Mixed theories of punishment hold that punishment is just only if two conditions are met:
Bentham and Kant offer purified theories that rely only on the desirable consequences of punishment or retribution for wrongdoing. The problem with each one is the sufficient condition for punishment.
Utilitarians like Bentham have to say that the desirable consequences of punishment are sufficient to show that it is justified (permitted, just, acceptable, etc.). That is a problem because punishing the innocent could have desirable consequences and so would be justified, according to the utilitarian’s proposed sufficient condition.
Kant’s idea that guilt is a sufficient reason for punishment also gives rise to trouble. There can be good reasons for deciding not to punish some guilty people. For instance, the social consequences of doing so might be terrible or there might be a better way to spend social resources. But Kant maintains that it would be unjust for the guilty not to get what they deserve.
Hart has a clever way of understanding the two conditions listed above. He sees them as answers to different questions about punishment. In his opinion, the utilitarians have the correct way of answers a question about what he calls the justifying aim of punishment while the retributivists have the right answer to what he calls questions about the distribution of punishment.
We spent a lot of time talking about the paragraph on the top of p. 12 because it is very hard to understand. Here it is
“Similarly the moral importance of the restriction of punishment to the offender cannot be explained as merely a consequence of the principle that the General Justifying Aim is Retribution for immorality involved in breaking the law. Retribution in the Distribution of punishment has a value quite independent of Retribution as Justifying Aim. This is shown by the fact that we attach importance to the restrictive principle that only offenders may be punished even where breach of this law might not be thought immoral: indeed even where the laws themselves are hideously immoral as in Nazi Germany, e.g., forbidding activities (helping the sick or destitute of some racial group) which might be thought morally obligatory, the absence of the principle restricting punishment to the offender would be a further special iniquity; whereas admission of this principle would represent some residual respect for justice though in the administration of morally bad laws.” (1959, 12)
As I understand it, what Hart was saying was that the retributivist’s answer to the question about the distribution of punishment is independent of the retributivist’s answer to the question about the general justifying aim of punishment.
In the case at hand, if the general justifying aim of punishment is that those who are guilty of moral wrongs should suffer, then the state should not punish anyone who violates these laws.
Do these laws meet the retributivist’s standard for the general justifying aim of punishment? Absolutely not. Even so, the retributivist’s standard for the distribution of punishment is still relevant. That is so because it is independently unjust to punish people who do not violate the law. That shows that the retributivist’s answer to the question about the distribution of punishment is independent of the retributivist’s answer to the question about the general aim of punishment. And that shows that the two questions are separate from one another.
The bulk of the article is taken up with a discussion of justifications, excuses, and mitigating factors. These are all reasons recognized in the law for not punishing those who have done things that normally are grounds for punishment.
Hart’s main points concern excuses. These are conditions that show that someone had special difficulty in complying with the law: mental incapacity, say, or ignorance of what one was doing. They are relevant to the distribution of punishment: we excuse those who are mentally incompetent or who act in ignorance, even though what they did was wrong. (By contrast, we think that those who kill in self-defense are justified. Killing is usually wrong but, in this kind of case, it is not.)
Hart said that Bentham’s attempt to give a utilitarian rationale for excuses was a failure: there could be good utilitarian reasons to punish the innocent. As Noah noted, the most that Bentham shows is that it would be pointless to threaten people with punishment who have excuses such as “I was just a child,” “I was insane,” or “it was an accident and so not in my control.” The threat of punishment cannot deter accidents or people who do not understand it. But punishment itself might deter crime anyway: people would be more careful if they knew they could not make a case for being excused. Hart concluded that the reason for excuses must be based on principles of justice that have nothing to do with producing good consequences.
Hart could have rested content with saying that it is simply unjust to punish people like this. But he went further than that and offered some reasons explaining why this is so and expanding on the value of distributing punishment on retributivist grounds. He claimed that it would only be fair for society to enforce its laws if people have the opportunity to avoid punishment; this is possible only if punishment is limited to voluntary actions by the mentally competent. He also argued that this way of distributing punishment makes sense for societies that place a high value on individual freedom (see Hart 1959, 21–22).
Hart, H. L. A. 1959. “Prolegomenon to the Principles of Punishment.” Proceedings of the Aristotelian Society, New series, 60: 1–26.