David M. Ryfe

(1)

D ÔES D ELIBERATIVE D ÊMOCRACY W ÔRK ?

David M. Ryfe

School of Journalism, Middle Tennessee State University, Murfreesboro, Tennessee 37132; email: dryfe@mtsu.edu

Key Words deliberation, deliberative democratic practice, political psychology, civic engagement

■ Abstract The growing literature on deliberative democratic practice finds that deliberation is a difficult and relatively rare form of communication. Each moment of a deliberative encounter raises significant obstacles in the path to stimulating greater intentional reflection on public issues. I explore these obstacles in the context of other empirical work in political and social psychology, small group communication, and public opinion. Taken together, these literatures explain why deliberation is difficult to achieve and sustain over time. They also suggest several rules that might assist practitioners in making deliberative democracy work better. Many of the obstacles to deliberative democracy raise questions about key theoretical constructs closely associated with deliberative democratic theory, including equality, legitimacy, autonomy, and reason. I conclude by suggesting that deliberative practitioners, empirical scholars, and theorists might gain from greater interaction.

INTRODUCTION

For most of its career, deliberative democracy has been something of a small, rarefied subfield of political theory. This “phase” of deliberative democracy (Chambers 2003) has not passed (see also Freeman 2000), but in recent years it has been supplemented by a more pragmatic impulse. Empirically minded students of deliberative democracy have turned to issues of implementation, institutional design, and evaluation (Abelson et al. 2002, 2003; Ackerman & Fishkin 2004; Bierle 1999; Denver et al. 1995; Dunkerly & Glasner 1998; Fishkin 1991, 1995; Gastil 2000; Gastil & Dillard 1999; Graham & Phillips 1998; Hibbing & Theiss-Morse 2000; Kim et al. 1999; Kraft & Clary 1991; Landeman 2002; Luskin & Fishkin 1998; Mackie 2002; Mansbridge 1980; Mullen 2000; Neblo 2000; Renn et al.

1995; Ryfe 2002; Simrell 1998; Smith & Wales 2000; Sulkin & Simon 2001).

These efforts do not supplant the need for normative theory. Indeed, to some extent the empirical literature is driven by the passion and vision of deliberative theorists. Nonetheless, one cannot ignore the fact that the empirical findings have been mixed. Under certain conditions, it appears that deliberation can produce more sophisticated, tolerant, and participative citizens (Fung 2001, Fung & Wright 2001, Gastil & Dillard 1999, Gastil et al. 2002, Luskin & Fishkin 1998, Sulkin

1094-2939/05/0615-0049$20.00 49

Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(2)

& Simon 2001, Walsh 2003), but these outcomes are not automatic and in fact may be rare (Button & Mattson 1999, Hendriks 2002, Hibbing & Theiss-Morse 2000, Holt 1999, Kuklinski et al. 1993, Mendelberg & Oleske 2000). Combined with the fact that institutionalizing deliberation can be quite costly (Rossi 1997), this finding suggests a need for more reflection. Assuming for the moment that deliberation is a sensible and normatively preferable way of making decisions, why is it so difficult in practice? What does the empirical literature contribute to contemporary debates within deliberative democratic theory? What does the literature on deliberative practice tell us about the conditions most likely to promote deliberation? At bottom, of course, lies the fundamental question: As a practical matter, can deliberative democracy work?

The literature on deliberative practice is still in its infancy, and its answers to these questions are by no means definitive. However, even in its current state, it is suggestive. By linking its findings with other empirical and theoretical work, I tease out the broad outlines of the central issues. I divide my discussion into three moments of the deliberative process: the organization of a deliberative encounter; the practice of deliberation within an encounter, and finally, the product of deliberative talk. Each of these moments raises practical challenges for organizers of deliberative initiatives. Who should participate, and how should they be contacted? Once they meet face to face, how do and should participants talk to one another? That is, what does deliberative talk look like? And what should be done with the product of deliberative encounters once groups have met? Should public officials use the product as an expression of public opinion? Should they be bound by the conclusions reached by deliberative groups? The way in which these moments work themselves out tells us something about the possibilities for deliberative democracy in actual societies. It also illuminates key conceptual dilemmas in normative theory. As I weave together the empirical scholarship and normative theory, I assess where we are in answering the fundamental question of whether or not deliberative democracy can work.

WHO PARTICIPATES?

As Pratchett (1999) observes, “there is nothing particularly new about public participation as a supplement to representative democracy” (p. 616). Officials routinely solicit public comment, hold public hearings, and issue public reports on their activities. But the promise of deliberative democracy hinges on more than public consultation. Deliberative democrats believe that ordinary people ought not only to be consulted but also to have a hand in actual decision making. However that role is captured (a subject I take up later), the idea is that public decisions ought to be influenced in some way by the citizenry that will be affected by them. This assumption, of course, makes it vital that such citizens participate actively in the process of decision making.

With this goal in mind, it would seem to be a simple matter of opening the policy- making process to greater citizen input. In practice, however, it is not simple at Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(3)

all. Participation has long been a minor consideration in most liberal democracies (Barber 1984, Bobbio 1987, Macpherson 1977). Moreover, at least in the United States, the kinds of civic associations that once connected ordinary people to the political process have withered considerably (Putnam 2000, Skocpol & Fiorina 1999). This disconnect between citizens and public officials has led to great cyni- cism and distrust on both sides (Hibbing & Theiss-Morse 1995, Nye et al. 1997).

Formidable psychological and structural barriers also impede public participation in policy making. People are, as cognitive psychologists like to call them, “cognitive misers.” For everyday reasoning, people prefer to use cognitive heuristics (also called information shortcuts, discussed below) to make reasoning relatively efficient but unreflective (Kahneman & Tversky 1983, Kahneman et al. 1982, Lupia et al. 2001, Mondak 1994, Mutz et al. 1996, Nisbet & Ross 1980, Sniderman et al.

1991, Taber et al. 2001). Especially in the face of difficult, complex issues, people seek to “pass the buck” in an effort to avoid responsibility for decision making (Festinger 1964, Fiske & Taylor 1991, Janis & Mann 1977, Tetlock 2001). Further, as we have known for 50 years, choices about public goods tend to create “collective action” problems (Downs 1957). Their sheer extensiveness and complexity make them difficult issues amenable to no easy answers. Their extensiveness means that any rational individual will seek to forego the burden of participation because she is not likely to directly affect the result, yet will share in its benefits even if she refuses to participate. Organizing deliberative initiatives in the face of these historical, cognitive, and structural facts is a daunting enterprise indeed.

To overcome these impediments, organizers of deliberative initiatives have two basic options when inviting public participation: They may advertise their initiatives locally and allow individuals to self-select, or they may take more active and direct recruitment steps (Button & Mattson 1999, Leroux et al. 1998, Renn et al. 1995, Ryfe 2002). The first option typically produces a “snowball” sample, in which interested individuals recruit from their social networks, these individuals recruit from their social networks, and so on until a group is composed. The second option usually involves some kind of representative sampling procedure, in which organizers create a group with a demographic profile that reflects the community. In neither case is the resulting group very large. Most deliberative initiatives involve no more than 20–30 individuals talking directly to one another.

A few try to link small groups into a larger conversation, often using information technologies such as the Internet, but for the most part the practice of deliberation is constrained to relatively small groups (Goodin 2000, 2004).

