I wish to explore the question of the relationship between desert theory and prison overcrowding. This requires an initial definition of desert theory. Desert, or retribution, is primarily concerned with the moral blameworthiness of an act. It focuses primarily on the crime, not the criminal. In determining the type and intensity of confinement, it focuses on the past act, not on the future criminality of the offender or others like him. It rejects deterrence, incapacitation, and rehabilitation on both moral and empirical grounds. Desert theory is based on a belief that it is immoral to impose undeserved suffering on some so that others may gain happiness. Supporters of the theory further argue that the data thus far accumulated by researchers is at best too ambiguous to show that any utilitarian goal can be achieved by punishment, and at worst, shows that in most instances these utilitarian goals have not been achieved.

Desert theory means, in a sentencing context, that all persons commit-ting the same crime should be sentenced to conditions which are similar in both type and duration; individual character traits, such as rehabilitation or recidivist potential, are irrelevant. But because desert theory is a morally based approach to criminality, it also requires that the substantive criminal law recognize excuses and defenses far beyond those recognized by the common law. Mens rea and a proper understanding of moral culpability and blame are essential as the basis for punishment. Defenses such as duress, necessity, diminished capacity, and perhaps even physical weaknesses such as premenstrual syndrome, would occupy a greater role in a desert-based substantive criminal law.

Thus defined, desert theory seems to increase the problem of over-crowding by threatening as a matter of principle to send every felon to prison, perhaps for a very lengthy period. I am grateful, therefore, for the opportunity to try to sift out for myself, and I hope, for others, the relation between desert sentencing and prison overcrowding.