One of the latest episodes in the war against ISIS was a raid by U.S forces that killed a leading figure and captured his wife. According to news stories, the intent was to capture Abu Sayyaf, believed to play a major role in financing ISIS via black market sales of oil, for the sake of valuable information he could provide about ISIS operations. Killing him, while better than nothing, was not the preferred outcome.
That raises an obvious question. Assuming the raiders had been successful, how did they expect to get at information in Abu Sayyaf's head? The most obvious conjecture is by torture—which the U.S. government claims not to engage in. An alternative possibility is by threatening his wife—also not, so far as I know, a tactic U.S. forces admit to using.
A central principle of economics is revealed preference. What people do provides more reliable information than what they say.
62 comments:
"One of the latest episodes in the war against ISIS was a raid by U.S forces that killed a leading figure and captured his wife."
-As Clay Claiborne points out, this is dubious: not even leading researchers of the Islamic State recognize this name.
http://claysbeach.blogspot.com/2015/05/was-isis-leader-abu-sayyaf-promoted.html
I thought the U.S. government does admit to engaging in torture, but they simply call it by some euphemism. "Enhanced interrogation techniques" or something like that. Everybody knows this means torture but deep down everyone also understands that it keeps them (at least marginally) safer.
By the way, I don't understand the argument that torture isn't effective. Surely, some high proportion of detainees divulge information under torture, and some significant proportion of those who crack must give up information that is useful and truthful. It's not like competent, imaginative torturers are in short supply. Torture must also have some deterrent qualities.
Does any active military really not use torture?
Under torture, prisoners will give up seemingly useful information whether they have any or not, parroting back bits of what we've told them we want to hear. The vast majority of prisoners don't know anything particularly valuable, so torturing gets you lots and lots of false leads that wastes time. Sometimes the true answer is "I don't know anything" - and we want to be able to HEAR that answer.
Torture is likely to reduce the chance of being voluntarily given info that you might have gotten by befriending the prisoner. It turns people who weren't previously our enemies (and their friends and families) into new enemies. There is the signaling problem that if we are known to torture, people will try harder to resist capture. There is the signaling problem that we can't credibly promise to STOP torturing once we start - as soon as somebody says something promising under torture, the lesson we learn is that we should torture them more to get more.
Then there's the problem that it turns us into monsters. And that professional interrogators tell us that in practice it doesn't work. And that every time some official has claimed it helped stop terrrorism, it turned out that they were lying.
I have trouble following discussions about torture because torture is so seldom defined. My definition is: something we won't do to our own soldiers in training, e.g. cut off part of their body. Some of the things labeled torture strike me as unproblematic like playing music they don't like, subjecting them to something they abhor that won't hurt them (e.g. menstrual blood) I think the issues differ depending on what techniques we are discussing.
"Some of the things labeled torture strike me as unproblematic like playing music they don't like,"
It can be if it is loud and incessant. The purpose of "playing music they don't like" is to leave the victim disoriented and deprive them of sleep. Sleep deprivation can kill a person.
Kevin,
Again definition. Sleep deprivation can kill, but also it can be what parents of an infant have which is disorienting sometimes, but not torture. My sense (but my problem is actually knowing) is the torture occurs, but sometimes people are talking about things I don't consider torture.
You guys are saying torture of *one* individual may not work. But what if you torture two individuals? Can't you easily cross match what they say?
I thought the U.S. government does admit to engaging in torture, but they simply call it by some euphemism. "Enhanced interrogation techniques" or something like that.
Recently, the military under President Obama admitted that the enhanced interrogation techniques performed under President Bush were torture. As of now, *those* techniques are officially off the table.
Under torture, prisoners will give up seemingly useful information whether they have any or not, parroting back bits of what we've told them we want to hear.
As Megan McArdle said, "arguing that something doesn't work isn't necessarily an argument for not doing it--it could just as easily be an argument for improving our technique. And if advances in brain scanning research let us develop a reliable lie detector, as seems possible in the relatively near future, then torture will work very, very well" ( http://www.theatlantic.com/business/archive/2009/04/does-torture-work/16520/ ).
