1. Dahl 1989 refers to modern pluralistic states as "polyarchies" so as not to give the false impression that any nation has reached the democratic ideal. [BACK]

2. There is no one-to-one correspondence between the models distinguished in Held 1987 and my own. Rather than analyzing every model in depth, I have simply tried to demonstrate the applicability of exit and voice to a wide range of models. [BACK]

3. For integrations of Athenian democracy into modern participatory theories, see Barber 1984; Mansbridge 1983. The Athenenian system failed by many modern measures of democracy, but it remains a vivid illustration of some democratic institutions no longer in use. [BACK]

4. For a relatively coherent and thorough description of a council system of government, see the "demarchy" outlined in Burnheim 1985. [BACK]

5. . See Barber 1984; Pateman 1970. [BACK]

6. Influential contemporary writings on deliberation include Cohen 1989, Fishkin 1991, and Gutmann and Thompson 1996. The idea of a "general will" is often traced back to Rousseau (e.g., by Mansbridge 1983). [BACK]

7. For example, the political theorist Mark Warren underscores the importance of effective collective exit: "Democratic authority can exist when an institutionalized possibility of challenge allows individuals to suspend judgment" (Warren 1996a: 57). [BACK]

8. "In framing a government which is to be administered by men over men, the great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself," The Federalist, No. 51, observes (Hamilton, Madison, and Jay [1788] 1947: 265). The Feder alist Papers are available on-line at http://www.mcs.net/~knautzr/fed/fedpa-per.html. Upon reviewing Madison’s less well-known writings, Robert Dahl found that even Madison recognized the necessity of following the majority’s will—even when it infringed upon the property rights of a minority. Madison remained worried about the tyranny of the majority, but he acknowledged that its rights to govern were fundamental (Dahl 1996: 296). [BACK]

9. Held 1987: 187. [BACK]

10. This capitalist model is a mix of Held 1987’s "competitive elitist democracy," the "rational choice" models of politics described by Hauptmann 1996, and the elitist theory of democracy presented by Schumpeter 1976. On the nature and impact of economic assumptions, see Schwartz 1986. [BACK]

11. Dahl 1989 provides a succinct description of American government as a form of "polyarchy," a term he developed to describe an actual institutional arrangement that only partially meets his criteria for the democratic process. [BACK]

12. Hirschman 1970: 2. [BACK]

13. Ibid.: 4. [BACK]

14. Ibid.: 34, 42. [BACK]

15. Ibid.: 39. [BACK]

16. Ibid.: 78, 79, 90. Also see 38–39, 77. [BACK]

17. Ibid.: 43; emphasis in original. [BACK]

18. In a later work, Albert Hirschman writes that "exit and voice were defined in my book as two contrasting responses," but he points out that in some cases exit and voice have "worked in tandem and reinforced each other" (Hirschman 1993: 175–77). In this book, I use the exit, voice, and loyalty model to describe two important response modes, and I devote relatively little attention to the trade-off between exit and voice. [BACK]

19. The most common use of Hirschman’s model is as a means of understanding behavior in economic and nonprofit organizations (see, e.g., Bender and Sloane 1998). [BACK]

20. On the expansion of economic assumptions into the social sciences and American culture, see Schwartz 1986. On the intrusion of economically inspired "public choice" theory into democratic theory, see Hauptmann 1996. [BACK]

21. Hirschman 1970: 27. [BACK]

22. Ibid.: 59–60. [BACK]

23. Ibid.: 24, 79. [BACK]

24. Ibid.: 28. [BACK]

25. Ibid.: 45–46. [BACK]

26. Ibid.: 83–84. [BACK]

27. See Hirschman 1993 on Germany; Lehman-Wilzig 1991 on Israel; Kato 1998 on Japan; Lyons, Lowery, and DeHoog 1993 on urban politics; and Sorensen 1997 on empowerment. [BACK]

28. Sorensen 1997: 558–59 [BACK]

29. Throughout this book I refer to "representatives." By this I mean popularly elected officials in relatively democratic political systems. Using Norberto Bobbio’s definition, a representative is "a person with two very specific attributes: someone who (a) enjoys the trust of the electorate by virtue of election, and so is responsible to them and cannot be dismissed; and (b) who is not directly answerable to the electorate precisely because he is called upon to safeguard the general interests of civil society and not the particular interests of any one group" (Bobbio 1987: 48). My model has more detail regarding individual candidate selection than would be necessary to examine proportional representation systems where officials are named after seats are allocated among victorious parties, although party evaluation is not entirely different from candidate evaluation. [BACK]

