The U.S. Supreme Court had a chance last week to address a serious defect in criminal cases, improve the quality of American justice and strengthen people's faith in the court system.

It did none of those things.

Instead, by an 8-1 margin, the court reaffirmed a rule that stretches back at least as far as a 1977">decision involving eyewitness identifications. According to the rule, if law enforcement officers appear to have used suggestive methods to obtain such an identification, a judge must assess its reliability before it may be introduced as evidence in a trial.

That process, as Justice Ruth Bader Ginsburg wrote last week in her majority">opinion, helps ensure a person's constitutional right to due process of law and serves as a deterrent to police rigging witness ID conditions to get the results they want. It also is an incentive for officials to ensure accurate identifications take criminals off the streets, rather than convicting innocent individuals.

But the court's ruling last week in Perry v. New Hampshire ignored two crucial developments.


First, more than 2,000"> studies into the mechanics and psychology of eyewitness identifications, all published since the court's 1977 decision, have found a wide range of variables that produce inaccurate eyewitness identifications. They include the presence of a weapon, consumption of alcohol or drugs, how long a witness watched what was happening, how long after a crime an identification is made and the race and age of an alleged perpetrator. They go well beyond suggestive conditions set up by police.

Second, despite the often-shaky reliability of eyewitness identifications and the opportunity of defense lawyers to expose flaws through cross-examination in court, juries place disproportionate faith in the IDs anyway.

For his 2011"> book, "Convicting the Innocent: Where Criminal Prosecutions Go Wrong," University of Virginia law professor Brandon Garrett researched the cases of 250 people who first were found guilty — some sentenced to be executed — and later found innocent through DNA evidence. He found 190 "who had been convicted based on eyewitness testimony," Mr. Garrett wrote last year in an opinion piece for Slate. "Nearly all the cases had eyewitnesses who were positive at trial that the defendant was the person they saw."


Barion Perry, who was convicted of theft in 2009, is serving three to 10 years in the New Hampshire State Prison. He was arrested after a questionable identification at 3 a.m. by a witness who could describe the suspect only as a young black man. Even so, she pointed to Perry, who was standing next to a uniformed police officer and was the only black man in the vicinity. She later could not pick Perry's picture out of a photo lineup.

Perry's lawyers had"> asked the Supreme Court to require judges to review all eyewitness identifications made under conditions that might be considered suggestive, whether police set them up or not. More and more states do just that. After a comprehensive review, for example, the New Jersey Supreme Court last summer issued new rules covering identifications, including instructions for judges to warn juries about the many factors that produce inaccurate ones.

Rather than acknowledge the significance of the abundant research, the Supreme Court, except for dissenting Justice Sonia Sotomayor, chose established process over fairness. The injustices are sure to continue.