Each option has its pluses and minuses. The major benefit of the snowball method is that it makes recruitment relatively easy and inexpensive. Because organizers may find participants within their own social networks, they need not expend a great deal of time or resources recruiting beyond their immediate field of vision.

However, this relative ease of access also marks the major drawback of this ap- proach: Self-selection often makes for homogeneous groups. Other work on civic participation reveals why this is the case. Research shows that civic participation is strongly correlated with belonging to social networks that privilege civic identities and make access to the political process relatively easy and frequent (Burns Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(4)

et al. 2001, Verba et al. 1995). Further, civic participation is closely associated with education levels, and this variable correlates with other indicators such as race and class (Conway 2000, Nie et al. 1996). The upshot is that if participants are allowed to self-select, those who participate are very likely to be white, college-educated, and middle-class.

There are reasons to be concerned about this tendency toward homogeneity.

Some are practical. Work on public talk and opinion shows that diversity can be a key indicator of a deliberative frame of mind (Huckfeldt & Sprague 1995; Knoke 1990; Krassa 1990; Leighley 1990; McLeod et al. 1999; Moscovici 1976, 1980;

Mutz 2002a,b; Mutz & Martin 2001; Nemeth 1986; Nemeth & Kwan 1985; Turner 1991; Walsh 2003). Individuals confronted by a greater diversity of ideas, either in the context of their own social networks or in face-to-face discussions with strangers, tend to be more open-minded, to learn more from others, and to engage in a deeper consideration of issues—in short, to be more deliberative. In contrast, homogeneous groups tend to privilege more intimate kinds of talk that make open discussion of political conflict difficult (Eliasoph 1998).

Central deliberative principles are also at stake. It is difficult to see how equality—a key principle of deliberative democracy—is achieved when deliberative groups are largely white and middle-class. Although every person who is engaged in deliberation may have an equal opportunity to speak, not every person is so engaged when groups are self-selected. Deliberative theorists also argue that participation socializes individuals into being more civic-minded and trusting of others (Haney et al. 2002). But if self-selected members of deliberative groups are already predisposed to being participative and civic-minded, deliberation becomes more a consequence than a catalyst of democratic socialization. Finally, the idea that deliberation increases the legitimacy of outcomes also seems to be threatened by this tendency toward homogeneity. Legitimacy hinges on a belief that all views have been expressed and considered—an unlikely situation given the makeup of most self-selected deliberative groups.

These problems have led some architects of deliberative initiatives to favor random, representative sampling as a method of group formation (Fishkin 1995, Gastil 2000). This process seems to ensure that, even if every person in a community is not given an equal opportunity to deliberate, at least every point of view is included. If the literature is correct, such diversity ought to improve the quality of talk in these encounters. Further, the representativeness of these groups seems to allow a degree of legitimacy that is unattainable via the self-selection method. As Fishkin (1995) puts it, a random, representative sample of Americans should possess the “recommending force of the public’s considered judgments”

(p. 170). Finally, even though this method involves relatively few individuals and thus would seem limited in its capacity to socialize people into civic-mindedness, the participants might return to their social networks as catalysts for greater civic involvement (Huckfeldt & Sprague 1995, Knoke 1990).

However, there are reasons to be cautious about this method of selection. Most obviously, a small group of people—even if randomly selected—cannot represent Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(5)

the views of a community of any size. Random selection may ensure inclusivity, but it does not grant representativeness (Burnheim 1985). Without this latter quality, the legitimacy of the group is open to question. Moreover, a paradox lies at the core of sampling methodology when applied to deliberative conversations.

When individuals represent others not present in a conversation, their views obviously precede participation in a deliberative encounter. Yet, deliberative democrats clearly wish and expect learning to take place. To the extent that learning takes place, individuals cease to represent the community from which they were drawn (Abramson 1994). Ironically, by fulfilling one deliberative principle (learning), the method short-circuits another (representativeness). There is also the question of which community to sample. The obvious answer is that the community consists of individuals who will be directly affected by a decision, but it is not always clear who those people are (Smith & Wales 2000). Moreover, there is little evidence that short-term participation in deliberative exercises—the preferred format of most initiatives that adopt random selection—spurs individuals to greater civic involvement (Kimmelman & Hall 1997, Delli Carpini & Keeter 1996). Thus far, the evidence indicates that such initiatives prompt short-term gains but little long- term civic activity. Finally, random selection can be dreadfully time-consuming and expensive. Fishkin (1995) reports that to get a woman in poverty to his 1996 National Issues Conference in Austin, Texas, he had to personally drive her to the local airport (p. 180). His more recent call for a national deliberation day includes paying participants $150.00 for their participation (Ackerman & Fishkin 2004).

In short, random selection may be neither pragmatically feasible nor normatively preferable.

It is not clear what to make of the conundrum of participant selection. It appears that many of those who would be eager to deliberate already possess the motivation and civic skills to participate in public life. Random sampling may compensate for some of the deficiencies of self-selection, but it has significant drawbacks of its own. Clearer, however, is the fact that deliberative theorists have been content to ignore these pragmatic issues. They simply argue that, as a theoretical matter, deliberation requires equality, and that, once achieved, equality will produce legitimacy. Any difficulty in realizing this equation is an empirical matter having to do with society, not with deliberative theory. One other response has been to focus less on the participation of ordinary people and more on the promotion of deliberation within representative institutions, particularly within the legal system [see Chambers (2003) for a review, Schroeder (2002) for a criticism]. The idea seems to be that if equality and legitimacy cannot be achieved via greater public participation, perhaps already-existing institutions can be made more deliberative and thus more diverse and legitimate. There is a certain logic to this move. But it jettisons a primary catalyst of the movement, that is, to involve greater numbers of ordinary people in the policy-making process.

In the absence of conceptual assistance from the theorists, deliberative architects continue to experiment with new institutional designs (Ackerman & Fishkin 2004, Fung 2004, Fung & Wright 2001, Leib 2004). Perhaps their efforts might be Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(6)

enlivened and even emboldened by greater effort on the part of the theorists to resolve the riddle of participation.

WHAT DOES DELIBERATION LOOK LIKE?

In laboratory experiments, psychologists have shown that deliberation often re- duces the consistency between attitudes and behavior among subjects. It can lead to decisions that not only conflict with expert opinion but also conflict with subjects’ own opinions—that is, decisions they later regret (Holt 1993, 1999; Wilson et al. 1989; Wilson & Schooler 1991). Other studies have shown that deliberation can cause participants to doubt that a “correct” decision is available at all (Armor

& Taylor 2003, Iyengar & Lepper 2000). And still others have found that participants may feel more anxious and frustrated about the issue under discussion after a deliberative encounter than before (Cook & Jacobs 1999, Button & Mattson 1999, Hendriks 2002, Kimmelman & Hall 1997).

Why should this be the case? We simply do not know. And we won’t know, I think, until we learn more about how people actually deliberate with one another.

Surprisingly, this issue remains something of a void in the literature. Delibera- tive theorists say quite a bit about what deliberation ought to look like. Following Habermas (1984, 1996), most assume that deliberation takes place through an ex- change of reasons. A participant defends a view by providing reasons; others probe the usefulness of this view through criticism; by reflecting together on the evidence for and against various views, free and independent participants come to accept what Habermas calls “the force of the better argument.” Other theorists challenge this picture of deliberation (Benhabib 1996, Moon 1991, Phillips 1999). They ask whether every individual must agree that one argument is better than all others.

