[11]
Hegel on the justification of Punishment

Dudley Knowles

My claim in this essay is that Hegel’s theory of punishment is a very near miss. It is fair to say that the theory is complex and unclear—what else could account for the variety of interpretations it has provoked, the amount of good work directed to its exegesis and evaluation? But I think a clear path can be traced through the textual thickets. Unfortunately, when clarity dawns, we are left with a conclusion that I believe cannot serve to justify punishment: the claim that the criminal himself cannot complain if he is treated in roughly the way he has treated others. I want to end the essay on a provocative note by suggesting that Hegel’s purpose, the justification of punishment within the normative framework of “Abstract Right,”1 could best have been fulfilled had he espoused a hypothetical social contract theory. Hegel’s refusal to do so is explicit. I suggest that he was confused.

Hegel wrote about punishment throughout his career; he discusses the problem in his early essay on The Spirit of Christianity and its Fate, in the essay on Natural Law, in the Jena manuscripts and the Nuremburg Propaedeutic. But the text on which I concentrate is the Philosophy of Right. His final statement of the principles of just punishment occurs at PR §220:

When the right against crime takes the form of revenge (see §102), it is merely right in itself, not in a form that is lawful [Rechtens], i.e. it is not just [gerecht] in its existence [Existenz]. Instead of the injured party, the injured universal now makes its appearance, and it has its distinctive actuality in the court of law. It takes over the prosecution and penalization of crime, and these thereby cease to be the merely subjective and contingent retribution of revenge and are transformed into the genuine reconciliation of right with itself, i.e. into punishment. Objectively, this reconciliation applies to the law, which restores and thereby actualizes itself as valid through the cancellation [Aufheben] of the crime; and subjectively, it applies to the criminal in that his law, which is known by him and is valid for him and for his protection, is enforced upon him in such a way that he himself finds in it the satisfaction of justice and merely the enactment of what is proper to him [des Seinigen].

A sound interpretation of Hegel’s views will clarify the conclusion he reaches here and outline the arguments that sustain it. The burden of argumentation is effected earlier in the book, in subsection 3, “Wrong” of “Abstract Right.” Let us turn in that direction.

CONTEXT: PERSONS AND RIGHTS

“Abstract Right” elaborates the moral perspective of persons, taking “person” to be a technical term denoting our first-shot (immediate) conception of ourselves as discrete, atomic individuals. The “person” is the familiar protagonist of modern liberalism, the agent conscious of himself merely as distinct from other individuals, the most skeletal, universally applicable mode of self-ascription. In the classical literature of political philosophy, Hegel’s concept of the person best articulates the target of modern communitarian writings. The category is almost as empty and formulaic as these critics of liberalism suggest.2

Sparse though the concept of the person may be, it has an appropriate normative content—the capacity for rights—and an appropriate normative expression—the language of rights, summarized in the command or imperative of right as ‘be a person and respect others as persons’ (PR §36). As they say nowadays, the right has priority over the good. At this stage of the dialectic, Hegel has nothing to say concerning characteristic human goals or ends.3 “In formal right, therefore, it is not a question of particular interests, of my advantage or welfare” (PR §37). The rules of right are Nozickian side-constraints on action: “The necessity of this right is limited to the negative—not to violate personality and what ensues from personality. Hence there are only prohibitions of right” (PR §38).4 “Abstract Right” therefore delineates the contours of a rights-based morality. The details are filled out in three movements: “Property” explains the necessity of rights to property and physical integrity; “Contract” advances the analysis to explain the structures of will manifest in interpersonal transactions; “Wrong” advances the analysis still further by explaining that rights are enforceable—and, in the particular case of criminal wrongs, how punishment is the appropriate response.5 Hegel’s account of punishment is thus lodged in a theory of rights. The point of the discussion of punishment is to show what is entailed by our concept of the person and its attendant doctrine of personal rights.

Of course, our prime interest may be in the philosophical problem of punishment; Hegel’s chief aim, by contrast, is to show how our thinking about punishment informs our understanding of ourselves as rights-bearers.6 His specific thesis is that “through this process of mediation [the righting of wrong, generally the infliction of punishment] whereby right returns to itself from its negation, it determines itself as actual and valid, whereas it was at first only in itself and something immediate” (PR §82)7 The righting of wrong, whether it is effected in civil law by adjudication or compensation, or in criminal law by punishment, establishes the rights that persons characteristically claim as actual and valid in a way that their mere assertion, howsoever universal, could never achieve.

It is tempting to read Hegel’s doctrine of punishment as the response to a difficult problem which all rights theorists face and which some have tackled explicitly.8 Rights theorists, minimally, demand respect for physical integrity, territorial mobility, and private property. Such rights demand protection and enforcement by coercive legislation that sanctions the punishment of offenders. Punishment standardly takes the form of capital or corporal punishment, imprisonment, fines, or compulsory social service—all of these being measures that in the normal case would constitute rights violations. Hence rights theorists propose, as protective of rights, coercive instruments that standardly would violate rights. Rights theorists, then, have a special problem concerning the justification of punishment and we should expect their discussion of both rights and punishment to reflect this.

That said, I am unclear whether Hegel himself notices the point. He certainly does not address the problem directly as a problem concerning rights—in fact, read naively, his claim that punishment is the criminal’s right suggests the opposite: punishment, far from constituting an incipient violation of the rights of the criminal, serves to respect them. On the other hand, exactly the same form of the problem emerges when it is couched in terms of the value of freedom. Rights are necessary to promote and protect freedom. The coercion of the criminal looks to diminish his freedom in just the way that the criminal limits and repudiates the freedom of his victim. As injuries to the criminal’s will, the standard modes of punishment promise to infringe on his freedom. As soon as we insist that rights are enforceable—in Hegel’s terms, “Abstract Right is a coercive right” (PR §94)—paradox threatens.9

THE STRUCTURE OF THE ARGUMENT

Let us extract from PR §220, quoted above, several theses that will form the agenda of our discussion:

First, punishment, as imposed by courts of law, is the genuine reconciliation of right with itself.

Second, this reconciliation has an objective characterization: summarily, the restoration of right.

