The Forum,the System,and the Polity: Three Varieties of Democratic Theory

Introduction

Deliberative democracy now stands at the core of democratic theory. As Goodin puts it, “We are in a period of high innovation, both within democratic theory and within democratic practice. The ‘deliberative turn’ has provided much of the impetus for both.” ¹ Innovation is accompanied by intramural disputes, as well as criticisms from other sorts of democratic thinkers—including a recent presidential address to the American Political Science Association. ² The intent of this paper is to ascertain where the theory and practice of deliberative democracy should now go in light of these disputes.

The argument will proceed in light of three images of deliberative democracy that seek to locate its essence in respectively a single forum (be it a theoretical ideal, a legislature, or a citizens’ forum), a deliberative system that specifies some division of responsibility or attributes across multiple sites, and an encompassing polity defined by integrative norms that can be more or less deliberative in their content and formation. The first two images are ubiquitous, though much about them (including their relationship) remains contested, the third is glimpsed in some recent contributions, but requires explicit statement, which it receives here for the first time. I will argue that the three images should be held in productive tension—but then we need to think about how practices that make sense in terms of each of them relate to the other two. The forum cannot do without the system if the very different micro benefits yielded by different sorts of forums are to be combined in meaningful fashion to have broad political effect. The system cannot do without the polity which provides a necessary democratic vantage point for the evaluation of deliberative systems. So while in this sense the polity image should take priority, this image proves not to give us the points of leverage for improving public deliberation found in the other two images. Advocacy, criticism, and internal dispute alike can be informed by empirical evidence, as well as by conceptual clarification and normative theorizing. I will attend to the evidence to the degree it helps clarify what can be expected of the forum, the system, and above all the polity.

In this light, the various “turns” that deliberative theory has taken over the years—institutional, empirical, and systemic—should not be seen in terms of a sequence of well-bounded generations of theories. ³ Rather, they are essential components of a productive dialogue about how democracy can and should be pursued in theory and practice.

Forums in Theory and Practice

Forum imagery was prominent in early explicit theoretical statements of deliberative democracy. Elster’s sympathetic critique of rational deliberation invokes an undifferentiated citizenry seeking unanimity. ⁴ Conceptions of democratic legitimacy that specify an ideal of “the free and unconstrained deliberation of all about matters of common concern” ⁵ imply a reference point (if only as an ideal) of an encompassing forum with a common-interest agenda. Cohen’s belief that “outcomes are democratically legitimate if and only if they could be the object of free and reasoned agreement among equals” again calls to mind a body of citizens seeking agreement. ⁶ Habermas’s “ideal speech situation”—often identified as a precursor of deliberative democracy—proceeds in the image of a single forum.

As soon as we try to apply this image of an ideal forum to real-world politics we encounter the problem of scale: large numbers of people cannot deliberate face-to-face. In practice, then, potentially deliberative forums work with representation of some sort. Representation might be along conventional electoral lines, constituting legislatures. Or it might involve stratified random sampling, constituting citizen forums, now better known as minipublics. Or a forum might bring together partisan advocates for different sides, constituting mediation or what are sometimes called “consensus-building” exercises (though what constitutes consensus in such exercises can be quite elastic). Authoritative forums may also be constituted on the basis of expertise rather than representativeness—think of constitutional courts. Different thinkers locate the best home for deliberation in each of these categories of forum. ⁷

“Deliberative democracy” is sometimes characterized as a “tool” for the design of forums. ⁸ But deliberative democracy should not be seen as a toolkit. For each forum type proves to exhibit not only strengths but also potentially fatal weaknesses from a deliberative and democratic point of view, meaning that none is sufficient actually to constitute the essence of a defensible deliberative democracy. Let me take a look at the details.