They wonder whether every individual must accept all of the reasons in support of an argument, or whether neutral standards even exist to allow individuals to make such a judgment. But in pursuing this debate, theorists remain silent about what deliberation looks like on the ground, where real people discuss concrete issues.

Perhaps more surprisingly, the empirical literature has not addressed the issue either. Researchers have been less interested in deliberation itself than in measuring its effects. Whether they use laboratory, survey, or participant-observation methods, the authors of most empirical studies assume that deliberation ensues when certain structural conditions (such as equality and autonomy) hold. By organizing interactions along these lines, they feel free to assume that deliberation takes place, thus allowing them to focus on measuring its effects. Typically, measurement takes the form of pre- and post-tests to ascertain changes in attitudes, beliefs, opinions, and learning. Findings such as more consistency in beliefs and opinions, or greater recall of factual information, are taken to indicate that deliberation has succeeded.

In the process, however, deliberation itself remains essentially unexamined.

This area is ripe for greater investigation. As a start, a few scholars (Lupia 2002, Mendelberg 2002; Rosenberg 2002, 2003a,b) have called for linking the Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(7)

study of deliberation with the “cognitive revolution” that has occurred in the social sciences during the past 30 years (Baars 1986, Gardner 1985, Johnson & Ermeling 1997, Simon 1992). As applied to political psychology, the concept of heuristics is perhaps the most important idea to come out of this tradition. In a nutshell, the idea is that, in any given situation, individuals will reason by using information cues.

Instead of taking in and evaluating all relevant information, individuals take an information shortcut, relying on some subset of information to make a judgment and discarding the rest.

Researchers have found this process at work in every level of choice making. For instance, research on mass opinion has shown that citizens make snap judgments on the basis of party identification, their liking for a candidate, group affiliation, personal ideology, media frames, elite cues, perceptions of likely winners and losers, and a host of other cues (Lupia et al. 2000, Mondak 1994, Mutz et al. 1996, Sniderman et al. 1991; for a review, see Lau & Redlawsk 2001). Several studies by Sears (1993, 2001) show that the mere presence of a symbol, notably race, can trigger reflexive, largely unconscious judgments. The mechanism of this process is in some dispute. Some heuristics appear to be attached to scripts stored in long- term memory: “I was raised Democratic and will always vote Democratic.” Others seem to be stored in short-term or “on-line” memory (Taber et al. 2001). That is, as individuals encounter a new stimulus, they process it on the spot, revising calculations as they go. The difference between the two methods can be important.

It can, for instance, determine whether and how much citizens are persuadable by immediate messages. For our purposes, the general point is key: As mass citizens, individuals rely on heuristics to unreflectively mobilize cognitive structures at their disposal.

Interestingly, a similar process has been observed in small group settings. Re- searchers have found that participants in small group interactions will work together to find some subset of information or cue that allows them to identify common knowledge. This information may be identified in several ways: through the influence of group leaders (Nye & Simonetta 1996, Ridgeway 1987), through the influence of individuals who have a strong motivation to achieve consensus (De Grada et al. 1999, Kruglanski 1996, Webster et al. 1997), through the accep- tance of group stereotypes and identities (Giles et al. 1987, Haslam et al. 1996, Maass et al. 1989, Maas & Arcuri 1996, Messick & Mackie 1989), or through perceptions of group consensus (Sunstein 2002). However it is discovered, this subset of information becomes an information shortcut, providing the group with a basis on which to select subsequent information unconsciously but efficiently (Davis et al. 1989; Gigone & Hastie 1993, 1997; Kameda 1991; Kerr & Kaufman- Gilliland 1994; Nemeth & Rogers 1996; Orbell et al. 1988; Schulz-Hardt et al.

2000; Winquist & Larson 1998; Wittenbaum et al. 1999). Small groups arrive at these information cues in a slightly different manner than do mass societies.

Mass-mediated cues require individuals to identify appropriate heuristics and personally mobilize pertinent scripts. In small groups, the same process occurs in social interaction. Individuals engage in microrituals of social behavior to avoid Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(8)

conflict, identify points of agreement, and reach consensus (Brown & Levinson 1987, Mulkay 1985, Pomerantz 1984, Schiffrin 1990, Sheldon 1992). The outcome is much the same: judgments based on information shortcuts that mobilize scripts and thus allow groups to reach unconscious rather than deliberate judgments.

This brings us to a key insight: Deliberation represents a disturbance of everyday reasoning habits. People prefer to rely on routine scripts to navigate through their social world. Being jolted out of these scripts is, generally speaking, a dis- concerting experience. This directly implicates emotions in the process of deliberation. Marcus et al. (2000) elaborate this point in a theory of the role of emotions in political judgment. According to them, human judgment is regulated by two emotional systems, which they call, respectively, the dispositional and the surveillance systems. The dispositional system regulates the domain of habit and routine.

Linked to conscious awareness and attached to procedural memory, the dispositional system monitors our interactions with familiar environments by adjusting emotional responses and calling up learned scripts in procedural memory. “In- sofar as. . .behavior. . .falls within the realm of learned behaviors,” they explain,

“the disposition system. . .play[s] a role in the initiation, adaptation, and control of the plan of action. Moreover, reliance on habits, most of which are developed without explicit reasoning, provides efficient and therefore reasonable solutions to the recurring tasks of daily life” (Marcus et al. 2000, p. 52). In contrast, the surveillance system monitors novel or threatening environmental stimuli. It kicks in when habits are disrupted and routines break down. Associated with feelings of anxiety and unease, the surveillance system makes us more attentive to our environment and to assessing new information. “When activated,” Marcus et al.

(2000) conclude, “the surveillance system shifts our conscious state away from the task at hand and toward an explicit consideration of what we should choose as the best course of action” (p. 58).

This cognitive shift may or may not be natural, but ordinarily people are reluctant to make it, precisely because it involves frustration and anxiety. This helps us to understand why individuals tend to be hesitant deliberators, preferring to “pass the buck” when they can and to rely on information short cuts when they cannot. It also helps us to understand why participation in deliberation may produce greater anxiety and frustration than other choice-making processes. It is unsettling to have one’s cognitive scripts disrupted, and it is even more frustrating to recognize that no new script is forthcoming, since decisions about public issues are necessarily complex and admit no easy answers. As the empirical research shows, this situation may engender greater sympathy for public officials, but it also often results in feelings of powerlessness and hostility toward ultimate decisions.

Why, then, should individuals let themselves be bothered enough to be jolted out of their everyday reasoning habits? This question highlights the crucial role of motivation in deliberative reasoning. A motivation is an incentive, or a drive, to do something. When we say that individuals succeed in deliberating, we mean that they have been motivated to overcome historical, structural, and psychological impediments to intentional reflection. Political scientists and psychologists have Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(9)

been interested in this subject, though not formulated in these terms, for some time [see Sorrentino & Higgins (1986) and Taber et al. (2001) for reviews]. They have shown, for instance, that individuals motivated to preserve prior beliefs are less willing to veer from conventional scripts (Festinger 1957). In recent years, the study of motivation has become more systematic (Kunda 1990). Psychologists have identified a continuum of motivational goals, ranging from maintaining prior beliefs to obtaining an accurate conclusion (Baumeister & Newman 1994, Kunda 1990). Within this continuum, deliberation is associated with accuracy goals. In- dividuals motivated to reach accurate conclusions are more likely to engage in an intentional consideration of symbolic stimuli. Thus, when we say that people deliberate, we assume that they are driven by a motivation to be accurate. What kinds of mechanisms might prompt this motivation?