Third, this reconciliation has a subjective characterization: it is applied to the criminal on the grounds that he endorses its application. As expressed at PR §100, punishment is “his right … a right for the criminal himself, that is, a right posited in his existent will, in his action.”

How are these theses related to each other? Each of them postures as a justification of punishment, so one might say: punishment is legitimate as the reconciliation of right with itself, and/or the restoration of right, and/or the recognition of right on the criminal’s part. It is fair to say that the variety of good (and some bad!) interpretations of Hegel’s discussion of punishment reflect the ambiguities of the “and/ors.” Some readers have concentrated on one or another of these items; some have discussed two, or maybe three. I try to make sense of all three, defending the following theses.

The first claim, that punishment is the reconciliation of right with itself, is programmatic: it announces that the justification of punishment that is to follow is an implication of the theory of rights that has preceded it. As I suggested above, the notion of rights, their universal assertion and moral force, frames the context for the discussion of punishment that follows.

The second and third claims, concerning objectivity and subjectivity, may be viewed as distinct arguments. I argue that, although each may be (and have been) advanced independently as distinctive of Hegel’s contribution to the philosophical problem of punishment, they work better in harness. This, of course, is Hegel’s intention; this is how ‘Objectivity’ and ‘subjectivity’ work for him. I shall outline the arguments severally, point out their limitations, and explain how these limitations are overcome when they are understood to imply each other. This is a nicely dialectical procedure, and, for Hegelians, it should be good news, supposing that the complex account is defensible. The bad news is that this defense is faulty. Were the flaws to be rectified, Hegel would be required to accept a version of contract theory that he did not fully understand, and which, to the extent that he did understand it, he positively rejected!

THE RECONCILIATION OF RIGHT WITH ITSELF

Mention of reconciliation calls to mind Michael Hardimon’s recent discussion of Hegel’s social theory.10 Reconciliation is achieved when those susceptible to the tensions of fulfilling both individual and social projects—tensions that may amount to alienation—find themselves at home in the world, and a world that is a home for its members is a world of freedom.

One might doubt that Hardimon’s analysis can have any grip on Hegel’s discussion of punishment. “Abstract Right” does not delineate a properly social world. But the persons who inhabit this hypothetical11 moral structure recognize norms of property and physical integrity that are interpersonally valid and engage in transactions that express a common, though contingently fixed, will. If the institutions of ethical life comprise the actual social world, we can think of “Abstract Right” as a quasi-social world. It is not the pre-social world of solitary individuals that Rousseau conjectures in the Discourse on the Origin of Inequality. As in the states of nature encountered in Hobbes’s Leviathan and Locke’s Second Treatise of Government, persons like us (rather than our ancestors) confront each other with a social problem: how to manage interactions so as to avoid conflict and effect cooperation. It establishes in the foreground of discussion the fact that persons like us tend to assert conflicting claims of right—hypostatized as “Wrong,” whether unintentional (nonmalicious wrong, PR §§84–86) or intentional (Fraud, PR §§87–89, or Crime, PR §§90–103)—which require arbitration or adjudication. Not only is “Abstract Right” a social world in this weak sense; as in Locke’s natural state, its members have a distinctive moral status; as persons, they have a capacity for rights on the basis of which they make claims to respect from each other. So “Abstract Right” is enough of a social world to make talk of reconciliation apt, if it has features that call for reconciliation.

Furthermore (and this is the point of my citation of the conclusion of Hegel’s discussion of punishment at PR §220), the hypothetical or abstracted construction of “Abstract Right” is integrated into the structures of ethical life chiefly in the first two figures of “Civil Society”—the “System of Needs” and the “Administration of Justice.” At this level of analysis, members of “Civil Society” identify their social world as one in which the rights that they claim as persons, as atomic, discrete, individuals (albeit integrated into families—male heads of families, for the most part) find institutional expression and recognition.

We can attempt an understanding of the reconciliation of right with itself by first exploring the conditions under which such a reconciliation is necessary. Formally, it is the fact of wrongdoing that calls for reconciliation. Wrongdoing is characterized as a show or semblance of right. Some sort of false claim concerning the principles of right is made by the wrongdoer. The wrong may be an unintentional error, calling for adjudication. Or it may be a fraud, wherein deception parades as respect for right, mutton is sold as lamb. Or it may, in the particular case of crime, involve a rejection of the principles of right. Hegel announces: “The initial use of coercion, as force employed by a free agent in such a way as to infringe the existence [Dasein] of freedom in its concrete sense—i.e. to infringe right as right—is crime” (PR §95).

The implication is twofold: the criminal’s use of coercion violates the particular right of the victim to the object of his legitimate claims, and hence his freedom, and in addition, denies his moral status as a person, his capacity for rights.12 The criminal, say a car thief, harms the victim, if not twice-over, then in two related domains, failing to respect both his specific right to the car he steals and the victim’s moral status as an owner, a person with rights.

There is a contradiction at the heart of crime or, if not exactly a contradiction, then a couple of competing claims. The victim is a person, asserting a capacity for rights and making a specific claim to be, for example, the owner of this car. The thief, who has driven it away, rejects the victim’s rightful claim and, by implication, his moral status as a person, a bearer of rights.

Unfortunately, at this stage in the argument of the Philosophy of Right, Hegel’s analysis looks as though it collapses into a string of metaphors that darken rather than clarify the position. Crime is a nullity as manifested by the subsequent nullification of the infringement when punishment is effected. Punishment is the negation of a negation, the actuality of right, the necessary mediation of right with itself as it achieves the cancellation (annulment, negation, Aufhebung) of its infringement. This is the sort of rhetoric that has given Hegel a bad name with impatient critics. As Wood points out, if this language is taken at face value, some very bad arguments are being sketched.13

If these phrases are to have any purchase on reality, they must be construed loosely as trailers for the closer arguments to follow.14 We must read Hegel as insisting that persons cannot be, know, or feel themselves to be at home in a world where conflicting claims are made regarding each other’s moral status and specific rights. The demand for the reconciliation of right with itself is the demand that the conflict or contradiction be sorted out—and, evidently, the way to do this is to vindicate one of the conflicting claims, advancing the analysis of rights by ascribing to them a force which was hitherto unperceived or unrecognized (as we shall see, on the part of the criminal). Better still, we can seek to demonstrate to one of the contending parties that they themselves must accept the decisive judgment as an implication of other things they affirm. This how Hegel proceeds. Mercifully, we can abandon the rhetoric and attempt to track down the arguments.

