Friday, September 27, 2013

The prisoners' other dilemma

A person is made up of multiple selves across time.  In a sense, you can never really punish the agent who committed a crime, because that agent is long gone.

Now, we care about our future selves, so the threat of punishment to our future selves deters us from doing bad acts today.  However, we frown on threats to our family members, even though they would also deter us; is there a qualitative difference?

You could say we are more closely "related" to our future selves than to our family members.  It's wrong to punish your brother for something you did, because he's so obviously not you.

Hypothesis: People who commit serious crimes are systematically less likely to care about their own future selves.  They are not so closely "related" to their future selves, and so do not take the future consequences as deeply into account, which is why they commit crimes.

If so, we penalize specifically the people for whom there is the biggest disconnect between the agent who committed the crime and the agents who must pay the price.

That makes me kind of uncomfortable.  But note that the legal system treats "crimes of passion" as a special case, presumably in recognition that the agent who committed a crime of passion is no longer around.

Drug addiction is especially susceptible to the problem identified here.  Strong addiction induces an obsessive focus on the present, and thus a disconnect between present and future selves.

Maybe it's wrong to punish your brother on your behalf because he clearly had nothing to do with the bank robbery, and didn't even benefit in any way.  By contrast, future versions of yourself may be beneficiaries of the robbery, and indeed might agree that current-you should have committed it.  Perhaps a sense of common agency comes from agreement over what should be done.

But in the case of drug addiction, that link is broken.  In general, drug addicts harm their own future selves by their present indulgence.  Any current smoking makes it both harder to quit and harder for the addict to satisfy the craving in the future.  The worst!


  1. An alternative issue is that punishments are not always issued for the purpose of deterrence, alone. A person who is prone to commit crimes of passion every once in a while may not be *that* self at the time of the trial, but he has a greater chance of returning to a similar state in the future, and thus, may be a danger to society. Putting dangerous people behind bars is an alternative motive to give harsh punishments for precisely these people who are not often themselves.

    I think there's another motive to punish hidden in your post - retribution. Somehow, issuing a harsh punishment for the one responsible for the crime is a measure of satisfaction for the victims and/ victims' families. If the agent doesn't seem like the self that committed the crime, I'm just as uncomfortable using this motive as I am the deterrence motive.

    To clarify, take the case of Ariel Castro. Two-faced man who was often not himself, and rarely revealed the side of himself that committed his crimes to the world. In this case, I think the lawyers got it right. It's better to send someone like this off to jail than to try and deter the next Castro (what could deter someone as deranged as that man?) or try to exact a measure of retribution that would (realistically) not come. More often than we think from an optimal crime setup, the point of punishment is to keep dangers to society away from the vulnerable.

  2. Thanks, Tony. That's the trifecta, isn't it? Deterrence, Retribution, and Prevention, let's call it. (Prevention = locking someone up so they can't commit future crimes).

    Deterrence and Prevention are both perfectly legitimate reasons to hurt some agents for the sake of social welfare, even if those agents didn't do anything wrong. That's what it is, though.

    The especially irritating thing about Deterrence, as opposed to Prevention, is that it penalizes precisely those people who weren't deterred. Suppose people have a distribution of types in [0,1] and for any given level of deterrence, there is some threshold type T such that only people with type t > T commit the crime. The value of the deterrence level is that it keeps certain people with t < T from committing the crime. Meanwhile, for t > T it is just a pointless punishment, not a deterrent. In an ideal world, if we knew people's types, we would make the punishment apply only to t < T. But in our world, we end up only punishing precisely the people with t > T.

    Prevention, by contrast, works properly. In cases where the punishment entails keeping the agent away from society, we lock up precisely the people at highest risk of committing future crimes.