within the Social Contract: Rousseau on Punishment

Rights with in the social contract: Rousseau on Punishment


Citizenship is not a license that expires upon misbehavior.

—Earl Warren writing for the Majority in Trop v. Dulles


In thinking about our own contemporary practices and justifications of punishment we would do well to turn to the history of political theory, in particular to Rousseau’s political theory. Such an inquiry will help to frame, in a way that contemporary normative theories of punishment have not, what is significant about the distinctive practice of state punishment. In my view this historical inquiry can help to correct a mistaken emphasis in the dominant approach to the subject in the past half-century

Specifically, much contemporary work on justifications of punishment has been pursued from the perspective of moral philosophy.1 Such inquires are concerned with the Tightness or wrongness of punishment from the perspective of utilitarian or retributive moral theory considered in isolation from the political question of legitimacy. As I have argued at length in another place, the problem with an exclusively moral as opposed to a distinctly political inquiry about punishment is that it addresses only the punishment deserved by criminals and ignores the particular context involved when the state is doing the punishing.2 In contrast to a broadly moral theory of punishment, a theory of punishment within the confines of political morality should address not only what is deserved but also which punishments the state rightly metes out. In other words, a political theory of punishment is concerned not only with how and when to punish but also with the question of who is administering a punishment. Such an inquiry would concern not just the issue of desert but, more fundamentally, that of the political legitimacy of state punishment.

I turn to Rousseau in an attempt to contextualize the question of how the State should punish within a wider conception of political legitimacy. I suggest that we look closely at Rousseau not only because his account of punishment is grounded in what the state can and cannot do, but also because he is arguably the premier theorist of the social contract. In turning to his views on punishment we can thus be guided in thinking about the right role for the state in punishment by a theorist concerned to subsume this question within the more fundamental question of political legitimacy and one who did not view punishment as a moral dilemma isolated from questions of politics.

In particular it is important that Rousseau viewed justifications of punishment as a central issue of political legitimacy. In contrast to questions about Tightness or justice, legitimacy concerns the issue of why and when the state rightly can or cannot coerce citizens against their will. Given that punishment is perhaps the most coercive of state actions it is a central issue for theories of legitimacy. But punishment is not just a paradigmatic issue for theories of legitimacy; it is also a hard case. Simply put most criminals do not consent to be punished. Yet the social contract rests on an idea of consent. Indeed, given that many criminals resist punishment, does this suggest that they must be beyond the bounds of state legitimacy and outside the social contract? Hobbes clearly thought the answer to this question was yes.

Rousseau, I will argue, suggests otherwise. He gives us an account not only of why legitimate punishment can be explained within the context of the social contract; he also explains why the notion of political legitimacy limits those punishments that can rightfully be meted out. In doing so he reframes normative questions of punishment from being solely about desert to questions about the state as a legitimate agent of punishment. Controversially, I will assert that he also gives an account not only of limits on state action, but also of the rights of criminals that stem from their membership in the social contract.

Rousseau is, in many ways, an unlikely ally of defenders of the rights of convicted criminals and thus of the notion that an account of political legitimacy might provide for some rights of criminals. Certainly, he seems an even less likely ally of opponents of capital punishment. Rousseau not only defends the death penalty, but he does so by appeal to a theory of punishment that seems hostile to the rights of convicted criminals. He claims not only that capital punishment is justifiable, but that the very criminals who are subject to such a penalty consent to die. He reasons in his “The Social Contract” that:

The social treaty has as its purpose the conservation of the contracting parties. Whoever wills the end also wills the means, and these means are inseparable from some risks, even from some losses. . . . Whoever wishes to preserve his life at the expense of others should also give it up for them when necessary. For the citizen is no longer judge of the peril to which the law wishes he be exposed, and when the principle has said to him, “[It] is expedient for the state that you should die,” he should die. Because it is under this condition alone that he has lived in security up to then, and because his life is not only a kindness of nature, but a conditional gift of the state. . . . The death penalty inflicted on criminals can be viewed from more or less the same point of view. It is in order to avoid being the victim of an assassin that person consents to die, were he to become one.3

The problem with Rousseau’s view is that actual criminals resist attempts by the state to kill them. They pursue legal appeals and often protest the state’s action in their final words. Rousseau attributes a belief to the condemned that they do not actually hold. Rousseau’s justification of capital punishment therefore suffers from a flaw of “false attribution.” He confuses his view of the justice of capital punishment with the views of the condemned.

