To: Catherine C.
From: Geoffrey Klempner
Subject: Free will and the justification for blame and punishment
Date: 17 November 2006 13:12

Dear Catherine,

Thank you for your email of 9 November, with your first essay for the Possible World Machine, in response to the question, 'In the light of the critique of "free will", can blame and punishment ever be rationally justified?'

This is an interesting essay. But I would agree with you hunch that there is something missing.

You have set things up very well, with some very good and believable examples. In some of these cases, it seems pretty clear that in committing canicide you have performed an action which merits punishment. In other cases, it seems pretty clear that you have performed an action which merits help, or treatment.

Courts of law have to deal with cases like this. Courts are set up for a practical purpose which somehow has to mesh with our ideas of justice. But how sound are these ideas in the light of the critique of the naive or common sense notion of 'free will'?

There are two issues here, which need to be separated. First, there is the all-or-nothing argument that determinism precludes free will. You can agree, or disagree: either way, you are then faced with the necessity of giving an account of the nature of justice and punishment.

On a 'hard' determinist view, punishment is just a way of altering controls. How we choose to alter those controls - either with a spell in prison or a course of treatment in a secure institution - will depend on our view of how amenable the culprit is to persuasion, their capacity to think rationally for themselves.

On a view which refuses to accept that human beings are merely cogs in a causal chain, the very same kinds of question will arise. Some persons demonstrate by their actions that they are capable of rationally deciding for themselves - are responsible for what they do - while others do not.

A move one can make here is to 'bracket' the metaphysical problem of free will and determinism. Whatever view you take on the metaphysical debate, you will still be faced with the same practical problems of knowing when or when not to punish.

(I am over-simplifying somewhat: determinists tend to argue that punishment is never justified and that all punishment is merely a form of 'treatment'. However, there will still be a choice between the best form of such 'treatment'.)

So the essay title is somewhat misleading. The 'light' cast by the critique of free will is not that 'blame and punishment can never be rationally justified', but rather, 'Every action has causal antecedents, and there is no sharp dividing line between the kinds of antecedent conditions which lead to responsible actions, and the kinds of antecedent conditions which lead to actions for which the agent is not responsible.'

In other words, your answer to the question could take the form of looking for some principled way of separating the two broad classes of case. In all probability, it will not be possible to do the separation neatly, but a fuzzy distinction is still a distinction. As demonstrated every day in the courts, there are easy cases and hard cases.

Your examples are sufficiently rich to give us a good purchase on the kinds of consideration that would be relevant. What is a 'fit of rage', road rage for example, or the rage which led to the awful case reported yesterday of a dispute between school children developing into murder by arson? Are we always responsible for our rages, or only sometimes? On what basis would one make a distinction? Likes and dislikes are one thing; the question is whether, for a normally rational person, one accepts that certain extreme provocations are sufficient to 'madden' us, i.e. deprive us of the responsibility for our actions.

Your reference to 'crime passionel' suggests that in France unfaithfulness was (I don't know if it still is) considered such a provocation. An alternative interpretation is that a husband who coldly planned and executed his wife's lover's murder would still be in a position to claim that it was a 'crime of passion'.

Drug dependency, partner abuse, inability to reason clearly each raise thorny issues (you cite the example of the drug addict whose reason is 'blurred' but a simpler example might be someone with a very low IQ).

Another class of cases which you don't consider concerns alleged cases of brainwashing, such as the famous Patty Hearst case. Then there are the less spectacular but more common examples where one person comes under the strong influence of another.

The more we look at the nitty gritty details, the more difficult it seems to make a meaningful distinction. As I've remarked, courts are set up with a practical purpose. Maybe one would be led, ultimately, the view that blame and punishment are never rationally justified - not by some clever metaphysical argument, but just by becoming aware of the depressing reality of cases like these.

You've made a good start, despite your misgivings. I look forward to your next piece of work.

All the best,

Geoffrey