How reliable is the forensic evidence collected and analyzed by crime scene investigators? Forensic science is an essential part of the criminal justice system and a staple of American TV culture. It’s easy to see the appeal. Clever investigators whose expert eye and powerful techniques for collecting and analyzing data compel powerful inferences about right and wrong. But PBS Frontline’s “The Real CSI” offers a very different picture. Through a joint investigation with ProPublica and drawing on a 2009 National Academy of Science study, Frontline looks at evidence of the reliability of forensic analysis, and finds a number of trusted techniques like fingerprinting and hair identification are much less certain than often assumed. To anyone who tries to apply the standards of scientific inference, the uncertainty characterizing forensics should be no surprise. Science is a messy often ambiguous business, which assumes lots of work will go down “blind alleys” and the most compelling inferences are often frustratingly uncertain.
The criminal justice system demands evidence that is decisive and will ultimately support a criminal case beyond a reasonable doubt. Scientists, by contrast, are in the business of reasonable doubt. Investigators and prosecutors cannot afford scientific uncertainty. In recent decades, the gaps between these ill-fitting institutions has been filled by specialists who acquired the attributes of science – illustrated by the empty credentialing process for forensic experts – while actually obscuring the uncertainty essential to meaningful scientific inference. A similar problem applies across many other fields – this is why management consultants tend to be such hucksters – but when theories like shaken baby syndrome or fingerprint identification are falsified the human consequences can be enormous. Another of many distressing examples is presented in this podcast based on a 2009 New Yorker article by David Gramm suggesting faulty arson investigations expertise resulted in a Texas man’s wrongful execution.
Lastly, more of a question, I wonder about the potential consequences for status of experts in American society. Again, the human consequences of bad forensic science are real. But, related and perhaps more importantly, because it makes good television drama, Americans otherwise skeptical of expert knowledge have embraced the forensic sciences. My unscientific study of these shows says the forensics expert often embodies the very essence of scientific professionalism – competent and committed to truth. Regardless of how courts and politicians respond to the mounting case for more realistic ways of thinking about forensic evidence, what happens when people realize the vaunted forensics specialist may be as much of a huckster as the management consultant they despise?