By Lindsay Markel, Assistant Director, Justice Brandeis Innocence Project, Schuster Institute for Investigative Journalism, Brandeis University
In response to a last-minute appeal by attorneys representing Georgia death row inmate Troy Davis, the Supreme Court refused to stay his execution. But the controversy over his case -- and the issues it has raised, particularly about the reliability of eyewitness identification -- will continue. Initially, nine eyewitnesses identified Davis as a cop killer, but seven of them have since recanted. There was no forensic evidence. Davis was convicted largely because of eyewitness testimony.
HOW RELIABLE IS EYEWITNESS IDENTIFICATION?
Discover the science that says eyewitness identification is highly unreliable -- and write about DNA exonerations in your own state in which inmates were wrongfully convicted based on faulty eyewitness identification -- sometimes by several people who were simply wrong, as DNA tests have proved.
Earlier this week, the Georgia Board of Pardons and Paroles denied Troy Davis clemency in his death sentence, ending a hearing before eyewitness science expert Jennifer Dysart had a chance to testify about the failings of eyewitness memories and identifications.
What Dysart would have likely told the parole board is counterintuitive, but well documented: we don't always see what we think we saw.
"The Troy Davis case was staged--pure theater," wrote Virginia Law School professor Brandon Garrett, who has studied the first 250 cases of DNA exoneration. And he didn't mean 'staged' in the 24-hour news cycle sense -- "By 'staged' I mean that the eyewitness evidence at the core of [Davis's] original criminal trial was, quite literally, staged by the police," he wrote.
Experts who have extensively studied the reliability of memory and eye witness identification have concluded in a number of peer reviewed studies that eyewitness evidence is highly unreliable.
This is counterintuitive -- and most people, including some criminal justice officials, simply don't know enough about this research.
Experts say that the human memory is malleable, easily corruptible evidence, and that eyewitnesses' recollections should be treated with the delicacy of any other crime scene evidence. Around the country, that doesn't always happen.
Here, a resource guide for covering eyewitness misidentifications and wrongful convictions in your jurisdiction.
Mistaken eyewitness testimony is the single highest contributor in sending the wrong person away for someone else's crime. In fact, mistaken eyewitnesses have played a role in 75% of the 273 DNA exonerations we've seen pop up steadily around the country in the last couple decades since we've learned the power of DNA to exculpate the innocent--and put away the guilty. That means 187 men and women were put in prison for something they didn't do because someone -- a fallible human being -- made a mistake. In 36% cases, more than one person made the same mistake, identifying the same wrong perpetrator.
But according to Garrett, what's perhaps most worrying is the fact that these criminal cases were perfectly normal -- there was nothing particularly extraordinary about most of them, and yet the system got it wrong.
"That makes these cases all the more troubling," Garrett wrote in his book Convicting the Innocent (Harvard University Press, 2011). "If there is no reason to think that anyone acted differently in these cases as compared to others, then similar errors may have convicted countless other innocent people and led to the guilty going free."
What this all suggests is that, despite best intentions on behalf of law enforcement, prosecutors, and especially eyewitnesses and traumatized victims, there's probably someone in your own area who has also been the victim of a fallible system -- and there's a story to be done.
With the nation's attention on Troy Davis, the 43-year old Georgia man convicted for murdering a police officer, was executed Wednesday night. He maintained his innocence from the beginning, and 7 of the 9 eyewitnesses who fingered him have recanted their testimony, and one of those eyewitnesses could possibly be the true perpetrator.
As in Davis's case, problematic witness identifications happen in jurisdictions all over the country.
Take Eric Sarsfield, who was wrongly convicted in Marlboro, Massachusetts after an unsure victim identified him as her attacker from one-man-lineup in which he was forced to wear the perpetrator's jacket, which had been left at the scene. Sarsfield was exonerated by a DNA test 9.5 years later.
Or there's Antonio Beaver, who was convicted in Missouri after a witness described a perpetrator wearing a baseball cap and with a "David-Letterman-like gap between his teeth." After a four-man lineup in which Beaver was the only man with any visible defects to his teeth (and one of two men with a hat) the witness identified Beaver. DNA showed him to be innocent after he served 10 years.
