Burning Justice

For decades, fire investigators walked into charred buildings in search of the same clues, the same subtle traces, that they thought indicated arson: the furniture and windows buckled by extreme heat, the burn patterns on the floor scorched by gasoline. Their methods weren’t scientific. Investigators relied on a set of assumptions—inherited knowledge passed from one generation to another—about how buildings burn. They used those pieces of evidence to send thousands of defendants to prison. And much of it was wrong.

In the past 15 years, the science of detecting arson has undergone a revolution. Research has shown definitively that many of the assumptions fire investigators once employed to distinguish an intentionally set fire from an accidental one were false.

Yet those disproved forensics sent Cameron Todd Willingham to the Texas death chamber on Feb. 17, 2004. His final claims of innocence—that he hadn’t started the 1991 house fire that killed his three daughters—were dutifully recorded before potassium chloride pumped through his veins and stopped his heart.

More than five years later, Willingham’s execution lies at the center of the biggest political scandal in Texas since Tom DeLay was roaming the halls of power. Nine of the nation’s top fire scientists have looked at the evidence against Willingham and concluded it was based on flawed and outdated assumptions, a collection of “old wives tales,” as one put it. These experts say the fire was likely accidental. If there was no arson, then there was no crime, and Willingham was innocent.

The story became even more explosive in late September when Gov. Rick Perry—in the middle of a tough re-election campaign—replaced three members of the Texas Forensic Science Commission, which was investigating the Willingham case, just days before an important hearing. Perry’s new appointees promptly canceled the meeting. John Bradley, the Williamson County prosecutor Perry placed in charge of the commission, has not committed to restarting or completing the Willingham probe.

Suddenly there was the whiff of a cover-up, and the makings of a national scandal. Critics accuse Perry of trying to subvert an investigation that might prove he oversaw the execution of an innocent man (he was told of the faulty forensics before the execution). Meanwhile, anti-death penalty activists—sensing that Texas (of all places) might become the first state to admit putting an innocent man to death—latched on to the Willingham case and began rapidly firing off email blasts and organizing protests.

Perry was all too happy to play along and frame the story as a debate over the death penalty. He told reporters recently that Willingham was a “monster” and that the analysis of the nation’s leading fire experts was nothing more than “propaganda by the anti-death penalty people.”

What has been lost in the heat of political scandal is the original and central issue: the veracity of the forensic arson evidence.

For a time, it appeared the Forensic Science Commission’s investigation into the Willingham case might improve the way fire investigators in Texas do their work. The Legislature created the commission in 2005 after scandals at crime labs around the state. The commission was charged with investigating innocence claims based on flawed forensics, and bringing to light any faulty practices that may have sent innocent people to prison. It’s one of the first such governmental bodies in the nation. For once, Texas was a criminal-justice innovator.

Arson cases seemed the perfect place to start. Like the cases of Willingham and Ernest Ray Willis, who like Willingham was wrongly convicted of setting a fatal fire— a 1986 blaze in West Texas—based on the same kind of flawed interpretation of burn patterns. (Willis, however, was luckier; he was exonerated in 2004 after 17 years on death row.)

Craig Beyler, a nationally respected fire expert hired by the commission to study Willingham and Willis, concluded in a recent report that he saw no evidence of arson in either case.

Nearly 800 people are serving prison sentences in Texas for arson-related crimes. Quite a few are probably innocent. This year, the Observer has investigated some of the state’s older arson convictions in its series, “Burn Patterns,” and so far found three men who were likely wrongly convicted and remain in prison. (You can read all three stories here, here, and here.)

It’s probably still happening. In some areas of the state, poorly trained fire investigators are no doubt still using some of the old, disproved forensics to accuse and convict innocent people of arson.

Sadly, there’s no helping Willingham now. His case has been closed irrevocably. But the Forensic Science Commission’s work might have (and hopefully might yet) help many people. The commission’s report could serve as a kind of official list of flawed and outdated arson evidence that could be applied to past and future cases.

And now it’s all in doubt. That may be the greatest injustice yet.