30. Relating this back to the notion of public trust in chapter 1, the exit, voice, and loyalty model of representation only functions properly at moderate levels of public trust. Declining public trust would not be a problem if it was already "too high" (Schudson 1998: 302). Supporting this interpretation of trust/loyalty, a study of presidential voting from 1968 to 1996 found that heightened public distrust benefited the out-party in a two-way race or the third-party in a three-way race (Hetherington 1999). [BACK]

31. This definition is taken from Gastil 1993: 24. It also has much in common with the critical group decision-making functions described and studied by

32. Mansbridge 1996: 125. On the pros and cons of emotion in argumentation, see Walton 1992. [BACK]

33. I develop this definition in more detail in Gastil 1993 and Gastil and Adam 1995. It also bears some resemblance to the definition developed by Knight and Johnson 1994. [BACK]

34. Cohen 1989: 22–23 [BACK]

35. Habermas 1979. This idea resonates throughout the literature on political persuasion and discussion. Lupia and McCubbins 1998 make only a loose connection between their work and Habermas's interest in deliberation, but their requirement that effective deliberation include procedures for "verification" is comparable to Habermas's stress on the right to "challenge validity claims." [BACK]

36. Stephen Esquith and Richard Peterson provide a useful account of how Rawls's notion of the "original position" has moved from an abstract philosophical notion to a more concrete approach to public life. In this view, "the original position describes the kind of representative citizen who would choose acceptable institutional principles for protecting and advancing freedom, equality, and reciprocity within a pluralistic world where only a limited moral consensus exists" (Esquith and Peterson 1988: 301). [BACK]

37. Rawls 1971: 358–59. [BACK]

38. See Burleson, Levine, and Samter 1984; Hirokawa, Erbert, and Hurst 1996. Though group discussion is more fruitful, on average, certain conditions facilitate optimal deliberation, and other factors can result in counterproductive discussion. I discuss those in detail in chapter 8. [BACK]

39. Rittel and Webber 1973. [BACK]

40. Dahl 1991 emphasizes this point in his criteria for the democratic process. As Mark Warren argues, "A theory of democracy should be oriented toward creating institutional environments that encourage the self-examination of preferences that brings them closer to needs" (Warren 1996b: 264–65). It is impossible to "know" objectively the enlightened interests of a person, let alone a society, but deliberation can at least aid people in arriving at more informed understandings of their interests. Warren points out that "the doctrine of revealed preferences" held by economists and rational choice theorists shows that even these theorists recognize that surface policy preferences may obscure underlying values and interests that will only "reveal" themselves after a period of reflection or deliberation. [BACK]

41. See Briand 1999: ch. 8; Warren 1992 [BACK]

42. Gutmann and Thompson 1996: 4. Briand 1999 argues along similar lines. [BACK]

43. Walzer 1991 and Putnam 1993, 1995a, and 1995b are among the most influential modern writings on civil society. [BACK]

44. The Senate recognition statistic comes from Fishkin 1995: 10. With regard to literacy, the U.S. Department of Education’s National Center for Education Statistics (NCES) conducted an extensive survey of adult literacy in 1992. The NCES estimated that roughly 21 percent of A mericans 16 and older

45. The behavioral theorist Albert Bandura (1977, 1986a) turned to the concept of self-efficacy when he found himself unable to explain why people who had learned certain behaviors failed to perform them when called upon to do so. Learning theory alone could not account for the reluctance of some individuals, who were low in self-efficacy, to respond appropriately to their environment. The cognitive psychologist Icek Ajzen (1991) abandoned the widely researched and well-supported theory of reasoned action that he had helped to develop when he found that attitudes and subjective norms failed to explain adequately the actions that people took. His substitute, the theory of planned behavior, added the concept of "perceived behavioral control," which is synonymous with self-efficacy. [BACK]

46. Such "internal" political efficacy (see Pollock 1983; Wolfsfeld 1986) is an important psychological resource and helps determine whether or not an individual will become an active citizen (Rosenstone and Hansen 1993: 79). [BACK]

47. Bandura 1986b introduced me to the concept of group efficacy. My own research on civic participation has found inconsistent evidence in support of the group efficacy concept, partly because it has proven difficult to measure compared to people’s assessments of their own abilities. See Gastil 1994b. [BACK]

