1991, p. 11) that state institutions will use deliberation to make policy and resolve disputes.These expectations are predicted to encourage citizens to demand additional opportunities for deliberation.These demands are expected to fuel the spread of democratic-deliberative procedures to more institutions in a "contagion" of (perceived) legitimation (Zucker, 1987, p. 446).
Summarized in broad terms, our model explains the normative legitimacy of deliberative-democratic procedures by furnishing grounds for preferring such procedures to non-deliberative processes.First, deliberative participation tends to yield salutary cognitive effects of a greater number and variety than participation in non-deliberative democratic processes.Second, all deliberative participants are considered capable of experiencing these cognitive effects.Third, democraticdeliberative procedures tend to express mutual respect, a value not emphasized as much by non-deliberative democratic procedures.

Extending Burkhalter et al.'s Model of Deliberation
This model explicates the role of legitimation in Burkhalter et al.'s (2002) selfreinforcing model of deliberation.Burkhalter et al. (2002, pp. 413-418) describe feedback loops whereby the cognitive consequences of deliberation lead to corresponding precursors.For our purposes, the most important of these loops involve deliberation's cognitive effects that feed back into one's future inclination toward deliberation.Deliberation's fruits motivate one to participate in the future, more optimistic that deliberation will function effectively, and more likely to deem deliberation an appropriate method.The model shown in Figure 1 reveals a series of predicted associations.Boxes represent core concepts: the procedural integrity of deliberative-democratic procedures, citizens' cognitive attributes influenced by deliberative participation, and the procedures' perceived legitimacy.The model begins with participants' experience of the procedural integrity of democratic deliberation.This integrity consists of elements such as selection processes yielding representative participants (e.g., legislative elections and relatively random selection of jurors), formal rules governing information intake (e.g., evidence rules for juries and legislative-hearings procedures), and reasoned, evidence-based deliberation among equals (e.g., jurors' deliberation about trial evidence and legislators' floor debate).Regarding the cognitive function, at the interpersonal level, deliberative participation is predicted to alter participants' cognitions: to enhance participants' understanding of issues and of others' interests and values, their deliberative skills, and their political efficacy.Regarding the symbolic function, primarily at the macro-social level, deliberative procedures having attributes of integrity are predicted to express widely shared democratic values to the public, through mass media and public education.Then citizens exposed to such value-expression are expected to accord perceived legitimacy in the form of "propriety" to deliberative procedures.
Through the ambassadorial function, the salutary cognitive effects of deliberative participation are expected to contribute to deliberative processes' perceived legitimacy.Participants, partly due to those cognitive effects, are predicted to serve as ambassadors of deliberation (Pearse, 2008, pp. 76) and thereby communicate their deliberative experiences to other citizens.In these interactions, these other citizens are predicted to perceive in the participants the salutary cognitive effects of deliberative participation and to view these participants as representatives of deliberative processes.These perceptions are predicted to cause these other citizens to judge favorably-i.e., to grant "propriety" (Stryker, 1994, p. 857) to-deliberative processes represented by the participants.
Reflective self-regulation and public demand for deliberation concern the reproduction of deliberative processes.In reflective self-regulation, deliberative experience is predicted to lead participants to increase their capacity to regulate their own behavior, such that they become models of deliberation themselves.The cumulative effect of reflective self-regulation is predicted to be the establishment of social norms of deliberative conduct, which are expected to render citizens generally more willing and able to engage in deliberation in the future.This prediction accords with empirical findings that deliberation can engender deliberative norms (Gastil, 2004, pp. 310-312, 323-325;Knobloch & Gastil, 2013, pp. 236-237).As opportunities for deliberation expand, more citizens can learn deliberative practices and thereby help to secure the procedural integrity of future deliberations.The repetition of this virtuous cycle reinforces the initial justification for the use of deliberative-democratic procedures.
There is a three-stage process by which a deliberative process's increasing perceived legitimacy is predicted to fuel public demand for deliberation, which is expected to strengthen deliberation's procedural integrity.First, soon after a deliberative event, a deliberative procedure is predicted to express democratic values through mass media to the broad public, eliciting perceived legitimacy in the form of "propriety" (Stryker, 1994, p. 857).Perceived legitimacy is predicted to be enhanced by other citizens' approval of deliberative procedures after observing, in "ambassadorial" (Pearse, 2008, p. 76) interactions, salutary cognitive consequences of deliberative participation in former participants.
