Confessions Not Always Clad in Iron

Sen. Larry Craig (R-Idaho) is trying to clear his name after confessing to a misdemeanor that he tried to solicit sex. Though not specific to his case, research shows that innocent people do falsely confess to crimes.
Sen. Larry Craig (R-Idaho) is trying to clear his name after confessing to a misdemeanor that he tried to solicit sex. Though not specific to his case, research shows that innocent people do falsely confess to crimes. (By Matt Cilley -- Associated Press)
By Shankar Vedantam
Monday, October 1, 2007

In the courts and in Congress, Sen. Larry Craig is fighting to withdraw his guilty plea to a misdemeanor charge that may suggest he tried to solicit sex from a man in June at a Minneapolis airport bathroom. Rather than resign yesterday, as the senator had promised and Republicans had hoped, Craig said he was going to stick around and try to clear his name.

In the courts of public opinion and on the late-night comedy circuit, Craig was convicted a long time ago. It boiled down to one simple thing: He confessed.

A decade of psychological experiments and observations, however, suggests that Craig's confession should not be seen as iron clad. While the senator's claims of innocence have been widely mocked, a large body of psychological research shows that innocent people do falsely confess to crimes. Some even believe they are guilty.

The psychological research says nothing about Craig's case in particular. Experienced interrogators might be right when they say that Craig's actions have all the marks of a guilty person. But what the psychological research conclusively undermines is the widespread notion that innocent people never plead guilty.

That assumption has informed centuries of law enforcement, and decades of movie plots and murder-mystery novels. The whole point in many investigations is to get the bad guy to confess. Laboratory experiments and dozens of case studies, however, show it is not hard to get innocent people to confess.

"Innocence is a state of mind that puts innocent people at risk," said psychologist Saul Kassin at Williams College, who has studied the phenomenon. Innocent people, Kassin found, are more likely to waive their constitutional rights to remain silent and to have a lawyer present. Innocent people also assume that innocent people do not get convicted, or that objective evidence will exonerate them. Nearly a quarter of all convictions overturned in recent years based on DNA and other evidence have involved false confessions.

While false confessions often involve the mentally disturbed -- John Mark Karr last year confessed to being involved in the murder of 6-year-old JonBenet Ramsey, despite convincing evidence that showed he was innocent -- Kassin and Gary Wells, a psychologist at Iowa State University, say the problem is not limited to the mentally ill.

In one experiment, Kassin asked volunteers to perform a challenging task on a computer but warned them not to touch the "Alt" key or risk damaging a computer. Volunteers were told that the computer had been damaged and were asked whether they hit the banned key. In reality, the volunteer did nothing wrong. Most volunteers denied it, but as the initial task they were given was made difficult, they became less sure because they were distracted. When researchers had confederates lie about having seen the volunteers hit the Alt key, the number of people who confessed went up to 100 percent. Every stage of increased pressure led ever larger numbers of volunteers to believe they were really guilty.

Craig said he confessed to a misdemeanor because he panicked and had hoped the plea would make the problem go away. Allegations about Craig's secret homosexual life had been the subject of an investigation by an Idaho newspaper, and the senator said he had been upset about the rumors. But he didn't plead guilty to everything. He refused to admit he ran his fingers under the stall separating him from the cop, which would have been a gross misdemeanor.

Interrogation experts such as Joseph Buckley, president of John E. Reid and Associates, a Chicago-based firm that has trained tens of thousands of interrogators, and Pete Blair, an assistant professor of criminal justice at Texas State University, say false confessions are rare. Blair said laboratory experiments don't say much about the real world -- where confessing to crimes can involve serious penalties -- and that Craig's case had all the hallmarks of a guilty politician trying to save his skin.

"Everybody who confesses tries to retract," added Buckley. "When you talk to your lawyer, he says, 'we should challenge that so we will say you were coerced or abused.' That is very, very common."

But psychologist Kassin said that when he heard a widely disseminated audio recording of Craig's confession to a police officer, he had "the unnerving sense that something was not right."

In a recent presentation at a meeting of the American Psychological Association in San Francisco, Kassin pointed to the 1989 "Central Park jogger" case where five young men confessed to violently raping a jogger in New York. One defendant later reenacted for jurors how he pulled off the victim's running pants. Another blamed peer pressure. Although there was conflicting evidence, confessions always make for powerful testimony, and all five were convicted.

Thirteen years after the crime, a serial rapist named Matias Reyes confessed to the crime and said he had acted alone. Reyes's DNA matched DNA taken from the victim, and he knew details about the crime scene that had not been made public. The five young men, who had spent more than a decade behind bars, were released. Detectives and commentators around the country remained unconvinced.

If the five men were innocent, they demanded, why in the world would they confess?

© 2007 The Washington Post Company