What Are Forums Good For? Justification versus Reflection

Deliberative democracy is characterized by Chambers as a “talk-centric” (as opposed to “voting-centric”) view of democracy. ⁹ This talk can involve the giving of reasons for or against positions, though positions can also be justified or criticized by the telling of stories, humor, or various sorts of rhetoric. For talk to take effect, listening is also necessary. ¹⁰ To have any deliberative point, listening needs to be accompanied by reflection on the content of positions, commitments, and preferences. For Goodin reflection is something that is done internally, by individuals (ideally those in positions of power) as they contemplate the interests of others. ¹¹ But reflection can also be a social process, as a collectivity together works through the positions it holds, such that publicly defensible reflections (which are not necessarily the same as persisting private desires and prejudices) come to the fore and are embedded in the group’s collective position. Different sorts of forums provide different degrees of possibilities and incentives for reflection of this sort; some provide obstacles.

There are reasons to suppose that justification and reflection do not sit easily together. Individuals who are good at making a case may not be very good when it comes to the possibility of changing their mind in response to what they hear from others. ¹² But it is not just a matter of personality, it is also one of institutional setting, and that is what I wish to emphasize here. It is hard for particular forums to embody high degrees of justification and reflection simultaneously.

We can note to begin a severe reflection deficit in contemporary parliamentary systems accompanying the partisanship that legislators are expected to show by their parties ¹³ —and by many of the mostly wealthy individuals, corporations, and groups that fund either parties or individual candidates for office. Partisan legislators rarely change their minds in response to the arguments of those on the other side of an issue—and are unlikely to admit it and act upon it if they do. Where then might we look for reflection, if not in a legislature?

Here we can turn to some empirical findings concerning how the deliberative virtues of respect and reflection are manifested in different sorts of forums. Many deliberative democrats have been attracted to the idea of minipublics composed of lay citizens, selected to be more or less representative of the larger population. Thousands have now been conducted in many parts of the world; there are a number of different models, including citizens’ assemblies, consensus conferences, citizens’ juries, and deliberative polls. ¹⁴ Initial enthusiasm for these forums celebrated the possibilities they enable for deeper deliberation than is normally possible in the public at large. Fishkin believes the conclusions of one kind of minipublic—the deliberative poll—should “have a recommending force for policy makers and the broader public” because “their considered judgments are the ones the public would support under good conditions for considering the problem.” ¹⁵ This sort of statement would fuel the fear of Lafont that minipublics could substitute for broader public judgment, and lead to giving up on mass democracy. ¹⁶ The fears are warranted for a further reason: mini-publics embody some deliberative virtues, but not others.

Especially when compared to legislative debate, minipublics seem to do poorly when it comes to the virtue of justification. Pedrini compares the level of justification in an online citizens deliberation with that in a legislative debate on the same immigration issue in Switzerland and finds that the level of justification in the latter is much higher. ¹⁷ Bächtiger and Parkinson compare findings from the Europolis deliberative opinion poll (an exercise in cross-national lay citizen deliberation) with debates in national legislatures, and again find the average level of justification much higher in legislatures. ¹⁸ Bächtiger and Parkinson conjecture that this justification deficit would be reduced (and possibly even disappear) for the rare long-lived mini-publics in which citizens become experts on the topic they are deliberating—such as the British Columbia Citizens’ Assembly on electoral reform, which met over a period of several months. ¹⁹ There is no evidence yet available on this conjecture.

Pedrini and Bächtiger and Parkinson also find that the deliberative virtue of respect for the positions taken by others is significantly higher in citizen deliberation than in parliamentary debate. ²⁰ Respect here indicates that what Gutmann and Thompson call reciprocity is taking effect. ²¹ Reciprocity means an effort to communicate in terms that those who do not share one’s framework can accept; explicit expression of respect for the points made by those with whom one disagrees is an indicator that reciprocity has been achieved. In the context of a very different kind of minipublic (the Oregon Citizens’ Initiative Review, which convenes citizens’ juries to deliberate referendum questions), Knobloch et al note that in drafting a final report, citizen-participants who ended on different sides of an issue helped those on the other side clarify their points in the final short document that is produced for wider dissemination. ²²