Second, deliberative procedures, having gained some perceived legitimacy, are predicted over time to become established as social facts, leading to perceived "validity" legitimation as the public orients itself towards those procedures (Walker, 2004, pp. 246-248).Third, accumulated perceived legitimacy is predicted to foster popular expectations (Beetham, 1991) that political institutions will employ deliberation, leading citizens to demand expanded deliberative opportunities.Such demand is predicted to encourage deliberation to spread to more institutions, in a "contagion" of (perceived) legitimation (Zucker, 1987, p. 446).
Thus our model illustrates how the cognitive and symbolic functions of deliberative procedures contribute to those procedures' perceived and normative legitimacy while encouraging interpersonal-and macro-social-level processes of reflective self-regulation and enhanced public demand for deliberation that foster the reproduction of deliberative institutions.We now explain the application of our model to jury and legislative deliberation.

Application to Deliberation in Juries and Legislatures
The jury system, which represents the longest-running deliberative body employing lay citizens (see Vidmar 2000, pp. 1-27), illustrates our model.Regarding the symbolic function, the jury's deliberative procedures are predicted to express values widely esteemed in democratic societies: popular sovereignty, 24 political equality, and mutual respect-since jurors engage in self-rule as equals 25 -and rationality, by requiring verdicts supported by reasons and trial 24 For example, the jury has long been considered an emblem of popular sovereignty (Dzur, 2012, p. 122;Ostwald, 1986, p. 5;Vidmar, 2000, p. 8), and several jurisdictions employ near-random selection to choose jurors (Duff, 2000, pp. 259-260;Jackson, Quinn, & O'Malley, 2000, pp. 291;Vidmar, 2000, pp. 8-9, 34).These procedures are likely to convey the ideas that (a) the jury is a representative microcosm of the public and consequently (b) the jury validly represents the selfgoverning public. 25For example, in some jurisdictions, each juror's vote has equal weight (Kassin & Wrightsman, 1988, p. 172).Thus the jury's deliberative procedures are likely to convey the idea that jurors, as evidence. 26Jury procedures' value-communication is predicted to be conveyed to the mass public by media trial coverage (Hans & Dee, 1991) and public-school civics instruction (Levine, 2000, pp. 228-232).Value-expression by the jury's deliberative procedures is expected to enhance those procedures' perceived legitimacy, i.e., "propriety" (Stryker, 1994, p. 857), as citizens are predicted to approve of those procedures for expressing values according with broadly shared norms (Suchman, 1995).As the jury process endures and becomes established as a social fact (Stryker, 1994), perceived legitimacy-i.e., "validity" (Walker, 2004, pp. 246-248)-is likely to increase as citizens orient themselves to that process.
Moreover, because jury procedures possess deliberative attributes-including gathering high-quality issue-relevant information, norms of equality and respect among jurors, a clear decision-making procedure often requiring a unanimous vote, and a final decision having policy influence (Gastil, 2008, pp. 151-172)participation in jury deliberation is predicted to yield many salutary cognitive effects described in the cognitive account of deliberative legitimacy.These include gains in jurors' knowledge of issues and solution options, changes in jurors' attitudes about trial-related issues (Consolini, 1992, pp. 7-8, 92-98), 27jurors' increased understanding of their own and other jurors' interests, values, and of jurors' collective interests (Gutmann & Thompson, 2004, pp. 153-154), 28perspective-taking and meta-consensus (Dryzek & Niemeyer, 2006, pp. 638-646), and enhanced political efficacy and willingness to be civically engaged (Consolini, 1992, pp. 8, 92-98;Gastil et al., 2010, pp. 23, 26;Hans et al., 2014, p. 699). 29urther, former jurors are predicted, partly due to the salutary cognitive effects of deliberative participation, to function as ambassadors by recounting their representatives of the public, due to their equal status represent the values of political equality and mutual respect arising from that equality (Pettit, 2012, p. 262). 26For example, many jury systems employ evidence rules controlling the flow of information to the jury and procedural rules restricting jurors' considerations to trial evidence (Lloyd-Bostock & Thomas, 2000, pp. 80-83;Vidmar, 2000, p. 14) and require that jurors support their verdicts with reasons, through special verdicts (Vidmar, 2000, p. 42) or general verdicts which must be informed by jurors' deliberations about evidence (Kassin & Wrightsman, 1988, pp. 172-173).Thus the jury's deliberative procedures are likely to convey the idea that rationality grounds democratic decision making as represented by the jury.
deliberative experiences to other citizens (Gastil et al., 2010, p. 127).In these interactions the latter are likely to observe in former jurors desirable cognitive effects of deliberative participation and to consider those jurors representatives of jury deliberation.These other citizens are likely to approve of-i.e., grant "propriety" (Stryker, 1994, p. 857) to-the jury's deliberative process due to its representatives' exhibiting desirable cognitive effects of deliberative participation (Suchman, 1995, pp. 578-579).
