¹ Parenthetical citations of Rousseau's works are to the following translations, followed by the volume and page number of the Pléiade Œuvres complètes: Of the Social Contract (SC), Discourse on Political Economy (PE), Considerations on the Government of Poland (GP), in The Social Contract and Other Later Political Writings, ed. Victor Gourevitch (Cambridge: Cambridge University Press, 1997); Second Discourse (SD) in The Discourses and Other Early Political Writings, ed. Victor Gourevitch (Cambridge: Cambridge University Press, 1997); Emile, or On Education (E), trans. Allan Bloom (New York: Basic Books, 1979); Political Fragments (PF), in The Collected Writings of Jean-Jacques Rousseau, vol. 4, ed. Christopher Kelly and Roger D. Masters (Hanover, NH: University Press of New England, 1994); Plan for a Constitution for Corsica (CC), in The Collected Writings of Jean-Jacques Rousseau, vol. 11, ed. Christopher Kelly and Roger D. Masters (Hanover, NH: University Press of New England, 1990).

² Rousseau does not provide a formal definition of the term “partial societies.” In a footnote referring to Machiavelli's History of Florence, he argues that not all forms of pluralism (what he calls “divisions”) are harmful. He classifies as harmful those that are “accompanied by factions and parties” and as beneficial those that are not (SC, 60; 3:372). I use the term “pluralism” to refer to neutral or positive divisions and “factions” to refer to the harmful divisions. I mainly employ “partial societies” when citing Rousseau with the caveat that the term may include both types of divisions.

³ Williams, David Lay, Rousseau's “Social Contract”: An Introduction (Cambridge: Cambridge University Press, 2014), 171Google Scholar. A welcome exception to this trend is Valentina Arena, “The Roman Republic of Jean-Jacques Rousseau,” History of Political Thought 37, Special Issue (2016): 1–31. Her paper contextualizes Rousseau's use of Rome as a way to respond to contemporaneous debates about separation of powers, particularly Montesquieu's discussion of Rome in the Spirit of the Laws. This article largely agrees with Arena's historically informed account, while defending the relevance of book 4 independently of the specific debates about Rome in eighteenth-century France.

⁴ Kelly, Christopher, “Sovereign versus Goverment: Rousseau's Republicanism,” Acta Politologica 10, no. 2 (2018): 22Google Scholar. Kelly's article offers a complementary account to mine. The only other sustained discussion of Rome has been highly critical, accusing Rousseau of oligarchic tendencies. See McCormick, John P, “Rousseau's Rome and the Repudiation of Populist Republicanism,” Critical Review of International Social and Political Philosophy 10, no. 1 (2007): 23CrossRefGoogle Scholar.

⁵ Shklar, Judith, Men and Citizens: A Study of Rousseau's Social Theory (Cambridge: Cambridge University Press, 1969)Google Scholar, 14 and “Rousseau's Two Models: Sparta and the Age of Gold,” Political Science Quarterly 81, no.1 (1966): 25–51.

⁶ Masters, Roger D., The Political Philosophy of Rousseau (Princeton: Princeton University Press, 1968), 351Google Scholar.

⁷ Melzer, Arthur, The Natural Goodness of Man (Chicago: University of Chicago Press, 1990), 79CrossRefGoogle Scholar.

⁸ Riley, Patrick, “A Possible Explanation of Rousseau's General Will,” American Political Science Review 64, no. 1 (1970): 87CrossRefGoogle Scholar.

⁹ Crocker, Lester G., Rousseau's “Social Contract”: An Interpretive Essay (Cleveland: Press of Case Western Reserve University, 1968), 49Google Scholar; Bloom, Allan, “Rousseau's Critique of Liberal Constitutionalism,” in The Legacy of Rousseau, ed. Orwin, Clifford and Tarcov, Nathan (Chicago: University of Chicago Press, 1997), 148Google Scholar.

¹⁰ Kelly, Christopher, Rousseau as Author: Consecrating One's Life to Truth (Chicago: University of Chicago Press, 2003), 122–24Google Scholar.

