Flawed FBI forensic testimony comes back to haunt US criminal justice system

Before DNA testing, FBI forensic hair analysts often gave faulty testimony, including in many death penalty cases. These mistakes have come back to haunt the justice system. Spencer Kimball reports from the US.

Deutsche Welle, 21 April 2015

Innocent until proven guilty. This principle is the cornerstone of the US criminal justice system. But what happens when an innocent defendant is found guilty based on faulty "evidence"? It's a common problem in America. More than 300 people have been exonerated by DNA testing nationwide since 1989. On average, they served 14 years in prison before being freed.

Half of these wrongful convictions were the result of bad forensic evidence, and more exonerations could be on the way. The Federal Bureau of Investigation (FBI) has admitted that 26 of its 28 hair analysis experts gave faulty testimony in at least 90 percent of the trials scrutinized in a recently completed review. That includes 32 trials in which defendants were sentenced to death. Nine have already been executed and another five died of other causes on death row.

The FBI conducted its audit in cooperation with the Innocence Project and the National Association of Criminal Defense Lawyers (NACDL), focusing on cases before DNA testing became a routine method of confirming or rejecting hair matches. In 2012, the Washington Post published an article questioning the reliability of hair matches, prompting the review. According to Brandon Garrett, a professor at the University of Virginia School of Law, the results are not surprising.

"A mass disaster has been brewing for years," Garrett, author of "Convicting the Innocent: Where Criminal Prosecutions Go Wrong," told DW via email. "When I examined the trials for the first 250 DNA exonerees, I found similarly invalid and unscientific testimony in over one half of the cases."

Overconfident analysts

According to the National Academy of Sciences, the problem with forensic methods such as microscopic hair analysis is that they ultimately rely on an individual judgment about whether a specimen matches a specific person. There's no standard protocol established by science to determine with certainty when there's a match.

More than 300 people have been
 exonerated by DNA testing nationwide
since 1989

"The simple reality is that the interpretation of forensic evidence is not always based on scientific studies to determine its validity," the report said. "This is a serious problem."

David Kaye, an expert on forensic evidence at Penn State University, believes that hair analysis can be useful in conjunction with other evidence. The problem is that analysts in the cases under review didn't qualify their testimony by saying that their conclusions were subject to human error and weren't definitive.

"Prosecutors, if they're the ones introducing the evidence as they usually are, want it stated as strongly as possible," Kaye told DW.

"You've got a laboratory in the FBI working with a prosecutor from the Department of Justice in which the FBI sits - they have a relationship," he said. "You want to please people. It's easy to see how people could be overconfident."

Post-conviction relief not guaranteed

Despite the FBI's flawed testimony, Kaye believes that most of the cases are unlikely to be overturned. There may be other evidence of guilt, and if the defense didn't object to the hair analysis during the original trial, the grounds for challenging the conviction could be weakened.

"The FBI, the Innocence Project, the National Association of Criminal Defense Lawyers are going around telling the old defense lawyers here's a problem with the testimony, but whether they're going to get relief from that is very unclear," Kaye said.

The problems persist today, according to Garrett with the University of Virginia School of Law. Though hair comparisons have become less common with the advent of DNA testing, analysts also use microscopic comparisons for ballistics, bite marks, fibers, tool marks, shoe prints and even fingerprints.

"DNA testing can only be done in a small percentage of criminal cases," Garrett said. "In far too many cases, unscientific subjective comparisons are used."