A Shared Capacity Account of Rousseau's General Will

Introduction

Rousseau proposes the idea of the general will as an answer for a problem regarding humans’ interdependence. Insofar as we depend upon others’ cooperation to meet our needs, we are subject to their wills and hence seemingly unfree. Rousseau suggests, though, that each person can enjoy the benefits of society and “nevertheless obey only himself and remain as free as before.”¹ The key is to be ruled by the general will. If all are subject only to the general will, and if the general will is the will of each citizen, then each citizen is subject only to his own will—and therefore free.

Despite the centrality of the general will to Rousseau’s political philosophy, and despite its continued influence on contemporary political philosophy, there is little agreement about what it is. Opinions vary sufficiently that some even question whether Rousseau is, in the title of one book, “Totalitarian or Liberal.”² In the background of these debates are two strands of thought in the Social Contract, each suggesting different accounts of the general will. On the one hand, Rousseau suggests that the content of the general will is determined procedurally: the general will is simply whatever people vote for when voting is carried out correctly. On the other hand, Rousseau says that the general will wills the common good, and the common good seems to be fixed independently of the voting procedure. Because it is difficult to see how any voting procedure could be guaranteed to will the common good,³ it is difficult to see how these two strands in Rousseau’s thinking relate.

Recent interpretations tend to take one strand as more fundamental and then accommodate the other as best they can. I call procedural accounts those that define the general will as a certain voting procedure.⁴ I call common good accounts those that define the general will in terms of the common good, independently of any vote.⁵ One popular group of common good accounts specifies that this understanding of the common good must be publicly shared.⁶

The privileging of one of the two strands of Rousseau’s claims about the general will sometimes stems from the assumption that one must be more fundamental. Evidence against one strand’s adequacy by itself is taken as evidence against that strand’s being fundamental,⁷ perhaps because taking the conjunction of each as fundamental seems ad hoc.

In this essay, however, I shall argue that there is nothing ad hoc about Rousseau’s invoking both procedure and common good, for in fact both are necessary to constitute a shared will. A will does not consist of just any decision-making procedure, nor is it simply a conception of the good. Rather, a will is a capacity to decide what to do in accordance with a conception of the good.⁸ Hence, a will consists of both a conception of the good and a process for determining how to act on that conception. Inasmuch as the general will is a will, it must have both conception and process. In Rousseau’s general will, I argue, the voting procedure is the process for deciding how to act on the conception of the common good. The general will’s content is not simply a conception of the common good or the result of a procedure; it is the result of a procedure insofar as it is carried out in accordance with a conception of the common good. This view unifies the two lines of thinking under a single conception of a will.

I call my view the shared capacity account: the general will is a shared capacity for self-determination in accordance with a publicly shared conception of the common good. To be governed by the general will is to be governed by the exercise of this shared capacity. “Shared” here ought not be understood in a loose sense, as when two people who can both whistle are said to share that ability.⁹ I mean the term strictly, as when we say that Supreme Court justices share the ability to decide cases.¹⁰ I can exercise my ability to whistle independently of yours, but a justice’s ability to decide cases can be exercised only along with the rest of the justices. I shall argue that the general will is a capacity shared in this strict sense.

I defend my shared capacity account with two main arguments. First, my account better accommodates Rousseau’s claims about the general will. Second, it makes compelling sense of the freedom that Rousseau claims we enjoy under the general will. Although some have tried to combine Rousseau’s voting procedure and the constitutive aim of the common good,¹¹ my account differs in explaining two things: how these two aspects fit together under the idea of a shared capacity for self-determination, and why Rousseau thinks participating in a voting procedure is essential to our freedom under the general will.

I begin by considering the shortcomings of procedural and common good accounts of the general will. As some of the textual grounds for my view are also the grounds for rejecting procedural and common good accounts, these views set the stage for my own.

A Procedural Interpretation

Some take Rousseau’s claim that “the tally of the votes yields the declaration of the general will”¹² to suggest a procedural interpretation of that will.¹³ On this view, the general will is just a certain voting procedure. The general will is everyone’s will in the sense that we all participate in determining it, and we all consent to abide by the results of the vote.¹⁴ Rousseau writes that “when an opinion contrary to my own prevails, it proves nothing more than that I made a mistake and that what I took to be the general will was not.”¹⁵ The procedural interpretation accounts for this claim—if the general will is just a certain voting procedure, the tally of the votes is incontrovertibly the general will’s content.

