THE QUALITY OF CITIZEN JUDGMENTS
For many reasons, the citizen panels I propose would make policy judgments of varied quality. Sometimes the panels would see an issue clearly, connect the policy options to their underlying values, and reach a decision in sync with the public's most enlightened understanding of its own interests. At other times, however, the panels might make decisions based on imperfect reasoning, low self-awareness, and possibly even faulty information. In those cases, the citizen panels would write flawed rationales for their policy decisions, and their decisions would often be contrary to the public's best interests.[1] No process can remove the possibility of such mistakes, but I have designed citizen panels to increase the likelihood of deliberation and sound judgment.[2]
The psychologist Scott Tindale raises concerns about the quality of deliberation in programs like the 1996 National Issues Convention. "Small groups," Tindale explains, "can sometimes lead to unexpected and occasionally problematic outcomes." In the worst groups, participants will polarize into extreme and possibly irrational points of view, and the majority will wear down the minority through pressure to conform and its superior numbers, which produce more speaking time and arguments for the majority position.[3] Separating out these problems, there are four hazards that merit examination: a flawed or unduly limited information base; low motivation to deliberate thoroughly; excessive social pressure to adopt the majority position; and a countervailing possibility of internal divisions that prevent the discovery of common ground.
INFORMATION
Insufficient or flawed information can result in suboptimal group decisions.[4] Groups that convene to discuss an issue can make outrageous statements and reach ridiculous conclusions because the group participants lack necessary information. In a group with a poor knowledge base, a single idea or assertion can hold sway, especially if there is no procedure for testing participants' beliefs against some external information source. For that reason, all of the deliberation programs presented in chapter 6 have some mechanism for infusing basic facts and important arguments into group discussions.
The citizen panels proposed in chapter 7 would be no different: participants would gain relevant information from informal discussions
MOTIVATION
Good information is useless, however, if panelists choose to ignore it. Citizen panels will fail if the participants lack sufficient motivation to deliberate. As explained in chapter 2, deliberation is hard work. Even with its alternative choices predeflned, a panel must carefully examine a problem, establish a set of criteria for judging alternatives, and then weigh the pros and cons. Only a highly motivated group can simultaneously sustain high levels of attention, critical scrutiny, and self-awareness over a period of days. If the panels cannot sustain a sufficient level of motivation, their final judgments are likely to suffer some of the same weaknesses as superflcial survey responses.[8]
The sources of motivation for citizen panels are straightforward. First, participants would be paid a considerable honorarium to demonstrate that panel organizers respect the time and effort required to participate. In past random sample forums, participants have appreciated this gesture, and citizen panelists would receive even greater compensation for their efforts. Second, panelists would be asked to perform for only a limited number of hours over a limited number of days. No citizen could sustain a high level of deliberation indeflnitely, but a clear deadline and a structured series of activities would make it possible for most participants to remain alert and constructive throughout their deliberations. Third, since some panelists might tend to become distracted and produce uncritical judgments if left to their own devices, the panel organizers would be present throughout the event. Motivation can be contagious, and with large panels, there would always be someone to
PRESSURE TO CONFORM
Panelists might push each other to maintain a high level of motivation, but the panels are also designed to reduce the pressure to conform to a particular group viewpoint. Groups are more likely to arrive at a premature and low-quality consensus if (a) the discussion concerns an ambiguous object or issue, (b) only one group member holds the minority view, (c) the group members are required to present their final judgment in public, and (d) the group is close-knit.[10]
First, the ambiguity of the panelists' discussion topics would be inevitable, for truly political dilemmas never have straightforward solutions. If there were an objective standpoint from which one could judge public decisions, society would have no need for politics. Psychological research suggests that this inherent ambiguity can increase pressure to conform, because of the difficulty of knowing the accuracy or appropriateness of one's own views.[11] To alleviate this pressure somewhat, the citizen panel process would present participants with clear task requirements and focus their discussion on relatively narrow issues, such as a single piece of legislation or one set of candidates.
Second, using citizen panels with flfty or more members reduces the likelihood that lone individuals will favor the minority position. Past research on small group behavior shows that even the encouragement of just one fellow dissenter can bolster the confldence and independence of a minority group member, who will then continue to present arguments in favor of his or her minority viewpoint. For this reason, minority influence is common in jury deliberation. In one study (Hastie, Penrod, and Pennington 1983), the final jury verdict did not correspond to the original judgment of a majority "verdict-favoring faction" on thirteen out of sixty-nine mock juries. Four wound up as hung juries, and nine reached agreement on a verdict opposite the initial majority preference.
Third, the use of secret ballots in citizen panel discussions would alleviate the pressure produced by the public presentation of a minority viewpoint. For this same reason, panel moderators would also prohibit public straw polling and encourage panelists to refrain from publicly stating early verdicts. In the same jury study cited above, Hastie found evidence that the postponement of preliminary judgment can aid deliberation in just this manner: "Juries that deliberate with an evidence-driven style, starting deliberation with a discussion of evidence rather than law and deferring formal voting until later in deliberation" tend to produce "more thorough and impartial assessments" and a better "inte gration of the evidence.[13]
The fourth source of group conformity pressure is high group cohesion. The political psychologist Irving Janis flngered cohesion as the culprit in Groupthink.[14] Countless scholars, organizational consultants, public flgures, and citizens have accused cohesive decision-making bodies of engaging in groupthink, as if the phenomenon were a common and almost inevitable result of group decision making. On the contrary, Janis found that groupthink only arose under particular circumstances and that many groups escaped it through well-crafted procedures. According to Janis, a group is more likely to engage in groupthink when it is cohesive, flawed in its organizational design (i.e., a homogenous membership isolated from other people and no neutral leader and procedures), lacking in self-esteem, and under extreme external pressure.