If torture became effective; and I mean highly effective, would you change your mind about whether it is acceptable? I assume you wouldn't, but then it seems silly to bring up the effectiveness if you consider the effectiveness irrelevant. Why mention it at all?
Max: Oddly enough, your point about a lie detector comes up, in a fantasy contet, in my second novel:
"Magister Coelus once spoke to me of the limits of magery to extract information without destroying the mage one used it against. It had not occurred to him that the combination of torture with truth telling, a simple magery familiar to our enemies as to us, provides a means of extracting information from a prisoner with no injury to his mind, only his body."
Ricardo: Your argument assumes that the two people haven't coordinated their story in advance. Also that, if they disagree, you know which one is lying.
On the general issue of torture ...
I think it's hard to argue that it is never morally justified. Not so hard to argue that, if it is classified as an acceptable practice, it will end up being used mostly in contexts where it isn't justified.
"Torture is the particular bane of the terrorist, just as antiaircraft artillery is that of the airman or machine-gun fire that of the foot soldier."
Bernard B Fall's introduction to Roger Trinquier's "Modern Warfare".
David,
Also that, if they disagree, you know which one is lying.
You can torture them until they agree.
I highly agree with the earlier comment that definitions matter.
The truth is that the general public doesn't know what's happening in Guantano Bay. We can be pretty sure, though, that it's things the administration doesn't want the general public to know about.
I think it's hard to argue that it is never morally justified. Not so hard to argue that, if it is classified as an acceptable practice, it will end up being used mostly in contexts where it isn't justified.
Well put. Torture may be one of those weird things where we're better off with the contradiction: it's best if we don't publicly acknowledge it but don't put too much effort into stamping it out either.
@Glen:
I acknowledged that the number of prisoners who, under torture, give up information that is both useful and true is less than 100%. But it is certainly not 0% either. At what percentage does torture become justified?
If the vast majority of prisoners don't know anything, that seems like it may be an argument to select prisoners for torture more strategically, but has no bearing on whether to torture those who are selected.
It seems circular to say that torture reduces the chance of being voluntarily given information. By definition, information gotten from torture is not given up voluntarily.
And what's this about befriending the prisoner? I can believe that could be effective some percentage of the time, but I find it much harder to believe that percentage will be higher or even about the same as for torture.
About turning us into monsters: context might be useful. I got on a plane at Gatwick airport in the summer of 2006. Another passenger on that plane had decided he would use liquid explosives to blow up the plane while it was in the air, killing himself, me, and everyone else on board. After sitting on the tarmac for 5 hours, we learned that he had eventually been discovered and detained. His planned actions, which would have killed well over 100 innocent people, were monstrous. If we tortured him (one person) and get information that would save the lives of hundreds more, I don't really see the monstrosity in it. Personally, even if we just tortured him as punishment I wouldn't lose sleep over it...but that's just me.
I think it's hard to argue that it is never morally justified. Not so hard to argue that, if it is classified as an acceptable practice, it will end up being used mostly in contexts where it isn't justified.
I like that formulation. I had considered linking to your previous post about torture of slaves in ancient Greece, but I find McArdle's statement more menacing.
I recently started a job in New York, which means I have a lot of time on the subways to read. I really need to read Harald and Salamander. And, for that matter, Machinery of Freedom.
"via black market sales of oil"
Why via the black market? Is it stolen oil?
"Is it stolen oil"
From the standpoint of anyone who doesn't regard ISIS as the legitimate sovereign of the territory they control, which I think describes pretty nearly every government in the world, yes.
Max mentions my old post on torture of slaves in ancient Greece. For those interested, it is:
http://daviddfriedman.blogspot.com/2014/12/torture-old-issue.html
That's it. And, reading it over, the part about Ancient Greece is a much smaller part than I remembered.
My definition is: something we won't do to our own soldiers in training
I quite agree with Norm. If it is something that you can get away with doing to your own troops in a training environment, then it is manifestly not torture. Techniques like water-boarding merely simulate drowning and I imagine it must be unpleasant as hell, which seems to be the point. But nobody actually drowns. For a captured terrorist engaged against us, I call that fair treatment.
