Why torture does not work

Interrogation is far too important to be left to amateurs. Obtaining actionable and reliable intelligence can be crucial to activities ranging from everyday law enforcement to preventing acts of terror.

That is why interrogation techniques should be based on brain and behavioural sciences, not on the fevered imaginings of Hollywood producers that are believed by politicians, supported by lawyers and carried out by amateur torturers.

Torture has been with us for all of human history — even if it has not always been called by that name.

Democracies, for example, tend to use torture secretly and prefer techniques that target core psychological, neural and physiological functions.

These methods — near-drowning, suffocation, shackling or stress positions to inflict physical pain, as well as sensory assaults such as freezing temperatures, loud noises or bright lights — often leave no physical evidence.

But they — together with psychological methods, including enforced nakedness, social isolation, threats using guns, drills or attack dogs, and fabricated assaults on a victim’s loved ones — can be devastating.

As abhorrent as these methods may be, they seldom lack defenders, who argue that they are needed to obtain information that can save lives.

Extreme stress, they argue, causes the subjects to reveal what they know.

But there is no evidence that this is true. In fact, torture undermines the very goals it is supposed to achieve.

Confessions elicited through torture can be voluminous, but they are just as often nonsensical.

Consider, for example, how many women confessed under torture that they were witches, or how the mere threat of torture induced Galileo Galilei to deny the proposition that the Earth travels around the sun.

Experienced interrogators uniformly repudiate torture, knowing that it does not yield usable, verifiable or actionable intelligence.

Indeed, numerous studies of military personnel, certain patient groups and normal volunteers demonstrate that chronic and severe stressors compromise psychological functioning, causing tissue loss in brain regions supporting memory (the hippocampal formation) and decreased activity in brain regions supporting intention, planning and regulating complex behaviour (the frontal lobes).

Extreme stressors also cause increased activity in brain regions associated with processing fear and threat-related information (which can trigger post-traumatic stress disorder).

Soldiers enduring sleep deprivation as part of their training show large decrements in psychomotor and general cognitive function, as well as profound memory deficits.

Sleep deprivation also profoundly and negatively affects mood, further compromising cognitive function.

Extended periods of sleep deprivation can cause polysensory hallucinations, psychotic-like episodes, and other neuropsychiatric phenomena.

There is no evidence whatsoever that sleep deprivation in any way enhances access to memories stored in the brain.

Studies conducted with patients in chronic pain or with volunteers on whom pain is inflicted demonstrate that physical suffering impairs cognition, memory and mood.

Suffocation or near-drowning are similarly problematic techniques.

Oxygen restriction reliably draws activity away from brain regions concerned with higher cognitive function and memory towards brainstem regions concerned with reflexive responses supporting immediate survival.

This militates against truthful recall and favours confabulation.

Humans are bad at detecting lies — often doing little better than they would if they had flipped a coin.

And they are markedly worse at lie detection when under heightened emotional strain.

There is no evidence that torturers are better at lie detection than anyone else; on the contrary, there is plenty of evidence that torturers or their superiors routinely disbelieve their subjects’ testimony.

To make matters worse, torture is traumatic not just for the victim, but also for the perpetrator.

Politicians who support torture never have to waterboard, starve or physically abuse prisoners personally. But somebody has to carry out their policies, and those who do are terribly affected by it, for reasons that are deeply rooted in our brain circuitry.

Humans have a specialised brain network (the “pain matrix”) that automatically and reflexively responds to distress, pain and despair in another.

Engaging in physical and emotional assaults upon the defenceless in order to elicit worthless confessions and dubious intelligence is degrading, humiliating and traumatising.

And when these acts are carried out at the request of a democracy, those who implement them have no secret society of fellow torturers to turn to for social support or comfort.

Even if the welfare of the victims is ignored, torture is not cost-free; it damages the perpetrators, corrodes democratic institutions and corrupts the rule of law.

Fortunately, the realisation that brain and behavioural sciences should be at the core of interrogation practice and intelligence work is gaining ground.

In the US, for example, recent legislation should help ensure that the best evidence-based practices will form the basis of non-coercive interrogation.

Lives can indeed by saved when reliable and truthful information is obtained quickly. And that is exactly why torture in all its forms should be rejected. 

The writer, professor of experimental brain research and Wellcome Trust senior investigator at the Institute of Neuroscience at Trinity College, is the author of “Why Torture Doesn’t Work: The Neuroscience of Interrogation”. ©Project Syndicate, 2016. www.project-syndicate.org