Scientists are accustomed to a landscape of shifting knowledge, while the justice system prizes certainty. But the criminal justice system, including forensic science, increasingly has been called upon to re-examine past certainties and to revisit what they once believed to be incontrovertible facts.
Take “comparative bullet lead analysis,” a technique used since John F. Kennedy’s 1963 assassination that “matches” a bullet or bullet fragments to a box or batch of ammunition manufactured at a specific time and place. The chemical analysis behind it hinged on what proved to be a false premise: that each batch of lead had a unique elemental makeup. In 2005, the FBI voluntarily abandoned the technique and every past conviction that conceivably turned on the analysis had to be reviewed.
The conviction of Jimmy Ates was the first thrown out after the FBI’s decision. A jury had found Ates guilty of the 1991 murder of his wife, Norma Jean; she had been shot seven times and their home was set on fire. For years prosecutors wrestled with what they felt was a lack of evidence. When they eventually tried Ates in 1998, among the evidence they did present was comparative bullet lead analysis.
Prosecutor Rod Smith’s closing argument at Ates’ trial typified trust in the analysis at that time: “Of all the millions and billions of bullets that are made by any given company in any given time frame, the bullets that killed Norma Jean were manufactured from the same batch that were found in the back room [of their home].”
Righting What’s Wrong in Criminal Justice
Wrongful convictions stem from the belated entrance of scientific rigor into the field of forensics, systemic problems, and the ubiquitous ‘human factor.’ In the coming weeks, a series of stories by crime author Sue Russell looks at why convictions go wrong, at the common reluctance to rectify error, and at innovations to better safeguard justice.
Stories so far:
The rejection of comparative analysis was one of several reasons the state conceded that Ates—a client of Innocence Project Florida—was entitled to a new trial. He was released from prison in December 2008 after serving 10 years. (His case did not have a storybook ending: In March 2011, the jury at his second trial again found him guilty and he is serving life without parole.)
Three bullet-analysis-centered verdicts were overturned between 2005 and January 2010 as the FBI’s review of all potentially affected cases slowly progressed.
“Now, I would have preferred that they didn’t do the review themselves but had an independent person do it,” says Barry Scheck, cofounder and co-director of the Innocence Project and a professor at Benjamin N. Cardozo School of Law at Yeshiva University, “but at least they did it.” He was particularly gratified to have his FBI contacts concur that “as a legal matter, the FBI laboratory had a duty to correct. They agreed with that principle.”
THAT THOSE IN CRIMINAL JUSTICE have a “duty to correct” is a phrase that Scheck finds himself invoking ever more frequently.
It’s a simple concept: if any lawyer puts on testimony in a U.S. court and later learns that the witness made a mistake, it’s their duty to correct it. “You must get up and say, ‘My witness made a factual misstatement,’” says Scheck. “You have an ethical duty to correct during the trial, after the trial, whenever it comes to your attention. The same thing, I maintain, is true for forensic scientists. If the forensic scientists realize that their prior reports are no longer correct, they must call the prosecuting authorities and tell them, ‘We can no longer stand behind the testimony of that analyst who previously gave evidence.’”
The Model Rules of Professional Conduct from the American Bar Associationalso require prosecutors to act to rectify wrongful convictions and to follow up on “new, credible and material evidence” that could suggest innocence. It’s their duty to disclose it promptly to the court or appropriate authority and to the defendant, and to “undertake further investigation, or make reasonable efforts to cause an investigation, to determine whether the defendant was convicted of an offense that the defendant did not commit.”
Cameron Todd Willingham was executed in 2004 for the 1991 arson murder of his three young children. The arson evidence against him was later debunked.
Scheck invoked the principle while pushing the Texas Forensic Science Commission to review the case of Cameron Todd Willingham. Willingham was executed in 2004 for the 1991 arson murder of his three young children after being convicted on debunked arson evidence.
The commission never was expected to exonerate Willingham posthumously, only to see if the original fire investigators had been negligent. Nine top fire experts were unequivocal in saying that the original analysis at the time of the fire was wrong and that there was no evidence of arson, so they had a duty to correct their testimony. But the then-fire marshal, Paul Maldonado, told the Texas commission that his office would stand by the original investigators’ roundly condemned analysis.
Maldonado’s stance was laughable, says arson expert John Lentini, given how outlandish the investigators’ claims were when held up to the industry standards prevalent even in 1991. Still, defenders of the original investigators argue that National Fire Protection Association’s manual of standards introduced in January 1992 was new when Willingham stood trial and the investigators didn’t know about it.
“Well,” says Scheck, “they should have known about it.” At some point after Willingham was convicted and before chemist Gerald Hurst’s last-ditch pre-execution report argued that there was no legitimate evidence that the Willingham fire even was arson, “Surely the Texas fire marshal knew that this was no longer valid science? And as soon as they realized that, our contention was they had a duty to correct. That’s exactly how we laid the issue out for the commission. They had a duty to correct and they failed to meet it.”
In 2010, Tarrant County medical examiner Nizam Peerwani—then a commission member and now its chair—announced he was calling for a review of all deaths related to arson in his own jurisdiction going back 20 years. It is now underway. “There is no legal requirement for retroactive review,” Peerwani said. “But there is some moral and ethical duty to do that.”
Last April, the commission’s report took a critical step forward by stating that there is a duty to correct among forensic scientists. The duty has been accepted by other jurisdictions, such as New York’s Forensic Science Commission, and says, Scheck, “I think this will become generally accepted in the forensic community.”