As for reliability arguments about the resulting information, you would never likely use it in isolation anyway. You would always seek to corroborate it from some other source to verify its accuracy before relying on it.
"If torture became effective; and I mean highly effective, would you change your mind about whether it is acceptable? I assume you wouldn't, but then it seems silly to bring up the effectiveness if you consider the effectiveness irrelevant. Why mention it at all?"
Just wanted to pop in to say something about this: Just because the person making the argument thinks torture is unacceptable regardless of whether or not it is effective doesn't mean the person making the argument thinks other people think the same way. They might reason other people may factor in efficiency. In that scenario, making a statement about efficiency stands a chance of swaying those people on their own terms.
(Mind you, I have no idea if that's what happened here, but the general case checks out for me.)
pinkgothic, unlike McArdle, I don't think people making the "we shouldn't torture because it's ineffective" argument are opportunists. I agree that they seem to believe it's a persuasive argument, but that doesn't mean it's the argument that convinces them.
Fundamentally, I have a hard time imagining the kind of person they think they're trying to persuade. I strongly suspect that, by and large, they've never met anybody who disagrees with them as far as torture is concerned so they're essentially debating imagined opponents.
I'm not sure revealed preference applies to counter-factual speculation.
Revealed preference applies to observed behavior. In this case I am assuming that the description of what was intended, capture rather than killing, is correct—perhaps it isn't.
If it is correct, we can ask what it means for the U.S. to go to a good deal of trouble to try to capture someone who had information that would be very useful to the U.S.
What are the economic arguments for and against torture? Shouldn't torture be a good idea in economic terms?
'How did they expect to get at the information in Aby Sayyaf's head? The obvious possibility is torture-'
A lot of professional military interrogators say they just keep asking different people till they find the ones who are both informed and willing to talk. Abu Sayyaf's disgruntled secretary, trusted henchman, wife- might have been the real hope for info.
The more torture is done, the more it becomes regarded as generally acceptable, the more people will do it, and the more chance that it will be used someday on you, or on someone you care about.
If you play nice, that doesn't mean that everyone else will play nice too. However, if you choose to play nasty, that tends to increase the number of other people who'll be willing to play nasty too.
That's why torture "is prohibited under international law and the domestic laws of most countries" (Wikipedia).
@Jonathan:
It's not like torture was invented ten years ago. If it's a prisoner's dilemma then the analogy should include the fact that both prisoners come from a long line of inmates who've been living in the prison for hundreds or thousands of generations.
Like I said, torture may be one of those weird things where it's better for everyone to verbally agree that it's bad and inexcusable and that only barbarians would even consider doing such a thing, while we all also understand that every goverment does it.
No, of course torture wasn't invented ten years ago: it was more commonly used in olden times when playing nasty was the norm.
"... only barbarians would even consider doing such a thing, while we all also understand that every goverment does it."
In other words, we all understand that we're governed by barbarians and hypocrites, who pass laws that they don't obey. That may be the case, but I don't regard it as satisfactory.
@Jonathan:
"That may be the case, but I don't regard it as satisfactory."
Yes of course; it's good to repeat that mantra. But it's also a scab we shouldn't be too hasty to pick off. Torture is nasty and barbaric and fuels our governments' lies etc. etc., but it ostensibly keeps us safer.
I don't believe that it keeps me safer. I'd prefer the laws against it to be strictly enforced.
In fact, if I could press a button and rid the world of all torturers, I'd be happy to do so.
I don't mind killing people: death is intrinsically painless. But deliberately inflicting pain on people is a symptom of a mind gone bad.
Torture seems like an obvious first choice to you because you live in a culture which is pro-torture.
http://amptoons.com/blog/2015/05/18/enough-with-the-torture-scenes-please/
Norm, speaking as a person who finds normal-loud music acutely painful, you don't know what you're talking about. You're also not thinking about the effects of serious sleep deprivation.
Yes, befriending people can work-- probably better if they don't have other social contacts. People want to boast, or complain, or correct errors.
Military interrogation is a profession-- as I understand matters, one way to do it is keeping the prisoner comfortable but doing relentless questioning and pushing about any omissions or contradictions.