¹¹ Shklar, Men and Citizens, 13; Trachtenberg, Zev M., Making Citizens: Rousseau's Political Theory of Culture (London: Routledge, 1993), 211–13CrossRefGoogle Scholar. Others see Sparta as partly a genuine ideal, but one only appropriate for a minority of small European city-states such as Rousseau's home country of Geneva and the island of Corsica. See Melzer, Natural Goodness of Man, 270–71.

¹² Putterman, Ethan, “Realism and Reform in Rousseau's Constitutional Projects for Poland and Corsica,” Political Studies 49, no. 3 (2001): 481–94CrossRefGoogle Scholar; Hanley, Patrick Ryan, “Enlightened Nation Building: The ‘Science of the Legislator’ in Adam Smith and Rousseau,” American Journal of Political Science 52, no. 2 (2008): 219–34CrossRefGoogle Scholar; Schaeffer, Denise, “Realism, Rhetoric and the Possibility of Reform in Rousseau's Considerations on the Government of Poland,” Polity 42, no. 3 (2010): 377–97CrossRefGoogle Scholar, and “Attending to Time and Place in Rousseau's Legislative Art,” Review of Politics 74, no. 3 (2012): 421–41CrossRefGoogle Scholar.

¹³ This point has been noted ever since Benjamin Constant distinguished modern liberty from the highly intrusive form of ancient political liberty: “Among the Spartans, Therpandrus could not add a string to his lyre without causing offense to the ephors.” See Constant, Benjamin, “The Liberty of the Ancients Compared with That of the Moderns,” in Constant: Political Writings, ed. and trans. Fontana, Biancamaria (Cambridge: Cambridge University Press, 1988), 311Google Scholar.

¹⁴ Riley, “A Possible Explanation”; Riley, Patrick, “Rousseau's General Will,” in The Cambridge Companion to Rousseau, ed. Riley, Patrick (Cambridge: Cambridge University Press, 2001), 124–53CrossRefGoogle Scholar.

¹⁵ Williams, Rousseau's “Social Contract”; Williams, David Lay, “The Substantive Elements of Rousseau's General Will,” in The General Will: The Evolution of a Concept, ed. Farr, James and Williams, David Lay (Cambridge: Cambridge University Press, 2015), 219–46CrossRefGoogle Scholar. For the most extensive reconstruction of both formal and substantive conditions for the general will, see Cohen, Joshua, Rousseau: A Free Community of Equals (New York: Oxford University Press, 2010)CrossRefGoogle Scholar.

¹⁶ Bertram, Christopher, “Rousseau's Legacy in Two Conceptions of the General Will: Democratic and Transcendent,” Review of Politics 74, no. 3 (2012): 403CrossRefGoogle Scholar. This view receives its most extensive elaboration in Sreenivasan, Gopal, “What Is the General Will?,” Philosophical Review 109, no. 4 (2000): 545–81CrossRefGoogle Scholar.

¹⁷ Bertram, “Rousseau's Legacy,” 403.

¹⁸ Crocker, Rousseau's “Social Contract, ” 70; Cullen, Daniel E., Freedom in Rousseau's Political Philosophy (DeKalb: Northern Illinois University Press, 1993), 120Google Scholar; Kelly, Rousseau as Author, 125–27.

¹⁹ For this mechanism to work, Rousseau tells us that, among other conditions, the people must be “adequately informed” and the citizens should have “no communication among themselves” (SC, 60; 3:371). The no-communication requirement has been the subject of controversy. See Grofman, Bernard, Feld, Scott L., Estlund, David M., and Waldron, Jeremy, “Democratic Theory and the Public Interest: Condorcet and Rousseau Revisited,” American Political Science Review 83, no. 4 (1989): 1317–40Google Scholar. For a summary of Rousseau's claims about deliberation and voting in the Social Contract and the Letters from the Mountain, see Kelly, Rousseau as Author, 119–22.

²⁰ As Rousseau puts it when returning to the discussion of suffrage in SC 4.2: “This presupposes, it is true, that all the characteristics of the general will are still in the majority: once they no longer are, then regardless of which side one takes there no longer is any freedom” (SC, 124; 3:441).

