The Victim: A Wrongful Conviction

It was Burton's testimony about the B-secretor evidence that had helped seal Ruffin's fate in 1982. In 2002, however—when she'd been retired for 14 years and dead for three—she would become his unwitting savior. His file was one of those still stored in boxes in a warehouse of the Virginia Department of Forensic Science. Finally, after two decades of trying to prove his innocence, Ruffin got a break.

Recently, as Meg and I sat at the tiny table in her kitchen leafing through the dozens of newspaper clippings she has collected on the case since it resurfaced in the news in December 2002, she said, "I've read everything printed, and followed it like crazy." During the two months after that first call from the prosecutor, she welcomed the certainty DNA would bring. But on February 12, when the prosecutor called again to say that Ruffin was innocent—the DNA matched a man named Aaron Doxie, a convicted rapist—she nearly collapsed.

I have to say, when I see the mug shots of Doxie and Ruffin at the time of the rape, I am stunned. "You've got to be kidding me!" I tell her. "They look absolutely nothing alike!"

"You don't think so?" she asks sincerely.

Not only is Ruffin taller, thinner, and several shades lighter, but the two men bear no facial resemblance at all. Being black men is the only thing they have in common. As an African-American, I can see this.

But such mistakes are surprisingly common. In an analysis of 200 prisoners exonerated by postconviction DNA in this country (there have been 205 cases to date), all but 15 involved rape and the leading cause of wrongful conviction (79 percent) was eyewitness misidentification. (Other reasons were faulty forensic evidence, false informant testimony, and false confessions.) Of the 200 exonorees in the study, to be published in January's Columbia Law Review, 14 had been sentenced to death, and 50 sent to prison for life. Almost all the eyewitness errors were made by strangers. When black men are accused of sexually assaulting white women, the odds of such errors are much higher—about five times the rate of misidentification within the same racial group, according to Peter J. Neufeld, cofounder of the Innocence Project, which works to exonerate wrongfully convicted prisoners through DNA testing. The reason has to do with how we respond to trauma, says Gary Wells, PhD, an eyewitness identification expert and professor of psychology at Iowa State University: "Under stress, the brain goes into fight-or-flight mode and devotes most of its efforts to survival, lessening its ability to form clear memories. When people visually process their environment, they're actually taking in much less information than we ever thought." In Meng's case, still traumatized by her ordeal, she ran into a man similar enough to her attacker, and he gave her a friendly look. "What luck," she says, "to smile at someone who's just been raped."

At first, when Meng learned Ruffin was innocent, she was terrified. "I was so afraid of what he might do to me," she recalls. "He had every reason to hate me."

But when she watched him on television walking out of prison a free man and he was asked what he would say to his accuser, he responded: "I don't fault you. I don't blame you. It was just an honest mistake."