Nevertheless, despite Rousseau’s reputation as a collectivist hostile to individual rights, and despite his apparent antagonism to the rights of the convicted, I will argue that Rousseau offers an important framework for theorizing about both the rights of accused criminals and the rights of the guilty. Furthermore, I will suggest why this framework offers a distinctive and important argument in defense of the rights of the guilty and against capital punishment. Rousseau offers an account of how the problem of punishment can be rethought from within a legitimate social contract. While liberal individualists, such as Locke, have often grounded rights in an account of morality independent of the social contract, Rousseau demonstrates that such rights are actually required by the social contract. While accounting for these rights within the social contract might posit a legitimate basis for state punishment, it also, as Rousseau suggests, offers a basis for limitations on punishment. Thus, Rousseau’s account can be seen both to legitimize state punishment in some cases and also to generate rights for convicted criminals.

The social contract serves as a foundation for rights related to the practice of punishment in two ways. First, Rousseau’s understanding of rights suggests, in ways that other rights theories have failed to, why convicted criminals should have rights against the state. His conception of rights contrasts sharply with the suppositions of theorists such as Hobbes, who viewed rights as inherent in the individual but failed to limit state action against individuals.4 Second, Rousseau’s account positions the rights of the guilty within wider policy considerations, in particular those of individual welfare. Specifically, I argue that for Rousseau, legitimate punishment depends upon the state’s guarantee of welfare rights. In the absence of welfare guarantees, there is no legitimate social contract for Rousseau. In such circumstances, the state lacks legitimacy when it punishes. Rousseau thus provides a more nuanced account of punishment that situates the practice within the wider realm of social justice issues.

In some ways my contribution here cuts against the grain of some themes in this volume. As Sarat and Umphrey suggest in their introduction, thinkers such as Foucault have been particularly suspicious of the way punishment and regulation function within the modern sovereign state to minimize the autonomy and integrity of individuals, in particular when the issue is coercive regulation or punishment. Indeed, the social contract tradition as a whole might be though on this line of reasoning to be part of a historical process that results in the justification of the discarding of individual interests and real rights against the state. Certainly such suspicions have been prominent in critiques of Rousseau, notably in Marx’s “On the Jewish Question.” Moreover, “citizenship” as a concept linked to sovereignty far from preserving the interests of the individual might be thought along these lines another device for regulations that ignore individual rights. Both Marcus Dubber and Alec Ewald, in their contributions to this volume, gesture toward this kind of Marxian critique.

But I believe that Rousseau himself has a powerful normative response to these worries. While Rousseau does not disregard the facts of coercion and regulation in the modern state, I will argue that his account of sovereignty, consent, and the general will offers a way of analyzing and potentially critiquing all state coercion whether under the guise of “regulation” or officially recognized as punishment. His is a standard for legitimate coercion, often failed by contemporary states in a variety of ways. Importantly, I will argue that the civic rights of the citizen are tied to limits on any state coercion regardless of its label. In short, in theorizing about how state coercion can be limited given its historical tendency to disregard the distinct interests of the individual, Rousseau serves as a worthy guide, despite his reputation as a collectivist who ignores the distinct interests of individuals.

I begin by closely examining Rousseau’s claim that the criminal consents to punishment, and by suggesting that his seemingly absurd and perverse claims are actually defensible. I go on to clarify two senses in which Rousseau’s understanding of consent serves to ground the rights of the guilty. First, I argue that the idea of consent itself legitimizes some punishment but also sets strict limits on when and how the state can punish. Second, I argue that Rousseau’s account of welfare rights also sets a limit on when the state may rightfully punish. In the final two sections, I consider the role of punishment outside the social contract in Rousseau’s theory and suggest an important contrast between Rousseau’s and Hobbes’s conceptions of punishment.

The Meaning of Consent

Rousseau’s theory of rights, specifically his account of the rights of convicted criminals, is implicit in his conception of consent. For Rousseau, it is necessary that criminals consent to punishment because all coercive acts of the state must be arrived at unanimously, which seemingly grants each citizen a veto over the coercive powers of the state. On Rousseau’s account, this need for unanimity is a fundamentally democratic one. If individuals subject to the laws of the state are to be regarded as the authors of that law, it is necessary that they affirm and consent to the law.

