Hayden’s Straw Man Argument on “Interrogation Deniers”

by Roger Koppl

In a Wall Street Journal op ed of 2 June 2011, General Michael Hayden, director of the CIA from 2006 to 2009, compares “interrogation deniers” to “birthers” and “truthers.”  Hayden’s op ed mischaracterizes the basic claim of those who say torture is not effective, substitutes insult for argument, and includes a non sequitur worthy of the old joke that that “military intelligence” is an oxymoron.

Hayden defines “interrogation deniers” as “individuals who hold that the enhanced interrogation techniques used against CIA detainees have never yielded useful intelligence.”  Talk about a straw man!  I suppose there must be some “interrogation deniers” as defined by Hayden, and I suppose some of them are out there floating in the wide waters of the Internet, waiting for someone to cut and paste.  But I don’t know of any examples, and their possible existence at the margins of public discourse has no bearing on the public question.   As Glenn Greenwald noted on May 4th,  “Nobody has ever argued that brutality will never produce truthful answers.”  No.  “[T]he point has always been — as a consensus of interrogations professionals has repeatedly said — that there are far more effective ways to extract the truth from someone than by torturing it out of them.”

The scholarly literature on torture tends to support the skeptics.  The arguments showing that torture is ineffective apply to “enhanced interrogation” too, if there is such a thing apart from torture.  Darius Rejali is probably the leading scholar on the inefficacy of torture. His important study Torture and Democracy shows that non-coercive interrogation generally yields more actionable intelligence and that the historical record does not support the idea that torture is a reliable technique of intelligence gathering.

I have argued here  and here that torture is ineffective in most empirically relevant cases. Torture does not “work” because the torturers cannot make a credible promise to stop when the truth is finally told.  This logic is not universal.  In chapter five of the Invisible Hook, for example, Peter Leeson has argued that pirates raiding a ship were sometimes able to torture their victims into revealing where at least some of the ship’s treasure is hidden.  In this case, however, the veracity of the report could be quickly checked and they wanted to get away once they had their booty, making further torture costly.  But in modern bureaucratic torture, intelligence gathered may take a long time to confirm or disconfirm and the marginal cost of further torture is essentially zero.

Hayden’s main evidence that torture “enhanced interrogation” works is the testimony of four CIA directors and “Mr. Obama’s chief counterterrorism adviser John Brennan’s statement that there’s been ‘a lot of information that has come out from these interrogation procedures that the agency has in fact used against the real hard-core terrorists.’”  Plus, he says, “when I was first briefed in 2007 about the brightening prospect of pursuing bin Laden through his courier network, a crucial component of the briefing was information provided by three CIA detainees, all of whom had been subjected to some form of enhanced interrogation.”  His evidence, then, is that the people responsible for it say it worked just fine.  Such evidence is self-serving, biased, and unreliable.  In a 2002 memo, the military’s Joint Personnel Recovery Agency warned that the use of “physical/psychological coercion in interrogation” yields unreliable information.  (See here and here.)

Rather than engaging the arguments that torture is less effective than non-coercive interrogation, Hayden resorts to insult.  He calls torture skeptics “interrogation deniers,” compares them to “birthers” and “truthers,” and labels their opinion “preposterous.”  As Rousseau is reputed to have said, “Insults are the arguments employed by those who are in the wrong.”

Hayden says that if “deniers” are “true to their faith” they should “insist” that the CIA “identify all the information derived directly or indirectly from enhanced interrogation. And then they should insist the agency destroy it.” If torture “enhanced interrogation” is unreliable, we should view any intelligence produced by it skeptically.  But it is a non sequitur to say that such skepticism means you want the CIA to throw away information.  Hayden leverages this non sequitur to suggest that the “deniers” are hypocrites.  Since no one is calling for the CIA to discard information, the “deniers” may “quietly concede to themselves that facts really do matter.”  Deep down, he suggests, even “interrogation deniers” know he is right.   Well not me.  Deep down, I think he is all wrong about torture “enhanced interrogation,” and I am not swayed by straw man arguments, insults, and non sequiturs.

17 thoughts on “Hayden’s Straw Man Argument on “Interrogation Deniers”

  1. Roger has done an excellent job sticking to the empirical issues. I admire his restraint.

    Torture is repugnant. There is a reason that coerced testimony is inadmissable in court.

    For those needing a refresher on the topic, I recommend watching “The Battle of Algiers.” The brutality not only destroys those being tortured, but those performing the torture.

  2. Excellent post. You make many good points and clearly the potential for the abuse of state power here is huge and very disturbing. However, I honestly struggle with this issue, as I suspect may do.

