Does Torture "Work"?

Nell Lancaster had a very good posting at her blog the other day, Torture: It's not about "intelligence gathering".

One of the most persistent and discouraging themes that crops up in discussions of torture is the question of whether it "works" or not. The people engaging this question make a fatally wrong assumption: that the goal of torturers is the same as that of legitimate interrogators -- to get reliable information useful for active, circumscribed military operations or police investigations.

But torture does something else altogether, and is designed to do so: it extracts false confessions. These confessions, along with the agony of the torture itself, serve the goals of limitless, lawless "war": to humiliate and break opponents, to divide them from supporters, to terrify those not actively in opposition into staying inactive, and, most importantly, to justify the operations of the dirty war within which torture takes place: commando raids, assassinations, spying, kidnapping, secret and/or indefinite (and unreviewable) detention, and further torture.

I think Nell makes some very good points, and they are especially applicable to the use of U.S. torture during the period we have lived and still living through, beginning with the large-scale revival of the U.S. torture program after 9/11.

However, I thought there was more to say about the issue of "false confessions," and on the issue of the so-called efficacy of torture in general. What follows is my comment to Nell's posting at A Lovely Promise (her blog) (links within have been added):

Re the torture argument.

The U.S. government spent serious money and decades thinking about and experimenting upon torture and other forms of controlling human behavior.

I think the issue is falsely separated into orthogonal realms where one supposedly tortures to gain information, OR one tortures to terrorize or gain control (this would include the idea of eliciting false confessions, as well).

It would be wrong to suppose that torture does not sometimes occur as an attempt to gain information. I worked in therapy with a former Central American insurgent who was captured, and then tortured to reveal the names of his comrades. The poor fellow did reveal names under torture, and suffered tremendous guilt as a result (and hence had come to see me).

I have also worked with torture survivors who clearly were tortured as a matter of social control and terror, and had no identifiable information or connection that could feasibly make them a possible source of intelligence. (I remember one case particularly well of a man from Egypt.)

I also have worked with some who were tortured and coerced to make false confessions.

I think that like all human behavioral and psychosocial phenomenon, the desire to isolate motivations into identifiable causal factors betrays our understanding of the situation.

The human psyche is internally divided, determining reality based on a complex set of assumptions, identifications with others (or with entities or causes), and a large retinue of defensive mental maneuvers to ward off all kinds of anxiety, including the anxiety of not knowing or not belonging.

The result is a psycho-social-emotional stew of motivational factors that defies any easy kind of categorization, and that is what I believe we have when we look at the purely human phenomena of torture.

It gets even more complicated when we look at the question of "false confessions." The latter can often be a mixture of fact and fiction. They can also be rendered for use in very complicated counter-intelligence schemes, so that it's not clear what kind of information was desired or not, and by whom.

The classic example of the latter is the paradigm case of the U.S. torture experience: the "confessions" by U.S. Air Force officers captured by the Chinese and North Koreans, who confessed to operating aircraft used in a secret U.S. program to use biological weapons during the Korean War.

Were these confessions true or false? If you believe torture always produces garbage, then it must be false. But empirically, it is not true that torture produces only bad intel. When CIA torturers have discussed their results publicly, as in Biderman's The Manipulation of Human Behavior, they make it clear that best results for accuracy happen in a thin band between normal interrogation and resistance and overt brutality or overuse of psychological techniques which collapse the mind of the victim.

In the case of the captured airmen, the U.S. denied any such use of bio weapons, despite the findings of an independent commission to study the issue. Only much later, in the early 21st century, have some legitimate historical examinations found that there may have been some truth in the airmen's confessions. (See Endicott and Hagerman, The U.S. and Biological Warfare.)

To conclude, it is not a question of the efficacy of torture to provide information, as you ably point out. The motives for torture, however, are complex, interconnected and over-determined, so that every instance of its use must be looked at in its cultural-political-historical context to see to what degree one causal aspect played a more or less significant role as against a number of other possible causes.

A final thought: by entering into the argument as to whether or not torture "works", we move farther away from the primary point, which is that torture is unacceptable and illegal no matter what the reason or the cause. Period.

Otherwise, it would be as if we were still arguing about cannibalism, with one group arguing whether or not it really provided nutrition or not.

For further thoughts, one can see an earlier essay I wrote on this topic,
Some Thoughts on Utilitarian Arguments Against Torture