7

The Right to Kill in Cold Blood

DOES THE DEATH PENALTY VIOLATE
HUMAN RIGHTS?

THIS ESSAY BEGAN life as a public lecture, and I have not tried to remove the informality of style appropriate to such an occasion. The essence of the argument is this: all punishment must be inflicted in cold blood; whatever damage we do to others not in cold blood is not punishment but self-defense or revenge; what we have a right to inflict in cold blood is a question of the rules of just social cooperation and especially the justice of the sanctions required to sustain those rules; it is here argued that the fundamental principle is that we may inflict whatever punishment is necessary to deter wrong-doing and is not disproportionate to the offense; I do not dismiss “pure” retribution as a goal of punishment, but I do not discuss it here. I invoke only a diluted concept of retribution in the form of the concept of a “proportionate” response to crime. The central issue of fairness is, then, that of the process by which alleged criminals are pursued and caught, questioned, tried, and sentenced; in the view taken here, the death penalty raises no questions of justice that are different in kind from those raised by other forms of punishment, but that death as contrasted with incarceration may be thought to raise questions about irreversibility that are different in degree. One reason for insisting that a failure of fairness in the criminal justice system makes all punishment suspect is that an undue attention to the death penalty (by both defenders and opponents) diverts attention from too many of the other injustices perpetrated in our name. Finally, the essay turns to the question of how we should discuss the penalty of death in particular, and I invoke the shade of Joseph de Maistre for the purpose.

I hope that you will be unable to detect whether I am an enthusiast for the death penalty, an opponent of capital punishment, or deeply ambivalent. You may have your suspicions about where my allegiances lie, but I shall do my best to leave you uncertain whether your suspicions are well founded. I ought perhaps to mention that one of my philosophical heroes, John Stuart Mill, not only believed in the indispensability of the death penalty as a deterrent, but also personally launched a private prosecution for murder against a brutal British governor of Jamaica, with every intention of having Governor Eyre hanged.1 What you should have no doubt about is my answer to the question I pose in my subtitle. My answer is that whatever objections one might have to the death penalty, they should not be objections to its compatibility in principle with any plausible account of human rights. If anyone wants to know only my answer to the question whether the death penalty must always and in principle violate human rights, they have it—it is no. But I doubt whether anyone does want to know only that. The 65 percent (today) to 85 percent (in the 1980s) of the American people who say in response to questionnaires that they believe in the death penalty already believe that it does not violate human rights, and do not need me to reassure them. Nor will the 20 percent or so who think that it is intrinsically barbarous and morally disgusting, and that it will inevitably be administered in a racially and economically biased fashion, much care whether some version of the death penalty might be operated without violating anyone’s rights. The very small percentage of the American people who share my professional interest in the history of arguments about human rights know without my telling them that lawyers, philosophers, and legislators over the centuries have argued not only that our rights do not include the right not to be subjected to the death penalty under appropriate conditions, but also—as did Immanuel Kant, G.W.F. Hegel, and some of their late nineteenth-century disciples in Britain and the United States—that criminals have a positive right to be executed.

So why do I chase this particular hare again? Partly because of the contrast between the light-heartedness of Governor George Bush over the likelihood that the death penalty has been carried out on Texans who were not guilty of the crime for which they were executed, and the anxieties of Governor Ryan of Illinois, who declared a moratorium on executions as soon as he became unsure that everyone on death row was properly there.2 Partly because it is a very striking fact that the United States is a country that takes rights seriously and yet refuses to subscribe to various international declarations of human rights because they state that the death penalty violates those rights. Partly because of something I do not myself fully understand. The European Convention on Human Rights (ECHR), which was first ratified in 1950 by a number of European states, including Great Britain, was modified in 1983 to include a protocol (the Sixth Protocol) prohibiting the use of the death penalty except in time of war by any of the states that were signatories to the convention. This would have been impossible at the date of its first creation. Not only Britain, but France, the Benelux countries, and most others as well not only had capital punishment on the statute books but actually carried out executions. The French, indeed, abolished executions by means of the guillotine only in 1981, having carried out the last execution in 1977. The British slowly relinquished the use of hanging, but after a near miss at abolition in 1947, further attempts to abolish the death penalty in the 1950s came to nothing, and a royal commission reporting in 1953 declared itself uncertain on the facts, but inclined to think that death was a “unique deterrent.” The discussion in those years never suggested that even if it was a unique deterrent, it violated the human rights of the executed prisoner. Lawyers assumed an existing consensus on the limits of lawful punishment; lawful punishment, inflicted only after a fair trial and a sound verdict, might be ineffective, might have bad consequences of one sort and another, but was not itself a violation of the rights of the condemned. It was assumed that the processes of the law were sufficiently respectful of the rights of the accused and of the condemned.

