Willingham case puts spotlight on forensic analysis
Posted Saturday, Apr. 30, 20110
BY YAMIL BERARD
yberard@star-telegram.com

One Oklahoma forensic chemist was so prolific in implicating criminal suspects by examining hair strands that she earned herself the nickname "Black Magic." But her technique was discredited years later.

Was the chemist then obligated to retract her reports to criminal justice officials?

"It's such an obvious thing," said Thomas L. Bohan, past president of the American Academy of Forensic Sciences. "You don't tell a lie. You don't fake your data." There's little disagreement that forensic scientists should correct outright mistakes. In fact, truth-telling is a fundamental principle in a profession that pays homage to rigor and cold objectivity. But gray areas abound, from a typographical error on a report to embellished testimony. When should a scientist correct? And what is the obligation when science evolves to the point that it makes a material difference in the original analysis?

Debate over this principle of "duty to correct" has taken center stage in Texas, as the Forensic Science Commission has examined the case of Cameron Todd Willingham, executed for setting a fire that killed his three children.

While the panel, so far, has sidestepped a finding on whether the fire investigators were negligent, it is urging the State Fire Marshal's Office to take steps that could lead to re-examination of hundreds of older cases where findings may have been based on obsolete science.

In a report, the commission told the fire marshal's office that it has a "duty to correct, duty to inform, duty to be transparent" in such cases.

But it is unclear whether the office will undertake such a review. The office, which has said it made no errors in the Willingham case, won't respond to the commission's report until the Texas attorney general rules on whether the commission has jurisdiction in the case, said Ed Salazar, an attorney with the office.

Some scientists see the whole debate over duty to correct as a smoke screen by "innocence reformers," not intent on truth-telling but on chipping away at public confidence in the criminal justice system.

"Most experienced forensic scientists during the course of their careers turn into a situation where they conduct really sound scientific testing but have their work really spun in the courtroom to meet a certain legal objective," said John Collins, executive director of the American Society of Crime Lab Directors. "When that gets out into the press, it looks like the scientist was a bad scientist."

He and others say the profession is reliable and trustworthy at its core.

If there are any doubts about forensic evidence, some scientists say, the adversarial court system will self-correct. Inconsistencies in test results, for example, can be challenged by defense attorneys, they say.

"You see people who are very comfortable saying, 'Hell, no, that's the defense attorney's problem to figure it out.'" said Amy Driver, a forensic scientist in Washington, D.C.

But she and other scientists find that position appalling and point to hundreds of exonerations based on reversals of faulty science. To them, correcting errors is a moral and scientific duty. Failing to do so leads to failures in the criminal justice system.

"It's one of the most despicable types of things when human lives are at stake," Bohan said.

The problem, to some, is that some scientists become entrenched, unable to admit mistakes.

"Some people develop a siege mentality: 'We're under attack,'" said Max Courtney, a Fort Worth forensic chemist.

Yet even scientists troubled by uncorrected mistakes acknowledge that it is unlikely that anyone can undertake a thorough re-examination of old cases.

Results may vary

A drug sample reaches a crime lab. The name on the chain-of-custody document is "Joe Blow." The lab test is positive for cocaine.

But if the name on the document should be "Sally Sweet," then an error occurred, said Aliece Watts, a forensic scientist in Euless.

Now imagine that the analyst doesn't do the test but reports the cocaine finding anyway, a process called drylabbing. That's not an error, said Watts, a former member of the Texas Forensic Science Commission. Drylabbing is a blatant lie, she said.

There's less debate over the obligation to correct in such a situation. The test is positive or negative. The name is right or wrong. The test was done or not.

"If it's a number that we're working with, DNA profiles, weights of drugs," Watts said, "that's pretty cut and dry. It's either that, or it's not."

But scientists can't be held responsible, she said, for use of techniques later shown to be imprecise. For example, for years scientists tested blood types to identify rapists. The tests were accurate but couldn't tie a particular person to a crime. Science evolved, and DNA tests became the gold standard.

Where the duty to correct really gets blurry is in methods that aren't as black and white. Those methods, such as analyses of hair fibers, tool markings and fingerprints, rely on observation and interpretation, and results may vary among scientists.

Fort Worth labs -- among many -- relied on hair fibers for years to match a person to a crime scene. But the technique was deemed unacceptable after studies made clear that hair strands can't be matched to a single person and can even vary on the same head.

Courtney said he has gone back to review some hair fiber examinations, but some cases have been untouched for years. Evidence can get lost and trial exhibits can become difficult to re-examine, he said.

Dr. Nizam Peerwani, a Texas forensic science commissioner who is also Tarrant County's chief medical examiner, has said labs are required to correct invalid or mistaken results, such as those he said were made in the Willingham case. In light of concerns that surfaced in that case, Peerwani has said he is reviewing arson cases his office handled.

DNA analysis

Crime labs have relied on methods that tie tire tread to tires, guns to criminals and blood spatter to crime scenes, but in 2009 the National Academy of Sciences cast doubt on a host of such interpretive methods, saying DNA analysis was the only proven way to match an individual to crime evidence. All other disciplines, the academy wrote, were more about opinion than fact and lacked protocols legitimized as "scientific."

Are scientists obligated to correct findings of those analyses? No law requires a scientist to amend a mistake or omission. And with very rare exceptions, there are no sanctions against those who do not make corrections, said Bohan, a forensic physicist who has a law degree.

Some scientists say there is no need to correct what amounts to differences in opinion among experts. It's not the scientist's fault, they say, if a jury doesn't grasp that findings are interpretations.

The Texas Forensic Science Commission rejected such arguments. In its report, it said that a courtroom is not well-suited to determining "scientific truth" and that the words a scientist uses can have a profound effect on jurors' understanding.

Take a firearms expert testifying about whether a bullet is a match to a gun. The expert says the bullet is a "match" to the gun. That means the bullet could have come from such a gun, but jurors may jump to the conclusion that the gun was used in the crime. Science has not yet shown that a single bullet can be matched to a single firearm, the academy study said.

The industry is working on concise wording to limit such ambiguities, said Collins, of the American Society of Crime Lab Directors.

That won't end the controversy, though, Driver said.

"It's almost never a problem with the science itself. It's a problem with bad forensic scientists applying the science in a way that would shock scientists in the field," said Driver, a firearms examiner and an expert on crime scene reconstruction.

A stigma is also attached to those who don't disclose key errors as soon as one surfaces, Driver said. Foot-dragging is akin to lying, she and others say.

"Head-on" is the way Arthur Eisenberg, one of the nation's top DNA forensic scientists, described the right approach to mistakes. If one occurs, "we're obligated to bring it to light ... as opposed to covering it up," said Eisenberg, director of the DNA Identity Laboratory at the University of North Texas Health Science Center in Fort Worth.

"We're human; mistakes can be made," he said. "It's how you address and handle those mistakes and what you do to correct them."

What's the duty of a forensic scientist? "To give the most accurate scientific information that we can," he said. "If we fail to do that, if a mistake occurs or an omission occurs, we've let ourselves and the legal system down."

Yamil Berard, 817-390-7705

Read more: http://www.star-telegram.com/2011/04/30/3039976/willingham-case-puts-spotlight.html#ixzz1L5VIq4TX