But in the very next paragraph Rousseau himself seems to deny the procedural account: “This [that the tally of votes declares the general will] 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 is no longer any freedom.”¹⁶ The voting procedure declares the general will, which constitutes our freedom, only when the majority has the “characteristics of the general will.” Hence, the voting procedure by itself does not define the general will’s content.

Gopal Sreenivasan tackles this apparent contradiction by incorporating those characteristics into a more sophisticated account of the voting procedure. Rousseau nowhere specifies all the “characteristics of the general will,” but Sreenivasan identifies several conditions for the vote to express the general will:

1. The subject matter of deliberation is perfectly general.

2. The conclusions of deliberation apply equally to all citizens.

3. All citizens participate in deliberation.

4. All parties to deliberation think for themselves.

Still, these conditions seem insufficient to guarantee that, as Rousseau puts it, “the general will is always upright and always tends to the public utility” (SC 2.3.1). Sreenivasan himself admits as much.¹⁷ He suggests that Rousseau has left the problem of filling out the conditions for us to resolve.¹⁸

However, Rousseau does not intimate that he leaves a problem for the reader, nor does he claim this list of conditions to be a guarantee that the vote aims at the common good. Indeed, Rousseau never gives a list of conditions at all. Sreenivasan draws his conditions from several different chapters, only one of which addresses how to get the vote to aim at the common good.¹⁹ If Rousseau meant for the general will to be defined by a procedure involving such conditions, it would have been more natural to list the conditions together. The textual evidence suggests that Rousseau did not intend for the content of the general will to be defined procedurally.

Moreover, it is difficult for procedural accounts to explain Rousseau’s claim that the general will is never “annihilated or corrupted,”²⁰ even when voters use legislative procedures to advance private interests. If constraints definitive of the general will procedure no longer obtain, would we not say that the procedure has been corrupted or annihilated? In what sense is the procedure not annihilated if it is no longer carried out properly? Only in the sense that it continues to exist in corrupted form. If the general will is a procedure with constraints, it is hard to make sense of Rousseau’s insistence that it is not annihilated or corrupted when those constraints are not honored.

In sum, Rousseau does not seem to be defining a procedure, and if he were doing so, he would appear to have failed by his own standard, for the procedure would not guarantee that the general will aims at the common good. Moreover, it is difficult to square a procedural account with Rousseau’s claim that the general will continues to exist even when the conditions for its expression fail.

These problems may lead one to prefer a common good account.

Common Good Accounts

Common good accounts define the general will in terms of the common good, independently of any voting results. The most popular kind of common good account defines the general will in terms of the people’s shared conception of the common good. For example, on Joshua Cohen’s interpretation,²¹ a general will is constituted by the citizens’ publicly sharing—knowing that they share—a conception of their common good, and by the citizens’ being motivated to prioritize that good over their particular interests in relevant circumstances.²² The general will itself is then the ability each individual has to act for reasons based in this shared conception of their common good.²³

Scholars debate how to understand Rousseau’s notion of the common good, and in particular, whether the common good is a function on private interest of some formal principle, such as universality, or whether it involves substantive principles.²⁴ Although this debate is important for the general will’s content, it is beyond this paper’s scope. What matters here is that common good accounts reject the claim that the content of the general will depends on the results of a voting procedure.

One advantage of common good accounts is that they can explain why the general will always aims at the common good, for the general will is defined in relation to that good. These accounts also explain why it is reasonable to agree to be governed by the general will: for something to be the general will is for it to be good for all citizens, at least in their best judgment. It may not be clear why I would agree to always comply with the results of a voting procedure when it offers little protection against a tyranny of the majority, but it is clear why I would agree to be governed by what is good for all.

Common good accounts can also give a persuasive reading of Rousseau’s claim, considered above, that the general will persists even when the vote is captured by private interest. In those cases, Rousseau says that the general will survives in the form of the individuals’ willing the common interest:

Each person, in detaching his interest from the common interest, sees clearly enough that he cannot separate them entirely, but his share of the public evil seems to him as nothing compared to the exclusive good he seeks to make his own. Except for this particular good, he wills the public good in his own interest just as strongly as anyone else.²⁵

Common good accounts with a shared conception of the common good accommodate this passage easily. Individuals retain their shared conception of the common good, but they subordinate it to their private interest.

On these views, however, the voting procedure Rousseau advocates becomes puzzling. We need not all participate in legislation for the laws to be for the common good. Hence, common good accounts sometimes attempt to downplay Rousseau’s insistence on universal participation in legislation.

