'Leading Fire Investigation Into the 21st Century'

At TexasMonthly.com, Brantley Hargrove has a substantial piece titled "Leading Fire Investigation Into the 21st Century" about the case of Sonia Cacy, a West Texas woman accused of arson who was found guilty based on flawed junk science. For the record, Cacy is currently represented in her habeas corpus proceeding by Gary Udashen, who is board president of my employer, the Innocence Project of Texas.

Hargrove framed the story in terms of the renewal and revival of arson science in Texas beginning when Gerald Hurst first analyzed expert testimony in Sonia Cacy's case to help her secure parole in the 1990s. He related the role of the Todd Willingham execution and the post-mortem political fiasco surrounding the Texas Forensic Science Commission as a spur to banishing bad science in Texas arson cases. And he described state fire marshal Chris Conneally's panel of experts reviewing old arson cases like Cacy's and Ed Graf's to identify bad science and reinforce the use of the good stuff.

Grits is happy to see wider coverage of these obscure but important topics. Go read the whole thing.

Waco conviction overturned over debunked arson testimony

In a curious per curium opinion, the Texas Court of Criminal Appeals today granted habeas corpus relief to Ed Graf (see Grits coverage here, here, and here), who was convicted based on expert testimony about arson that was substantively demonstrably false. The case is notable both for a) the quick turnaround and b) the lack of dissent among a court that was divided 5-4 on similar questions in Ex Parte Robbins, Overton and Henderson. Perhaps with SB 344 by Whitmire passing the Texas Senate this week in direct response to their recent decisions, the court decided these old arson cases were the wrong venue to quarrel over whether debunked science provides a basis for relief.

MORE: At the Dallas News (March 28), see, "Force arson science to apply in criminal court," and from Dave Mann at the Texas Observer, see "The Arson Files: Ed Graf wins a new trial."

Judge recommends habeas relief in conviction based on junk arson science

Following a habeas hearing in district court discussed earlier on Grits, Judge George Allen in McLennan County has recommended a new trial based on expert testimony that the arson investigation in the two-decade old case was profoundly flawed. See related coverage:

• Texas Observer: After serving 25 years, Ed Graf may finally receive a new trial 
• Texas Observer: Judge recommends new trial for Ed Graf; see also the Observer's 2009 coverage of the case.
• Waco Tribune-Herald: Judge's okay puts Graf one step closer to new trial in murder
• KWTX-TV: Judge recommends new trial for local man convicted of sons' fiery deaths

Graf's case will be the first time the Texas Court of Criminal Appeals will consider whether a defendant convicted based on junk arson science may be entitled to habeas relief, or at least the first since the whole controversy over the Todd Willingham case brought international attention to the subject. But precedents at the high court are all over the map regarding whether such convictions can be overturned when the science changes or improves, and several judges on the CCA think it shouldn't be allowed. The Texas Legislature this spring will consider a bill (expected to be filed next week) to close that gap in state habeas law to clarify that new, potentially exonerating scientific evidence can provide a basis for post-conviction relief. The measure is one of two recommendations from the Timothy Cole Advisory Panel on Wrongful Convictions that the Legislature has not yet acted upon (the other is to require police to record custodial interrogations). For Graf's sake, and others similarly situated, I hope it passes this time around.

This morning Grits is headed to the Forensic Science Commission meeting where they'll hear a presentation on the status of the arson review being conducted by the state fire marshal and my employers at the Innocence Project of Texas. After the politicized brouhaha surrounding the Todd Willingham case, it's a welcome relief to see these topics addressed in a calmer, more thoughtful and systematic fashion. But unless and until the Legislature or the court clarifies that false convictions can be overturned based on new, exonerating science, it's impossible to envision how the end-game will play out, either in Graf's case or others involved in the arson review.

Prosecution failed to dispute expert testimony at habeas hearing on arson science

Dallas News editorialist Sharon Grigsby has a blog post about Ed Graf's habeas hearing in Waco last week, complementing Waco Tribune-Herald coverage over the weekend. Grigsby, who was childhood friends with Graf's sister, pointed to similarities between expert testimony in Graf's habeas appeal and the expert conclusions in the Todd Willingham case, providing links to documentation from Graf's fire experts and the Craig  Beyler report debunking arson investigator testimony used to convict Willingham. As Grits suggested last week, both cases involved the use of arson "indicators" that are no longer considered valid in the field, and in both cases the bulk of the probative evidence was hauled  off to the dump before it could even be examined. From Grigsby's post:

The arson investigators at the time had little understanding about, among other things, how fire burns. Bluntly, they seemed clueless about the science involved. One example: The arson investigator thought fire can burn straight down, through a plywood floor, when that’s just not the case.

Rodger [Jones], who has written a lot of editorials about the Cameron Todd Willingham case, says he sees remarkable similarities between the Graf reports and reports done on the controversial Willingham case. The initial arson investigators in Willingham’s case had the same poor grasp of the basics of fire science as they did in the Graf case.

In my opinion, the new analysis in the Graf case indicates the same use of “old wives’ tales” used by arson investigators in reaching their conclusions.

