“I beat myself up a lot” about having confessed, Mr. Lowery said in a recent interview. “I thought I was the only dummy who did that.”
But more than 40 others have given confessions since 1976 that DNA evidence later showed were false, according to records compiled by Brandon L. Garrett, a professor at the University of Virginia School of Law. Experts have long known that some kinds of people — including the mentally impaired, the mentally ill, the young and the easily led — are the likeliest to be induced to confess. There are also people like Mr. Lowery, who says he was just pressed beyond endurance by persistent interrogators.
New research shows how people who were apparently uninvolved in a crime could provide such a detailed account of what occurred, allowing prosecutors to claim that only the defendant could have committed the crime.
An article by Professor Garrett draws on trial transcripts, recorded confessions and other background materials to show how incriminating facts got into those confessions — by police introducing important facts about the case, whether intentionally or unintentionally, during the interrogation.
To defense lawyers, the new research is eye opening. “In the past, if somebody confessed, that was the end,” said Peter J. Neufeld, a founder of the Innocence Project, an organization based in Manhattan. “You couldn’t imagine going forward.”
The notion that such detailed confessions might be deemed voluntary because the defendants were not beaten or coerced suggests that courts should not simply look at whether confessions are voluntary, Mr. Neufeld said. “They should look at whether they are reliable.”
Professor Garrett said he was surprised by the complexity of the confessions he studied. “I expected, and think people intuitively think, that a false confession would look flimsy,” like someone saying simply, “I did it,” he said.
Instead, he said, “almost all of these confessions looked uncannily reliable,” rich in telling detail that almost inevitably had to come from the police. “I had known that in a couple of these cases, contamination could have occurred,” he said, using a term in police circles for introducing facts into the interrogation process. “I didn’t expect to see that almost all of them had been contaminated.”
Of the exonerated defendants in the Garrett study, 26 — more than half — were “mentally disabled,” under 18 at the time or both. Most were subjected to lengthy, high-pressure interrogations, and none had a lawyer present. Thirteen of them were taken to the crime scene.