Comparing a legislative debate and a citizen conference, both in Germany on the issue of embryonic stem cell research, Landwehr and Holzinger find that communication in the citizens’ forum had a much higher dialogical quality than that in the Bundestag, where speakers mostly gave prepared statements—but did not respond to each other. ²³

Respect, reciprocity, and dialogical responsiveness are instrumental to reflection, because all involve taking seriously a position that is different from one’s own. We do have considerable evidence from empirical studies of minipublics that reflection does occur and does have an impact on the positions that participants hold. Now, the mere fact of preference change signifies little: it might be for bad reasons (such as social pressures). And if preferences do not change, but are better informed and more soundly backed, that is surely a positive reflective outcome. Niemeyer has developed an index of “intersubjective consistency” which measures the degree to which people agreeing on inner reasons (values and beliefs) also agree on expressed preferences. ²⁴ Frequently there is a weak relationship before deliberation, a stronger relationship afterward. Niemeyer attributes this (using the terminology of Kahneman) to a move from “fast” to “slow” or more considered thinking, as individuals renounce “cognitive shortcuts” and process the dimensions of an issue more fully. ²⁵ Obtaining direct evidence of the degree to which similar sorts of effects obtain in legislative debate is very hard—minipublics are much easier to treat as experimental settings, with systematic interviews with participants done before, during and after deliberation in order to gather necessary information. But the very fact that legislative debate generally features justification without dialogue suggests such effects would be less likely therein.

Connecting Forums

Partisan legislatures and minipublics do, then, exhibit very different kinds of deliberative virtues and deficiencies, suggesting that those who believe the essence of deliberative democracy should be sought in either minipublics or legislatures are mistaken. ²⁶ Is it possible to link minipublics and legislatures so as to preserve the virtues while overcoming the deficiencies? Consider for example a jury trial. We do not expect the lawyers advocating the two sides to be good at reflecting upon their relative merits; that task is assigned to the jury. The advocates for the two sides are not of course allowed in the jury room. A jury trial features two physically distinct yet procedurally linked chambers: one of justification, one of reflection (and decision).

But what of democratic political systems, as opposed to the delivery of criminal justice? Here we can look more closely at the idea of a deliberative system. Any system is a set of differentiated yet linked components that can be interpreted in the light of some common purpose. For a deliberative system, the purpose in question might be the generation of democratic legitimacy. ²⁷ Mansbridge et al refer to epistemic (problem-solving), ethical (mutual respect), and democratic (inclusion) functions of deliberative systems. ²⁸ For Mansbridge et al., “the system should be judged as a whole in addition to the parts being judged independently” ²⁹ (I will argue below that Mansbridge et al fail to provide any effective way to assess the whole). The parts in question might involve “everyday talk” ³⁰ or abstractions such as “public space” and “empowered space’ ³¹ or a sequence of settings such as party caucus, legislative debate, election campaign, and post-election negotiation. ³² It is possible to think of linking different forums with different virtues in this light.

Leydet suggests that the listening and reflection deficit in legislatures does not necessarily matter in light of a model of deliberation that is distributed across different sites. ³³ While in an ideal distribution Leydet believes that mass publics would judge the advocacy of legislators, in practice that does not happen, as mostly inattentive publics are only likely to be reached by plebiscitary reason (via a media which has no incentive to correct matters). Leydet points out that unlike a jury mass publics do not have the power to decide on particular issues; so the advocacy of legislators (unlike lawyers) does not have to engage seriously with the arguments of the other side.

