1. This observation appears in Rousseau’s discussion of democracy in The Social Contract, bk. 3, ch. 4; see Rousseau [1762] 1967: 70. For a detailed account of Rousseau’s political thought, see Shklar 1969. Cronin 1989 and Mansbridge 1983 also view Rousseau as a powerful influence on subsequent writings on direct democracy. [BACK]

2. O'Leary 1996: 29. [BACK]

3. Burnheim (1985). [BACK]

4. Cronin 1989: 197; initiative examples are from p. 199. [BACK]

5. September 1, 1994, survey by Yankelovich Partners for Time/CNN cited in The Public Perspective 9, 2 (February–March 1998): 45. The full question wording was, "Would you favor or oppose having a national referendum system in which all citizens voted on proposals that deal with major national issues—voting either for them or against them—before they become law?" [BACK]

6. Cronin 1989: 230. [BACK]

7. Because initiatives have the same problems as elections for public office, they might also benefit from the solutions I propose for representative government. I examine this possibility in some detail in chapter 7. [BACK]

8. Guinier 1994: 14–15. [BACK]

9. Ibid.: 15. Engstrom and Barrilleaux 1991 shows how the Sisseton-Wahpeton Sioux have used the system in South Dakota to win a seat on the school board. Probably because they have used it effectively, in exit polls the Sioux voters rated the system positively, whereas Anglo voters had more mixed reviews of cumulative voting. Still 1992 and Brockington et al. 1998 also provide evidence that cumulative voting can facilitate minority representation. [BACK]

10. This description is adapted from Erdman and Susskind 1995: 35–36. [BACK]

11. Ibid.: 4. [BACK]

12. Ibid.: 45. [BACK]

14. Amy 1993 makes a persuasive case for the use of proportional representation in the United States. Among other advantages, one undeniable benefit relative to the winner-take-all system is its elimination of gerrymandering, because no district lines need to be drawn. In my view, proportional representation is superior to the present system, but I believe it does not solve the problem of exit and comes at the cost of a voter’s direct connection to individual public representatives. On the brief history of proportional representation in Ohio, see Barber 1995. [BACK]

15. Benjamin and Malbin 1992 presents a wide range of voices on term limits, as well as the texts of term-limits ballot initiatives in Oklahoma, Colorado, California, and Washington. Kamber 1995 provides a synthetic inventory of arguments against term limits. [BACK]

16. John Hibbing and Elizabeth Theiss-Morse found that most supporters of term limits are firm in their convictions. Only 13 percent changed their views after hearing the following counterarguments: "Some people opposed to term

17. Many elected officials leave office after a decade, anyhow. Hibbing’s (1993) study of careerism in Congress found that despite the increased professionalization of that institution, on average, members serve only twelve years. On the other hand, that means that the typical representative has five successful reelections. Gerald Benjamin and Michael Malbin (1992: Appendix B) provide similar data showing the gradual turnover of most state legislators. The problem isn’t turnover so much as it is accountability (and a clear public voice as guidance) while in office. [BACK]

18. "Fundamental distrust of government is represented by the call for term limits for elected officials" (Miller and Shanks 1996: 508). [BACK]

19. Proponents make the more modest claim that the people elected under term limits would be "average" citizens who would better represent the public by virtue of having a primarily nonpolitical identity. As of yet, there is no evidence of this difference. [BACK]

20. In an experimental analysis of term limits, William Bianco (1996) makes it clear that they by no means give the public more trustworthy representatives. If citizens expect representatives to be responsive to the electorate, term limits undermine the electoral connection; yet if citizens expect representatives to be untrustworthy, term limits simply give elected officials more freedom to do as they choose. "Term limits may stiffen the backbones of elected officials," Bianco writes, "enabling them to make hard choices. The problem is that constituents often do better with a representative whose backbone is subject to their recall" (166–67). Victor Kamber (1995: 53) points out that term limits encourage potential opponents to pass up opportunities to challenge incumbents, since they can wait a few years and run in an open-seat race. Thus, a twelve-year limit on senators serving six-year terms nearly guarantees an incumbent a twelve-year tenure. [BACK]

