1. Two Interpretations of the General Will

Two camps emerge on whether the formal constraints described in the Social Contract are enough to realize the general will.Footnote ⁷ Those constraints are relatively straightforward and reflect different forms of generality:Footnote ⁸

Some find these sufficient for the general will's expression. Roger Masters describes the general will as “a formal requirement which must be fulfilled by the laws which constitute any legitimate regime.”Footnote ¹² John T. Scott argues that Rousseau “abandons all substantive standards of justice and law.”Footnote ¹³ Leo Strauss finds the general will so devoid of normative content that “cannibalism is as just as its opposite,”Footnote ¹⁴ and Tuck has recently emphasized the power of the simple vote in establishing the general will.Footnote ¹⁵ All of these formal readings locate the general will's uprightness in a set of procedures and aims.

Formal readings struggle, however, to reconcile Rousseau's positivism with the fact that popular judgment is “not always enlightened” (SC 154; 3:380). The people can err. Their votes can be procedurally legitimate but substantively illegitimate. This has led scholars to seek what Christopher Bertram calls a “transcendent fact of the matter” as to what assemblies should choose.Footnote ¹⁶ On these “transcendent” accounts, law is legitimate only when it is normatively correct. The nature of this correctness varies. Some take the general will to exist “out there.” Jacob Talmon finds in the concept a “mathematical truth or Platonic ideal,” while Jon Mandle posits a “hypothetical outcome of an idealized assembly.”Footnote ¹⁷ Others see a commitment to certain principles. Judith Shklar identifies a “will against inequality.”Footnote ¹⁸ David Lay Williams observes across Rousseau's work a conformity to “an eternal idea of justice,” leading Williams to conclude that “an objective or even transcendent conception of justice is part of the core meaning of the general will.”Footnote ¹⁹ Transcendent accounts maintain that the general will is always right because it embodies right values.

Both approaches capture something essential about the general will. Formalists rightly observe that legitimacy in the Social Contract is always and only a matter of voting. Rousseau gives no mechanism by which to check an assembly's decision against an external standard of right. Indeed—in a vital phrase—he finds that the laws “merely record our wills” (SC 153; 3:379).Footnote ²⁰ “And what is a Law?” he asks in Letters Written from the Mountain. “It is a public and solemn declaration of the general will, on an object of common interest” (LM 232; 3:807–8). Yet Shklar and Williams are also right to note that the general will always tracks certain values. By definition it cannot be unjust (LM 231; 3:807). Rousseau stresses that the general will tends toward “the preservation and welfare of the whole and each part” (PE 143; 3:245), toward “equality” (SC 145; 3:368) and “the public utility” (SC 147; 3:371). This suggests that the will does contain fixed content.

Rousseau's three distinct conceptions of the general will help us to resolve this apparent impasse. These conceptions address different agents, the will of the body politic operating differently from that of the assembly or individual. Rousseau's first discussion of the volonté générale in Political Economy postulates an implicit will of associations to promote the good of their members. Crucially, the text employs only this conception. Nowhere does it describe a general will shaped by popular expression or debate. Indeed, Rousseau even imagines “the large town of the world”—which could never gather to deliberate—acting as a body politic in which “the law of nature is always the general will” (PE 143; 3:245).

The “Geneva Manuscript” and the Social Contract introduce two new conceptions. One is the general will declared by an assembly. When “the qualities of the general will are still in the majority” (SC 201; 3:441) that majority's vote must be the general will. Rousseau also identifies a personal general will that each citizen finds “within himself” that is directed toward the common good (SC 199; 3:438; see also GM 79; 3:286). This public-spirited will checks a “particular will” bent toward private gain. Personal general wills are the source of the assembly's declared will and become the foundation of legitimate law.

Rousseau's shifting usage reflects a broader incompatibility between Political Economy and the Social Contract. Interpreters often take the former to reflect Rousseau's considered views; Williams takes as definitive the claim in Political Economy that leaders “need only be just in order to be sure of following the general will” (PE 148; 3:251).Footnote ²¹ The difficulty, as we shall see, is that one searches in vain for such comments in later works. There are other gaps. The idea of a general will of humanity is rejected outright in the “Geneva Manuscript” and never mentioned in the Social Contract. Political Economy never alludes to moral freedom or a general will possessed by individuals, while the Social Contract would be unrecognizable without these. Rousseau's use of a single phrase—the general will—to describe three discrete forms of volition has obscured the development of his thought between texts.