Leydet’s solution is to move the mass public to later in the deliberative sequence, and promote dissent within parties as a way to get legislators to take the arguments of the other side seriously and respond to them. She recounts the case of a 2013 debate in the UK House of Commons, where—unusually—the government could not rely on its own backbenchers to support a motion authorizing military intervention in Syria. However, on Leydet’s account this kind of situation only produces a higher quality of justification in the legislature, as leaders are forced to respond to the arguments of their opponents (rather than talking past them). But Leydet can only hope that the larger public is paying attention; and she provides no mechanism whereby enlightened public opinion might make itself felt in the deliberative system.

Leydet takes the institutional structure of politics—and parliamentary democracy in particular—as given. But once we start thinking about institutional reconfiguration, some richer possibilities emerge for correction of reflection deficits.

To begin, we might think about the role minipublics can play in deliberative systems, given their role is not to substitute for public judgment. As Hendriks argues, minipublics should be integrated into deliberative processes in the larger civil society. ³⁴ We might note, contra Lafont’s fear, that it is not the conclusions of minipublics that should be inserted into larger political dialogue, but rather the arguments. ³⁵ In practice, this does not always happen. Karpowitz and Raphael in a content analysis of the reports of minipublics find substantial variation on a continuum from “decisional” (emphasizing conclusions) to “dialogical” (emphasizing reasoning). ³⁶ Dialogical reports should undermine the bad symbolic arguments that often dominate the larger public sphere ³⁷ and promote arguments that are reflectively defensible (including, crucially, to the members of the minipublic who do not agree with the argument in question). In this sense, the place of the minipublic in the larger system should be seen as deliberation making rather than anything like decision making. ³⁸ Anything communicated out of the minipublic has survived reflection therein. As Niemeyer puts it, minipublics can be both propagators and regulators of arguments. ³⁹ Of course, members of any larger public may still choose to reject the points made by any minipublic. If they do so reflectively, that is no problem from a deliberative point of view.

Linking Chambers of Justification and Reflection

We can be fairly confident that minipublics manifest the deliberative virtue of reflection, even as they fall short on justification. Yet whatever their virtue in this sense, they would still be vulnerable to Pateman’s criticism that they are one-shot affairs that are “not part of the regular political cycle in the life of a community,” and as such leave democratically deficient political systems untouched. ⁴⁰ Pateman believes that the stress on minipublics hurts the pursuit of participatory democracy. From within the deliberative democracy camp, Chambers likewise believes that emphasizing minipublics means giving up on mass democracy. ⁴¹ One obvious response here would involve the institutionalization or even constitutionalization of minipublics so that they became a permanent part of the arrangements of government (later, I will show how a deliberative systems image can respond to Patemans’s worries about broad participation). Along these lines, Leib imagines the establishment of a fourth “popular” branch of government in the United States composed of citizens’ juries deliberating legislation. ⁴² While interesting as a speculative exercise, it is not feasible at the US federal level, where the Constitution is so hard to amend even in minor fashion, let alone in the creation of a whole new branch of government.

The more general need here is for any deliberative system to involve moments of justification and moments of reflection that are not necessarily in the same location and engaged equally by the same people. Indeed, it may be too much to expect a location dominated by justification to take reflection seriously. So perhaps what is needed in any system are a “Chamber of Justification” and a “Chamber of Reflection.” This resonates with the idea that deliberation may benefit from the sequencing of different sorts of moments—be it within a single forum or across different forums. ⁴³ This proposal is also consistent with the old idea that an upper house in a legislature should be a house of review. In this light, the problem with legislatures such as those of Australia and the United States (at the national level) is that the Senate as well as the House is dominated by partisan advocacy—so there are two Chambers of Justification. If the United States federal government has a Chamber of Reflection it is perhaps the Supreme Court—but this institution too has become increasingly divided on predictable partisan lines. Moreover, it is not a democratic institution, given that its members are appointed for life, so relying on something like a Supreme Court for reflection in the system means attenuating its overall democratic quality. The same might be said for the House of Lords in the United Kingdom. While Parkinson praises the Lords for its deliberative qualities (particularly when it comes to inclusiveness and the examination of arguments), ⁴⁴ it suffers from the same democratic problem as the Supreme Court: members appointed for life.

