Rousseau's political thought remains a central reference point for democratic theory. Participatory democrats such as Barber (1984) and Pateman (1970) appeal to Rousseau's ideas about democratic legitimacy and the supposed transformative effects of participation, arguing that representative and liberal political institutions that determine the legislative agenda undermine democratic self-government. Critics of strong democracy, in turn, target Rousseau as the primary exponent of a theoretically and practically untenable populism. Within social choice theory, for example, Riker (1982) famously argues that Rousseau's theory of the “general will” falls prey to paradoxes of voting that make trying to identify the common good through voting meaningless, and suggests that agenda-setting is a pervasive feature of political and social organizations. This debate over democratic legitimacy and institutions touches on a broad spectrum of issues in contemporary democratic theory and political science more generally.
In his recent article “Rousseau on Agenda-Setting and Majority Rule,” Ethan Putterman (2003) examines through an analysis of Rousseau's political theory how the democratic principle of popular majority rule might be reconciled with agenda-setting by legislative experts. Putterman argues that Rousseau accomplishes this reconciliation through a novel separation of powers between the legislative and the executive powers where the sovereign people delegates the exclusive power to initiate laws to the executive, or “government.”
Putterman's attention to the institutional features and actual operation of the state outlined in The Social Contract is a welcome departure from most interpretations. Too often scholars view Rousseau's treatise on the “principles of political right” as mere principles, yet fully half of the work is devoted to the art of government. Rousseau's theory of the separation of powers is “perhaps the most original aspect of his political teaching” (Masters 1968, 335), but it is widely overlooked or misunderstood. Participatory democrats, for example, tend to ignore the fact that he divides executive and legislative power. Indeed, they usually fail to mention that if Rousseau famously insists on democratic self-legislation, he argues just as emphatically against democratic self-government. If the participatory interpretation overlooks the separation of powers, so too does the opposed reading that sees Rousseau's sovereign people as inactive or “immobile.” This reading is typified by Shklar (1969), who argues that “the sovereign does very little” (181). For Shklar, the institutions of Rousseau's state are designed “to prevent all change,” and she therefore almost entirely ignores the separation of powers and only briefly discusses the government, characterizing it as the “active agent of destruction” of the state (209–11). Finally, interpreters who do take note of Rousseau's separation of powers often argue that this delegation of power seriously undermines democratic sovereignty (Cullen 1993; Fralin 1978; Johnston 1999). Putterman steps into this debate to analyze the consistency between Rousseau's political principles and the institutions he recommends, bringing Rousseau into contemporary debates over democratic principles and political institutions in a potentially fruitful way.
Despite the virtues of his approach, however, Putterman's interpretation is based on a misunderstanding of Rousseau's theory of the separation of powers. “As brief as Rousseau is on this point,” Putterman (2003) claims, “he is consistent that a body other than the sovereign must initiate the law” (467). Rather than letting the sovereign both initiate and ratify the law, he claims, Rousseau chooses “to subdivide the legislative power between representatives and the majority” (466). In arguing that Rousseau delegates the exclusive power to initiate law to the government, however, Putterman identifies as a solution to the problem of democratic self-legislation what Rousseau sees as the greatest danger to it: the transfer of agenda-setting power to the government. Rousseau does not argue that the legislative power must be divided or subdivided, and he stresses throughout his work that the government must be restricted to executing the law.
At issue here is not merely the correct interpretation of Rousseau's political theory, for Putterman's argument raises far-reaching questions concerning the compatibility of democratic principles and institutions. In this article, I analyze Rousseau's theory of the separation of powers and demonstrate that Putterman is incorrect in his claim that Rousseau invests the executive with the agenda-setting power of legislative initiative. Throughout I follow Putterman in using the term “agenda-setting” to refer to the power of legislative initiative, although this power is more properly a form of “gatekeeping,” which is merely one form of “agenda-setting.” If Rousseau rejects such agenda-setting power by the executive as undemocratic, his theory may fall prey to a related issue central to contemporary democratic and social choice theory: the problem of preference aggregation through voting in the absence of agenda-setting institutions. I therefore conclude by examining how Rousseau anticipates and addresses the paradox of voting. Overall, my aim is to show that Rousseau's treatment of both agenda-setting and voting throws light on issues of principles and institutions still at the center of democratic theory and practice.
ROUSSEAU'S ANTI–AGENDA-SETTING AGENDA
—I warn the reader that this chapter should be read carefully, and that I do not know the art of being clear for those who are not willing to be attentive.
