Honor and Revenge: A Theory of Punishment

By Whitley R.P. Kaufman

This book addresses the problem of justifying the institution of criminal punishment.   It examines the “paradox of retribution”: the fact that we cannot seem to reject the intuition that punishment is morally required, and yet we cannot (even after two thousand years of philosophical debate) find a morally legitimate basis for inflicting harm on wrongdoers.  The book comes at a time when a new “abolitionist” movement has arisen, a movement that argues that we should give up the search for justification and accept that punishment is morally unjustifiable and should be discontinued immediately.  This book, however, proposes a new approach to the retributive theory of punishment, arguing that it should be understood in its traditional formulation that has been long forgotten or dismissed: that punishment is essentially a defense of the honor of the victim.  Properly understood, this can give us the possibility of a legitimate moral justification for the institution of punishment.​

• Language: English
• Publisher: Springer
• Release date: Aug 28, 2012
• ISBN: 9789400748453

Book preview

Whitley R.P. KaufmanLaw and Philosophy LibraryHonor and Revenge: A Theory of Punishment201310.1007/978-94-007-4845-3_1© Springer Science+Business Media Dordrecht 2012

1. The Problem of Punishment

Whitley R. P. Kaufman¹ 

(1)

Department of Philosophy, University of Massachusetts Lowell, Lowell, MA, USA

Abstract

This chapter sets out the problem of punishment. Punishment involves inflicting massive harm (loss of life, freedom, etc.) on many millions of people, yet we have no clear idea whether the practice is morally justified. Indeed, we have no clear idea as to the very purpose of punishment: do we punish to prevent crime? Or to give the criminal his just deserts? Or some combination of these two? Or something else? It is simply extraordinary that we continue to inflict these tremendous harms on people without even having a clear idea of why we are doing it. Even worse, the two leading theories of punishment (deterrence and retribution) are themselves morally problematic. Deterrence, a theory based on the utilitarian moral theory, is flatly inconsistent with basic moral principles in that it permits harming people as a means to a greater good. But retribution seems no better, for it seems to be gratuitous, pointless harm inflicted after the crime is already finished, and hence seems morally incomprehensible. The moral situation is so bad that a growing movement called Abolitionism has begun to call for the elimination of punishment, on moral grounds. It is the aim of this book to provide a new defense of the retributive theory that is consistent with moral principles.

The institution of criminal punishment is in a scandalous state, and what is worse, we largely refuse to acknowledge it. It is certainly true that there is constant debate over such issues as whether the death penalty should be eliminated, whether we punish too hardly, what constitutional rights prison inmates have, and whether sentencing policy is race-biased. Yet we ignore the far more fundamental problem, namely that we do not know whether punishment itself is morally justified. Even worse, we do not have any clear idea of just what the purpose of punishment is supposed to be. Is it to give wrongdoers their just deserts? To deter others? To prevent criminals from committing future crime? To reform them? Or some combination of some or all of these? Needless to say, if we do not know why we punish, we cannot even begin to ask the question whether punishment is morally justified. Even worse, as we will see, none of the proposed explanations for why we punish appear to provide a moral justification for punishment, a fact that has long been recognized by moral philosophers. Hence the current intolerable situation, in which we have no good basis for believing that punishment is morally justified and ample reason to suggest that it is not. Yet there is no evidence that this problem is the least concern for any but a small number of moral philosophers. It is precisely the purpose of this book to address the problem of punishment and raise the question as to how punishment might or might not be morally justified.