²¹ Grofman, Bernard and Feld, Scott L., “Rousseau's General Will: A Condorcetian Perspective,” American Political Science Review 82, no. 2 (1988): 567–76CrossRefGoogle Scholar; Grofman, , Feld, , Estlund, , and Waldron, , “Democratic Theory and the Public Interest”; Barry, Brian, Political Argument (London: Routledge and Kegan Paul, 1965)Google Scholar.

²² The standard version of this theorem relies on four strict assumptions: (1) the decision involves a binary choice between two mutually exclusive outcomes (e.g., guilty/not guilty, ratify/reject, candidate A/candidate B), one of which is better than the other according to a shared standard; (2) votes are statistically independent (i.e., the probability of voter 1 choosing option A is independent of the probability of voter 2 choosing the same option); (3) voters have identical competence; (4) voter competence is higher than 0.5 (i.e., each voter has a higher than random or 50-50 chance of choosing the correct outcome). For a more extensive discussion of the application of the CJT and its extensions to Rousseau's Social Contract, see Oprea, “Voting in the Roman Republic: An Alternative Model for Epistemic Democrats” (unpublished manuscript).

²³ Kelly, Rousseau as Author, 120.

²⁴ Schwartzberg, Melissa, “Voting the General Will: Rousseau on Decision Rules,” Political Theory 36, no. 3 (2008): 403–23CrossRefGoogle Scholar.

²⁵ Although Rousseau mentions the Athenian legislator Solon alongside Numa and Servius, he never investigates the details of the Athenian model in the Social Contract. In the First Discourse, Athens appears as a paradigmatic example of the corruption produced by the arts and sciences.

²⁶ Shklar, Men and Citizens, 31.

²⁷ Ibid.; SC 4.7.

²⁸ Melzer, Natural Goodness of Man, 95. He adds that “his [Rousseau's] model here, as in most things, is Sparta” and cites the Second Discourse for the uniqueness of the Spartan model “where the law attended principally to the education of children and where Lycurgus established morals which almost allowed him to dispense with adding law” (95); see also SD, 182; 3:187–88.

²⁹ Kelly, Rousseau as Author, 76.

³⁰ Rousseau certainly admired both Sparta and its legislator. Rousseau describes Lycurgus as a quasi-divine figure who could “work in one century and enjoy the reward in another” (SC, 68–69; 3:381).

³¹ John T. Scott describes the role of the legislator as “a higher form of agenda-setting” where one works to “create the restrictions on the domain of preferences.” See Scott, John T., “Rousseau's Anti-Agenda-Setting Agenda and Contemporary Democratic Theory,” American Political Science Review 99, no. 1 (2005): 143CrossRefGoogle Scholar.

³² Crocker, Rousseau's “Social Contract, ” 13.

³³ Herodotus, , The History, trans. Greene, David (Chicago: University of Chicago Press, 1987), 625CrossRefGoogle Scholar; Plutarch, Life of Lycurgus, in Plutarch's Lives, ed. Arthur Hugh Clough, trans. John Dryden (New York: Modern Library, 2001), 59. For a fuller account of ancient Spartan demographics and the decline in the number of full citizens, see Cartledge, Paul, The Spartans: The World of the Warrior-Heroes of Ancient Greece, from Utopia to Crisis and Collapse (New York: Overlook, 2003)Google Scholar and Figueira, Thomas J., “Population Patterns in Late Archaic and Classical Sparta,” Transactions of the American Philological Association, no. 116 (1986): 165–213CrossRefGoogle Scholar.

³⁴ For a list of Rousseau's ancient and contemporary sources concerning Rome, see Arena, “Roman Republic of Jean-Jacques Rousseau,” 18–24.

³⁵ Plutarch, Life of Lycurgus, 74.

