Eyewitness misidentifications have long been recognized as a major — probably the major — cause of wrongful convictions. Now, Brandon Garrett, a law professor at the University of Virginia, has become perhaps the first person to analyze the factors that helped cause the misidentifications.
In his book Convicting the Innocent, Garrett analyzes 250 cases in which lawyers at the New York City-based Innocence Project used DNA evidence to exonerate people convicted of serious crimes — most commonly, rape, murder or rape and murder. Garrett started his legal career as an associate at the law firm of Barry Scheck and Peter Neufeld, who founded the Innocence Project at Cardozo Law School in 1992 with the mission of applying what was then relatively new DNA forensics to challenge cases of possibly wrongful convictions.
Garrett found that eyewitnesses misidentified exonerees in 190 of the cases (76 percent). Based on the trial records or other materials he located for 161 of those cases, Garrett found evidence of what he called police “contamination” of identifications in the vast majority: 125, or 78 percent. The contamination, Garrett found, occurred through suggestive remarks, biased lineups or one-person “showups” — presentations of suspects to witnesses soon after crimes are committed.
Garrett also found that despite positive identifications at trial, 57 percent of the witnesses earlier had not been certain at all — “a glaring sign,” he writes, “that the identification was not reliable.” In all, Garrett said only 12 percent of the identifications were free of one or the other of those signs of possible error.
Out of 250 exonerees, 17 had been sentenced to death and another 80 to life in prison; overall, they served an average of 13 years in prison. The vast majority were minorities. All but four were male. Twenty-four were juveniles at the time of their alleged offenses; 18 had mental disabilities.
A disproportionate number of the rape cases involved cross-racial identifications, which eyewitness researchers say are more prone to error than same-race IDs. Out of 171 exonerees convicted of rape, 84 were African-Americans or Latinos accused by white victims.
Garrett notes that DNA test results identified the actual perpetrators in 112 of the 250 cases (45 percent), most often through a “cold hit” or a match in law enforcement DNA data banks. “The damage caused by these wrongful convictions extends far beyond the suffering of the innocent,” Garrett writes. “Dozens of criminals continued to commit rapes and murders for years until DNA testing identified them.”
— Kenneth Jost