Merged Senate Report on CIA Torture Program

[RandFan]

60 Minutes

Ali Soufan in 2011: Enhanced interrogation is ineffective

 

[sunmaster14]

The CIA was illegally torturing people before 9/11?

Are you serious? Do you really think that the CIA wasn't torturing people during the Korean War, or during the Vietnam War, or during the Cold War in general? One poster here alluded to a program called Project MKUltra. I had never even heard of it, but it appears that it did quite a bit of experimention with torture. I had heard about the Phoenix Program during the Vietnam War though. That was pretty gruesome. Also, the French had plenty of experience torturing people during the Algerian War, so I'm sure the CIA learned a lot from that too. Torture has actually been pretty common throughout history, and, yes, our side did it quite frequently during WWII despite intimations to the contrary here. The way in which the CIA's program in 2002-2003 was an aberration was that, for the first time ever, the CIA demanded some serious CYA authorization from the administration before proceeding. In the old days, there wouldn't have been any of that legalistic haggling.

 

[Crossbow]

Fair point. My evidence is that the CIA claims it works. These are experienced people who decided they needed more extreme measures to extract information, and they went to great lengths to obtain the proper authorities. What is Ken's evidence that there is no evidence that torture works? To begin with, he has to prove my evidence is flawed.

Of course the CIA is going to claim that torture works when it is the CIA who is doing the torture, and of course the CIA is going to claim that they had the proper legal authority to do the torture when they were the ones who were doing the torture.

This process is much like the one Scarlett used in Gone With the Wind when she felt sorry for getting caught doing the wrong thing instead of feeling sorry for actually doing the wrong thing.

Anyway, if torture works so well and if the torture was legally justified, then the CIA should be using torture all of the time instead of just once in a great while.

 

[Tony Stark]

Are you serious? Do you really think that the CIA wasn't torturing people during the Korean War, or during the Vietnam War, or during the Cold War in general? One poster here alluded to a program called Project MKUltra. I had never even heard of it, but it appears that it did quite a bit of experimention with torture. I had heard about the Phoenix Program during the Vietnam War though. That was pretty gruesome. Also, the French had plenty of experience torturing people during the Algerian War, so I'm sure the CIA learned a lot from that too. Torture has actually been pretty common throughout history, and, yes, our side did it quite frequently during WWII despite intimations to the contrary here. The way in which the CIA's program in 2002-2003 was an aberration was that, for the first time ever, the CIA demanded some serious CYA authorization from the administration before proceeding. In the old days, there wouldn't have been any of that legalistic haggling.

What I meant was were the specific people in the CIA who conducted/ordered post 9/11 torture torturing people before 9/11. If so, what exactly did this torture accomplish besides making them criminals. If not, they didn't really have any special expertise.

 

[sunmaster14]

What I meant was were the specific people in the CIA who conducted/ordered post 9/11 torture torturing people before 9/11. If so, what exactly did this torture accomplish besides making them criminals. If not, they didn't really have any special expertise.

I'm speaking of institutional experience and knowledge. People who learned from torture passed on their information to the next generation and so on. It's not necessary for the people who first conducted the research or gained the experience to still be at the CIA, or even alive. I think that if torture really didn't work, then the CIA as an institution would have taught its employees that it didn't work. I mean there's really no good reason for using it if you know it doesn't work. It's not only demoralizing, but it puts your people in legal jeopardy too. And it's quite clear that the CIA was very anxious from the very beginning about putting their people in legal jeopardy. They knew that the legal and political environment was a lot different in 2002 than it was in 1972 or 1962 or 1952.

 

[RandFan]

Due process and prohibition of cruel and unusual punishment exist for a reason. Fundamental human rights are not merely officious. They in fact result in better societies.

 

[marplots]

I don't think this argument about "does it work/doesn't it work" is using the same idea of the role of torture on both sides. The one side seems to have an idea that torture is going to be used to extract specific information - like someone's pin code or where the bomb is hidden. Others would say it's just one piece of the larger picture, a link in a chain meant to develop intelligence generally. In this view, saying we could have got the info in another way is always true - a case is built up from many elements, no one of which is the "smoking gun." Any of them could be missing and it is quite possible the information would be found. This puts torture in a more minor, adjunct role, instead of the star of the show.