I do not know why sentiment against the death penalty moved so swiftly and completely in Europe, moved rather reluctantly and incompletely in Britain, and swung dramatically back and forth in the United States. In the United States, opinion was firmly in favor of the death penalty in the 1950s, swung against it to the extent that supporters were in a minority in the 1960s, then swung back again until at present it is supported by 64 percent of those polled, after attaining a high point of support at 84 percent.3 One needs to begin with a well-known but underappreciated paradox. In no country where capital punishment has been abolished has it been abolished with the support of the electorate. To put it the other way round, in no country has the electorate demanded the abolition of capital punishment and then seen legislators follow public opinion by abolishing it. There are many countries where capital punishment no longer enjoys the support of the majority of the population; in all countries, the number who support it varies quite dramatically according to whether the most recent event in the public mind is the murder of a small child by a sexually deranged adult or the execution of a probably innocent person. But in no Western society has the death penalty been abolished because the public at large demanded its abolition.4 The common pattern is that it is abolished because legislators decide that it is either not a deterrent or that too many doubtful convictions have occurred, or that it is for whatever reason distasteful, and the electorate then gets used to it not existing; as that happens, so legislators get used to not trying to restore it, and the public eventually goes off the whole idea. But principled opposition to capital punishment is not a common trait in democratic electorates. Nor is principled support for it important in Europe. That punishment ought to reflect some version of “an eye for an eye” is felt by many people, but it is not argued for by pressure groups or politicians. This contrasts with other issues that attract similar degrees of passion when they flare up in the public mind. In Britain, there is a persistent though low-key campaign by right-to-life groups to have the rules on abortion tightened up and to have the 1967 act regulating abortion either repealed or made much more restrictive. Similarly, pro-gay groups persistently campaigned for many years to have the age of consent made the same for homosexual and heterosexual relations—it is now sixteen in all cases—and MPs introduced legislation to this effect from time to time. Nothing of the sort happens with respect to the death penalty. Two rather halfhearted attempts at reinstatement shortly after repeal are all there have been. Large numbers of the public insist, when sufficiently excited and goaded on by interviewers from the tabloid press, that sex offenders should be hanged—preferably after prolonged torture culminating in castration; but there is no steady, persistent pressure on the legislature to reintroduce even the sanitized executions by lethal injection that American states have instituted.5

This returns me to the oddity of the European Convention on Human Rights. Although Canada and some other not very European countries have observer status at the Council of Europe, the United States cannot do so, because the retention of the death penalty is inconsistent with membership. (Actually, that is not strictly true; as always with treaties that commit states to acknowledging the provisions of a convention, there is room for derogation and reservation. Britain for some years preserved as a matter of law, though not of practice, the death penalty for those rare persons who committed piracy with violence on the high seas or committed treason, a crime that included the seduction of the sovereign’s daughter as well as betraying one’s country to an enemy in time of war. Even in its present, apparently absolute form, the convention allows states to use the death penalty in time of war, a concession not allowed for the use of torture, which is banned absolutely.) The Sixth Protocol, proposed in 1983, does not go into details about the change of heart on the death penalty, but simply notes that in the thirty years since the convention was initially promulgated, European opinion had turned against the death penalty and most European countries had given up its use.6 This does not amount to saying that at some time during that thirty years they had noticed, discovered, or decided that the death penalty was a violation of human rights. Nor, in fact, is the chronology of the death penalty’s falling into desuetude quite as neat as that suggests. There was a considerable gap in many countries between their ceasing to execute anyone in practice and their abandoning the penalty in theory; the Swedes last executed anyone in 1910, but abolished the penalty only in the 1950s, while Eire took from 1954 to 1990 to bring abstentionist theory into line with abstentionist practice. Britain last executed anyone in 1964—the victim on that occasion, a man called Joseph Hanratty, may well have been innocent, but even if not, there had always been enough anxiety about the matter to make it an appropriate case to pause on—and the penalty was abolished for everything except piracy and treason in 1973; only in 1998, when Britain prepared to incorporate the convention into British law, did Parliament take the last step and abolish it entirely. Nobody noticed, and there was no evidence that either the royal family or our mariners slept less easily in their beds and hammocks. Part of what lay behind the movement in Europe was a change of heart within the Catholic Church; traditionally, the church had no qualms over the death penalty, but for the past thirty years or so, whether the pope was a liberal like John XXIII or a conservative like the present pope, the church has been opposed to execution and has exposed itself to numerous rebuffs from American states by appealing, and usually in vain, for reprieves for convicted criminals on the verge of execution.

Now, it is, of course, true that if we define human rights in positivist terms—that is to say, as those rights that are protected by some applicable convention or other and accepted by the government within whose jurisdiction the question arises—then it is true of all citizens of those states that are signatories to the ECHR and members of the Council of Europe that the imposition of the death penalty on them and there would violate their human rights. And conversely, for citizens of the United States in the United States, it would not. The situation of European nationals in American courts would be conceptually problematic in this view, though in law, their position is covered by the Vienna Convention. Failures of individual states to take any notice of the convention have recently placed the U.S. government in some difficulty on the matter. But a positive account of human rights is highly implausible. The thought that the human rights of Europeans are different from the human rights of Americans empties the concept of human rights of all meaning. The point of talking about human rights is to claim that what lies behind the positive rights set out in a convention are deep rights, rights that ground the convention rights rather than simply flowing from them or reflecting them. Human rights are what we used to call “natural rights” or “moral rights” or just “rights.”