Why not then replace upper houses with chambers of randomly selected lay citizens? Their task would be to deliberate and reflect, not to legislate. In 1998 a proposal by the Demos think tank in the United Kingdom suggested something like this in the context of debates over House of Lords reform. The title of the pamphlet, “The Athenian Option,” reminds us that democratic assemblies were once constituted in this fashion. Currently unicameral legislatures such as those of New Zealand, Nebraska, and Queensland provide an opportunity to experiment. Of course thinking of an upper house in this light suggests a general-purpose chamber, somewhat different from the single-issue basis on which minipublics are normally constituted. But there is no reason why the entire chamber would need to deliberate every piece of legislation; creative institutional design could reduce the burden on individual citizen-members by having subsets of the whole handle particular pieces (or involve ways to constitute different forums for different issues).

A less thoroughgoing way to make minipublics part of the “regular political cycle” has already been institutionalized in the state of Oregon, in connection with citizen-initiated referenda, whose results are binding on the State government. In the Oregon Citizens’ Initiative Review process established in 2009, a citizens’ jury is constituted to deliberate a referendum measure—such as the legalization of medical marijuana or mandatory minimum sentencing for particular crimes. The jury then writes an evaluation that goes into the Voters’ Pamphlet that is sent to all voters in the state. This evaluation contains the jury’s view of key arguments both for and against the measure. ⁴⁵ Here we have a process clearly involving justification, reflection, and decision. Justification is made initially by proponents and opponents of the measure in public campaigns, reflection is done initially by the citizens’ jury, and decision is made by the voters in the referendum. Ideally, good arguments (on either side of the measure) survive, while bad arguments do not. This sequence—first justification, then reflection, then decision—is actually much more logical than seeing minipublics as providing inputs into legislative debates or executive decision. For the latter sequence implies moments of reflection precede moments of justification, which then lack the benefit of reflection prior to decision.

Beyond minipublics, ways to promote reflection might also be sought in bodies of unaffected strangers (in a particularly good position to exercise what Sen calls “open impartiality”), ⁴⁶ constitutional courts, non-partisan elected bodies, moderated online processes, or indeed the exceptional individuals Goodin believes should exercise “democratic deliberation within” their own minds, caring about the interests of others. ⁴⁷ Such institutions and practices could be deployed in such a way that they do not leave the regular cycle of politics untouched.

The Challenge of Inclusion

However much minipublics or their reflective alternatives became part of the regular cycle of politics, participatory democrats such as Pateman might remain unsatisfied because any moments of reflection introduced into the system would not entail the more thoroughgoing participatory development necessary to challenge the core institutions of a limited and unsatisfactory democracy. ⁴⁸

Lafont doubts that simply inserting the conclusions of minipublics into broader public communication will bring larger numbers of people into deliberative exchanges, given the “highly defective deliberative context of a public sphere full of sound bites, demagoguery, and manipulative misinformation.” ⁴⁹ Lafont gives no indication as to what else might help when it comes to this dismal condition of the public sphere.

How then to meet the challenge of deliberative inclusion and so respond to critics such as Pateman and Lafont? The best efforts of participatory democrats such as Pateman notwithstanding, recent decades seem to have yielded only a few participatory exemplars with some staying power (such as participatory budgeting) and many short-lived experiments. Moreover, as Fishkin points out, in practice the maximization of participation will often come at the expense of equality—because the more motivated and engaged are more likely to want to participate. ⁵⁰ But Pateman’s implicit point that democratization needs somehow to promote inclusion of the citizenry at large—rather than (more or less demographically representative) samples of them—is well taken.

