Posted by Steven Fabian, Sonoma County Criminal Defense Lawyer
As she walked under the freeway overpass, a man grabbed her from behind and ripped at her clothes. She smashed him in the face with a CD player and ran across the street. He caught her again, threw her into bushes near a stoplight and sexually assaulted her. When she saw another car on the street, she broke free and ran to it. The driver let her in.
Police were called. Although the victim was unable to provide enough details for a composite sketch of her attacker, she remembered the truck.
A similar vehicle was spotted and the trail led detectives to Uriah Courtney, a 25-year-old construction worker from North Park. His picture was put in a photo lineup and the victim, although not completely sure, picked him out. So did another witness.
Courtney was arrested. The main evidence against him at trial was the eyewitness testimony. He was convicted of kidnapping and rape and sentenced to life. He told the victim in court he didn’t do it and said he hoped the truth would one day come out.
It has. Recent DNA testing on the victim’s clothing cleared Courtney and pointed to a felon who lived about three miles away from where the attack occurred. Charges were dismissed and Courtney freed from prison after serving eight years.
The case raises questions about whether enough is being done to guard against mistaken eyewitness identification, long known to be a leading cause of wrongful convictions.
Hundreds of law-enforcement agencies across the country have adopted reforms in recent years aimed at improving ID accuracy. But the key ones haven’t taken hold in San Diego County, where there have been five wrongful convictions involving faulty eyewitnesses, according to the National Registry of Exonerations, a database run by law schools in Michigan and Illinois.
The database lists every known exoneration in the country — 1,168 at last count — since 1989. According to its records, mistaken ID’s were a contributing factor in 43 percent of all cases — and in 81 percent of the sexual assaults.
Another compilation, by the Innocence Project, a national litigation and public-policy group that has a chapter in San Diego, calls eyewitness misidentification “the single greatest cause of wrongful convictions nationwide.” It played a role, the Innocence Project said, in more than 75 percent of cases later overturned through DNA testing.
“Obviously I’d like to see changes,” Courtney said during a news conference last month. “Misidentification puts too many people away, and I happened to be one of them.”
A steady stream of exonerations has prompted government officials from coast to coast to pay more attention to 30 years of social science research showing how unreliable eyewitness accounts can be, especially in traumatic situations such as a violent crime.
The result has been a push for reforms aimed in particular at one of the most enduring symbols of police detectives at work, at once recognizable to any fan of mystery books, TV cop shows and Hollywood thrillers.
As six-packs go, this one’s no party.
To police, a six-pack is a group of mug shots — one is of the suspect, the other five are “fillers” — shown to an eyewitness. Agencies everywhere use them, either with actual photos or pictures on a computer screen.
Ideally, the photos are of people who look alike in terms of age, race, build, hair color and facial characteristics. They should be presented in a way that doesn’t suggest who the suspect is. Those viewing the photos should be told it’s OK to say no one matches the person they saw committing the crime.
But we don’t live in a perfect world, and mistakes get made, both in photo lineups and live lineups. Detectives can inadvertently give clues about which one the witness should pick. Sometimes it’s obvious from the photos — in one rape case, the suspect was the only one whose picture had an “R” written on it.
Once made, mistakes can be hard to undo. In trials where there is no significant physical evidence — and sometimes even in those where there is — an eyewitness account can be the most persuasive thing presented to a jury, especially if it comes from the victim.
“Studies have shown that what a witness will do with a six-pack is pick the person who looks the closest to the one they saw,” said Jan Stiglitz, a professor at California Western School of Law in San Diego and co-director of the California Innocence Project based there.
“Then if they do a live lineup, the witness tends to identify the person not from the crime, but from the six-pack. And then the witness sees the guy in the courtroom, and each time that choice gets reaffirmed. So when it comes to trial, the witness is absolutely certain that the defendant is the person they saw.”
In the Courtney case, when the victim picked him out of the photo lineup, she said she was 70 percent sure he was the rapist, according to Alissa Bjerkhoel, his Innocence Project attorney. At the trial, the victim was 100 percent certain. She said seeing him in person made her more confident.
Another witness in the case also picked Courtney out of a photo lineup. The problem with that one, Bjerkhoel said, is that the witness is Hispanic and Courtney is white. Cross-racial identifications are known from research to be problematic; people tend to see those of other races as looking alike.
David Greenberg, a San Diego chief deputy district attorney, said prosecutors are aware of the potential pitfalls with eyewitness identifications and take those into account when deciding how to handle a case. He stressed that wrongful convictions are rare.
“Even though we want to be perfect, and that is what we strive for, you’re never going to get there because we’re human,” he said. “It’s not a completely infallible system. Nobody wants to see anybody in a situation where they don’t belong.”
“Double-blind sequential” sounds like something from a science lab out at Scripps, but it’s actually a description of what reformers say is the best way to do lineups to guard against mistaken eyewitness identification.
In double-blind, the detective presenting the photos to the witness doesn’t know which one is of the suspect, so he or she can’t influence the choice. And the photos are presented one after another, not all together, to guard against the witness comparing the pictures to each other and picking the one that most resembles who they saw commit the crime.
“Those two changes make the biggest difference,” said Gerald Uelmen, a law professor at Santa Clara University and executive director of the California Commission on the Fair Administration of Justice.
Created in 2004 to address the growing number of wrongful convictions in the state, the commission reviewed studies and heard from experts on eyewitness identification before recommending adoption of double-blind sequential lineups.
Legislation to that effect was passed twice but then vetoed by former Gov. Arnold Schwarzenegger, who heeded concerns from law enforcement lobbying groups that the measures would restrict the flexibility of detectives trying to solve crimes. A new bill is pending.
Several states have adopted reforms, including Minnesota, North Carolina, Wisconsin and New Jersey. In California, Santa Clara was the first of a handful of counties to go in that direction, approving protocols for all police agencies to follow.
Jeffrey Rosen, the district attorney there, said prosecutors were convinced by research showing double-blind sequential reduces false identifications.
“We thought it would decrease the overall number of identifications, but in fact that didn’t happen,” he said. “What happened is it got a lot more accurate.”
Greenberg, of the local DA’s office, said prosecutors here leave it up to the individual law enforcement agencies to decide what kinds of lineups to use. “If we see them doing something improper, we’ll bring it to their attention, but we’re not going to tell them which way to go,” he said.
The Sheriff’s Department, which investigated the Courtney case, is taking a look at double-blind sequential. “If our review supports the use of this technique to help prevent misidentification,” said Central Investigations Lt. James Bolwerk, “we will implement it and update our policy accordingly.”
Four years ago, the San Diego Police Department was one of four agencies nationwide to participate in a field test of double-blind sequential. The results showed a decrease in the number of mistaken identifications. Investigations Capt. Terry McManus said detectives are adopting the use of sequential six-packs, but aren’t sold yet on the need to do them in a double-blind manner.
Innocence Project attorney Bjerkhoel said it’s not surprising to her that law-enforcement agencies are hesitant to embrace change. “In the majority of cases, cops are getting it right and the eyewitnesses are picking the right people,” she said. “Because these are procedures that they’ve used for a long time, and wrongful convictions are the exception to the rule, they figure the process is working good enough.”
She likened wrongful convictions to plane crashes, which are also rare. “When a plane crashes, you figure out what went wrong and you fix it and you make sure the next plane doesn’t crash,” she said. The Courtney case “should be a turning point for any of the law enforcement in San Diego to say we gotta make sure this type of plane crash doesn’t happen again.”