Cohen argues, for example, that direct participation in lawmaking “is about preserving sovereignty, about ensuring its stability” and is not “a defining condition in the conception of sovereignty itself.”²⁶ I find it difficult to square this interpretation with passages in Rousseau such as the following: “The instant a people gives itself Representatives, it no longer is free; it no longer is.”²⁷ Cohen would have to read this claim as hyperbolic to the point of obscuring Rousseau’s meaning.

Or consider this passage:

Sovereignty cannot be represented for the same reason that it cannot be alienated; it consists in its very essence in the general will, and the will does not admit of being represented: either it is the same or it is different; there is no middle ground. The people’s deputies therefore are not and cannot be its representatives, they are merely its agents… Any law the People has not ratified in person is null; it is not a law. The English people thinks it is free; it is greatly mistaken, it is free only during the election of the members of Parliament; as soon as they are elected, it is enslaved, it is nothing.²⁸

Presumably Parliament’s will is capable of holding and prioritizing, at least in principle and for some (perhaps short) amount of time, the publicly shared conception of the common good. Rousseau is not arguing here that Parliament is unlikely to do so stably, as Cohen suggests. He is saying that Parliament’s will is not the English people’s will for the same reason that my authorized agent’s will is not my will: they are simply not the same (identical) capacity of willing. If Rousseau’s concern were with “preserving sovereignty” and “ensuring its stability,” as Cohen claims, he ought to have said that the people’s deputies cannot be relied upon to represent it, not that they “are not and cannot be its representatives.”

Rousseau makes the same point earlier in the Social Contract: “I say, then, that sovereignty, since it is nothing but the exercise of the general will, can never be alienated, and that the sovereign, which is nothing but a collective being, can be represented only by itself; power may well be transferred, but not will.”²⁹ Again, the problem is not that no one can be relied on to act in the sovereign’s interests. It is that the sovereign is one particular collective being, and that any agent purporting to represent the sovereign is not that collective being—and hence cannot exercise that collective being’s will.

Only after this point does Rousseau add that no one can be relied on to will the same things as the general will: “Indeed, while it is not impossible that a particular will agree with the general will on some point, it is in any event impossible for this agreement to be lasting and steady.”³⁰ The “indeed” or “in fact” (en effet) suggests that Rousseau is making an additional point beyond the one made previously:³¹ not only is a representative’s will not the people’s will, it cannot even be relied on to agree with it.³² Cohen’s interpretation, which reads this final point onto the other passages that argue that transferring a will is impossible, is not very plausible.

David Lay Williams suggests a different approach to Rousseau’s procedure. On his account, “the general will is the citizens’ willing of…the [objectively defined] common good,” and the legislative procedure is the “exercise” of the general will, sovereignty.³³ But it makes little sense to speak of the citizens’ exercising their willing of the common good. Willing is already an exercise; one does not exercise an exercise, one exercises a capacity. Because Rousseau himself says that sovereignty is “the exercise of the general will,” it seems that Williams ought to modify his account of the general will.³⁴

Even setting this problem aside, Williams’s account does not explain Rousseau’s claim that, in obeying the general will, one obeys no one but oneself.³⁵ How does the fact that the voters willed the common good make the law I opposed my will? The answer could be that the majority tends to be right.³⁶ The law would be my will because it is what I willed de re, the common good, even though I (mistakenly) thought that it was not. But that explanation makes no appeal to my participation in the vote. As long as I will the common good and the legislator legislates the same, the law is my will de re, regardless of whether I participate in the legislation. Once again, then, it is hard to see why everyone must participate in legislation.

I have argued that procedural accounts of the general will do not respect Rousseau’s claim that the general will aims at the common good and that common good accounts do not respect Rousseau’s claim that a vote by all the citizens is necessary to declare the general will. We should prefer an interpretation that combines these two seemingly opposed strands. I offer such an interpretation in the next section.

A Shared Capacity Account

A will is a capacity for determining oneself to act.³⁷ Hence, in this section, I propose that Rousseau thinks of the general will as a kind of shared³⁸ capacity for determining one’s shared self to act. I also suggest several features that a shared capacity of willing must have. These features account for Rousseau’s insistence on both a voting procedure and a constitutive aim, the common good.