Marcus et al. (2000) give the short answer: things that make us uncomfortable.

Other empirical research provides a longer answer. So far, researchers have found three conditions that tend to motivate individuals to adopt a deliberative frame of mind: accountability, high stakes, and diversity. Experimental work has shown that individuals who are told that they will have to discuss their judgments publicly are more likely to process more information more objectively (Tetlock 1983, 1985).

Related to this notion, but less well documented, is the idea that perceptions of consequences will also influence motivation (Taber et al. 2001). If consequences are perceived to be great and direct, then individuals ought to expend more energy to get decisions right. Finally, as discussed above, other work has shown that deliberation is more likely in diverse groups. Moscovici (1976, 1980) finds that, under some conditions, minority group members can offer novel views that spur majority members to learn—that is, to veer from established scripts toward a deeper consideration of new ideas (see also Nemeth 1986, Nemeth & Kwan 1985, Turner 1991). Similarly, Huckfeldt (1986) and Huckfeldt & Sprague (1995) argue that political participation increases the diversity of one’s social networks. Moreover, there is some evidence that diversity of social networks prompts a more deliberative frame of mind (Mutz 2002a,b; Mutz & Martin 2001). By taking people out of their comfort zones, these conditions may instigate more considered judgment.

So far so good. But does being uncomfortable in and of itself prompt deliberation? Not quite. Even within such structurally promising situations, some individuals will express disappointment with a deliberative encounter while others will come away perfectly satisfied. Why should this be the case? As a way into this question, we might note that all three of our motivating prompts involve psychological assessments of self in relation to other people or to one’s environment.

One becomes motivated to deliberate when one is accountable to others; when one perceives oneself to be threatened by others or by one’s environment; when one encounters others different from oneself. In other words, motivation is a culturally and socially constructed drive.

At first glance, this insight seems to violate the cognitive model of reasoning with which we began. As Gardner (1985) observes, for much of its career, the cognitive revolution has been spurred by an assumption that all humans reason Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(10)

in the same standard ways, regardless of context. “Nearly all cognitive scientists have conspired to exclude from consideration such nontrivial factors as the role of the surrounding context, the affect aspects of experience, and the effects of cultural and historical factors on human behavior and thought” (p. 387). And there is more than a grain of truth in this assumption. At least a basic structure of human reasoning is the birthright of every human [Hutchins (1980); for criticisms, see Bruner (1990), Putnam (1981)]. However, a growing body of work argues that more complex forms of reasoning are culturally and socially conditioned (Cole 1996, Conover & Searing 2002, D’Andrade 1995, Kuklinski 2001, Kuklinski &

Hurley 1996, Nye & Brower 1996, Resnick et al. 1991, Shweder 1991, Sniderman et al. 2001). Mounting research suggests that the “mere presence” of others is enough to alter a person’s cognitive activity (Levine et al. 1993). Other evidence shows that the way individuals mobilize and organize memory in complicated cognitive tasks differs across cultures (Cole & Scribner 1974, Scribner & Cole 1981, Shweder 1993, Stigler et al. 1990). Cognition, in other words, is not solely hard wired; it also involves cultural software. To accept this idea is not to enter a world of total cultural relativism. Remember, all humans share a basic architecture of the mind. Nor is it to lapse into a form of determinism in which individuals simply bear their inherited culture; after all, people must learn and adapt shared mental models to their own lives. Rather, it is to accept and perhaps deepen an image of culture offered by Geertz (1973) more than 25 years ago: Culture is a

“web of significance,” which people living in communities collectively spin, and which, we might add, is mapped onto the brains of its individual members (p. 5).

How ought this meeting of cognitive and cultural theory change our view of deliberation? Absent empirical research, it is difficult to know with any precision, but Bruner (1986) gives some helpful suggestions. As we know, deliberative theorists tend to conceive of deliberation as argument, a form of communication that Bruner describes as a “formal mathematical system of description and explana- tion.” This view is based on a conceptual preference for argumentation and logic as legitimate forms of deliberative discourse. But of course political argument is never detached from its social and cultural circumstances (Laden 2001). Rather, as Bruner (1986) puts it, deliberation always “deals in human or human-like intention and action. . .[that] strives to put its timeless miracles into the particulars of experience, and to locate the experience in time and place” (pp. 12–13). To the extent that deliberation combines cognition (the act of making sense) with culture (the act of making meaning), it probably looks more like storytelling than argumentation.

As Fisher (1999) puts it, “the idea of humans as storytellers indicates the general form of all symbol composition: it holds that symbols are created and communi- cated ultimately as stories meant to give order to human experience and to induce others to dwell in them to establish ways of living in common, in communities. . .”

(p. 271). To the extent that narrative functions as a “metacode” (White 1980, p. 7), which organizes symbolic stimuli into recognizable patterns, it captures the sense of the terms script, schema, or model (see also Mink 1978). Stories bring order to human experience. But the notion of narrative connects cognition with other Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(11)

elements—norms and identity—that are also central to community. As Ricoeur (1980) observes, once organized, stories may be repeated; repeated often enough, they become tradition; and tradition is the basis of community. Put another way, beyond cognitive understanding, the ultimate force of storytelling is moral and constitutional. It is moral in the sense that stories tell us not only what happened but also what ought to have happened (White 1980). It is constitutional in the sense that it produces the conditions for self-identification (Somers 1994). Thus, stories place motivation—a central element in attaining and sustaining a deliberative mindset—into human experience; they provide a footing, to borrow a term from Goffman (1974), on which understanding and self-understanding might take place all at once.

If we grant that deliberation looks like storytelling, then we have to rethink the theoretical link between reason and autonomy. In deliberative theory, the use of reason is seen to promote autonomy to the extent that individuals freely and independently engage in an exercise of critical reflection. This assumption has led many empirical researchers to look skeptically at any outside influence on individual reflection (Bartels 1998, Edelman 1993, Manheim 1991, Parenti 1999). But this skepticism only makes sense if we assume that humans share basic cognitive structures that compel them to deliberate in the same ways regardless of context.

Research in social and cultural psychology does not support this view. Instead, it suggests that deliberation is associated at least as much with community as with autonomy. Recent work in deliberative theory has begun to develop this perspective (Connolly 2002, Devereaux 2000, Laden 2001, Mouffe 2000, Rosenberg 2002).

It implies that for deliberation to be successful, motivating conditions such as accountability, high stakes, and diversity must be set in a cultural context that enables roles, identities, and norms compatible with a deliberative frame of mind.

A literature on participation in social movements perhaps demonstrates this idea best. Gamson (1992) observes that groups motivated by stories of injustice develop a deeper, richer, more reflective texture of talk. They track arguments more closely, offer more of their own views for the group’s consideration, and ultimately feel and act more connected to public life. In other words, such stories catalyze a sense of self that overcomes disincentives to participate in public life, and they form a barrier to disappointment about ultimate outcomes. As Mansbridge (1980) found in her study of Vermont town meetings, people are loath to participate in political conversations with their fellow citizens—so much so that, as one participant put it, “I wouldn’t say a word. . .unless they got me madder’n hell” (p. 60). Being

“madder’n hell” may motivate people to deliberate. But it is not enough to ensure that the subsequent interaction will be successful. Rather, the empirical literature on reason suggests that anger must be accompanied by a commonly shared narrative that promotes deliberative roles and norms.

