Junk arson science helped secure capital murder conviction

Walter Reaves, an attorney out of Waco (see his blog here), is becoming the go-to defense lawyer in Texas on arson cases involving antiquated junk science, taking the issue on at a very high level after his client Cameron Todd Willingham was executed based on bad arson science. Here's a story about a case he's working on with the Baylor Innocence Project as reported by KXXV Channel 25 (June 2):

New scientific developments in arson analysis may help set free a man convicted of capital murder more than 20 years ago.

Ed Graf was convicted of killing his two stepsons by burning them in a backyard shed in 1986. But the burn patterns on the floor that had given prosecutors reason to believe it was arson by gasoline, is now being revisited by scientists using newer knowledge of fires developed in the last two decades.

Attorney Walter Reaves and law students with the Baylor Innocence Project dug up the case and asked Dr. Gerald Hurst to re-examine the old evidence.

"Burns on the floor, holes in the floor, had been misinterpreted as being the accelerant poured like gasoline," said Dr. Hurst, "When in fact those patterns were caused by a phenomenon called flashover."

Flashover occurs when all combustible material in an enclosed area ignites almost simultaneously. Hurst said that objects in the shed, like the fold-up cot, could have brought the shed to the point of flashover in a matter of three minutes. Hurst even called the polyurethane foam in the mattress "solid gasoline."

But the original prosecutor, Vic Feazell, said that circumstantial evidence pointed strongly to Graf. "We had so much circumstantial evidence that I believe the jury would have convicted Ed Graf even without the arson evidence."

The circumstantial evidence he refers to is the fact that Graf did not feed the boys or fill their cereal boxes in the last days like he normally did, that he bought life insurance for them only months before the incident, and waited until the last minute to buy them new school clothes, all signs that the prosecutors said that meant Graf knew the boys would soon be gone.

In addition to evidence against Graf's character, original fire investigators also pointed to the burn patterns they thought were caused by gasoline, and the latch that had locked the boys in the shed. That physical evidence from the fire, however, is now being refuted by Hurst.

"In the period of the Graf case it was totally subjective," Hurst said in regard to the burn patterns. "But it was subject to a bunch of old wives' tales that people really believed in."

Don Youngblood, the original defense investigator, added, "It's been shown now that the assumptions made under the testimony by the fire experts was faulty, that the conclusions reached were not accurate."

Dave Mann at the Texas Observer recently covered this case in depth, concluding "It’s ... difficult to say conclusively that Graf is innocent. What does seem clear is that, given the botched forensics in the case, he never should have been convicted."