This requirement of unanimity appears throughout Rousseau’s political philosophy. For example, if property rights are to be justified within the social contract, Rousseau demands that the owner receive the “explicit and unanimous consent from the human race” to that right. The most difficult application of this theory, however, concerns its application to punishment. If sustaining unanimity is a requirement of the social contract, Rousseau must show that even criminals, who are subject most obviously to state coercion, consent to their punishments. But again, it seems that this requirement of unanimous consent in regard to punishment is impossible to satisfy. Specifically, the problem with unanimity is that it gives a veto power to any one individual within society, which is particularly problematic in the case of punishment. As I suggested in the introduction, the condemned, and criminals more generally, do not often willingly agree to their punishment. Therefore, it is unrealistic to think that unanimity could be achieved with respect to individual criminal punishments.

However, I want to suggest that Rousseau’s notion of unanimous consent is not subject to the obvious objection that actual criminals do not consent to their own punishments. By consent, Rousseau is not referring to actual empirical consent. Rather, he espouses a hypothetical notion of consent, which seeks to reconcile the individual will of the condemned with what he calls the “general will.”

In “The Social Contract,” Rousseau argues that in societies with a legitimate social contract, unanimous consent is necessary, “at least on one occasion.” In particular, he refers to the founding moment of the social contract to legitimize the authority of the terms of the social contract over every citizen.5 While it would be impossible to satisfy unanimity on every occasion, citizens could unanimously consent to a set of constitutional principles governing the makeup of the social and political system. Rousseau emphasizes, however, that in the absence of such a social contract, unanimity would be required to justify every act of legislation: “In fact, if there were no prior convention, then, unless the vote were unanimous, what would become of the minority’s obligation to submit to the majority’s choice, and where do one hundred who want a master get the right to vote for ten who do not?” Here, Rousseau’s explanation of unanimity suggests a distinction between the justifications of particular acts of the state within the context of a legitimate social contract and those that fall outside of it. Rousseau’s point is that state punishment in a society without a social contract could not be legitimate because unanimity would be impossible to achieve each time an issue of punishment arose. However, there is potential justification for punishment in a society with a legitimate social contract. In this type of society, unanimity is not required to implement every social policy or institution; rather, a more general concept of legitimacy is required. On my view, unanimity on this “one occasion” should not be understood to imply that legitimate polities must have the actual assent of all their citizens, even at the moment of founding. Such an understanding of unanimity would be unworkable given the likely scenario in which at least one individual would dissent from the content of the social contract. Thus, a more nuanced interpretation, despite Rousseau’s use of the term “explicit” consent, would entail some idea of hypothetical, as opposed to actual, consent.

The issue of whether a criminal consents to punishment depends on whether the punishment adheres to the kind of hypothetical consent necessary in the formation of a legitimate social contract. Hypothetical unanimous consent is not an account of actual procedure, nor is it reliant on what persons actually say or do. Rather, it is a principled account of which guarantees are necessary to ensure that all citizens could consent to a particular set of constitutional principles. An ideal of hypothetical consent asks us to place ourselves in a contractual situation where we must determine which laws are justifiable provided that we ourselves and our fellow citizens are uniformly free and equal. Although such a contractual situation would never occur, it serves as the thought experiment by which the legitimacy of laws is measured. I am suggesting that modern accounts of the social contract such as those defended by Habermas and Rawls have their antecedents in Rousseau.6

The major requirement of the social contract, according to Rousseau, is respect for the “general will.” The general will is often confused with a democratic procedure, such as majority rule. However, I believe Rousseau’s text suggests that while a majoritarian procedure might be compatible with the general will—because, for instance, it “tends towards the general will”—the general will is rightly understood as an ideal standard of “sovereignty” distinct from any particular procedure.7 This distinction is evident in Rousseau’s contention that while majority will might err, the “general will can never err.” The distinction between majoritarian procedure and general will is also evident in the multiple places in the text where Rousseau speaks of the general will as requiring some substantive protections for individual rights, even in the face of popular decisions to violate them. Rather than serving as a particular democratic procedure, the general will aims to protect and preserve each individual’s interest within the social contract. Although it does not require actual unanimous consent, it does ask us to consider whether each individual subject to coercion, when placed in a hypothetical initial contract situation, would agree to a particular policy. In this hypothetical situation Rousseau is imagining that the contractors would seek to find terms of the social contract that respected the freedom and equality of each individual citizen.8 This hypothetical consideration of what individuals would say about any policy in the situation of the original contract is a way of considering the distinct interests of all citizens when making general legislation.