What will we find when we devote greater attention to deliberation as a form of discourse? The cognitive literature suggests that we will find deliberation to be episodic, difficult, and tentative. Within any particular interaction, deliberation may ebb and flow as participants alternately resist and accept the challenge of Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(12)

deliberation. And we will find that deliberation sustains itself through these con- versational eddies by means of the coconstruction of identities and values that keep people motivated and engaged. This picture of deliberation does not destroy the image proffered by deliberative theorists. It does, however, suggest some revisions.

Habermas may be correct that deliberation is a natural human talent, but it is not easy to cultivate and maintain. The key to successful deliberation lies in the manner in which individuals collectively account for problems. As the initial studies show, it is as likely that groups will talk their way out of deliberating as it is that they will hunker down to do the difficult work of sifting through the choices that lie before them. Successful deliberation not only helps groups evaluate choices but also provides the cultural glue that keeps them engaged in the task. It is left to subsequent research to identify particular keys, strategies, or patterns of talk that assist in this outcome.

THE PRODUCT OF DELIBERATIVE TALK

Suppose for a moment that a deliberative initiative has overcome all the obstacles we have discussed so far. Its creators have devised a sampling procedure that ensures equality among its members and the legitimacy of its deliberations. It has facilitated conversations among its members in such a way that participants feel motivated to do the hard work of intentional reflection, cognitively able to handle its complexities, and culturally empowered to believe that their work can make a difference. Even after accomplishing all of this, the initiative can still go awry.

Once a group has deliberated and reached its judgment, that choice must then enter the political system in which policy decisions are made. In the United States, this means that deliberative choices enter a pluralist system populated by, among others, elected officials, bureaucrats, and representatives of interest groups. The possibility of realizing a deliberative democracy depends in part on successfully linking deliberation to this political system.

Empirical researchers have examined three responses to this issue [see Leroux et al. (1998) and Rowe & Frewer (2000) for discussions]. First, a deliberative initiative may be designed to avoid an explicit linkage between deliberation and policy making. National Issues Forums (NIFs) take this form. NIFs invite participants to deliberate on an issue using materials provided by forum organizers.

These materials are quite explicit that the goal of the discussion is education, not policy making. Participants are encouraged to see the process as an opportunity to learn and reflect, not to offer guidance to policy makers.

Second, in the consultative mode, representative bodies may be mandated to consult, but not abide by, the outcome of a deliberative initiative. Deliberative polls and citizens’ juries are examples of this mode (Fishkin 1995, Gastil 2000).

Typically, representative bodies use these formats to gauge public opinion. In this mode, a deliberatively made choice serves as one input among others for a policy maker’s consideration.

(13)

Third, in the decision-making mode, policy officials are explicitly bound by the decisions of deliberative groups. When used, the decision-making mode usually involves the relevant “stakeholders” in a systematic process of reflection on a defined set of policy options (Button & Ryfe 2005, Ryfe 2002).

Interestingly, I have come across few examples of this last mode in the United States [but see Fung & Wright (2001) and Renn et al. (1995) on similar initiatives in other countries]. That is, the “ordinary citizen” representing only her views rarely makes an appearance in deliberative initiatives. Instead, most initiatives focus their efforts either on education or consultation. Put another way, most initiatives imagine that the ultimate impact of deliberation is on public opinion and not the policy-making process. This evident reluctance to incorporate citizen deliberation more fully into policy making indicates a structural ambivalence within deliberative democracy about the relationship between talk and action.

Warren’s (2001) work on associations and democracy illuminates this ambivalence. We know that people prefer a psychological state in which they can rely on familiar cognitive routines and scripts. When dislodged from this condition, they will engage in more considered reflection, but they will also seek a return to equi- librium. Warren notes that different kinds of associations will help them achieve this goal in different ways. Some associations, such as social service providers, will simply take care of their problem without deliberation. Other advocacy groups will champion their cause in other venues. Deliberative associations—groups that foster and promote deliberation as a way to solve common problems—have a particular profile. They tend to be political groups; they are relatively easy to exit;

and they favor talk as a medium of decision making (p. 65). This profile is difficult to maintain because an emphasis on deliberative talk can easily inhibit a desire for political action (and vice versa). On the one hand, because deliberative groups are easy to exit, those that stress action will tend to become cognitively homogeneous as those who think differently from the growing group consensus exit [on the psychology of this process, see Gollwitzer (1990) and Gollwitzer &

Bayer (1999)]. On the other hand, groups that seek to preserve deliberative talk will either avoid “political” issues altogether or choose issues that allow a limited range of disagreement (Eliasoph 1998). Finally, a group may focus on process more than outcomes. And this is precisely what we see happening in the deliberative field. Along with NIFs, Warren (2001) observes that such foundations as the Pew Charitable Trusts and the Kaiser Family Foundation also fit this profile (p. 164). Each of these foundations dedicates itself to improving the critical skills of citizens. In so doing, however, they also work against the basic cognitive logic that motivates people to deliberate in the first place. Without feeling that the stakes are high, or that they are accountable for an outcome, individuals will be less willing to engage their critical faculties.

Warren’s work shows, I think, that deliberative associations face a daunting challenge when seeking to link their efforts to the wider political system. To the extent that a group favors deliberation above all else, it will tend to avoid or constrain explicit linkages to the political system. By contrast, to the extent that a Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(14)

deliberative group seeks real political action, it risks losing its diversity of views.

This is not to say that associations will necessarily fall into these traps, only that the structural conditions of conventional politics make such an outcome as likely as not.

One might suppose that the consultative mode, requiring policy officials to consult but not accept deliberatively made choices, can avoid these dilemmas.

But this mode raises its own paradox. As Warren notes, deliberative groups tend to believe that political choices are legitimate only to the extent that they are made in a deliberative fashion, but this principle conflicts with the terms on which officials and policy experts evaluate public decisions. For members of these “strong publics” (Fraser 1989), decision making involves an artful compromise between the technicalities of issues and the politics of interest-group bargaining. Given this perspective, it is not surprising that they chafe at the values implicit in the deliberative model, i.e., that decisions are legitimate only if they arise from open discussion among equals. Consider how policy officials and experts might respond if a deliberative group came to a decision on where to put a waste-management facility, ignoring the political realities of a local community and overlooking a crucial technical detail of how waste-management systems work. In this situation, policy experts might rightly look on a deliberative outcome with suspicion and even contempt. Ironically, a choice that enjoys the legitimacy conveyed by a deliberative process may well lack political legitimacy. At the same time, were policy officials to summarily dismiss a deliberative group’s judgment, one can understand why its participants might come away from the process more disenchanted with politics than ever. Observations of real deliberative initiatives find that these reactions are quite common (see Button & Mattson 1999, Hendriks 2002, Kimmelman & Hall 1997, Ryfe 2002). This work suggests that the term “consultation” elides important differences in the assumptions and expectations that deliberative groups and policy officials bring to public decision making. If these differences are not accounted for, a deliberative initiative can end unhappily for all involved.

These practical challenges raise the conceptual issue of how to coordinate deliberation with representative democracy. It is one thing to argue abstractly that contemporary politics might be reinvigorated with greater deliberation and participation. It is quite another to make interactions between ordinary people and policy makers actually work. As with the issue of participation, deliberative theorists sometimes respond to these difficulties by retreating into the abstractions of proceduralism or constitutionalism. But, just as practitioners of deliberative democracy would do well to buttress their conceptualization of what they do, it seems to me that deliberative theory can only be invigorated by closer contact with empirical realities.