Dallas County District Attorney Craig Watkins honors it. In January 2007, during his first week in office, Watkins saw a man exonerated. “I attended the actual innocence hearing and apologized to the man for the injustice he had endured,” Watkins recalls. “The media made a big deal over the elected district attorney apologizing for a wrongful conviction. I didn’t really understand why that was such a big deal because it just seemed like the right thing to do.”
According to both Watkins and the Texas Code of Criminal Procedures, it’s all about seeking justice, not convictions. Watkins promptly posted copies of Article 2.01 of the Texas Code of Criminal Procedure in assistant district attorney workrooms: “It shall be the primary duty of all prosecuting attorneys, including any special prosecutors, not to convict, but to see that justice is done.” They are words to live by for Watkins, who says “while that usually means convicting those who are guilty of committing crimes, it also means freeing the innocent.”
Texas passed a law allowing inmates to request post-conviction DNA testing in 2001. But pre-Craig Watkins, most requests were denied. “Couple that,” he says, “with the fact that at the time, our office had more than 400 claims of post-conviction innocence waiting to be reviewed. I began to think that perhaps the problem was much more widespread than we previously thought.”
He and his then first assistant Terri Moore took a bold and controversial step in establishing the Conviction Integrity Unit, or CIU, “the first of its kind in the country at a prosecutor’s office.” The prosecutors also partnered with the Innocence Project of Texas to review claims of innocence, eventually expanding that partnership to include other innocence groups.
Various innocence groups have brought cases to the Dallas County CIU’s attention for possible reinvestigation but his team, says Watkins, has done all the “critical leg work.” His office’s subpoena power and resources “far exceed what a defense attorney or advocacy group can actually do on a case. So it really is the district attorney’s responsibility to reinvestigate a legitimate innocence claim if a person has been wrongfully convicted.”
Florida’s Innocence Commission studies wrongful conviction causes and potential preventative measures and can review cases once a wrongful conviction has already been established. The Connecticut Advisory Commission on Wrongful Convictions also reviews cases and recommends reforms. But with common causes of wrongful convictions now clearly established, the next challenge is to move beyond conducting further similar studies to concrete action. Experts suggest that with limited resources available, the focus should now be on reviewing cases and on formulating specific proposals for reform.
There are many suggestions for improvements, such as requiring checklists at key points en route to a trial. Barry Scheck cites Atul Gawande’s book The Checklist Manifesto: How to Get Things Right in which Gawande suggests a team is only as strong as its checklists. Scheck credits some of the Dallas CIU’s success to its use of pre-trial and post-indictment checklists. Medical field reformers have shown the way after themselves following organizational principles and practical remedies first seen in business and industry. Famously, Johns Hopkins’ critical care specialist Peter Pronovost created a five-point checklist system in 2001 to prevent intravenous line infections, a big cause of preventable deaths. A year after it was introduced, the hospital’s infections rate at the 10-day mark dropped from 11 percent to zero.
Formal checklists put less reliance on fragile human memory, too.
Pre-trial and post-indictment checklists tailored to specific case types like homicides or sexual assaults, geared to be used a month before trial, for example, could help both prosecution and defense avoid errors of omission in required disclosure and evidence-sharing. The New York City Bar’s Criminal Courts Committee and Criminal Justice Operations Committee recommend checklists as a way to improve prosecutors’ compliance in the disclosure of evidence to defense counsel.
And the American Bar Association has called for local jurisdictions to adopt written checklists.
WELL-DESIGNED TECHNOLOGY and equipment could also further error prevention, says Itiel Dror of the University College of London Institute of Cognitive Neuroscience who has looked at this issue in the context of medicine. “Just as it’s impossible to start a car when it is in gear,” he says, “a workstation could issue prompts and forms that need to be filled out or saved in the correct spot before the user can move on.”
Training that shifts from focusing solely on avoiding errors to detecting and recovering from them could be hugely helpful in criminal justice and forensics, says Dror. The approach works well in medicine and “it is very applicable to policing as well as forensics and other aspects of the criminal justice system, but has yet to be applied.”
It’s essential that training allows students to experience making errors themselves. “Like if you tell your students/kids to ‘save’ on the computer so they do not lose their work if it crashes,” says Dror, “they rarely do it (if at all) until they lose a file. Then they remember! That is the power of error. Experiencing even a small and symbolic failure forms salient mental representations that shape future behavior.”
It helps for instructors to make errors for their students to spot and use them as teaching moments for error recovery strategies. Dror also uses what he calls “sabotage” techniques to make student errors more likely. Medical students, for example, can be “set up” to have the wrong expectations so they’re likely to make mistakes in diagnosis and treatment plans.
Students can also be challenged by competing to see who can identify the most errors in a videotape in the shortest time. And in medicine and criminal justice alike, compiling a videotape library of errors then transforming them into interactive learning tools helps keep brains engaged. (Click here, and then on “medical surgery,” for an interactive taste.)
Former homicide investigator and law enforcement trainer David Taylor routinely uses crime scene photographs in training to show that things aren’t always as they appear. One depicts a dead man lying on the floor, shot once in the head, once in the back. “I ask, ‘What’s that on the floor circled in yellow?’ Everyone immediately says, ‘It’s a bloody footprint!’ I tell them it’s not and I let them think about it. Finally, it dawns on them. The shot in the back was incapacitating. The victim rolled over on the floor then was shot again in the back of the head. It’s the victim’s blood in the yellow circle, not a suspect’s footprint.”
Putting some of the many possible reforms into practice could go a long way towards restoring public confidence in the American criminal justice system.