1. When used without further qualification, "majority rule" in this paper always means "simple majority rule".

2. A similar story for Norway is told in F. Sejersted, "Democracy and the Rule of Law", in J. Elster and R. Slagstad, eds., Constitutionalism and Democracy (Cambridge, England: Cambridge University Press, 1988), 131-52.

3. References to the American proceedings will be cited from the three volumes of M. Farrand, ed., Records of the Federal Convention (New Haven, Conn.: Yale University Press, 1966). References to the French proceedings will be cited from volumes 8 through 30 of the Archives Parlementaires. Série I: 1789-1799 (Paris: 1875-1888).

4. A good exposition is A. Sen, Collective Choice and Social Welfare (San Francisco: Freeman, 1970), chaps. 5 and 5*.

5. Here I draw on B. Barry, "Is Democracy Special?", reprinted in his Democracy, Power and Justice (New York: Oxford University Press, 1989), 24-60. For the idea of focal point, see T. C. Schelling, The strategy of Conflict (Cambridge, Mass.: Harvard University Press, 1960), chap. 2.

6. Karl Marx, The Eighteenth Brumaire, in Marx and Engels, Collected Works (New York: International Publishers, 1979), vol. 11, 166.

7. Barry, "Is Democracy Special?", 56. The same idea underlies Tocqueville's Democracy in America.

8. For a survey, see E. Spitz, Majority Rule (Chatham, N.J.: Chatham House Publishers, 1984), chap. 8 and passim.

9. See, for instance, the essays in R. G. Frey, ed., Utility and Rights (Oxford: Blackwell, 1985).

10. For the idea of rights as trumps, see notably R. Dworkin, Taking Rights Seriously (London: Duckworth, 1977).

11. One might try to avoid the regress by fixed-point reasoning. Consider the proposal that x percent of the votes (x "e 50) is sufficient for a law to be passed. The percentage of members of the assembly who agree with the proposal is some decreasing function f(x). There must then exist some self-sustaining percentage of x* such that f(x*) = x*. This is the percentage that should be adopted by the assembly. However, this procedure must itself be voted by simple majority.

12. Mounier, Archives, 8:555; Mirabeau, ibid., 8:538; Clermont-Tonnerre, ibid., 8:574; Robespierre, ibid., 26:124.

13. J. M. Thomson, Robespierre (Oxford: Blackwell, 1988), 134ff.

14. F. Furet, La révolution 1770-1870 (Paris: Hachette, 1988), 104.

15. L. Fuller, The Morality of Law (New Haven, Conn.: Yale University Press, 1969), 79-81.

16. In fact, even when the government cannot choose the date of the election, it can still manipulate the economic conjectures so as to make them more favorable. To overcome this problem, one could time elections randomly, as proposed by A. Lindbeck, "Stabilization Policy in Open Economies with Endogenous Politicians", American Economic Review: Papers and Proceedings 66 (1976): 1-19. I note as a curiosum that in the debates of the Assemblée Constituante the proposal was made (30: 97) to have periodical constitutional conventions called in a quasi-random manner, namely, by linking them to the death of the monarch. However, in these cases, as in many others, the advantages of randomization are easily seen to be offset by its drawbacks. See generally chap. 2 of my Solomonic Judgements (Cambridge, England: Cambridge University Press, 1989). An important exeption is randomization in the judiciary, as further discussed below.

17. Madison. Records, 1:135.

18. For an analysis of the distinction between interest and passion among the American founders, see M. White, Philosophy, The Federalist, and the Constitution (New York: Oxford University Press, 1987), chap. 7. For general discussion, see S. Holmes, "The Secret History of Self-Interest", in J. Mansbridge, ed., Beyond Self-Interest (Chicago: University of Chicago Press, 1990), 267-86, and my Sadder but Wiser? Studies in Rationality and the Emotions, forthcoming from Cambridge University Press.

19. Madison, Records, 2:123.

20. Bergasse, Archives, 9:116. Strictly speaking, this example is out of place. It shows how certain solutions to the problem of majority rule might fail to work, rather than instantiating that problem itself.

21. When such short-sighted behaviour is induced by passion rather than representing the permanent subjective interest of the actor, it belongs to the next category. For quotations in this paragraph: Randolph, Records, 1:51; Randolph, ibid., 1:59; Hamilton, ibid., 1:289; Madison, ibid., 1:421; Madison, ibid., 1:430; Morris, ibid., 1:512; Lally-Tollendal, Archives, 8:516; Grégoire, ibid., 8:567; Sieyès, ibid., 8:597; Robespierre, ibid., 9:81.

