In Texas, where being “tough on crime” is pure political mantra, guilt is usually assumed. I have reported extensively about DNA and non-DNA exonerationns. Texas ranks fourth nationally. A fairly staggering statistic considering that 1 in 22 people in Texas are in jail, prison or under court supervision. The job confronting organizations like The Innocence Project is monumental. There are a myriad of reasons for false convictions. Lying prosecutors. Lying cops. Lying and sometimes mistaken eyewitness testimony and not the least of all…junk science used to convict.

You may recall the case of Cameron Todd Willingham who was executed for an arson fire that killed his children. Governor Rick Perry called him a “monster” and shuddered a state forensics commission that was about to announce Willingham was likely innocent of the charges that led to his execution.  On the day in 2004 that her first cousin Cameron Todd Willingham was scheduled to be executed in Texas, Patricia Cox had good reason to believe he was innocent. Willingham was convicted of murder for killing his three daughters in a late-1991 house fire following a prosecution that was based almost entirely on a forensic fire analysis performed by an assistant fire chief and a state deputy fire marshal in the days after the blaze. Two weeks before Willingham’s execution date, though, Cox managed to get famed arson analyst Gerald Hurst to take a fresh look at the case.

Working pro bono, Hurst reviewed the evidence and drafted a study that devastated the original findings. While Hurst’s conclusions relied on his own skills, arson forensics had become far more advanced and standardized in the previous 13 years. “A contemporary fire origin and cause analyst might well wonder how anyone could make so many critical errors in interpreting the evidence,” Hurst’s report began. Hurst found that there was, in fact, no arson at all. Cox and Willingham’s attorney rushed the report to the office of Governor Rick Perry on Feb. 13, just four days before the scheduled execution, Cox says. 260xStory

Little is known about why the Hurst report did not compel Perry to use his authority to delay Willingham’s execution for further review. The governor’s office has resisted legal efforts to pry loose Perry’s most sensitive deliberative documents on the Willingham execution. “The governor’s office has been a giant black hole of information,” says Walter Reaves, Willingham’s post-conviction attorney. Cox’s frantic correspondence with the governor’s staff in the days and hours before Willingham’s death, as described by Cox and detailed by Perry’s office to TIME, provide a new window on those events.

In 1991 Willingham, then 24, was a poor white unemployed auto mechanic who sometimes drank too much. His young wife Stacy had been out at the Salvation Army picking up Christmas gifts for the kids on the morning of the fire. His trial lasted just two days, focusing almost exclusively on the forensics, and Willingham remained in prison until his 2004 execution. Willingham, his family and some close friends were among his only allies until the Chicago Tribune reviewed the case in the months after his death, which drew in death-penalty opponents and even more press. Then David Grann explored the Hurst report in an extraordinary September 2009 article in the New Yorker. According to the report, what the initial investigation said were “pour patterns” of gasoline or another accelerant in the house were instead simply the scars typically left behind by a conventional, hot, racing, “post flashover” fire. Hurst also found no evidence that the fire started in multiple locations, a telltale sign of arson.

Hurst disagreed with the original investigators’ determination that charred wood under an aluminum door threshold suggested that an accelerant had burned there. “Liquid accelerants can no more burn under an aluminum threshold than can grease burn in a skillet even with a loose-fitting lid,” Hurst wrote. He also flatly dismissed the first investigators’ finding that strange patterns in the glass of Willingham’s windows, called “crazed glass,” proved the use of an accelerant. Instead, Hurst noted that crazed glass forms when firefighters spray cold water on hot windows when extinguishing a fire. Finally, Hurst said, the early investigators misunderstood the significance of “V pattern” heat scars on walls. “Most of the conclusions reached by the fire marshal would be considered invalid in light of current knowledge,” Hurst wrote.


Since the execution, scientific opinions have accumulated in the conclusion that there was no arson at the Willingham home in 1991. After Willingham’s death, the New York–based Innocence Project commissioned a panel of five arson experts to take another look at the case. The group’s 2006 report matched Hurst’s findings that the evidence showed a conventional fire, not arson. “The artifacts examined and relied upon by the fire investigators … are the kind of artifacts routinely created by accidental fires that progress beyond flashover,” the panel’s report said.

In 2008 the Texas Forensic Science Commission, charged with guarding the integrity of Texas forensics investigations, started to probe the case. The commission hired Craig Beyler, an arson investigator with a résumé nearly as impressive as Hurst’s. Beyler’s Aug. 17, 2009, report echoed Hurst’s findings. “A finding of arson could not be sustained,” Beyler found. But in late September 2009, just days before the commission was set to hear testimony from Beyler — and when Perry was within a few months of facing U.S. Senator Kay Bailey Hutchison in what was expected to be a heated primary — Perry dismissed the chairman of the commission, Austin defense attorney Sam Bassett, and two other commission members. That dampened the commission’s investigation.