2. Transcendence in Political Economy

The general will of Political Economy is unique in two respects: it is an implicit feature of the body politic (rather than a possession of individuals), and deliberation is neither necessary nor sufficient for its expression. Rousseau's chief concern is that rulers be able to learn and legislate from this general will, though, as we shall see, even they need not know the general will so long as their actions align with its aims. Rousseau credits the “great and luminous principle” (PE 143; 3:245) of the general will to Denis Diderot's essay Droit naturel, published in the same 1755 edition of the Encyclopédie as Rousseau's essay. That reference reflects more than collegiality: Diderot's influence pervades Rousseau's early political writings, first as inspiration and then as foil. But Diderot's general will never earns its name; it is general, but it is never shown to be willed. Patrick Riley rightly remarks that Diderot's earlier talk of a morale universelle better captures the concept's significance in guiding moral reflection.Footnote ²² It is everywhere comprehensive and comprehensible, found in the laws and practices of both “civilized” and “barbarian” peoples and even in the “tacit conventions of the enemies of the human race.”Footnote ²³ It functions as a “pure act of the understanding . . . in the silence of the passions,” without which concepts of goodness, justice, humanity, and virtue remain opaque.Footnote ²⁴ For Diderot, the general will illuminates.

In Political Economy Rousseau accepts all of this. But he adds a key innovation: where Diderot had seen only the will of the human race, Rousseau finds that every association has a general will. In “the large town of the world” there is, he agrees, a “law of nature [that] is always the general will” (PE 143; 3:245), but this global society is just the largest of many nested associations, each with a will that is general toward members and particular toward outsiders. The human race itself thus has a general will—and so does every collectivity within it.

While Diderot and Rousseau agree that the general will offers an implicit standard of right, Rousseau doubts its efficacy. In the absence of positive law, universal right is weakly motivating and dimly perceived. “The feeling of humanity evaporates and weakens as it is extended over the whole world” such that “interest and commiseration must in some way be confined or compressed to be activated” (PE 151; 3:254). Leaders must therefore make the people love the fatherland and the freedom of its laws (PE 152; 3:255). Rousseau's epistemic concern presents more difficulty. Diderot had claimed that even brigands can consult the general will. Rousseau demurs. To follow the general will, “it must first be known” and “above all well distinguished from the private will, starting with one's own, a distinction that is always extremely difficult to make and on which only the most sublime virtue can shed sufficient light” (PE 145; 3:247–48). Political Economy gives this task of discernment not to citizens but to their leaders. This passage, like so much of the text, is directed toward authorities. Rousseau reinforces this at the end of the paragraph just cited, when he notes that it is “through law alone that leaders must speak when they command; for as soon as a man claims to subject another to his private will independently of the laws, he immediately leaves the civil state” (PE 145; 3:249). And even executive discernment may be optional as “leaders know very well that the general will is always for the side most favorable to the public interest—that is, for the most equitable; so that it is only necessary to be just and one is assured of following the general will” (PE 148; 3:251). A statesman could dispense entirely with the work of discerning the general will so long as he acts, by instinct or accident, in accordance with justice.

Can we harmonize this account with that of the Social Contract? Political Economy is a thoroughly republican text—the vox populi is the vox dei—but popular deliberation is given no essential role in either expressing or enacting the general will. If Rousseau can posit the general will of a human race which could never assemble, there can be no necessary bond between deliberation and the general will. Even near the end of the text, where he mentions in passing that taxation should be decided “through a general will, by majority vote” (PE 163; 3:270), this simply expresses “the consent of the people or its representatives” (PE 163; 3:270; emphasis mine), and does nothing to establish the necessity of popular deliberation.