³⁶ In Emile, Rousseau suggests that any educational project must choose between raising a man and a citizen (E, 39; 4:248). On the face of it, the education of Emile is that of a man. However, identifying two mutually exclusive models of political institutions complicates this commonly employed dichotomy in Rousseau's work. Scholars have long drawn a sharp contrast between the simplicity of natural man and the unnatural subordination to the community characteristic of the civil man. While this contrast is justified, the plurality of models of citizenship allows for the possibility that Emile's education could be compatible with a Roman rather than Spartan political model. Unfortunately, a full exploration of the connections between the Roman model and Emile is beyond the scope of this article.

³⁷ In fact, Rousseau describes Helot relations as a subspecies of Spartan foreign policy given the implied state of war between Spartans and Helots (SW, 176; 3:608). Although slavery was also extensively practiced in Rome, the enslaved population did not constitute a permanently enslaved racial group. Instead, slaves could earn their freedom and even acquire rights of citizenship.

³⁸ Rousseau says that “at the same time as he remedied the existing inequality he forestalled its future recurrence” (SC, 128; 3:445).

³⁹ He does, however, note that such a method would be unlikely to apply to a contemporary context given the high degree of social stability and moral virtue required to prevent it from producing disastrous consequences: “Where is the modern people whose devouring greed, unsettled spirit, intrigue, constant comings and goings, perpetual revolutions of fortune would let such an establishment last twenty years without overthrowing the whole state?” (SC, 131; 3:448). Unlike the division by tribes, which both multiplied the number of partial societies and equalized their power, the division by centuries left itself more vulnerable to factions through the unequal political power it bestowed on the wealthy.

⁴⁰ One issue to note in assessing the Roman voting assemblies is that they simultaneously fulfilled the role of sovereign and government. Rousseau called the latter role a “usurpation” (SC, 132; 3:449–50). His discussion of the two assemblies and their voting rules therefore includes both matters of political right (i.e., sovereign assemblies with universal suffrage) and maxims of government (e.g., the election of specific magistrates) without drawing explicit distinctions with respect to the relevant voting rules. For the purposes of this article, I follow his practice.

⁴¹ “But it must be noted that here the corporate interest begins to guide the public force less in accordance with the standard of the general will” (SC, 93; 3:407).

⁴² It is this feature that leads McCormick to accuse Rousseau of oligarchic tendencies. See McCormick, “Rousseau's Rome,” 23. Arena defends Rousseau's preference for the comitia centuriata as grounded in his defense of undivided sovereignty against Montesquieu's separation of powers. Her discussion of the means to correct the imbalance of power resembles this one. See Arena, “Roman Republic of Jean-Jacques Rousseau,” 13–15.

⁴³ According to Rousseau, “the censorship, stronger than this institution, corrected for its vice” (SC, 131; 3:448). Another check on the wealthy was the use of the lottery to determine which of the centuries would cast the first (publicly announced) ballot (i.e., the centuria praerogativa). For a discussion of debates concerning the democratic nature of the centuria praerogativa, see Arena, “Roman Republic of Jean-Jacques Rousseau,” 20–22.

⁴⁴ As a legislator, the work of Servius is dually removed from the explicit attention of the public. First, the electoral consequences of the divisions are difficult to discern in advance. Second, the military justification of the division further disguised the gerrymandering intentions: “In order that the people might less readily discern the consequences of this last form, he pretended to give it a military cast” (SC, 130; 3:447).

⁴⁵ Arena makes a similar point about the importance of morals and the censorship, but focuses on the role of the censors in influencing public opinion rather than their influence over the shape of electoral districts. See Arena, “Roman Republic of Jean-Jacques Rousseau,” 14.

⁴⁶ “No theoretical checks, no form of government, can render us secure. To suppose that any form of government will secure liberty or happiness without any virtue in the people, is a chimerical idea” (James Madison, speech of June 20, 1788, in The Debates in the Several State Conventions on the Adoption of the Federal Constitution … and Other Illustrations of the Constitution, ed. Jonathan Elliot, 2nd ed. [1836], 3:537, http://oll.libertyfund.org/titles/1907, accessed April 22, 2018).

