On Torture
A much discussed topic in the US lately is the question of torture - whether there are situations in which torture should be a legitimate interrogation method - and if yes, how to define these situations, and what should the allowed limits of torture.
I will now address the question, but I will on purpose delay stating my verdict on it, for the following reasons:
Firstly, I want to “filter out” the people who believe the question itself is a taboo, and shouldn’t be discussed at all; secondly, if you’re reading this and thinking “Before she says anything, I need to know whether she is for torture or against it” - I want you to take notice that this is what you’re doing, and to understand that it is wrong: you are setting yourself up for - how this cognitive bias is called? When someone assigns more weight than the argument merits to arguments which support their opinion, and vice versa? That thing.
How important is the question of the effectiveness of torture as an interrogation technique? I think that, surprisingly, it is not very important. Here’s why: as long as we agree that torture is not a “100% magic cure” that automatically enables us to prevent all the evil plans of all terrorists - and as long as we agree that there is a non-zero chance that torturing a person will provide us with information that will help to prevent a certain bombing - then it is not very important to try and estimate what this chance really is. Remember: we don’t really compare various interrogation techniques to determine which ones work best. No, that’s not the question here at all. The question is: assuming all else failed, should we resort to torture, or do nothing? If we discuss only effectiveness, then, no matter how ineffective torture is, it is more effective than doing nothing at all, right?
Also, I don’t think the effectiveness question can be really answered. I mean, do we have data that shows that a person, on average, has X% chance to give valuable information under torture? I don’t think it is possible to gather such data - security concerns, and also too many variables involved. I would be extremely curious to learn about such an estimate, if it exists - I suspect it is very low, because the prior is low - not many people have valuable information to share to begin with. Of course, interrogators sometimes can estimate this prior - the probability that a given subject possesses valuable information. Still, I don’t think someone can put a figure (even in probability terms) on the chances that a person who knows something of interest will “break” under torture - there are just too many variables there. So, we’ll just have to accept that torture gives us some chance (not high, but not zero, either) to prevent the deaths of some people in certain situations.
Now, what to make of this? Does this mean that, in some situations, torture should be legal?
It is a purely moral question, I think. On the one hand, we have that torture has a chance to save lives - on the other hand, we have the imperative that torture is wrong. Which one wins?
An important thing to note is that questions like this one can be answered, but not reasoned about. These are axioms; they can’t be backed up by logical arguments. Imagine two people who both agree that burning kittens is wrong. But how wrong? Person X would burn a kitten for a cure for cancer, and person Y would burn a kitten for a new house. “Everything has its price”, sure - but what is the right price for anything is exactly what constitutes our moral principles. Person X would be right to call person Y an asshole; but he wouldn’t be able to back this opinion up with reason. If you ask him, why one price is okay and another is way too little, he’ll just say “Because.” And another point I want to stress is: if person Y tries to say “But what makes X a better person than me? We both agreed that burning kittens may be justified!” - the asshole is wrong! He’s making the slippery slope fallacy. The price matters. A lot.
These two points are especially relevant to the torture dispute. We all know the hypothetical ethical quizzes of the form “Would you rather save N people with probability X% or save K people with probability Y%?” If you think that the answer depends on the specifics - in particular, on the numbers X,Y, K and N - then, in order to be morally consistent, you must admit that there exist (at least theoretically!) situations in which torture is the right choice. (Assuming that you agree that torturing someone is at most as morally wrong as killing them. This statement is debatable, I know, and I have a lot more to say about it; but at least for terrorists/ticking bombs situations, I think it holds.)
But the step from the theoretical to the practical is the most important. Do situations that justify torture really exist in practice? Let's try the extreme ones first, and take it from there. Suppose you uncovered an impeding bombing, and the terrorist in on his way. You have captured his coordinator. You are fairly certain that this guy knows the place and time of the attack. If he tells you, you could save everyone... oh, wait. Doesn't work. Why would he tell you the truth? You have no way of checking him. An “ideal” situation for torture as an interrogation practice is, for example, when a robber tortures a victim for the safe combination. He can immediately enter any combination the victim names - and if it is wrong, he'll just keep hurting the poor guy until he breaks and names one that works. In ticking bomb situations, I suspect it is usually different. The most crucial pieces of information are the ones you can't verify until it's too late.
But never mind that - we're trying to construct an extreme scenario, right? So suppose, for the sake of the argument, that you can verify the information. You are fairly certain that around N people will die if you don't get the information; you have the guy who has it; and you can check whatever he says, and keep torturing him until he either breaks or the time runs out. With probability X%, he will break. Do you do it?
For me, the answer is: depending on N and X, possibly yes. And no, I don't know the precise limits - but for sufficiently large N, the minimal X can be rather small. To name an extreme example: if the bombing is of 9/11 magnitude, and we're fairly certain that the guy has the information that can make it or break it - yes, then I'd say torture him. Or her. But what if it's a car bombing that would kill around 20 people? Then, I'm not sure... But probably still yes. But what if we're not so certain that the guy has the information? Or, what if we can't check him? In short, what if X is low?
Then, my answer is a firm NO. And, remember, the situation we discuss is still rather extreme! It has nothing in common with the “let's waterboard a bunch of people for a few hours a day for months, just because they might know something” situation. To make myself perfectly clear: I think that the motherfuckers responsible for that shit should all be in jail. No question about that for me. I'm not making the slippery slope fallacy: just because I can name some hypothetical situations in which I'd justify torture, it sure as hell doesn't mean that I justify it by default. It sure as hell doesn't mean that I agree with the “We shouldn't tie our officers' hands with laws and regulations, because then they won't be able to do their jobs efficiently” approach! Hell, yes, we should! We do want our officers to think very hard before they decide to torture or kill someone, damn it! We do want them to remember that there will be an investigation and they will be prosecuted, if they're wrong. That's the whole point of having laws. And CIA should not be above the law - despite the fact that it would doubtlessly allow the organization to run so much more efficiently, oh boy (the very thought of this “efficiency” is giving me chills).
In Yura's recent post where he complains about the people who regard the very topic of torture as a “critique stopper”, I suspect that Yura is mistaken regarding these people's intentions. Chances are, they were simply person X trying to explain person Y why he is an asshole - and failing, because moral axioms cannot be justified by reason. Instead of trying, they should have just said “We have an irreconcilable ethical disagreement” (a polite way of saying “You're an asshole”), and leave it at that - instead of coming up with appealing to emotion shots like “Imagine your loved one being tortured”.
By the way, maybe I should explain why I think that “Imagine your loved one being tortured” is a terrible argument - as well as its mirror argument, “Imagine your loved one in a building that will fall down if we don't stop them”. The thing is: we're discussing government policy - and government policy should be impersonal. For example, if I had to kill 10000 random innocent people in order to save Danny's life, I would definitely do it - exactly as I would expect any one of these 10000 people to kill me and Danny and all the others in order to save her boyfriend. But I don't want either of these decisions to be a government policy. The law (unlike particular people) should treat everyone's loved ones equally - and therefore it has no choice but to go by the numbers.
So, to sum up: I could accept torture being legal in certain extreme situations (the conditions for which should be specified in the law). I could also accept the stance that torture should never be legal, ever, under no circumstances. With both of these I could agree. The stance that says we shouldn't look too carefully into how our security is ensured, and we should just let our officers do their jobs and find us some other issues to vent about? This is a stance with which I cannot agree, and which I, in fact, find appalling to the extreme. Sorry, Yurchik.