It is often supposed that a theory of punishment predicated on desert lends support to the death penalty. What leads to this assumption is a prior thought about the appropriate punishment for murder: If we are to punish murderers as they deserve, we will inflict on them what they inflicted on their victims, namely death. This association between a desert-based theory of punishment, known as retributivism, and the death penalty appears not only in academic writings on the subject, but in popular views of punishment as well. Public rhetoric in support of the death penalty, for example, is nearly always retributivist. Politicians urging its use in a particular case will more readily speak of justice and desert than of future dangerousness or setting an example for others. They evidently think the retributivist argument for death more appealing than the utilitarian arguments that might be made in its favor.
In my view, however, the faith that death penalty proponents place in the retributivist theory of punishment is misplaced. In this essay I argue that retributivism fails to justify the use of death as punishment, and, moreover, that a desert-based theory of punishment is particularly ill-suited to such a task. I shall not argue against retributivism as a theory of punishment per se. Although there may be reasons of a more general nature to reject retributivism, I will not attempt to make the more general case here. My more limited suggestion is that even if retributivism succeeds in justifying the practice of punishment overall, it cannot provide a compelling reason for including the penalty of death in that practice.
The present essay constitutes one piece of a more general argument against the death penalty. If I am correct that capital punishment cannot be justified on retributivist grounds, the death penalty proponent would need to find a different basis on which to argue his case. The most promising alternative line of argument appears to be general deterrence: The death penalty is justified against this murderer, he will have to argue, in order to deter other people who might murder in the future. But as I argue elsewhere, even if we grant the questionable empirical assumption that the death penalty really does deter, it is extremely difficult to justify the death penalty on the basis of deterrence alone. (1) If I am correct that retributivism does not lend support to the death penalty, the death penalty proponent will find it substantially more difficult to argue the merits of that form of punishment, for he is unlikely to find support for his position elsewhere in political philosophy.
The implications of this larger claim for the death penalty are worth spelling out. If the death penalty cannot be affirmatively justified, it cannot be permissibly imposed. Why? More specifically, why not simply be agnostic about its moral permissibility, and allow practical considerations to determine whether we should employ it? The answer has to do with a fundamental assumption about the nature of punishment, namely, that punishment is a harm or evil to the person on whom it is inflicted. (2) Although this assumption is a standard one in the literature on punishment, its implications have not been adequately noticed by legal philosophers and criminal law scholars. (3) If punishment, including the death penalty, is an evil that stands in need of justification, then the burden is on proponents of the death penalty to justify its use, otherwise the death penalty must be assumed to be impermissible. Thus this relatively uncontroversial assumption about the nature of punishment has powerful implications for a debate about the moral permissibility of the death penalty: the death penalty may not be permissibly imposed unless it can be affirmatively justified. It also follows that the death penalty opponent need not show that the death penalty is morally unacceptable to make his case. …