⁴⁷ One concern is the potential corruption of the people that comes from turning its attention from matters concerning law to questions about the execution of the laws in particular circumstances: “Nothing is more dangerous than the influence of private interests on public affairs, and abuse of the laws by the government is a lesser evil than the corruption of the lawgiver, which is the inevitable consequence of particular considerations” (SC, 91; 3:404). In the case of Athens, another model of multiplying the number of partial societies and controlling the inequalities among them, the constant preoccupation of the people with elections, bestowing honors, ostracisms, and passing particular degrees, including degrees concerning foreign policy, resulted in corrupting the sovereign: “the people no longer had a general will properly so called; it no longer acted as a sovereign but as a magistrate” (SC, 62; 3:374). There are a number of relevant differences between Athens and Rome explaining why this disadvantage was more problematic for the Athenians. A full discussion is beyond the scope of this article.

⁴⁸ Another concern Rousseau discusses is the influence of populist leaders. Rousseau follows Machiavelli in noting the use of Roman religion (i.e. auguries and prodigies) for political purposes. In this case, Rousseau notes the senate's use of religion to thwart the influence of populists such as the Gracchi.

⁴⁹ Rousseau claims that the assembly by curiae did not include every eligible Roman: “although every citizen was enrolled in a tribe, far from everyone was enrolled in a curia” (SC, 130; 3:447).

⁵⁰ Former slaves were also generally enrolled in the urban tribes.

⁵¹ Arena notes that although Rousseau's Rome includes a number of historical inaccuracies, his claims about the assignment of freedmen to urban tribes, the lower prestige of those tribes, and the changes in the significance of tribal membership were generally accurate. See Arena, “Roman Republic of Jean-Jacques Rousseau,” 11, 15.

⁵² While Rousseau does not fully spell out the basis for his critique of the censors, this explanation offers a plausible interpretation to an otherwise puzzling comment in the text.

⁵³ Of course, he does leave open the possibility that such a rule cannot be found, but such an interpretation of the Social Contract would require an explanation for why Rousseau decided to provide so much rich detail, both theoretical and empirical.

⁵⁴ Kelly, “Sovereign versus Government,” 29.

⁵⁵ As Rousseau puts it in a fragment titled “Parallel between Sparta and Rome”: “Ever ready to die for his country, a Spartan loved the fatherland so tenderly that he would have sacrificed freedom itself to save it. But the Romans never imagined that the fatherland could outlive freedom or even glory” (PF, 63; 3:543).

⁵⁶ For further information, see Williams, Rousseau's “Social Contract, ” 26–34.

⁵⁷ Shklar, Men and Citizens, 155.

⁵⁸ Schaeffer, “Attending to Time and Place,” is right to argue that even what I have been calling the Spartan model requires a form of social engineering that goes beyond indoctrination to develop political judgment and adaptability in the face of unexpected circumstances.

⁵⁹ Ibid.; Putterman, “Realism and Reform.”

⁶⁰ For a further reason to suspect that the Corsican model relies primarily on eliminating pluralism, one can turn to Rousseau's foreword. Here, Rousseau contrasts institutional “mechanisms” which focus on “chains” and “shackles” to keep governments from abusing their power with the type of nation building that shapes the character of a people: “Nevertheless, there is something much better to do, that is to form the nation for the government” (CC, 123; 3:901).

⁶¹ Williams, David Lay, ”Modern Theorist of Tyranny? Lessons from Rousseau's System of Checks and Balances,” Polity 37, no. 4 (2005): 443–65CrossRefGoogle Scholar.

⁶² He calls it a “truly grand and noble luxury, to the inconvenience of which I am fully sensible, but which at least elevates souls instead of debasing them, gives them sentiments, resilience, and which was not abused among the Romans as long as the Republic endured” (GP, 188; 3:965). Note that Rousseau suggests manumission and eventual incorporation of the serfs into the government, which would require further imitation of the Romans in the design of electoral institutions.

⁶³ “Now the law, which is but the expression of the general will, is indeed the resultant of all of the particular interests combined and in balance of their large number. But corporate interests, because of their excessive weight, would upset the balance, and should not be included in it collectively” (GP, 206; 3:984).