The latter view would have torture as a useful tool, even if it didn't "get the goods" in a clear an unambiguous manner. It might not even reveal any specific fact and still have a use as evidence that some suite of facts is true in a cumulative fashion.

It could also have a role as a threat. If information is given to avoid torture, no torture happens - does the torture get credit or not?

The general mechanism is valid, or seems so. "Tell us about X, or Y will happen." That's a common enough pattern. We see it in cop interrogations all the time - sometimes implied, sometimes stated outright: "The judge will go easier on you if you come clean now, before we get the DNA tests back."

The root issue about whether it works or not then has to do with what we take "works" to mean. As usual, it comes down to expectations and metrics. With two different operative definitions, both sides of the issue can be correct.

 

[RandFan]

The root issue about whether it works or not then has to do with what we take "works" to mean. As usual, it comes down to expectations and metrics. With two different operative definitions, both sides of the issue can be correct.

Granted. Now, the CIA and the FBI are both very good at compiling data. Both know very well how to apply bayesian statistics and other models coupled with proper controls to obtain valid inferences. Where are those studies? That something could be beneficial is not evidence that it is.

That said, why can't we lie to detainees? Do we really need to torture people to prove that we could in fact torture them?

One of the reasons I stopped supporting "enhanced" interrogation techniques was the absence of data to support its necessity.

 

[thaiboxerken]

Exactly which statement(s) in my post were wrong? And I don't suppose you can point to any research or credible source which corroborates your claim that torture has never produced any useful intelligence, can you?

It is very hard to provide evidence of a negative. For one who pretends to value logic so much, you should know what you ask is illogical. It's like asking for evidence that unicorns don't exist.

 

[marplots]

Granted. Now, the CIA and the FBI are both very good at compiling data. Both know very well how to apply bayesian statistics and other models coupled with proper controls to obtain valid inferences. Where are those studies? That something could be beneficial is not evidence that it is.

That said, why can't we lie to detainees? Do we really need to torture people to prove that we could in fact torture them?

One of the reasons I stopped supporting "enhanced" interrogation techniques was the absence of data to support its necessity.

I have no idea if it works or not, under either definition. But here's how I'd investigate it -

All those dudes/dudettes in SERE training. You give them a secret to keep and while you are torturing them (which you are going to do anyhow as part of the training), see if you can get them to spill the beans.

That still won't meet all the objections, but it's at least a way to experiment.

I think the stronger position would be: "I don't care if it works, it's morally repugnant and we shouldn't do it."

 

[jimbob]

I don't think this argument about "does it work/doesn't it work" is using the same idea of the role of torture on both sides. The one side seems to have an idea that torture is going to be used to extract specific information - like someone's pin code or where the bomb is hidden. Others would say it's just one piece of the larger picture, a link in a chain meant to develop intelligence generally. In this view, saying we could have got the info in another way is always true - a case is built up from many elements, no one of which is the "smoking gun." Any of them could be missing and it is quite possible the information would be found. This puts torture in a more minor, adjunct role, instead of the star of the show.

The latter view would have torture as a useful tool, even if it didn't "get the goods" in a clear an unambiguous manner. It might not even reveal any specific fact and still have a use as evidence that some suite of facts is true in a cumulative fashion.

It could also have a role as a threat. If information is given to avoid torture, no torture happens - does the torture get credit or not?

The general mechanism is valid, or seems so. "Tell us about X, or Y will happen." That's a common enough pattern. We see it in cop interrogations all the time - sometimes implied, sometimes stated outright: "The judge will go easier on you if you come clean now, before we get the DNA tests back."

The root issue about whether it works or not then has to do with what we take "works" to mean. As usual, it comes down to expectations and metrics. With two different operative definitions, both sides of the issue can be correct.

Well, being most generous to the idea of torture it might be good for getting information that is difficult to otherwise get easily - however it is significantly harder to get accurate information.

The CIA say it worked but can't give any examples. An FBI interrogator said it stopped one source of information and gave an example. There is at least one other case where a tortured Al Quaeda suspect cracked, then falsely confessed as well as falsely claiming that Iraq had been supplying Al Quaeda with WMDs.