Aside from any other issues, people who are pro-torture seem to assume that torture can be done rationally, and torturers actually aren't good at that.
Probably of interest: _Torture and Democracy_ by Darius Rejali, about the evolution of no-marks torture. One point from the book is that torture is so much fun and so easy that organizations that torture tend to give up on rational methods of investigation.
@Nancy Lebovitz:
You've got it backwards: I think torture should be a last choice (but still a choice--just, one we don't readily fess up to; that's right, I'm IN FAVOR OF MAINTAINING A CONTRADICTION), and I live in a culture which is anti-torture. (Which, as I said, is a separate matter from whether we actually torture. We do.) (If we weren't anti-torture, then where does the pressure to close down Gitmo come from? Even the word "torture" is avoided as being too taboo. If we're anti-torture, then what's up with that?)
If torture can't be done rationally, then why should "professional military interrogation" be doable any more rationally? In general I think we should assume some basic level of irrationality to all human endeavors and go from there.
I don't know about this point in Rejali's book. It seems more like torture is something that people who join torturing organizations are eager to do (after all, you get to personally inflict pain on an enemy of your country, with your own safety guaranteed), but which they quickly burn out on. (Imagine the difference between a torturer who just started his first day on the job and one who just finished his second week.) Then they go on to claim that torturing doesn't work, etc.
It's possible to come up with imaginary situations in which torture might be the best choice; but, in the real world, people can and should do without it. The cost of using it (making the world a nastier place to live) exceeds the value derived from it.
Interrogation should consist of asking people questions. If they don't answer, OK. You can threaten them with death, if they deserve death anyway; but that should be the extent of it.
I don't willingly pay taxes to support sadists.
@Jonathan:
It's possible to come up with imaginary situations in which torture might be the best choice...
I've got a real-life situation. It's the plane I was on in the summer of 2006, when a Muslim guy tried to blow us all up with liquid explosives. (Thank God he was detained.)
It'd be good to know what organization he was a part of and what their plan was, and how they come up with their plans. (Several other flights that day removed Muslim passengers carrying liquid explosives.) It'd be good to know how their organization is structured, how it's funded, how they recruit, what they plan to do next. People who work in international intelligence probably have loads of questions I haven't even thought of.
It seems to me that someone willing to blow himself to bits in the name of his false god isn't likely to become friends with his interrogators or divulge information under threat of death.
Let's pretend that the fact that someone somewhere is being tortured means, by itself, that the world is in general a nastier place to live. I'd still rather live in that world than be dead because bits of me are falling from the sky.
Power Child: In the situation you describe, your life was saved without torture.
Yes, it would be nice to know plans, organization, etc., but who knows whether such information would be obtainable through torture or how much use it would be.
If you're willing to torture someone for the faint chance of getting that sort of dubious information, I'm afraid I'd watch fragments of you falling to Earth without much sympathy.
The world is cleaner without bad people, and people who torture others are bad people.
People in first-world countries live in comfort and security. Their backs are not against the wall. They have no urgent need to authorize desperate measures to protect themselves.
Yes, they have a tiny chance of dying through some terrorist attack. They have much more chance of dying from road accident, cancer, alcoholism, etc. I'd suggest paying attention to what matters.
Power Child, I was replying to David Friedman's "Assuming the raiders had been successful, how did they expect to get at information in Abu Sayyaf's head? The most obvious conjecture is by torture".
The reason interrogation (of the sort I described) is more possible to do rationally is partly because it's a rational process of information and deduction and partly because it isn't as much fun.
@Jonathan:
It's possible that the terrorist who wanted to kill me was detained because of information gotten from the torture of someone else.
It's easy to believe that torture does not always yield useful true information. It's hard to believe that torture never yields it. It's also hard to believe that torture almost never yields it.
(Forgive me for repeating the syntax pattern, but...) It's easy to believe that people who torture others as part of their duty in an organization are doing bad things. It's hard to believe that this makes them bad people, even if they eagerly volunteered for that duty. Again, imagine the 19 year-old recruit who looks forward to getting his hands on real live enemy combatants--the ones that his comrades are risking their lives to defeat--with virtually no threat of physical harm to himself. His instincts are normal--part of the darker side of ourselves we'd rather not think about, perhaps--but not evil.