What I claim is that the infliction of death is not a violation of the human right of a convicted felon under appropriate conditions and, in the converse, that individuals and society as a whole have the right to inflict such a sentence, but I am also going to argue that this has moral force as a defense of the death penalty only if many other rights are in place. There are many rights of the accused and convicted that have to be respected, and many conditions that must be in place before we may justly exercise our right to kill in cold blood. I go this long way round for something other than the pleasure of performing philosophical arabesques. In the summer of 2000, an essayist in the New Republic, engaging in the journal’s favorite pastime of irritating such liberals as still read it, argued that the rise of DNA testing in capital cases might backfire on the liberals who had demanded it. They, he observed, had demanded DNA testing on the assumption that it would show condemned felons to be innocent; but what if it showed that they were guilty? Would that not open the way to their execution?7 To which the answer, of course, is that pro tanto (to that extent) it must. To remove one obstacle to execution is to remove one obstacle to execution. Why liberalism should be thought to be inconsistent with elementary logical inference is not clear to me, nor why the New Republic thinks itself so clever in pointing out a tautology. But the New Republic and I are at one on the logic. The right not to be executed—nor to suffer any other penalty—for a crime one has not committed is the most basic right that must be respected if we are to have the right to take a life in cold blood. If guilt is established without doubt and by just and transparent means, the most important obstacle to execution has been cleared away.

Let us approach this conclusion from the opposite direction: the right to punish anyone for anything, a right that Robert Owen and most anarchists have denied that we possess, but that I shall defend. The first step is to distinguish punishment and self-defense; the right to self-defense is the right to retaliate against immediate attack and, by extension, to do whatever is needed to prevent an assailant from immediately attacking us.8 The (potential) victim of an attack may do whatever is necessary under the circumstances and must use his or her best judgment as to what that is. That best judgment is subject to scrutiny. If you are kicked in the leg by a five-year-old child, you cannot shoot her dead and pass that off as your best judgment; you have committed murder. If you are attacked by a madman with a knife, on the other hand, you will be excused the death of your would-be assassin on the grounds that there was no way of knowing quite what he might have done, and there is no upper bound to what you may need to do to protect yourself. There is no settled doctrine about these limits, as attested by that lamentable case in Louisiana five or six years ago in which a householder shot dead a Japanese schoolchild who knocked at the door to ask for directions, and was acquitted of all wrongdoing by a local jury. John Locke, the theorist of mild and constitutional government, insisted that defending our property in the absence of government assistance licenses the use of deadly force. Someone who set out to rob us might do us worse harm still, and any degree of self-defense is licit.

But the right to inflict punishment is not identical with the right to self-defense. Punishment is what we threaten ahead of time and inflict after the event, but it is not whatever damage we may happen to do to someone from whom we are defending ourselves. We may properly describe the institution of punishment as part of society’s battery of means of defending itself, but that is a different matter. The right of self-defense is not the same as the right to inflict punishment. Punishing people is not merely something that we do in cold blood, but something that we morally—and almost logically—must do in cold blood. First, the inflicter of the penalty must be authorized to impose the penalty and see to its carrying out; courts are so authorized, lynch mobs are not. Locke thought individuals were so authorized in the state of nature; Hume thought Locke’s view made no sense. But both saw the same difference between doing something that could qualify as punishing and doing what amounted simply to harming someone else. Locke was eager to insist that governments could acquire only rights that individuals gave them. If governments had a right to punish, they must have got it from individuals who transferred it to them. Hume thought the institution of punishment just grew in the same way as property rights and government generally. Second, the recipient of the penalty must be guilty of an offense and must have been found guilty by a reputable procedure. The point of a system of punishment is not primarily to respect the rights of the criminal; but none of us can pursue our purposes while violating the rights of others, not even the rights of criminals, so when we purport to punish someone, we must not violate their rights in the process. Third, the purpose of punishment is to secure the rights of the noncriminal and to requite a proper measure of justice on the wicked. Reliable systems of punishment based on proper adjudication and sentencing are as good a way as anyone knows to reduce the insecurity and uncertainty of a world where we are vulnerable to theft, assault, sexual assault, and fraud on the part of others, without creating new forms of insecurity as bad as or worse than those we prevent. This is what Herbert Hart many years ago called the general justifying aim of punishment.9 If you follow Locke and say that individuals have a right to punish by the light of nature, what does this mean? It means that they may threaten others with harm that they will inflict on them if they damage other people’s legitimate interests; but it must be noticed—what Locke agreed his readers might think very strange—that the persons threatening punishment are behaving as if they were, so to speak, little governments before any real government existed. This is because of the fourth point: punishment must be aimed at any breach of the law, no matter who is harmed. We say not that if someone hurts us, we shall retaliate, but that if they hurt anyone, we shall punish them. The implausibility of imagining that persons bereft of government will in fact consult John Locke before they act is obvious enough. We do not think that individuals will act on behalf of others as energetically as on behalf of themselves; indeed, we think they will act too energetically on behalf of themselves and not at all on behalf of others—that what we call “punishment” will in fact be revenge when our own interests are at stake, and nonexistent otherwise.