Jean-Jacques Rousseau (The Social Contract [SC] III.1)
Rousseau includes the above admonition to the potentially inattentive reader at the head of his discussion of government in Book III of The Social Contract. The warning is unique to the treatise and is meant to expose a particularly pernicious error in politics. Up to this point in the work, Rousseau has insisted that law can legitimately come only from the sovereign people, but now he is about to argue that the power to execute the law should be delegated to a separate body, the “government.” This delegation of power nonetheless poses a threat. Rousseau's warning is animated by his fear that the people will lose their sovereign authority, either when the executive illegitimately assumes the legislative power or when the citizens mistakenly alienate their right to representatives. One of Rousseau's primary purposes in Book III of his treatise is, therefore, to alert the people to the danger inherent from the government and to suggest ways in which that body can be kept within its sphere of executing the law rather than making it.
Given Rousseau's insistence on the separation between legislative and executive powers, Putterman (2003) asks the right question in response to the overly democratic reading of Rousseau that would invest both the legislative and the executive powers in the people: “Why is a separation of powers among lawmakers even necessary?” (460–61). Putterman is correct that Rousseau is dividing political power, but the answer he provides to the question is based on a misunderstanding of the separation of powers and Rousseau's reasons for arguing that it is necessary.
The Separation of Legislative and Executive Power
In the chapter to which he attaches his warning to the inattentive reader, Rousseau distinguishes between the legislative and the executive powers. The relationship between the sovereign people and the government has proved “unfathomable in politics,” according to Rousseau: “Here, all legislators have gone astray” (Rousseau 1978, 168; see also Letters Written from the Mountain, Letter 8, Rousseau 1990–, 9:257). His delineation of the legislative and executive powers requires equally clear and distinct terminology. In explaining his novel theory of the separation of powers he nonetheless chooses to use familiar terms, giving old words new meanings. His warning at the head of the chapter therefore alerts the reader to pay attention to usage. Not only do Rousseau's terms often not correspond to our usage today, posing an obstacle to proper understanding of his work, but also they are sometimes at odds with the usage of his own time. Such is the case for the general term he uses to name the executive power: “government.” The term “government” in Rousseau's time, as often in ours, includes the power to make law. Indeed, the sovereign legislative power is the main attribute of “government” in this understanding. This is the case even for Montesquieu, the theorist who most influenced Rousseau's theory of the separation of powers (see Montesquieu [1748] 1951, III.1). Rousseau complains, however, that the government “has been incorrectly confounded with the sovereign, of which it is only the minister” (SC III.1). His theory requires a radical separation of powers that restricts the government to the execution of the law.
The primary purpose of the separation of powers in Rousseau's theory is to maintain the generality of the laws from the particularity necessary in their execution by delegating the executive power to a distinct body. “We have seen that the legislative power belongs to the people and can belong only to it,” he explains at the beginning of his discussion of government (SC III.1; see II.2). Since the laws emanating from the general will of the people must be general in their form as well as in their source (II.6), the power to execute the laws cannot belong to the people in its capacity as sovereign (III.1). A separate body is needed to execute the laws and give movement to the body politic. This is the “government.” Rousseau gives several alternative names to the body and its members: “The members of this body are called magistrates or kings, that is, to say governors; and the body as a whole bears the name of prince” (III.1). His terms amount to putting entirely new wine into old bottles, and he thereby intimates that kings and princes have hitherto misunderstood—or have abused—their proper function.
Rousseau consistently emphasizes the limited functions and subordinate nature of the government. The government is “an intermediate body established between the subjects and the sovereign for their mutual communication and charged with the execution of the laws and the maintenance of civil as well as political freedom” (SC III.1). The government is “only the minister” of the sovereign. “It is absolutely nothing but a commission, a function in which, as simple officers of the sovereign, they exercise in its name the power that has been entrusted to them by the sovereign, and that the sovereign can limit, modify, and take back whenever it pleases” (III.1). As he explains earlier in the work when discussing the fact that the law much come from the sovereign people, “the government must not be confounded [se confonde] with the sovereign, but must be its minister” (II.6, n.). The government and the sovereign must not “become merged” (se confondre) when it comes to the making of law: “The people that is subject to the laws ought to be their author” (II.6). In his more detailed discussion of the role of government in the essay on “Political Economy,” where he also stresses the need to distinguish the sovereign and the government, Rousseau (1978, 211) characterizes the government as attending to the execution of the laws and to “administrative and economic details” under the guidance of the general will expressed through law (216). Nowhere in his treatments of government does he suggest that it shares the legislative power with the people.