Bassett said of Perry, “I think he was very uncomfortable with what we were doing in that investigation.” One lingering question has been whether Perry was dismissive of the Hurst report in the hours before Willingham’s execution. Cox says this is where the governor erred. Cox’s contact in Perry’s office at that time was Mike Schofield, then Perry’s assistant general counsel. Cox has nothing but kind words to say about Schofield. “For the 10 days prior to the execution, I talked with him daily,” Cox says, describing Schofield as polite, engaged and prompt in returning her calls. “He told me he would be the one to brief the governor on the morning of the execution.”

Explorer V: Inside Death Row NGC-US: Ep. Code 4082

Cox, who took notes on her interactions with the governor’s office, talked to Schofield on Feb. 16, the day before that briefing. She says she was convinced that Schofield understood the importance of Hurst’s findings. “Mike Schofield told me he had the Hurst report in his possession,” she says of that talk. “I know that for a fact.” Schofield said he had discussed the Hurst report with other attorneys serving the governor, Cox says.

Schofield did not respond to efforts to contact him through his associates and Perry’s office. The next day, Feb. 17, Cox went to work as usual. She waited anxiously for the phone to ring with news of a last-minute stay. The call never came. Cox talked to Schofield again the next day, after Willingham was dead. “He did give his condolences to myself and the rest of the family,” she recalls. She asked about the briefing with Perry. “I asked him, ‘Did you take the Hurst report?’ He told me he absolutely did,” Cox says. “He said the governor felt like, in light of the time that had passed since the conviction, a stay in the execution would only delay an event that would ultimately happen anyway.”

The closest Perry has come to explaining his decision to proceed with Willingham’s execution was covered in a Sept. 18, 2009, Dallas Morning News story. In the article, Perry seems to dismiss the advances in arson forensics since 1992 and belittle the qualifications of experts who had studied the Willingham case more recently. “I’m familiar with the latter-day supposed experts on the arson side of it,” Perry said. The newspaper reported that Perry formed quotation marks with his fingers to emphasize his skepticism. Lucy Nashed, Perry’s deputy press secretary, confirms that Perry’s office did obtain a copy of Hurst’s report, but not until the day of the execution, and that Perry’s staff had to retrieve a copy from the Texas attorney general late that afternoon. “It was only through the efforts of Governor Perry and his staff that the report was obtained at all by our office,” Nashed says.


Regardless of when it arrived, Nashed confirms that Perry was aware of the Hurst report and its contents. “As with any execution, the governor was previously briefed on all of the facts of the case. He was briefed on this report once our office obtained it late that afternoon,” she says. “After the document — which was four pages — was requested, obtained and reviewed, Governor Perry shared the opinion of the state and federal judiciary that the report was not enough to merit a stay of execution,” she says. Nashed is referring to the frantic efforts by Willingham’s post-conviction attorney, Reaves, to get various courts to look at Hurst’s report prior to Perry’s last-minute review. The courts, however, were barred from reviewing anything that wasn’t classified as new evidence, rather than new science applied to old evidence, so the courts rejected Hurst’s findings on a technicality. “The stuff I put in that report had never been reviewed before,” Hurst says. “The courts are not technical. They don’t know what is new. We just know a hell of a lot more about post-flashover fires than they did then.”

Fast forward to today. NBC News asks a new question about an old case.  Is Douglas Boyington an innocent man?  The fire started in the apartment building’s utility shaft. The space was narrow, wood-framed and as tall as a chimney. Water pipes and electric lines snaked through it, then out, into the three-story brick building. By the time the first fire investigator arrived, at 7:05 p.m. on August 16th, 1988, bright yellow flames had burst through the roof. Black smoke billowed overhead. “There was fire everywhere,” the investigator, Hickman Garrett, later recalled, according to court documents. No one was hurt, though the 24-unit building — part of a sprawling apartment complex in Pasadena, an oil refinery city southeast of Houston — was destroyed. Douglas Boyington, a 32-year-old chemical tank driver whose parents lived in the apartment complex, and who witnesses saw there that day, was charged with starting the fire. On March 2nd, 1989, after a four-day trial, Boyington was convicted and sentenced to 75 years in prison.

In a recent interview at the state penitentiary in Huntsville, Boyington, 57, said what he always has — that he was at a nearby bar celebrating his birthday when the fire broke out. “I’m innocent,” he said. Now, an alliance between two unlikely partners — the state fire marshal and the Innocence Project of Texas — may help prove him right. When Boyington was convicted, fire investigation was still a largely unscientific endeavor. Experts now know that widely accepted standards for what was considered clear evidence of arson — for example, that black smoke and chipped and discolored concrete, or “spalling,” indicated that a fire was set using an accelerant — were based on flawed interpretations of how fires behave. “There was no science,” said John DeHaan, who has written several editions of the seminal text book, Kirk’s Fire Investigations.

In Texas, the state fire marshal’s office and the Innocence Project of Texas, a nonprofit dedicated to exonerating wrongfully convicted people, are now reviewing old cases where unsound methods may have helped secure a conviction. Such a partnership is unprecedented, and it is transforming the relationship between science and the law, helping turn Texas into “a model” for the country, as DeHaan put it. “Nobody in the criminal justice system wants to look back at these cases to see if they’ve made a mistake,” he said. Like DNA testing, which has helped exonerate hundreds of innocent people who were convicted of crimes since 1989, re-examining arson cases could have broad implications. “This truly is the future of innocence work,” said Jeff Blackburn, the founder of the Innocence Project of Texas.

