March 16, 2009 — Flawed testimony by forensic experts contributed to the conviction of innocent defendants, according to a new study co-written by University of Virginia Law School professor Brandon Garrett.
The findings are featured in an article, "Invalid Forensic Science Testimony and Wrongful Convictions," published in the March issue of the Virginia Law Review. The article will be available today on the Social Science Research Network, and on the Virginia Law Review.
Garrett and Peter Neufeld, co-director of the Innocence Project, studied the transcripts of 137 trials in which prosecution forensic analysts testified, and the defendants were exonerated years later by post-conviction DNA testing.
The pair found that in 60 percent of those wrongful conviction cases, forensic analysts gave "invalid testimony that overstated the evidence," Garrett said.
"What we mean by 'invalid' is simply that the testimony was unscientific or contrary to empirical data," he said. "Just because a wrong statistic was offered, does not mean that the testimony necessarily caused the wrongful conviction. However, these powerful examples support efforts to adopt and enforce scientific standards governing forensic reporting and testimony."
The flawed testimony uncovered by Garrett and Neufeld included erroneous or unsupported testimony about the accuracy and results of forensic techniques, including hair comparison, bite-mark comparison, fingerprint comparison and even DNA testing.
The study originated with a request to Garrett — who conducted previous research on wrongful convictions — from a National Academy of Sciences committee examining the needs of the forensic science community, asking him to present at one of the committee's public hearings.
Garrett and Neufeld then spent more than a year compiling and analyzing trial transcripts from the cases of people later exonerated by DNA evidence. Several scientists and forensic scientists also reviewed the categories used for analysis and examined transcripts in particular cases.
The majority of the cases were rape cases from the 1980s, and many included testimony about forensic techniques that are still used today, said Neufeld, co-founder of Innocence Project, a national litigation and public policy organization that uses DNA testing to exonerate wrongfully convicted people and seeks to reform the criminal justice system to prevent injustice.
The National Academy of Sciences report, "Strengthening Forensic Science in the United States: A Path Forward," was released in February, and recommends the establishment of a national institute of forensic science, an independent scientific entity to adopt and enforce standards for forensic report writing and testimony.
"With the exception of nuclear DNA analysis ... no forensic method has been rigorously shown to have the capacity to consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual or source," the National Academy of Sciences report says.
In their study, Garrett and Neufeld found that forensic analysts often testified that a particular piece of evidence — such as a hair or a fingerprint — was closely connected to the innocent defendant, despite the fact that no scientific data permitted analysts to reach such conclusions.
For example, in one case an analyst told a jury that only 5 percent of the population had a certain type of hair pigment discovered at a crime scene, and that the defendant was among them. But there is no empirical data about the frequency of particular hair pigments, Garrett said.
"These trial transcripts were fascinating to read, because in retrospect we know that all of the defendants were innocent," he said. "Yet few have looked at these records. Even after these wrongful convictions came to light, crime laboratories rarely conducted audits or investigations to review the forensic evidence presented at the trial."
The study doesn't speak to the state of mind of analysts; it's impossible to tell from trial materials whether they were simply inexperienced or poorly supervised or acting in bad faith, Garrett said.
"Nor do we know how many cases involved concealment of forensic evidence," he said. "In 13 exonerees' cases it has since come to light that forensic evidence was concealed that would have helped to prove innocence. Similarly, we do not know from reading trial transcripts in how many cases errors were made in the laboratory, although in a few exonerees' cases in which the underlying evidence was re-examined, gross errors have come to light."
The study also notes that the criminal justice system is not well-suited to prevent unscientific testimony. Reasons include that the presentation of forensic evidence is almost entirely one-sided, Garrett said.
"Defense counsel rarely cross-examined analysts concerning invalid testimony and rarely retained experts, since courts routinely deny funding for defense experts." Only 19 of the eventual exonerees whose cases were examined had defense experts. "Prosecutors, moreover, presented erroneous accounts of the forensic evidence during closing arguments," Garrett said.
The study's authors agree with the National Academy of Sciences report's assessment that a set of national scientific standards should be established to ensure the valid presentation of forensic analysis. Neufeld called the report "a major breakthrough toward ensuring that so-called scientific evidence in criminal cases is solid, validated and reliable."