Rousseau's discussion of the different forms of government revolves around the question of which form will best maintain the separation of legislative and executive powers. For example, in his rejection of democracy as a form of government he emphasizes the main reason for separating the legislative and executive powers: investing the same body with both powers would lead inevitably to the corruption of the general will by the temptation of particular considerations. “If there were a people of Gods, it would govern itself democratically. Such a perfect government is not suited to men” (SC III.4). His discussion of monarchy likewise points to the frailty of human nature as a danger, in this case the danger that the king will exceed the bounds of his executive function. “Kings want to be absolute,” Rousseau declares, and he subtly underscores the threat from those who have considered themselves “sovereign” by avoiding the term in the chapter on monarchy (III.6; Gildin 1983, 114). Finally, in recommending elective aristocracy as his preferred form of government, Rousseau stresses not only that it is most liable to lead to government by those with probity, enlightenment, and experience, but also that it “has the advantage of distinguishing between the two powers,” legislative and executive, that he wants to keep separate (SC III.5).
If the first half of Book III of The Social Contract is devoted to defining “government,” delineating its limited functions, and examining the forms it might take, the second half of the book is concerned with preventing the usurpation of the sovereign power by the government (see Gildin 1983, 127). Rousseau views the expansion of government power, and its concentration in an ever smaller body, as the main cause of the decline and death of the body politic. The dissolution of the state, he explains, comes about “when the prince no longer administers the State in accordance with the laws and usurps the sovereign power” (SC III.10). He is similarly concerned that a people fallen away from virtuous concern with the public good will cede its power to legislate to representatives, an illegitimate alienation of authority. “Sovereignty cannot be represented for the same reason it cannot e alienated. It consists essentially in the general will, and the will cannot be represented” (III.15). In order to ensure that the people preserves its sovereign power, Rousseau argues that it must assemble as often as possible (III.12–14). When the people assembles, the power of the government ceases (III.14), and the assembly is opened by asking whether the sovereign wishes to maintain the same form of government and whether it wishes to leave the administration in the hands of those currently responsible for it (III.18). This argument about the cessation of the executive power alone provides persuasive evidence that the deliberations of the sovereign people are not limited by any agenda set beforehand by the government. Indeed, like the decision to maintain or abolish the present form of government, the institution of government is itself a law that cannot have been put on the agenda by the body it creates (III.17). As Rousseau (1990–) explains in a fragment, “If laws exist before government, they are therefore independent of it; the government itself depends on laws since it is from them alone that it derives its authority; and far from being their author or master, it is only their guarantor, administrator, and at most, interpreter” (4:28).
A Critique of Putterman's Interpretation
Putterman claims throughout his article that Rousseau invests the agenda-setting power of legislative initiative with the government, but he acknowledges that the evidence is sparse. He is surprised that “Rousseau offers just two brief cryptic comments on the positive features of legislative representation in his mature state—and both within the context of rejecting the sovereignty of representatives.” He conjectures that Rousseau did not elaborate on this purported agenda-setting power because he wanted to obscure the significance of the government's power (Putterman 2003, 461–62; see also 460, n.2). Examination of the evidence Putterman cites for his claim reveals not only that the government in Rousseau's political system does not legitimately have the power to initiate laws and set the legislative agenda, but that the desire of the government to possess that power poses a constant threat to the people's legislative authority.
The first passage Putterman adduces for his claim that the government possesses agenda-setting power comes from the chapter of The Social Contract “On Deputies or Representatives” (III.15). Rousseau explains there:
Sovereignty cannot be represented for the same reason it cannot be alienated. It consists entirely in the general will, and the will cannot be represented. Either it is itself or it is something else; there is no middle ground. The deputies of the people, therefore, are not nor can they be its representatives; they are merely its agents [commissaires]. They cannot conclude anything definitively.
This passage is an inauspicious source for wisdom on the power and role of the government since Rousseau is not even discussing the government's (proper) function in this context. Rousseau's argument in the chapter under analysis is that the existence of representatives or deputies is a sign of a degenerated state. His specific point in the passage is that if the citizens decide to “name deputies and stay home,” their supposed representatives should be considered as mere agents (commissaires) who make proposals for the people to consider in its capacity as the legislative authority. He is not describing the legitimate role of the government in his own system, for he is decisively not recommending representation in any form. His intention is rather to warn against an improper understanding of delegated legislative power, and his argument is at best a suggestion of how to make the best of a bad situation.