Rousseau gives no general account of the will, let alone a shared will, but three features seem essential to his view. First, a will is something that determines itself to act (or not to act). Second, a will determines itself in accordance with an understanding of its circumstances and possibilities. Rousseau takes the understanding or judgment as a faculty separate from the will,³⁹ but the will determines itself on the basis of that understanding or judgment: “The general will is always upright, but the judgment that guides it is not always enlightened. It must be made to see objects as they are, sometimes as they should appear to it, shown the good path it seeks…”⁴⁰ Third, as we see in this quotation, the will determines itself to act in pursuit of some constitutive aim. For the general will, that aim is the common good—that is, Rousseau holds a version of the “guise of the good” thesis, at least for the general will.⁴¹

To summarize: if we are to share a capacity of willing in the fullest sense, we must 1) share the ability to determine ourselves to action, by 2) together consulting a shared understanding, in accordance with 3) a shared aim of the common good.

These three requirements account for Rousseau’s insistence that the general will features a shared voting procedure and aims at the common good. To share the ability to determine ourselves to action (requirement 1), we must participate in the decision to act—that is, we must vote. To share in the consultation of a shared understanding (requirement 2),⁴² we must participate in the deliberation⁴³ preceding the vote. And to share an aim (requirement 3), we must share a conception of the common good. Thus, we need not choose between defining the general will in terms of a procedure or in terms of the common good. If we define it as a shared capacity of willing, both the procedure and the guidance by a publicly shared conception of the common good follow.

In addition to accounting for procedure and aim, my shared capacity account harmonizes with other passages suggesting that the general will is a shared capacity. For example, when the social contract is signed,

At once, in place of the private person and each contracting party, this act of association produces a moral and collective body…which receives by this same act its unity, its common self, its life and its will…As for the associates, they collectively assume the name people and severally call themselves Citizens as participants in the sovereign authority.⁴⁴

Notice the analogy between the body, unity, self, life, and will of the people, on the one hand, and those of an individual on the other. Rousseau seems to say that the general will plays the same role in the life of the people that an individual will plays in the life of the individual: it is a capacity for determining itself to act, one in which the citizens participate.

Rousseau further describes the general will as something that can be exercised: “Sovereignty, since it is nothing but the exercise of the general will, can never be alienated.” Capacities are exercised. He continues: “The sovereign, which is nothing but a collective being, can be represented only by itself.” ⁴⁵ The general will belongs to a collective being, and so if the will is a capacity, it would seem to be a shared capacity, one which the citizens collectively constitute.

Neither procedural nor common good accounts describe a shared capacity of willing in this strict sense. A mere voting procedure has no constitutive aim. A procedure without an aim is no more a will or a willing than a coin toss. Consider what we might say of someone’s behavior in a temporary state of dementia or insanity: “What they do is not really their will.”⁴⁶ Why not? Because their behavior is not properly guided by their understanding of the good. Procedural accounts of the general will simply do not describe a will.

While procedural accounts do not describe a will at all, common good accounts do not describe a shared will. Suppose my partner and I know that we share a conception of our common good, and that we prioritize it. One day, though, on the basis of this shared conception of the common good, my partner decides to replace the TV without telling me. It would be strange to call replacing the TV our action, even if I later recognized that our shared conception had motivated the action. We did not decide to replace the TV; my partner decided to replace the TV. Only if we had both participated in making the decision—minimally, for my part, by consenting to the idea—would it have been our action. Similarly, for a law to be legislated by our will, it is not sufficient that the law arise from a publicly shared conception of the common good, or even that we all know that it so arises. It can be legislated by our will only when we decide it together.⁴⁷ Without a shared process for determining ourselves to act on a conception of the common good, neither that good itself nor a publicly shared conception thereof is a shared capacity of willing.

Because they do not treat the general will as a shared capacity, procedural and common good accounts fail to solve the problem that the general will is meant to solve in the first place: to show how I can be subject only to my own will. Procedural accounts fail because they describe not a will but a procedure. How can I obey only myself if what rules me is not my will because it is not a will at all? Common good accounts fail because they do not show the ruling will to be mine. They settle for affirming that I obey a will conforming to my conception of the common good.

Only a shared capacity, involving both procedure and aim, is both a will and, as I shall now explain, mine, because it is ours. This account takes Rousseau’s tendency to describe the general will as a shared capacity seriously, and it makes sense of Rousseau’s associating the general will with both a procedure and a constitutive aim, the common good.

