Effective? Who cares?

Matt responds to my post on whether or not torture works thusly:

Megan counterproses that "people take the hard stance and say 'Yeah, torture may still work, but we still shouldn't use it because it's wrong.'" I think Megan thinks that people from the "torture doesn't work" camp are arguing in bad faith, but I'm really not. I don't think it makes any sense at all to say that there's a categorical moral against smashing people's fingers with a hammer or whatever other depraved acts of torture you may care to imagine. After all, I believe (as most people believe) that it's sometimes morally praiseworthy for the state to have its agents kill people with bullets, bombs, mortar shells, etc. so there's surely some end such that torturing someone would, if effective, be a just method of achieving that end.

The difference is that despite the horrors of war, there's a very strong argument to be made that if good people systematically disavowed war-making as a practice that bad guys would run roughshod over us. When Hitler's tanks start rolling across Europe, someone's got to shoot back. By contrast, I don't see any examples of societies using routinized legal torture to gain a decisive advantage over their foes or any evidence that the current era of torture has been a net positive in fighting al-Qaeda. To say that a method of investigation works "provided that you can verify the information" is, after all, merely to beg the question. Consulting a psychic works provided that you can verify the information, but spending person-hours chasing down the psychic's "leads" isn't going to make the country safer.

I don't think that Matt &c are arguing from bad faith. Indeed, I agree with them that torture in most cases isn't very effective. I don't agree with what we might call the "strong case" against the efficacy of torture, which is that it never works. If I have, say, a kidnapper from whom I want to get the location of his victim, and I can credibly promise to shoot out his kneecap if the victim isn't where he says she is, then I think I have a reasonable chance of finding out where the victim is by torturing him. If he knows, it is clearly in his best interest to tell me rather than a) playing dumb or b) giving me false information.

Now, those cases might be rare. Which is why many people, I assume Matt included, make what we could call the "weak case" against torture, which is that it generally isn't that effective. But I don't think that this is a very good argument to deploy if your goal is, as mine is, a legal ban on torture by the US government. The weak case doesn't prove we shouldn't use torture; it just proves that we should limit it to cases, such as the above hypothetical, where there is a reasonable likelihood that it will be effective. I doubt the rules for doing so would be as complicated as, say, the New York City building code.

The other problem with the weak case is that torture can theoretically be made more effective. Those brain scans are real; a workable machine might be less than a decade away. (It also might well not; the history of science journalism is littered with the corpses of "next big things" that turned out not to, y'know, actually work.) If you cannot make the case against legal torture without resorting to efficacy arguments, what the hell do you do if it becomes pretty damn effective?

My position is that even if it is 100% effective--in the sense of producing only true information--we should ban it. I don't trust anyone, not myself and certainly not the state, with the power implied by sanctioned torture. I don't want to live in a state that tortures people. And I don't think you need an efficacy argument to make that case.