What is the general will?

Philosophical Review

Volumn 109, Issue 4, 2000, Pages 545-581

  Sreenivasan, Gopal ^a  

^a NATIONAL INSTITUTES OF HEALTH   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 1342298665     PISSN: 00318108     EISSN: 15581470     Source Type: Journal    
DOI: 10.1215/00318108-109-4-545     Document Type: Article

References (76)

1. The relevant constraints will be elaborated in section 4 below. As we shall see, the effect of these constraints is to ensure that the decisions in question-that is, the general will-always promote the common interests of the community.
2. Jean-Jacques Rousseau, On the Social Contract, trans. Donald A. Cress (Indianapolis: Hackett, 1987). References to this work will be given in the text by book, chapter, and paragraph number (for example, 1.6.9).
3. On the latter, see Patrick Riley, The General Will Before Rousseau (Princeton: Princeton University Press, 1986). To go by Riley's account, the importance of locating Rousseau in this longer tradition lies largely in the appreciation of certain interesting, but rather general, thematic and idiomatic connections between Rousseau's discussion and the writings of Male-branche, Bayle, Montesquieu, and others.
4. In principle, of course, there need be no trade-off between analytical acuity and interpretative breadth. This is illustrated by Joshua Cohen's admirable discussion, with which I am largely in agreement, so far as it goes; see his "Reflections on Rousseau: Autonomy and Democracy," Philosophy and Public Affairs 15 (1986): 275-97.
5. This conception of the general will finds expression, for example, in Riley's General Will Before Rousseau, 253: "Soon enough one finds Rousseau saying that the general will-the will one has as a citizen, when one thinks of the common good and of civic membership-is 'always right'.
6. Cf. John Charvet, The Social Problem in the Philosophy of Rousseau (Cambridge: Cambridge University Press, 1974), 127;
7. Andrew Levine, The General Will (Cambridge: Cambridge University Press, 1993), 19.
8. "[T]he law is merely the declaration of the general will" (3.15.8).
9. The conditions in question may, for example, instantiate an invisible hand mechanism. I revisit the significance of this point below in section 5, near note 56.
10. Notice that I do not say that the general will can be distinguished from the promotion of the common good. That would be going too far. If the majority's votes are sufficiently divorced from the promotion of the common good, Rousseau does disqualify them from declaring the general will. "Finally, when the state, on the verge of ruin, subsists only in an illusory and vain form, when the social bond of unity is broken in all hearts, when the meanest interest brazenly appropriates the sacred name of the public good, then the general will becomes mute. Everyone, guided by secret motives, no more express their opinions as citizens than if the state had never existed; and iniquitous decrees having as their sole purpose the private interest are falsely passed under the name of laws" (4.1.5). This merely underlines the point I have already granted: the general will is intimately related in some way to the common good.
11. Cf. Cohen, "Reflections on Rousseau," 277-79.
12. If an associate's true self cannot be identified with his common self, then there will be cases in which his true self does not want what his common self wants. In these cases, the associate will not be obeying his true self in obeying the general will, thereby violating Rousseau's requirement that he obey only himself.
13. Strictly speaking, all that is required is that an associate's true self comprise the whole of his common self. It does not matter if his true self comprises more than this. I shall spell this out in section 3 below. In general, however, I shall omit the qualification 'at least weakly' when speaking of the identification of an associate's true self with his common self.
14. Positive freedom contrasts with negative freedom, which can be defined as the absence of interference or constraints of various kinds. This distinction has been much discussed, but I think my simple statement will suffice for our purposes. The classic discussion is Isaiah Berlin, "Two Concepts of Liberty," in his Four Essays on Liberty (Oxford: Oxford University Press, 1969), 118-72.
15. See also Charles Taylor, "What's Wrong with Negative Liberty?" in his Philosophical Papers, vol. 2 (Cambridge: Cambridge University Press, 1985), 211-29.
16. As I define it in the text, and in Rousseau's explicit statement here (1.8.3), [positive] freedom is identified with what is often simply called "autonomy." On other occasions, however, Rousseau seems to be implicitly drawing on a stronger notion of positive freedom-stronger in the sense that it also requires conformity with what is objectively good for one (for example, 4.2.8, quoted in the text below). This stronger notion collapses the distinction between positive freedom and obedience to one's true self. Not surprisingly, perhaps, Rousseau's implicit appeals to it come in contexts where he equates freedom with obedience to the general will. I shall not pursue the question of how best to understand these different aspects of Rousseau's notion of positive freedom; but neither shall I place any special weight on the distinction between positive freedom and obedience to one's true self.