Rousseau also emphasizes the insufficiency of deliberation: “An assembly is all the less necessary because it is not sure its decision would be the expression of the general will . . . and because it is rarely necessary when the government is well intentioned” (PE 148; 3:250–51). To reconcile this position with later texts, we would need to show that variations were merely rhetorical. Perhaps Rousseau simplified for the lay readers of the Encyclopédie, or simply suppressed his more radical views on citizen agency. Or perhaps, as Robert Derathé suggests, his explicit focus on government (rather than sovereignty) explains his attention to executive power and inattention to the agency of citizens (PE 142; 3:244).Footnote ²⁵ Perhaps the arguments of Political Economy only sound different.

Yet the logic of the texts rejects such reconciliation. As noted, popular deliberation in Political Economy is neither necessary nor sufficient for the general will's expression: unnecessary because executives can independently discern it and because associations can be too scattered to deliberate; insufficient because assemblies err.Footnote ²⁶ The Social Contract, by contrast, finds suitably situated deliberation to be both necessary and sufficient in a well-ordered society, such that the general will is known when and only when the people declare it. Assuming “the qualities of the general will are still in the majority” (SC 201; 3:441)—a key caveat—the majority's vote is always conclusive.

We should expect such contradictions since Political Economy is no contractarian text. Like the Social Contract it relies on a législateur to make society well ordered, but Political Economy also requires this figure to enact laws. Leaders can be oblivious of the polity's implicit general will so long as justice obtains. The laws must reshape wills to make men “what one needs them to be.” Newborns must “learn to deserve to live” (PE 155; 3:260). Not by coincidence does this essay feature Rousseau's strongest use of the body politic metaphor in which men become mere appendages, their identity lost; not by accident is Sparta invoked in the same breath (PE 143; 3:245).Footnote ²⁷ The text largely ignores citizen agency and never mentions the moral freedom made possible through rightful contract. We are far here from taking men as they are.

Like Rousseau's subsequent works, Political Economy links legitimate law with the general will. It also makes a case for wise leadership. But this text remains unique in claiming that the general will exists beyond public declaration and can be enacted by authorities who need not even know it. Both claims were soon abandoned.

3. The Transitional “Geneva Manuscript”

Masters has noted that in “the first draft of the Social Contract, it had been quite clear that the ‘general will’ was a modern version of a Platonic ‘Idea’ or ‘Form.’”Footnote ²⁸ This is certainly the text's dominant conception, but not its only one. The “Geneva Manuscript” preserves an experimental moment in Rousseau's thought that locates legitimacy in both the implicit will of the polity and the declared wills of individuals. This section examines two of the text's singular features: its introduction of the remaining two conceptions of the general will, and its conflicted account of the legislator's epistemic authority. In borrowing his account of law from Political Economy, Rousseau gives the Legislator too much epistemic authority, and his revisions to the “Geneva Manuscript” suggest that he wished to avoid this in the Social Contract by grounding legitimacy in citizen discernment and will. We revisit the draft, then, both for its idiosyncratic account of political right and to see why Rousseau was right to reject it.

The unpublished manuscript gives familiar concepts unfamiliar forms. Its rendition of the social contract requires placing one's will under the general will (GM 82–83; 3:96–97). Wholly different is the “fundamental problem” it aims to solve (GM 82–83; 3:96–97). The text never names moral freedom. These differences make it all the more surprising that there exists no comprehensive study comparing Rousseau's published and unpublished drafts. C. E. Vaughan's view in 1915 that the differences between texts are functionally “of no great importance” remains widely accepted, such that Robert Wokler could argue that the “style and form” of the text was the source of its unsuitability.Footnote ²⁹ Stylistic explanations do not withstand scrutiny, however, for just as the argument of the “Manuscript” contradicts Political Economy, so too does much of the Social Contract contradict the “Manuscript.” Indeed, not a single chapter of the “Manuscript” was left unchanged for publication.