To protect against groupthink, the citizen panels would have a diverse membership, neutral procedures, and a professional moderator. The most important procedure regarding groupthink might be exposing the panel's deliberation to outside scrutiny on the third day of deliberation. In addition, whenever a panel's deliberations appeared to converge on a consensus ahead of schedule, the panel moderator could randomly assign a citizen to play the role of devil's advocate to spark critical examination of the panel's emerging decision.[15] Some citizen panelists may lack self-esteem and feel outside pressure, but not to the degree that Janis observed in his case studies of groupthink. In addition, the panelists are likely to form cohesive bonds, as have the participants at past random sample forums; however, social psychological and small group communication research has found that there is no simple correspondence
Differences among participants in communications skills and style are also relevant to conformity. Individuals vary in their abilities to express themselves and process new information rapidly, and they also have different styles of listening and speaking. Were these variances randomly distributed across the population, they would reduce the quality of deliberation but produce no pressure toward conforming to a particular viewpoint. Eloquence would come from all directions. Some critics of deliberation, however, argue that "status inequalities and regular patterns of social oppression might intrude when Americans deliberate." In this view, males, whites, and more highly educated participants tend to dominate in juries and classroom discussions, so groups might conform to their views by force of social power more than argument.[17]
Evidence supporting this view comes from jury trials, however, and citizen panels differ from juries in many important respects. Jury discussions are shielded from public view, have a relatively loose structure, and do not use an outside moderator. Each of these features makes it easier for dominant social groups to wield power. Citizen panel deliberations are subject to external review and comment, so they lack the insulation that facilitates unjustifled judgments. Because panel deliberation has guidelines and structured activities promoting debate, it is harder to silence opposition. Like all random sample forums, citizen panels also employ a moderator, who reinforces discussion rules and encourages broad participation.[18]
Mitigating factors such as these are important considerations when considering the power of social status on group conformity and bias. Careful study of group interaction has found that the impact of social status varies tremendously across different groups. For example, a careful meta-analysis of seventy-five studies of group leadership found that males emerged as leaders more often than women, but this effect was reduced if groups met outside a laboratory, undertook a gender-neutral task, or had complex social relations among participants. If groups had more than one meeting, the effect of gender on leadership emergence all but disappeared.[19] In sum, social status differences in communication styles and skills are real, but carefully designed group activities can diminish—or eliminate—their potential to force group conformity to a particular viewpoint.
FINDING COMMON GROUND
While guarding against conformity, the citizen panels will need to avoid the personal conflicts and internal schisms that can polarize groups into extreme factions and create a deadlock. This problem is especially important in the case of political deliberation, because American politicians, the media, and citizens share a cultural tradition of adversarial democracy.[20] Interest groups tenaciously defend their positions in public debates, and giving ground is more a sign of weakness than a virtuous gesture. Candidates for public office—and the parties backing them—create a climate of hostile conflict through irrelevant personal attacks, exaggerated or misleading policy contrasts, and empty but evocative symbols.[21] The media plays its part by giving undue attention to posturing and demagoguery, as opposed to more respectful dialogue. By framing politics as the competitive pursuit of self-interest, the media feed into a spiral of political cynicism.[22] From the right come such soothing pet names as "baby killer," "feminazi," and "tree hugger," and the left replies with "woman hater," "Bible thumper," and "greedy capitalist," among others.[23]
How could deliberation and consensus-building take place in such a climate? Citizen panels would engage in meaningful discourse and careful listening because of the nature of their participants. Though immersed in a culture of political conflict, most citizens show a willingness to speak and listen respectfully in face-to-face interactions with strangers. Personally, I have found that even at the most bloodthirsty public rallies, I can strike up a reasoned conversation with an apparent adversary if we can stand just beyond the crowd to hear each other speak. We usually come away understanding one another much better and even agreeing on certain issues. When citizens step into the far more hospitable setting of a random sample forum, such as the National Issues Convention described in chapter 6, they become even more likely to play the sober role of juror. As Roderick Hart and Sharon Jarvis concluded after studying transcripts from the convention, participants "showed themselves respectful of—but not cowed by—the issues of the day and respectful of—and not cowed by—their own heterogeneity. They used a variety of strategies to build bridges to one another.[24]
Despite participants' varied viewpoints, random sample forums usually move toward consensus because participants come together as a group to compensate for the unfamiliarity of their setting. They develop a thin but shared group identity around the very structure and official norms of the jury or public forum, which is all they have in common.
In sum, citizen panels can be expected to reach reasonable decisions, given an adequate information base and group procedures that mitigate against both conformity and internal polarization. In some respects, the citizen panel design is more elaborate than those used in criminal and civil juries, which deliberate in private without any procedural structure. Despite their lack of safeguards against faulty information-processing and social pressure, conventional juries perform quite well. As Valerie Hans and Neil Vidmar conclude in Judging the Jury (1986): "The data from hundreds of studies of jury trials and jury simulations suggest that actual incompetence is a rare phenomenon. Juries do differ sometimes from the way judges would have decided, but it is on grounds other than incompetence."[27] In a similar manner, one could expect citizen panels to deviate from some conventional political judgments, but that is part of their purpose—to reveal the judgments that average citizens reach when provided with roughly the same information base as policy experts and public officials.