We begin by giving a preview of the overall argument. The present chapter sets out the problem of punishment in its starkest terms, demonstrating that at least prima facie, the practice of punishment appears obviously morally unjustified. In Chap. 2, we examine whether punishment can be justified as a means of preventing crime, arguing that the policy of deterrence (punishing Joe to deter Jim) is morally unacceptable, but that other crime-prevention purposes such as incapacitation or moral reform can be morally defended, albeit under strict limits. However, crime prevention does not get to the heart of what we mean by genuine punishment, which is essentially retributive in nature: inflicting punishment as a response to wrongdoing. Chapter 3 takes on the Retributive Theory of punishment, assessing the many attempts to justify retribution and demonstrating that all of them fail, and fail badly. Not only do they fail to provide a moral justification for retribution, but they do not even state a coherent purpose for which punishment is inflicted. We need a fresh start in making sense of retributive punishment. Chapter 4 examines and rejects the attempts to combine utilitarian and retributive theories in order to create a mixed theory that purportedly would have the best of both theories.

Chapter 5 begins a new approach by calling into question the almost universally accepted claim that retributive punishment is essentially distinct from revenge; instead, we argue that the two are essentially identical, and that the very purpose of retribution is to gain revenge on wrongdoers. The question then becomes whether revenge is morally justified. Chapter 6 presents the case that revenge can be understood as the defense of a crucial personal value, one’s honor or self-respect. Further, we argue that the defense of honor provides the only plausible explanation of why we punish retributively, and that it provides at least in principle a potential moral justification of punishment. However, the case is only prima facie and not dispositive. Chapters 7 and 8 address the question of the nature and value of honor and whether it is a sufficiently important value to ground the practice of retributive punishment. We argue that this question does not have an easy answer, contrary to the view of both sides of the currently polarized debate. Explaining the retributive impulse as the defense of honor, at the very least however, provides a basis for reframing the debate so as to escape the current stalemate.

Thus the central aim of this book is to help find a way out of the present impasse in the punishment debate. In fact, there are two distinct impasses to be dealt with. First, there is the endless back and forth between utilitarian and retributive theories (as well as various unsuccessful attempts to combine them into a single theory). If the present account is correct, we can understand punishment as having more than one purpose, including both crime prevention and retribution, and each of them can (within strict limits) provide a moral foundation for the practice. Second, there is the current impasse between the Abolitionists who tend to hold that punishment is obviously unjustified, and the retributivists who hold that punishment is self-evidently morally justified. By understanding the true underlying purpose of retribution, it is my hope that we can lay the foundation for a more productive debate between these two extreme views.

But before we get there, we will need to establish just how serious the problem of punishment is, and furthermore that it is not merely an academic exercise but one that is an urgent and compelling social issue. This fact becomes obvious once one considers the vast scope of harm caused by punitive practices in ours and every society. The institution of criminal punishment involves the infliction of human suffering on a massive scale. There are some two million people currently incarcerated in the United States, with seven million total persons under the supervision of the criminal justice system in one way or another (on parole, probation, etc.). The fact that punishment falls disproportionately on minority races and the lower social classes only exacerbates the problem. Only warfare exceeds in magnitude the level of socially approved, deliberately inflicted harm against people. War, however, is considered an exceptional, emergency state of affairs, a last resort given the enormity of the harm it causes. In this respect, it is quite unlike the institution of punishment, which is taken to be an ordinary, normal institution, one that it is widely assumed will always be with us. It is not too utopian to look forward to a future age of world peace through mutual agreement of nations, when war will be a relic of the past. Yet there is no such expectation for punishment. Indeed, we do not have an idea of what sort of changes could bring about a world in which punishment is no longer needed. Finally, to place the two institutions in stark relief, in war we tend to see most of the harm done as regrettable and undesired, whereas most people view the harm inflicted in punishment as a positive good, something that is deserved, that – paradoxically – makes the world better, not worse.