Nonetheless, this objection itself shows what punishment really is and therefore what we have a right to impose on others, and it shows what accused and convicted people have a right to. To dogmatize, all of us have both a right and a duty to make a system of punishment, but not a system of simple revenge, work as it should. We owe this to other people, and we owe it to them as part of a cooperative and fair system of mutual help for the sake of security and prosperity. Everyone has a right to benefit from this system unless they forfeit that right in whole or in part by some violation; and among their rights are some they cannot forfeit because other human beings can never have the right to treat them unjustly, cruelly, or wickedly. What we have the right to inflict by way of punishment is now the crucial topic. On the face of it, we can inflict whatever meets two requirements; the first that it is effective in deterring and adequate in retributing misconduct; the second that it does not violate rights that a criminal cannot forfeit or imply rights on the part of the punisher that nobody can have. Innumerable governments have violated this second condition. The Spanish government of the seventeenth century made it a capital offense to import books from foreign publishers; nobody had a right to impose such a sentence, and being so sentenced violated the criminal’s rights. However fastidious the legal process by which one’s guilt was ascertained, the offense was not (morally speaking) an offense, and the penalty not proportionate.

If we take these thoughts in sequence, we reach the following position. Because the infliction of punishment is itself a bad thing to the extent that it is the causing of misery, the amount of punishment to be inflicted must be the minimum necessary to achieve the goals of deterrence and retribution.10 To retribute to someone more than is proper is itself a violation of justice and therefore of their rights; it is also to cause more misery than strictly necessary, though that is a utilitarian argument rather than an argument from rights, and sustains an argument from rights only once you add that it is unjust that anyone should suffer more than is absolutely necessary to achieve a defensible goal. To threaten by way of punishment more than is necessary to secure compliance (or effect just retribution) is to declare yourself ready to violate the rights of anyone whom we punish in fact. The question whether it is right to threaten to do what it would be wrong to do in fact was much debated during the 1950s by Catholic critics of nuclear deterrence; many theorists of deterrence claimed that in order to keep the peace, it was necessary to threaten to launch a nuclear war if the Soviet Union launched the first strike, even if we knew that no good purpose would be served by retaliation and doubted whether we would actually do it. Much elegant deterrence theory rests on the thought that it is good to lead your opponent to believe that you are stark raving mad and might retaliate from atop a radioactive ash heap. Many Catholics held that if it would be wicked to launch a retaliatory strike, we had no right to threaten to do it.

There is an analogous situation vis-à-vis the death penalty in the following sense. If the death penalty is not much of a deterrent in the case of murder and armed robbery, it surely would be in the case of parking offenses; and yet everyone is clear that to threaten death for overstaying one’s time on a meter would be wicked—even if we were perfectly sure, on the one hand, that the deterrent would be so effective that nobody would ever overstay and, on the other, that if they were to do so by some mischance, we would find some way of remitting the sentence. It follows that if the death penalty is out of proportion to the crimes for which it is inflicted, or if some much lesser threat would secure everything that a system of deterrence needs to secure, the death penalty would be illicit as a punishment for those crimes too. In short, punishment is not just something threatened for the sake of deterrence, but must also be an apt response to the crime for which it is threatened. The overstayed parker is guilty only of mildly inconveniencing the rest of the public, if that, and a small pecuniary penalty is adequate. We have, you might say, a right to threaten that much, and nobody can have a right not to be threatened with that much, but we all have a right not to face the threat of more than that.

My own view is that some argument of this sort is cogent; perhaps more to the point in this context, it is felt by most people to be cogent. It can be given some unusual philosophical tweakings, however. Immanuel Kant, for instance, held that the right to punish was implicit in a hypothetical social contract—one that nobody had in fact signed, but that we could imagine everyone signing and could imagine embodying the conditions of mutual obligation that lie at the basis of social cooperation. This contract allows each of us to invoke the assistance of everyone else in preventing others from violating our rights so long as we do our part to secure their rights as well; the contract guarantees our freedom from assaults on our lives, persons, and property as long as we cooperate in securing the same freedom for others. Since this vision of the role of the state implies an essentially defensive role for it, there is no suggestion that the state’s task is to make us better, more amenable, more morally virtuous, or adherents of the true faith; all those things are tasks for individuals working under the protection of the state, but not under its instructions. The role of punishment in this scheme is striking. We commit ourselves to the contract on terms that in effect instruct others to inflict on us for breaches of the law the penalties we stand ready to inflict on others. Not to punish criminals would be a dereliction of moral duty, and Kant observes in a footnote that a society that knew it would perish from the earth in the middle of the night should ensure that the murderers in its jails were hanged before the society was destroyed.11 Even in Texas, this might be thought a bit fierce, but the grounds for Kant’s views are not as odd as the conclusions they ground. To be part of a law-abiding society is to be part of a scheme of rights observance; in effect, we are estopped from complaining that others hold us to standards of rights observance that we have advertised ourselves as accepting. We do not become outlaws by violating the rights of others, tempting though it is to say that we forfeit our immunity to the assaults of others if we break the rules that protect them against the assault of ourselves. What we forfeit is only our immunity to whatever punishment is properly imposed by the agents of the criminal justice system. We become criminals, not outlaws, which is to say that we are still within the ambit of law, and are treated as beings with rights, not beings without them.