Inclusion might seem something that minipublics cannot help with; but we should not summarily dismiss their inclusion-promoting role lightly. Now, for the public deliberation-making task mentioned earlier, the size of the minipublic does not really matter; there is no point in seeking (for example) the “millions of voices” (a slight exaggeration) that help legitimize relatively large exercises promoted under the 21st Century Town Meeting Model. ⁵¹ However, it does matter that all relevant arguments, stories, and subjective stances are present; or to put it another way, the forum should aspire to good discursive representation, where all relevant discourses are represented. ⁵² (An exception to this principle can be made when the discourse in question seeks mainly to undermine possibilities for effective societal reflection.) Discursive representation is itself a kind of inclusion—though of discourses, rather than persons. But it could also go some way toward promoting inclusion in the larger public sphere to the degree its presence in a smaller forum would increase the chances that this forum would generate a product that resonates with a larger public’s subjective dispositions, thus drawing sections of that public into the deliberative system. Still, the inclusion mechanism here is a bit tenuous and modest.

Inclusion in the Deliberative System

Inclusion is central to the very idea of a deliberative system: the subtitle of Parkinson and Mansbridge’s book on deliberative systems is “deliberative democracy at the large scale,” and the move to scale is accomplished not just by linked representative institutions, but also by (for example) recognizing roles for everyday talk, public conversations, and political activism. ⁵³ When Mansbridge et al talk about the “democratic” functions of deliberative systems, it is inclusion they have in mind. ⁵⁴

At one level the connection is obvious: the more sites recognized in a deliberative system (be they social movements, lobby groups, protests, social media, election campaigns, referendums, exercises such as participatory budgeting, minipublics, everyday conversation, or legislatures), the more opportunities there are for inclusions of different sorts.

However, a systemic approach to inclusion cannot be additive: that is, we cannot simply add up bits of inclusion achieved in different locations in a deliberative system and assume that they support each other and contribute to the inclusiveness of the system as a whole. As a corollary of the deliberative systems axiom that intrinsically non-deliberative acts can have deliberative consequences for the system as a whole (the standard example is strategic activism that gets hitherto suppressed questions on the political agenda), we should be alive to the possibility that particular inclusions could have consequences that turn out to be exclusive. Consider the inclusion of internet trolls who drive out others from political debate. Or should we welcome into a deliberative system those whose main concern is to strengthen hierarchies based on race, wealth, or nationality? While in a forum it is easy to see how the power of deliberation itself can curb and civilize exclusionary practices, the available mechanisms are much weaker in a deliberative system.

We should also be alive to the possibility that particular exclusions may have systemic consequences that prove to be inclusive. For example, Dryzek et al deploy comparative historical analysis to argue that the exclusion of environmentalists from influence on public policy in some cases contributed to the development of a lively and competent green public sphere that eventually yielded both more effective and more inclusive policy making. ⁵⁵ Karpowitz, Raphael, and Hammond show that restricting participation in a citizens’ forum to the disempowered can have positive effects on the participants’ competence and appreciation for diverse arguments. ⁵⁶ Dovi, though not working in a deliberative systems idiom, suggests that the exclusion of privileged oppressors (notably racist groups) can be democratically defensible. ⁵⁷ The dynamics of inclusion and exclusion are amenable to analysis within a deliberative systems framework, for example, as we see how exclusions in one place can promote inclusions in another. What is still missing is some metric to assess the performance of the system as a whole on the inclusion criterion. Mansbridge et al say “the system should be judged as a whole” ⁵⁸ but do not show how this can be done when it comes to inclusion—or indeed on any other deliberative or democratic criterion. This absence applies equally to the criteria of justification and reflection (the key aspects of deliberative authenticity). Or to put it another way, a deliberative systems approach cannot easily tell us the degree to which a system is actually deliberative. If some solution cannot be found to this problem, then the systemic turn in deliberative democracy is in trouble.