There is yet another oddity about punishment. If an alien visitor from another planet were to arrive here and attempt to understand our society, he would likely be perplexed by the very strange beliefs of humans in regards to justice. He would presumably not be surprised at the fact that humans express outrage at the many serious wrongs inflicted on others: killing, rape, beatings, mutilation, physical abuse, theft, kidnaping, and so forth. But it is our response to such wrongs that might puzzle the visitor, for we in the name of justice inflict the very same sorts of physical harm on the wrongdoers. Thus consider what we do to criminals: we kill them, take away their freedom, confiscate their property, or physically beat them (while in our society corporal punishment has largely disappeared, it is still practiced in much of the world, and throughout most of our history beatings, mutilations and torture were regularly part of punishment). It is hard to think of a sort of wrong that we do not in turn inflict on criminals. Notably, the Thirteenth Amendment to the United States Constitution prohibiting slavery makes a single exception: Neither slavery nor involuntary servitude, except as a punishment for crime…shall exist in the United States. Virtually every major wrongful act committed on another human being becomes morally permissible under the rubric of punishment. (The one apparent exception is rape, but prison rape has become so common that it has widely come to be seen as an expected element of punishment and is even frequently used as a threat by police officers and prosecutors. Further, one can even find moral philosophers who call for punitive rape: thus Angelo Corlett declares that we ought to develop a rape machine that can be programmed to reproduce…the kinds of experiences that a rape victim had (2006, 9).) But how can the worst wrongs become morally right dependent on the context, or on who is doing them? If these acts are wrong when the criminal does them, how can they become right when the state does them?

Even worse, the most extreme historical examples of deliberate cruelty inflicted by one human being on another have been, arguably, those inflicted under the name of justified punishment. Foucault begins his book Discipline and Punish with the story of the regicide Damiens in eighteenth century France, who was sentenced thus: the flesh will be torn from his breasts, arms, thighs, and calves with red-hot pincers, his right hand, holding the knife with which he committed the said parricide, burnt with sulphur, and, on those places where the flesh will be torn away, poured molten lead, boiling oil, burning resin, wax, and sulphur melted together and then his body drawn and quartered by four horses and his limbs and body consumed by fire, reduced to ashes, and his ashes scattered to the winds (1979, 3). Oscar Wilde, himself punished for his homosexuality, observed: As one reads history…one is absolutely sickened not by the crimes the wicked have committed, but by the punishments the good have inflicted.¹ The very fact that our Bill of Rights explicitly prohibits cruel and unusual punishments demonstrates how concerned the founders were about the many possibilities of cruelty inflicted in the name of punishment.

In light of this situation, it is astonishing to recognize that we have no clear consensus as to why we punish people. It stands to reason that one should have only the weightiest and most profound reason for inflicting harm on people, all the more so the greater the scope and severity of that harm. Yet there is an oddly disturbing lack of urgency in investigating the purpose of punishment, and thereby being able to ask the question whether our rationale is sufficiently morally important. When asked, many people will say that the purpose of punishment is crime prevention. Yet as we will see, there is surprisingly little clear evidence of the deterrent effect of punishment, nor much public interest in establishing whether or not there is a real deterrent effect. Again, basic moral principles would seem to dictate that, if deterrence were the purpose of punishment, we would make sure to have extremely strong and even definitive evidence that punishment does deter, as well as detailed evidence as to how much deterrence is produced by a given level of punishment. Consider by way of comparison the extraordinary burden is placed on the testing of pharmaceuticals, so as to avoid even very small risks that the drugs might cause harmful effects on people. Punishment involves harm that is certain and that is massive, and yet we make little if any effort to determine just what the purported benefits are and whether they are worth the costs. This is all the more remarkable given how extraordinary those costs (economic as well as personal) are, something on the order of $75 billion of taxpayer money per year in the United States for incarceration alone, and far more than that for the total cost of the system.

In fact, there is good evidence (presented in Chap. 2) that most people, even when they say deterrence is the purpose of crime, really in fact have a very different purpose in mind, that of retribution. The retributive justification of punishment is that the criminal deserves to be punished simply because he has committed a crime – regardless of the deterrent or other beneficial consequences of punishing him. The only justification of punishment, on this view, is that the criminal deserves to be punished. But this rationale is even more perplexing (a matter to be discussed in detail in Chap. 3). The retributivist insists that punishment is not to be justified by any future social benefits it produces, but by desert alone. But this almost seems to concede – bizarrely – that punishment has no purpose at all, making it if anything even less comprehensible (and also effectively insulating it from the need for moral justification). The idea that we should harm a wrongdoer simply because he deserves it is simply mysterious, lacking in any apparent rational or moral basis.