Comfort and security do not last if you're not willing to take a few "desperate measures" to defend them. There's a balance there of course, but it's definitely one of those things where the libertarian habit of pushing things to their logical conclusions does not work.
Also, you've committed the fallacy of "You're arguing about X, that must mean you think X is the most important thing in the world and that you don't care about A, B, C, and D."
@Nancy Lebovitz:
Oops, my mistake. It wasn't clear who you were responding to before.
As I said though, I don't think most torturers consider torturing fun after a few days on the job (maybe when they start to realize it isn't like in the movies). I imagine that in real life it indeed takes a lot of planning, patience, deduction, cross-referencing, etc. And, importantly, I'd bet that torture is almost always used in conjunction with other forms of interrogation, even including friendship-building. (A movie example, but in Zero Dark Thirty they alternate between torturing a guy and treating him as if he were a valued advisor, complete with offering him fine food and cigarettes.)
Power Child: It's not hard for me to believe that torturers are bad people. There are many other things they could choose to do in life.
And no, I haven't alleged that you believe terrorist attacks to be "the most important thing in the world". However, you clearly believe them to be important enough to justify torture, and I disagree with you.
If you lived in Syria, say, I might have more sympathy with your outlook. People there really do have their backs up against the wall: survival is a major problem.
Does libertarian theory have anything to say about torture? About the treatment of prisoners in general?
My understanding of libertarianism is that it's mainly about not using force on other people, which would seem to rule out torture.
A libertarian can use force as a defensive measure against aggression; but torture seems a purely aggressive use of force to me. As far as I know, it's normally used on a helpless victim who has already been subdued and tied up.
@Jonathan:
Are torturers bad people? Seems like a chicken and egg question. It's plausible that the job selects for a disproportionately large percentage of bad people, but I still don't buy that these are anywhere near a majority of people who do the job. If your argument is simply that engaging in torture makes you a bad person, then you must believe 100% of torturers are bad people and I just have to disagree with your premise.
Just because we aren't on the brink of extinction does not mean it's a bad time to think about our continued survival. Things can change rather quickly. If in a generation or two we found ourselves in Syria's position, then I'd say it's probably too late to just start thinking about survival.
"If in a generation or two we found ourselves in Syria's position..."
Syria's position is civil war. If the USA finds itself in that position in the future, I don't think throwing a bit of torture into the mix is going to help.
When the USA actually experienced civil war in the past, I don't have the impression that either side was much into torture. At least, it doesn't seem to feature in the accounts I've read.
Yes, I believe that engaging in torture makes you a bad person. You disagree, I shrug. Disagreement among people is a common fact of life.
Unspeakable Acts, Ordinary People-- I've only read a little of the book, but the author seems to conclude that torturers are people who are really good at rationalizing what they do.
Are soldiers, policemen, and prosecuting attorneys also just ordinary people who are really good at rationalizing what they do?
Casino managers must also be good at rationalizing what they do--and they're not protecting anyone or anything except their bank accounts.
I am sorry if this has been discussed. I think our attitude about torture is one that gives us the ability to deny it and clears our conscience. We pretend torture is not given any moral authority in our democracy and sleep better at night. Or, to quote a famous movie line: "You want me on that wall. You need me on that wall."
@anonymous:
The notion that we don't in fact "need them on that wall" (torturing enemies to get information that will help protect us, and also perhaps as a deterrence against other enemies) is only true if you accept as true the claim that torture is completely or almost completely ineffective.
@Power Child- 'Does any active military really not use torture?'
Say you have a million-man army. One percent at least will be the scum of the Earth and love torture- not just waterboarding but the fun stuff with pliers and genitals. That's ten thousand volunteers hustling to questioning prisoners. Again, a lot of soldiers of excellent character to start with, will be warped by the horrors of war. The military has a lot of institutional experience with keeping these dogs on a leash. Partly hypocrisy and PR, but mostly because the other soldiers hate scum to the point of fragging or other self-help, even to the point of careful administrative procedure.