21. Jacobson 1997: 54. [BACK]

22. Makinson 1994: 3. [BACK]

23. A move toward "video direct mail" through cable television is anticipated by Abramson, Arterton, and Orren 1988: 113–15. This and the Internet could both facilitate "narrowcasting" to voters. [BACK]

24. The Gallup survey was conducted October 29–30, 1996, with 714 registered voters. This result and similar findings have been assembled by Public Campaign at http://www.publicampaign.org/pollsumm.html. [BACK]

25. Gallup poll, January 31–February 2, 1997, cited in The Public Perspec tive 9, 2 (February–March 1998): 47. [BACK]

26. Thomas E. Mann and Norman J. Ornstein, "Credibility for a Collapsed System," Washington Post, December 16, 1996. [BACK]

27. The study was conducted by the Committee for the Study of the American Electorate, and its findings are available at http://tap.epn.org/csae/spendlim.html.

28. See League of Women Voters, "5 Ideas for Practical Campaign Reform," July 21, 1997, available at http://www.lwv.org. [BACK]

29. Budesheim, Houston, and DePaola 1996 found that issue-oriented attacks on an opponent work well with an audience in the advertiser’s in-group, and character-based attacks work well within an advertiser’s out-group. The ingroup identifies with the advertiser’s ideological similarity, whereas the outgroup sees no such similarity and responds best to attacks with no issue relevance. [BACK]

30. Ansolabehere et al. 1997 [BACK]

31. On the Buchanan ad, see Cappella and Jamieson 1997; on the gubernatorial race, see Pfau and Louden 1994. [BACK]

32. Associated Press, "Judge Blocks Taft TV Ad," Elyria Chronicle-Telegram, October 11, 1998, C1. [BACK]

33. Information on the Project is available at http://www.campaignconduct.org. This website has copies of the Maine and Washington codes of conduct. [BACK]

34. This argument was offered to me by Estus Smith of the Kettering Foundation. [BACK]

35. The quotation comes from an editorial written by Anne Smead, the chair of the League of Women Voters Ohio Education Fund. The League partnered with many other organizations to sponsor the Ohio Project on Campaign Conduct. See "A Proposal to Stop Negative Campaigns," Cleveland Plain Dealer, May 8, 1998. [BACK]

36. After the election 31 percent percent of all residents surveyed reported reading the guide, and 34 percent of those who reported voting said that they had read it. See the University of New Mexico Institute for Public Policy’s Quar terly Profile of New Mexico Citizens 9 (Winter 1997): 3. [BACK]

37. Surprisingly, the national office of the the League of Women Voters knew of no formal research on the use of the voters' guides, as I learned in a November 9, 1998, telephone interview. It did, however, lead me to contact a league member in Maryland, who found that direct mail distribution of voters' guides slightly increased voter turnout in traditionally low-turnout precincts (Watkins 1998). [BACK]

38. November 13, 1998, interview with Cathy Mitchell, the Initiative and Ballot Pamphlet Coordinator for the California Secretary of State. The use figure comes from Magleby 1984. [BACK]

39. For a comprehensive guide to on-line voting guides and other political websites, see Browning 1996. On late contributions, see California’s Late Contribution Watch at http://www.calvoter.org/cvf/96gen/aboutlcw.html. [BACK]

40. Rosen 1997: 22. [BACK]

41. For a correlation of .32 to .35 between "reading news in the newspaper" and political knowledge, see Delli Carpini and Keeter 1996: 182–83. [BACK]

42. Dubois, Feeney, and Costantini 1991 [BACK]

43. Adam Nagourney and Marjorie Connelly, "In Poll, New York Voters See Senate Race as Too Negative," New York Times, October 27, 1998, A1, A23. [BACK]

44. Chaffee and Rimal 1996: 277–78. [BACK]