CONCLUSION

Does deliberative democracy work? The empirical literature answers this question with a qualified yes. Even if it is a natural human capacity, deliberation is not easy.

It seems to require a mixture of knowledge/skills, motivation, and civic identity. It Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(15)

is difficult to create conditions to bring these elements together. It is perhaps even more difficult to sustain them once they are created. At every step of this process, basic concepts such as equality, legitimacy, reason, autonomy, representation, and democracy are at stake. But although deliberation is difficult and fragile, it is not impossible. On my reading, five mechanisms seem to be particularly associated with successful deliberation: rules, stories, leadership, stakes, and apprenticeship.

1. Rules. “Fully public democratic conversation takes place,” Schudson (1997) writes, “in settings where talk is bound to be uncomfortable. . . . Such talk is threatening enough to require formal or informal rules of engagement”

(p. 306). Precisely because people seem disinclined to deliberate—and if at all, not for very long—explicit rules must prop up deliberative initiatives.

Rules of equality, civility, and inclusivity may prompt deliberation even when our first impulse is to avoid it. They may institutionalize deliberation as a routine process. Once institutionalized, they ensure that deliberation continues over time, perhaps even across generations. Finally, during actual exchanges, rules help participants ensure that their judgments are reflective and based on a full range of information.

2. Stories. Deliberative theorists sometimes seem to adopt an “if we build it they will come” mentality. If we infuse a context with the right procedures, and organize an encounter to conform to the right norms (equality, civility, etc.), then deliberation ought to take place. But rules may mean little if individuals do not feel accountable for outcomes, and even less if participants are not imbued with a civic identity that harnesses them to the task even when the going gets tough. Successful deliberation seems to require a form of talk that combines the act of making sense (cognition) with the act of making meaning (culture). Storytelling is one such form of talk. Stories anchor reality by organizing experience and instilling a normative commitment to civic identities and values. Once set, stories function as a medium for framing discussions (Farr 1993).

3. Leadership. Leaders provide important cues to individuals in deliberative settings. They can steer small groups toward nondeliberative conversations by insisting on the salience of particular cues. Alternatively, they can keep groups on a deliberative track when their members prefer to slip into routine and habit. At the level of mass society, leaders often manipulate cues to achieve personal political goals. Indeed, Zaller (1992) argues that public opinion is largely a product of elite cues as they are transmitted by the news media. If this is the case, it stands to reason that leaders who engage in more thoughtful rhetoric may prime citizens to adopt a more deliberative posture.

Like rules, leaders may act as sea walls against the tide of routine habits of reasoning.

4. Stakes. Individuals are more likely to sustain deliberative reasoning when outcomes matter to them. Tetlock’s (1983, 1985) experimental work on accountability suggests as much. In a different way, Forester (1999) argues for Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(16)

much the same principle. His work on deliberation and environmental planning finds that individuals who are included in a policy-making process from the beginning become more invested in the process than individuals brought in at the end to choose among a range of predetermined options. Put simply, deliberation works best when individuals are invested in the outcome.

5. Apprenticeship. Few people would characterize public life at any level of American society as “deliberative.” Moreover, as I have been at pains to show, people tend to prefer nondeliberative forms of reasoning. Therefore, ordinary people have little experience with deliberation, and presumably little skill. This insight has prompted calls for renewed civic education. But what form should this education take? Basic political knowledge is necessary, but it is not a sufficient spur to deliberation. The arts of rhetoric are another obvious subject, but deliberation is a way of doing politics rather than a way of speaking about political subjects. We know that deliberation is shaped by culture and society. This implies that abstract forms of argument are not as central to deliberation as theorists sometimes imply. Instead, we might do well to imagine education as a form of apprenticeship learning (Lave 1988, Lave & Wenger 1991), in which individuals learn to deliberate by doing it in concert with others more skilled in the activity. In apprenticeship, new skills emerge from the sensuous but guided activity of deliberating in real contexts and not from the rote recall of information. One might enable such learning in any context of public decision making, simply by establishing deliberative mechanisms, providing effective leaders, and guiding ordinary people through the process.

These five mechanisms stand out as critical to the successful design of deliberative initiatives. There are probably more, and to discover them, a great deal more research remains to be done. Despite its breadth, the empirical study of deliberation is not yet very rich or deep. More integration across disciplinary boundaries would be useful. Political scientists have made great strides in recent years by linking the study of opinion formation to cognitive science and psychology. But other literatures—in communication, sociology, social psychology, linguistics, and anthropology—remain almost untapped. Moreover, the theory of deliberative democracy needlessly remains removed from its practice. Theorists and applied researchers alike would benefit from greater interaction. We need to know more about the specific political contexts in which deliberation is likely to succeed. Work by Fung (2004) in particular has begun to grapple with this question, but it is an area ripe for further investigation. Finally, and perhaps most impor- tantly, we must learn more about what deliberation actually looks like. It simply will not do to place the very practice under investigation into a black box. Psy- chologists and small group communication scholars provide hints about the nature of deliberation as a form of communication. Political scientists might follow their lead by investigating deliberation in the natural political contexts in which it takes place (e.g., Walsh 2004). Extending our research agendas in these directions can Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(17)

only enhance our understanding of the possibilities and limitations of deliberative democracy.

The Annual Review of Political Science is online at http://polisci.annualreviews.org

LITERATURE CITED

Abelson J, Eyles J, Smith P, Martin E, Martin FP. 2002. Obtaining public input for health systems decisionmaking: past experiences and future prospects. Can. Public Admin. 45:

70–97

Abelson J, Forest PG, Eyles J, Smith P, Martin E, Gauvin FP. 2003. Deliberations about deliberative methods: issues in the design and evaluation of public participation processes.

Soc. Sci. Med. 57:239–51

Abramson J. 1994. We, the Jury. New York:

Basic

Ackerman B, Fishkin J. 2004. Deliberation Day. New Haven, CT: Yale Univ. Press Armor D, Taylor S. 2003. The effects of mind-

set on behavior: self-regulation in deliberative and implementational frames of mind.

Pers. Soc. Psychol. B 29:86–95

Baars B. 1986. The Cognitive Revolution in Psychology. New York: Guilford

Barber B. 1984. Strong Democracy: Participa- tory Politics for a New Age. Berkeley: Univ.

Calif. Press

Bartels L. 1998. Democracy with attitudes. Pre- sented at Annu. Meet. Am. Polit. Sci. Assoc., Boston

Baumeister R, Newman L. 1994. Self-regulation of cognitive inference and decision processes. Pers. Soc. Psychol. B 20:3–19 Benhabib S, ed. 1996. Democracy and Differ-

ence: Contesting the Boundaries of the Po- litical. Princeton, NJ: Princeton Univ. Press Bierle TC. 1999. Using social goals to evaluate

public participation in environmental deci- sions. Policy Stud. Rev. 16:75–103

Bobbio N. 1987. The Future of Democracy: A Defense of the Rules of the Game. Transl. R.

Griffin (from Italian). Cambridge, UK: Polity Brown P, Levinson S. 1987. Politeness: Some

Universals in Language. Cambridge, UK:

Cambridge Univ. Press

Bruner J. 1986. Actual Minds, Possible Worlds.

