Debunked science has kept Suzette Welton in prison for 17 years

Second of three columns

Junk science convicted Suzette Welton of setting the fire that killed her teenage son, but 17 years later the Alaska Court System won't let her out of prison unless she proves she is innocent, which is next to impossible.

In 2002, when the judge in her case sentenced her to 99 years in prison, he made it clear that he disagreed with the jury's verdict.

"I am haunted by the specter, the possibility, that I am sentencing an innocent woman. I'm haunted by that," Judge Milton Souter said. The case, at the end of his 24-year career, was the hardest he had ever judged.

He could have overruled the jury if he found the verdict was completely unjustifiable, but decided not to after balancing the question for three agonizing months.

His reason was the "very clear evidence" of arson. Experts said someone set the fire in Welton's rented house. A process of elimination pointed to her as the culprit.

But it turns out that the evidence of arson was not clear at all. Fire investigators relied on debunked methods — "old wives' tales," a national expert calls them — which have been discredited scientifically, leading to dozens of exonerations across the country.

[Part 1: Unjustly imprisoned for 30 years, Donna Armey can't get Alaska's courts to listen]

With no arson, there was no crime. The rest of the case against Welton consisted of a few wisps of circumstantial evidence.

But, as I wrote in my last column, Alaska's tough laws, and our lack of an executive clemency process, make it nearly impossible for innocent people to get out of prison before finishing their sentences. It might not be enough for Welton to show there was no evidence of arson.

"It doesn't actually prove innocence," said Appeals Court Judge David Mannheimer, at a hearing last month, speaking to an attorney for the state.

Mannheimer asked, "Is there a procedural vehicle in Alaska law for a defendant to come back and basically ask for a new trial because there was reason to distrust the scientific evidence that the state relied on, even though it doesn't rise to the level of proof of actual innocence?"

Diane Wendlandt, an attorney with the Office of Criminal Appeals, responded, "If it does not rise to the level of actual innocence, if it merely casts some sort of doubt, then I would argue that no, there is not."

'A decent, loving, caring person'

In September 2000, Welton, then 37, lived with her two teenage sons and 6-year-old daughter in Wasilla. She was a recently divorced child care worker. Even her sentencing judge called her, "a decent, loving, caring person."

She had bought life insurance on her sons, but not the term insurance or double indemnity that would maximize benefits if they died. Instead, she bought whole life insurance that is normally used as a savings vehicle. On the night of the fire, she slept downstairs with the uninsured daughter.

The fire started upstairs. Jeremiah Welton, then 17, escaped, but his 14-year-old brother, Samuel, died. Both boys appeared to have taken a sleep-inducing drug found in the allergy medication Benadryl, which was also found in beverages they drank that evening. The prosecution said their mother drugged them.

The insurance and Benadryl are the primary evidence against her.

Investigation showed that the boys had abused drugs, including sleeping medications. Samuel had played with fire in the past and was unhappy at the new school he was attending due to the divorce.

Fire investigators for the state of Alaska determined the fire was started by an accelerant such as gasoline. Their evidence was the marks that were left, a hole in the upstairs floor, and firefighters' belief that the fire was especially hot. It reignited when they put water on it.

But chemical testing didn't find any accelerant and the kind of signs the Alaska investigators relied on had been debunked years earlier in scientific research.

Accelerant-caused fires do not burn hotter. Accelerants are not necessary to cause reignition. Accelerants do not normally burn holes in floors. Most important, burn patterns can emerge the same way from accelerant fires and nonaccelerant fires.

These facts had been published in professional standards by the National Fire Protection Association. But the old methods continued to be used by nonscientist fire investigators.

At Welton's trial, a nationally known scientist testified for the defense, Dr. Vytenis Babrauskas, who had done some of the original experiments that debunked fire investigators' unreliable methods.

But today, state attorneys say the fact that he testified is a reason to keep Welton in prison. The reasoning goes that since Babrauskas exposed flaws in the case at the time, and the jury still convicted, then Welton must be guilty.

Legally, trials are assumed to be infallible if they are done according to proper procedures.

The concept is called finality. Essentially, it says that when a case ends, unless the trial was improper, new information is irrelevant. U.S. Supreme Court rulings have placed finality above justice, allowing an execution in 2009 rather than looking at new evidence.

In a 1993 decision, Justice Antonin Scalia explained the refusal to consider new facts, writing, "There is no basis … for finding in the Constitution a right to demand judicial consideration of newly discovered evidence of innocence brought forward after conviction."

Finality is said to bolster the public's respect for the courts (although surveys show fewer than 1 in 4 of Americans do have confidence in the criminal justice system). Finality also saves work for judges and money for prosecutors and prevents the loss of face that comes from admitting mistakes.

