Let's Talk Torture

It's not easy to talk torture. First, you have to get around the verbal pitfalls. Are we talking about rules, or a lack of rules? Are the people being interrogated detainees, or prisoners? Terrorists, or suspected terrorists? Prisoners of war, or enemy combatants?

I'm going to try to avoid those verbal traps by jumping straight to the HOTM. Which we can get to via two questions: (1) what is torture; and (2) does it work.

What is torture? It's easy to define torture at the margins, but the margins are always easy. An ocean view room at the Ritz Carlton Maui is not torture. Having your fingernails pulled out is.

But somewhere in the middle things become much more subjective. Is keeping someone in an uncomfortably cold room torture? What if the room has inadequate light? Is interrogating someone for four hours straight torture? Eight hours? Twenty-four? Intruding into someone's personal space? Yelling in his face? What about solitary confinement? Silence vs noise? Insults? Threats of violence?

Let's put aside the first question for a moment and examine the second. And then we'll see how they're linked.

Doe torture work? That is, does it lead to effective intelligence? I'm not sure. I've never tortured someone and I've never been tortured. But my guess is that the answer is, it depends. On the kind of information you're after; the amount of information you already have (by which you can judge the quality of what's extracted); the kind of torture employed; perhaps most of all, on the strengths, weaknesses, determination, susceptibility, and other idiosyncrasies of the person being tortured.

In other words, the honest answer to the question of "Does torture work?" is probably... sometimes.

Okay, then. Why not just employ the practice wholesale? When it doesn't work, nothing lost. When it does, much is gained.

Let's back up there a minute, and take a look at that notion of "nothing lost." If torture really did cost nothing, we'd be doing it all the time. And we wouldn't stop at waterboarding, either... as Ving Rhames' character Marsellus Wallace put it in Pulp Fiction, we'd just "get medieval." That we're not suggests some inherent awareness of torture's inherent costs as well as its possible benefits.

I can imagine several costs. First, the practice brutalizes its practitioners and by extension the society that condones it. Second, it spreads easily because it's an easy substitute for more exacting, nuanced, and effective forms of interrogation. Third, it produces a lot of false data along with the occasional actionable intelligence because the person being tortured will offer up anything to make the torture stop.

Let's look at two extremes. First, we torture wholesale. By doing so, we extract every bit of meaningful intelligence the subjects have in their minds. But we also incur enormous costs, as detailed above, including so much false information that the real leads are largely obscured and rendered useless in the process. Conclusion: wholesale torture doesn't make sense.

Second, we do everything we can to make interrogations pleasant. We keep the subjects in five star hotel rooms; we feed them three gourmet meals a day; we submit questions to them only with their express consent and desist the moment they tell us they're tired. We incur none of the costs detailed above. But we get no useful intelligence, either.

It seems, then, that somewhere in the middle lies an optimum balance -- tactics tough enough to be effective (what is torture?) without being counterproductive (does it work?). And the fact that those tactics are effective without being counterproductive enables our society to justify them, and to bear them, as an ugly necessity.

Torture offers benefits, but also costs. Opponents ignore the former; proponents, the latter. The uncomfortable truth lies somewhere in the middle.