Commentators have generally overlooked the formulation of the general will as a means of preserving individual interests and personal rights. However, Rousseau’s very first mention of the general will reveals its importance as a way of thinking about individual rights. In the “Discourse on Political Economy,” Rousseau clearly states that for any policy to be consistent with the general will, it must account for the distinct interests of the individual. While potentially compatible with majoritarian procedures, the general will can never require the sacrifice of the individual for the common good. Rousseau thus rejects any interpretations of the general will in terms of a mere aggregation of preferences. Consider the standard as it is elaborated in the following passage, likely the first in-print exposition of the idea of the general will:

It is no more believable that the general will would permit a member of the state, whoever he might be, to injure or destroy another member than that the fingers of a man in his right mind would put out his eyes. . . . In effect, is it not the commitment of the body of the nation to provide for the maintenance of the humblest of its members with as much care as for that of all others? And is the welfare of a citizen any less the common cause than the welfare of the entire state? . . . [If] this means that the government is permitted to sacrifice an innocent person for the welfare of the multitude, I hold this maxim to be one of the most despicable that tyranny has ever invented . . . . For far from [its] being the case that one individual should die for all, all have committed their goods and their lives in defense of each of them, so that individual weakness would always be protected by public force, and each member by the entire state.9

I take this passage to suggest that, regardless of what is required by hypothetical consent and the general will, policies and institutions must protect individual interests. Importantly, it uses punishment as a paradigmatic example of how the general will places limits on the state. Even when the common good would derive benefit from the punishment of innocents, such an act could never be legitimate, nor could it be consented to wholesale. For instance, utilitarians often face what has become known as the “sheriff problem.” They are asked to imagine that a sheriff could save hundreds of lives that would otherwise be lost in an impending riot over the state’s failure to catch a criminal if she would only frame an innocent individual, ultimately quelling the disturbance. While utilitarians, for whom the aggregate good is central, often have trouble explaining why such an action would not be justified, Rousseau rules out this kind of sacrifice in his explanation of the general will. For Rousseau, the common good cannot replace or counterbalance the most basic interests and rights of the individual.

In my view, Rousseau’s idea of unanimity, and its requirement for individual consent, is best understood as a precursor to modern contractualist political theory. Such accounts posit a need to justify all state actions involving coercion in a manner that respects the distinct interests of all individuals. For instance, Rawls’s principle of liberal legitimacy requires that coercion be justifiable to all citizens.10 In a variation on this theme, Tim Scanlon suggests a right of each citizen to reject any coercive policy.11 Much has been made of the distinction between this modern contractualism, which heavily weighs the distinct interests of individuals, and the tradition of the social contract. I think, however, that with the understanding of unanimity and the general will that I have laid out in this section, we can come to see a much closer relationship between these modern theories and that of Rousseau.

Specifically, Rousseau’s contention that the criminal must consent to his or her own death can be read as a requirement of political morality that any particular punishment be compatible with his or her reasonable interests. The question, therefore, is not whether criminals actually consent to their own deaths, but whether, when they are regarded as ideal free and equal citizens, they would consent to the terms of their punishments. Of course the fact that a criminal has violated the terms of the social contract makes him or her less than an ideal citizen. Rousseau’s point, however, is that such people are not exiled from the social contract. Instead, they are entitled to be treated as citizens. In Rousseau’s view it is necessary to use hypothetical consent to discern how the state best continue to treat such “bad citizens” in a manner consistent with the requirements of the social contract. Ultimately, consent for Rousseau is the way the state can continue to balance the distinct interests of the criminal against the interests of the aggregate society.

The Rights of the Guilty within the Social Contract

So far, I have suggested why Rousseau’s notion of consent is hypothetical as opposed to actual. I have argued, moreover, that this idea of consent seeks to preserve individual interests and rights. We are now in a position to reexamine the notion that guilty criminals consent to be punished through the lens of hypothetical consent. I want to argue that, although this requirement seemed perverse at first glance, it is actually protective of individual rights.

Only gold members can continue reading. Log In or Register to continue