The result is a big, long-term complex trend towards building leashes inside the military. Meanwhile, actual trained information-gathering professionals have built a good record- Talking With Mr Charlie, for a well-written example of a professional who just plain talked to thousands of people and compared what they said.
On the other hand, our feckless post-60's governing class, which shirks military service as a political religion, twaddles about 'not really torture' if soldiers do it in training, or if Science. Cowards and fools and traitors.
@Bruce:
Well put.
"The most obvious conjecture is by torture—which the U.S. government claims not to engage in."
Is that so obvious? Perhaps, the intent was to bribe Sayyaf, where "bribe" might include giving him a more comfortable environment for serving his prison sentence for his terrorist crimes. In fact, since ISIS presumably tries to prevent Sayyaf from selling his information, one could make a libertarian argument that trying to capture Sayyaf was just a way to help him access information markets. (Ok, this last sentence may be pushing it.)
The most obvious conjecture is by torture—which the U.S. government claims not to engage in.
Or maybe they intended to interrogate him without using torture. Given that the raid could potentially be for either purpose, I don't see how it's a revealed preference for the former over the latter.
I'm glad you post here. I'm sickened by the other posters.
@Josiah Neeley: Thank you! Your comment spurred me to go back and re-read the original post. You know, I think you're right, I think Dr. Friedman made an unjustified leap there. As he said, killing Sayyaf was not the preferred outcome -- because dead he could give no information of any kind.
You can conclude that this reveals a preference for torture or violent extortion only if you are sure that those responsible believe Sayyaf would be impervious to any other kind of interrogation. They might well believe this. (And they might be right.)
But I don't see how we could know that, and it doesn't accord with what I (think I) know about tradecraft. History is full of people in positions of trust and responsibility who have been turned by surprisingly trivial mechanisms.
Yeah, if you think there's any form of non-torture interrogation that can work, then that would be a justification for capturing rather than killing a high-level member of ISIS. I wouldn't be the least bit surprised if we are still torturing people (it turns out only disposable enlisted men from West Virginia ever get in any trouble for that sort of thing), but I don't think this story is strong evidence one way or another.
I think David's formulation of the argument against torture is probably the strongest one. Specifically, we're not going to be able to allow the government to torture only in the contrived ticking time bomb case--if we let it use torture then, we also let it use torture much more broadly. In practice, the people we know who have been tortured or died under interrogation were overwhelmingly not involved in some kind of ticking time bomb situation. Instead, they were mostly low-level nobodies who were being interrogated to help us hold down Iraq or Afghanistan.
The analogy I'd use here is RICO and civil forfeiture--both created to go after a small set of really serious organized crime figures, both now very widely used in very different sorts of cases. If we actually end up institutionalizing torture, we'll see its use become very widespread, until we're torturing people to get information on tax avoidance schemes or potential political embarrassments. (How many people would we have tortured to keep the Snowden leaks from happening?)
Good points, NOTA.
What do you people think of Steve Keen's claim that the axioms of revealed preference are wrong?http://www.debtdeflation.com/blogs/2011/08/04/behavioral-finance-lecture-01-debunking-revealed-preference/
Do you think his arguments are any good?
Was the civil forfeiture-esque use of torture on low-level nobodies for questionable reasons widespread before the big international crackdown on it? If not, isn't that evidence that it wouldn't be again in the future?
Several years ago, I had an exchange with Cook in the comments at skepticalscience (or whoever responds to comments; he represented himself as the person running the site). I asked for clarification on the statement on the site that the fact of increases in global temperatures was evidence that it was caused by human activity.
We had a long exchange, it which I tried to explain that a fact cannot be evidence for its cause. I was astounded that this guy, who claimed to be a climate scientist, not just disagreed with me, but could not understand what I was saying. He just restated that one of the primary lines of evidence to show that humans were causing the increase in global temperatures, was that global temperatures were increasing.
I eventually gave up, because I seemed to be having a discussion with someone who could type words, but didn't understand what they meant.
Post a Comment