This is plainly correct, and it makes sense of some of the things that distress critics of capital punishment, such as the suicide watch that is kept on condemned criminals. If the only thing we were interested in was that condemned criminals should be dead, we would presumably not mind how they became dead—and quite a lot of hard-boiled commentators have suggested that leaving a bottle of something lethal readily to hand would be a cheap way of disposing of the condemned who chose to go of their own volition. It would, however, be a lawless way of proceeding, and for those who care that punishment should be a lawful undertaking, this is intolerable, however perverse the results appear. But this is running ahead of things, because the more crucial question is how people actually get onto death row in the first place. Here, issues of rights violation really rear their heads.

The first thought is that if capital punishment is felt to be special, as it plainly is in Western societies nowadays, then all the usual rules have to be observed with extreme scrupulousness. What are they? Begin at the beginning; we all have a right not to be condemned for any crime that we have not committed, even if this is barely overstaying at a parking meter. How fastidiously this right has to be observed before there is a serious injustice is another matter, however; to complain that my rights were seriously violated by a parking authority that did not maintain its parking meters in tip-top condition would be going a bit far. To complain that my rights were seriously violated by a state that hired known drunks to defend me in a capital case would be putting the matter quite lightly. This point runs right through the whole of criminal justice, and I want to say two things that may seem surprising. The first is that capital cases are not, to my mind, as special as the public inclines to think, and the second is that the issues raised by capital cases run disquietingly through the whole criminal justice system. Let us start with the banal thought that we have a right to be tried on one issue only, namely, whether we committed the crime of which we stand accused. There is no law against being black, against having a low IQ, or against being poor. It is, however, inconceivable that any well-off white person of reasonable intelligence would run much risk (not, that is to say, no risk at all, but very much less than a poor, not very bright black person of comparable wickedness) of suffering the death penalty in the United States. Nor is this primarily a racial matter; as the case of O. J. Simpson sufficiently illustrates, anyone with really substantial means need not worry unduly that they will suffer execution if they put their minds to escaping it. I ought to say that this is not a uniquely American phenomenon, nor one relevant only to the death penalty. The prisons of Britain contain disproportionately many young black men, and the criminal classes are overwhelmingly drawn from the ranks of the poor and the educationally subnormal. But these are banal observations, so why are they worth noticing?

If the justice of punishment depends on its being part of a general scheme of social control under which we all observe the law in order to share in the benefits of general law-abidingness, there must be some rough equity in how we are requited for our forbearance. Contrary to some radical views, the poor, the intellectually unadvantaged, and members of low-status groups certainly get something from the legal system, and in some ways they get more than is apparent. If the law is scrupulously observed, they are protected from exploitation and especially from being bamboozled by the quick and the clever, as well as from being subject to arbitrary ill-treatment at the hands of the majority. They may have little property to protect, few contracts to enforce, and meager estates to bequeath, but they have their lives, limbs, and security to protect, as much as anyone else. Nonetheless, they are, as Bentham observed, required to keep their hands off the property of the rich, they are obliged not to take shortcuts to obtain the property of others, and they ought to be adequately recompensed for this abstention. One form that that recompense might properly take is being guaranteed the real, as distinct from the nominal, protection of the law. So far as the law of property goes, individuals who have little property must have what they do possess fastidiously protected, and the widow’s cottage must be as safe from harm as the plutocrat’s mansion. The life and times of John D. Rockefeller, John Jacob Astor, and Henry Clay Frick suggest that we do not always live up to that. When it comes to the criminal law, the thought is even starker. Your life and liberty must not be at stake because your means are slender. So one right we must have in place before the death penalty can be imposed without a violation of rights is a genuine right to free and equal justice, and that must mean that the quality of legal assistance available to the badly off must be no worse than that available to the well off.