I'm speaking of institutional experience and knowledge. People who learned from torture passed on their information to the next generation and so on. It's not necessary for the people who first conducted the research or gained the experience to still be at the CIA, or even alive. I think that if torture really didn't work, then the CIA as an institution would have taught its employees that it didn't work. I mean there's really no good reason for using it if you know it doesn't work. It's not only demoralizing, but it puts your people in legal jeopardy too. And it's quite clear that the CIA was very anxious from the very beginning about putting their people in legal jeopardy. They knew that the legal and political environment was a lot different in 2002 than it was in 1972 or 1962 or 1952.

If that was the case, why did they hire private contractors with no experience of actual interrogation - just of designing training regimes to help withstand interrogation?

There are reasons why the CIA turned to outsourcing. Few of them are good. The CIA had no experience in running detention centres. It was handed this mission abruptly a few days after 9/11. The Senate report stresses how “unprepared” it was. So it sought assistance from outside its ranks. The psychologists asked to conceive a set of “coercive interrogation techniques” had previously worked at a US Air Force school, training pilots to withstand the treatment they might face if taken prisoner or hostage. This was the basis on which they won this grim contract.

Meanwhile a couple of science blog posts in the guardian:
here (from 2010)

One of the interesting features of the torture debate is that many in the military and intelligence communities seem decidedly unconvinced about the effectiveness of torture. Ali Soufan, a former FBI special agent with considerable experience interrogating al-Qaeda operatives, pointed out in Time that:

When they are in pain, people will say anything to get the pain to stop. Most of the time, they will lie, make up anything to make you stop hurting them. That means the information you're getting is useless.

He isn't alone in this assessment – a number of former intelligence people have expressed similar views, and his words are echoed by the US Army Training Manual's section on interrogation, which suggests that:

…the use of force is a poor technique, as it yields unreliable results, may damage subsequent collection efforts, and can induce the source to say whatever he thinks the interrogator wants to hear.

or this about one of the CIA's claims:

The situation is further clouded by the fact that members of the George W. Bush administration made claims for the effectiveness of torture that have later been proven to be untrue. One such claim was that the water-boarding (simulated drowning) of Khalid Shaikh Mohammed produced vital information that allowed them to break up a plot to attack the Liberty Tower in Los Angeles in 2002. Slight problem - in 2002 Shaikh Mohammed was busy evading capture in Pakistan.

Well, no. Suppose I start beating you around the head, demanding that you tell me that Justin Bieber is in fact a supremely talented artist. Eventually, although it may take several days of torture to get there, you'll tell me what I want to hear, but that doesn't make it true.

The second major problem is that human memory just isn't reliable. Take a bunch of witnesses from any major news event: a bombing, 9/11, a car crash, wherever. The more people you interview, the more different stories you'll get, because our recall of past events isn't always very accurate. On top of that, there is a vast body of scientific literature telling us that one way to make a person's memory even less reliable is to deprive them of sleep, or put them under great stress, or otherwise confuse them. You know, like you do with torture.

In a stunningly obtuse piece of mismanagement, the same psychologists the CIA had contracted to engage in the torture were also assessing their own effectiveness, as detailed on page 473:

The CIA Inspector General Special Review states that CIA ‘psychologists objected to the use of on-site psychologists as interrogators and raised conflict of interest and ethical concerns.’ According to the Special Review, this was ‘based on a concern that the on-site psychologists who were administering the [CIA’s enhanced interrogation techniques] participated in the evaluations, assessing the effectiveness and impact of the [CIA’s enhanced interrogation techniques] on the detainees. In January 2003, CIA Headquarters requested that at least one other psychologist be present who was not physically participating in the administration of the CIA’s enhanced interrogation techniques. According to [Redacted] OMS, however, the problem still existed because ‘psychologist/interrogators continue to perform both functions.

’

 

[thaiboxerken]

I think the stronger position would be: "I don't care if it works, it's morally repugnant and we shouldn't do it."

Yea, but the rebuttal is "if it can save a million lives...."

 

[marplots]

Yea, but the rebuttal is "if it can save a million lives...."