How a Shared Will Can Be My Will

Any interpretation of the general will must account for the purpose given to it by Rousseau: to enable “each, uniting with all, nevertheless [to] obey only himself and remain as free as before.”⁴⁸ When a capacity of willing other than my own gives me the law, I am obeying another’s will in some important sense, even if that will is guided by a publicly shared conception of the common good and is known by all to be so guided. In other words, another’s act of legislation is not an act of my will. By contrast, our shared capacity of willing is my will. When I obey our shared will, I obey only myself.

I have taken for granted thus far that our shared will is my will: what is ours is mine—not mine alone, of course, but mine and yours together. At the very least, what is ours is more mine than what is yours alone. But what is ours also seems to be less mine than what is mine alone. Given how many people may belong to a political community, the general will can appear to be mine in a very meager sense, and consequently, the freedom we enjoy by being part of the general will can likewise seem meager. Suppose that there were no danger that a representative assembly would violate our publicly shared conception of the common good. What freedom would I gain by being ruled by our shared will, rather than simply by the representative assembly?

The challenge is to explain the sense in which a shared capacity of willing is an individual participant’s will, such that being ruled by the shared capacity constitutes non-negligible freedom. Rousseau does not explicitly discuss this challenge, but insisting on direct participation in legislation suggests that he has some answer in mind. In the remainder of this section and the next, I characterize the conditions under which a shared capacity of willing is an individual participant’s will.

Answering this question not only rounds out my interpretation of how a shared capacity of willing offers us freedom, but also buttresses my argument for the shared capacity account. For I shall show that many of Rousseau’s claims about the general will can be explained by two premises: 1) the general will is a shared capacity of willing (this premise I defended in the previous section), and 2) the general will is shared in such a way that it counts as the will of each individual as well (this point is essential to Rousseau’s account of freedom under the general will). That from these premises we can explain various features of the general will should reinforce our confidence that we are correct to attribute these premises to Rousseau.

I begin with a general question: when is our activity my activity? For example, when is our playing a melody my playing a melody? Playing the same melody in unison is one possible answer to this question, but not the only one. We might take turns contributing notes to a melody, for example. Under certain conditions it would be reasonable to say both that we are playing that shared melody, and that therefore I am also playing that melody (though not by myself, of course).

A first condition for our activity to be my activity in the fullest sense is that I have the capacity to do what we are doing: I am playing the melody that we are playing only if I as individual have the capacity to play a melody in the first place. If I am unmusical and just hit certain keys on an instrument as directed by someone else, we may play a melody, but I am not playing that melody, for I lack the capacity to do so. I am simply hitting a key as directed. The general will meets this condition: our activity of willing can be my activity of willing in the fullest sense only because I have a will.

A second condition is that my capacity for performing the activity be used in the process. If playing a melody involves shaping the duration, volume, and timbre of each note in relation to others such that the whole is musical, then I must use my capacity to shape the notes in these ways if I am to play a melody in the fullest sense. Suppose that you simply record me playing a few notes without regard for duration, volume, or timbre, and then later you supplement the recording with notes of your own. In this case, I am not using my capacity for playing a melody; rather, I am simply using my capacity for playing notes. It may be true that we played a melody in some sense, but I was not playing that melody. In the same way, for the general will to be my will, my capacity of willing must be used in the general will, and not just some sub-capacity like that for deliberating or for consulting the understanding.⁴⁹ Rousseau’s general will meets this demand: the general will 1) deliberates by 2) consulting the common understanding to 3) arrive at a shared decision, and I participate in the entire process when I 1) deliberate by 2) consulting the common understanding to 3) arrive at a decision through our votes.

A third condition is publicity: for our playing a melody to be my playing a melody, I must recognize that we are playing a melody together. If I am wearing noise-canceling headphones, for example, and so am unaware of the contributions my melody makes to a broader melody, we may play a melody together but I am not playing that melody. For my capacity to play a melody is not being directed at the melody we are playing. Similarly, I must know that I am contributing to the general will if the general will is to be my will.⁵⁰ Hence the legislative process and its constitutive aim must be publicly known for the general will to be the constituents’ will.

One theme uniting each of these three conditions is that, for our activity to be my activity, our activity must not use my activity as mere input; what I do cannot be just material for the construction of something different from what I am doing. Rather, I must be a full participant and co-shaper of the entire shared activity.

Civil Man

For me to enjoy full freedom under the general will, it is not enough that the state does not treat my participation in the general will merely as input. I must also not treat my participation as output. Only then, as I argue in this section, is our will my will. As we shall see, this thought explains several of Rousseau’s most striking claims about the general will.

