Holding your hands up and saying "we got things wrong" is a difficult thing to do. But that's exactly what the FBI and US Department of Justice have had to do this week, the initial results were released of an ongoing review of thousands of criminal cases in which FBI scientists' testimony may have led to wrongful convictions – including for some people now on death row.
In the 1980s and 1990s, forensic scientists routinely compared hairs under the microscope, looking for physical similarities that might indicate a hair recovered from a crime scene came from a suspect. This still happens today, but it is usually backed up by mitochondrial DNA testing.
Two years ago, the FBI launched an inquiry into 2600 convictions, including 45 death-row cases, from the 1980s and 1990s, looking for instances where FBI analysts may have exceeded the limits of science when presenting their conclusions in court – saying a hair matched to the exclusion of all others, for example, rather than simply saying it looked similar.
This week it was reported that of the 10 per cent of cases reviewed so far, the "vast majority" contained errors. As a result, 136 defendants, including two on death row, will receive letters informing them of their right to DNA testing as a means of proving their innocence. This is in addition to 23 letters that went out last year, including to 14 people on death row.
This could just be the start. Some think that the FBI's admission could trigger a landslide of similar cases involving state labs that also performed hair analysis – many of which were trained by the FBI. It could even ripple out to other areas of forensic science that don't fall under the current review.
Already, state labs in Texas, North Carolina and New York have launched similar reviews of convictions involving hair analysis. "I think it's just the tip of the iceberg," says Peter Neufeld, co-director of the Innocence Project in Washington DC, which fights to exonerate wrongfully convicted people through DNA testing. "There's no question that it's going to affect state labs as well."
Of the 317 DNA exonerations in the US so far, 75 of the original convictions included testimony from hair analysts – and of these, 70 involved experts from state labs, rather than the FBI lab. "Most had some kind of scientific errors in their testimony," Neufeld says.
"Nobody who attended any of the FBI classes was told to over testify, that's not part of the training," says Max Houck, a former FBI hair analyst who taught some of these classes and now heads the Department of Forensic Sciences in Washington DC. "However, our fear was always that we teach these people for two weeks, and they would go back to their laboratories with a certificate of completion and be told: 'Great you're qualified to do this – here's your caseload.'"
Yet hair analysis is just one of many forensic disciplines that hinge on using a microscope to visually compare two samples and declare a match. Ballistics, fibre analysis, tyre and shoeprint comparison and tool and bite-mark analysis all take a similar approach. All came under heavy criticism in a landmark report on the state of forensic science published in 2009.
"This review is likely to have an effect on any discipline where they didn't have a statistical reference to estimate the chances of another person being a match," Neufeld says. He believes it could even filter across to disciplines with a more robust statistical basis such as fingerprinting, but which have been exposed as flawed in recent years.
Of course, in many of the cases under review, there may have been additional evidence that might have seen the person convicted. "The review does not evaluate if a potentially erroneous statement was material or relevant to the conviction," an FBI spokeswoman told New Scientist.
Neufeld admits that we don't yet know how many convictions were gained on the back of erroneous statements. "In almost no case is a person convicted on just a single piece of evidence," he says. However, he adds that before DNA testing, a matching hair would often be a crucial point that prosecutors would emphasise in their closing remarks. "Combined with an unreliable eyewitness statement, it could create a perfect storm," he says.
If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.