Over time the accumulated perceived legitimacy of the jury's deliberative procedures is expected to prompt citizens' expectations (Beetham, 1991) that deliberation will be used in policy making and dispute resolution.Those expectations are expected to lead citizens to demand more chances for deliberative participation.Such demands are predicted to encourage wider dissemination of democratic-deliberative procedures among state institutions, whose perceived legitimacy is thereby likely to grow (Zucker, 1987).
Reflective self-regulation is also expected to foster deliberative institutions' reproduction.Jurors' deliberative experience is predicted to increase jurors' capacity to regulate their own conduct to conform to deliberative norms and thereby enable jurors to become models of deliberation.In aggregate and over time such reflective self-regulation is predicted to strengthen social norms of deliberative conduct, which are expected to foster citizens' willingness and capacity to participate in deliberative processes.As deliberative opportunities increase, more citizens are predicted to be equipped for deliberative participation and their competent participation is likely to bolster the procedural integrity of deliberative processes-including the jury process.
Moreover, the symbolic and cognitive functions of the jury's deliberative procedures furnish grounds for those procedures' normative legitimacy.First, deliberative procedures tend to express mutual respect, a value not emphasized as much by non-deliberative democratic processes.Second, participation in jury deliberations is predicted to yield, in jurors, salutary cognitive effects greater in number and variety than those predicted to be produced by participation in nondeliberative political processes.Third, all participants in jury deliberation-and thus all citizens, where jurors are chosen through near-random selection-are deemed capable of experiencing these desirable cognitive effects.
Thus our model helps to explain associations between the jury's deliberative procedures, those procedures' perceived and normative legitimacy, and the social reproduction of deliberative institutions at interpersonal and macro-social levels over short and long durations.
Our model applies to legislative deliberation as well as it does to jury deliberation, with the following exceptions.Regarding the symbolic function, the legislature's deliberative procedures are predicted to express values similar to those expressed by jury procedures-popular sovereignty, political equality, mutual respect, and rationality-but the particular features of the procedures differ from the jury's deliberative procedures. 30Regarding the cognitive function, the particular deliberative attributes of legislative procedures that are predicted to enhance the salutary cognitive effects of deliberative participation differ somewhat from those of jury procedures.These differences concern methods of gathering issueinformation-namely, legislative hearings and legislative research office reports (Bessette, 1994, pp. 50-51;National Conference of State Legislatures, 2006)and norms for reasoned argumentation during legislative debate (Bessette, 1994, pp. 51-52;Pedrini, 2014, pp. 265-266;Steiner et al., 2004, p. 90).Moreover, legislators' ambassadorial function differs from jurors' since the former occurs mainly in interactions with constituents (Gutmann & Thompson, 2004, pp. 30-31).

These jury and legislative examples suggest that our model is somewhat realistic:
The associations predicted by the model seem consistent with social conditions in contemporary, culturally heterogeneous democratic polities, and some of these associations accord with empirical evidence.This empirical support in particular suggests that the model's explanations for the perceived and normative legitimacy of democratic-deliberative procedures are rooted in those procedures' operation among people who employ them in their roles as citizens.
The foregoing explanations suggest that our model satisfies the specified evaluative criteria (Bohman, 2004;Galston, 2010;Shoemaker et al., 2004).The 30 Regarding popular sovereignty and equality, in many democracies laws give each citizen an equal number of votes, and establish a legislative chamber whose composition reflects proportional representation or each of whose members represents a roughly equal number of citizens (Dahl, 1986, pp. 195-196).Regarding mutual respect, legislative procedures in many jurisdictions express respect directly to citizens, or indirectly through legislative-decorum regulations requiring legislators, as citizens' representatives, to treat each other respectfully (Gastil, 2008, pp. 125-132).The former include procedures giving citizens a right to petition the legislature (Beetham, 2006, pp. 74-77), requiring legislators to treat citizen-witnesses at hearings with respect, and requiring respectful treatment of citizen-visitors to legislative offices and townhall meetings (Gastil, 2008, pp. 185, 193).These procedures are likely to communicate the ideas that (a) the public through its representatives controls the state and so governs itself, (b) each citizen shares equally in that control, and (c) each citizen deserves respect due to his or her status as an equal citizen.Legislative deliberative procedures are also expected to communicate the value of rationality, by providing for legislation to be based on evidence gathered during public hearings and arguments offered during debates (Gastil, 2008, pp. 124-136).
