When a former high-ranking Justice Department official speaks of a “revolution” in criminal justice, with the whole field turning toward science, could it mean less failed evidence in the future? What does it mean for those concerned with faulty forensic science?
Laurie Robinson served as Assistant Attorney General in both the Clinton and Obama Justice Departments, where she oversaw the Office of Justice Programs (OJP), the research, statistics and criminal justice assistance arm Justice. That made her remarks to the Delaware Center for Justice the other day worth noticing. According to the Wilmington News Journal, “[w]e’re seeing something akin to a revolution in criminal justice in this country,” said Robinson, now on the faculty of George Mason University. “We’re at an important crossroads, one where ideology has taken a back seat, and science and pragmatism have come to the fore.”
Robinson’s web page at George Mason says her tenure at OJP “was marked by a focus on science and evidence-based programming.” She was in Delaware to discuss the state’s re-entry programs and other initiatives to reduce recidivism, and few would disagree that those important programs need scientific and statistical support. But I wonder whether Robinson would be as optimistic about science’s role in forensic methods, which have played a role in about half of all wrongful convictions across the U.S. Surely, forensic science needs “science and evidence-based” support and examination — badly.
It has now been more than four years since the release of the landmark 2009 National Academy of Sciences report Strengthening Forensic Sciences in the United States: A Path Forward, which found that except for DNA and chemical analysis, most of what we think of as forensic science isn’t science at all. In that time, little seems to have changed; scandals in crime labs continue to pile up in jurisdictions across the country (see my posts about lab scandals just this past year in Massachusetts and Minnesota, for example). In a particularly compelling piece of writing in the Huffington Post, Radley Balko discusses the long-running crime lab scandal in Mississippi, and puts it in context: Mississippi’s scandal “is just the latest in a long, sad line of such stories” that Balko has already chronicled.
What’s to be done? As a start, I argued in chapter 7 of Failed Evidence, all federal money that goes to law enforcement should carry with it a requirement for compliance with best practices in police investigation and forensic science. Not all law enforcement agencies run their own forensic labs (and as the NAS report said in Chapter 6, “Improving Methods, Practice, and Performance in Forensic Science,” labs should be independent of law enforcement.). But for those that do, compliance with standards that would avoid systematic error, human biases, fraud, and improper scientific testing and testimony should be mandatory.
That would start a revolution right there. Because what law enforcement agency could afford to just turn down federal funding, in budgetary times like these?
Failed Evidence 
The problem with much of the research, particularly in the area of psychology is that it purports to be science when it is simply pseudoscience despite the fact that it is based upon ‘research’. For example in the area of detection of deception facial action coding systems claim to detect deception but in fact are only able to detect stress which often is not associated with deception. There are no distinguishing expressions of deceivers or liars. Perhaps this is why Dr. Ekman and associates do not want to test their assertions in court. The problem is that we are being duped into believing ‘experts’ can shape social behaviour and security, often to the detriment of the public. Science in the area of human behaviour is fallible. Rue the day that it sways our courts.