THE RESTORATION OF RIGHT

Hegel tells us explicitly that punishment is the restoration of right. The claim is first made at PR §99: “Thus, an injury to the latter [the criminal] as an existent will is the cancellation [Aufheben] of the crime, which would otherwise be regarded as valid, and the restoration of right.” It is repeated, as we have seen, at PR §220, where the “genuine reconciliation of right with itself” is effected objectively by legal punishment, which “restores [the law concerning rights] and thereby actualizes itself as valid.”

This reading of Hegel’s theory of punishment has produced a dense and useful literature in recent years. In 1971, David Cooper explained Hegel’s talk of punishment restoring the right as the assertion of a logical or conceptual thesis. It is a necessary condition on rights being successfully asserted that their violation be punished. “Unless people are generally apprehended and punished for preventing others doing x, there is reason to suppose that the latter do not have the right to do x.”15 Although Cooper stresses the conceptual connection between the assertion of rights and the application of just punishment, Peter J. Steinberger argues that the conceptual connection has not been drawn tightly enough. Since it relies on empirical judgments to the effect that only the punishment of offenders (i.e., hard treatment, and not, for example, public denunciation) can vindicate the rights which the criminal challenges, it misrepresents the logical or conceptual groundings of Hegel’s defense of punishment.16 Allen Wood notices Hegel’s talk of restoring the right and associates it with Feinberg’s expressivist theory.17 He is ambivalent about attributing this theme to Hegel, claiming that Cooper’s explication fails on all grounds: it is not explicitly stated in the texts, and, if it were, it would be a poor argument.18 Stephen Houlgate has challenged Wood’s rejection of a conceptual thesis, claiming that “a [criminal] violation cannot therefore be allowed to stand but must be negated so that the necessary validity of right is restored.”19 In the same volume, Wood replies forcefully, challenging the ‘conceptual’ interpretation of the restoration of right theme and rejecting the argument, once more, as “just no good.”20

We can begin to adjudicate these disputes by paraphrasing Hegel’s statement of the position at PR §97. If an ostensible right is violated and the violator is not punished (supposing him to be known and available for punishment, etc.), we must regard his deed as innocent; if he has taken some property, we must regard that property as his. Contrariwise, if the criminal is punished, the status quo ante crime is publicly restored; both the victim’s moral status and his specific rights are vindicated. The deed cannot be both a crime and right.21 It is either not a crime or not right—and the response, by way of condonation or punishment, demonstrates one’s judgment of it. Are these conceptual claims? As they used to say, it all depends what you mean by “conceptual.” Cooper amplifies his interpretation by the employment of the Austinian apparatus of performative utterances, so perhaps this argument is best seen as the application of conceptual analysis of the Ordinary language’ variety.22 This kind of analysis evidently takes on board ground-level empirical beliefs (or “presuppositions”) on the part of those whose discourse is being analyzed—which is why Steinberger feels Cooper’s account is mistaken. (Steinberger’s own contribution is an even more pernickety employment of this philosophical genre.) Wood gestures toward Feinberg’s expressivist theory as a close relation, but, so far as I can see, does not endorse this reading, since he believes that a consequentialist concern for the restoration of rights cannot be reconciled with other, central, propositions of Hegel’s theory, including notably the claim that punishment is retributive in intent. (Other critics have claimed that Hegel’s account becomes teleological if the purpose of punishment is to restore the right—a version of rights- or freedom-consequentialism, perhaps.)23 Houlgate believes that Hegel adduces a conceptual rather than a causal connection between punishment and the actualization or validation of rights, claiming that it is that of the form of logical necessity integral to Hegel’s speculative practice.

It should be evident by now that these issues cannot be sorted out, at the level of interpretation, without the commentator’s coming to some explicit conclusions concerning the methodology of Hegelian argumentation. It looks as though no advance is possible without exploring the deepest veins of Hegelian scholarship. If what we are faced with is conceptual analysis of a familiar sort, as Wood points out, the conclusion will be parochial and conservative—but then perhaps “we” can find agreement about how “we” think in “our” parish, and maybe this enterprise can afford “comprehension of that truth concerning right, ethics, and the state [which] is as old as its exposition and promulgation in public laws and in public morality and religion” (PR Preface ¶5).

If, by contrast, we are tracking the path of a speculative logic, we still have to unpack the metaphors that fill the conceptual space of “the reconciliation of right with itself,” and what better way of doing this can there be than to find a valid argument? One point should be agreed on by all: that Hegel’s argument, even as paraphrased lengthily above, is elliptical. As such, we should expect any plausible reconstruction of it to read the texts at their most capacious and go beyond them if the argument requires supplementation. And, pace Wood, a plausible reconstruction of the filling Hegel’s argument evidently requires need not, of itself, amount to the reconstruction of a plausible argument or one that is consistent with everything else Hegel says on the topic.

It is on these grounds that I am content to endorse the drift of Cooper’s original reading. Following my earlier remarks about Hegel’s focus on our understanding of rights and his interest in what a philosophical examination of our practice of punishment contributes to that enterprise, I state the conceptual truth at the heart of the restoration of rights argument as follows:

Rights are not properly recognized (actualized) as valid claims, binding on others, unless their violation is met with punishment wherever possible.

As conceptual truths go, this is, as Wood implies, parochial. It applies to our world, supposing that alternative responses would not serve the purpose of public recognition. For all I know, there are other worlds wherein a public judgment of wrongdoing may suffice to restore the right. But then, I suspect (and this is suggested by Tunick’s example),24 public judgment would amount to denunciation, and this in turn would be regarded as hard treatment and hence as a measure of punishment. I speculate, on the basis of the sort of common sense that is acknowledgedly fallible, that in our world rights cannot be protected, right cannot be restored, by nonpunitive communications.

