Jailed by Bad Science

Brandon Mayfield and his children outside a federal courthouse after his release from custody, Portland, Oregon, May 2004.
Don Ryan/AP Images
Brandon Mayfield and his children outside a federal courthouse after his release from custody, Portland, Oregon, May 2004. Mayfield was arrested after fingerprints on a bag found near the site of the 2004 Madrid bombings were mistakenly identified as his.

Anyone who watches crime shows knows that twenty-first-century police and prosecutors have at their disposal an array of modern forensic techniques—ways of analyzing hair, fibers, paint, clothing, firearms, bloodstains, and even bitemarks—that can “scientifically” establish guilt or innocence. Or can they? It has become increasingly apparent that most of these techniques are in fact unscientific, involve a great deal of guesswork, and too frequently result in false convictions. Of the more than 2,400 proven false convictions since 1989 recorded by the National Registry of Exonerations, nearly a quarter involved false or misleading forensic evidence.

The precursors of most of the forensic techniques used today were originally developed by police labs as helpful investigative tools, with no claim to being hard science. But starting in the first quarter of the twentieth century, the information obtained by use of these tools was introduced as substantive evidence in criminal cases by lab technicians (or sometimes ordinary police officers) portrayed as highly qualified “forensic experts.” These “experts,” few of whom had extensive scientific training, nonetheless commonly testified that their conclusions had been reached to “a reasonable degree of scientific certainty”—a catchphrase that increasingly became the key to the admissibility of their testimony in court. Such testimony went largely uncontested by defense counsel, who lacked the scientific and technical training to challenge it.

This began to change somewhat in the late 1980s, when DNA testing was developed by scientists applying rigorous standards, independently of the criminal justice system. It proved to be far more reliable in establishing guilt or innocence than any of the forensic techniques that preceded it. DNA testing not only helped to convict the guilty, but also led to the exoneration of hundreds of felons, many of whom had been convicted on the basis of faulty forensic evidence.

The leader in this has been the Innocence Project, founded in 1992 by Peter Neufeld and Barry Scheck at Cardozo School of Law. Using DNA testing, the Innocence Project has proved in court that more than 360 people (at last count) who had been convicted of crimes such as murder and rape (and had served an average of fourteen years in prison) were actually innocent. In over 40 percent of these cases, false or misleading forensic science was a major factor in the wrongful convictions. DNA testing, because it was so good, exposed how bad much other forensic evidence was.

Meanwhile, the Supreme Court in 1993 gave federal judges the responsibility to act as “gatekeepers” for the admissibility of scientific and…


This is exclusive content for subscribers only.
Get unlimited access to The New York Review for just $1 an issue!

View Offer

Continue reading this article, and thousands more from our archive, for the low introductory rate of just $1 an issue. Choose a Print, Digital, or All Access subscription.

If you are already a subscriber, please be sure you are logged in to your nybooks.com account. You may also need to link your website account to your subscription, which you can do here.