Reporters around the country can use the Troy Davis case not only as a catalyst to write about such cases in their own area, but also to inform the public about the dangers of tainted identification procedures.
Here's an example of bringing the Troy Davis story home to Michigan, by Detroit Free Press columnist Jeff Gerritt, who wrote that Davis' case is "an injustice Michigan should remember."
To aid in your reporting, below is a curated list of reliable resources that reporters can use to cover this story and others surrounding this phenomenon.
Eyewitness reform factsheet:
The Innocence Project has laid out a set of "best practice" reforms for eyewitness identification procedures. These are based on decades of social and cognitive science research, as well as what the criminal justice system has learned from what went wrong in those 187 cases where a witness identified the wrong perpetrator. Here, the experts explain the science behind how witnesses can get it wrong.
Factors to be taken into consideration when evaluating eyewitness statements, according to Garrett, Convicting the Innocent: witness confidence at the time of the identification (not at trial), opportunity to view the attacker, degree of attention paid to the perpetrator, the passage of time since the crime, cross-racial identification which increases the likelihood of a mistaken identification, suggestive remarks from those administering identification procedures, stacked lineups which single out any particular suspect, and the age of the witness (children are especially vulnerable to suggestion).
Eyewitness reform in your state:
Enter your state name in this Google search to see how up-to-date on best practices your state or jurisdiction is, or if they're paying attention at all.
Innocence projects in your area:
For experts in your area, see this state-by state listing of innocence efforts. The directors of those projects will be able to give you more information about the biggest issues in your in your city, state or region.
There is a listserv for innocence project directors. To contact the list, email email@example.com and we will forward your request to this list ASAP.
Eyewitness misidentification in your state:
The Innocence Project has a map function which will allow you to isolate the DNA exonerations in your state, organized by the underlying cause.
For example, here's a listing of Massachusetts exonerees who were wrongly identified by eyewitnesses:
The data on eyewitnesses:
Gary Wells is a leading expert in the area of eyewitness identification. His federally funded studies explore the relationship between memory and errors in identifications. Wells's website is a treasure trove of information on eyewitness science, and he has written prolifically on the subject.
Federal guide on eyewitness procedures:
The Justice Department, through the National Institute of Justice, published a trainer's manual for law enforcement dealing with eyewitness identifications: "Eyewitness Evidence: A Trainer's Manual for Law Enforcement" (Sept. 2003).
Their general manual, "Eyewitness Evidence: A Guide for Law Enforcement" was published in Oct. 1999. Both guides were developed by the Technical Working Group for Eyewitness Evidence (TWGEYEE), a multidisciplinary group of content-area experts from across the United States and Canada created to develop recommended procedures for law enforcement use in investigations involving eyewitness evidence.
A case study: The Ronald Cotton case:
One of the better known, but most poignantly told victim misidentification stories is that of Jennifer Thompson and Ronald Cotton. In 1984, a man broke into then-college student Jennifer Thompson's apartment. While he raped her, she focused all her energy on noticing everything about her rapist so she could escape and help put him away.
"I would do everything in my power to memorize my attackers' face," she said. "Throughout the course of the rape, I struggled to look into the face of the person who was destroying my life. I needed to know what his hair looked like, was his skin dark or light. Did he have any scars, tattoos, piercings, things he could not alter later? His voice, his age, weight and height all mattered."
Then she identified Ronald Cotton, and he went to prison. But Cotton didn't do it, and DNA proved it 10 and a half years later. In an amazing example of forgiveness and compassion, Thompson and Cotton are now friends, co-authors of the book Picking Cotton, and vocal advocates of eyewitness identification procedure reform.
 That's 36% of the first 250 exonerations which Brandon Garrett examined in depth in his book Convicting the Innocent. In real numbers, that means that about 90 innocent people were wrongly identified by at least two mistaken eyewitnesses.
 Testimony of Jennifer Thompson to the Maryland Commission on Capital Punishment September 5, 2008
CONTACT: Lindsay Markel
Assistant Director of the Schuster Institute for Investigative Journalism
and the Justice Brandeis Innocence Project at Brandeis University