Are we talking hypothetical or real life?
Because the response is going to be, "Well, get back to me when you are about to save a million lives and we can talk about it. Until then, no torture."

 

[thaiboxerken]

Are we talking hypothetical or real life?
Because the response is going to be, "Well, get back to me when you are about to save a million lives and we can talk about it. Until then, no torture."

It's always a hypothetical when it comes to defending torture, since there is no evidence that it works or has ever saved lives.

 

[RandFan]

I think the stronger position would be: "I don't care if it works, it's morally repugnant and we shouldn't do it."

Yet I defended torture because I believed it was effective. Was that a morally defensible position? No but it was the rationalization I used. People on this very thread are engaged in the same rationalization.

It's morally repugnant and there is no evidence that it works. Once you accept those two propositions you have no leg to stand on. That it "could" work is a non-starter when it comes to immorality.

 

[The Norseman]

Fair point. My evidence is that the CIA claims it works. These are experienced people who decided they needed more extreme measures to extract information...

My evidence is not just that the CIA claims torture actually produced useful intel in this case, but also that very experienced people at the CIA believed that it would before the program even started (hence the push to get the proper authorization). The CIA had had decades of institutional experience with this stuff. They had done all kinds of experiments and tests, most of which have never seen the light of day... ...I have little doubt that many people at the CIA honestly believed that it was an effective way to extract information from an uncooperative and resilient detainee...

There's more like this but I decided this is representational enough for my purposes.

The literally exact same thing is said for polygraphy, a now well-known junk science. The FBI, CIA, etc., all claim it works and continue to use it every day of the year. But, when evidence is gathered to try and support the claims, it fails on every count. The science doesn't lie. The conclusions don't lie. But the CIA does.

 

[jimbob]

Yet I defended torture because I believed it was effective. Was that a morally defensible position? No but it was the rationalization I used. People on this very thread are engaged in the same rationalization.

It's morally repugnant and there is no evidence that it works. Once you accept those two propositions you have no leg to stand on. That it "could" work is a non-starter when it comes to immorality.

It's stronger than that: not only is there no evidence that it works; there's plenty of evidence that it is counterproductive

 

[Upchurch]

There's more like this but I decided this is representational enough for my purposes.

The literally exact same thing is said for polygraphy, a now well-known junk science. The FBI, CIA, etc., all claim it works and continue to use it every day of the year. But, when evidence is gathered to try and support the claims, it fails on every count. The science doesn't lie. The conclusions don't lie. But the CIA does.

Not to mention, as I did a few posts back, that the CIA employed and trained psychics fromantic the early 70s to the mid-80's or 90s as part of Project Stargate. That the CIA uses it or says it works is no proof that it does.

 

[sunmaster14]

There's more like this but I decided this is representational enough for my purposes.

The literally exact same thing is said for polygraphy, a now well-known junk science. The FBI, CIA, etc., all claim it works and continue to use it every day of the year. But, when evidence is gathered to try and support the claims, it fails on every count. The science doesn't lie. The conclusions don't lie. But the CIA does.

Having been subjected to polygraph tests many times in my life, I'm of the opinion that various government agencies use it as an employee screening tool not because they believe it works, but because they believe that most potential employees believe it works, and it encourages them to be more forthcoming about uncomfortable and embarrassing things in their past (it worked for me, at any rate).

That said, your point about government institutions clinging to scientifically unsound practices is well-taken.

 

[sunmaster14]

Not to mention, as I did a few posts back, that the CIA employed and trained psychics fromantic the early 70s to the mid-80's or 90s as part of Project Stargate. That the CIA uses it or says it works is no proof that it does.

In reading the wiki page on this project, it seems the government was reasonably logical and systematic about the whole thing. Admittedly, it probably went on longer and cost more money that it should have, but that's kind of par for the course:

In 1995, the defense appropriations bill directed that the program be transferred from DIA to CIA oversight. The CIA commissioned a report by American Institutes for Research that found that remote viewing had not been proved to work by a psychic mechanism, and said it had not been used operationally. The CIA subsequently cancelled and declassified the program.

ETA: Also, don't conflate evidence with proof.