This reading can be supported in a way that Cooper does not attempt. Punishment can only restore the right if the institutions whereby punishment is effected constitute the means of public recognition of rights. Plausibly, if not speculatively or logically or conceptually-in-all-possible-worlds, public recognition requires public institutions. And so it proves. In the institutions of “Civil Society” that administrate justice, “Abstract Right” is posited objectively as law, universally promulgated and intelligibly codified, publicly dispensed in open court following trial by jury.25 The practice of punishment, following court proceedings that have established the fact of criminal behavior, makes it clear to all parties (victim, criminal, and the general public) that the rights of the matter are as the law states them to be. The fact that the restoration of right is achieved objectively, in a public institutional process, emphasizes one element of the reconciliation of right with itself. In Hardimon’s language, the social world is, and is known, felt, and affirmed to be, a home for rights-bearers. The victim’s rights are vindicated, his moral (now legal) status as a person with the capacity for rights is affirmed, the public’s interest in countering a danger to society is satisfied (PR §218). But what of the poor criminal?

If the criminal is left out of the picture, the objection I mentioned earlier to the restoration of rights theory of punishment, that it is, at bottom, consequentialist, is telling. Even if, instead of the usual suspects (happiness, pleasure net pain, preference satisfaction, objective list), the value to be maximized is conformity to rights, or freedom, it is hard to see Hegel’s account as distinctively retributivist. This is not because retributivism and consequentialism are inherently incompatible. Retributivism is too broad a philosophical church to permit this inference.26 Rather, it is because the restoration of rights thesis seems to articulate the social functionality of punishment and seems to ignore that aspect of punishment which is directed toward the particularity of the specific violation. The dealings that the punitive agency has with the criminal seem to be secondary to the efficacy of punishment as the instrument of social purposes. If we read Hegel in the fashion that Cooper recommends and I have reconstructed, we must conclude either that the theory, as articulated thus far, does not accomplish its retributive intent, as Wood argues,27 or that it is incomplete. This latter is my claim. The restoration of right is a public function: the objective face of the reconciliation that the fact of crime necessitates. The subjective aspect of the reconciliation of right with itself is revealed in the response of the criminal. To repeat: “reconciliation applies … subjectively … to the criminal in that his law, which is known by him and is valid for him and for his protection, is enforced on him in such a way that he himself finds in it the satisfaction of justice and merely the enactment of what is proper to him” (PR §220). We need to understand these ambitious claims. And we need to show how they amplify, rather than contradict, Hegel’s concern for the restoration of right.

THE CRIMINAL’S RIGHT TO PUNISHMENT

The criminal objectively denies the right of the victim to his property, rejects his moral status as a person with a capacity for rights, and, in “Civil Society,” repudiates the regime of rights as this is actualized, that is, institutionalized in the workings of the economy and the justice system by which it is regulated. Whatever the criminal says he is up to, this is what his action bespeaks. This is what Hegel calls, teasingly, “the right of the objectivity of the action … to assert itself as known and willed by the subject as a thinking agent” (PR §120), as distinct from the moral subject’s right of intention. The rational agent knows that his actions disclose his intentions since he employs the same rational schema as others do to understand the actions of his fellows. And he knows how they will interpret his actions when he commits a crime. Irrationality (and, consequently, absent or diminished responsibility) is the cost of dislocation between the specification of his own intention given by the agent and the description of his action given by fully informed observers. So the criminal, if he is a real criminal, either knows or is culpably ignorant of the fact that he is striking both at his victim’s rights and those recognized by his society. Whether or not he is alienated, he is in conflict with his society and reconciliation is called for.28

Let us return to Hegel’s treatment of punishment in “Abstract Right” to see how these facts are treated there. To focus discussion, I cite the paragraph we shall dissect:

The injury [Verletzung] which is inflicted on the criminal is not only just in itself (ana since it is just, it is at the same time his will as it is in itself, an existence [Dasein] of his freedom, his right); it is also a right for the criminal himself,29 that is, a right posited in his existent will, in his action. For it is implicit in his action, as that of a rational being, that it is universal in character, and that, by performing it, he has set up a law which he has recognized for himself in his action, and under which he may therefore be subsumed as under his right. (PR §100)

This argument suggests a variety of characterizations that have both a classical provenance and contemporary support. Punishment is justified since the rights violator has forfeited his rights.30 Punishment is justified on the basis of the criminal’s consent.31 Punishment is justified as the issue of a social contract.32 None of these positions, as outlined by their ancient and modern protagonists, quite captures Hegel’s approach. But their variety alerts us to the possibility of rewriting his argument through this tradition. A successful reconstruction will conclude, as each of the above approaches attempts to do, that the criminal accepts the necessity of his own punishment. Thus it will present a process of practical reasoning that can successfully be imputed to the criminal.

The clearest way of reading the argument of PR §100 is stated carefully by Wood. It works back from the statement that the criminal’s action sets up a law under which the criminal is himself brought. Through his action, the criminal is saying (in the sense of PR §120) and knows or ought to know that he will be taken as saying (mens rea), “It’s fine to violate the rights of the victim.” The implication of the law implicit in the action is that it is fine to violate the rights of the criminal in the same measure. The punitive response is justifiable ad hominem, in accordance with the principle of his own actions, a principle which, as a rational agent, he cannot reject. This conclusion is stated explicitly in the Propädeutik: “It [the principle or law of the criminal’s action] is valid only for the one who committed it because he alone recognizes it by his action and no one else. He himself, therefore, is essentially subject to this principle or ‘Law’ and it must be carried out upon him.”33

The difficulty, or rather, one difficulty,34 with this argument is that it does little more than dress up the thought that the criminal himself cannot complain if he is treated in the same fashion that he treats the victim. As put, this thought has much to be said for it, but it is hard to see how it can be acceptable to the punishing agency. The criminal’s act, we recall, also has the dimension of a ‘negatively infinite judgment’; he denies the victim’s capacity for rights—and, in “Civil Society,” the whole regime of right, as this applies to its members. The punishing agency cannot be thought to assert this as the principle of its action, and, explicitly, it does not. Punishment recognizes the criminal’s own right; through it he “is honoured as a rational being” (PR §100R). If this is true, how can the criminal’s law be applied validly by the victim or the courts? One might say: because his law is valid for him alone, it is applied to him alone. But this is to lose the validity of the license which the criminal act yields to the punisher. And in “Civil Society,” it would not be applied to him alone. It would be applied to any criminal, in accordance with the law and its proper procedures. The last thing the punishing agency should be doing is adopting the moral perspective of the criminal.