17. For an interesting discussion of this question, see Frederick Neuhouser, "Freedom, Dependence, and the General Will," Philosophical Review 102 (1993): 363-95.
18. See the helpful discussion of real interests in Raymond Geuss, The Idea of a Critical Theory (Cambridge: Cambridge University Press, 1981), chap. 2.
19. Suppose, for example, that I am very thirsty and someone hands me a large, cool drink. Unbeknownst to me the drink is poisoned. It is plausible to say that having this drink is not in my interest, despite the fact that I very much want it. Thus, sometimes wanting something is not sufficient to establish that it is in my interest. Now suppose that I suffer from a curable illness for which the standard treatment, though effective, is extremely painful. Imagine that I do not want to undergo the treatment. Here it is plausible to say that undergoing the treatment is in my interest, despite the fact that I very much do not want to do so. Thus, sometimes wanting something is not necessary to establish that it is in my interest.
20. For the impatient, there is a short alternative argument for adding this restriction. We have already seen, in note 8, that under some conditions majority voting is disqualified from declaring the general will (4.1.5). Hence, to identify the general will with the community's deliberative decisions, we need some way of filtering these decisions so as to exclude instances formed in the italicized conditions. The restriction in the text accomplishes that. Of course, this argument does not help us to specify the relevant constraints in much detail. But that is the price of impatience.
21. Previously, I said only that the community's deliberative decision "can be," or is "sometimes," sufficient to declare the general will. The restriction permits us to dispense with these qualifications.
22. Notice that two important features distinguish this assumption from the view which I said, in section 1, was undermined by even a modicum of realism, namely, that the majority, in voting, always intends to promote the common good. First, (FA) does not require an intention to promote the good; it only requires that implementing the community's decisions have promoting the good as its actual effect, however they may have been intended. Second, the content of the constraints implicit in (FA) distinguish the decisions in question from majority voting as we know it.
23. I am assuming that some of one's interests can be inconsistent with others. If that is not true, then the whole issue of subordination simply does not arise. "In fact, each individual can, as a man, have a private will contrary to or different from the general will that he has as a citizen. His private interest can speak to him in an entirely different manner than the common interest" (1.7.7; cf. 4.1.6).
24. This follows from (HC) together with the associate's commitment to taking permission by the general will as a necessary condition of his true self's wanting some course of action.
25. Notice that, insofar as individuals are subject to these requirements, Rousseau's fundamental problem doubles as a practical problem. Moreover, as we are in the midst of seeing, its practical aspect plays a critical role in securing Rousseau's theoretical solution.
26. One difficulty with this assumption is that it seems to presume that individuals have reliable beliefs about their (critical) interests, not to mention reliable beliefs about the impact of various political arrangements on the promotion of these interests. This difficulty can be relieved to some extent, since it may suffice for individuals to have reliable beliefs about what forms of deliberation can be counted on to identify their interests. But some difficulty remains, as Rousseau clearly recognized (2.6.9). I discuss this point further in sections 4 and 5.
27. In emphasizing the congruence between an individual's obeying only himself-that is, his positive freedom-and the satisfaction of his interests, I am differing with Cohen ("Reflections on Rousseau," 283-85). To some extent-how far exactly, I am not sure-this is due to our differing construals of various terms, notably 'interest'. Cf. Rousseau: "I will always try in this inquiry to bring together what right permits with what interest prescribes, so that justice and utility do not find themselves at odds with one another" (1.0.1).
28. Let us assume, for the sake of concreteness, that this common interest in healthcare is consistent with a system of private medicine only if that system is firmly circumscribed. (Everyone's interest would be satisfied, for example, by a system of only public medicine or by a mixed system of public and private medicine, where the scope of the latter was firmly circumscribed; but many people's interest would be frustrated by a system of only private medicine or by a mixed system, where the scope of private medicine was not firmly circumscribed.) Then there is some tension between O's interests. Of course, this tension might be relieved: perhaps O's personal interest can be satisfied by a firmly circumscribed system of private medicine; or perhaps O's own interest in affordable quality healthcare is guaranteed, by other facts, to be satisfied by a system of only private medicine. But, then again, the tension might not be relieved.
