Wednesday, July 06, 2005

Limits of reason

Economists look at the issue of the decade and the rational choice theorists conclude that suicide terrorism is ... irrational (via Marginal Revolution):

Suicidal terrorism is a far stronger counter-example to narrow self-interest. From a nonevolutionary point of view, it is impossible to reconcile the two. No matter how much you receive for your services, it does you no good if you are not alive to consume anything. Furthermore, if you get paid first and die later, there is an end-game problem. A selfish agent would take the money, then do everything in his power to back out.

This paper, by GMU economist Bryan Caplan, does conclude that there is a form of rationality to be found in this irrationality, which he unsurprisingly dubs rational irrationality. In other words, even most would-be terrorists would rather not kill themselves and opt out at the last minute. And some of those that do kill themselves are operating under extreme circumstances in which irrationality makes some kind of rational sense.

But he ultimately concludes that most of them are in the grip of flat-0ut irrational irrationality.

An interesting exercise, which also concludes that there are temptations to appeasement when deterrence is ineffective. But didn't we know this stuff already? Isn't game theory more useful when people have not been driven beyond reason?