July 25, 2007 -- University of Virginia law professor Brandon Garrett’s groundbreaking study examining the cases of the first 200 wrongly convicted people later found innocent through DNA testing reveals fundamental flaws in the criminal system. Courts not only failed to redress trial errors over years and sometimes decades, but they often found innocent appellants to be guilty, and sometimes denied relief even after DNA testing proved innocence, Garrett says.

The study, to be published by the Columbia Law Review in January 2008 as the first comprehensive inquiry to explore the cases, examined evidence introduced during the appellants’ initial trials, each claim they raised during their appeals, how courts ruled on them, and how DNA testing ultimately freed them. Garrett has personal experience with the issue; he formerly litigated wrongful conviction cases as an associate at the law firm Cochran, Neufeld & Scheck in New York City.

“Although our system of criminal appeals is intended to protect against wrongful convictions, this study shows how appellate courts misjudged innocence,” Garrett says. “A third of those with written decisions received rulings in which judges claimed that errors were harmless due to outweighing evidence of guilt in their cases. Some courts judged innocence particularly poorly, in retrospect. Nine percent of innocent appellants had a court deny a new trial, ruling that there was ‘overwhelming’ evidence of their guilt.”

“This study shows what an uphill battle many of the Innocence Project's clients faced in the courts,” says Peter Neufeld, co-director and co-founder of the Innocence Project, a nonprofit organization formed to assist prisoners who could be proven innocent through DNA testing. “In our experience, many exonerees suffered terribly unfair trials, were convicted based on unreliable evidence, and this should have been apparent long before they had DNA testing. Just like others in the criminal system, judges can learn valuable lessons from these DNA exoneration cases about the importance of ensuring rigorous and fair dispositions.”

Few innocent appellants in the study brought innocence claims—just 25 percent of those with written decisions. “Those that brought innocence claims failed,” Garrett says. “One reason for this is that the Supreme Court has not yet recognized any constitutional right entitling a petitioner to relief on the grounds of actual innocence.”  Yet even those who brought state law innocence claims failed.

The majority, 86 percent, never received any relief during their appeals before they obtained DNA testing. However, even after they obtained DNA testing, some innocent appellants still had trouble convincing a court to vacate their conviction. “Courts denied 12 innocent appellants relief despite at least preliminary DNA test results exonerating them,” Garrett says. In 41 cases (20 percent), appellants received a pardon from their state executive, often because they lacked any available judicial forum.

“Many of these exonerees had trouble obtaining access to DNA testing due to traditional limits on post-conviction motions brought after a statute of limitations,” Garrett says. “In response, most states have passed statutes making DNA testing more widely available post-conviction. However, those statutes often include severe restrictions that most of these exonerees could not have satisfied. Decades into the DNA era, we still do not make DNA testing available to all of those who could use it to conclusively prove their innocence.”

Criminal appeals brought before DNA proved innocence resulted in a high 14 percent reversal rate, but Garrett found that rate consistent with the reversal rate in other rape and murder convictions not involving DNA evidence. 

“Approximately half of these reversals were innocence-related, indicating a high degree of factual error in the most serious criminal prosecutions,” Garrett says.

The innocent group, all male save one, included 22 juveniles and 12 mentally handicapped people; 71 percent were minorities. The vast majority of exonerated rape convicts (73 percent) were black or Hispanic, while studies show only about 37 percent of rape convicts are minorities. Fourteen were sentenced to death, including seven who confessed, three of whom were mentally retarded. In the entire innocence group, only eight pled guilty.

The 200 exonerees were convicted based on eyewitness identifications (79 percent), forensic evidence (55 percent), informant testimony (18 percent), and false confessions (16 percent).

“Surprisingly few innocent appellants brought claims regarding those facts, such as challenging an eyewitness identification for being suggestive, even though in 56 cases, the victim’s identification was the sole noncircumstantial evidence supporting the conviction,” Garrett says. 

Of those with written decisions, 45 appellants challenged none of the central facts supporting their conviction during their appeals. Garrett suggests this could be due to ineffective appellate lawyering, but also because “factual claims require expensive investigation, and the standards for such claims ensure that success is rare. 

“We need to embrace changes to our criminal system to facilitate the accurate development of factual evidence at all stages of the process — not just in DNA cases, but in the vast majority of cases in which DNA testing cannot be conducted,” Garrett says.

Adopting more dependable eyewitness identification procedures across the nation, videotaping identifications and interrogations, evaluating the use of jailhouse informants, improving or establishing oversight of crime labs, and enabling access to independent forensic experts would all provide courts and juries with more reliable evidence, Garrett says, and would then empower appellate courts to better remedy wrongful convictions when they happen.

• Reported by Mary Wood