This response to Hegel also dresses up an old saw: two wrongs don’t make a right. If this play on the universal character of the criminal’s deed is all that Hegel manages at PR §100, it should be judged a weak and unconvincing argument. I think we can find an alternative that is stronger and more satisfactory.

A HYPOTHETICAL CONTRACT MODEL OF PUNISHMENT

To do this we return to the perspective of “Abstract Right” and note that all rational agents claim the moral status of person, assert their capacity for rights, and endorse the imperative of right commanding them to be persons and respect the personality of others. Respecting the personality of others requires one to respect their rights claims. All of this is clear to the criminal. If it is not, he is not a rational agent responsible for his actions. I claim that we can develop from these initial postulates a full justification of punishment on the hypothetical contract model. Thus:

  1. Individuals claim rights against each other and recognize that others claim equivalent rights against themselves.

  2. They see no prospect of others respecting their rights while they themselves are immune to the rights claims of others.

  3. They suspect that others may attempt to become free riders on the convention of respect for rights, since they understand that the attractions of wrongdoing with impunity on their own part is enhanced by the predictability of others’ behavior—which they may be able to exploit.

  4. Hence, they demand a guarantee of good faith in the principles of rights from those others who avow them, and they are willing to give such a guarantee themselves.

  5. The guarantee which is universally offered and taken up is a recognition of the legitimacy of punishment exacted against criminals; a guarantee taken up against criminals on the part of all contractors, against themselves, of course, should they turn out to be criminals.

  6. They accept that punishment may take the form of actions which, in other contexts, would amount to a violation of their rights.

  7. Hence, those who wish their rights to be promoted and protected are willing to alienate their rights should they, themselves, violate the rights of others.35

How much of this argument can we recognize in Hegel’s texts? At first sight, admittedly, not much. Point 1 certainly is a Hegelian thesis, and so is the conclusion at point 7. This explicates the thought that, both in itself and explicitly, punishment is the right of the criminal. In some sense, the criminal consents to his punishment. What of the steps in between? I would be prepared to defend point 2 as a Hegelian position. Besides being one of the sources of the universality explicit in the imperative of right, it is a clear implication of the demand that law be universally valid and universally known through its promulgation in a public legal code and its prosecution in transparent legal processes (PR §§209–211, 215–217). We can take it, too, that the rational agent will endorse the principles of “Abstract Right” on which the administration of justice is, in part, founded. Similarly, I think point 6 can be defended as an implication of the Hegelian texts, which stress that the appropriate measure of punishment will be equal to the crime in point of its value (PR §§101, 214). But that is as far as the texts will take us.

Why did Hegel reject the sort of argument that Beccaria, and following him, Rousseau, had provided (although they disagreed on the specific issue of capital punishment)? Why was he so hostile to a contract argument in favor of punishment? The answer to these questions is that he was hostile to the social contract model of the ontology of the state, of the relation between the state and its constituent citizens, and could not separate the philosophical problems of the legitimacy of sovereign authority and the legitimacy of state punishment. See how swiftly Hegel’s discussion of Beccaria on punishment at PR §100 modulates into a discussion of the social contract account of the state.36 Hegel may well be correct in his criticism of the implications of contract theory for an acceptable social and nation-state ontology, and by implication for an acceptable account of political obligation. But his central doctrines concerning punishment are lodged, as we have seen, in his discussion of our moral nature as persons, distinct and discrete bearers of rights. Thinking of ourselves as persons, as Hegel insists that we must, we may have quite sufficient cognitive-cum-rational resources to work out, from the perspective of the rational agent who turns out to be a criminal, that his liability to punishment, her legitimate punishability, is the inevitable normative consequence of his criminality.

All that is necessary for each person to reach this conclusion is that he review the claims he (and, by implication, each other person) is prone to make, the duties he (and all others) are likely to demand (and they and he to accept on their own part), together with some tariff of punishment as the recognized normative consequence of anyone’s actual failure to respect the claims made and fulfill the duties they entail.37 This argument does not rest on a contract theory of the state or a contract theory of rights. In principle, it is quite open to anyone to reject contract theories in these fields (and contract theories of justice and morality, too) and still endorse a contract theory of punishment.

The contract is, of course, hypothetical. It binds not as a real contract might, but solely in virtue of the independent persuasiveness of the premises it contains.38 The version of the contract argument that I give above details the major elements of such a deduction as is available to the criminal contemplating his punishment, and, so far as I can see, it can be advanced without reference to our political standing as citizens of a nation-state. If this is right, Hegel was quite wrong to disallow this model of practical reason as germane to the deduction of our duties in respect of punitive institutions. His own favored argument, which has criminals, through their actions, laying down a law under which they in turn are brought, employs premises that are no less individualistic, although it exhibits the striking flaw of reaching the wrong conclusion.

Disregard for the moment Hegel’s obliviousness to the resources of this model of practical reason when defending punishment. Ask instead: What is its plausibility, what are its limitations? I will not defend its plausibilty here, but its usefulness is clear. It displays how the criminal, on condition of his rational agency, must accept as legitimate the punishment his actions call forth. This is exactly the conclusion Hegel’s argument strains for. Its weakness is exactly the same as the weakness Wood notices in respect of the incompleteness of the consent version of Hegel’s theory as he interprets it: Hegel’s argument does not give “a positive reason why the state should actually inflict punishment. That is, granting that punishment does no injustice, we might still ask if there is any good reason for the state to punish.”39 Now this is a limitation on Hegel’s theory following Wood’s interpretation. If, on the other hand, Hegel were to have employed the contract theory I have recommended, he would have a perfectly straightforward reply. At this point in the dialectic he could bring together the ‘objective’ and ‘subjective’ strands of argument.

