Brandon L. Garrett’s Convicting the Innocent is a systematic investigation of what went wrong in the cases of the first 250 people to have their wrongful convictions overturned by DNA evidence. Garrett uses the lessons of those cases to present a powerful case for a number of improvements to our criminal justice system that would dramatically reduce the number of wrongly convicted. In a conversation we had here last fall, Garrett discussed the phenomenon of false confession, and explained some of the obstacles to the implementation of the reforms he suggests. Below, he tells the story of how an innocent man came to spend twenty-two years in prison for a crime that he didn’t commit.
Tens of thousands of suspects are identified by eyewitnesses in line-ups each year. How accurate are those identifications? Eyewitness misidentifications were the most common type of flawed evidence in the cases of the first 250 innocent people cleared by DNA tests. In Chapter Three of my new book, Convicting the Innocent, I describe what often goes wrong. Police use live line-ups and, far more often, photo arrays to test the memory of an eyewitness to a crime. The remarkable image below captures what can go wrong in those identification procedures. Look at this photo. The victim originally told the police that her attacker was “well built” and had a “round face.” Does any person stand out? The man on the far right seems to be the only one who would fit that description.
And yet, confronted with this line-up, the victim wrongly identified an innocent young man placed right in the middle, John Jerome White. In a dramatic moment at the trial, the victim pointed to John Jerome White and identified him as her assailant. The transcript reads:
How did the victim come to identify the wrong man at trial?
In addition to initially describing her attacker as a black man, “well built” with a “round face,” she also described him as clean shaven with short hair. White was very thin and had a mustache. He did not have a round face and he was not “well built.” The victim was elderly and had not been wearing her prescription glasses when assaulted in the dark in her home, but that alone does not explain how she came to identify White. Procedural flaws also contributed to her misidentification. More than a month and a half after the crime, police asked her to look at photographs. She identified White in a photo array. That array was not documented. However, we do know that she was not completely certain at the time.
Police wanted to make sure, so they conducted a live line-up. White was the only man from the photo array to be repeated in that line-up. Psychologists have found that repeating just one person in a second identification procedure reinforces eyewitness confidence, even when they are wrong. The detective also told her, before the second line-up, that “he had caught somebody,” implying that they had the attacker in person this time. Psychological studies show how an encouraging comment like that can increase a witness’s confidence and cause error. Instead, the eyewitness should be instructed that there may or may not be a suspect in the array or line-up.
In this case, we can see how reinforcement from the police, even if completely well-intentioned and designed to reassure the eyewitness and double-check her identification, had the opposite effect of generating false certainty. She remembered his face from the first photo array, so at the live line-up she looked straight ahead and identified John Jerome White, as she later did again in court. At trial, White told the judge, “I know I didn’t rape that lady.” However, based on that eyewitness identification and some unreliable forensic evidence, he was convicted and sentenced to life in prison.
The woman’s attacker was indeed present in the live line-up depicted above. Only years later did DNA evidence prove that the man on the far right, the one who matched her original description, actually committed the rape that the police had convened this line-up to solve. He was not the suspect at the time, though. He just happened to be in jail, and was selected to help fill out the line-up. This coincidence illustrates how mishandled identification procedures can lead to the conviction of innocent people, even when victims are confronted with the real perpetrators.
John Jerome White spent more than 22 years in prison until he was exonerated by DNA tests in 2007. The powerful footage below shows White just after his release:
Seventy-six percent of the 250 DNA exonerees were identified by eyewitnesses. In most of those cases, just like in John Jerome White’s case, eyewitnesses had earlier been uncertain but became certain by trial. Unsound and suggestive identification procedures played a troubling role. Police used unnecessary show-ups, where they presented the eyewitness with just the defendant. Stacked line-ups made the defendant stand out. Suggestive remarks told the eyewitness who to identify, to expect a suspect in a lineup, or confirmed their choice. Further, almost half of the eyewitness identifications were cross-racial, like that in White’s case; psychologists have long documented how eyewitnesses have greater difficulty accurately identifying persons of another race.
One of the themes of Convicting the Innocent is that insights from wrongful conviction cases and also from science offer improved criminal justice procedures that can prevent prosecutions from going wrong. There are a set of well-recognized best practices, grounded in decades of social science research, which can reduce eyewitness misidentifications. Eyewitnesses should be told that the attacker might not be present in the line-up. Their initial confidence level should be documented (because by the time of trial it may change). The most crucial reform is double blind administration of identification procedures; the police officer administering a photo or live lineup should not be aware of whom the suspect is, and the witness should be told that the officer does not know. Despite decades of research, police, prosecutors, and courts have not typically adopted those best practices. As in so many other areas that I discuss in the book, such as false confessions and invalid and unreliable forensics, DNA exonerations have made the need to improve our system far more salient. If our criminal justice system does not adopt accurate procedures for correctly identifying the guilty and excluding the innocent, these tragic wrongful convictions are bound to recur.
(You can read a review and summary of Convicting the Innocent by the Director of the Texas Innocence Network here. To receive updates on news related to the problem of wrongful conviction, you can connect with the Convicting the Innocent page on facebook.)