29. Compare Cohen: "As contractors, individuals regard themselves as free. From this standpoint, an allegiance to the common good is simply one possible allegiance among others, and lacks any special authority," "Reflections on Rousseau," 284.
30. Indeed, more generally, everyone (including O) has every reason to demand that none of his laws frustrate any of his interests unless another of his own interests makes it necessary.
31. "Thus, in order for the social compact to avoid being an empty formula, it tacitly entails the commitment-which alone can give force to the others-that whoever refuses to obey the general will will be forced to do so by the entire body" (1.7.8).
32. I shall assume that every individual in the state of nature has an interest in affordable quality healthcare. The assumption allows us to continue using the same examples as before, while varying the composition of the groups being discussed.
33. If we retain the substantive assumption made for the sake of concreteness in note 24, we would therefore have a case of inconsistency (or tension) between two common interests. Otherwise put, we would then have the same tension as before between O's interest in private medicine and his interest in affordable quality healthcare, but this would now count as a case of tension between two common interests. The substantive assumption in question here is hard to evaluate, since the situation is underdescribed. But I think we have anyhow to assume that cases of inconsistency between common interests cannot arise (cf. 4.1.1; 2.1.1). If they can, there will be serious problems for Rousseau's argument.
34. I have simplified the discussion by considering only the extreme cases in which either no one or everyone in the rest of the group shares O's interest in private medicine. But this does not prejudice anything important, since the number of those who share an interest that is not strictly common to the group only makes a difference when the number is large enough (and the interest important enough) to warrant forming a breakaway group on that basis. Even then, no difference is made unless the group does break away, in which case we thereby reach the opposite extreme (in which everyone shares the interest in question).
35. A similar treatment can be given to O's other personal interests, since the argument has been purely structural. Some of these interests will obviously be consistent with his common interests. In these cases, the issue of subordination does not arise. Cf. note 19.
36. I add 'that the law aims to satisfy' because nothing in the argument has shown that the laws must satisfy all of an associate's common interests. Each law must satisfy some interest a given associate has. Now, since this applies to every associate, each law must actually satisfy some common interest. But all this is consistent with there being some common interests that no law satisfies. The reason for this is that what ultimately drives the argument is concern for an individual's obeying himself alone rather than concern for satisfying his interests; and while the congruence between these ends is real, the latter end extends beyond the former. Cf. note 23.
37. A law, for example, that firmly circumscribes the scope of any system of private medicine.
38. As presented, the explanation's completion leaves one hostage to fortune, namely, it depends on individuals' reliably believing that the general will identifies their interests correctly (cf. note 22). I shall discuss Rousseau's attempts to satisfy this further requirement in sections 4 and 5 below.
39. As a convenience, I shall often speak of the "community's deliberation" in place of the more cumbersome "community's deliberative decision, when implemented.
40. For a will to be general, it need not always be unanimous; however, it is necessary for all the votes to be counted. Any formal exclusion is a breach of generality" (Bk. 2, n. 1). We shall discuss this point in section 5.
41. I have omitted a clause to highlight my point. The clause reads, "and that it loses its natural rectitude when it tends toward any individual, determinate object." The final sentence quoted in the text follows a sentence with multiple clauses (of which I have quoted the penultimate and omitted the ultimate). I read that final sentence as qualifying all of the clauses in the preceding sentence, and not simply its ultimate clause.
42. This is what matters because what the explanation in section 3 ultimately requires on this point is that potential associates always be able (a) reliably to believe that the general will promotes their common interest, and (b) reliably to distinguish their common interests from their personal interests, so that each can act on his commitment to subordinate the latter to the former, as well as insist that others do likewise. However, (a) carries (b) in its train, since-given (a)-the commitment to subordinate one's personal interests to one's common interests can be replaced without loss by the commitment to submit to the general will. So (a) is really all that is needed.
