Skip to content, or skip to search.

Skip to content, or skip to search.

“I Did It”


Frank Sterling in 1992.  

But, of course, they do. In recent years, the use of DNA evidence has allowed experts to identify false confessions in unprecedented and disturbing numbers. In the past two decades, researchers have documented some 250 instances of false confessions, many resulting in life sentences and at least four in wrongful executions. Of the 259 DNA exonerations tracked by a major advocacy group, 63 of them—or one out of every four—was found to have involved a false confession. Counting just the homicide cases, the proportion shoots up to 58 percent of all exonerations. Even this number could be an underestimate. “Most of the documented false confessions have been in highly publicized murder cases,” says Steven Drizin, of Northwestern Law School’s Center on Wrongful Convictions. “There is no reason not to think the same tactics would be as effective if not more effective in lesser cases, where the punishment that could flow from a confession would be less.” False confessions appear to be particularly common in New York State, in which twelve of the 27 DNA-based exonerations have turned out to be based on bogus admissions of guilt.

Researchers who study false confessions say the roots of the problem lie in the interrogation tactics themselves. The most influential such method is the Reid technique, a decades-old nine-step procedure designed to isolate and persuade a suspect to reveal his deceptions. Virtually every police department in the country has been influenced, directly or indirectly, by the Reid technique. Its defenders see it as the cornerstone of good police work, but its detractors say it places too much power in the hands of interrogating officers. In light of the new research documenting the scope of the problem, reformers in New York and elsewhere are calling for a wholesale reevaluation of the way the police question suspects. Frank Sterling’s story should help their cause; it demonstrates just what can go wrong with the science of interrogation.

In 1940, a burly, clean-cut Irish Catholic cop named John E. Reid was thinking of quitting the Chicago police force. Reid was tough, a former guard on the DePaul University football team, but was never comfortable carrying a gun. At the last minute, he applied for a transfer to a desk job at the Chicago crime lab. He arrived in the midst of a technological revolution in police work. In 1931, a presidential panel known as the Wickersham Commission had exposed abuses brought by the “third degree,” the use of force by police to extract confessions. Police across the country had held suspects’ heads underwater, hung them out of windows, and beaten them. In 1936, the Supreme Court decision Brown v. Mississippi—the brutal case of three black men who were beaten and whipped until they confessed—effectively outlawed confessions brought by brute force. Crime labs like Chicago’s began developing new, more scientific means to solve cases: ballistics, document examination, and lie detection.

As much as anyone, John Reid can be credited with leading American law enforcement into the modern age. Reid’s advances began with the lie detector. In 1945, he designed a chair that used inflated rubber bladders to detect a subject’s jitters. In 1947, he essentially created the modern polygraph procedure with the “control-question technique,” a way of measuring a suspect’s reaction to provocative questions. That same year, Reid left the crime lab and founded John E. Reid & Associates, which went on to train scores of polygraph analysts, including members of the CIA and the Mossad.

Reid’s most influential work focused on the art of the interrogation. Soft-spoken and sincere, he had a knack for gently persuading suspects to confess. “It was almost a priestlike approach,” says George Lindberg, who worked for Reid for thirteen years. “He’d hold your hand and say, ‘You should really get this off your chest.’ ” Reid played an important role in a number of high-profile Chicago murder trials, and other cities shuttled him in as a closer for their most sensitive cases. He was credited with personally helping to solve some 300 murders and coaxing 5,000 thieves to confess. Some in law-enforcement circles called him the most famous name next to J. Edgar Hoover. Reid’s aim wasn’t always true—in 1955, he got a Nebraska man named Darrel Parker to admit to killing his wife, and the real killer confessed 33 years later—but his faith in his own ability, and in the professionalization of his craft, led him to believe interrogations could be systematized to the point of being foolproof. “It’s almost as if every crook reads the same book on what to do and say to give themselves away,” he liked to say.

In 1962, Reid and his mentor, a Northwestern Law professor named Fred Inbau, co-wrote the first edition of Criminal Interrogation and Confessions. Criminologists and law historians credit their method with defining the culture of police-interrogation training for the past half-century. The procedure basically involves three stages meant to break down a suspect’s defenses and rebuild him as a confessor. First, the suspect is brought into custody and isolated from his familiar surroundings. This was the birth of the modern interrogation room. Next the interrogator lets the suspect know he’s guilty—that he knows it, the cops know it, and the interrogator doesn’t want to hear any lies. The interrogator then floats a theory of the case, which the manual calls a “theme.” The theme can be supported by evidence or testimony the investigator doesn’t really have. In the final stage, the interrogator cozies up to the subject and provides a way out. This is when the interrogator uses the technique known as “minimization”: telling the suspect he understands why he must have done it; that anyone else would understand, too; and that he will feel better if only he would confess. The interrogator is instructed to cut off all denials and instead float a menu of themes that explain why the suspect committed the crime—one bad, and one not so bad, but both incriminating, as in “Did you mean to do it, or was it an accident?”


Current Issue
Subscribe to New York

Give a Gift