The last word

We have ways of making you talk

Laurence and Emily Alison: spent hours listening to terrorists talk
Hannibal Lecter: a controlling interviewee

“If you find yourself having a go at someone, ask yourself: ‘What am I achieving by this?’ Because they will stop talking to you”

Expert interrogators know that torture doesn’t work – but until now, says Ian Leslie, nobody could prove it. By analysing hundreds of top-secret interviews with terror suspects, two scientists have revolutionised the art of extracting the truth


In 2013, a British man was arrested for planning to kidnap and brutally murder a soldier. The suspect had posted messages on social media in support of violent jihad. In a search of his residence, the police had found a bag containing a hammer, a kitchen knife and a map with the location of a nearby army barracks. Shortly after his arrest, the suspect was interviewed by a counter-terrorist police officer. The interviewer wanted him to provide an account of his plan, and to reveal with whom, if anyone, he had been conspiring. But the detainee – we will call him Diola – refused to divulge any information. Instead, he expounded grandiloquently on the evils of the British state for 42 minutes. When the interviewer attempted questions, Diola responded with scornful, finger-jabbing accusations of ignorance, naivety and moral weakness: “You don’t know how corrupt your own government is – and if you don’t care, then a curse upon you.”

Watching a video of this encounter, it is just possible to discern Diola’s desire, beneath his ranting, to tell what he knows. In front of him, a copy of the Koran lies open. He says he was acting for the good of the British people, and that he is willing to talk to the police because, as a man of God, he wants to prevent future atrocities. But he will not answer questions until he is sure that his questioner cares about Britain as much as he does: “The purpose of the interview is not to go through your little checklist so you can get a pat on the head. If I find you are a jobsworth, we are done talking, so be sincere.” The interviewer, who has remained heroically calm in the face of Diola’s verbal barrage, is not able to move the encounter out of stalemate, and eventually his bosses replace him.

When the new interviewer takes a seat, Diola resumes his inquisitorial stance. “Why are you asking me these questions?” he says. “Think carefully about your reasons.” The interviewer does not answer directly, but something about his opening speech triggers a change in Diola’s demeanour. “On the day we arrested you,” he begins, “I believe that you had the intention of killing a British soldier or police officer. I don’t know the details of what happened, or why you may have felt it needed to happen. Only you know these things, Diola. If you are willing, you’ll tell me, and if you’re not, you won’t. I can’t force you to tell me – I don’t want to force you. I’d like you to help me understand. Would you tell me about what happened?” The interviewer opens up his notebook, and shows Diola the empty pages. “You see? I don’t even have a list of questions.” “That is beautiful,” Diola says. “Because you have treated me with consideration and respect, yes I will tell you now.”

For years, any debate over what constitutes effective interrogation has been dominated by a pervasive folk belief in coercion. From NYPD Blue to 24 and Zero Dark Thirty, we are trained in the idea that interrogators get the job done by intimidating, demoralising and, when necessary, brutalising their subjects. Steven Kleinman, a former army colonel and one of the US military’s most experienced interrogators, told me it is not just the public that is influenced by popular narratives: “Politicians, policy-makers, senior military officers – people who have never conducted interrogations are somehow just convinced they know what works.”

In 2003, Kleinman tried to stop his fellow soldiers from conducting abusive interrogations of Iraqi insurgents; he later became the first military officer to speak out against such practices. He did so not just because he thought they were wrong, but because he thought they were stupid. Kleinman believes that coercion is counterproductive, because it destroys the trust that underpins a successful interview. But conventional wisdom in military and law enforcement circles has been very hard to shift. This is because it is difficult to prove what works. High-stakes interrogations take place in secret, and have rarely been available to objective researchers. A body of scientific literature supports Kleinman’s view, but most of it is based on laboratory experiments, in which students are asked to pretend they have just robbed a bank and interrogators are asked to believe them. These are easily dismissed by practitioners as academic game-playing.

Now, two British researchers are quietly revolutionising the study and practice of interrogation. Earlier this year, in a meeting room at the University of Liverpool, I watched a video of the Diola interview alongside Laurence Alison, the university’s chair of forensic psychology, and Emily Alison, a professional counsellor. The Alisons, husband and wife, have done something no scholars of interrogation have been able to do before. Working with the police, who allowed them access to around 1,000 hours of tapes, they have observed and analysed hundreds of real-world interviews with terrorists suspected of serious crimes. No researcher in the world has ever laid their hands on such a haul of data before. Based on this research, they have constructed an unprecedentedly authoritative account of what works and what does not, rooted in a profound understanding of human relations.

Pausing the Diola video, Emily grimaced. “I call this one ‘the Hannibal Lecter interview’,” she said. “He wants a piece of the interviewer. When I watched this tape the first time I had to switch it off and walk away. I was so outraged, my heart was pounding. Of course, if you’re in the room, it’s 1,000 times worse.” Laurence nodded. “As the interviewer, you’re bound to have an emotional response,” he said. “What you want to say is, ‘You’re the one in the fucking seat, not me.’ He’s trying to control you, so you try and control him. But then it escalates.” The moment that an interrogation turns into an argument, it fails. “You need to remember what your purpose is,” said Emily. “You’re seeking information. You’re not there to speak on behalf of the victims or the police. If you find yourself having a go at someone, ask yourself: ‘What am I achieving by this?’ Because they will stop talking to you.”