The System in the Polity

To drive home the difficulty a deliberative systems approach has in generating such assessment, consider the reductio ad absurdum advanced by Owen and Smith. ⁵⁹ Taking seriously the axiom that intrinsically non-deliberative practices can have positive systemic deliberative consequences, Owen and Smith raise the specter of a defensible deliberative system that does not actually contain any deliberation—only a series of nondeliberative practices, each of which has positive systemic consequences. (The corollary is surely an indefensible deliberative system composed of fully satisfactory deliberative parts.) One possible solution to their challenge would be to specify the necessity of good deliberation only in some particular privileged location within a system (for example, among “decision-makers” ⁶⁰ )—but that would be an effective renunciation of system-level evaluation.

Owen and Smith’s own solution to the problem they identify can be found in what they call a “deliberative stance.” While sharing Mansbridge’s commitment to the centrality of “everyday talk,” ⁶¹ the deliberative stance reveals itself when citizens are “engaged in the mutual exchange of reasons oriented as if to reaching a shared practical judgment” (emphasis in original). ⁶² Deliberative democrats should in this light cultivate the “challenging and fragile” deliberative stance in society as their primary task, and “consider the ways in which the system enables (or obstructs) the deliberative stance on the part of citizens.” ⁶³ Thus deliberative systems can be evaluated in terms of their impact on the prevalence of the deliberative stance among the citizenry—at least in principle. This does not mean deliberative democrats should turn their back on the system; as Owen and Smith also note, an analysis of how public deliberation (motivated by the deliberative stance) affects collective decision is still required. ⁶⁴

In practice, matters are not so clear. How do we know if citizens are adopting a deliberative stance, or if they are not? Is it better for many citizens to adopt it in a small way, or for fewer citizens to adopt it in a big way? More generally, given the deliberative stance seems to be an attribute of individuals, how do we move from individuals to (complex) collectivities? Should we just assume that more individuals adopting the stance at some threshold level is enough to render the collectivity deliberative? Is the deliberative stance the same for all people in all places, or is cultural variation in its content allowed? Is it vulnerable to criticism on the grounds raised long ago by Sanders that deliberation (narrowly defined as reason giving through argument) embodies a particular communication style that privileges some people and marginalizes others? ⁶⁵ Owen and Smith give no examples that would help clarify such questions and make the deliberative stance concept operationalizable in the assessment of deliberative systems.

We do however need some reference point or level of judgment that is comprehensive or in Owen and Smith’s words “expansive” in order to be able to assess deliberative systems in their entirety. There are two other clear candidates.

The first is the venerable idea of the public sphere, a communicative realm where norms and cultures can be constructed and maintained. The problem here is that the idea of the public sphere has almost always been used by deliberative democrats in the sense of politicized civil society distinct and somewhat separate from formally constituted public authority (such as the state, or an international governmental organization). This treatment begins with Habermas’s classic history of the early bourgeois public sphere in Europe, where the emerging bourgeoisie were excluded from state power by monarchies and landed aristocracies, developing a political society of their own in response. ⁶⁶ Some more recent treatments of the deliberative system see the public sphere—or public space—as part of the system. ⁶⁷ In short, the public sphere as understood by deliberative democrats is not encompassing enough to be a reference point for the evaluation of deliberative systems, but would rather involve privileging one location within deliberative systems as a source of assessment. A more encompassing interpretation of the public sphere would, however, overcome this problem.

The second candidate is the idea of a comprehensive deliberative culture as developed by Böker, which she defines as “all of the informal, unsteered, norms, expectations, meanings, and customs that drive citizens’ attitudes and behavior in the political arena.” ⁶⁸ A specifically deliberative culture would manifest “the norms of deliberative democracy (such as fairness, equality and the right to justification).” ⁶⁹ Böker turns standard notions on their head by treating culture as a driver of change, rather than an inherited set of features. Linking deliberative culture to deliberative systems, she sees culture as the essence of deliberative democracy, as something that is causally prior to institutions within the system. In this sense, deliberative culture is the dynamic counterpart of the deliberative stance proposed by Owen and Smith. Culture should drive institutional development and for Böker can only “self-evolve” from the bottom up. ⁷⁰ The idea can be used to evaluate deliberative systems (or at least their components) if as she says “system components contribute positively to deliberative democracy if they enhance the openness, inclusiveness and autonomy of the social-political space in which democratic cultural processes evolve.” ⁷¹