Even assuming that Rousseau was describing the role of the government in this passage, the inferences Putterman draws from it go well beyond what is actually written. Putterman's positive claim that the government possesses agenda-setting power is based on Rousseau's negative statement that the “deputies” or “agents” of the people “cannot conclude anything definitively.” Putterman seems to assume that if these commissaires “cannot conclude anything definitively,” they must be proposing something. That is, they must propose laws. Furthermore, rather than arguing that the government can propose laws for the people's consideration, which is a strained but plausible interpretation of this passage under the present assumption, Putterman claims that only the government may initiate laws. To be sure, Putterman (2003) argues that the government may not actually make the laws: “Laws proposed by commissaires are illegitimate and the expression of an alien will unless and until each is met with the express approval of the majority” (464). While it is true that Rousseau insists that the sovereign people alone can make law, he does not positively argue that commissaires, magistrates, or anyone else has the power to propose laws.
Finally, since I have emphasized Rousseau's insistence on precise terminology, let me note that the chapter “On Deputies or Representatives” is the only place in The Social Contract where Rousseau uses the terms “representative,” “deputy,” and “agent” (commissaire). By sequestering these terms, he indicates that they are not synonymous with the proper term for members of the government, “magistrates.” Putterman (2003) nonetheless uses the term “commissaires” throughout his article interchangeably with “magistrates,” and does the same with “deputies” and also occasionally with “representatives” (e.g., page 459 [abstract]), although he does distinguish between commissaires and representants at one point (461; see also 467). At a minimum, Putterman thereby introduces unnecessary confusion into his interpretation by blurring the terms Rousseau wants to keep separate, like the roles they identify.
The second passage from The Social Contract Putterman cites on behalf of his interpretation also comes from an argument about the illegitimate usurpation of the people's legislative power. The passage in question concludes the chapter “That the General Will Is Indestructible”:
I could make many comments here on the simple right to vote in every act of sovereignty, a right that nothing can take away from the citizens, and on the right to give an opinion, to make propositions [proposer], to analyze, to discuss, which the government is always very careful to allow only to its members. But this important subject would require a separate treatise, and I cannot say everything in this one. (IV.1; trans. altered)
The intended irony of this passage is manifest from its context. Rousseau argues in this chapter that the general will endures in a state even as the social bond weakens and private interests begin to triumph over the public interest. “Does it follow from this that the general will is annihilated or corrupted? No, it is always constant, unalterable, and pure. But it is subordinate to others that prevail over it” (SC IV.1). In this context, then, he argues that the people maintains its rights— even when the government attempts to prevent it from doing so. Hence the sarcastic conclusion: “The government is always very careful to allow only to its members” the power to give opinions, make propositions, analyze, and discuss. If the statement were limited to stating that the government is careful to retain the right “to make propositions” (proposer), then it might possibly be read as sincere. But surely the other rights he lists—giving an opinion, analyzing, and deliberating— belong to the citizens, and not to the government alone. They have the right to propose laws and then analyze, discuss, and vote on them. This interpretation is strengthened by the very next chapter, where Rousseau discusses voting and insists that the citizens must have “public deliberations” (IV.2). Far from indicating that the government has agenda-setting power, then, Rousseau's text is a warning that the government will try to usurp the people's exclusive right to make law.
To be fair, the irony of this passage has eluded interpreters besides Putterman, including Derathé, who voices perplexity over Rousseau's apparent limitation of the legislative right in his editorial note to the passage in the Pléiade edition (Rousseau 1959–95, 3:1492–93, n.1–439; see also Cullen 1993, 215–16). Gildin is closer to a correct interpretation, I think, when he glosses Rousseau as saying that the government “seeks” to arrogate these rights to itself, and suggests that the author's words “do not amount to an enthusiastic endorsement of the practice they describe.” He goes too far, however, when he concludes that “Rousseau's remarks suggest that there are many possible arrangements regarding the right to propose and to debate laws compatible with his principles” (Gildin 1983, 159; see also Melzer 1990, 173). Bertram (2004, 173) is still more tentative, concluding, “Certainly it looks like an unsatisfactory basis for ascribing to Rousseau the position that the government should dominate, since so much of the discussion of Book III has centered on the dangers of this very possibility.” In line with my own interpretation, Kelly argues that Rousseau is “warning against this practice as a symptom of the effort of governments to usurp legislative power.” He further points out that Rousseau never places restrictions on the right of the people to give opinions, make propositions, analyze, and discuss legislation, and suggests that the “important subject” requiring a separate treatise refers to the conditions for preserving democratic deliberation (Kelly 2003, 125–26).
