Maybe this makes sense if you think of the punishment as having a deterrent component and a retributive component. Someone who attempts a crime but fails needs to be deterred from future crime, but does not owe the same magnitude of retributive debt to society.

I'll note that despite the above I'd probably prefer a justice system that focuses on deterrence and rehabilitation, and omits retribution.

+1. There's no material societal benefit in retribution. (True, it makes some people feel good, but so does burning a cross in someone's yard.) The purpose of a justice system is to keep us safe. Someone who hires a hitman should be punished, so that future people don't do it. Whether they succeed is irrelevant.

A counterargument I can imagine is that if we punish the unsuccessful hiring of a hitman more lightly, then future hirers are more likely to err on the side of a hitman with a lower probability of success. But that seems improbable to me -- one doesn't hire a hitman to see them fail.

> The purpose of a justice system is to keep us safe. Someone who hires a hitman should be punished, so that future people don't do it. Whether they succeed is irrelevant.

One small step further: someone who plans to hire a hitman should be punished, so that future people don't even think about it. Whether they manage to hire one is irrelevant.

Now you're punishing a thought crime and equating that with successful murder even.

Merely thinking about hiring a hitman is bad, yes, and should be punished, yes.

But I would not equate it with hiring one. That's because the expected damage from planning to hire a hitman is smaller than that from actually hiring one. A year of punishment hurts the criminal the same regardless of which of those two scenarios holds, but the social benefit of punishing them is less. Therefore we should punish them less. (Unless we don't care about the welfare of someone who's been convicted at all. I think that's a sick position to hold -- even setting aside the possibility of false positives -- but it has its adherents.)

A real eye-opener for me regarding these questions of weighing the costs and benefits (i.e. the economics) of punishment was Thomas Friedman's book Law's Order. Until I read that, I thought the whole question was just a political fight. And while it certainly is a political fight, there's a much more objective kind of analysis available -- in fact, it's the norm in economics.

David D. Friedman actually, Thomas Friedman is the columnist for the New York Times. His newest book is awesome by the way.

Ooh. You're referring to Legal Systems Very Different From Ours[1]? That does look fascinating.

And thanks for the correction.

[1]http://www.daviddfriedman.com/Legal%20Systems/LegalSystemsCo...

It is very, very good.

We call it a sting operation, not a thought crime.

> Maybe this makes sense if you think of the punishment as having a deterrent component and a retributive component.

It also makes sense if one rejects retribution as illegimate but accepts proportionality as a basis for limiting on punishment. (Many doctrines of criminal punishment accept both retribution as a goal and proportionality as a limit, so...)

Proportionality on outcome or proportionality on intent? If the assassin's gun misfires nobody is harmed. A punishment proportional to the outcome would be no punishment at all. But a punishment proportional to intent should not distinguish between the scenarios where the gun does and does not misfire.