Cambridge, MA: Harvard Univ. Press Bruner J. 1990. Acts of Meaning. Cambridge,

MA: Harvard Univ. Press

Burnheim J. 1985. Is Democracy Possible? The Alternative to Electoral Politics. Cambridge, UK: Polity

Burns N, Schlozman K, Verba S. 2001. The Pri- vate Roots of Public Action: Gender, Equal- ity, and Political Participation. Cambridge, MA: Harvard Univ. Press

Button M, Mattson K. 1999. Deliberative democracy in practice: challenges and prospects for civic deliberation. Polity 31:

609–37

Button M, Ryfe D. 2005. What can we learn from deliberative democracy? In Where the Public Meets: Assessing Innovative Meth- ods for Promoting Citizen Deliberation and Dialogue Across the Globe, ed. J Gastil, P Levine. Boston: Jossey-Bass. In press Cole M. 1996. Cultural Psychology: A Once

and Future Discipline. Cambridge, MA: Har- vard Univ. Press

Cole M, Scribner S. 1974. Culture and Thought:

A Psychological Introduction. New York:

Wiley

Connolly W. 2002. Neuropolitics: Thinking, Culture, Speed. Minneapolis: Univ. Minn.

Press

Conover P, Searing D. 2002. Expanding the envelope: citizenship, contextual methodolo- gies, and comparative political psychology.

See Kuklinski 2001, pp. 89–114

Conway M. 2000. Political Participation in the United States. Washington, DC: CQ Press.

3rd ed.

(18)

Cook FL, Jacobs LR. 1999. Deliberative Democracy in Action: Evaluation of Amer- icans Discuss Social Security. Washington, DC: Pew Charit. Trust

D’Andrade R. 1995. The Development of Cog- nitive Anthropology. Cambridge, UK: Cam- bridge Univ. Press

Davis J, Kameda T, Parks C, Stasson M, Zim- merman S. 1989. Some social mechanics of group decision making: the distribution of opinion, polling sequence, and implications for consensus. J. Pers. Soc. Psychol. 57:

1000–12

De Grada E, Kruglanski A, Mannetti L, Pierro A. 1999. Motivated cognition and group interaction: need for closure affects the con- tents and processes of collective negotiations.

J. Exp. Psychol. 35:346–65

Delli Carpini M, Keeter S. 1996. What Ameri- cans Know About Politics and Why It Mat- ters. New Haven, CT: Yale Univ. Press Denver D, Hands G, Jones B. 1995. Fishkin and

the deliberative opinion poll: lessons from a study of the Granada 500 television program.

Polit. Commun. 12:147–56

Devereaux M. 2000. Cultural Pluralism and Dilemmas of Justice. Ithaca, NY: Cornell Univ. Press

Downs A. 1957. An Economic Theory of Democracy. New York: Harper

Dunkerly D, Glasner P. 1998. Empowering the public? Citizens’ juries and the new genetic technologies. Crit. Public Health 8:181–

92

Edelman M. 1993. Contestable categories and public opinion. Polit. Commun. 10:231–42 Eliasoph N. 1998. Avoiding Politics: How

Americans Produce Apathy in Everyday Life.

New York: Cambridge Univ. Press

Farr J. 1993. Framing democratic discussion. In Reconsidering the Democratic Public, ed. G Marcus, R Hanson, pp. 379–91. University Park: Penn. State Univ. Press

Festinger L. 1957. A Theory of Cognitive Dis- sonance. Stanford, CA: Stanford Univ. Press Festinger L. 1964. Conflict, Decision, and Dis- sonance. Stanford, CA: Stanford Univ. Press Fisher W. 1999. Narration as a human commu-

nication paradigm: the case of public moral argument. In Contemporary Rhetorical The- ory, ed. JL Lucaites, CM Condit, S Caudill, pp. 265–87. New York: Guilford

Fishkin J. 1991. Democracy and Deliberation:

New Directions for Democratic Reform. New Haven, CT: Yale Univ. Press

Fishkin J. 1995. The Voice of the People: Public Opinion and Democracy. New Haven, CT:

Yale Univ. Press

Fiske S, Taylor S. 1991. Social Cognition. New York: McGraw-Hill

Forester J. 1999. The Deliberative Practitioner:

Encouraging Participatory Planning Pro- cesses. Cambridge, MA: MIT Press Fraser N. 1989. Unruly Practices: Power, Dis-

course, and Gender in Contemporary Social Theory. Minneapolis: Univ. Minn. Press Freeman S. 2000. Deliberative democracy: a

sympathetic comment. Philos. Public Aff. 29:

371–419

Fung A. 2001. Accountable autonomy: toward empowered deliberation in Chicago schools and policing. Polit. Soc. 29:73–103 Fung A. 2004. Empowered Participation: Rein-

venting Urban Democracy. Princeton, NJ:

Princeton Univ. Press

Fung A, Wright EO. 2001. Deepening democracy: innovations in empowered participa- tory governance. Polit. Soc. 29:5–41 Gamson W. 1992. Talking Politics. Cambridge,

UK: Cambridge Univ. Press

Gardner H. 1985. The Mind’s New Science:

A History of the Cognitive Revolution. New York: Basic

Gastil J. 2000. By Popular Demand: Revitaliz- ing Representative Democracy Through De- liberative Elections. Berkeley: Univ. Calif.

Press

Gastil J, Deess EP, Weiser P. 2002. Civic awak- ening in the jury room: a test of the connec- tion between deliberation and political par- ticipation. J. Polit. 64:585–94

Gastil J, Dillard J. 1999. Increasing political sophistication through public deliberation.

Polit. Commun. 16:3–23

Geertz C. 1973. Thick description: toward an interpretive theory of culture. In The Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(19)

Interpretation of Cultures, pp. 3–32. New York: Basic

Gigone D, Hastie R. 1993. The common knowledge effect: sharing information and group judgment. J. Pers. Soc. Psychol. 65:959–74 Gigone D, Hastie R. 1997. The impact of infor-

mation on small group choice. J. Pers. Soc.

Psychol. 72:132–40

Giles H, Mulac A, Bradac J, Johnson P.

1987. Speech accommodation theory: the first decade and beyond. Commun. Yearb.

10:13–48

Goffman E. 1974. Frame Analysis: An Essay on the Organization of Human Experience.

Cambridge, MA: Harvard Univ. Press Gollwitzer PM. 1990. Action phases and mind-

sets. In Handbook of Motivation and Cog- nition: Foundations of Social Behavior, ed.

RM Sorrentino, ET Higgins, 2:53–92. New York: Guilford

Gollwitzer PM, Bayer U. 1999. Deliberative versus implemental mindsets in the control of action. In Dual-Process Theories in So- cial Psychology, ed. S Chaiken, Y Trope, pp.

403–22. New York: Guilford

Goodin R. 2000. Democratic deliberation within. Philos. Public Aff. 29:81–109 Goodin R. 2004. Reflective Democracy. New

York: Oxford Univ. Press

Graham KA, Phillips SD, eds. 1998. Citizen engagement: lessons in participation from local government administration. Monogr. Can.

Public Admin. No. 22. Toronto: Inst. Public Admin. Can.