Rousseau's great discovery in the “Manuscript” is that the general will must be understood and expressed by citizens. Though he now rejects outright the general society of mankind entertained in Political Economy (GM 81; 3:288), he retains that text's Diderotian universal morality. Quoting directly from Diderot's Droit naturel, Rousseau claims that no one will deny that “the general will in each individual is a pure act of the understanding, which reasons in the silence of the passions about what man can demand of his fellow man and what his fellow man has the right to demand of him” (GM 80; 3:286). But he doubts its efficacy for the two reasons already present in Political Economy. It cannot motivate: “It is false that in the state of independence, reason leads us to cooperate for the common good out of a perception of our own interest” (GM 79; 3:284). Natural man must be shown “how his personal interest requires his submission to the general will” (GM 80; 3:286). The epistemic obstacles are even more insurmountable, for outside civil society “it would be impossible for him ever to apply these principles with certainty in a state of things that would not allow him to discern [discerner] either good or evil, either the decent man or the wicked one” (GM 77; 3:282). This faculty of discernment is itself formed through the experience of citizenship:

Individuals—and not, as in Political Economy, leaders alone—are now tasked with “generalizing ideas” and discerning the boundaries of particular and general wills, exercising a faculty formed through the habit of “judging and feeling within society.” The published Social Contract will further elaborate the role of these in deliberation.

This passage also reflects the jurisprudential cast of the general will in the “Manuscript.” Rousseau ends the text's first section by asking how the implicit general will is discovered: “Will it always be evident? Will the illusions of private interest never obscure it? Will the people always remain assembled to declare it, or will they rely on private individuals who are ever ready to substitute their own interest?” (GM 98; 3:309–10). Only the law can satisfy such concerns. In a rhapsodic passage copied almost verbatim from Political Economy, Rousseau praises law as the “most sublime of all human institutions.” Law rectifies weak wills by “subjugat[ing] men in order to make them free.” It enlightens their understanding by acting as a “celestial voice that tells each citizen the precepts of public reason, and teaches him to behave according to the maxims of his own judgment.”Footnote ³⁰ “It is not enough for everyone to be subject to the general will,” Rousseau concludes. “To follow it, one must know it” and from this epistemic demand arises “the necessity for Legislation” (GM 99; 3:310).

Here we encounter the draft's unresolved dilemma, for if the law already embodies the general will, the potential circularity is obvious: the effect needs to provide its own cause, the “blind multitude” must write laws according to principles they do not yet see (GM 103; 3:317). To avoid this, Rousseau introduces the Legislator. The “Manuscript” explicitly restricts his godlike authority to drafting the laws and directing the general will itself by persuasion (GM 103; 3:317). And yet this introduces a new problem: if the general will is revealed only through his laws, then the people will have no other access. In effect his laws simply become the citizens’ general will. Either the Legislator assumes complete epistemic authority over the general will, or citizens must have another source by which to discern its outlines.

The epistocratic possibilities clearly trouble Rousseau.Footnote ³¹ In one of the most heavily edited passages of the “Manuscript,”Footnote ³² he wonders whether a polity could refuse rulership “to someone who is known to be the master and in whom are combined the public confidence and force” (GM 101; 3:314). If such a figure were given rule, subjection would be “conditional” on this rulership remaining “for the good of all,” with the people being vigilant as to “whether the will of the Prince really is the general will, a question of which the People is the sole judge” (GM 102; 3:315). Yet these assurances assume that the people can exercise independent judgment and can recognize their good outside the lawgiver's persuasion. But as we have seen, in the “Manuscript” his powers allow for no such independence, no Archimedean point for evaluation or critique. Everything the people know of the general will has been conditioned by the Legislator and his laws.

This dilemma highlights two of the text's essential features. First, it reveals Rousseau's increasing interest in citizen agency. Earlier in the text he had, for the first time, identified a general will “in each individual” (GM 79; 3:286) such that “each individual can, as a man, have a particular will contrary to or differing from the general will he has as a Citizen” (GM 84; 3:364).Footnote ³³ In the chapter on the Legislator he asserts that “today's law should not be an act of yesterday's general will, but of today's, and we have engaged ourselves to do not what everyone has willed, but what everyone now wills” (GM 103; 3:316). This last phrase—“what everyone now wills”—underscores the text's new perspective. In the Encyclopédie, neither Diderot nor Rousseau had shown interest in personal volition. Now Rousseau's citizens possess and enact their general wills. These notes of personal agency are rare in the “Manuscript” and sit awkwardly beside the powers of the Legislator, but when carried through to the published edition they form a familiar refrain.