    It is unquestionably the case that torture is repugnant but the difficulty I suppose is that, for many people, their moral intuition is that a failure to do everything possible to save innocent lives from evil plots can also be repugnant, particularly where the failure may be motivated by moral vanity or moral cowardice. I suppose that the ethical situation may also be substantially more complicated for consequentialists than for natural rights libertarians.

    Greenwald states:

    “Nobody has ever argued that brutality will never produce truthful answers. It is sometimes the case that if you torture someone long and mercilessly enough, they will tell you something you want to know. Nobody has ever denied that. In terms of the tactical aspect of the torture debate, the point has always been — as a consensus of interrogations professionals has repeatedly said — that there are far more effective ways to extract the truth from someone than by torturing it out of them.” (emphasis mine)

    I understand the game theoretic arguments regarding the effectiveness of torture, but that is not the argument he is making in the quote. I am honestly not sure what he means when he says that “there are far more effective ways to extract the truth”, particularly if one is going to adopt an empirical or consequentialist approach. Does he mean that there are means to extract the truth that are faster?

    Also, as someone who has tried hard to understand this issue, I am uncomfortable with the equation of “enhanced interrogation techniques” (whatever that means) with “torture”. As I recall from reading that I had done several years ago, my impression is that not all enhanced techniques would fall within the legal definition of torture (as defined by international treaties). Am I wrong there? If I am not, one is then talking about various degrees of discomfort. Now, subjecting someone to discomfort short of torture without justification can certainly be morally objectionable but perhaps the ethical assessment can be influenced by context. I am wondering in particular whether an argument can be made, in, for example, the tiresome “ticking time bomb” scenario, that enhanced interrogation techniques short of torture are a legitimate form of self-defense, i.e., they are made necessary only by one’s opponents’ previous aggressive (or, in this case, planned aggression).

    Regarding the effectiveness of “enhanced“ techniques in these situations, I note that the 2002 memo does not state that any use of “physical/psychological coercion in interrogation” yields unreliable information. It states that:

    “The application of extreme physical and/or psychological duress (torture) has some serious operational deficits, most notably, the potential to result in unreliable information. This is not to say that the manipulation of the subject’s environment in an effort to dislocate their expectations and induce emotional responses is not effective. “ (original emphasis)

  3. Suppose a person is tortured and in the course of it says both true and untrue things. Can the “interrogator” distinguish ex ante? And with what kind of reliability?

    Now suppose a certain situation (eg, the whereabouts of bin Laden) is resolved by information coming from various sources. It would then be easy to go back to some tortured fellow’s statements and cherry-pick those that are consistent with what we now know ex post.

    This last phenomenon does not mean that the information arising out of torture is useful ex ante. Its ex ante reliability (probability of being true) could be very low.

    So one has to be careful about all statements regarding the usefulness of information gathered by torture.

  4. @Mario. But if the process of establishing the value of information (or the likelihood of its being true) is one of corroboration and the information extracted under some form of duress (torture or discomfort short of torture) is one of the corroborative elements, does the information not have value despite the fact that, in isolation, it is not possible to determine the value of the information ex ante?

    BTW, in the quote at the end of my comment above, the word “extreme” (in the first line) should be underlined (as it was in the original text).

  5. While Roger’s post is eloquent and well-reasoned, I don’t find it dispositive. True, the state can easily concoct justifications for the use of torture, and can find convenient definitions of “enhanced interrogation techniques” in order to excuse unprincipled action. The presumption should always be against coercion and inhumane treatment of any person.

    On the other hand, is the issue that black and white? Is water-boarding, for instance, as (to our knowledge) applied recently by the US, against individuals who clearly do not fit the historical legal definition of an enemy soldier, and who intentionally target innocents, necessarily categorically wrong? The danger of investing such power in the state is clear, but it seems to me it can and should be weighed against these other considerations.

    Individual liberty is a value, and I believe one of the first order, but it is not the only one. As only one example, would that more of those who so value individual liberty should recognize the individual right to life of a person not yet born, and find the moral indignation to argue as forcefully against legal abortion as against torture by the state. I don’t want to equate apples and oranges. Far from it–the objective evil of abortion exceeds that of the methods being defended by General Hayden by orders of magnitude.

  6. Some of us do not believe that the fetus is a person. Hence abortion is not an aggression against persons within the ordinary meaning of the word. Torture is.

  7. Hmm. Here’s a question: who’s less deserving of our aggression – an innocent and (temporarily) potential person or a not so innocent terrorist?

  8. David:

    You offer some really thoughtful remarks. Thanks for that. I’ll try to address them in order.