Finally, elsewhere in his writings Rousseau does discuss the power of government to propose laws in his native Geneva, but Putterman wisely rejects this evidence. Like the first passage analyzed above from The Social Contract, concerning representatives, these discussions have the remedial purpose of at least limiting the illegitimate powers usurped from the citizenry. In the relevant passage from the Dedication to Geneva from the Discourse on Inequality, Rousseau (1990–) writes: “I would have desired that, in order to stop the selfish and ill-conceived projects and the dangerous innovations that finally ruined the Athenians, everyone did not have the power to propose new laws according to his fancy; that this right belonged exclusively to the magistrates” (3:5). This passage is cited by Fralin (1978, 54) in his analysis of Rousseau and representation as evidence that the government properly possesses the power to propose legislation in Rousseau's state (see also Melzer 1990, 173). Putterman (2003) nonetheless persuasively argues that this evidence offers “soft ground” for such an interpretation given that Rousseau's intentionally idealized portrait of his native city is an implicit critique of the illegitimate usurpation of the legislative power by the Petit Conseil (461–63). I would add that Rousseau's (1990–) argument here is that no one should have “the power to propose new laws according to his fancy [fantaisie],” since for him legitimate laws must have a specific form and formality. He is not arguing that the citizens do not have the right to propose legislation, for in the preceding paragraph he asserts that the “the right of legislation [droit de législation]” should be “common to all citizens” (3:5), and, given the formulation, this right presumably includes the right to propose proper laws.
The potential evidence from the Letters Written from the Mountain (esp. Letter 9, Rousseau 1990–, 9:283–86) is similarly unpersuasive. Putterman does not refer to these passages in his analysis, but Melzer (1990) does in his considerably more tentative discussion of the possible role of the government in initiating legislation (173, 212; see also Gildin 1983, 159). Rousseau's concern in this context, however, is to wrest as much power as possible from the Petit Conseil, which claims the sole right to legislate, and if he recognizes its “negative right” over the laws, restricting the body to this function would amount to an effective reduction of its legislative power. He therefore strenuously insists that the right to debate proposals belongs to the people and he argues that the people also has the right to enact laws not proposed by the magistrates (Letters Written from the Mountain, Letter 7, Rousseau 1990–, 9:250). The most promising passage for the alternative interpretation, cited by Melzer, comes when Rousseau agrees with his opponent that some form of veto power “very reasonable,” at least in the case put forward, and explains that it would be impossible for a democratic constitution to “maintain itself if the legislative power could always be set into motion by each of those who compose it” (Letter 9, 9:285). First, in the Genevan case Rousseau is discussing there are two legislative bodies that “compose” the legislative power, and he is therefore discussing their concurrence and a possible “negative power” of one body on another. Second, as with the passage from The Social Contract where he objects to anyone being able to propose new laws “according to his fancy,” Rousseau is once again emphasizing that the initiation of legislation cannot “always be set into motion,” but can legitimately occur only under certain circumstances and in certain forms (see SC III.13). In any case, Rousseau's acknowledgment of the power to propose laws possessed by the Genevan government does not constitute a division of legislative power he would otherwise approve.
In sum, then, the evidence from The Social Contract and elsewhere in the author's writings does not support Putterman's claim that the government in Rousseau's state possesses the exclusive right to propose legislation. In my analysis of the passages Putterman cites for his interpretation, I have shown that Rousseau's intention is quite the contrary: he is warning the people that the government will try to usurp its legislative authority by claiming the sole right to propose and debate legislation or, when he seems to acknowledge such a right, he is trying to limit the damage from what he otherwise considers an illegitimate arrangement. If Rousseau nowhere positively argues that the government should possess the right to propose laws, much less the exclusive right to do so, the most plausible alternative argument would be that, since he nowhere positively asserts that the government can never possess the power of legislative initiative, the government may under certain conditions propose laws for the people's consideration. Such an argument is plausible, and is similar to the interpretations offered by Gildin (1983) and Melzer (1990). If I am correct about the available evidence, however, this interpretation suffers from the fact that it argues from a negative; it is a dog that did not bark in the dark. The dog may indeed sometimes bark, for Rousseau does not definitively state that the government can never propose laws. Given that he explicitly states that the people alone possesses the legislative power, however, the more compelling interpretation is that in a legitimate state the right of legislative initiation belongs to the people as part of its legislative power.
ROUSSEAU AND CONTEMPORARY DEMOCRATIC THEORY
From Rousseau's perspective, as well as from the perspective of his contemporary democratic followers, agenda-setting is a threat to popular sovereignty. I have argued against Putterman that Rousseau does not delegate the agenda-setting power of legislative initiative to the government. If I have removed Rousseau from the frying pan of this agenda-setting power, however, I may have done so only to throw him into the fire of the paradoxical problems of voting identified by contemporary social choice theory. In order to broaden the inquiry into the relevance of Rousseau's thought to contemporary democratic theory begun by Putterman through his discussion of agenda-setting and majority rule, then, in conclusion I suggest how Rousseau anticipates and addresses the related issue of voting in the absence of agenda-setting institutions such as legislative “gatekeeping.”