Deliberative culture is in Böker’s eyes a normative concept informed by deliberative theory, which constrains the way a culture can be deliberative. What then are we to make of the observable variety of defensibly deliberative cultures? ⁷² The thickness of the norms of a deliberative culture in Böker’s terms may skew evaluation—and sit uneasily beside the cultural self-evolution she also prizes. In addition, there may be times (notably but not solely when it comes to transnational deliberative systems) when the idea of a single culture overstates the conceivable degree of normative integration. A deliberative system can in such cases co-exist with multiple deliberative cultures.

As the dynamic counterpart of the deliberative stance, the deliberative culture concept is also subject to some of the same questions about moving from individuals to collectivity if (as for Böker) culture is defined as a driver of individuals’ “attitudes and behavior.” However, this problem could be solved very easily by treating culture in the intersubjective terms of the ethnographer as opposed to the individualistic terms of the political scientist. Such a move would helpfully separate deliberative culture from what is in Johnson’s terms the “moribund” tradition of political culture research. ⁷³

The deliberative stance, the public sphere, and the idea of a deliberative culture clearly bear a family resemblance to each other. All take an encompassing approach in contrast to the differentiated orientation of deliberative systems analysts, without quite providing an adequate solution to the pressing need for a means to evaluate the democratic credentials of deliberative systems in their entirety. So is it possible to do better? We can approach an answer from the idea that any deliberative system can be conceptualized as embedded in a polity, though their boundaries need not coincide. Here, it makes sense to use the term polity rather than society, because the latter has dimensions that are not necessarily political. In addition, a polity may exist when a society does not; for example, in transnational governance on issues such as climate change, biodiversity, or finance. A society and a polity alike are defined by normative integration: they have norms that regulate the interactions of their members. Integrative norms should not be taken as given, but can themselves be the subject of contestation as they evolve with time. However, there must be a degree of normative integration—not just contestation all the way down—because without it, there is actually nothing to contest.

The integrative norms that define a polity are contextually specific, and can be substantive or procedural. They are substantive when (for example) there is a widely shared discourse—such as sustainable development in global environmental affairs, or neoliberalism in the economic affairs of Anglo-American countries since the 1980s. Aside from shared discourses, substantive integrative norms can take the form of a meta-consensus on the legitimacy of disputed values, the credibility of disputed beliefs, the range of acceptable options, or the nature of disputed choices. ⁷⁴ Integrative norms are procedural when they concern how actors should relate to each other, be it peacefully or violently, in cooperative or adversarial fashion, through bargaining or argument. The polity image therefore enables response to critics who believe the deliberative systems approach is too forgiving of non-deliberative practices—to the point of abandoning core deliberative commitments. ⁷⁵

In this light, deliberative systems can be evaluated in terms of their impact upon the deliberative well-being of normative integration in the polity. We ought to be looking for competent, critical, inclusive, and egalitarian communicative action in the definition of integrative norms.

Evaluating the System in the Polity: Examples

Let me give some illustrations to show what such evaluation means in practice.

Conclusion

Deliberative forums only make sense in a deliberative system. A deliberative system can only be evaluated in comprehensive democratic terms through reference to the idea of a deliberative polity. At the same time, forums are some of the best ways of ensuring that deliberative systems feature the justification and reflection necessary for deliberative authenticity; and systems are vital when it comes to thinking about how to apply deliberative ideals in inclusive fashion at large scale. Yet ultimately it is in terms of the polity that the democratic contributions of forums and systems need to be assessed.