To treat my vote as mere output is to regard my contributions to deliberation and my voting as expressing my individual will on the matter (regardless of whether it is aimed at the common good). I may take input about the common interest from public deliberation, but I conceive of what I do primarily in terms of my own activity, and not in terms of any collective activity: I gather ideas; I make up my mind; and I declare my opinion. In this case, my vote expresses my will: if I had my way, we would act accordingly. If my vote is counted, we may say that the general will is partly mine, just as we may say of the owner of a voting share in a company that the company and its decisions are partly hers—as a shareholder, she has a small share of control over the company. Here, the shared will is each individual’s will in the weak sense that each individual partly owns the shared will. Call this thin freedom the “shareholder’s freedom.”

On the other hand, suppose I view public deliberation primarily as a collective activity in which I am involved, an opportunity for us to make up our mind. In this case, I conceive of my participation in public deliberation not primarily in individualistic terms, as a means of making up my own mind about the public good and of proselytizing and voting for the views I take to be correct; instead, I conceive of my participation as merely part of a process by which we arrive at a collective decision. On this approach, my vote does not express what I will that we do, but rather gives an opinion about what is in the common interest. If I had my way, we would not simply act in accordance with my vote, for the goal of my deliberation is not for me to determine my will, but for us to determine our will. My mind is made up only when our mind is made up, and our mind is made up only when we tally the votes. I have my own thoughts, of course, but I do not take them to determine my will about what is to be done. My will is determined by the outcome of the vote.

Here, we may say that the shared will is mine not just because it (partly) belongs to me, but also because I belong to it. My individual will is part of the whole, not just because it is a vote included in the whole, but because my will is completed and determined by that whole. When taken independently of the whole, my will is indeterminate with respect to legislative questions, and yet it is active with the whole in answering them. Hence, the general will is my own will in a much more robust sense than in the case of the citizen with shareholder freedom, because she views her vote as mere output.

Consider again the analogy to music. When I play a melody with someone, my aim is not simply to play my notes well, but to play the melody well. If the other player straggles and I play at my preferred tempo, our notes will not combine well. For the sake of playing a melody, each player must make accommodations for the other. If one is the sort of player who cares only that the music be played well according to one’s own understanding, then one will experience the accommodations one makes as a failure to get one’s way. But if one’s aim is to give a performance that every performer can regard as theirs to the greatest extent possible, then accommodations are not deviations from one’s will but rather essential to what one wills. In the same way, if one regards one’s participation in the general will in collective rather than individualistic terms, one views the outcome of the process not as compromising one’s own will, but rather as expressing the group’s will, of which one is a part.

Rousseau puts it this way:

When a law is proposed in the People’s assembly, what they are being asked is not exactly whether they approve the proposal or reject it, but whether it does or does not conform to the general will that is theirs; everyone states his opinion about this by casting his ballot, and the tally of the votes yields the declaration of the general will. Therefore when the opinion contrary to my own prevails, it proves nothing more than that I made a mistake and that what I took to be the general will was not it. If my personal opinion had prevailed, I would have done something other than what I had willed, and it is then that I would not have been free.⁵¹

What I will in legislation is expressed not by my vote but by the tally of the votes, just as the melody I play with you comprises not just the notes I play but all the notes we play together. My will qua individual, Rousseau suggests in the last sentence, is not that we act according to my opinion, but that we decide together. The will to regard one’s vote as belonging to a collective willing process, rather than as mere output, is essential for making our will my will.

To will to be part of a general will requires a special kind of character. Rousseau describes the citizen with such a will as having a changed nature:

He who dares to institute a people must feel capable of, so to speak, changing human nature; of transforming each individual who by himself is a perfect and solitary whole into part of a greater whole from which that individual as it were receives his life and his being; of adulterating man’s constitution in order to strengthen it; of substituting a partial and moral existence for the independent and physical existence we have all received from nature. In a word, he must take from man his own forces in order to give him forces that are foreign to him and which he cannot use without the help of others.⁵²

Citizens have life and being only as part of a greater whole. Taking away citizens’ solitary nature involves bringing them to see participating in group decision-making not as their acting individually—as shareholders—but instead as their making up their collective mind.

