Science plays a growing role in crime detection and prosecution, but experts at four universities say a lot can go wrong in the lab, and many people may be wrongly convicted based on bogus claims. Now, the University of Virginia has launched a blog to share research on the subject.
The analysis of DNA and hair samples, footprints and bite marks may all be used in court to convict people of crimes they did not commit. Even fingerprints can mislead according to Law Professor Brandon Garrett.
“Watching a lot of CSI, you might think fingerprints – that’s really old-fashioned. Instead what we found was that people really thought fingerprinting was perfect, and it never occurred to them that there are a lot of steps in the analysis, and that the fingerprint you find at a crime scene may be smeared and deteriorated, It didn’t matter what was aid in the courtroom. If there was a fingerprint match, the jurors thought it was great evidence of someone’s guilt.”
In fact, he adds, fingerprints and many other forms of evidence are subject to interpretation by people who can be wrong, and what’s said in the courtroom isn’t always true.
“You can’t say with a fingerprint that one in a million people would have this unusual marking on their finger or one in ten. We don’t know. There are not population data bases, and the same is true of bite marks and tool marks and fiber comparisons.”
Even the Federal Bureau of Investigation, seen as a leader in law enforcement technology, is backing away from some prior claims.
“The FBI has just been re-opening huge numbers of old cases involving hair testimony. Instead of just saying, ‘We looked at these hairs under the microscope. They’re similar,’ they would make claims that, ‘It could only have come from the defendant.’ There was no scientific basis for saying stuff like that. There are number of states that have responded to this by saying, ‘We have a lot of old hair cases too, and all of our hair people were trained by the FBI, so you now have Texas, Massachusetts and two other states re-opening old hair cases, but then you have to wonder, ‘Well how about Virginia?’ I’ve read wrongful conviction cases in Virginia with overstated hair testimony. Texas is doing an audit of all the old bite mark cases, because there are so many questions raised about just how flimsy bite mark analysis is.”
And thousands of cases in Texas and other states are being re-opened because of errors in the way the statistics are calculated for DNA mixtures.
That’s why UVA has joined forces with Iowa State, Carnegie Mellon and UC Irvine to form the Center for Statistics and Applications in Forensics Evidence or C-SAFE. With a $20 million grant from the federal government, they’ll promote research and education to improve the accuracy of forensic evidence – how it’s processed in the lab and explained in the courtroom. To spread the word of its findings, UVA has launched a blog – Forensics Forum – and plans to organize talks and conferences on the subject.