For some 2,000 years, the debate over punishment has shifted back and forth between these two competing and contradictory theories, the deterrence and the retributive theory, with no sign of resolution in sight. Is the purpose of punishment prospective or forward-looking, i.e. aimed at future good consequences; or is it retrospective or backward-looking, i.e. aimed only at giving the criminal what he deserves? Even worse, it is far from clear whether either of these theories presents a morally justifiable reason for inflicting punishment. As we will see in Chap. 2, the deterrence theory presupposes a deeply problematic moral theory, utilitarianism or consequentialism. It is a major, probably decisive, objection against consequentialism that it permits intentionally harming one person for the benefit of others, contrary to what morality allows. But this is just what the deterrence theory does: permit harming Peter to benefit Paul and Mary. Hence most moral philosophers have concluded that the consequentialist theory is morally unacceptable; if so, a fortiori, the consequentialist justification of punishment is morally unacceptable.

But the retributive theory is no better morally, and arguably even worse. For it outright rejects the idea that one may justify inflicting harm on a person for the sake of the social good. Its insistence that there be no further social benefits from punishment has made it seem to critics that it ends up holding that punishment is imposed for no reason at all (as we will see in Chap. 3). The idea that the proper response to a person committing an unjustified harm, is to harm him in return, is far from obvious. And it has long been subject to a devastating objection. As Plato long ago recognized, you cannot undo evil once it has been done: the purpose of punishment is not to cancel the crime—what is once done can never be made undone—but to bring the criminal and all who witness it to complete renunciation of such criminality (Plato 1989, Laws 934a). In the Protagoras, the sophist Protagoras voices the following critique of retributivism:

In punishing wrongdoers, no one concentrates on the fact that a man has done wrong in the past, or punishes him on that account, unless taking blind vengeance like a beast. No, punishment is not inflicted by a rational man for the sake of a crime that has been committed – after all, one cannot undo what is past – but for the sake of the future, to prevent either the same man or, by the spectacle of his punishment, someone else, from doing wrong again (Plato 1989, Protagoras 324a).

The French philosopher Blaise Pascal in his Pensees put the objection even more bluntly (we may call this Pascal’s Principle): Must one kill to ensure there are no evildoers? That makes two evildoers instead of one. (Faut-il tuer pour empêcher qu’il n’y ait des méchants? C’est en faire deux au lieu d’un) (Pensee 911–659). Pascal goes on to cite Augustine for the principle that one should conquer evil with good, not with more evil. If killing is wrong, it is rather mystifying how killing the killer can be a legitimate response, for all it accomplishes is that there are now two wrongs (two killings) instead of one. Rather than cancelling the evil, it only seems to add to the evil in the world. It would seem to violate the moral commonplace, two wrongs don’t make a right.

It is surprising however how often defenders of the retributive theory of punishment dismiss or ignore this problem. Consider one recent example, philosopher Angelo Corlett’s 2006 book-length defense of retributive punishment, Responsibility and Punishment. Corlett barely mentions Pascal’s Problem in his book, and only in the context of whether the death penalty is justified. There, Corlett briefly considers the Pointlessness Objection to capital punishment, according to which capital punishment is a waste of time and effort because inflicting death on the criminal cannot bring back, say, the life taken by her (2006, 137). Corlett’s reply is quite astonishing. Rather than attempt to explain what the point of punishment is, he launches into a blistering ad hominem attack on anyone who would even raise such a problem, for such people demonstrate a clear case of a lack of moral fortitude and character sufficient to hold everyone responsible for their actions (id.). Corlett’s attitude is admittedly extreme, but illustrative of the defensiveness of retributivists in response to Pascal’s Principle. It is not enough to assert that one must inflict suffering in order to hold someone responsible for their misdeeds. The very question is why it is permitted – let alone required – to harm someone in response to their wrongful actions. It is if anything the indifference of this kind of retributivist to the need for moral justification for deliberately inflicting harm on others that raises questions about moral character.