    On the ethics of the issue, I am personally a consequentialist. Thus, the efficacy of torture is among the empirical questions that matter for me in deciding if it is ever okay to torture. In the “tiresome ‘ticking time bomb’ scenario,” as you so eloquently put it, we usually assume away uncertainties and indirect consequences that exist and matter in reality. If we were really sure we had our bomber, whom we knew for sure to be a really bad person, and we could be sure torture would in fact generate the life saving information without creating an increased probability of the state using torture inappropriately in the future, and so on, THEN, the moral calculus would favor torture. But the “if” part is not empirically plausible. In reality, we have uncertainty about who the guilty persons might be, what they know, and so on. Even when there is no doubt about whether you’ve got the bad guy (as seems to be the case with Khalid Sheikh Mohammed) we would have reason to doubt that torture would generate reliable information, we would have reason to fear the precedent set by torturing him, and so on. We’re into a big topic, but my comments in this paragraph may suggest the sort of logic that leaves me opposed to torture and to any coercive technique that might be considered to fall short of torture, whether we call is “enhanced interrogation” or something else.

    When Greenwald says, “there are far more effective ways to extract the truth”, I think he just saying that non-coercive techniques are more effective overall. That’s what I took him to say, and I tend to think the overall context of his remark supports that reading.

    I tried to write my post in such a way that my points hold up no matter how (or whether!) we distinguish between “torture” and “enhanced interrogation.” Then you ask whether it might be “a legitimate form of self-defense” to use “enhanced interrogation techniques” in the ticking time bomb scenario. As you can see from my earlier remarks in this comment, I would agree, but only under the unrealistic hypothetical that we definitely have the right person and so on. Anyway, my original post steered clear of the ethical issues as Jerry O’Driscoll notes.

    Finally, you quote the JPRA memo’s conclusion which seems to concede that coercive techniques can work. But the language there is vague, I think, in contrast to the rather clear statements of the rest of the memo. I tend to think it’s a CYA add-on. However that may be, the memo tends to show that there is and has been less unanimity within the about coercive interrogation than Hayden’s op ed seems to suggest. That lack of unanimity, in turn, suggests that his attempt to paint “deniers” as zany is not really right.

  9. Arg. In my last paragraph it should say “less agreement within the *military* about coercive interrogation.

  10. James:

    I think your comment addresses the question of whether torture is necessarily ruled out morally or might under some conditions be morally justified. Your comment is about values. My original post, however, was about the efficacy of coercive interrogation, not its morality. I happen to think the two issues are linked, as my answer to David reveals. But I’m not sure your comment really connects to my original post.

  11. Mario: “Some of us do not believe that the fetus is a person. Hence abortion is not an aggression against persons within the ordinary meaning of the word.”

    So, if someone does not believe Muslims are people, are they good to go with torturing them?

  12. Gene,

    It is not a matter of what people believe; it is a matter of what is the case. Muslims are persons; fetuses are not. But this is not really on topic, although I responded to the person who brought it up. END.

  13. There is another point, surely, that should be made. What if the person being interrogated is actually innocent and knows nothing?

    How much torture (or enhanced interrogation) techniques are “enough” to decide that he, in fact, knows nothing?

    What if the interrogator concludes that the fact that he hasn’t “revealed” anything demonstrates just what a fanatic he must be, so more determined “methods” must be applied to get the”truth” out of him?

    Or what if he “talks,” but he just makes up things to get them to stop the torture? And when they discover that what he has told is not true, they decide, again, that this shows how fanatical and dangerous the fellow “really” is? So more torture is applied?

    Until when? He is dead? Permanently harmed, physically? Driven into madness?

    And how many innocent men are worth doing this to, in the name of, maybe, getting “valuable” information from an actually guilty party?

    This is a Pandora’s Box that opens back into the Inquisition, and the attempt to find out who “really” is a witch or an agent of the Devil, or a heretic.

    One of the great achievements of 19th century (classical) liberalism was its triumph in fostering conceptions of the “rules of war,” the treatment of prisoners, and respect for the life and property of non-combatants.

    One of these advancements was formal limits on the use of physical force on prisoners, just as in domestic criminal matters torture was increasingly banished from court proceedings as a bases of evidence.

    Or course, especially in war times, it was, no doubt, violated and not respected to the “letter of the law.” But the principle was recognized, and use of such methods — even in the heat of conflict — was not condoned and was considered illegal and a breach of proper conduct.

    We should see the return to torture methods as another example of what Veale was chronicling in his book on modern warfare, an “Advance to Barbarism.”

    Richard Ebeling

  14. Right on, Richard. It seems to me that the American defenders of torture (or supposed “harsh” techniques coming shy of torture) tend to adopt the posture that they really *know* they got the bad guys. Maybe that’s fair in one, two, or three rather special cases. But once you let the genie out of the bottle, it’s hard to get it back in. For example, at least one report claims that “American soldiers were copying” the torture “tactics” of the TV show “24 Hours.”


    This report is one very small piece of evidence tending to support the view that the practice of torture (or near-torture) tends to spread once a precedent is set. Abu Graib would seem to provide further evidence.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s