But aside from the utopian quality of that aspiration, does it pick out the needs of defendants in capital cases only? This is where things get more complicated. It is generally assumed that capital cases are special, but I want to raise some doubts about that. You may think that this is British callousness, and that only in a country where they used to execute fourteen-year-olds for sheep stealing and where the theft of five shillings was capital—and indeed, where the families of those condemned to hang used to pay the hangman a few pence to be allowed to pull on the feet of their relatives to ensure a quicker and more merciful death—could anyone doubt that capital cases are special. Now I do not deny that the death penalty has become special, inasmuch as those who are eager to see it imposed have gone to enormous and expensive lengths to ensure that it can be—it being, on average, three times dearer to kill a criminal than to keep him in jail for thirty-odd years—and the intricacies of what the Supreme Court does and does not permit have become pretty extraordinary. But this is what you might call factitious specialness; it results from people thinking that the death penalty is special, and it provides no independent reason to think that it is. The most obvious difference between death and all other penalties is that the death penalty has a finality that no other punishment has. Even this, however, ought not to be exaggerated. When inmates of Angola, the Louisiana State Penitentiary, go in on a life sentence without parole, they stay in Angola until they die, as much as if they were going in only until they are executed; like those who are executed, they will leave only in a coffin. Nor is the death penalty final in the sense of being passed and implemented speedily so that the whole thing is over and done with at once. We cannot today draw the same sharp distinction as we can between the finality of the old English system, in which there was no appeal and if the judge did not exercise mercy, you were hanged at the conclusion of the assizes, and “Angola finality” extending over forty years. The average stay on death row is almost two decades, so what those who are executed in fact receive is a life sentence followed by a lethal injection. Perhaps the finality we care about most is symbolic: the sentence expresses our conviction that you cannot be rehabilitated, that we have finally given up on you. Or perhaps it is merely precautionary: the only thing we are interested in is making sure you do not do whatever it was again, and this is an effective way of achieving that.

The thought that this is the sort of finality that many people want is supported by the fact that supporters of the death penalty moderate their support if offered the alternative of life without parole. There surely are some people whom we just want out of the way; and death, on the one hand, and throwing away the key, on the other, are not very different ways of achieving that. The sort of finality that disturbs critics, on the other hand, reflects the fact that death is irreparable and that it is too late to change our minds about guilt or about the justice of a sentence once we have killed the person sentenced to die. That is both irrefutable and important, but it also suggests some of the reasons that lead me to say that all punishment can raise similar issues. Consider Ian Gordon, acquitted in October at the age of seventy of a murder he almost certainly did not commit fifty-two years ago. He has an IQ in the eighties, and when he was eighteen, he was shy, hesitant, eager to please, and just the sort of person who would confess to a crime he did not commit in order to be helpful. You cannot exactly give him back his life. Nor can we do very much for another not very bright middle-aged man who has been in jail for twenty-seven years because he will not admit to having committed the murder for which he was sentenced at the age of seventeen. Since he seems not to have committed it, one can hardly blame him for saying so, but no parole board will release him until he “comes to terms” with his crime. It is not clear that these men’s plight is very much less than that of the person who has been executed, and it is not clear that the finality with which their ordinary lives were terminated was very much less than the finality of death. We think, intuitively, that killing people is very much more dramatic than locking them up, and I do not want to deny that it is. But I do want to say that we now think so for reasons that have more to do with our attitudes toward the death penalty and less to do with what we actually inflict on the criminal who is punished.

Let us leave finality and rest on this: although there is nothing about the death penalty that means, intrinsically, that we may not inflict it, and that the victim of it has a right not to have it imposed on him or her under any circumstances, there are a set of rights to just treatment under a system of criminal law that constrain what can justly be done to anyone. We all have a right to be subject only to a fair system of investigation, arrest, interrogation, and prosecution. This might perhaps cover a limited amount of so-called racial profiling, in the sense that it is not unfair—if it is done scrupulously and sensitively—to keep a particular eye on people who are more likely than the rest of the population to have committed some crime or other. Known homophobes cannot complain, in this view, if an obviously homophobically motivated crime is followed by some scrutiny of their actions, just as the discovery of a road-death victim with a great smear of white paint on their clothes rightly sends the police looking for white cars and panel vans. It does not license a priori assumptions about the likelihood that any given murder was committed by a member of some racial or income group. A person who can reasonably claim that the police set out to stitch him up or that they did not question their own assumptions about who was guilty of the crime can reasonably claim that their rights have been violated. (Of course, they may be guilty nonetheless.) This applies to all crimes and all penalties and, only to the extent that the death penalty is special, applies with special force to capital cases.

The process of chasing, catching, interrogating, and processing for prosecution is one part of the story. Since some people are unusually vulnerable to ill-treatment and unusually inept in self-defense at this stage, the foregoing arguments apply more sharply to them, which is why I made some tart remarks about the educationally subnormal. As for the trial itself, I have already said that the same considerations apply with even more force. If we are to jail people, let along hang them, we must be sure that all that can be said on their side has been said. This entails defense lawyers no less competent and no less well provided for than the lawyers for the prosecution. I perhaps ought to say that I do not think of a good lawyer in quite the way that many Americans do in this context. I do not think that the adversarial theory of justice means that the task of a defense lawyer is to get his client acquitted no matter his guilt or innocence. Lawyers are agents of the system of justice as much as or more than they are agents of their clients. They are officers of the court as well as hired help for the dubiously well behaved. It is easy to confuse two distinct thoughts: the first that the best way of ensuring justice is to operate an adversarial system of justice so that each side can properly scrutinize the other side’s case, and the other that justice is whatever results from having two sides fight it out as best they can. The second is not true; a criminal trial is not a boxing match—in the latter, the object is to batter your opponent into a condition where he cannot continue to fight, but in the former, the object is to discover an objectively true answer to a part-factual, part-legal question: did the accused commit the crime of which he is accused? Just as the finding of this answer is frustrated if the defense lawyers see it as their job to trip up the prosecution by any means possible and to get the jury on their side by any degree of grandstanding, misrepresentation, and whatever lies or near lies that the judge will let them get away with, so it is frustrated—and much more frequently—if the accused has poor representation. It is in this sense that my own answer to the question whether the hundred and forty people executed in Texas under the auspices of Governor Bush had their rights violated would be yes. They may or may not for the most part have been guilty, and perhaps even have been guilty as charged of the crimes for which they were sentenced, but they did not have an adequate defense, and they were not pursued, arrested, interrogated, and prosecuted under a squeakily clean system. It was not their being put to death but their being put to death as part of that process that was the violation of rights, and the rights that were violated would have equally been violated if they had been jailed for twenty years instead.