This indifference to the central issue is arguably also evidenced by the otherwise inexplicable obsession in Western culture with one form of punishment in particular: the death penalty. A recent letter to the editor in the New York Times reads:

Thank you to Lincoln Caplan for his editorial about abolishing the death penalty. We are a disgrace to all of Western civilization for allowing the existence of such punishment. I hope that in 50 years our children and grandchildren will be horrified that we permitted the death penalty, which should be considered as barbaric as chain gangs (Jan. 8, 1912).

But if we do not know why punishment in itself is justified, it is hard to see why we should single out capital punishment as if it were the single and unique problem with our punitive practices – as if locking someone in a cage for the rest of his life were not equally barbaric. It is almost as if our culture uses opposition to the death penalty as a means of evading the much bigger problem, why punishment of any sort is justified at all. Perhaps opposing the death penalty gives us a (false) sense of taking the moral high ground (though it is interesting that the letter writer includes chain gangs as well, as if admitting the problem goes deeper than merely the death penalty). Similarly, the vehemence with which a large segment of the population defends the death penalty perhaps suggests that they realize that much more is at stake, i.e. the very legitimacy of punishment itself.

We are thus in the odd position of having only two candidates for a theory of punishment, neither of which appears the least bit morally acceptable. To make matters worse, scholars fail to agree on the psychological question as to whether either of the two (or some combination of the two) actually describe the reason for which people generally support the institution of punishment. That is, punishment suffers from a double problem: a descriptive and a normative question. We are unclear just what the reason is for which we punish, and whether that reason is morally sufficient to justify punishment. What is remarkable is that this highly problematic situation persists even after 2,000 years of debate on the topic, and regarding a practice that is apparently a cultural universal, found in all societies in all times and places.

Every once in a while someone claims to have come up with a new, alternative justification of punishment, for example the currently popular expressive theory of punishment, to be discussed in detail later on. But these theories inevitably turn out to be either incoherent or merely another name for the deterrent or retributive theories. It seems highly unlikely that at this late stage in the debate someone will discover a radically new theory of punishment. It would be odd indeed if we were to discover only now the rationale for a practice that has been going on for thousands of years and debated for just as long. At the same time, the case for the two leading theories, deterrence and retribution, seems extremely weak. If neither has made a convincing defense by now, one can reasonably wonder if they ever will (the various attempts will be discussed in detail in the next two chapters). As Alvin Goldman explains, the paradox of punishment is that punishment seems from a moral point of view to be both required and unjustified (Goldman 1994, 30).

Yet it is the goal of this book to escape the paradox of punishment. The strategy here is not to discover a new theory of punishment, but rather to revisit the two classic theories, deterrence and retribution. The argument is that we in modern times have come to misunderstand both of these theories, and that understood in their traditional form they can be seen not to violate basic moral intuitions. However, the emphasis will be on one of these theories in particular, the retributive basis of punishment. Retribution is, we will argue, the essence of punishment, indeed almost definitional of punishment. The biggest challenge then will be to face a further and perhaps even more difficult problem: making sense of retribution. For retribution has its own paradox: it seems both intuitively required and morally indefensible.

1.1 The Paradox of Retribution

As if things were not bad enough, the retributive motive for punishment presents us with yet another paradox. The problem is that, whatever the claims about crime prevention, deterrence, and other forward-looking goals of punishment, deep down the intuition that most people appear to have is that punishment is retributive. People should be punished because they deserve it, and if there are further benefits in the way of crime prevention, that is a bonus but not the motivating reason for punishment. The paradox of retribution, in the words of J. L. Mackie, is that on the one hand, a retributive principle of punishment cannot be explained or developed within a reasonable system of moral thought, while, on the other hand, such a principle cannot be eliminated from our moral thinking (1982, 3). The idea of retribution, paying back evil for evil, seems absurd, pointless, even juvenile: if your brother hits you first, the childish impulse is to hit him back in order to get even, and one’s parents of course tell you that this is wrong and you may not hit back. But how is retributive punishment any different? Retribution seems obviously wrong, and yet it seems to be the ineradicable basis of punishment.

