A Colorado man imprisoned for 15 years says he was wrongly convicted of a triple shooting inside an Adams County bar because of faulty eyewitness identification.
James Garner, 51, is challenging his conviction with help from the Korey Wise Innocence Project. He claims that three witnesses misidentified him as the shooter in a crowded bar and that new science around the unreliability of eyewitness identification should invalidate his 2012 conviction.
None of the three witnesses could pick Garner out of a photo array shortly after the shooting. They did not identify him as the shooter until three years later, during Garner’s jury trial, when the three victims, all brothers, pointed to him at the defense table and said for the first time that he was the shooter.
“That’s an enormous red flag,” said Brandon Garrett, a professor of law at Duke University. “People’s memory does not get better over time.”
Garner is due in Adams County District Court on Monday for the beginning of a two-day hearing on his petition for post-conviction relief. He is arguing both that his attorneys failed to adequately represent him during his jury trial and that the science around eyewitness identifications has changed enough that it critically undermines the original evidence against him.
Garner doesn’t dispute that he was at the bar at West 52nd Avenue and Pecos Street on Nov. 22, 2009. He was there celebrating a birthday with several friends. Around 1:30 a.m., a dispute broke out between Garner’s group of friends and a separate group at the bar, according to court records.
Someone fired at least five shots, and three brothers in the other group were wounded, the records show. All three survived.
After the shooting, almost everyone in the bar ran away, including Garner. He dropped a pair of glasses in the chaos and his girlfriend dropped her phone — police later used those items to identify Garner.
The three wounded brothers and the bar’s staff stayed until police arrived, court records show. Witnesses initially gave varying descriptions of the shooter. One said the shooter was a white bald man; another said the shooter was a Hispanic man with short black hair, another witness described the shooter as thin with a mustache.
Police never found the gun used in the shooting and did not make any arrests for several months, until they eventually identified Garner as having been present in the bar. Officers showed the witnesses a lineup of photos of five random people and Garner, to see if they could identify Garner as the shooter.
None were able to. One witness said Garner was “possibly there,” and then identified one of the random people as the shooter. Another witness also picked out a random person as the shooter.
Not until Garner’s jury trial did the three brothers say they were certain the man at the defense table was the shooter that night. Garner was convicted of assault and related charges and sentenced to 32 years in prison. He will be eligible for parole in 2029.
But since his conviction, new science has shown that witnesses should not be shown a suspect’s face more than once, said Jeanne Segil, an attorney for Garner with the Korey Wise Innocence Project, an organization within the University of Colorado Boulder that provides free legal services to people who claim to be wrongfully convicted.
“A first test really can be quite reliable,” she said. “And the finding that the brothers did not identify Mr. Garner, that is actually probative of innocence. That non-identification is looked at differently today.”
In court filings, prosecutors with the 17th Judicial District Attorney’s Office have argued the changes in eyewitness identification research do not constitute new evidence in Garner’s case.
“Since this evidence does not concern a debunked forensic process, and was widely available at the time of trial, it does not qualify as newly discovered evidence,” the prosecution wrote in a July 2023 motion.
A spokesman for the DA’s office did not return a request for comment on Garner’s petition for post-conviction relief.
Garner’s case went to the Colorado Supreme Court in 2019, which upheld the validity of the witnesses’ in-court identifications in a 4-3 decision, finding that the identifications of Garner were not “impermissibly suggestive.” The three dissenting justices found that first-time, in-court witness identifications should be handled with more legal guardrails to ensure a fair process.
Mistaken eyewitness identifications have played a role in roughly two-thirds of the hundreds of cases in which DNA later exonerated the person convicted of the crime, said Garrett, the Duke professor.
“The overall pattern in these DNA exonerations is you have very confident eyewitnesses at trial — and we now know they were wrong,” he said.
Jurors tend to trust confident in-court witness identifications even though a broad body of research shows they are unreliable, Garrett added.
“It’s counterintuitive when you are on a jury, because you think, ‘This is a crime, someone was focusing, someone is trying to pay attention,’” he said. “And in a courtroom sometimes the witnesses are really confident, because they’ve been prepared by the attorneys. What happens in the courtroom may be compelling, but it is courtroom theater, not memory.”
Garner is innocent, Segil said. He hopes the hearing Monday and Tuesday will start the process of erasing his conviction, she said.
“He’s been really deeply frustrated, as anybody in his position would be,” she said. “He’s looking forward to getting back into court and having a chance to make his case. He’s been waiting for this for a really long time.”
Sign up to get crime news sent straight to your inbox each day.