Forensic science: validation and incompetence

By Anna Sandiford 12/02/2010


U.S. academics at UCLA are being granted funds to consider error rates in latent fingerprint evidence.  Some would ask whether or not this sort of exercise should have been completed loooong ago.  As with any other area of applied science, regular review should be undertaken.  Unfortunately, this is not something that necessarily occurs in forensic science, partly because some agencies aren’t keen on their databases being examined (see
Crime DNA databases should be independently examined
).

Last year, the United States National Academy of Sciences issued a damning report on the state of forensic science in U.S. crime laboratories.  The report basically stated that the accuracy and reliability of practically all forensic science methods, ranging from glass to fingerprints, had not been established adequately through rigorous scientific scrutiny.  At the end of this month (Feb 2010), the American Academy of Forensic Sciences (AAFS) is holding its annual scientific meeting, which is tellingly entitled, Putting Our Forensic House in Order: Determining Validation and Expelling Incompetence.’

So, not only is forensic science in the States being shaken upside down until the grotty bits drop out of its pockets, someone somewhere is getting paid to do the work that should have been done long ago, on an ongoing basis. Let’s hope one of the resolutions that arises from the AAFS meeting is that forensic science techniques should be reviewed regularly. The problem is of course that most people working for the prosecution don’t really want any technique investigated too deeply in case problems are found. Once that happens then other convictions might be called into question, which could throw the whole system into disorder, incurring enormous expense and all the obvious associated problems.

To my mind, that’s not a good enough reason not to do it.  If people are going to be sent to prison based, even in part, on scientific findings then the science must be robust and reliable (as well as other legal issues such as relevant and repeatable).  If science is reviewed regularly and the law takes that into account then it should be possible to work out a system whereby the courts can be sure that the science is up-to-date, which in turn adds to the strength of science in court.  It also might present the current stink that’s going on in Texas over the inadequate forensic science presented in the case of Cameron Todd Willingham (Wikipedia link), who was executed for allegedly killing his children in a house fire he was supposed to have set.  The science has since been shown to have been wrong.

The other thing to remember is that if the overall outcome of the US review and the work being carried out to address those issues identifies some real problems, the implications could be felt throughout court systems worldwide — including New Zealand.