The classic contemporary critique of Rousseau that raises the paradoxes of voting in the absence of agenda-setting institutions is Riker's Liberalism Against Populism (1982). Riker discusses the problems of preference aggregation, cycling, and agenda-setting faced by any voting system, including Rousseau's “populist” theory. Drawing on Arrow (1963), he shows that where there are at least three voters and three choices, different outcomes may result from identical values depending on the voting rule. Given this “paradox of voting,” there is no actual or identifiable common good, or “general will.” Similarly, different orderings of preferences would lead to “cycling,” where voters cycle through different, but unstable majorities. Under these conditions there is a pervasive opportunity for manipulating the outcome of voting through agenda-setting, or controlling the order in which the alternatives are considered. If Rousseau is aware of the pervasiveness of agenda-setting and tries to prevent the government from controlling the legislative agenda, is he similarly conscious of the paradoxes of voting?
There have been a number of responses to Riker's argument and his critique of Rousseau, including skeptical investigations of whether deliberative democracy can remedy the problems of aggregative democratic institutions identified by social choice theory (Knight and Johnson 1994) and arguments about the efficacy of deliberative democracy based in part on a critique of the social choice approach (Dryzek 2000). The most interesting attempts to clarify Rousseau's thought with the findings of social choice theory in mind have centered on interpreting the “general will” in light of Condorcet's jury theorem (Grofman and Feld 1988; see also Barry 1965, 292–94, and Runciman and Sen 1965). These attempts characterize voting in Rousseau's theory as the expression of opinions about what is in the common interest, or the identification of the general will, and therefore adopt an “epistemic” conception of voting (see Coleman and Ferejohn 1986). This approach to the general will is useful, and has good textual warrant in Rousseau's writings. Before the citizens give an opinion on the common good, however, they have to will or desire it. The measures Rousseau proposes to increase the probability that the general will is realized in this sense are aimed less at the citizens' knowledge than at their preferences.
In his own solution to something like the paradox of voting identified by social choice theory, Rousseau restricts the domain of the preferences, or wills, that can be or are likely to be expressed by the citizens through voting. The first domain restriction Rousseau imposes is a formal one, for he argues that laws must be general in form: “Thus just as a private will cannot represent the general will, the general will in turn changes its nature when it has a particular object; and as a general will it cannot pass judgment on either a man or a fact” (SC II.4). The subject of the law must be “general like the will that enacts,” and “there is no general will concerning a particular object” (II.6). This is precisely Rousseau's logic for delegating the execution of the law to a separate body. A number of contemporary democratic theorists, in part following Rawls (1971), address the issues raised by social choice theory by suggesting analogous restrictions on what can legitimately considered in deliberation (Dryzek 2000, 42–47; Gutmann and Thompson 1996).
The second way in which Rousseau restricts the domain is to limit the variety of preferences that are likely to be expressed in his state. That is, if the possibility of cycling occurs where there are at least three voters and three alternatives, one could solve the problem by reducing the number of voters or alternatives to two or fewer. Rousseau tries to accomplish this by increasing the homogeneity of the citizens, which would limit the variety of preferences. In principle, this strategy would approach a situation where there is a uniform preference-ordering among all citizens or, more realistically, one sufficiently dominant preference or preference-ordering. As Trachtenberg (1993, 256) notes, Arrow (1963) himself recognizes this possible solution to his paradox when he explains that “the possibility of social welfare judgments rests upon a similarity of attitudes toward social alternatives” (69). The question for Arrow would be whether such a strategy would violate his condition of citizens' sovereignty, which dictates that the social welfare function is not to be imposed (28–30).
Rousseau's awareness of the strategy of domain restriction can be seen from his treatment of different “wills” present in the same person or group of people. The most general case is the conflict he acknowledges between the potentially competing wills each person has as an individual and as a citizen: “Indeed, each individual can, as a man, have a private will contrary to or differing from the general will he has as a citizen.” In order to solve or at least ameliorate this conflict, the general will one has as a citizen must be made to dominate the particular will one has as an individual. In Rousseau's infamous formulation, the person whose particular will is predominant “will be forced to be free” (SC I.7). His point here is clear enough for the present purposes without entering into the long debate over the meaning of this shocking phrase: the citizens all have the same dominant preference, to prefer the common good as a citizen. Rousseau's basic idea can translated into the language of single-peaking preference curves, for as Riker and Ordeshook (1973) explain, “reflect a cultural uniformity about the standard of judgment, even though people differ about what ought to be chosen under that standard” (105).