43. This gloss (the 'that is') is not strictly correct, since it mislocates the 'always' in moving it inside the scope of what the associates must believe. An associate who believed, for example, that, on some occasion in the past, the general will did not promote the common interest could never again believe (HC) exactly, as it says that the general will always promotes the common interest. But that is not a problem. What each associate must do is reliably believe, occasion by occasion, that the general will promotes the common interest; ideally, he will also do so on every occasion.
44. Naturally, these parts must also be appropriately related. But given how-in the context of (GW) and (FA)-each is to be secured, I do not think this raises a special problem.
45. This assumption rests on the truth of (FA): the first part of what really matters here.
46. Bernard Grofman and Scott Feld, "Rousseau's General Will: A Condorcetian Perspective," American Political Science Review 82 (1988): 567-76.
47. Cf. Brian Barry, Political Argument (London: Routledge and Kegan Paul, 1965), 292-93.
48. Grofman and Feld, "Rousseau's General Will," 570; emphases in the original.
49. On the Condorcetian interpretation, however, the citizens' identification of their common interest does proceed by means of an explicit attempt on their part to identify their common interest, presumably in order to promote it. In this respect, the Condorcetian interpretation is a species of the over-simple account of the general will criticized in section 1. I consider it here simply as an example of a unified treatment of the two stipulations laid down in 2.3.4.
50. Grofman and Feld, "Rousseau's General Will," 571; emphasis in the original.
51. Quite apart, that is, from its being a species of the over-simple account. To begin with, the assumption (stated in the quotation flagged by note 43) about the average citizen's competence is quite strong. But even waiving this, it is not clear that the content of that assumption is consistent with Rousseau's ban on partial societies. Consider the following scenario: A subset of community members have an extremely low competence-"by competence we mean the individual probabilities of making the 'correct' (dichotomous) choices" (Grofman and Feld, "Rousseau's General Will," 570). As a result, the average competence in the community falls below the required threshold of 0.5. But the average competence in the rest of the community stands above 0.5, and hence the jury theorem applies to their deliberations. (It is therefore false that "this faction is equivalent to a single voice deciding things" and that "thus the benefits of large numbers are lost completely" (ibid., 571)). So, if the votes of the low competence subset did not count, the jury theorem would also apply to the deliberations of the whole community. It seems to me that the motive behind applying Condorcet's jury theorem to Rousseau's case favors permitting (perhaps even requiring) the rest of the community to exclude the votes of the low competence subset. Yet that would be precisely to permit the operation of a partial society. In fact, in their contribution to a later discussion of their original article, Grofman and Feld effectively concede this point, by accepting that maximizing the judgmental competence of the community as a whole may require assigning "negative weights for those with competence less than .5." However, they do not seem to appreciate the significance of this concession.
52. See their "Democratic Theory and the Public Interest: Condorcet and Rousseau Revisited," American Political Science Review 83 (1989): 1332ff.
53. But when intrigues and partial associations come into being at the expense of the large association, the will of each of these associations becomes general in relation to its members and particular in relation to the state" (2.3.3).
54. The ban on partial societies, for example, is not actually an absolute ban: "If there are partial societies, their number must be multiplied and inequality among them prevented" (2.3.4).
55. That is, the general will includes any such decision. Strictly speaking, recall, the general will is the totality of unrescinded decisions made by the community when its deliberation is subject to constraints (i)-(iv).
56. It is these credentials, considered in isolation, that prompt the over-simple account of what the general will is.
57. This is also made clear by the passage (2.6.9) in which he introduces "the necessity of having a legislator.
58. I have modified Cress's translation of this clause, since it is incorrect and misleading. 4.2.9 reads thus: "Ceci suppose, il est vrai, que tous les caractères de la volonté générale sont encore dans la pluralité: quand ils cessent d'y être, quelque parti qu'on prenne il n'y a plus de liberté." Cress has 'free' where I have 'in it'.
59. I do not think it is going too far to say that (HC) describes a characteristic effect of the general will: that it always promotes the common interest. This effect is characteristic because it is in virtue of having it that the general will fulfills the role Rousseau assigns it within his theory, namely, to preserve each associate's obedience to himself alone. The truth of (FA) is the condition on which this characteristic effect always resides in the relevant majority.
