False moral dilemma on torture

It was put forth during the Bush administration as the US pursued terrorist intelligence-gathering that torture was necessary in order to gather information to thwart terrorist plots and thereby save lives. Some objected that it is morally reprehensible to torture, which sets up an interesting moral dilemma: is the suffering of one worth saving many lives. A Utilitarian ethicist would respond yes (since the suffering of the many victims is clearly outweighed by the suffering of one or a few), many other ethical schools of thought would respond no, and this would lead to an interesting ethical debate. Alas, such a debate rests on a false premise, namely that torture is actually an effective and reliable way to extract information, and thus there really is no dilemma.

A survey of movies, tv shows and books can provide ample examples of torture used in many different ways to extract reliable information. The hero has capture some minion of the antagonist, the minion is reluctant, the hero applies pain, and suddenly the minion is talking. This genuinely realistic scenario can lead us to think that pain makes people talk and if the person is still reluctant under minor torture, then we just need to apply more torture and yet more torture and then they'll talk. This is the foundation for a longstanding view, which made torture in Roman times and the Middle Ages an instrumental part of jurisprudence. Torture was considered to make people unable to lie. Thus, confessions extracted under torture were actually considered more reliable by courts. This is a good example of what is called "naive psychology," namely the understanding of human psychology that the average person develops through experience. Naive psychology is generally accurate in most normal situations, but for less common situations our naive psychology becomes generally inadequate and sometimes grossly mistaken.

In this case, though our naive psychology can lead us to an inaccurate view, nonetheless, I think our naive psychology is sufficient to help us understand why torture doesn't work, simply if we reflect upon it a bit more. Let's imagine three scenarios. Scenario one: a person active in terrorist organizations bearing considerable information is captured by US officials, he is at first reluctant to give information, they apply torture, he starts confessing what he knows. Scenario two: another person active in terrorist organizations bearing considerable information is captured, he too is reluctant to give information but is persuaded under torture and tells them all he knows, unfortunately officials think he is either lying or is hiding something, they apply more torture, now he starts making stuff up just to get them to stop, and he starts telling them the names of real but innocent people. Scenario three: a person with little or no connections to any terrorist organizations is captured and is completely honest and forthright about knowing nothing, officials think he is hiding something, they torture him, he finally cracks and starts making things up to get them to stop and he again starts telling them the names of real but innocent people. Of these three scenarios only scenario one actually provided accurate information, and scenario one can easily slip into scenario two, and scenario two leads to innocent people being captured, which creates scenario three, which is probably the most common. This is the type of thing that happened with the many centuries of anti-witch efforts in Christian Europe. People would be suspected of witchcraft and consorting with the devil. They would be captured, tortured, and ultimately made to confess. And then they would be forced to name others. Since, of course they weren't a witch and didn't consort with the devil and thereby knew no other witches, they would provide the names of other innocent people. And the cycle would continue. And I can guarantee that all three of these scenarios still do happen in our anti-terrorist efforts today.

Torture is a good way of extracting confessions. You can probably get a person to confess to just about anything under torture. You can get someone to confess to having kidnapped the Lindberg baby, assassinated Kennedy. sold nuclear secrets to Russia and brought Hussein to power if you just tried hard enough. But if you're intending to get reliable information for use in intelligence gathering, torture is very unreliable. You don't know whether the person is being honest or is simply saying things to get the torture to stop. Vanity Fair had an article last month about how unreliable our information-gather through torture has been.

Generally the most effective way to get information from people is to earn their trust. In the case of terrorist suspects, that would mean putting moles in their prison cells: people from their native country who will get friendly with them and get them to start telling their secrets. We humans don't really like keeping secrets and love having an outlet for information that's bottled up. This requires a bit more subtlety and a bit more patience. Torture is a way of getting quick results, which is why it is more attractive to the military and the government. With torture they can point to better productivity--namely more information and more confessions extracted--even if most of the information is junk.

There is very little morally objectionable about using moles to get to people to confess, and its more reliable. Thus, as it turns out in this situation, the most reliable method and the most ethical method are one and the same. No dilemma here.

Update: Julian Sanchez over at Cato lays down a number of the unseen disadvantages to using torture for interrogation, bolstering the case that torture really isn't a pragmatically sound option.