In order to encourage the dominance of the general will and also to limit the range of likely preferences, Rousseau argues that civic-mindedness must be instilled through education and civic religion and common customs, opinions, and mores (see SC II.7, II.12, IV.8). If the citizens all think of themselves as being the same, then their individual preferences become common preferences. “Why is the general will always right and why do all constantly want the happiness of each,” Rousseau explains, “if not because there is no one who does not apply this word each to himself, and does not think of himself as he votes for all?” (II.4). Relative equality among the citizens further reinforces this mutual identification (see I.9, II.11), for “the private will tends by its nature toward preferences, and the general will toward equality” (II.1). Rousseau's advice to eliminate or disperse “partial societies” has a similar intention, for these factions multiply preferences and preference-orderings (II.3; see also Grofman and Feld 1988). Overall, then, limiting the heterogeneity of preferences makes the restriction of the domain to generalizable preferences even more effective both as a solution to the paradox of voting and as a way to prevent agenda-setting.
The most elaborate treatment of the issue of competing preference-orderings comes, revealingly enough, in Rousseau's discussion of a problem posed by creating a separate executive power:
We can distinguish three essentially different wills in the person of the magistrate. First, the individual's own will, which tends only toward his particular advantage. Second, the common will of the magistrates, which relates uniquely to the advantage of the prince [i.e., government]; which may be called the corporate will…. Third, the will of the people or the sovereign will, which is general in relation to the State considered as a whole and in relation to the government considered as a part. (SC III.2; trans. altered)
“In perfect legislation,” he explains, “the particular or individual will should be null; the corporate will of the government very subordinate; and consequently the general or sovereign will always dominant and the unique rule of all the others.” In other words, if the individual will is rendered “null,” in principle there are only two preferences with a determinate preference-ordering. But there is a problem, for nature does not follow the dictates of politics. The “natural order” that derives from the primacy of natural self-love or self-interest produces a “preference-ordering” of these three wills that is “exactly the opposite” to that required by politics, with the will of the individual predominant, the corporate will secondary, and the general will weakest of all (III.2). It is the inevitable ascendancy of this natural order, especially among the magistrates, that Rousseau credits with the death of the body politic. He attempts to ameliorate this problem both through institutional means, keeping the executive power within its bounds and insisting that the people assemble often to exercise their general will, and through cultural means, inspiring magistrates with the same public-spiritedness that must reign among the citizens.
Rousseau's awareness both of agenda-setting and of the solution to the paradox of voting through domain restriction comes to light in the chapter from The Social Contract that ends with the second passage examined above in which he sarcastically warns against the government's desire to usurp the power to propose laws. The beginning of the chapter is more hopeful:
As long as several men together consider themselves to be a single body, they have only a single will, which relates to their common preservation and the general welfare. Then all the mechanisms of the State are vigorous and simple, its maxims are clear and luminous, it has no tangled, contradictory interests; the common good is clearly apparent everywhere, and requires only good sense to be perceived. Peace, union, and equality are enemies of political subtleties…. (SC IV.1)
In such a world, there is no room for subtle manipulations such as agenda-setting and voting is unproblematic. But Rousseau is not naive. Having painted his picture of peasants under an oak tree reaching immediate and unanimous decisions, he warns that the government is all too careful to accord to themselves the right to give an opinion, to make propositions, to analyze, to discuss (IV.1). The art of political manipulation is a threat to the democratic society he portrays, but he also knows that his portrait is idealized. “If there were no different interests, the common interest, which would never encounter any obstacle, would scarcely be felt. Everything would run smoothly by itself and politics would cease to be an art” (II.3, n.).
Everything does not run smoothly of itself, of course, and Rousseau is well aware of this. If part of his agenda is to prevent agenda-setting by the government in the form of legislative initiative, Rousseau's argument that individual self-interest or self-love is at the core of human nature does not allow him to entertain political perpetual motion. He calls for a higher political art that is a higher form of agenda-setting. This is the art of the great legislator, who must create the restrictions on the domain of preferences that make self-legislation in the absence of agenda-setting possible (SC II.7). Rousseau says that his role is “extraordinary,” and it is both extraordinary in the sense of being necessarily outside the normal course of the law— otherwise such “law-making” would be illegitimate by Rousseau's standard of self-rule—and extraordinary in its difficulty (II.7). Since the political spirit and therefore political institutions tend to decay, Rousseau also suggests that analogous interventions into the political machine are necessary. “Just as the regimen of healthy people is not suited to the sick, one must not want to govern a corrupt people by the same laws that are suited to a good people” (IV.4). In his extended discussion of Rome (IV.4–7), for example, he argues that different methods of voting were necessary for different purposes and in different times, and suggests that divisions of power between the patricians and the plebians became necessary, as did the institution of the tribunes. The erection of these institutions and their actions can be understood as forms of agenda-setting. Such agenda-setting is possible, even necessary, for the continued working of Rousseau's state. However, they are legitimate if and only if they create the conditions necessary for the active and effective exercise of the sovereign people through voting. As with the power of legislative initiative he refuses to delegate to the executive, Rousseau is ever cognizant of the danger they pose to popular sovereignty.