In Emile, Rousseau writes that this change gives civil man a relative existence:

Civil man is only a fractional unity dependent on the denominator; his value is determined by his relation to the whole, which is the social body. Good social institutions are those that best know how to denature man, to take his absolute existence from him in order to give him a relative one and transport the I into the common unity, with the result that each individual believes himself no longer one but a part of the unity and no longer feels except within the whole.⁵³

To denature a will involves more than aiming that will at the common good rather than at private interest. It also involves disposing the will to act in the legislative process as a contributor to the collective judgment, as “part of the unity,” rather than as an independent capacity, or as “one.” The natural will is determined by its own perspective; the citizen’s will when legislating⁵⁴ is determined by the whole’s perspective. As citizen I recognize the collective decision as my own because the result completes the activity in which I am participating.

Natural man can regard the general will as his will only in the shareholder sense, because his particular will is already complete by itself: he takes the majority opinion’s resistance to his own will as a foreign imposition. But civil man’s will on legislative questions is indeterminate until the votes are in. He may differ from the majority in his opinion on what will achieve the common good, but that opinion he holds to be of secondary importance: he has a partial, relative will with respect to the body politic. In a robust sense, therefore, the general will is each civil man’s will.

The secret to allowing “each, uniting with all, nevertheless [to] obey only himself and [to] remain as free as before”⁵⁵ is thus to transform the self whom each person obeys. I cannot live in society with others and still obey only my natural, individual self. But if my self transforms into a partial self belonging to and completed by the whole of my society, then to obey myself in matters of legislation is no longer to obey my individual will; it is to obey the general will of which I am a part. In the state, civil man is subject only to his own will—because he has been reconstituted so that the collective will is his.

Of course, civil man does not will the deliberation’s outcome unconditionally. If the vote does not meet the conditions we have considered, including that of being guided by a publicly shared conception of the common good, then it is not the decision of the shared will, hence nor of his will. Moreover, because Rousseau argues that the common good involves a sphere of civil freedom in which neither law nor others interfere,⁵⁶ commitment to make decisions about the laws together does not imply commitment to make all our decisions together. Civil man still has a private will.

In the previous two sections, I have been arguing for two conclusions at once. First, the features of the general will as described by Rousseau derive from two basic thoughts: 1) the general will is a shared capacity of willing, and 2) the general will is so related to the individual wills composing it that the general will is the individual’s will. These two thoughts account for the requirement that there be a publicly shared conception of the common good, a shared voting procedure, and shared deliberation, as well as for the requirement that voters treat their votes as mere opinions. Additional features⁵⁷ of the general will could be derived from these thoughts, but I shall not derive them all here. The point is simply that my interpretation of the general will as shared capacity explains many of Rousseau’s claims about it.

The second conclusion is that being ruled by a shared will, when it is characterized by these two thoughts, constitutes a kind of freedom for participants. Participants in the general will fulfill the conditions that make a shared activity also the participant’s activity in the fullest sense. Such participants may then reasonably claim that the shared will is their will and that they obey only themselves, as Rousseau claimed.⁵⁸ My interpretation of the general will as shared capacity fits the role the general will is supposed to serve: to make citizens free.

The difference between merely being ruled in accordance with a public conception of the common good and being ruled by a shared will in which I participate is that, in the latter case, I can regard legislation as my own activity; in the former, I can regard it as at most an activity I endorse. The difference becomes especially salient when I hold a minority opinion about the best means to achieve our common good. In that case, voters who can identify their will as uniting with others to form the legislative will are co-authors of the law. By contrast, participants who can regard themselves only as shareholders are, in at best an attenuated sense, authors of laws they disagree with; non-participants are not authors at all.

Passages that Might Raise Doubts

Some passages about the general will may appear to challenge my interpretation. First, Rousseau says that even when people vote their private interest, the general will is never annihilated but “is always constant, unalterable, and pure.”⁵⁹ On my interpretation, Rousseau is asserting that the general will as shared capacity is “constant, unalterable, and pure.” We should not understand procedural failures as corruption of the capacity, but rather as hijacking of the voting procedure—which is only one part of the capacity—for another purpose. In the same way we might view a madman’s behavior not as his will, but as the hijacking of his faculty of desire by madness.⁶⁰ Rousseau conceives of the will as a capacity for action in accordance with the understanding and the constitutive aim of the common good. Since the common good is the general will’s constitutive aim, that will cannot be corrupted. What does not aim at the common good is simply not the general will. As long as citizens recognize that they share a common interest (and Rousseau insists that they do, even when they subordinate it to private interest), then the capacity to deliberate and vote on that interest remains. Hence, the general will is neither annihilated nor corrupted.