Second, Rousseau must have found defective the draft's epistocratic potential. He discards long passages invoking an implicit general will: the early chapter challenging Diderot's formulation (GM I.2); the section claiming that the general will is known only through law (GM I.7); the treatment of the lawgiver just discussed (GM II.2); a passage from II.4 introducing a “rational natural right” that provides a “fundamental and universal law of the greatest good of all” (GM 113–14; 3:329–30). Each edit cuts off a potential route by which the general will could be known outside deliberation.

We find further evidence in the material added or reworked for publication. Even Rousseau's minor edits suggest greater individual agency: the revised “fundamental problem” adds that each “obeys only himself,” and the contract no longer requires putting one's will under the general will.Footnote ³⁴ The discussion of alienated sovereignty in the Social Contract now reads as if written to explicitly reject epistocracy: “I say, therefore, that sovereignty, being only the exercise of the general will, can never be alienated, and that the sovereign, which is only a collective being, can only be represented by itself. Power can perfectly well be transferred, but not will” (SC 145; 3:368). In the “Manuscript” this chapter (I.4) had raised the possibility of epistocratic rule; now the Social Contract cuts off such possibilities. This new formulation remains compatible with Rousseau's preference for aristocratic or epistocratic government.Footnote ³⁵ But while governing authority can be transferred, sovereignty cannot, for if “the people promises simply to obey, it dissolves itself by that act; it loses the status of a people. The moment there is a master, there is no longer a Sovereign, and from then on the body politic is destroyed” (SC 145; 3:269). Whatever his doubts about democratic government, Rousseau's views on sovereignty are thoroughly democratic.

Finally, consider the Legislator. In the Social Contract we find his task somewhat softened: he no longer must take “all” men's forces or “mutilate” their constitutions.Footnote ³⁶ With Rousseau's removal of the passage from Political Economy on law, the lawgiver's purview extends only to making society well ordered, helping citizens generalize their wills and grasp their interdependence. Emphasizing this further circumscription of authority, Rousseau adds to the published edition that “even the Decemvirs never took upon themselves the right to have any law passed solely on their authority. Nothing that we propose, they said to the people, can become law without your consent” (SC 156; 3:382; emphasis in original). These revisions do not erase the Legislator's power, but they do buttress the constraints on his influence. After the founding, citizens must discern their good through deliberation alone.

In sum, the “Manuscript” offers a suggestive snapshot of Rousseau's evolving uses for general wills. The text shows far more attention to citizen agency than had Political Economy and marks Rousseau's shift to a contractarian account of legitimacy. Like the Social Contract, it asserts that citizens must know and will their common good. But where the “Manuscript” relies exclusively on the Legislator and his decrees to provide such knowledge, the Social Contract will describe a system of deliberations constrained to generality.

4. The Personal Will of the Social Contract

The distance between the “Manuscript” and the Social Contract can be measured in a single clause: where the former reads “The general will is always right; it is never a question of correcting it. But it is necessary to know how to consult it properly” (GM 99; 3:311), the latter claims that “The general will is always right, but the judgment that guides it is not always enlightened” (SC 154; 3:390). Here the general will derives not from wise laws, but from personal judgments formed and tempered in deliberation.

The Social Contract goes further than the “Manuscript” in examining the general will that each citizen finds “within himself” (SC 199; 3:438). Indeed, here Rousseau grounds legitimacy exclusively in the public declaration of personal general wills. It also sketches the institutions needed to publicly declare a general will.Footnote ³⁷ Rousseau's new position is not without costs. He must show how fallible citizens can consult and will their general wills, and must name the institutional and social constraints powerful enough to ensure that the deliberations of a well-ordered polity always enact a general will. His account proceeds in two stages: first showing how the general will can in principle be derived from individual interests; then identifying the institutions needed for citizens to carry out such work in practice. Indeed, Rousseau grounds these arguments entirely in common interests. Once a society is well ordered, the people need neither natural laws nor epistemic authorities for the ongoing declaration of the general will. The deliberative settings outlined in the Social Contract allow citizens to refine their judgment, express their general wills, and experience moral freedom.

