Unreliability of Experts

Forensic scientists had a strong tendency to overestimate their ability to draw correct conclusions from evidence.

Erik Eckholm reports that expert testimony about bite marks, considered trustworthy enough to send people to prison, has turned out to be extremely unreliable. Mark Cheney was sentenced to life for murder "after a dental expert testified that it was virtually certain that his teeth had caused marks on an arm of the victim, a drug dealer who was stabbed to death." After 28 years in prison, he was recently freed because "studies have shown that dental experts cannot reliably claim that a bite wound was caused by a particular individual. They cannot even consistently agree on whether wounds were caused by bites at all."

Forensic science more broadly is in turmoil as prosecutors, defense lawyers and judges confront evidence that many long-used methods, like handwriting analysis and microscopic hair comparisons, were based more on tradition than science and do not hold up under scrutiny. Even fingerprint and certain kinds of DNA matches are not quite as certain as many once believed, scientists say. ...
Put under rigorous scrutiny, some forensic tools, including comparisons of lead chemistry in bullets and the matching of aural voice prints, have already been largely discarded. The accuracy of many supposed signs of arson, like burn patterns that seemed to be caused by a liquid, has also been disproved.

This year, the Federal Bureau of Investigation admitted that examiners at its vaunted crime lab had for decades overstated the reliability of microscopic hair matches.
- Erik Eckholm, Lives in Balance, Texas Leads Scrutiny of Bite-Mark Forensics, NYT, Dec. 12, 2015

Show php error messages