The State Fire Marshal’s Office stands behind its controversial conclusion that Cameron Todd Willingham started the house fire that killed his three children in 1991, contradicting arson experts and scientists who insist the agency relied on bad science in its investigation.
In a pointed letter to the Texas Forensic Science Commission , which is nearing the end of a contentious review of the Willingham arson investigation, Fire Marshal Paul Maldonado defended his agency’s handling of the case that led to Willingham’s execution in 2004.
In July, the commission announced a tentative finding that investigators employed “flawed science” — including now-debunked beliefs that certain fire behaviors point to arson — to conclude that Willingham intentionally set fire to his Corsicana home.
But Maldonado said his agency’s investigation remains valid, even after modern, scientific arson standards are applied.
“We stand by the original investigator’s report and conclusions,” Maldonado said in his Aug. 20 letter to the commission. “Should any subsequent analysis be performed to test other theories and possibilities of the cause and origin of the fire, we will of course re-examine the report again.”
Maldonado’s letter was among 11 received last month after the commission solicited final comments before members prepare to write their report on the Willingham investigation at a specially called meeting Sept. 17 in Dallas. The four-member Willingham subcommittee will have draft language for the rest of the seven-member panel to consider, but much of the discussion on final wording will take place during the open meeting, Chairman John Bradley said Wednesday.
Some letter writers criticized the State Fire Marshal’s Office, which provides arson investigation help to communities across Texas, for declining to acknowledge shortcomings in the Willingham case.
“For the fire marshal’s office to permit (Willingham) to remain imprisoned during all those years leading up to his execution, in spite of knowing for most of those years that the compelling testimony \u2026 provided at trial was solidly wrong, is to partake in unethical scientific behavior of the most extreme kind,” wrote Thomas Bohan, a 35-year forensic physicist from Maine.
The fire marshal’s office compounded its “unethical, negligent behavior” by failing to review similar arson cases that may have been influenced by false or misleading arson indicators, Bohan said.
The letters to the commission highlight an intriguing fight over the scope of its Willingham inquiry, which is being closely watched by both sides of the death penalty debate.
One side, led by the Innocence Project of New York, is pushing for a broad inquiry that could cement the primacy of science-based fire investigations while serving as a launching pad to review convictions that may have been based on now-discredited investigative techniques.
The other side, including Corsicana officials and some fire investigators, seeks a limited review focused on one question: Did investigators follow standard practices when they sifted through the rubble of Willingham’s home in 1991 and 1992? They argue that investigators cannot be faulted if they applied commonly used techniques to assess the fatal fire.
Thus far, the Texas Forensic Science Commission appears to be engaged in a limited review, a course it set after Gov. Rick Perry shook up the panel’s membership a year ago, replacing the former chairman, Austin defense lawyer Sam Bassett, with Bradley, the Williamson County district attorney.
At the commission’s quarterly meeting in July, the Willingham subcommittee announced that it could not find that investigators engaged in misconduct or negligence because they were merely following standard practices for arson investigations of the era.
Based on that finding, many of those who submitted letters to the commission tackled a version of the question: What did investigators know, and when did they know it?
Nobody disputes the notion that fire science has come a long way since 1991, largely because of the National Fire Protection Association’s “NFPA 921: Guide for Fire and Explosion Investigations.” Published in 1992, after the Willingham fire, the guide marked the first concerted effort to bring the scientific method to fire investigations, but it took several years to catch on.
Barry Scheck, co-founder of the Innocence Project, said the fire marshal who investigated the Willingham fire, Manuel Vasquez, should have known that his methods had been deemed unreliable by scientists and leaders of his profession. That information “was readily available” in 1991 even if NFPA 921 had not yet been published, Scheck said in an Aug. 20 letter to the commission.
“Simply because ‘everyone else was doing it’ does not make (his) actions reasonable or not negligent,” Scheck wrote.
Craig Beyler, a fire scientist hired by the commission to review the investigation into the Willingham fire, wrote a 2009 report that was highly critical of Vasquez’s findings and testimony. Beyler’s report, and a follow-up letter sent to the commission Aug. 3, concluded that Vasquez relied on techniques that industry texts had already dismissed as folklore, such as:
• Vasquez testified that wood burns at 800 degrees, meaning an accelerant must have been used to reach the 1,200 degrees necessary to melt an aluminum threshold. But accelerant fires are no hotter than wood fires, and both can reach 2,000 degrees, Beyler said.
• Vasquez testified that “puddle configurations” and burn patterns on the floor could only have been caused by burning liquids. In reality, Beyler said, such patterns are typical in rooms, like those in the Willingham home, that were fully involved in a fire.
• Vasquez determined that a severely cracked porch window indicated a fast, hot fire due to accelerants. It is more likely, Beyler said, that the “crazed glass” resulted from firefighters hitting hot glass with water.
Maldonado, who became state fire marshal in 2004 after rising to assistant chief for the Austin Fire Department, acknowledged that his agency used many of the principles and practices espoused by NFPA 921 when Vasquez — who died in the mid-1990s — investigated the Willingham fire.
Attached to Maldonado’s letter was a point-by-point analysis showing that Vasquez’s arson finding can be supported by NFPA 921, which says melted aluminum, burn patterns, broken glass and other fire phenomena “may also be caused by ignitable liquids.”
The attachment also suggested that commission members take into account that Vasquez’s conclusions were based on a personal review of the fire scene and interviews with Willingham, who offered conflicting accounts of the fire.