1. See, e.g., Russell, Bertrand, A History of Western Philosophy (Oxford: Oxford University Press, 1945).Google Scholar

2. Schumpeter, Joseph, Capitalism, Socialism, and Democracy (New York: Harper and Row, 1942)Google Scholar; Berelson, Bernard, “Democratic Theory and Public Opinion,” Public Opinion Quarterly 16 (Fall 1952): 313–30.CrossRefGoogle Scholar

3. Riker, William, Liberalism Against Populism (San Francisco: W. H. Freeman, 1982).Google Scholar

4. Ibid., p. 238.

5. For a discussion of this concept in social choice terms, see Young, H. P., “Condorcet's Theory of Voting,” American Political Science Review 82 (1982): 1230–44.Google Scholar

6. Berlin, Isaiah, Two Concepts of Liberty (Oxford: Oxford University Press, 1958).Google Scholar

7. The General Will expresses the community's opinion (which is definitionally correct) about this collective good. Given that each individual is assumed to want what is best for the community (as opposed to what is best for the self), then if the results of voting are contrary to one's preferences, then those preferences are simply wrong. In other words, a citizen may want (“will”) what is best for the society, but be mistaken about the substance of that desire, so that they may not be aware of their own will. Hence, they must be “forced” to follow the collective will, which is truly their own.

8. Kendall, Willmore, John Locke and the Doctrine of Majority Rule (Urbana, IL: University of Illinois Press, 1941).Google Scholar

9. It might be argued that the General Will cannot be reduced to majority rule or voting games in general. However, it must be understood that the aggregation problem applies to any collective preference (or judgment) that depends upon the preferences (or judgments) of individuals; see note 23 below. Thus, the cycling issue applies whenever the wishes, preferences, interests, judgments, values, or opinions of citizens are thought to define, measure, or discover the General Will. Accordingly, even if the General Will somehow emerges without resort to voting, the fact remains that as long as it is in any way a function of individuals, cycling is a problem.

10. See Kelly, Jerry, Arrow Impossibility Theorems (New York: Academic Press, 1978), for a review of the many variants of the general impossibility result.Google Scholar

11. Riker, Liberalism Against Populism; for similar sentiments, see also Nelson, David, The Justification of Democracy (New York: Holt, 1980)Google Scholar; and Dahl, Robert, A Preface to Democratic Theory (Chicago: University of Chicago Press, 1956), chap. 2.Google Scholar

12. The difficulty with arbitrary outcomes is that individuals may be compelled to accept decisions which are not functions of the General Will, which implies that they are forced to accept laws which they have not assented to. However, if all social decisions are made (passively or actually) by unanimity, this problem cannot manifest itself in that arbitrary outcomes cannot obtain under unanimity.

13. Carmines, Edward and Stimson, James, “Two Faces of Issue Voting,” American Political Science Review 74 (1980): 78–91.CrossRefGoogle Scholar

14. Downs, Anthony, An Economic Theory of Democracy (New York: Harper a Row, 1957).Google Scholar

15. Schattschneider, E. E., The Semi-Sovereign People (New York: Holt, 1960).Google Scholar

16. The a priori probability of cycling increases as the number of alternatives increases. If the set of available alternatives is kept reasonably small, then the likelihood of intransitivities may be (relatively) low.

17. Riker, Liberalism Against Populism.

18. Barry, Brian and Hardin, Russell, Rational Man and Irrational Society? (Beverly Hills, CA: Sage, 1982).Google Scholar

19. Mckelvey, Richard, “Intransitivities in Multi-dimensional Voting Models and Some Implications for Agenda Control,” Journal of Economic Theory 12 (1976).CrossRefGoogle Scholar

20. More recent findings in the literature actually suggest that the theoretical instability of voting games may be radically less pernicious than this result suggests. In particular, McKelvey's work on the “uncovered set” suggests that outcomes may be bound to a small set of points in the issue space such that all outcomes are majority preferred to all points outside the set (Mckelvey, Richard, “Covering, Dominance, and Institution Free Properties in Social Choice,” American Journal of Political Science 30 [1986]: 282–314.CrossRefGoogle Scholar). If so, then cycling ma be virtually irrelevant, in that it will be contained to a very small area of the choice set.

21. Schofield, Norman, Grofman, Bernard, and Feld, Scott, “The Core and Stability of Group Choice in Spatial Voting Games,” American Political Science Review 82 (1988): 195–212.CrossRefGoogle Scholar

22. Institutions predicated upon the General Will may not be widely affected by cycling for two additional reasons. First, it may be that atomistic preferences evolve into consensus through a process of social accommodation, given that consensus itself may be a valued commodity. In other words, if individuals value agreement, their preferences may change so as to produce consensus (see Mates, Elaine, “Paradox Lost — Majority Rule Regained,” Ethics 84 [1978]: 282–314).Google Scholar This may be particularly applicable to the two-stage process suggested above, in that unanimity becomes a necessary part of the decision procedure. The second argumentative road would appeal to the literature on structurally induced equilibria. The literature suggests that organizational structure may eliminate cycles or confine them (even when theoretically capable of including all alternatives) to a relatively small portion of the choice set (see, for example, Shepsle, Kenneth, “Institutional Arrangements and Equilibria in Voting Models,” American Journal of Political Science 23 [1979]: 27–59).CrossRefGoogle Scholar Thus the theoretical instability of politica games may not manifest itself to the degree (or with the potentially dire consequences) one might otherwise expect.

23. Coleman, James and Ferejohn, John, “Democracy and Social Choice,” Ethics 97 (1986): 6–25CrossRefGoogle Scholar, suggest that it is possible to avoid much of the difficulty flowing from the voter's paradox by assuming that the General Will is not defined by the results of voting. In those situations where a cycle has obtained, it is not necessary to suggest that the General Will is incoherent, but rather that voting has simply failed to reflect it. A similar argument is made by Grofman, Bernard and Feld, Scott, “Rousseau's General Will: A Condorcetian Perspective,” American Political Science Review 82 (1988): 567–78CrossRefGoogle Scholar, in their application of Condorcet's jury theorems to Rousseau. There are two serious problems with this approach. One is that cyclically produced outcomes will be erroneously accepted as indicative of the General Will (and thus the populace will be falsely obliged to accept that result). The second is that the argument implicitly assumes that the voting mechanism can be incoherent, but the General Will (which is not defined in terms of voting games) cannot. This view fails to realize that the aggregation problem applies not just to voting but to any attempt — metaphysical or actual —to derive a collective judgment from individual judgments. Simply put, if the General Will is a function of individual judgments or preferences, then the aggregation problem applies to the concept itself, not just to attempts to measure it.

24. Riker, Liberalism Against Populism.

25. Riker, William, “The Paradox of Voting and Congressional Rules for Voting on Amendments,” American Political Science Review 82 (1988): 349–66.Google Scholar