Rousseau begins by modeling the implicit general will. His infamous mathematical metaphor attempts to bring analytical rigor to the implicit will posited in Political Economy. The task is to find what individuals objectively share: “There is often a great difference between the will of all and the general will. The latter considers only the common interest; the former considers private interest, and is only a sum of particular wills. But take away from these same wills the pluses and minuses that cancel each other out, and the remaining sum of the differences is the general will.”Footnote ³⁸ Suppose that a polity needs a turnpike and must decide how to pay for it. Each citizen's will has two elements: a particular will bent toward free riding, and a general will favoring an equitably shared burden. From these we can infer the outcomes of various decision procedures. Begin with the “will of all,”Footnote ³⁹ which considers only each person's desire to free ride. Under pure majority rule, the first proposal to let most people free ride will win. But suppose that we put each particular will to a vote: Citizen Jones proposes that she free ride at our expense, Citizen Perez proposes that he do the same, and so on. So long as everyone else continues to vote in their own interest—and forms no coalitions—all such proposals will fail. Particular wills thus “cancel” each other out. Only the set of general wills remains. These differ only in indexicality: I want the equitable burden primarily for my benefit, you want it for your benefit, and so on. Sum these small remaining differences to find the polity's general will.

Yet such abstractions are useless for real politics. The implicit general will cannot be made explicit. No citizen has the omniscience to deduce it, and no decision procedure could reliably express it.Footnote ⁴⁰ Rousseau's solution to the fundamental problem must fit the limits of human understanding and will. He thus elaborates, in passages scattered throughout books II and IV, a model of deliberation in which the general will is expressed when and only when citizens pronounce judgment on whether a proposal is good for each. Citizens must discern the common good for themselves and have strength of will to vote for it. Procedural constraints aid this effort by promoting generality:

Rousseau summarizes in book IV: “Each one expresses his own opinion . . . by voting, and the declaration of the general will is drawn from the counting of the votes” (SC 201; 3:441). A fuller understanding of these compressed comments requires that we examine in more detail the nature of the common good, the quality of citizens’ opinions, and the purposes of deliberation.

Rousseau finds that “what generalizes the will is not so much the number of votes as the common interest that unites them” (SC 149; 3:374). He never defines the “common good” or “common interest” and uses the two terms interchangeably. He claims that “the greatest good of all” consists in “these two principal objects: freedom and equality” (SC 162; 3:391; emphasis in original), but this is too nebulous, and the “greatest good” (le plus grand bien) suggests a ranking absent from the “common good” (le bien commun). Far more helpful are his comments on how common interests are perceived: “For if the opposition of private interests made the establishment of societies necessary, it is the agreement of these same interests that made it possible. It is what these different interests have in common that forms the social bond, and if there were not some point at which all the interest are in agreement, no society could exist” (SC 145; 3:368). The common good denotes the set of shared interests. It is perhaps a small set—Rousseau remarks in Émile that “what particular interests [intérets particuliers] have in common is so slight that it will never outweigh what sets them in opposition”Footnote ⁴¹—but the force of these interests is enough to make the contract both desirable and feasible. That contract sets the limit of legitimate law at the common good. Laws must be good for each.

Citizens opine on whether a proposal “does or does not conform to the general will that is theirs” (SC 201; 3:441). I have been using “discernment” to designate the faculty needed to judge these matters, since to evaluate conformity, one must first be a competent judge of one's own good and that of others. My usage follows Rousseau's choice of discerner to denote such powers of discrimination; we have seen how the “Manuscript” makes physical security a prerequisite of the ability “to discern either good or evil, either the decent man or the wicked one” (GM 77; 3:282).Footnote ⁴² In turning to politics, Political Economy had stressed the near impossibility—even in civil society—of making such distinctions, which are “always extremely difficult to make and on which only the most sublime virtue can shed sufficient light” (PE 145; 3:247–48). The Social Contract adds nuance. Prior to the Legislator's work, man “has difficulty perceiving the advantages he should obtain from the continual deprivations imposed by good laws” (SC 156; 3:383). After society is well ordered—once the general will becomes the ordering principle of law—conditions change. Under ideal conditions the common good “is clearly apparent everywhere, and requires only good sense to be perceived” (SC 198; 3:437). Yet even when the state is near collapse, when “the basest interest brazenly adopts the sacred name of the public good” (SC 198; 3:438), some element of discernment persists. The citizen “sees perfectly well” that his good is bound to others’ and therefore “wants the general good in his own interest” (SC 199; 3:438). The contents of that general good have simply become harder to know.