model is parsimonious, involving only symbolic and cognitive accounts of deliberative legitimacy.The model has a reasonable scope, covering the perceived and normative legitimacy and reproduction of democratic-deliberative procedures.The model has the desired explanatory power as it helps to explain perceived and normative legitimacy of democratic-deliberative processes at the interpersonal and macro-social levels and over shorter and longer time-spans.The model yields predictions regarding associations among deliberative processes, their symbolic and cognitive effects, those processes' perceived and normative legitimacy, participants' behavior, public demand for deliberation, and those processes' longevity.The model's propositions are falsifiable.Claims about normative legitimacy can be challenged by identifying non-deliberative theories of democracy that emphasize the value of mutual respect to a similar extent or that predict participation to yield a comparable number and variety of salutary cognitive effects in participants and that consider all participants capable of experiencing those effects.Further, all of the associations posited by the model can be refuted by empirical testing.The model is also realistic, in that its propositions seem consistent with the social reality of modern, culturally diverse democratic polities, and some of those propositions accord with empirical evidence.Finally, the model exhibits theoretical modesty: Its claims are relatively narrow and it excludes far-reaching claims about consequences of deliberative participation.

Conclusion
In sum, symbolic-cognitive proceduralism is a parsimonious and realistic theoretical model that helps to explain both the perceived and normative legitimacy of deliberative-democratic procedures.This model extends Burkhalter et al.'s (2002) self-reinforcing account of democratic deliberation by explaining how the symbolic and cognitive functions of deliberative procedures foster both kinds of legitimacy, which in turn contributes to the social reproduction of deliberative institutions.This model might supplant less realistic justifications for democratic deliberation.Deliberative democracy has considerable intuitive appeal for a public that applauds Jon Stewart's deliberative critique of conventional politics (Gastil, 2008, pp. 43-46).More explicit justifications matter, however, since fault-lines in the theoretical foundation of deliberative democracy can threaten its institutional viability.Such concerns are warranted, since deliberative democracy faces challenges as it becomes a widespread practice.Participatory processes declaring themselves "deliberative" without sufficient procedural integrity (Ganuza & Baiocchi, 2012, pp. 8-10) may undermine the credibility of the larger deliberative-democratic project.
Even if our model represents a significant advance, it has limitations.Our evaluation of the epistemic account of deliberative legitimacy does not address "purely procedural" versions of that account that do not rely on a procedureindependent standard for assessing deliberative-decision quality (Bohman, 2006, pp. 185-188;Gutmann & Thompson, 2004, p. 42;Peter, 2009, p. 128).Such "purely procedural" versions of the epistemic account may actually harbor procedure-independent criteria for evaluating deliberative decisions, as Estlund (1997, pp. 176-181;2008, pp. 85-97) claims.Further analysis of "purely procedural" epistemic theories could explore this possibility, and whether these theories are subject to Barber's (1984) critique or should instead be considered for possible incorporation into our model.In addition, our approach does not account for informal deliberative conversation (Kim & Kim, 2008, pp. 53-65).Zhang and Chang's (2014, pp. 137-140) research suggests that cognitive effects of participation in such conversation could contribute to perceived legitimacy that fosters the reproduction of informal deliberation.
Even so, our model paves the way for a program of empirical research exploring the role of perceived legitimacy and its related concepts in shaping deliberative processes.This research should be conducted at both small-group and macrosocial levels of analysis and cover both short-term and long-term time-spans to capture the full range of deliberation's impacts.Also, researchers should treat perceived legitimacy as mediating associations between deliberative procedures, on the one hand, and deliberative cognitions and conduct, demand for deliberative participation, and the sustainability of deliberative-democratic processes, on the other.Such investigations should clarify which of the paths in our model prove most potent as means of securing legitimacy.

Figure 1 Figure 1 .
Figure1summarizes our model in terms that permit an empirical scholar to trace three important sets of causal paths, one through the individual and the individual's self-reflection, a second through social symbolism and (perceived) legitimating social feedback, and a third from the individual participant to the social, perceived legitimating process.23