THE ALIGNMENT OF “OBJECTIVITY” AND “SUBJECTIVITY”

I have defended the claim that punishment is justified as restoring the right, but argued that this insight is not distinctively retributivist and, in any case, may be lost on the criminal.40 Next, I claimed that the criminal may be made to accept the legitimacy of his punishment through a contract argument, but suggested, following Wood, that the argument is incomplete. Two things are needed to remedy these putative defects. In the first place, the contract argument requires a lemma to the effect that the criminal must recognize that his punishment is legitimate insofar as it procures the restoration of right—a goal that he himself endorses as appropriate for the law to effect. All citizens accept the validity of the goal of the restoration of rights, not because this is a valuable social function of punishment, but because it is necessary for the protection of the rights that they themselves claim. So the state must punish criminals if it is to serve the purpose of protecting rights.41 Pace Wood, this amounts to a positive reason for the state to punish, and not merely a license. Second, viewing the matter from the perspective of the concern to restore the right, this becomes a properly retributive function just as soon as it is realized that the specific institutions that effect the restoration do so in a manner that enlists the rational endorsement of citizens. The transparency that is at the heart of the administration of justice serves to embed the reasoning of the contractor in the institutions that identify, prosecute, and punish criminality. The restoration of right is accomplished by social mechanisms that demonstrate to all, honest and criminal citizens alike, the nature of their rights, their concomitant duties, and the penalties to be imposed for noncompliance. The institutions make explicit, through their laws, processes, and punitive regime, the rights for which persons demand protection. Notoriously, this knowledge, explicit in the alignment of subjectivity and objectivity, is not available to the criminal in “Abstract Right” where punishment takes the form of revenge and revenge breeds vendetta (PR §102). But it is available in “Civil Society.”

The restoration of right is achievable wherever the administration of justice is so structured as to make transparent to citizens the rights they may justly claim, the duties they must fulfill, and the penalties they will incur for noncompliance. In “Civil Society,” the reconciliation of right with right, which is necessitated by the fact of crime, is achieved by the rule of law, serving ends which all citizens endorse as promoting their interests and dispensed in courts of law which make that endorsement evident to reflective, rational agents. The hypothetical contract device explicates the structure of practical reason by which all rational agents, criminals included, can be presumed to accept the legitimacy of punishment. By embedding the necessity of the restoration of right within the practical reason of all rational agents, and by describing how the administration of justice articulates the self-understanding acquired by these practical reasoners, we deflect the outstanding objections. Against all his instincts, but following through on his most convincing arguments, Hegel should have employed a hypothetical contract argument to defend the institution of punishment.42

NOTES

This essay has been improved through my having to reflect on the criticisms of the Hegel Society of America referees. They may disagree! But thanks are due for their detailed comments. Thanks are also due to those who helped me improve the essay following its presentation at the HSA meeting. Mark Tunick, Robert Berman, Robert Williams, and Stephen Houlgate were particularly helpful in getting me to change and clarify the argument.

1. I stress “the normative framework of ‘Abstract Right’,” meaning by this the concept (Begriff) of “Abstract Right,” ignoring Hegel’s conclusion that the concept cannot be actualized since a positive realm of “Abstract Right” would be riven by vendetta (PR §§102–103). As we shall see, the concept of “Abstract Right” is actualized within “Civil Society.” Citations of PR refer to G. W. F. Hegel, Elements of the Philosophy of Right, ed. Allen W. Wood, tr. H. B. Nis-bet (Cambridge: Cambridge University Press, 1991).

2. I have in mind the views of Charles Taylor and Michael Sandel, among others. See Charles Taylor, “Atomism,” in Philosophical Papers, vol. 2 (Cambridge: Cambridge University Press, 1985), and “Cross-Purposes: The Liberal-Communitarian Debate,” in Liberalism and the Moral Life, ed. N. Rosenblum (Cambridge, Mass.: Harvard University Press, 1989); Michael Sandel, Liberalism and the Limits of Justice (Cambridge: Cambridge University Press, 1982).

3. This is one reason why “Abstract Right” is abstract.

4. Allen W. Wood fails to notice the careful logic of Hegel’s initial presentation of his theory of rights, and this infects his reading of Hegel on punishment. Discussing David Cooper’s interpretation, he writes, “In the theory of abstract right developed in PR §§34–80, he regards abstract right as conferring on persons both a permission and a warrant to exercise arbitrary choice within a limited sphere (PR §§38, 41). Only later does he focus attention on the claim that others should respect this sphere (cf. PR §49R).” Allen W. Wood, Hegel’s Ethical Thought (Cambridge: Cambridge University Press, 1990), p. 111. PR §38, as quoted above, with its characterization of the rules of right as negative, as prohibitions, shows that this is a mistake—as does the imperative of right itself (PR §36), enjoining us to “respect others as persons,” which PR §38 glosses in negative terms. The ‘permission or warrant’ Hegel characterizes at PR §38 is not merely correlative to the duty of noninterference; the “positive form of commandments of right is, in its ultimate content, based on prohibition.”

5. It is not a coincidence that this structure mirrors that of Nozick’s entitlement theory of justice, which consists of principles governing acquisition, transfer and rectification. See R. Nozick, Anarchy, State and Utopia (Oxford: Blackwell, 1974), pp. 150–51. In “Abstract Right,” Hegel is presenting the core doctrines of atomistic liberalism. Although he traces the origins of this style of morality to the Roman world, in effect he is presenting a digest of modern, that is, post-Reformation, rights theory as one (but pace Nozick et al., only one) element of the way moderns think about ethics. Pace the communitarians, by contrast, although rights theory is not the whole story concerning our ethical obligations, it is ineliminable. Hegel insists, surely correctly, that we just do claim rights nowadays, seeing ourselves as distinct individuals. The metaphysical blankness of the concept of ‘person’ is remedied by its amplification, as rights are specified, and as other modes of self-identification are added. It is certainly not rejected as conceptually inchoate.