60. That is why I described (GW') as a provisional summary.
61. The closest he comes is 4.1.6, where he implies that this is normally how associates will vote: "Thus, instead of saying through his vote it is advantageous to the state, he says it is advantageous to this man or that party that this or that view should pass." But that is still not to say that their so voting is a necessary condition of declaring the general will.
62. The answer is not obvious. If, for example, the deliberative context contained some kind of invisible hand mechanism, whereby the members' voting simply in terms of their own interests led to the promotion of their common interests (cf. note 58), the answer might well be negative. Indeed, the context of the passage in which Rousseau permits members to vote in terms of their own happiness is consistent with the suggestion that he had some such mechanism in mind (2.4.5, quoted near note 7 above).
63. Cf. Hilail Gildin, Rousseau's Social Contract (Chicago: University of Chicago Press, 1983), 59-61.
64. John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971);
65. Jiirgen Habermas, Moral Consciousness and Communicative Action, trans. C. Lenhardt and S. Nicholsen (Cambridge: MIT Press, 1990), 43-194.
66. Rawls, for example, includes Rousseau's constraints (i) and (ii) among his five "formal constraints of the concept of right" (Theory of Justice, 130-36; cf. 146-47). His "original position," moreover, is supposed to incorporate an invisible hand mechanism: "the combination of mutual disinterest and the veil of ignorance achieves the same purpose as benevolence" (148). Habermas follows Rousseau (3.15.3), as Rawls does not, in requiring that the deliberations actually be conducted by real life individuals (Moral Consciousness, 94 and 122). However, Habermas does allow for representation.
67. See, for example, Jiirgen Habermas, Knowledge and Human Interests, trans. J. Shapiro (Boston: Beacon Press, 1971).
68. Cf. Geuss, The Idea of a Critical Theory, chap. 1.
69. "When, therefore, the opinion contrary to mine prevails, this proves merely that I was in error, and that what I took to be the general will was not so" (4.2.8). In adducing this passage, I am relying on there being some connection between the general will and the common interest-for example, (HC). It is possible that 4.2.9 (quoted between notes 51 and 53) is meant to qualify the claim quoted here, but I think that qualification has more to do with freedom than with correct judgments of interest.
70. "Whatever is good and in conformity with order is such by the nature of things and independently of human conventions" (2.6.2).
71. This is better because it avoids the needless paradox of asking why the associate does not believe and disbelieve the same thing at the same time. If an associate votes against some bill, we may suppose that he does not believe that that law is required to promote his common interests. The question is whether, short of changing his vote, there is some sense in which he might, without contradiction, nevertheless accept that that law is so required, for the purpose of crediting his true self with wanting to obey it. See also note 63 below.
72. It may be helpful to expand a little bit on this subjective condition. In discussing the general will, I suggested two senses in which one might accept that a decision by the community promotes one's interest, namely, believing it and accepting that one is mistaken not to believe it. But these do not exhaust the range of senses available.
73. Suppose, for example, that X understands herself as identified with her political community, and therefore as bound to obey its decisions, even those she believes do not promote her interest. X's self-understanding nevertheless depends, let us also suppose, on her believing that the community's decisions do, on the whole, promote her interests and that her interests are taken as seriously as those of everyone else. Now say X believes that some community decision does not promote her interest. As long as this belief does not move X to revise her self-understanding, the obligation to obey the decision is one that X will accept, and accept on the basis of understanding herself as identified with the community. I think it is reasonable, under these circumstances, to count X's obeying this decision as a case of obeying herself-at least if we suspend concern about whether the objective condition is also satisfied.
74. Hence X's self-understanding, together with the fact that X omits to revise it in the light of the decision at hand, suffices for her to satisfy the subjective condition. (I have adapted this example from Charles Taylor's instructive discussion in "Liberal Politics and the Public Sphere," in his Philosophical Arguments (Cambridge: Harvard University Press, 1995), 276-78.)
75. Naturally, one may also wish to weaken the objective condition in a symmetrical fashion-so that it only requires, for example, that the community's decisions do, on the whole, promote X's interests.
76. For an illustration of what may be gained from taking Rousseau's insight seriously, see my "Judicial Review and Individual Self-Rule," Revista Argentina de Teoría Jurídica 2 (2001), forthcoming. Available online at (http://www.utdt.edu/departamentos/derecho/publicaciones/rtjl/ primeraspaginas/index.htm).