The Courts Beginning to See That Eyewitnesses Are Fallible

Article excerpt

Byline: Robert Rocklin

As you walk down the street, you see someone knock a person down, take his bike and run off; you talk to the police at the scene. Months later, the police ask you to view a photo lineup.

The police officer says, "I will now show you several photographs. The person who committed the crime may or may not be in any of these photos. Tell me immediately if you recognize the person who committed the crime." You identify the suspect, and he is convicted of assault and robbery.

That kind of eyewitness identification is a mainstay not only of television shows and movies, but also of criminal prosecutions. But despite the widespread use of such sure-fire "I saw it with my own eyes" identifications, social science research consistently has demonstrated that eyewitness identifications often are wrong.

Indeed, it has been estimated that up to one-third of the 75,000 annual eyewitness identifications in criminal investigations are wrong. Of the first 250 exonerations involving DNA evidence, 190 involved eyewitnesses who were wrong - despite, in many cases, complete confidence on the part of the eyewitnesses.

In short, eyewitness identification is not all it is cracked up to be.

Courts often haven't paid much attention to what social scientists have learned about eyewitness testimony. Last week, however, the New Jersey Supreme Court recognized what research psychologists and others have been trying to tell the legal community for decades.

In a 135-page opinion issued on Aug. 24, the court explained that New Jersey's procedure for determining whether eyewitness identification evidence would be allowed in court had to be revised in light of decades of research calling into question the reliability of such evidence.

New Jersey courts - like the federal courts and most state courts, including Oregon's - had been using a test announced by the U.S. Supreme Court nearly 35 years ago. In its ruling last week, the New Jersey court set out a new procedure for judges to use in determining whether eyewitness identification testimony should be kept from the jury.

Under the new procedure, a criminal defendant first must demonstrate that police used a procedure that was suggestive - that is, acted in ways that made it more likely that the witness would identify the person that the police suspected of the crime.

If a defendant can meet that burden, the court must then hold a hearing to examine all factors that could have affected the reliability of the identification, such as poor lighting or the presence of a gun at the scene, or the use of suggestive police procedures when the identification was made. …