Can Reforming Eyewitness Testimony Prevent Wrongful Convictions?

Jun 30, 2016

After Jennifer Thompson was raped, police in North Carolina showed her a series of photos and asked if she saw the man who did it. In one photo, she was sure she saw her rapist’s face.

That face belonged to Ronald Cotton. But a decade later, DNA evidence proved to Burlington Police and prosecutors that Cotton was innocent. Someone else named Bobby Poole had committed the crime.

“Ronald and I both knew that we had been victims,” she said, at a keynote speech about reforming eyewitness testimony-gathering at Yale Law School on June 27. “We had been victims of the system that failed. We had been victims of Bobby Poole. And it wasn’t just us: six other women were raped that fall and spring before Bobby Poole was ever apprehended.”

At the first-ever National Symposium on Eyewitness Identification Reform, Thompson remembered old techniques for handling witness testimony. Those techniques gave everyone, including her, false confidence that Cotton was the perpetrator: police officers told her when she picked Cotton that she’d selected the suspect, which made her feel affirmed, like she “was a ‘good victim.’”

“All the players were trying to do the right thing,” she says. “There wasn’t gross misconduct, there wasn’t negligence on the police department, this was a case where every person involved was a good player—with the exception of [Bobby Poole].”

According to the National Registry of Exonerations, more than 1,800 people like Cotton have been cleared of false convictions in the United States since 1989. The symposium, run by the Hartford-based Justice Education Center, had one goal: prevent stories like Thompson’s and Cotton’s from happening again. Police officers, lawyers and advocates were there to talk about new techniques that could prevent false convictions.

Connecticut has an ugly history with wrongful convictions. But in 2011, the state set up a new task force that would help change that. The task force set up rules for police departments that would make eyewitness testimony, at least, more reliable.

The rules included “double-blind identification procedures,” which mean that when a witness is asked to peruse photos to spot a potential perpetrator, neither the witness nor the person administering the test knows who police suspect may have committed the crime. They also include getting information from the witness about how confident he or she is in her decision; in a properly-administered test, a confident witness is much more likely to be right than a witness who harbors doubts. The techniques mandated in Connecticut also include a statement from police telling the witness that even if she doesn’t see anyone of interest in the photos, police will keep investigating.

“It’s really become a model,” says Darcy McGraw, with the Connecticut Innocence Project—a state-funded group that helps people in prison overturn false convictions. Now, she says, other jurisdictions are trying to implement similar rules with their own police forces.

One of those techniques discussed at the symposium was videotaping any procedure when police officers ask witnesses to identify a potential suspect. Richard Colangelo is the state’s attorney for the Stamford-Norwalk district.

“Videotaping the procedure is a great policy to have so it takes away the argument—hey, did somebody do something wrong,” he says. Many police departments in Connecticut aren’t doing that yet, Colangelo adds. “It’s something we should strive to get to.”

Another technique: asking someone who’s totally unrelated to the investigation to show witnesses the photos, so that the detective officer can’t accidentally lead the witness. That’s a policy already followed in most social science tests and drug trials; it’s called double-blind testing.

Double-blind testing for eyewitnesses is now mandatory in Connecticut, thanks to the 2011 task force. But Middletown Police Captain Gary Wallace says the symposium showed him how to make it better.

“A lot of times, just out of sheer convenience, we use our own detectives—just because they’re in the bureau, the witness is up in the bureau,” he says, adding that the detectives are not the same ones assigned to the case, but because they’re involved in the same office and may see witnesses coming or going, they could develop a whiff of impropriety. Those whiffs are what he’s trying to stamp out. “What we don’t typically do is go find a patrol officer and say, ‘Hey, can you conduct this double blind?’ That’s something I would like to do because it keeps the process honest.”

In Connecticut police like Wallace are tinkering with a vastly improved system. But McGraw, with the Innocence Project, says one problem hasn’t been addressed: what’s going to happen to people who already convicted from often-faulty eyewitness testimony.

“It doesn’t affect cases in the past where we believe we have problematic identifications and we believe we have wrongful convictions,” she says.

McGraw hopes that Connecticut will start a new task force, like the one that decided on the rules for how police officers should gather eyewitness testimony—a task force that will decide what the rights are for people who’ve been convicted from badly-gathered eyewitness information. She hopes people who were convicted from bad testimony will get the right to a new trial.