Now, one response to this sort of argument—and one that I am more sympathetic to than you might suppose—is that no system of criminal justice can be squeaky clean in the sense I am demanding. Hard-bitten cops and criminal lawyers sometimes observe that it is warfare out on the mean streets and that a fastidious concern for justice on the battlefield is likely to get you killed. Let us take the analogy seriously. Those who make war on civilized society cannot complain of injustice if society makes war on them. But that thought is covered by the right of self-defense and immediate retaliation. Persons who adopt a posture of incipient aggression cannot complain if they are taken seriously, and a cokehead with a toy gun who has the bad luck to wave it at an armed shopkeeper suffers bad luck but no injustice if he is shot. But even in warfare there are rights in plenty to be considered. The first is that noncombatants have the right to be immune from aggression; too casual an attitude to victims caught in the cross fire and too casual a view of what evidence we need before we decide that someone is an incipient enemy remove the case from lawful self-defense and turn it into wanton aggression. And there are further arguments behind that one. One is that we should not accept the hard-bitten view too swiftly. Far from turning ordinary criminal justice into domestic warfare, the task of civilized states has always been that of trying to make real warfare more like domestic criminal justice. We try—and we do not wholly fail—to conduct war as a measured exercise in deterrence and retribution rather than an all-out exercise in slaughter and incapacitation; the notion lying behind the attempt to do so is in part that we want enemy societies to rejoin the society of nations in due course. If that is an ideal, then too much dwelling on the idea of the control of crime as a sort of domestic warfare is unhelpful.

In short, I end where I started. There are no grounds for saying that the death penalty is always and everywhere a violation of rights. Few people think, to take a concrete instance, that Adolf Eichmann’s rights were violated when Israel hanged him, even if they think that Israel might have done better to hold the trial and then let him go, or at any rate to have found him guilty and not hanged him. A death penalty imposed for those crimes, if any, for which it is uniquely effective as a deterrent and to which it is not disproportionate as a penalty does not violate a criminal’s rights so long as the criminal was arrested, tried, and convicted according to the strictest rules of justice, and so long as the criminal was of sound mind and fully responsible for their actions. This, in effect, is to say that I agree with the Supreme Court’s decision in Furman v. Georgia in 1972, but that I would take rather more persuading than the Supreme Court has lately taken that the appropriate degree of fastidiousness has been maintained.12 My guess is that no society that persistently thought as the majority in Furman thought would in fact impose the death penalty, but it might, and especially in time of war. If it did not, I suspect it would be because any society that thought in that way would mind about things in addition to rights—such as compassion and forgiveness and reconciliation and rehabilitation—and would simply not wish to employ capital punishment. But there are many things we have the right to do that we do not do, and that we think ourselves right not to do.

Which brings me to a last, entirely informal coda on the question of the death penalty in its own right. I have said that all punishment works by depriving people of things they value, whether money, liberty, or life itself; that we employ the threat of such deprivation as a deterrent to lawbreaking; and that what we may threaten is restricted by a constraint of retributive appropriateness such that the harm we inflict on the wrongdoer must not be disproportionate to the harm he or she inflicts. These principles allow the defender of the death penalty to claim that death is a proper harm to threaten for crimes of a sufficient degree of gravity—murder in our society, but blasphemy in others, corruption in others, and drug dealing in many. My purpose has been to argue that it is all the things that may go wrong between the commission of a crime and the execution of the offender that should concern us. But I now turn briefly to some speculation on why the death penalty is felt by so many people in the United States to be peculiarly appropriate. Notice the limitation of time and place; in societies where incarceration is difficult and expensive, physical penalties such as beating, mutilation, and execution have an obvious utilitarian justification they do not have for us.

In the American context, a trawl of pro-death-penalty Internet sites, some of which are of considerable sophistication and display an impressive grasp of the empirical literature, suggests that above all they are animated by the conviction that it is a divine commandment to kill murderers. The role of religion in the debate is thus interestingly confused. Thirty-nine Christian denominations have made formal and official declarations against the death penalty, including those not otherwise noted for their cultural liberalism, such as the Southern Baptists; but there seems to exist among the population at large a fundamentalist and literalist determination to take the words of the Old Testament at face value and to repudiate the attempts of church leaders to invoke Christ’s gospel of forgiveness in condemning the death penalty.