48. For Albert Hirschman, who claims that loyalty develops out of perceived personal influence and argues for the rational basis of loyalty versus faith, outcome expectancy almost seems to be a part of loyalty itself. On the other hand, Hirschman clearly distinguishes outcome expectancy from loyalty, which is defined as the willingness to "trade off the certainty of exit against the uncertainties of an improvement in the deteriorated product" (1970: 77–78). It appears that Hirschman is arguing that outcome expectancy influences but does not determine one’s loyalty. [BACK]

49. I discuss this at length in Gastil 1994b. For a demonstration of outcome expectancy’s significance in politics, see Finkel, Muller, and Opp 1989. Ajzen and Fishbein 1980 demonstrates the variable’s general significance across diverse social contexts. [BACK]

50. This should be qualified somewhat. Hirschman correctly points out that it is enough for a person to believe that his or her actions or the actions of a fellow citizen with similar complaints will prove influential. "[A] customer must expect that he himself or other member-customers will be able to marshal some influence or bargaining power" (Hirschman 1970: 40–41). [BACK]

51. "System responsiveness" is also called "external efficacy," but I have avoided that term to prevent confusion with self-efficacy. Despite conceptual similarities, external efficacy is only weakly associated with trust in elected incumbents per se (see Craig, Niemi, and Silver 1990). [BACK]

52. On efficacy and expectancy, see Ajzen and Fishbein 1980. Gastil 1994b presents a more complete presentation of my view of the significance of these

53. This "psychology" phrasing is taken from Paul Weithman (1995: 317), who eloquently summarizes the definition of deliberative democracy given by Cohen 1989. [BACK]

54. Lyons et al. 1993. In adapting Hirschman’s model to study Japanese party politics, Junko Kato indirectly demonstrates the importance of civic neglect, noting that a critical variable is "willingness to act when one recognizes a problem with organizational performance" (Kato 1998: 859). Kato chooses to redefine loyalty as such willingness, but I prefer to use the original conceptualization of loyalty and view inaction as a third kind of response. [BACK]

55. Robert Dahl’s "Criterion of Enlightened Understanding" (1989: 127) addresses one of the concerns of pluralist democracy’s radical critics, such as Samuel Bowles and Herbert Gintis (1987: 127), who criticize limited conceptions of democracy for "taking preferences and interests as pregiven." In a rejoinder to criticisms, Dahl acknowledges that his criterion is "vague." In general, it means that "a process for making collective decisions that incorporate reasons based on systematic research is superior to one in which reasons are based merely on conventional wisdom" (Dahl 1991: 230). [BACK]

56. Deliberative democratic theorists are often concerned with the difficulty of judging what policies best meet the needs of the larger public (e.g., Briand 1999; Fishkin 1995; Mathews 1994). The idea of tacit knowledge as a common cause of behavioral choices comes from Giddens 1984. [BACK]

57. An alternative is to give up on deliberative democracy and place strict limits on the scope of government (see Somin 1998). In this and the following chapters, I argue that a system using citizen panels can simulate nationwide deliberation in the short run and develop the psychosocial underpinnings of actual widespread deliberation in the long run. [BACK]

58. Bessette 1994 argues for designing political institutions favorable toward official deliberation even if this means compromising other principles. For example, "sunshine laws" that expose official debate to public scrutiny may undermine officials' opportunities for honest, searching deliberation. Other advocates of deliberation, such as Joshua Cohen (1989), stress the importance of "transparent" political institutions for maintaining governmental legitimacy. My goal is to create powerful incentives for official deliberation. Whether and when such deliberation takes place is less important than the correspondence of official actions with the public’s own deliberative judgment. If some officials substitute the public’s reflective decisions for their own, individual deliberation, that would be only a minor loss. [BACK]

59. On these and other conceptions of representation, see Pitkin 1967 [BACK]

60. Bianco 1994 points out that it is not enough for representatives to deliberate and expect the public to endorse their judgments. Even if the deliberation is held in the open, it is unlikely that citizens will pay attention to its details. Those who do watch or read about the details of officials' deliberation will still remain skeptical of representatives' true motivations. "Deliberation, to have any effect at all," Bianco argues, "must be aimed at educating the public, not members of Congress…. Measures designed to increase congressional de