What the Justice System Gets Wrong About Eyewitness Testimony

Photo: Courtesy of Crown Publishing

Do you see a person in the courtroom here today that was the person who came in your apartment that night?”

The victim looked around the room.

Yes, sir.”

The Meriwether County prosecutor chose his words carefully. This was a pivotal moment in the trial: an opportunity to tip the case. If he played it for maximum theatrical effect, the Georgia jury would be his.

The woman on the stand waited and they watched — this elderly woman, who had been hurt so badly that a rape kit could not be used; whose face, partially paralyzed, still carried the frozen terror of the crime.

If you would, please, ma’am, come out of the witness stand, and if you would just go point out that person.”

She stepped down, just as he asked, and, in full view of the entire court, raised her hand.

That’s him.”

John Jerome White was convicted and sentenced to life in prison.

At trial, White had been adamant that he was not the man who had broken into the woman’s house in the early morning hours of August 11, 1979, beaten her, and sexually assaulted her. He was not the one who had then rifled through her purse for seventy dollars in cash, yanked the telephone cord out of the socket, and walked out the back door. “I know I didn’t rape that lady,” he insisted.        

The victim had identified White as her attacker, not once but three separate times. As any D.A. will tell you, producing a positive ID by the victim is one of the most effective ways to lock a man up. In the words of Supreme Court Justice William J. Brennan Jr., “There is almost nothing more convincing than a live human being who takes the stand, points a finger at the defendant, and says, ‘That’s the one!’ 

But John Jerome White was not the one. DNA tests conducted in 2007 excluded him as the perpetrator.

Yet there was a still darker twist to this case. With White identified, the file was closed. The authorities never looked for the real perpetrator and had no idea how close they had been to catching him — not decades later, after he had raped another Meriwether woman, but way back in 1979. It was then, just a few weeks after the assault, that the victim had stood at the police station in front of the five men pictured below.

The victim had no trouble picking him out — the one holding the number 3. As she explained, she was “almost positive” that he was the perpetrator.

It was an awful mistake. But it was only half of the story. As it turned out, she made two fateful errors that day. Standing before her, just two spots to the right, was her real attacker: James Edward Parham, number 5, round-faced in jeans and a striped shirt, glancing off to the side. At the police station, she had looked upon the true perpetrator and picked out an innocent man.

How did the victim get things so wrong? Research on memory can give us some answers. Some are obvious in retrospect: She was seventy-four years old and not wearing her glasses when she was attacked. The only light came from the closet in an adjoining room. And before leaving, the perpetrator gave her a pillow and told her, “Hold this to your face until I get out.” Research shows that a witness’s eyesight and age, the viewing duration and distance, and the lighting all play a role in whether a memory is encoded accurately.

But many of the reasons are less intuitive. For instance, one study showed that a person’s identification of someone they saw under full moonlight was only as accurate as flipping a coin — which is to say, not at all. And, overall, the factors that play a role in our encoding of a memory are much more influential than we’d imagine. Simply by altering the conditions in which a witness viewed a person, researchers were able to boost identification accuracy to 86 percent or drop it to 14 percent.

The legal system amplifies the problem by treating many of these important variables as entirely irrelevant. In White’s case, for example, one source of the victim’s misidentification may have been that the victim was white and her attacker was black. Research suggests that people are 50 percent more likely to make an error in identifying a person from another race, although individuals who have a lot of contact with the other race tend to be more accurate. The same is true of identifying someone of a different age. But because of its commitment to nondiscrimination, the law generally doesn’t acknowledge or address this reality — and most police officers, judges, and jurors don’t even know it’s an issue.

Researchers have also shown that our memories can be affected by mental or physical stress. It’s difficult to simulate the fear and anxiety associated with being the victim of a real crime, but scientists have gotten creative, drawing upon other experiences that may generate similar feelings. In one study, for example, participants were taken on the “Horror Labyrinth” tour of the London Dungeon, in which they passed through a dark maze with a screaming skeleton, frightening music, disorienting mirrors, and an actor in a dark robe who blocked the path of visitors. Participants who did not find the experience distressing were more than four times better at identifying the actor out of a nine-person lineup than those who reported high levels of anxiety and had increased heart rates.

So, when a criminal uses a gun in a robbery, it doesn’t just encourage his victim to comply; it also makes it less likely that she’ll remember his face. There is more to it, though, than generalized fear; our poor memory also comes down to where our attention is drawn during a holdup. When a weapon is aimed at us, it tends to dominate the scene, and if our eyes are glued to the barrel, we are going to struggle to identify the suspect later on.

Compounding the problem is that not only do jurors fail to appreciate the significance of poor witness viewing, but they also depend on dubious factors to decide whether an identification or account is accurate. For instance, jurors are two to three times as trusting of extremely confident witnesses as they are of less confident ones, and these impressions affect their behavior: the more confident the witness, the more likely it is that a jury will convict. When a woman like the victim in the White case has the courage to come forward to identify the man who raped and beat her and says she’s certain, we feel assured that we have the right man.

We also need to keep in mind that the wrongfully convicted are not the only ones who end up suffering from flawed witness procedures. When an innocent person is locked up, the police stop looking for the true offender. Indeed, we know of at least 230 misidentification cases like White’s, in which an innocent person was exonerated and the real perpetrator eventually identified through DNA. In almost half of those cases, the perpetrator committed other crimes after the investigation was closed.

When memories are distorted and erroneous details about a crime are passed to the authorities, police officers and prosecutors lose days, weeks, and months investigating false leads and building cases against the wrong people. They fail to attain justice for victims and waste limited resources that are desperately needed elsewhere. Improving witness procedures does not pit the interests of defendants against law enforcement; it serves our common goal of achieving fairness and efficiency in criminal law. Indeed, many police officers are actually quite supportive of reforms. As a spokesman for the Georgia Bureau of Investigations explained at the time of White’s exoneration, “Nobody in law enforcement wants to arrest the wrong person.”

Excerpted from UNFAIR: The New Science of Criminal Injustice by Adam Benforado. Copyright © 2015 by Adam Benforado. Excerpted by permission of Crown Publishers, an imprint of the Crown Publishing Group, a division of Penguin Random House, LLC. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.

The Legal System Doesn’t Get Eyewitness Accounts