6. Robert R. Williams notices this point. See Hegel’s Ethics of Recognition (Berkeley, Los Angeles, and London: University of California Press, 1997), pp. 155–57. “Wrong clarifies the substantial nature of right” (p. 155). See also I. Primoratz, justifying Legal Punishment (Atlantic Highlands, N.J.: Humanities Press, 1989), p. 74.

7. The point is repeated in Hotho’s notes: following punishment, “right acquires the determination of something fixed and valid. … Whereas right previously had only an immediate being, it now becomes actual as it returns out of its negation; for actuality is that which is effective” (PR, Addition to §82).

8. Locke and Rousseau tackle the problem head on. See J. Locke, Second Treatise of Government, §§6–9, 16–19, 87–88 and elsewhere; J.-J. Rousseau, The Social Contract, bk. II, ch. V. Nozick notices the difficulty and skates over it quickly, suggesting that perhaps a social contract argument promises a solution. Anarchy, State and Utopia, pp. 137–38. In “Punishment and Rights,” Punishment and Political Theory, ed. M. Matravers (Oxford: Hart, 1998), I discuss a range of solutions to this problem.

9. Lewis P. Hinchman claims that Hegel is addressing this problem. See “Hegel’s Theory of Crime and Punishment,” The Review of Politics, vol. 44, 1982, p. 543. Allen W. Wood hints at such a reading. See Hegel’s Ethical Thought, pp. 109–10, as does Robert Williams, Hegel’s Ethics of Recognition: “Although in general coercion is a violation of right, sometimes coercion is necessary, and if necessary, then justified. Our question is, How is coercion of freedom—which is strictly speaking impossible—nevertheless possible?” (p. 158).

10. Michael O. Hardimon, Hegel’s Social Philosophy: The Project of Reconciliation (Cambridge: Cambridge University Press, 1994), pp. 95ff.

11. “Hypothetical” gives us yet another meaning of abstract in “Abstract Right.” The norms of “Abstract Right” require a “Civil Society” for their actualization.

12. Crime “constitutes a negatively infinite judgement’” (PR §95). In the paragraph cited, this is taken as a claim that crime negates not merely my will embodied as it is in property, but also “the universal and infinite element in the predicate ‘mine’—i.e. my capacity for rights.” At PR §218, where crime is considered in the context of “Civil Society,” the scope of the infinite judgment is widened, so that we are to see the criminal act as striking at society, presumably at the whole normative regime. In consequence, criminal action is viewed as “a danger to society.” Which actions constitute a danger to society beyond violations of individual rights? This issue is discussed by Mark Tunick, Hegel’s Political Philosophy (Princeton, N.J.: Princeton University Press, 1992), pp. 110–13.

13. Allen W. Wood, Hegel’s Ethical Thought, pp. 112–13.

14. It may be objected that these sayings are not metaphors, that they characterize the structure of Hegel’s argument in terms of his distinctive speculative logic. I have no doubt that this is what Hegel thought he was doing. At this point, I don’t want to pick a fight with those who insist that there is a distinctive and genuine logic here that constitutes good argument. On the other hand, I do want to insist that the argument which is being sketched is unpacked in the detail of the discussion that follows. I want to allow Hegel the opportunity to be persuasive to those for whom the speculative logic is a mystery, since this was a dialectical strategy he himself was ready to exploit, in both the text and the lectures on the Philosophy of Right. (I make this comment in order to address objections strongly put at the University of Georgia meeting.)

15. David E. Cooper, “Hegel’s Theory of Punishment,” in Hegel’s Political Philosophy: Problems and Perspectives, ed. Z. A. Pelczynski (Cambridge: Cambridge University Press, 1971), cited at pp. 162–63.

16. Peter G. Steinberger, “Hegel on Crime and Punishment,” American Political Science Review, 1983, vol. 77, pp. 858–70.

17. Joel Feinberg, “The Expressive Function of Punishment,” in Philosophical Perspectives on Punishment, ed. Gertrude Ezorsky (Albany: SUNY Press, 1972).

18. Allen W. Wood, Hegel’s Ethical Thought, pp. 111–13.

19. Stephen Houlgate, “Hegel’s Ethical Thought,” Bulletin of the Hegel Society of Great Britain, no. 25, 1992, cited at p. 12.

20. Allen W. Wood, “Reply,” Bulletin of the Hegel Society of Great Britain, no. 25, 1992, cited at p. 44.

21. “It would be impossible for a society to leave a crime unpunished— since the crime would then be posited as right” (PR §218A). Mark Tunick illustrates this point with a good story from the Upper Congo, concluding that “without the old woman’s response [of denunciation] there would have been no crime: not because if nobody discovers it happened it didn’t happen, but because if nobody declares it’s wrong, it’s not.” For further details, see M. Tunick, Hegel’s Political Philosophy, pp. 78–80. Likewise, if technical violations of the law go openly unpunished, as was the case in Scotland before the law on consensual homosexual acts was brought in line with the reforms effected in English law, we may judge that no wrong is committed. Where prosecution is capricious and arbitrary—this is the early history of boxing in Britain; sometimes the magistrates stopped the fights, sometimes they sat in the front row—the law is an ass because the right is indeterminate.

22. Allen Wood notices this, commenting that the methodology is as defunct as Hegelian speculative logic! Later, he claims, intriguingly, that the “conceptual argument” “deliberately leaves unasked the critical qestion whether ‘our’ talk, and the ‘form of life’ it reflects, can be given any rational justification … [hence] it embodies a fundamentally Anti-Hegelian attitude toward philosophy.” See “Reply,” pp. 43–45.

23. For another example of this line of criticism, see S. I. Benn and R. S. Peters, Social Principles and the Democratic State (London: George Allen and Unwin, 1959), p. 177.

24. See n. 21.

25. For these details (and more), see PR §§209–229.

26. In conversation, Mark Tunick convinced me of this.

27. Allen Wood, Hegel’s Ethical Theory, pp. 110–12.

28. There is a general difficulty in the background here concerning Hegel’s characterization of the content of the mind of the criminal. What exactly must we be able to claim about what he knows about what he is doing? Crime is not, after all, nonmalicious wrongdoing. Is it enough that we make true judgments concerning what the criminal ought to know? These are standard problems in applying the doctrine of mens rea, and I shall have more to say about this later.