Emily met Laurence at the University of Liverpool in 1996, shortly after arriving in the UK from her home in Wisconsin. She had applied to join the Madison police force, which she saw as a stepping stone to the FBI, but opted at the last minute to take a Master’s in “investi-gative psychology” – the application of psychology to police work (Liverpool was then one of the few institutions in the world to offer it). “This wasn’t long after The Silence of the Lambs,” said Emily. “I wanted to be the new Clarice Starling.” Laurence was a PhD student in the department of forensic psychology, and already a rising star. In 1998, Emily and Laurence got married. Laurence continued his academic career. Emily joined the Cheshire probation service, and later started a consultancy, helping social workers counsel families afflicted by domestic abuse. Alongside their day jobs, the Alisons started helping the police with criminal cases. “It might be, can you help us with this rape in Bath, or a murder in West Mercia,” Laurence said. The police often wanted to know the best way to interview a suspect or witness, usually after an initial attempt had gone badly. Laurence would ask for assistance from Emily, who knew a lot about interviewing difficult people, thanks to her background in counselling. The Alisons would read transcripts, watch footage and sometimes monitor interviews, searching for a way to get the interviewee to open up.

In 2010, Laurence was contacted by a US government agency that was commissioning research into interrogation. The High-Value Detainee Interrogation Group, or HIG, was set up in 2009 by President Obama, who was keen to signal a clean break from the Bush administration, which had sanctioned abusive interrogations in Iraq and Afghanistan. Housed within the FBI, the HIG’s purpose is to ground interrogation in science. Its chief researchers were particularly interested in Britain, whose counterterrorist police have earned a reputation for being sophisticated interviewers. In 1992, after public enquiries into two miscarriages of justice involving IRA attacks had revealed abusive interrogation practices, Parliament passed laws stipulating all interviews be recorded, and making it a right to have a solicitor present. With the option of coercion removed, the British police were forced to think harder about the best way to obtain information. In a minor but significant change, they stopped using the confrontational word “interrogation”, and replaced it with “interview”.

The HIG invited Laurence to apply to them for research funding. “I said, I don’t want to do research on students. I want to look at the real thing and extract what works.” He set his sights on an audacious goal: persuading the UK counterterrorism unit to give him access to video of its interviews with suspects. Two years and more than 100 phone calls later, in 2012 he was granted access to 181 interviews, a total of 878 hours of tape. They included Irish paramilitaries, al-Qa’eda operatives, far-right extremists, incompetent bunglers caught up in something they didn’t understand, and highly dangerous operatives.

The tapes were housed in a secure police facility and the Alisons were not permitted to move them. They took turns to visit. Each interview had to be minutely analysed, every aspect of the interaction classified and scored. They included the counter-interrogation tactics employed by the suspects, the manner in which the interviewer asked questions, the demeanour of the interviewee and the amount and quality of information yielded. Data was gathered on 150 different variables in all, and the resulting statistics analysed. The most important relationship was between “yield” – information elicited from the suspect – and “rapport” – the quality of the relationship between interviewer and interviewee. For the first time, a secure, empirical basis was established for what had, until then, been something between a hypothesis and an insider secret: rapport is the closest thing interrogators have to a truth serum.

Despite its reputation among elite practitioners, “rapport” has been vaguely defined and poorly understood. It is often conflated with simply being nice – Laurence refers to this, derisively, as the “cappuccinos and hugs” theory. In fact, he observes, interviewers can fail because they are too nice, acquiescing too quickly to the demands of a suspect, or neglecting to pursue a line of purposeful questioning at a vital moment. The best interviewers are versatile: they know when to be sympathetic and when to be direct and forthright. At the heart of the Alisons’ model is an insight from a neighbouring field – that of drug counselling. Emily, who worked as a counsellor in Wisconsin, noticed that interrogations failed or succeeded for similar reasons as therapeutic sessions. Interrogators who made an adversary out of their subject left the room empty-handed; those who made them a partner yielded information. She concluded that the detainee, like the addict, wants to feel free, despite or rather because of their confinement, and that the interviewer should help them do so. One of the Alisons’ most striking findings is that suspects are more likely to talk when the interviewer emphasises their right not to. “The more pressure you put on a person, the less likely they are to speak to you. You need to make them feel responsible for their choices,” said Laurence.

Studies of interrogation are often preoccupied with the question of how to detect deception, but even a lie is information; the hard problem for an interrogator is a suspect who says nothing. Terrorists can say nothing in a variety of ways. Irish paramilitaries were trained to focus their gaze on a spot on the wall and remain utterly silent. Some suspects give only monosyllabic answers or stick to scripted responses, or simply turn their chair around, presenting the interviewer with the back of their head. Islamic extremists, like Diola, are prone to long ideological rants. All such tactics can get under the skin of the interviewer, throwing him off his plan by goading him into anger. Skilled interrogators are adept at managing their own responses.

An interview fails when it becomes a struggle for dominance, in which the interviewee’s way of asserting himself is to tell his interviewer nothing. “In a tug of war, the harder you pull, the harder they pull,” says Laurence. “My suggestion is, let go of the rope.” I thought back to how Diola’s second interviewer had opened him up: “Only you know these things Diola. I can’t force you to tell me – I don’t want to force you. I’d like you to help me understand. Would you tell me about what happened?”

A longer version of this article first appeared in The Guardian. © Guardian News and Media Ltd.