The recent history of the criminal punishment debate exhibits this paradox. Starting in the late nineteenth century and lasting well into the twentieth century, a major reform movement in criminal justice aimed to eliminate the traditional retributive basis for punishment as a barbaric, superstitious relic, grounded, it was argued, in outmoded religious beliefs, and replace it with a modern, rationally-based, and scientific model of punishment. The new model would be utilitarian and even technocratic; punishment is an evil, like any suffering, and should be avoided except where necessary. It could only be justified, therefore, where it would predictably lead to measurable social benefits in the form of crime prevention or rehabilitation of the offender. Punishment policy, on this view, should therefore be removed from the democratic process, since the masses merely want blood, and put in the hands of the expert sociologists, psychologists, and criminologists who could determine on a scientific basis what punishments would be likely to produce the desired beneficial results. Retribution became stigmatized as merely an emotional reaction but not a rational basis for a progressive criminal policy.

This technocratic movement reached its height in the 1950s and 1960s, and its success was such that many experts declared the death of the retributive theory of punishment. David Dolinko has collected some of these declarations:

Fifty years ago, a defender of retributivism acknowledged the general belief that the retributive view is the only moral theory…which has been definitely destroyed by criticism. A decade later Justice Black, writing for the Court, declared retribution no longer the dominant objective of the criminal law. In 1966, a Manual of Correctional Standards published by the American Correctional Association flatly announced: Punishment as retribution belongs to a penal philosophy that is archaic and discredited by history. The first edition of a well-known criminal law text summed up the view as of the early 1970s: retribution was the oldest theory of punishment, and the one which is least accepted today by theorists (1992, n. 1).

Philosopher Walter Kaufmann expressed the widespread disdain for retributivism in 1975: As long as traditional Christianity flourished, retributive justice did too. When the faith in hell and the Last Judgment lost its grip, Jefferson and Kant…still tried to save the faith in retributive justice by providing a new, rationalist foundation for it. While men still had the old religious faith in their bones, such efforts seemed to have some plausibility; but no more (1975, 52).

The predictions of the demise of retribution, however, proved not only premature but utterly misguided. The revival of retributivism in the 1970s was sudden and overwhelming and it quickly regained its place as the dominant theory of punishment. It is difficult to know what caused this revival, though the increasing crime rates of the 1970s and the perception of increased social disorder, along with a loss of faith in the technocratic, expertise model of government, helped create a climate conducive to the return of the retributive impulse. The dramatic loss of belief in the effectiveness of rehabilitative punishment was noted by Francis Allen in 1979: In a remarkably short time a new orthodoxy has been established asserting that rehabilitative objectives are largely unobtainable and that rehabilitative programs and research are dubious or misdirected.² The California Determinate Sentencing Legislation in 1976 explicitly rejected the rehabilitative motivation for punishment (Zimring and Hawkins 1995, 9). However, as Zimring and Hawking note, this disillusionment with the rehabilitative ideal did not appear to have been based on any new empirical findings discrediting the idea of rehabilitation (id.). Rather, it seems to have been based on a general philosophical shift towards the retributive rationale for punishment in dealing with crime, regardless of the effectiveness of rehabilitation.