Discernment, then, requires certain epistemic conditions. Rousseau's formal constraints provide just such an environment. To know what is good for each, one must have trustworthy data. When citizens present opinions in good faith, each draws on knowledge of their own good, the good of their kith and kin, and the common good. The “large number of small differences” maximizes the data available to deliberators. If factions form, however, “there are no longer as many voters as there are men, but merely as many as there are associations” (SC 147; 3:371–72). Discrete opinions are cloaked by coalitions. Votes are now decided not by collective judgment but by numerical force. Partial societies degrade the available information and must be suppressed, even to the point of restricting communication among deliberators.

The “no communication” clause has been the source of much confusion. Interpreters often take this to require something approaching silence; perhaps Rousseau even hopes to banish discourse entirely.Footnote ⁴³ Nothing in the present argument hinges on whether political discussion occurs within or without the assembly; it is enough that Rousseau makes the assembly a site of citizen judgment and agency.Footnote ⁴⁴ But the Social Contract and later works resist an interpretation that restricts all communication. Rousseau is not averse to dialogue. Indeed, after arguing against factions, he proceeds in the next chapter to treat délibérations and discussion in parallel (SC 149; 3:374). Cohen has highlighted Rousseau's indicative, if not dispositive, comments decrying Genevan restrictions on the right of citizens to voice opinions.Footnote ⁴⁵ And in a footnote to his subsequent Letters Written from the Mountain, Rousseau defines his terms so as to require dialogical communication:

Deliberation names the entire process of collective “weighing,” from the “first round” of reason giving to the “last round” of voting, just as juries deliberate both in weighing evidence and in reaching a verdict. (Vote tallies, as we see below, become a key input for personal judgment.) Notably absent in all these passages is room for the negotiations by which factions might form, which suggests that the “communication” to be restricted is that which is needed to trade votes.Footnote ⁴⁶ Inner reflection, the sharing of reasons, and voting are the only aims of Rousseau's assemblies.

Armed with a better grasp of how Rousseau employs these concepts, we can now turn to how formal procedures convert the general wills of citizens into the declared will of the polity. Rousseau's analysis is clearest when showing how these procedures fail. He identifies two errors in book IV. The first is epistemic. Assemblies are asked whether a given motion is “advantageous” to the body politic (SC 199; 3:438) and, as we have seen, under ideal circumstances that advantage is “clearly apparent everywhere.” But conditions are not always so favorable. Disagreements arise. Supposing that I am outvoted, can I simultaneously “be free and forced to conform to wills that are not [my] own?” Rousseau rejects the question for misunderstanding the deliberative exercise:

Since the assembly aims only to establish commonality—and not matters beyond citizen competence, such as the nature of justice or optimal public policies—Rousseau finds competent majorities to be infallible. If everyone discerns well and wills generally, the vote will be unanimous. In less ideal conditions the competent majority decides the question, not from numerical might, but because no guide is more reliable. Observe how even the character of the deliberation provides useful data: “The more harmony there is in assemblies, that is, the closer opinions come to obtaining unanimous support, the more dominant as well is the general will.” But if the assembly faces long and tumultuous debates—here we should note that Rousseau's concern is the tenor of debates and not their presence—one has good reason to suspect “the ascendance of private interests” (SC 199; 3:439).

Voting does not simply express citizens’ discernment. The vote also refines their judgment. The declared and personal general wills are coconstitutive: discernment informs the personal general wills that determine the public declaration, and that declaration in turn shapes how citizens perceive the common good. Deliberators update their prior beliefs. This explains why, in a rarely noted line from book II, Rousseau claims that “there can never be any assurance that a particular will conforms to the general will until it has been subjected to the free votes of the people” (SC 156; 3:383). The knowledge gained from a vote (to say nothing of a series of votes) would be indispensable, particularly in the absence of modern opinion polling. The outcome of the vote itself has epistemic value. The democratic mind introduced in the Social Contract informs—and is informed by—public reason.

