I happened to read a book published in 1965, the year Britain legislated to end the death penalty, titled Murder Followed by Suicide, by the distinguished criminologist, D.J. West. For forty years up to that date, about a third of homicides had been followed by the suicide of those who committed them.
Most people who committed homicide followed by suicide were highly disturbed psychologically, if not outright mad. For example, in killing their families they imagined that they were saving them from a worse fate. They were not the kind of people who would be deterred by anything, including the death penalty.
Here was a natural experiment. I hypothesized that if the death penalty acted as a deterrent, the homicide rate would increase but the proportion of homicide followed by suicide, which in absolute numbers would remain more or less the same, would decrease. My friend, the criminologist David Fraser, looked at the actual figures and found that this was indeed the case. Some sane people who might otherwise be inclined to kill managed to control themselves knowing that they might be executed if they did.
For the death penalty to deter, it was not necessary for it to be applied in every case. Although the death penalty for murder was mandatory in Britain, it was commuted in nine cases out of ten. All that was necessary for it to deter was that execution was a real possibility. We shall never know whether the death penalty would have deterred even more if it had been applied more rigorously.
Does its deterrent effect, then, establish the case for the death penalty, at least in Britain? No, for two reasons. First, effectiveness of a punishment is not a sufficient justification for it. For example, it might well be that the death penalty would deter people from parking in the wrong place, but we would not therefore advocate it. Second, the fact is that in all jurisdictions, no matter how scrupulously fair they try to be, errors are sometime made, and innocent people have been put to death. This seems to me the strongest, and perhaps decisive, argument against the death penalty.
Against this might be urged the undoubted fact that some convicted murderers who have been spared death have gone on to kill again, and this will continue to be so. Victims of those who murder a second time are probably more numerous than those executed in error. Therefore, utilitarians might argue, even if mistakes are sometimes made, that the death penalty overall would save lives. (Let us disregard the fact that those murderers who go on to murder a second time would not necessarily have been executed after their first murder, for nowhere are all murderers executed.)
The argument holds only if utilitarianism is accepted as a true ground of ethics. But few of us would accept that it is. It might be that hanging the wrong person after the commission of a terrible crime would have a better social outcome than hanging no one at all, provided only that it was never publicly known that the wrong person had been hanged: but we would still be horrified at the prospect. Moreover, in practice, the execution of the innocent, once it is known, serves disproportionately to undermine faith in the justice system. And surely it is true that for the state to kill an innocent man is peculiarly horrific.
