Re-Trial of Ed Graf’s Flawed Arson Case Begins in Waco

by Dave Mann Published on Monday, October 6, 2014, at 1:22 CST

Photo courtesy The Waco Tribune-Herald

Ed Graf in 1988

Ed Graf is once again on trial for a crime that many experts say he likely didn’t commit.

Jury selection for Graf’s re-trial on capital murder charges began this morning in a Waco courtroom. Graf was convicted in 1988 of starting a fire in a shed that killed his 8- and 9-year-old stepsons. He served 25 years of a life sentence before his conviction was overturned last year. Graf has maintained his innocence, and much of the physical evidence in the case supports his claims. Three nationally known fire scientists and the State Fire Marshal’s Office’s expert panel have examined the forensic evidence against Graf and concluded there’s no proof that Graf committed arson. And yet he’s once again on trial for his life.

In 2009, the Observer was the first media outlet to examine the flaws in Graf’s conviction, part of a series on faulty arson cases. Since Graf was convicted a quarter century ago, the field of fire investigation has advanced considerably. Many of the indicators investigators once used to distinguish an intentionally set fire from an accidental one—and to convict hundreds of people of arson, including Graf—have been disproven. Similarly, nearly all the forensic evidence that convicted Graf of arson in 1988, such as burn patterns in the wooden shed, has now been debunked.

At a January 2013 hearing, fire scientist Doug Carpenter eviscerated the case against Graf, testifying that, “There is no evidence to be able to formulate a valid hypothesis that this is an incendiary fire.”

In March 2013, the Court of Criminal Appeals, the state’s highest criminal court, agreed and overturned Graf’s conviction. At that point, it was up to McLennan County prosecutors to decide whether to free Graf or attempt to re-try him.

A re-trial seemed unlikely given that nearly all the physical evidence had been undermined and the prosecution’s own expert agreed there was no proof of arson.

In June 2013, the State Fire Marshal Office’s Science Advisory Workgroup—a panel of six experts examining old arson cases in the wake of the Cameron Todd Willingham fiasco—also concluded there was no evidence of arson. The State Fire Marshal’s Office sent McLennan County DA Abel Reyna a letter notifying him that the office had changed its classification of the Graf case from “arson” to “undetermined.” If there was no arson, then no crime was committed, and Graf is, by definition, innocent.

Yet Reyna and the DA’s office plunged ahead with a new trial of Graf, who, having spent 25 years in state prison, remains incarcerated in the county jail.

Prosecutors will have little physical evidence on their side. They will have to rely on the circumstantial evidence in the case, including the life insurance policies Graf took out on the kids that he sought to collect after the fire.

In that sense, the re-trial will test whether prosecutors can win a conviction with circumstantial evidence alone. It will also test just how far Texas has come in dealing with arson cases: Graf is the first disputed case to be re-tried since the Willingham controversy and since the state began its review of older arson convictions.

Given the stakes, the proceedings aren’t off to an encouraging start. The re-trial was delayed twice because prosecutors’ original files have gone missing. At a bizarre hearing last week, former DA Vic Feazell, who prosecuted Graf in 1988, traded accusations with his ex-wife about what happened to the files. Though attorneys on both sides have located copies of the files, the originals are still missing, and it’s unclear if any relevant evidence has disappeared as well.

Meanwhile, the re-trial has garnered national attention. National Public Radio will air a story on the Graf re-trial on “All Things Considered” this afternoon.

Prosecutors may begin presenting their circumstantial case late today or tomorrow morning. I’ll be at the trial this week and will post updates in this space.

Rhonda Trest

Dave Mann, who are you to say Ed Graf likely did not commit the crime? The circumstantial evidence was overwhelming. There is so much more to the story. Talk about biased reporting. Shame on you.
- Dave_Mann
  
  Hi Rhonda,
  Thanks for your comment. Please allow me to elaborate. I’ve been reporting and writing about this case for many years. It’s not me saying he likely didn’t commit the crime—it’s the many experts who have examined this case and found no physical evidence that points to Graf’s guilt.
  
  You’re right: There is a lot of circumstantial evidence. I reported extensively on that in my original story on the case:
  
  http://www.texasobserver.org/3062-victim-of-circumstance/
  
  The problem with circumstantial evidence, according to the attorneys and experts I’ve talked to, is that there are multiple explanations. Graf taking out an insurance policy may seem suspicious, but it doesn’t prove he did the crime. Neither does him returning their school clothes or not refilling the cereal and all the other pieces of circumstantial evidence. They could be coincidences. Those things can cast suspicion, but to prove a crime, you do need some physical evidence. Even at the original 1988 trial, would Graf have been convicted without the arson evidence?
  
  The key points of the prosecution’s theory—that Graf drugged the kids, locked them in the shed and poured gasoline—have all been disproved. The autopsies showed no drugs in their systems. The shed door had to be open or the fire likely would have died out from lack of oxygen. And the the fire scientists who’ve examined the evidence now say there’s no evidence of gasoline.
  
  So the key points of physical evidence that convicted Graf have been disproved. In the absence of any other physical evidence, is it not reasonable to conclude that Graf is, in fact, innocent? Several experts have told me that is the likely scenario. And the CCA doesn’t overturn convictions unless there’s something seriously wrong with the case, and the judges think the person is likely innocent. So that’s why I wrote what I wrote. I didn’t say he definitely is innocent—because we simply don’t know that for sure. But the lack of any physical evidence makes that the likely scenario.
  
  This was an unimaginable tragedy. This is very difficult case, and I appreciate that there are strong feelings on both sides here. I’ve tried to stick with the physical evidence—and the physical evidence seems to indicate that it’s unlikely Graf committed the crime for which he was convicted.
  - Rhonda Trest
    
    I have known all parties involved in the case personally since before the murders occurred. Ed Graf is as guilty as an be in my opinion. I don’t care how unlikely you think it is. There is no need for you and I to continue this particular debate because we will never see eye to eye. Unfortunately, I do agree that Graf may very well be acquitted. That will be another tragedy. God will be his judge one day regardless. I do not believe reporters should choose sides. That is why I wrote in the first place.
Chuck Bruce

I doesn’t matter what people’s opinions are…e.g., “Ed Graf is as guilty as can be in my opinion.” What matters is the evidence – the *science* has shown that the carbon monoxide levels found in the children’s blood was much too high for them to be dead before the fire (as proposed by the prosecution). The more likely scenario is that these two kids were playing with fire (which is something that 8 and 9 year old kids will do) in the closed shed and were overcome by the carbon monoxide from the fire that they probably started themselves. It’s terribly tragic and horrible – but it is more likely based on the evidence. At a minimum, the science (which is sound!) establishes a reasonable doubt in the guilt of Mr. Graf. In our judicial system, that should be enough get him acquitted.