Does torture work? Trump says yes, but science says no

More than 80 per cent endorse torture if they are personally close to the victims

Torture has been widely used over thousands of years of human history and, despite being officially banned on moral grounds, torture can be sanctioned to this day even in highly developed countries. Systematic torture was employed in CIA detention centres under president George Bush's administration 2002 to 2008. And president Donald Trump now says he is open to bringing back torture to extract information from terrorists because "people at the highest level of intelligence" have told him torture works.

However, the evidence in this area, reviewed by Raj Persaud MD and Peter Bruggen MD (Psychology Today, January 26th, 2017) contradicts what Donald Trump was told. Torture does not work.

The main problem with using torture as a means of obtaining reliable information is illustrated by the following true story. In 1630, as the European witch hunts raged, the German Duke of Brunswick asked two Jesuits to examine how the Inquisition used torture to get information from accused witches. The Jesuits reported that all was fine. “The Inquisition only arrests people implicated by the confessions of other witches.”

The sceptical Duke then brought the two Jesuits to witness a woman being stretched on a rack. He addressed the woman: “You are a confessed witch. I suspect these two men of being warlocks. What do you say? Another turn of the rack executioner.” The woman screamed in pain and then gasped out: “You are quite right. I have seen them at the Sabbat. They can turn themselves into goats, wolves and other animals. Several witches have had children by them.”


The Duke turned to the dumbstruck Jesuits and asked: “Shall I torture you now until you confess?” One of the Jesuits, Friedrich von Spee, was so impressed that he became an influential opponent of witch trials.

Reliability of information is of the upmost importance as false information can lead to disastrous consequences, the Iraq war for example. Torture is very effective at getting people to talk but very poor at eliciting accurate information. In the words of one of Saddam Hussein's former torturers: "You can always make someone talk . . . The problem is what they say."

A survey of interrogation methods used by the FBI, active duty military, state police, and others was carried out by psychologists Allison Redlich and others (Applied Cognitive Psychology, Volume 28, Issue 6, 2014). The interrogators reported that rapport and relationship-building techniques were the most effective regardless of the interrogation goal. Confrontation techniques were the least effective and least often used.

Another survey of law-enforcement interrogators of high-value detainees across five countries carried out by Jane Goodbody-Delahunty and others (Applied Cognitive Psychology, Volume 28, Issue 6, 2014) concluded that coercive interrogation techniques have no impact on information disclosure but rapport-building efforts (showing concern, affection, using humour, demonstrating kindness and respect, making accommodation comfortable, etc) do elicit disclosure, often quite early in the interrogation process.

There are basic psychological/biological reasons why torture doesn't work, as explained in a recent book (Why Torture Doesn't Work, Harvard University Press, 2015) by Shane O'Mara, Professor of Experimental Brain Research at TCD. Professor Mara outlines how thinking, memory and mood are negatively affected by fear, starvation, thirst, sleep deprivation, immersion in freezing water, and various other commonly used torture techniques. Torture sufferers predictably produce deeply unreliable information.

But what about the ‘ticking time-bomb predicament?’ In this scenario a massive bomb is set to explode in a densely populated city and cause catastrophic loss of innocent lives. The terrorist who planted the bomb has been captured but won’t talk. To have any chance of saving lives the bomb must be neutralised very quickly and there is no time to establish rapport with the terrorist. Is torture permissible in this case?

People were asked this question in a study by Shannon Pouck and Lucian Conway (Journal of Applied Security Research, Vol. 8, 2013). Relatively fewer people would support torture if the innocent lives at risk were strangers than if they were close loved ones. About one third of the public support torture against terrorists but when participants in the study felt personally close to the victims, more than 80 per cent of them endorsed torture, liberals and conservatives alike. How would you deal with the terrorist in the "ticking time-bomb" scenario?

William Reville is an Emeritus Professor of Biochemistry at UCC