In conclusion, we should ask what we can learn from Rousseau's approach to the problems of agenda-setting, especially the paradoxes of voting central to contemporary democratic and social choice theory. If the solution to the paradox of voting requires a high degree of cultural homogeneity dependent on a sort of higher-order agenda control, is this solution possible or desirable in the modern state? If Rousseau himself admits that the strict separation of powers necessary to keep the government within its bounds of executing the law is inherently fragile, is democratic self-legislation sustainable? Let me suggest two answers. First, the trade-offs Rousseau makes us consider in his treatment of these issues underscore the fact that these questions have no easy answers. Attempts to take up Rousseau's thought as a cudgel on behalf of participatory democracy while dismissing the paradoxes of voting identified by social choice theory or to dismiss his thought as naive democratic theorizing ignorant of the issue of agenda-setting alike underestimate the thinker. They perhaps also underestimate the difficulty of the theoretical and practical issues in contemporary democratic and social choice theory. Second, the stark, even intemperate, solutions Rousseau proposes can illuminate the questions and problems perhaps less evident in more familiar forms. For example, his approach to the paradox of voting through domain restriction takes less dramatic form today in analyses in political science of the tendency of issues and ideology to become arrayed along a single dimension, sometimes through agenda-setting by parties and politicians. Like agenda-setting institutions themselves, this tendency might be considered a happy example of how seemingly intractable theoretical problems of voting are resolved in practice, as lamentable restrictions of democratic deliberation, or as an issue requiring further reflection and study. The questions of institutions and voting Rousseau addresses in his theory are still with us today.
%20-%20https://sketch.com%20--%3e%3ctitle%3e3ECBBF50-FD08-412A-943D-2CFC5E025A16%3c/title%3e%3cdesc%3eCreated%20with%20sketchtool.%3c/desc%3e%3cg%20id='Stylesheet'%20stroke='none'%20stroke-width='1'%20fill='none'%20fill-rule='evenodd'%3e%3cg%20transform='translate(-926.000000,%20-668.000000)'%20id='BUTTONS'%3e%3cg%20transform='translate(100.000000,%20597.000000)'%3e%3cg%20id='BTN-/-RIGHTS'%20transform='translate(816.000000,%2065.000000)'%3e%3cg%20id='ICON-/-Rights'%20transform='translate(10.000000,%206.000000)'%3e%3cpath%20d='M13.5752841,0%20L1.34943182,0%20C0.59375,0%200,0.586666667%200,1.33333333%20L0,22.6666667%20C0,23.4133333%200.620738636,24%201.34943182,24%20L17.6505682,24%20C18.40625,24%2019,23.3866667%2019,22.6666667%20L19,5.36%20L13.5752841,0%20Z%20M13.5752841,1.89333333%20L17.0838068,5.36%20L13.5752841,5.36%20L13.5752841,1.89333333%20Z%20M1.34943182,22.6666667%20L1.34943182,1.33333333%20L12.1988636,1.33333333%20L12.1988636,5.36%20C12.1988636,6.10666667%2012.7926136,6.69333333%2013.5482955,6.69333333%20L17.6235795,6.69333333%20L17.6505682,22.6666667%20L1.34943182,22.6666667%20Z'%20id='Shape'%20fill='%23595959'%20fill-rule='nonzero'%3e%3c/path%3e%3ccircle%20id='Oval'%20stroke='%23595959'%20stroke-width='1.5'%20fill='%23595959'%20cx='9.5'%20cy='13.5355339'%20r='5.5'%3e%3c/circle%3e%3cpolyline%20id='Path-2'%20stroke='%23FFFFFF'%20stroke-width='1.5'%20transform='translate(9.500000,%2013.035534)%20rotate(45.000000)%20translate(-9.500000,%20-13.035534)%20'%20points='8%2016.0355339%2011%2016.0355339%2011%2010.0355339'%3e%3c/polyline%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
Comments
No Comments have been published for this article.