Habermas J. 1984. The Theory of Communica- tive Action. 2 Vols. Transl. T McCarthy (from German). Boston: Beacon

Habermas J. 1996. Between Facts and Norms:

Contributions to a Discourse Theory of Law and Democracy. Transl. W Rehg (from Ger- man). Cambridge, MA: MIT Press

Haney B, Borgida E, Farr J. 2002. Citizen- ship and civic engagement in public problem- solving. Res. Micropolit. 6:225–52

Haslam SA, Oakes P, Turner J, McGarty C. 1996. Social identity, self-categorization, and the perceived homogeneity of ingroups and outgroups: the interaction between social

motivation and cognition. See Sorrentino &

Higgins 1986, pp. 182–222

Hendriks C. 2002. Institutions of deliberative democratic processes and interest groups:

roles, tensions and incentives. Austr. J. Public Admin. 61:64–75

Hibbing J, Theiss-Morse E. 1995. Congress as Public Enemy. Public Attitudes Toward American Political Institutions. Cambridge, UK: Cambridge Univ. Press

Hibbing J, Theiss-Morse E. 2000. Deliberation as a source of system delegitimation and pop- ular disharmony. Presented at Annu. Meet.

Midwest Polit. Sci. Assoc., Chicago Holt L. 1993. Rationality is hard work: an alter-

native interpretation of the disruptive effects of thinking about reasons. Philos. Psychol.

6:251–66

Holt L. 1999. Rationality is hard work: some further notes on the disruptive effects of de- liberation. Philos. Psychol. 12:215–19 Huckfeldt R. 1986. Politics in Context: Assimi-

lation and Conflict in Urban Neighborhoods.

New York: Agathon

Huckfeldt R, Sprague J. 1995. Citizens, Pol- itics, and Social Communication: Informa- tion and Influence in an Election Campaign.

New York: Cambridge Univ. Press

Hutchins E. 1980. Culture and Inference. Cam- bridge, MA: Harvard Univ. Press

Iyengar SS, Lepper M. 2000. When choice is demotivating: can one desire too much of a good thing? J. Pers. Soc. Psychol. 79:995–

1006

Janis IL, Mann L. 1977. Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment. New York: Free

Johnson D, Ermeling C, eds. 1997. The Future of the Cognitive Revolution. New York: Ox- ford Univ. Press

Kahneman D, Slovic P, Tversky A, eds. 1982.

Judgment Under Uncertainty: Heuristics and Biases. New York: Cambridge Univ.

Press

Kahneman D, Tversky A. 1983. Choices, val- ues, and frames. Am. Psychol. 39:341–50 Kameda T. 1991. Procedural influence in small-

group decision making: deliberation style Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

(20)

and assigned decision rule. J. Pers. Soc. Psy- chol. 61:245–56

Kerr N, Kaufman-Gilliland C. 1994. Communi- cation, commitment, and cooperation in so- cial dilemmas. J. Pers. Soc. Psychol. 66:513–

29

Kim J, Wyatt R, Katz E. 1999. News, talk, opinion, participation: the part played by conver- sation in deliberative democracy. Polit. Com- mun. 16:361–85

Kimmelman D, Hall G. 1997. A critical study of civic deliberation and the political process.

Walt Whitman Cent. Culture of Politics and Democracy. Rutgers Univ., New Brunswick, NJ

Knoke D. 1990. Political Networks: A Struc- tural Perspective. New York: Cambridge Univ. Press

Kraft M, Clary B. 1991. Citizen participation and the NIMBY syndrome: public response to radioactive waste disposal. West. Polit. Q.

44:299–328

Krassa MA. 1990. Political information, so- cial environment, and deviants. Polit. Behav.

12:315–30

Kruglanski A. 1986. A motivated gatekeeper of our minds: need-for-closure effects on inter- personal and group processes. See Sorrentino

& Higgins 1986, pp. 465–96

Kuklinski J, ed. 2001. Citizens and Poli- tics: Perspectives from Political Psychology.

Cambridge, UK: Cambridge Univ. Press Kuklinski J, Hurley N. 1996. It’s a matter of in-

terpretation. See Mutz et al. 1996, pp. 125–44 Kuklinski J, Riggle E, Ottati V, Schwarz N, Wyer R. 1993. Thinking about political tol- erance, more or less, with more or less in- formation. In Reconsidering the Democratic Public, ed. G Marcus, R Hanson, pp. 225–47.

University Park: Penn. State Univ. Press Kunda Z. 1990. The case for motivated reason-

ing. Psychol. Bull. 108:480–98

Laden A. 2001. Reasonably Radical: Deliber- ative Liberalism and the Politics of Identity.

Ithaca, NY: Cornell Univ. Press

Landeman M. 2002. Opinion quality and pol- icy preferences in deliberative research. Res.

Micropolit. 6:195–221

Lau R, Redlawsk D. 2001. Advantages and dis- advantages of cognitive heuristics in political decision making. Am. J. Polit. Sci. 45:951–71 Lave J. 1988. Cognition in Practice: Mind, Mathematics and Culture in Everyday Life.

Cambridge, UK: Cambridge Univ. Press Lave J, Wenger E. 1991. Situated Learning:

Legitimate Peripheral Participation. Cam- bridge, UK: Cambridge Univ. Press Leib P. 2004. Deliberative Democracy in Amer-

ica: A Proposal for a Popular Branch of Gov- ernment. University Park: Penn. State Univ.

Press

Leighley JE. 1990. Social interaction and contextual influences on political participation.

Am. Polit. Q. 18:459–75

Leroux T, Hirtle M, Fortin LN. 1998. An overview of public consultation mechanisms developed to address the ethical and social issues raised by biotechnology. J. Consum.

Res. 21:445–81

Levine J, Resnick L, Higgins ET. 1993. Social foundations of cognition. Annu. Rev. Psy- chol. 44:585–612

Lupia A. 2002. Deliberation disconnected:

what it takes to improve civic competence.

Law Contemp. Probl. 65:133–50

Lupia A, McCubbins M, Popkin S, eds. 2000.

Elements of Reason: Cognition, Choice, and the Bounds of Rationality. Cambridge, UK:

Cambridge Univ. Press

Luskin R, Fishkin J. 1998. Deliberative polling, public opinion, and representative democ- racy: the case of the National Issues Con- vention. Presented at Annu. Meet. Am. As- soc. Public Opin. Res., St. Louis, MO Maass A, Arcuri L. 1996. Language and stereo-

typing. In Stereotypes and Stereotyping, ed.

CN Macrae, C Stangor, M Hewstone, pp.

193–226. New York: Guilford

Maass A, Salvi D, Arcuri L, Semin GR. 1989.

Language use in intergroup contexts: the lin- guistic intergroup bias. J. Pers. Soc. Psychol.

57:981–93

Mackie G. 2002. Does deliberation change minds? Presented at Annu. Meet. Am. Polit.

Sci. Assoc., Boston

Macpherson CB. 1977. The Life and Times Annu. Rev. Polit. Sci. 2005.8:49-71. Downloaded from arjournals.annualreviews.org by University of Kanas - Lawrence & Edwards on 11/04/05. For personal use only.

David M. Ryfe

D OES D ELIBERATIVE D EMOCRACY W ORK ?

David M. Ryfe

INTRODUCTION

WHO PARTICIPATES?

WHAT DOES DELIBERATION LOOK LIKE?

THE PRODUCT OF DELIBERATIVE TALK

CONCLUSION

D ÔES D ELIBERATIVE D ÊMOCRACY W ÔRK ?