This is not to deny any of the obvious empirical evidence that opinion alters when the death penalty is compared to other punishments such as life imprisonment without parole, that enthusiasm for the death penalty drops when it is explained that it is not cheap, or that it is not particularly effective as a deterrent. One may be motivated by a religiously based morality but still pull back from its implications if they are at odds with other values or with common sense. Still, we need some explanation of the fact that some Americans, and particularly Americans in some parts of the United States, are attached to the death penalty for reasons that evidently do not move their fellow countrymen in Michigan or New Jersey. Supporters of the death penalty who concede that it is too expensive a form of punishment concede the point; they do not welcome it. This is different from the 20 percent of the sampled population who shift from supporting the death penalty to opposing it when they are asked to compare it with life imprisonment without the possibility of parole. They plainly hold a “throw away the key” view of the matter, and if assured that the key can be thrown away more cheaply without killing people, they take the alternative. The rest think death is what counts.

Together with the anecdotal evidence of the websites, this suggests that the familiar finding that the United States is the outlier among developed countries in being more or less immune to secularization has implications in penology too. It may also be related to a feature that the United States shares with Britain. In both countries, criminals are more inclined toward a strongly retributive and generally rather violent view of punishment. Felons support the death penalty more strongly than the population at large; this leads some advocates of the death penalty to suggest that this is “horse’s mouth” support for the deterrent effect of the penalty, and leads opponents to suggest that it confirms that human beings are poor judges of their own motivation. At all events, what we might conclude is that simple, firmly structured worldviews facilitate the thought that death is the apt and appropriate penalty for murder in particular. One might observe in passing that the Old Testament catalogue of death-penalty offenses, though shorter than the list of 222 that the British penal code ran to in the eighteenth century, is a great deal longer than that of the modern supporters. Bestiality and adultery are not offered up as candidates for stoning and incineration.13 Nor is blasphemy.

But this raises the question of what it is about the culture that sustains this selective but energetic espousal of a religiously based approach to punishment. One possibility is that it reflects a form of cultural pessimism. This is not to say that it is pessimistic about economic growth or about the possibility of making large fortunes or about a good deal else. What, then, might it be pessimistic about? Perhaps about the innate sinfulness of mankind, about the immovability of the stain of evil that wicked actions place upon us. It is hard to give a clear account because the cultural view itself so colors the things we are describing; but the thought I have in mind is this. A wholly secular view tends to be forward looking and not entirely judgmental: given that a criminal committed crime x, what are the odds that they will do it again, what is the risk that letting this one out will weaken the deterrent effect of the penalty, and so endlessly on. The nonsecular view is that the person had stained their soul, and might seek divine mercy but could not look forward to divine justice with any optimism. When I say the cultural view colors the phenomenon, I mean that the secular mind tends to see the past as in some sense eradicable and forgettable, whereas the religious mind tends not to. If people have incurred the stain that merits death—not that death erases it—then death it is. You might agree that people have become changed characters since their crimes, and much else besides, but you would harden your heart in the interests of preserving the rigid connection between the crime and the penalty.

As to why we must kill criminals, the best account was given by a two-thirds mad aristocrat and emigré some years after the French Revolution. In his St. Petersburg Dialogues, Joseph de Maistre gives a particularly disgusting account of a public execution in which he imagines a man being broken on the wheel—that is to say, being stretched out across a cartwheel and being killed by having all his limbs systematically smashed by the executioner. The scene ends with the executioner walking through the crowd, which draws aside at his passing; “Justice drops a few coins into his hand, from a distance,” says Maistre. Then he goes on to argue that honor, glory, and civilization itself rest on “this incomprehensible agent.” It is precisely the fact that killing people is so strictly forbidden that makes it the thing to do when we wish to make a point; the man who does the killing would be a murderer or a butcher if he were not an agent of the state. It is this that means we cannot quite approve of the executioner, no matter how skilled he may be. It also means that we—collectively, that is to say—cannot quite make up our minds about the way we want death inflicted. If it is to be an exemplary spectacle, it should be public, and the degree of brutality involved is not especially important. There is a degree of cruelty beyond which the motives of the state might be impugned, so limits must be observed; but making it easy for the criminal is not the object of the exercise—if it were, he might be given poison unawares. (Nevada did try in the 1920s to asphyxiate the condemned by pumping gas into their cells while they slept; it did not work.) If it is not to be an exemplary spectacle, the temptation is to medicalize the whole business as if it were a form of euthanasia—as when people executed by lethal injection have their arm swabbed with antiseptic before the fatal needle is inserted. That defeats several of the objects of the exercise, but I would not suggest that the choice before us is to embrace public and exemplary hanging on the one side or to give up on the death penalty entirely on the other. The social costs associated with public execution are just one further topic that has been left unexplored in this already overextended essay.