Mistakes Were Made - Carol Tavris [70]
The interrogator’s presumption of guilt creates a self-fulfilling prophecy. It makes the interrogator more aggressive, which in turn makes innocent suspects behave more suspiciously. In one experiment, social psychologist Saul Kassin and his colleagues paired individuals who were either guilty or innocent of theft with interrogators who were told they were guilty or innocent. There were therefore four possible combinations of suspect and interrogator: You’re innocent and he thinks you’re innocent; you’re innocent and he thinks you’re guilty; you’re guilty and he thinks you’re innocent; or you’re guilty and he thinks you’re guilty. The deadliest combination, the one that produced the greatest degree of pressure and coercion by the interviewer, was the one that paired an interrogator convinced of a suspect’s guilt with a suspect who was actually innocent. In such circumstances, the more the suspect denied guilt, the more certain the interrogator became that the suspect was lying, and he upped the pressure accordingly.
Kassin lectures widely to detectives and police officers to show them how their techniques of interrogation can backfire. They always nod knowingly, he says, and agree with him that false confessions are to be avoided; but then they immediately add that they themselves have never coerced anyone into a false confession. “How do you know?” Kassin asked one cop. “Because I never interrogate innocent people,” he said. Kassin found that this certainty of infallibility starts at the top. “I was at an International Police Interviewing conference in Quebec, on a debate panel with Joe Buckley, president of the Reid School,” he told us. “After his presentation, someone from the audience asked whether he was concerned that innocent people might confess in response to his techniques. Son of a gun if he didn’t say it, word for word; I was so surprised at his overt display of such arrogance that I wrote down the quote and the date on which he said it: ‘No, because we don’t interrogate innocent people.’” 30
In the next phase of training, detectives learn to become confident of their ability to read the suspect’s nonverbal cues: eye contact, body language, posture, hand gestures, and vehemence of denials. If the person won’t look you in the eye, the manual explains, that’s a sign of lying. If the person slouches (or sits rigidly), those are signs of lying. If the person denies guilt, that’s a sign of lying. Yet the Reid Technique advises interrogators to “deny suspect eye contact.” Deny a suspect the direct eye contact that they themselves regard as evidence of innocence?
The Reid Technique is thus a closed loop: How do I know a suspect is guilty? Because he’s nervous and sweating (or too controlled) and because he won’t look me in the eye (and I wouldn’t let him if he wanted to). So my partners and I interrogate him for twelve hours using the Reid Technique, and he confesses. Therefore, because innocent people never confess, his confession confirms my belief that his being nervous and sweating (or too controlled), or looking me in the eye (or not) is a sign of guilt. By the logic of this system, the only error the detective can make is failing to get a confession.
The manual is written in an authoritative tone as if it were the voice of God revealing indisputable truths, but in fact it fails to teach its readers a core principle of scientific thinking: the importance of examining and ruling out other possible explanations for a person’s behavior before deciding which one is the most likely. Saul Kassin, for example, was involved in a military case in which investigators had relentlessly interrogated a defendant against whom there was no hard evidence. (Kassin believed the man to be innocent, and indeed he was acquitted.) When one of the investigators was asked why he pursued the defendant so aggressively, he said: “We gathered that he was not telling us the whole truth. Some examples of body language is that he tried to