29. Stephen Houlgate puts it to me that “a right for the criminal himself” mistranslates “ein Recht an den Verbrecher selbst.” The Knox translation, reading “a right established within the criminal himself” (G. W. F. Hegel, Hegel’s Philosophy of Right, tr. T. M. Knox (Oxford: Clarendon Press, 1952), is better because it does not carry the implication that “a right for [my italics] the criminal himself” is a right of which the criminal is fully aware, or, perhaps, even stronger, actively demands. I can’t say whether the translator’s “for” carries the implication of full self-knowledge which is distinctive of Hegel’s für sich. The precise point is that the criminal’s action bespeaks his existing will. Neither the original nor its translation should be taken to entail that the criminal explicitly asserts the law of his action, as it were giving a running commentary on what he is doing. It is implied that he ought to accept such a law in accordance with the “right of objectivity of the action” (PR §120). Such an implication can properly be represented as the conclusion of a hypothetical contract argument, of which more later. See n. 28 supra.

30. John Locke, Second Treatise of Government: “[The criminal] so far becomes degenerate, and declares himself to be quit the Principles of Human Nature and to be a noxious creature” (§10); “In so revolting from his kind to that of Beasts and by making Force which is theirs, to be his rule of right, he renders himself liable to be despised … as any other wild beast or noxious brute” (§172). For a modern argument to the effect that the rights of the criminal, which would otherwise prohibit boundary crossings, are forfeit, see Alan H. Goldman, “The Paradox of Punishment,” Philosophy and Public Affairs 9 (1979).

31. J.-J. Rousseau, The Social Contract: “the death penalty inflicted upon criminals may be looked on in much the same light [as conscription]: it is in order that we may not fall victims to an assassin that we consent to die if we ourselves turn assassins” (bk. II, ch. IV). For a modern treatment, see C. S. Nino, “A Consensual Theory of Punishment,” Philosophy and Public Affairs 12 (1983).

32. Cesare Beccaria, On Crimes and Punishments, and Other Writings, ed. R. Bellamy (Cambridge: Cambridge University Press, 1995). For a modern treatment, see Jeffrie Murphy, “Marxism and Retribution,” Philosophy and Public Affairs 2 (1973).

33. G. W. F. Hegel, The Philosophical Propaedeutic, ed. M. George and A. Vincent, tr. H. V. Miller (Oxford: Blackwell, 1986), p. 31. G. W. F. Hegel, Philosophische Propãdeutik, Sämtliche Werke, ed. H. Glockner (Stuttgart: Frommann Verlag, 1971), vol. 3, p. 68. This argument is emphasized by Igor Primoratz in Banquos Geist. Hegels Theorie der Strafe. Hegel-Studien, Beiheft 29, 1986, and fully endorsed in Justifying Legal Punishment (Atlantic Highlands, N.J. and London: Humanities Press, 1989), ch. 4, especially pp.76–79. Robert Williams, Hegel’s Ethics of Recognition, also follows this line: “Hegel believes that retribution is just because retributive punishment is simply the reversal of the offense; that is, it demands that the principle of the transgression be applied to the offender” (p. 168). Neither Hegel nor his interpreters see that this is a very bad argument. If the principle of the transgression endorses the violation of rights, that should disqualify it as a principle to be applied by any other agency. The act of punishment should make this disqualification explicit rather than employ the invalid principle to vindicate punishment. By contrast, the thought that the criminal cannot legitimately complain is very useful when it is the appropriate measure of punishment that is being considered, as Primoratz sees (Justifying Legal Punishment, chs. 4.5, 5.2).

34. There are others. How do we get over the problem of the criminal who explicitly disavows the imputed saying? If we answer that he ought to recognize that he will be taken to be saying this, we are adducing his hypothetical consent to the proposition: he would accept it if he were rational, fully informed, sincere, and apprised of the logical consequences of other propositions he affirms. This takes us halfway down the road of hypothetical contract, wherein rational agents take into account what others, as well as themselves, may reasonably reject as the basis for agreement on principles.

35. I defend this argument in “Punishment and Rights,” where I argue that it shores up weaknesses in both forfeiture and consent arguments to the conclusion that the criminal wills his own punishment.

36. Contract theories of the state are criticized in PR at §§75 and 258 in addition to the above.

37. If this claim is hard to stomach, remember that Hegel himself insists that the justice of punishment can be demonstrated on the basis of the very limited normative resources available to the person under “Abstract Right.” “The state is not a necessary condition of justice in itself,” Hegel claims at §100R.

38. Famously, this point is made by Ronald Dworkin, “The Original Position,” in Reading Rawls, ed. N. Daniels (Oxford: Blackwell, 1975), p. 18. Despite Hegel’s knowledge of the contract tradition, especially as employed by Rousseau, Kant, and Fichte, I don’t think he managed to unravel the striking differences between the actual and hypothetical forms of the argument. Had he done so, he would have seen the aptness of the hypothetical contract structure for modeling the arguments of “Abstract Right” and the sections of “Civil Society” that discuss the protection of persons’ rights under the “Administration of Justice.” The version of hypothetical contract theory I employ here derives from T. M. Scanlon, “Contractualism and Utilitarianism,” in Utilitarianism and Beyond, ed. A. Sen and B. Williams (Cambridge: Cambridge University Press, 1982). For an application of this style of normative ethics that is useful in thinking about punishment, see T. Nagel, Equality and Partiality (Oxford: Oxford University Press, 1991), ch. 4.

39. Allen Wood, Hegel’s Ethical Theory, p. 116.

40. To be specific, the implication is not that this is a bad argument. I believe it’s a good one. Rather, it looks inconsistent with other claims that Hegel makes about the retributive intent of punishment.

41. To deflect objections: this is not the only purpose of the state. Rather it is just the best way of securing that freedom which is required by the fact of our personhood.