Unqualified Offerings

The torture piece has generated some reactions that deserve amplification. The Illuminated Donkey avers,

I have to admit I’m a bit unclear about Henley’s argument. His leading point seems to be that the track record of government terrorist investigations simply doesn’t warrant entrusting the authorities with that sort of power, with the Higazy case being the current example (though it appears that coercive interrogative methods approaching torture were used to elicit some sort of confession).

I’m not really sure how he moves from that point to the doomsday scenario he describes, however. His argument seems to be that if we take the step of allowing torture in order to prevent potential attacks, there will still be situations (imagine a worst-case Higazy scenario) where this leads us to make wrong choices, hypothesizing a case where a torture-induced false confession leads agents down a blind alley with disastrous consequences. But even if this were the case, it’s not as if investigation is the zero-sum game he describes, where we have to throw all of our resources at one location. Some of his statements indicate that he has a moral position against using torture, while others indicate that his take is a more practical one.

I do have a moral position against using torture. In fact, the mere notion that an American feels the need to convince other Americans not to torture makes me so mad I could just fucking spit. I abominate torture because of all that self-ownership stuff. Torture is the violent attempt to destroy another’s self-possession; it attempts to make an animal of a person and to own that animal. That is vile.

Not that everyone necessarily cares! Viz. Alan Dershowitz and his “torture warrant” enthusiasm, which only shows how “the rule of law” has come to mean “rule by lawyers” – so long as the professional apparatus in which Alan Dershowitz has status, power and a whole lot of money controls the torture action, what could be wrong with that?

See? I’m spitting, just like I warned everyone. Let me try again.

The primary argument in favor of an official resort to torture is utilitarian, and Americans are a practical people. So any hope of opposing torture means meeting utilitarian arguments with utilitarian objections. The utilitarian argument for is that the magnitude of possible harm from terrorist attacks are so great that to forego torture might mean a nuke destroying a city or a plague annhilating the country – sometimes called the “ticking bomb” scenario. My utilitarian counter-argument was paraphrased aptly by a reader I’ll call Air Force Virginia, just in case it’s not okay for active-duty military personnel to go around being quoted on websites: “See, I thought it was obvious–after all, torture doesn’t *work* if your goal is to get accurate information. It only works if your goal is to get your victim to say a particular thing, and you must of course tell him what you want to hear so he can say it.”

Pause to clarify a couple of things:

1) I am mad. I am not mad at the Donkey. The Donk merely offers an opportunity to clarify arguments that might have been drafted better in the first place, something of a trend here lately. (See “Wilderness of Carters” below.) The Donk, when you can keep him off basketball, is on the side of the blogging angels.

2) I don’t argue that torture will never give you accurate information, and I doubt AF Virginia would either. I do argue that it goes awry in the “tell them what they want to hear” way often enough to render it useless for the purposes of the argument actually advanced – that is, the “ticking bomb.” Hence our title: Over the years there have been experimenters who believed they had taught apes to use sign language or dolphins to squeak recognizable english. “Clever Hans” was a horse in 1900s Berlin who could do math. A nice capsule of the case appears on a website whose author abjures quoting without permission. But it’s short. Anyone who does not know the case should read it.

Hans could not do math. But he could interpret cues of which his testers were unaware, so if his testers knew the answer to a problem, Hans could pick that answer. It made them happy, and that he could do. Once someone thought to use testers who didn’t know the answers to the problems, Hans, well, Hans’ mathematical abilities regressed to the equine mean. So Koko the Gorilla, so Alpha and Beta in Day of the Dolphin.

It should be pretty clear what happens if the animal is keying on the tester’s cues and the tester “knows” the wrong answer to the problem: the animal chooses the tester’s preferred, wrong answer.

Torture makes animals of human beings. It reduces them to hindbrains attached to way too many nerves. “Successful” torture leaves a “client” whose only concern is to stop the torture, and the only way to do that is to satisfy the torturer. There is no guarantee whatsoever that a truthful answer satisfies the torturer even if the victim has a truthful answer to give. But you can bet that confirming a preexisting hunch will, and you can bet that the victim has a desperate incentive to figure out from verbal and nonverbal clues what you think he should be telling you.

Which just restates at much greater length what I said telegraphically the first time. The new business is very serious, The Donkey’s argument that “it’s not as if investigation is the zero-sum game he describes, where we have to throw all of our resources at one location.” I argue that while an investigation is not a zero-sum game, it is a game of diminishing returns – there are finite resources and they can be wasted. Even the largest investigation has finite resources to bring to bear and finite egos directing it. Most of all, in the “ticking bomb” scenario, it has finite time to accomplish its goal. Clara Thomas Boggs wrote, anent wrongful convictions,

To me, it is very clear that a wrongful conviction begins at the scene of the crime, then after the first errors are made, all other decisions are made to protect the initial one. This is where I would begin making things right — at the investigative level.

Note that Boggs does not say that the errors have to be willful or malicious, nor even that the “protection” needs to be. Once investigators have a theory (this “grad student” and his radio are connected to the attack on the trade center, e.g.), they will tend to evaluate evidence on the basis of how well it fits the theory. This can become, in paranoid enough hands, positively rococo. Viz. the history of CIA Counterintelligence Chief James Jesus Angleton and “the prophet Golitsyn,” detailed in Unqualified Offerings touchstone text Wilderness of Mirrors, by David Martin.

And that is how I “move to the doomsday scenario.” The torture enthusiasts are trying to sell us torture on the basis of doomsday scenarios – the “ticking bomb.” We shouldn’t buy, for the reasons set forth above.

Civilian Virginia (Postrel) weighed in on the issue too the other day. She agrees that torture should never be legitmized as a legal investigative technigue. I worry that her proviso, “If circumstances are so compelling that such cruel and corrupting practices seem justified, the torturers should face the possibility of punishment” may create some dire incentive problems. If the possible punishments are slaps on the wrist, you get a fair amount of torture. If the punishments are draconian, you get a fair amount of coverups. And if the unofficial direction from on-high is to torture regardless of the risk of punishment, you get cynicism in the ranks. But I take her point, I think, that for all this theorizing, real investigators may find themselves with an honest-to-god ticking bomb scenario on their hands and an obviously guilty party in front of them with testicles just waiting to be squeezed. Who can say we wouldn’t squeeze them ourselves in that case?