Book IV addresses a second error in which weak wills lead to selfish voting. Here citizens know rightly but act wrongly. Rousseau gives this volitional error a procedural solution:

Even here, at the end of the civil state, discernment persists. The wayward citizen knows that his well-being depends on the common good. His error is letting his particular will dominate. In book I, Rousseau had notoriously asserted that the free rider who “comes to view what he owes the common cause as a free contribution, the loss of which will harm others less than its payment burdens him,” must be “forced to be free” (SC 141; 3:364). Book IV, in revisiting this error, locates the fault in a weak will. Rousseau's remedy is to remind citizens to vote only on whether a motion is “advantageous to the State,” helping to ensure that general will of citizens “is always questioned [and] always answers” (SC 199; 3:438).

This passage is remarkable for capturing all three of Rousseau's conceptions. The constancy of the implicit general will comes not from natural law or Platonic truth but from inextinguishable personal wills. The citizen “wants the general good in his own interest just as vigorously as anyone else” because the common good is definitionally good for each. This ensures an upright declaration when the question is properly put to well-intentioned citizens. The fixedness of the personal general will also explains why Rousseau's solution to selfish voting requires nothing more than enforcing the citizen's previous agreement to will generally, binding him like Ulysses to his chosen freedom.

In these passages from book IV, Rousseau laments self-interested voting, echoing an earlier claim that “nothing is more dangerous than the influence of private interests on public affairs” (SC 173; 3:404). Yet in book II he grounds all political right in self-interest: “Why is the general will always right and why do all constantly want the happiness of each, if not because there is no one who does not appropriate this word each to himself, and does not think of himself as he votes for all?” (SC 148–49; 3:373; emphasis in original). We may reconcile these comments by noting where the text attends to preferences rather than judgments. The error discussed in book IV is a preference-based vote in which individuals ask whether “it is advantageous to a given man or to a given party for a given motion to pass” (SC 199; 3:438). The resulting tally would be the disastrous will of all, the sum of private preferences. By contrast, in book II (and again in book IV) Rousseau concerns himself with the sources of judgment. In reasoning about the common good, we necessarily refer to our own interests: How could a proposal be good for all if it is not also good for me? Our experiences provide the anecdata that inform the judgments pooled in voting. That vote needs no veil of ignorance: indeed I must know my interests to discern correctly. Rousseau suggests that when I use it well, self-interest can aid my discernment. Only when it bends toward avarice do I err.

At stake in these moments of discernment and willing is nothing less than our freedom. This is obviously true of civil freedom, which requires that the contract be maintained. Less conspicuous are the ties to moral freedom. Rousseau names this freedom only once in the Social Contract, describing how the contract allows for inner mastery: “The impulse of appetite alone is slavery, and obedience to the law one has prescribed for oneself is freedom” (SC 142; 3:365). Book IV ties this freedom to deliberation so closely that the moment one loses discernment, one is enslaved: “If my private opinion had prevailed, I would have done something other than what I wanted. It is then that I would not have been free” (SC 201; 3:441).Footnote ⁴⁸ When either ignorance or selfishness undermine my general will, I not only subvert my deeper desire to see the common good realized but also abandon moral freedom. Discernment allows us to obey—in both a political and moral sense—only the law we give ourselves.

This is Rousseau's new model of the democratic mind. The exercise of democratic sovereignty—the assembling, the voting—is built on a “union of understanding and will” (SC 154; 3:380) in which citizens define their good and enact general laws without deference to universal justice or epistocratic authorities. In doing so they will themselves free. The Legislator of the Social Contract is reduced to the preparatory role of a “mechanic who invents the machine” (SC 155; 3:381); once a society is well ordered, the people decide legitimacy.Footnote ⁴⁹ This gives them “their will alone as rule” and leaves them as free as before (LM 232; 3:807). Arthur Melzer has observed that for Rousseau, the “whole activity of voting in the popular assembly is nothing but a means (if also the only and essential means) for expressing, consulting, trying to make contact” with our general wills.Footnote ⁵⁰ We can now add that democratic processes help form those wills as well.