The nature of the death penalty, its irrevocability, together with the safeguards necessary to avoid errors in its application are such that it can never become an effective means of crime prevention or of societal protection. This point will be illustrated through a discussion of three aspects related to the application of the death penalty:

More than two hundred years ago Beccaria noted that severity and certainty of punishment are hard to reconcile. In this he was referring to the inverse relationship that often exists between the two variables: the more severe the punishment, the less certain it is. The same point was well stated in what is known as the "Rejected Preamble" of Sir Samuel Romilly's Bill of 1808:

"Whereas the extreme severity of penal laws has not been found effectual for the prevention of crimes; but, on the contrary, by increasing the difficulty of convicting offenders, in some cases affords them immunity and in most cases renders their punishment extremely uncertain".

This rule applies to capital punishment more than to any other. In my study of the deterrent effect of capital punishment¹⁵ I compiled data on charges and sentences for murder, manslaughter and for other violent offences in Canada during the period 1881-1967. The average yearly conviction rate for murder proved to be the lowest among all crimes of violence.

Capital punishment is known to exercise an inhibitory effect on juries and judges in capital cases. Those who are familiar with the administration of justice know that retaining the death penalty, especially if it is made mandatory, reduces the likelihood that indicted offenders will be convicted. In this way, the menace of the death penalty tends more to protect the accused through intimidation of the jury than to protect society through the conviction of the murderer. In a democratic, fair system of justice it is doubtful that the death penalty can ever reach the certainty level necessary for it to operate effectively as a deterrent. This difficulty was outlined in the introduction to the British criminal statistics furnished by the Home Office

"In consequences of the strong proof of guilt necessary for conviction of crimes punishable by death, the proportion of acquittals for murder is higher than for most other crimes, and an acquittal in such a case does not necessarily imply failure to detect the perpetrator of the crime".⁴⁴

One of the paradoxes of the death penalty is that if it is made mandatory it results in a high acquittal rate and leads to many murderers being let free. If, on the other hand, it is made discretionary then its application becomes arbitrary, discriminatory and erratic.

If capital punishment is really a unique deterrent then a scarce and sporadic use of it would undoubtedly weaken its deterrent value by reducing the possibility and the threat of execution. It is in this fact that the real paradox of capital punishment lies. If it is used lavishly it loses its horror, people become accustomed to it and are no longer affected or deterred by it; if it is

rarely applied, then the probability of incurring it sinks to insignificance in the minds of potential offenders.³³ This led him to declare that "the death penalty probably can never be a deterrent. Its very life seems to depend on its rarity and therefore on its ineffectiveness as a deterrent".

Another paradox of the death penalty relates to the nature of the crimes for which it is provided. Where it is wanted and used, that is for crimes of violence and sex, it is not likely to be effective. Where it might be effective (such as for rational economic crimes) it is not wanted.