In Boyington’s case, Garrett concluded that an accelerant was used before the fire was out. The blackness of the smoke and brightness of the yellow flames — and the intensity of the burn — “showed me there was something other than common combustible materials which had aided this fire,” Garrett said. This conclusion was further supported by the presence of spalled concrete. The inquiry into Boyington’s case began last year, after the Texas Forensic Science Commission completed an exhaustive report on the investigations into two deadly fires that helped convict and sentence to death two men, Cameron Todd Willingham in 1992 and Ernest Ray Willis in 1987. In 2004, Willingham, who had been charged with setting the fire that killed his three daughters, was executed; Willis was exonerated. The report revealed that the cases against them relied on a host of indicators that were once accepted signs of arson, but which science had disproven.

Blackburn wanted to see how many Texans had been convicted using these indicators, so the group began canvassing every Texan serving prison time for arson charges. After sifting through dozens of questionnaires and case files, the project compiled a list of nine inmates. Boyington was on that list. Though it wasn’t necessary, Blackburn hoped to work with state fire officials to re-examine those cases. “Being on the outside looking in is not a good way of getting people out of prison,” he said. So Blackburn reached out to Chris Connealy, the recently appointed state fire marshal. When the two met at a beer garden in downtown Austin, Blackburn, a blunt criminal defense lawyer from the Texas Panhandle, didn’t expect much; he had tried before to work with the fire marshal’s office to conduct a review of old cases, but he’d never gotten any traction.


Blackburn was immediately impressed with Connealy, a cool-headed career fire fighter. “The guy’s approach had nothing to do with maintaining the prestige of the fire marshal’s office — of making them look good by never admitting they were wrong,” he said. “That is a common attitude in law enforcement here and everywhere. That is the single biggest obstacle to getting innocent people out of prison.” Connealy explained the resulting partnership this way: “We have the same goal. We want to make sure that justice is served.” After meeting with Blackburn, Connealy began assembling an advisory group to oversee the review that included respected arson authorities like DeHaan, David Icove, an engineer and former arson profiler with the Federal Bureau of Investigation, and several others. In the months since, the panel added another case to the Innocence Project’s list, and has completed five reviews. In three, arson investigators were found to have relied on flawed science; in two, the original investigations were deemed adequate.

In Boyington’s case, the fire marshal’s advisory group concluded that the investigation that helped put him in prison did not meet modern scientific standards. The presence of spalled concrete in the apartment building did not mean that an accelerant had been used, the panel concluded. Nor did samples taken from the scene in 1988 indicate the use of an ignitable liquid. The panel also found that there was a failure to account for how the apartment building was built — with the shaft containing the electric lines and water pipes in its center core. Pasadena Fire Marshal David Brannon said that such construction was no longer allowed in the city. “There is nothing to impede the growth of fire,” he said.

Based on the advisory group’s findings in the Boyington case, Connealy wrote a letter this summer to the Harris County District Attorney saying that the fire could no longer be considered arson. “The cause of the fire should be listed as undetermined,” Connealy wrote.  A spokesman for the Harris County District Attorney’s Office declined to comment for this story. Brannon also declined to discuss the case with NBC News. In a statement, he wrote, “The case involving Douglas Boyington has been reviewed and upheld by an appellate court three times.” But Brannon has pushed for increased arson training for investigators in his department.   In a West Texas conviction where the advisory group also found flawed science, the district attorney there is challenging the experts’ authority to review the case. In a letter sent to the Texas Attorney General in October, District Attorney Rod Ponton argued that the fire marshal’s office should not make “sweeping legal pronouncements on 20-year-old criminal cases.”


The Attorney General is expected to issue an opinion in that case, which could affect others under review, including Boyington’s, within the next several months. Regardless, Boyington’s lawyers will ask for a hearing to determine their client’s innocence — the start of a potential exoneration process — in the coming months. They could be helped by a new state law that makes it easier for inmates who believe they were convicted using outdated science to challenge their convictions. Boyington is anxious to clear his name — something that he once had all but given up on. About a decade ago, he was paroled and had gone back to work driving trucks. He forgot to check in with authorities, a violation of his parole. During a traffic stop, a police officer asked for his ID — and Boyington knew that likely meant heading back to jail. “I was fed up,” he recalled. “I’d done enough time down here behind something I didn’t do.”

After turning over his license, Boyington punched the gas on the truck. With police racing after him, he stalled on a highway median, jumped out of the vehicle and began shuffling through his pockets, as if he had a weapon.  “I tried to get them to take me out,” he said. Just as one officer shouted “gun,” another tackled Boyington. Shortly after, he was back in prison. “I’ve done had enough for a crime I didn’t commit,” he said. “I’m tired.” The problems in Texas are not isolated or limited to one area…one person or always easy to identify. It simply falls under a cultural belief that being “tough on crime” solves all of the states criminal justice issues. It is under this political umbrella that prosecutors have been allowed reckless latitude to go after anyone they want while appeals courts have turned a blind eye to obvious miscarriages of justice.

About these ads