Criminal justice system based on outdated practices?

It's hard to think of anything that technology or science has changed more robustly or where the potential for improvement is greater than in the criminal justice world. Over the course of more than a century, police investigations and court cases have benefited from a long series of gains, most of them courtesy of scientific advances that bring improved knowledge or technology or both.

Fingerprint analysis. DNA. Blood typing. These and other tests help sort through crime scene evidence and chime in on who is responsible when bad things happen. But like all tools, the results are only as good as the skills being wielded to use them. It's not like television, where universal truth is black and white and readily discernable in a guaranteed-accurate lab report. Some forensics experts really are better than others. And there are, in fact, variations in standards and training and natural aptitude to do analysis, in what different states require, in how they interpret results, in how findings can be used and more.

The only thing absolutely certain on this topic is that it's hard to be absolutely certain on this topic. Without question, each year some individuals are wrongly convicted based on interpretation of forensic evidence. The National Registry of Exonerations, a joint effort of University of California Irvine, Michigan State University and University of Michigan law schools, counts 2,021 people convicted of crimes who have subsequently been exonerated. There likely are other innocents who never had legal grounds to challenge their convictions.

We're at an interesting time in the evolution of forensic knowledge and analysis, and the stakes are a bit breathtaking: For someone who might be wrongly accused, it's potentially a loss of freedom. For communities, the ability to find and incarcerate a criminal might have major ramifications for safety or other measures of public well-being.

Americans should pay attention to recent events because we all have a stake in how well our criminal justice system works.

- U.S. Attorney General Jeff Sessions has decided that the Department of Justice will get rid of its nonpartisan National Commission on Forensic Science. Erin E. Murphy, a New York University School of Law professor, noted in a New York Times column that Sessions also dumped ongoing "review of closed cases for inaccurate or unsupported statements by forensic analysts, which regularly occur in fields as diverse as firearm and handwriting identification, and hair, fiber, shoe, bite mark and tire tread matching, and even fingerprinting analysis."

- Differences in standards continue to be challenging, including uneven training and different jurisdictional rules. There's even a word-choice issue. When someone testifies that it's "virtually impossible," "highly likely" or "fairly certain," for example, what does that even mean unless there are standards on which a judge or jury can rely? Efforts to establish standards not just in word choice, but in how experts look at evidence, appear to be stalling or even reversing.

- Not all technology analyzes what happened. In Wisconsin a software algorithm called Compas helps judges decide sentences based on its prediction of how someone who is convicted might behave. Its proprietary (that means no one gets to know what it includes) algorithm predicts future actions. The Wisconsin Supreme Court has been asked to look at it, after the algorithm predicted a defendant posed "a high risk of violence, high risk of recidivism, high pretrial risk" and that conclusion was a factor in his sentence, although neither he nor his lawyer was privy to the report or how the conclusion was reached.

As hard as it is to prove reliably what has happened, it seems harder still to vet the future.

It all makes me nervous. I'm well past believing that someone who hasn't done anything wrong has nothing to fear. We've seen too much evidence to the contrary. It's worth fighting to get it right, because that's the only time that justice is actually justice.

Email: lois@deseretnews.com

Twitter: Loisco

Credit: By Lois M. Collins Deseret News