This same passage, though, may suggest a second interpretive puzzle for my view. Rousseau speaks of the general will as existing within each individual: “Even in selling his vote for money [the citizen] does not extinguish the general will within himself, he evades it.”⁶¹ One might think that, because the general will is something within each of us, its content cannot depend on a vote’s outcome. The vote, after all, does not happen within me. Cohen takes a similar passage⁶² to imply that “individual members of a group with a general will can themselves each be said to have a general will.”⁶³ If this were the case, then the shared capacity account would have a problem, because it insists that the general will is a single will shared by many.

Reference to the general will, however—la volonté générale, and take note of the definite article⁶⁴—as within each citizen is compatible with there being only one token general will that exists within every citizen, as opposed to one general will per citizen. We may say that the supreme judicial power exists in all members of the Supreme Court, without meaning thereby that each member has their own supreme judicial power. Rousseau’s language here is compatible with there being only one shared general will.

A passage about the will of government (the executive power), another group will, suggests how such a shared will manifests itself in each member:

For the body of the Government to have existence, a real life that distinguishes it from the body of the State, for all of its members to be able to act in concert and to assume responsibility for the end for which it is instituted, it has to have a particular self, a sensibility common to its members, a force, a will of its own that tends to its preservation.⁶⁵

Rousseau says that to act in concert, a government’s members need a shared “particular self.” That shared self manifests as a common sensibility, a force or will of its own. Its members share a single token sensibility, force, and will that inheres in all those members as a group.

This aspect of Rousseau’s view on government applies to the general will as well. That will, as we have seen, is the common self, life, and will of the people.⁶⁶ The people’s shared sensibility, force, and will presumably manifests itself at least partly in individual members’ attitudes and behaviors; in their shared conception of the common good, for example. This manifestation in the individual is what I take Rousseau to refer to in speaking of the general will within the citizen and is compatible with the thesis that the general will is a shared capacity in the strict sense.

Finally, a passage we have already considered might seem to favor a common good account over my own. Rousseau writes,

How will a blind multitude which often does not know what it wills because it rarely knows what is good for it, carry out an undertaking as great, as difficult as a system of legislation? By itself the people always wills the good, but by itself it does not always see it. The general will is always upright, but the judgment that guides it is not always enlightened.⁶⁷

This passage seems to suggest that the general will has an objective content, the common good, and that the content is independent of the process of deliberation and voting, which seem to merely be ways for the general will to come to know what it wills. And so it may seem that the general will is not a shared capacity after all.

The shared capacity account can allow that the will has some content independently of deliberation and voting. That content is its constitutive aim, the common good. But we must take care in jumping from the claim that the general will wills the common good to the claim that the general will wills X, where X is some particular law necessary for the common good. Suppose I will to adopt the best diet, and the best diet happens to be Mediterranean. It thus follows that I will to adopt the Mediterranean diet de re, but not that I will to adopt the Mediterranean diet de dicto. For I may not know that the best diet is in fact Mediterranean.

In the above passage, Rousseau is speaking of what the general will wills de re: there are laws that are in fact for the common good, and since the general will wills the common good, it wills those laws de re. But the general will does not know what laws are in fact for the common good, so it does not will them de dicto. In the same passage, Rousseau later speaks of the general will as being guided and taught by the understanding and “seduced” by particular wills; these remarks suggest that there is some sense in which the general will’s content depends on what seduces, guides, or teaches it.⁶⁸ Only when the general will is guided to specific laws by the people’s understanding does it will those laws de dicto. Hence, the general will’s legislative content always depends on a process of deliberation concluding in a vote.

Conclusion

I have argued for a shared capacity account of the general will on two grounds. First, this account explains Rousseau’s insistence that the general will involves a certain procedure, and that by its nature it aims at the common good. These two thoughts, which are troublesome for rival views, fit together naturally in the idea of a shared capacity of willing. Second, that idea harmonizes with a compelling account of freedom. A shared capacity of willing may be said to be each participant’s will, and this view allows for a more robust account of freedom than is available on rival interpretations of the general will. This is because it explains the thought that my will has made the prevailing decision even when my opinion was contrary.

This merging of a people’s wills is, by Rousseau’s own acknowledgment, a task for a god—but the possibility and attraction of a shared will may be sensed in our experience of sharing a will with family, friends, or colleagues. We can experience a taste of Rousseauvian freedom wherever there is a will to decide the terms of cooperation together, according to our publicly shared conception of the common good.⁶⁹