But a flood of exonerations by new science has shown that trials are not infallible. They're not even very good at finding the truth. With the errors accumulating, the National Academy of Sciences issued a report in 2009 showing problems with almost all the so-called forensic sciences that criminal courts rely on, except for DNA.

Some techniques, such bite- or tool-mark analysis, had no scientific basis at all, the report said. Even fingerprint analysis had serious problems. And the way crime labs do forensic tests and testimony was questioned, too. On arson, the academies pointed to previous work that had already found deep flaws.

American courts relied for decades on unfounded forensic testimony because the legal system is not a self-correcting system, as is the scientific method. Rather than a process of open truth-seeking, courts operate according to the medieval process of adversaries facing off, like the champions who once fought with swords to settle disputes.

In court, the adversarial process means dueling technical experts try to persuade judges and juries who are unqualified to understand their scientific arguments. What these contests often determine is who has the better lawyer.

In Welton's trial, Babrauskas, a highly respected fire scientist, faced cross-examination by a skillful prosecutor who got him to confusedly and inaccurately admit he was not an expert on fire cause and origin. His testimony sounded tentative and his technical explanations became muddy and hard to follow.

To rebut his testimony, the state brought in David Campbell, a retired fire investigator from North Carolina without a relevant scientific degree but 3,000 investigations to his name. A noted public speaker, he clearly and simply bolstered the state's case in language anyone could understand.

The defense attorney didn't challenge Campbell's expertise.

In fact, court records show Campbell is a controversial figure in the fire investigation community, criticized by scientists for continuing to teach traditional techniques after they were debunked and fighting the new professional standards transforming the field.

Those critics won the day nationally. By now, they have exonerated dozens of arson convicts around the country, some after serving as long as 45 years in prison.

Fire scientist and engineer Douglas Carpenter worked on some of those cases. In 2006, he served on a panel of four highly credentialed fire scientists who delved into a set of case studies to highlight the problem.

The bad cases included the conviction of Cameron Willingham, a Texas man executed in 2004 for setting a fire that killed his children while his wife was Christmas shopping. Willingham continued to proclaim his innocence while strapped to the gurney in the execution chamber.

The panel determined Willingham's case showed no sign of arson. That evidence was similar to Welton's case.

Publicity about the case studies helped change the Texas legal system, as I'll discuss in my next column.

In 2013, Carpenter reviewed Welton's case at the request of Fred Dewey, a mostly retired Anchorage attorney who took her on six years ago for token state compensation. Dewey was traveling to the East Coast for a wedding and made a side trip to Carpenter's office in Maryland.

After a preliminary review, Carpenter found the same errors in Welton's case as in the Texas case and many others. A friend of Welton paid his $3,000 fee. But for a full review and court testimony, the fee would be $14,000, which she couldn't raise.

Dewey asked a Palmer Superior Court judge for Carpenter's fee, but she dismissed the case and turned down the money in March 2014. He appealed that ruling and finally got in front of the appeals court last month.

Dewey got to speak just 30 minutes. Wendlandt, arguing for the state, didn't even use her full 30 minutes to say the lack of evidence of arson was not enough to prove Welton's innocence.

When I told Carpenter that, he said, "Any fire that occurs without any other indication is an accidental fire. An accidental fire would be one where a person is innocent."

He said the alternative, of proving a fire wasn't arson, is impossible. "You can't prove a negative," he said.

Welton lives at Hiland Mountain Correctional Center doing everything she can to improve herself. She runs with volunteer coaches, helps other women with fitness and weight loss, finished a program in culinary arts, played in an orchestra, and gained graduate degrees in Christian ministry.

When she stood at sentencing 15 years ago, she said she would stay alive long enough to vindicate herself. She still remains insistently positive.

"By the grace of God, he's the only thing that has ever kept me sane in here, because I know I have a higher purpose. I'm not going to let anything destroy that. They might have destroyed everything else, but I'm not going to let them get me," she said.

The Court of Appeals is considering whether to pay Carpenter to do a full report. Welton's attorney, Dewey, said that decision could take up to two years.

Assuming the answer is yes, the case would go back to the Superior Court for a hearing on the completed report. That judge would have six months to rule. An appeal of that ruling could take another two and a half years. And then that could be appealed to the Supreme Court.

"She could be an old woman by the time this is over," Dewey said.

This system is unjust. In my next column, I'll discuss ways to improve it that are used in other states.

Next: Alaska could release innocent prisoners. Other states do.

The views expressed here are the writer's and are not necessarily endorsed by Alaska Dispatch News, which welcomes a broad range of viewpoints. To submit a piece for consideration, email commentary@alaskadispatch.com. Send submissions shorter than 200 words to letters@alaskadispatch.com or click here to submit via any web browser.