22. See, however, Madison's letter to Jefferson of October 24, 1787, for a discussion of religious sects as a threat to the freedom of worship.

23. See also chap. 6 of my The Cement of Society (Cambridge, England: Cambridge University Press, 1989).

24. See, for instance, G. Wood, The Creation of the American Republic (New York: Norton, 1972), 188-96.

25. See notably Sieyès, Archives, 8:595.

26. Spitz, Majority Rule, 149-52.

27. My mother tells me that when she was a girl, she and her friends were obsessed by the following hypothetical choice: given that you could only have two out of the three properties of intelligence, kindness, and beauty, which two would you choice? Similarly, the discussion in the text suggests that democracies can only have two out of the following three virtues: enabling deliberation, limiting passionate popular majorities, and limiting self-interested legislative majorities.

28. V. Bogdanov, "Britain: The Political Constitution", in V. Bogdanov, ed., Constitutions in Democratic Politics (London: Gower, 1988), 53-72. It has been argued, however, that the violations of the unwritten British constitution under Mrs. Tatcher have created the need for a formal bill of rights (L. Siedentop, "Tatcherism and the Constitution", Times Literary Supplement, January 26, 1990).

29. M. J. C. Vile, Constitutionalism and the Separation of Powers (New York: Oxford University Press, 1967), 189.

30. G. Vedel, "The Development of the Conseil Constitutionnel", forthcoming in E. Smith, ed., Constitutional Justice under Old Constitutions.

31. For applications of the motive-opportunity distinction to political affairs see chaps. 9 and 10 of White, Philosophy, The Federalist, and the Constitution, and chap. 4 of my Political Psychology (Cambridge, England: Cambridge University Press, 1993).

32. A similar practice obtains in Great Britain: "Under the Parliament Act of 1911, as amended by the Parliament Act of 1949, a non-money bill can be passed into law over the opposition of the House of Lords if it has been passed by simple majority in two consecutive sessions of the House of Commons and one year has elapsed between the second reading of the Bill in the Commons in the first sessions and its third reading in the Commons in the second session". (J. Jaconelli, "Majority Rule and Special Majorities", Public Law [1989] 587-616, at 597.)

33. J. N. Eule, "Temporal Limits on the Legislative Mandate", American Bar Foundation Research Journal [1987] 379-459, at 394. Eule goes on to say, however, that "even in such a system . . . there remain moral and political restraints on the legislative alteration of constitutional doctrine".

34. John Potter Stockton in debates over the Ku Klux Klan Act of 1871, as cited in J. E. Finn, Constitutions in Crisis (New York: Oxford University Press, 1991), 5. On the general theme of self-binding, see chap. 2 of my Ulysses and the Sirens, rev. ed. (Cambridge, England: Cambridge University Press, 1984) and S. Holmes, The Paradox of Democracy, Forthcoming from University of Chicago Press. On the theme of constitutional self-binding, see my "Intertemporal Choice and Political Theory", in G. Loewenstein and J. Elster, eds., Choice over Time, (New York: Russell Sage Foundation, 1992). For a discussion of the putative paradoxes involved in self-binding, see P. Suber, The Paradox of Self-Amendment (New York: Peter Lang, 1990).

35. I simplify. The use of qualified majorities can also act as a restraint on momentary passions, because with a larger proportion of people required for the decision, the chances are better that not all of them will be caught up in the collective frenzy. Unamendable clauses, in particular, offer a perfect protection against impulsive rashness.

36. Tocqueville warned against the excessively stringent amendment procedures proposed for the French 1848 constitution. "I have long thought that, instead of trying to make our forms of government eternal, we should pay attention to making methodical change an easy matter. All things considered, I find that less dangerous than the opposite alternative. I thought one should treat the French people like those lunatics whom one is careful not to bind lest they become infuriated by the constraint". See A. de Tocqueville, Recollections: The French Revolution of 1848 (New Brunswick, N.J.: Transaction Bookks, 1990), 181. The implication, whether intended or not, is that by making change easier one reduces the desire for change. A similar motive was adduced by the government of East Germany in 1989: by making it clear that everybody could leave for the West, they hoped to ensure that nobody would want to do so. Also, it has been argued that the possibility of divorce makes marriages more stable rather than less: "We thought we were trying our marriage-knots more tightly by removing all means of undoing them; but the tighter we pulled the knot of constraint the looser and slacker became the knot of our will and affection. In Rome, on the contrary, what made marriages honoured and secure for so long a period was freedom to break them at will". (Montaigne, Essays 2:15.)

37. Mason, Records, 2:309.

38. See notably G. Ainslie, Picoeconomics (Cambridge, England: Cambridge University Press, 1992).