Notably, reported the Tribune-Herald, the McLennan County District Attorney declined to put on his own expert witness at the habeas hearing on Friday, leaving testimony by the two experts brought by Graf's legal team unrebutted. (In the interest of full disclosure, Graf's attorneys work with Grits' employers at the Innocence Project of Texas, though I have no knowledge of the case beyond published news reports.) Without the arson investigator testimony, Waco attorney Walter "Skip" Reaves told the paper, “I feel at minimum we will get a new trial out of (the hearing), and if we do, then I don’t think they’ll re-try it because I don’t think there is any evidence of arson.” If that prediction sounds optimistic, it's certainly bolstered by the fact that the McLennan DA failed to dispute the validity of expert testimony at the habeas hearing.

These old arson cases present a challenge for the Court of Criminal Appeals, which has struggled in determining whether convictions based on false forensic testimony may receive habeas corpus relief. However, without the culture-war circus surrounding capital punishment in the Willingham saga, cases like Graf's allow for a more thoughtful consideration of how to secure justice when junk science is presented as fact in the courtroom.

First case under consideration in Texas arson review

The Dallas Morning News editorial board offered praise this week for the arson review being conducted by the state fire marshal and my employers at the Innocence Project of Texas in an item ("Duty to correct arson cases," Jan. 9) which opened thusly:

Breakthroughs take time in criminal justice reform, and they get messy, but they are no less impressive when they happen.

Just this week in Houston, the state fire marshal’s office sat down with outside experts to pore over a short list of old arson cases suspected of using junk science to put someone behind bars. One of those suspect cases, from the Central Texas town of Hewitt, is on a separate review track in McLennan County. The district attorney there has cited “serious and complex issues” involving arson forensics in the murder conviction of Ed Graf, who will get a hearing Friday on a writ to reopen his 26-year-old case.

All this traces back to the noisy early days of the Texas Forensic Science Commission and its first case, the arson-murder conviction of Cameron Todd Willingham, executed in 2004. Critics were prone to calling reformers out of bounds, grandstanders who were out to undermine Texans’ support of the death penalty.

Those critics need to take a look today. The fight was a righteous one and has yielded a kind of systematic re-examination of the science in arson convictions that is unprecedented in the nation.

The News pointed to a Waco Tribune Herald report from December ("Experts to testify in Ed Graf arson/murder case from 1980s," Dec. 23) on the first arson case to undergo joint review by IPOT and the fire marshal, that of a man named Ed Graf. A hearing on Graf's habeas writ was scheduled in Waco for today. The Trib story opened:

Two of the nation’s leading fire science experts are scheduled to come to Waco next month for a hearing that will help decide the fate of a Hewitt man who claims he was wrongfully convicted of arson murder.

But the hearing could have implications far beyond Ed Graf’s case, said Jeff Blackburn, chief counsel for the Innocence Project of Texas.He said the state’s criminal justice system is starting to come to terms with the idea that junk science contributed to a number of wrongful convictions in recent decades.

But the state’s highest criminal court has not yet developed a uniform and fair way of handling such injustice, he said.

The hearing in Graf’s case, scheduled for Jan. 11, will be the first post-conviction hearing in Texas where attorneys will present evidence to show faulty fire science was used to secure an arson conviction, Blackburn said.

If he and Waco attorney Walter M. Reaves Jr. are able to provide the level of proof they think they can, Graf’s case could well be the one that finally causes the Texas Court of Criminal Appeals to set a precedent that offers appropriate relief to people ensnared by bad science, he said.

“The law in this area is complicated and generally terrible, but the facts of Ed Graf’s case are not,” Blackburn said. “It shows the stark possibilities of the way science can be misused and abused in a courtroom. The guy shouldn’t have been convicted and deserves a new trial.”

The DA and Graf's ex-wife still believe he is guilty, but the arson "science" in Graf's case eerily tracks the outdated and debunked techniques in the Todd Willingham debacle. Again, from the News:

The commission’s final report — while not commenting on Willingham’s guilt or innocence — said prosecutors relied on arson investigators who had a poor understanding of fire science and learned their craft when there were no uniform standards.

Those very themes surfaced in the Graf case, in reports compiled by experts retained by the Innocence Project of Texas. Prosecutors put Graf away for life after his two stepsons burned to death in a frame storage shed behind his house.

Yet the state arson investigator had little grasp of how fire burns and employed “old wives’ tales” in reaching conclusions, one expert wrote. Worse, there was no “crime scene” to examine, since the burned-up shed had been knocked down by volunteer firefighters and hauled off to a dump.

That's exactly what happened in the Willingham case, where probative evidence was simply hauled off to the dump and the main arson indicators cited by investigators are no longer considered valid.

Grits has no connection to IPOT's legal work on Graf's case and knows nothing more about it than has been reported in the press. But I've spent a great deal of time analyzing the revolution in arson investigation that's occurred since the early '90s, and feel strongly that the possibility of error in some of these older cases was high. For that reason, Grits agrees with the Morning News that "Graf, at the very least, deserves another day in court." And today's the day.