I don’t think I would be going out on too much of a limb if I observed that most of us find the idea of torture to be repellent. As social creatures it is natural for us to shy away from inflicting harm on our fellows, empathy stays our hand when ruthlessness might otherwise help us achieve our personal goals. This holds true both in normal social interactions and in the extreme situations of conflict. It requires a significant expenditure of effort for us to overcome our inbuilt aversion to causing pain.

Even so torture can be rationalised as being for the “greater good” such as in the hypothetical situation which illustrates when torture might be a reasonable recourse known as the “ticking time bomb” scenario. In this case it is supposed that a terrorist has planted a bomb in a populated area, the terrorist is captured but the authorities have not been able to ascertain the location of the bomb which might go off at any time killing hundreds or possibly thousands of innocent people. If the terrorist refuses to co-operate, so the logic goes, it is then not only permissible to resort to torture to extract the information it might even be obligatory in order to prevent greater harm.

This situation however contains a hidden assumption that if untrue leaves the entire argument moot, namely, torture leads to accurate information. If this assumption is false then the justification falls apart. Last week in the journal Trends in Cognitive Sciences Shane O’Mara argued exactly this point. The main thrust of O’Mara’s argument asserts that prolonged stress causes changes in the brain that make it difficult for the subject to accurately recall memories and make the implantation of false memories and confabulation more likely.

False memories can be laid down when a subject incorporates what they are told into their own memory as if it actually happened when it did not. Simple repetition of information can induce false memories in normal subjects, assuming that it does not also happen in stressful situations is not  particularly reasonable.

Confabulation is the involuntary creation of false memories in response to questioning when the frontal lobe of the brain is impaired. The subject believes what they are saying and so is not lying but the information may not be accurate in any meaningful way. Events from the subject’s past and imagination can be jumbled together without the ability to tell the difference between them, when or if they happened. As prolonged stress can have negative effects on the frontal lobe confabulation could be a real danger in torture situations.

In addition the subject will also be conditioned that while they are talking the extreme interrogation techniques are stopped, thus talking represents safety. There is no extra inducement to truthful speech. In this situation there is no immediate method of ascertaining the truth of the subject’s words and so as long as the subject continues to talk further interrogation is not necessary.

Given these objections and certainly in the absence of reliable data to refute these points the justification for the use of torture simply dissipates. I for one find this information very reassuring.