In addition, the discrediting of the utilitarian theory of ethics –a landmark is John Rawls’ A Theory of Justice (1971) – helped undermine the theoretical foundations for the consequentialist justification of punishment. The rise of conservative politics (and of religion) also contributed to the revival of retributivism, which is after all the traditional basis of punishment, though liberals too became skeptical about the way in which the scientific, medical model of punishment lent itself to abuses by the all-powerful state (an influential 1975 film portraying the medical mistreatment of social outcastes is One Flew Over the Cuckoo’s Nest). And with the collapse of the rehabilitative and consequentialist justification of punishment, a vacuum was created into which there was little else available but the retributive theory to fill. Whatever the specific historical causes, it quickly became obvious that the retributive impulse had not disappeared but had merely been temporarily suppressed. It would seem hard to deny that the retributive impulse remains what it has always been, the fundamental guiding motivation for punishment both today and in all historical periods. Hence the paradox: we punish for retributive purposes, yet we cannot make moral or rational sense of this practice.

1.2 The Incoherence of Public Policy: A Muddle of Theories

The revival of retributivism did not eliminate the deterrent theory, and the two continue to coexist in an uneasy and perhaps incoherent combination. Current criminal law policy documents typically attempt to evade the problem by simply asserting that punishment aims at both retribution and crime prevention. But this is no solution at all since the two goals are inconsistent and contradictory; retributivism declares that one may punish only on account of a past wrong and only proportionately to the severity of that wrong. The consequentialist theory says one may punish only when it will produce net overall social benefits, and the theory does not demand that punishments must be proportionate. Hence the two theories will give quite different and opposing recommendations in many cases. (And this is not even to mention the problem that each of the rationales appears to be morally impermissible.)

Consider for example the Model Penal Code (MPC), published in 1962 at the height of the utilitarian influence on the criminal law. The MPC exemplifies the technocratic ideal of criminal justice and the implicit suggestion that traditional retributivism is a legacy of a religious, pre-scientific society; it declares in the opening section that one of its central goals is to rationalize punishment: to advance the use of generally accepted scientific methods and knowledge in the sentencing and treatment of offenders. It adopts an essentially utilitarian theory of punishment, accompanied by weak retributive constraints. That the foundational idea is utilitarian is made clear by the Explanatory Note to Section 1.02(1), which declares that the dominant theme is the prevention of harm. However, its drafters did recognize one of the damning objections to utilitarianism, the fact that it permits disproportionate punishment. Hence it incorporates as another goal a constraint against such punishment: to safeguard offenders against excessive, disproportionate or arbitrary punishment. Thus while it does not endorse retribution as a positive goal of punishment, it does endorse the retributive principle as a negative constraint: that one should not be punished more than one deserves.

This position is often called negative retributivism, a view that seemed at the time to provide a solution to the problem by combining the best features of utilitarianism and retributivism. Negative retributivism is the account according to which the purpose of punishment is crime prevention, but this goal must be constrained by the retributive principles demanding that only the guilty be punished, and that they not be punished disproportionately. We will examine the idea of negative retributivism in detail in Chap. 4. For the moment, we may note that it does not appear to provide a coherent solution to the problem of punishment. Utilitarians can object that it is irrational (and even superstitious) to constrain this rational goal of crime prevention by arbitrary restrictions on the attainment of one’s goal. Retributivists will object that one cannot simply adopt the negative constraints of retributivism without also accepting the basic retributive idea, that the purpose of punishment is to give wrongdoers what they deserve, not to aim at crime prevention – and moreover that utilitarianism is an immoral basis on which to justify inflicting harm on people. Hence the theory of negative retributivism is no longer widely accepted.

The Federal Sentencing Guidelines, introduced in 1987, adopt a different strategy.³ Recognizing the inherent contradiction between the two theories, it gives up on the attempt to combine them. In the Introduction, the commission acknowledges the difficulty of stating the purpose of punishment: A philosophical problem arose when the Commission attempted to reconcile the differing perceptions of the purposes of criminal punishment. The problem of course is the lack of agreement on the purposes for which we punish. The commission continues: "Most observers of the criminal law agree that the ultimate aim of the law itself, and of punishment in particular, is the control of crime. Beyond this point, however, the consensus seems to break down. Some argue that appropriate punishment should be defined