DNA evidence at heart of recent exoneration cases

RICHMOND, Va. -- Ann Meng was at work when she saw a man who made her blood run cold.

It was him. The man who had broken into her home and raped her. There was no question in her mind.

She called the police and later picked him out of several lineups. Julius Ruffin was convicted of the attack in Norfolk in 1982 and sentenced to five life terms in prison.

"When I looked at Julius Ruffin, I was sure -- 100 percent sure," said Meng, now 56. "I had flashbacks when I looked at him."

But Meng was wrong; Ruffin was innocent. And so were four other Virginians convicted of separate rapes based on faulty victim identification and cleared in recent years thanks to DNA testing.

The pattern of incorrect identifications in the five recent exonerations is not unique. Experts say mistaken eyewitness identification has been a factor in about three-quarters of the 170 wrongful convictions overturned thanks to post-conviction DNA testing.

"Traditional eyewitness identification procedures are prone to witnesses ... misidentifying the actual suspect," said Stephen Saloom, policy director of the legal-aid group the Innocence Project. "Because of the compelling nature of an eyewitness identification, it can have a tremendous impact on the direction of a police investigation."

Ruffin, Marvin Anderson and Arthur Lee Whitfield, all convicted of rapes in the 1980s, were cleared after evidence saved by the late forensic scientist Mary Jane Burton was found and DNA testing revealed they were not the perpetrators. They served a combined 59 years behind bars.

Even before DNA testing had been invented, Burton, who worked in the Virginia state crime lab from 1974 to 1988, saved bits of the evidence she tested and inserted them into case files. They were discovered in a storage facility in 2001.

On Dec. 14, Gov. Mark R. Warner announced that two additional men were exonerated thanks to the samples. Warner has now ordered a review of all Burton's samples.

Willie Davidson, released in 1992 after 12 years in prison, was convicted largely because of faulty identification by the victim, a 66-year-old woman who had served as a grandmotherly figure to him, said his attorney, James Broccoletti.

Davidson, now 49, lives with his mother, and works in a low-income job, his attorney said.

"He was in hell for 12 years and purgatory for 13," Broccoletti said. "He's been denied opportunities in life ... jobs, housing opportunities, friends have shunned him, neighbors have looked at him. ... He's spent more than half his life tainted in this shroud of guilt."

Phillip Thurman, 50, was convicted in 1985 of raping, abducting and assaulting an Alexandria woman. The victim and a witness both incorrectly identified Thurman as the perpetrator. He was sentenced to 31 years in prison and paroled in February.

Prosecutors said DNA tests on the samples in Meng's case revealed a convicted rapist, Aaron Doxie III, was likely responsible for her attack. Doxie is already serving three life sentences for a separate sexual assault. Prosecutors did not pursue charges against him because Burton and other key witnesses are dead.

The reliance by juries on eyewitness identification is dangerous because it is so often flawed, said eyewitness identification expert Gary Wells, an Iowa State University psychology professor.

"When people visually process their environment, the fact is that they're actually taking in much less information than we ever thought," Wells said. "We have this type of evidence here that is at once easily prone to error, and simultaneously highly persuasive to juries."

The detail in recollections of crime witnesses can be affected by poor viewing conditions, limited time with the perpetrator and fear, Wells said. In stressful situations, the brain goes into "fight or flight" mode and devotes most of its efforts to survival, lessening its ability to form clear memories.

Conventional police lineup procedures, in which witnesses are shown groups of people or photographs at one time, can add to the problem. Many researchers believe this can lead witnesses to compare one person's image to another and identify someone who looks "most like" the offender instead of "exactly like."

Lineups administered by officers who know the suspect is can also be dangerous, as officers can inadvertently give cues as to who they want the witness to pick, Wells said.

Wells developed a lineup technique called the "double-blind sequential" method, in which witnesses are shown individual mug shots or people one at a time by an officer who does not know who the suspect is. The technique is being adopted by a growing number of police departments nationwide.

Advocates of the technique say it can help reduce the tendency to compare and settle and eliminates the possibility of an officer unintentionally influencing a witness' choice in a lineup.

In 2001, New Jersey became the first state to adopt the technique. A handful of jurisdictions -- including Boston, Minneapolis and Santa Clara County in California -- also have adopted them. Virginia Beach has used the technique for several years.

Virginia's General Assembly this year passed legislation requiring law enforcement agencies to have written policies for conducting lineups and the Virginia State Crime Commission asked law enforcement agencies to adopt the double-blind sequential procedure.

Workers at Virginia's Department of Forensic Science have already begun sifting through thousands of files for Burton's samples. The testing will be conducted by an independent lab and may take up to two years to complete, said director Paul Ferrara, who expects more exonerations.

"It's hard to imagine that there won't be more -- probably many more," he said.

That's all the more reason the justice system needs to address the issues surrounding its reliance on eyewitness testimony, said Meng, who has become an advocate for the wrongly convicted.

Prosecutors said DNA tests on the samples in Meng's case revealed a convicted rapist, Aaron Doxie III, was likely responsible for her attack. Doxie is already serving three life sentences for a separate sexual assault. Prosecutors did not pursue charges against him because Burton and other key witnesses are dead.

While The Associated Press does not typically identify victims of sexual assault, Meng has spoken publicly about her ordeal and requested that her name be used.

"I still think about it a lot, but I'm the kind of person that tries to take a lesson and tries to be part of the solution," Meng said. "We need to build into our justice system some more controls so this won't happen."