Skip to content

Todd Willingham’s witch trial: the ignorant investigation

Cameron Todd Willingham with his daughter Amber

Cameron Todd Willingham with his daughter Amber, manipulating us from the grave.

I could handle being here for something I did, but to be persecuted like this for nothing I shall never understand. No God who cared about His creation would abandon the innocent.


Willingham was a monster. He was a guy who murdered his three children, who tried to beat his wife into an abortion so that he wouldn’t have those kids. Person after person has stood up and testified to facts of this case that quite frankly you all aren’t covering.


Well, never mind how many weeks months it’s been, I’m gonna try to pick up the thread from my last post. The theme is still the criminal justice system — how it fails and who it fails. It’s quite a can of worms, and for someone like me who’s had the good fortune to avoid much contact with the cops and the courts, it’s quite an education. What I’m writing should be read in that spirit, as notes on a continuing education.

Last time the eye-opener was Alan Gell — how little it took to railroad him onto death row and how much it took to get him out. But eventually the built-in safeguards worked. They worked and worked, in fact, for about 10 years while Gell relaxed in jail. But hey, it could’ve been worse. He could’ve been in Texas and ended up like Cameron Todd Willingham.

Until his house burned down a few days before Christmas 1991, Todd Willingham lived with his wife and three young daughters in Corsicana, a small town in northeast Texas. He’d gone back to sleep, he claimed, after his wife left the house that morning. About an hour later a shout from the two-year-old woke him up. The house was full of smoke. He yelled for her to get out and made his way to the childrens’ bedroom but couldn’t locate the twins, who were about a year old. All three girls died.

In “Trial By Fire,” the masterful account of Willingham’s case that ran in the New Yorker last September, David Grann describes how injustice was piled on top of catastrophe. Fire inspectors quickly concluded that the blaze was arson and that Willingham’s story of waking up and getting out of the house was a fabrication. He had no compelling motive for either arson or murder but the authorities decided that he was nonetheless “a man without a conscience whose serial crimes had climaxed, almost inexorably, in murder.” At his trial the prosecutor portrayed him as a monster. A jury agreed and sentenced him to death.

As time was running out for Willingham, a scientist and inventor named Gerald Hurst agreed to look at the forensic evidence used to convict him. Hurst’s work had been instrumental in freeing Ernest Ray Willis, another Texas death-row inmate, and he was amazed at how closely the two cases paralleled each other. He rushed out a report debunking the evidence against Willingham. This time it didn’t do the trick — Willingham was executed in Feb. 2004 — but his reports on both cases found their way to a couple of reporters at the Chicago Tribune. They consulted with several other fire experts and at the end of the year published the first exposé on Willingham. The Innocence Project and the Texas Forensic Science Commission followed with detailed reports analyzing both the Willingham and Willis cases. The unanimous opinion of the nine fire investigators who’ve reviewed Willingham’s case is that the forensics is riddled with scientific misconceptions and that no legitimate proof of arson was presented to the jury. It’s a sobering result, but in a way not so surprising. Comprehensive scientific standards for fire investigators weren’t published until 1992, and on many points those standards contradicted the conventional wisdom that had long circulated in the field.

The Willingham case got a burst of attention last fall in the wake of Grann’s article and the release of a report prepared for the Forensic Science Commission by Craig Beyler. It turned into a campaign issue for Texas Gov. Rick Perry, since back in 2004 he signed off on Willingham’s execution in spite of the report from Hurst. At the end of September Perry staged a heavy-handed intervention to rein in the Forensic Science Commission and push any serious consideration of the Willingham case off its agenda. It was a sleazy move, but it seems to me that the governor’s opponents were about as indifferent as he was to the commission’s actual purpose, not to mention the folks most likely to benefit from it — the poor saps who might otherwise be railroaded by “junk science” (there’s a fine post on Grits for Breakfast about the gap between what the committee was prepared to do and what everyone wanted to argue about).

Gerald Hurst

Gerald Hurst: “The proof that they used for arson… is in the report and in the testimony and that is not proof. So yeah, Willingham was innocent.”

The unavoidable hot-button question is the subheading of Grann’s article — “Did Texas execute an innocent man?” I like the way Gerald Hurst answered it on Nightline: “The proof that they used for arson… is in the report and in the testimony and that is not proof. So yeah, Willingham was innocent.” The evidence that the fire was intentional was the foundation of the case against Willingham, and if that evidence was grossly misinterpreted and misrepresented, then his guilt was not proved and he was legally innocent. No doubt it would be quite a coup if Texas was forced to acknowledge that. It’s only the admission that would be a breakthrough, though, since any argument about the innocence of a convict is, like the lottery, heavily rigged in the State’s favor. Ernest Willis managed to win the argument when the stakes were his life, and what he showed was that the courts in Texas are quite capable of sending an obviously innocent man to death row and of fighting like hell to keep him there until the bitter end. If Willingham was the first innocent man (or woman) they managed to execute, it was just dumb luck.

Nothing suits the law-and-order machine and its apologists better than a debate focussed on something so easy to doubt and impossible to prove as Willingham’s factual innocence. The physical evidence is long gone. The only thing that the original forensic work proves is that the investigators were, at best, misinformed and irresponsible — it would be giving them way too much credit to imagine that they proved anything about the fire. I don’t know what kind of evidence might be filed away in the archives, but most of the other evidence at the trial was part of a narrow-minded effort to scare up suspicion and loathing. One thing that it’s good for is undermining any claims that Willingham was factually innocent. Otherwise it stinks, and it always has — it’s never been necessary to get into the forensic details to pick up the foul smell this case gives off. That’s the argument I’m going to try to make, anyway.

It’s true that the legal system is way outside my area of expertise (whatever that is), so I might be missing something big. Maybe I fundamentally misunderstand what counts as proof or what counts as justice. I’d worry about it more if the authorities who built and prosecuted the case didn’t act like justice was just a matter of tossing out a piece of human garbage in the quickest and most self-righteous way possible. They seem incredibly narrow-minded, and as far as I can tell they were completely untroubled about compounding the misery of a family that was already deeply traumatized. The ultimate risk, if they made a false charge stick, was that a man would be executed merely for surviving the horrible accident that killed his children — a truly obscene miscarriage of justice. So it seems like they’d embrace their burden of proof beyond a reasonable doubt, if not out of common decency then because of the prosecutor’s statutory obligation “not to convict, but to see that justice is done.” They couldn’t honestly expect Willingham’s meager defense to hold them to it. But no, they did their very best to make all the reasonable doubt just go away — they put on a witch trial.

The fire inspectors set the whole process in motion and they gave the proceedings a veneer of rationality. Actually, though, they were priestly figures, and when they conjured up the crime and the defendant’s guilt, it was presented and accepted as a matter of faith. The crime was horrific, too — little children intentionally burned alive, which happens to be the kind of thing witches do. To make it seem plausible, the jury was offered a lot of highly suggestive testimony that was supposed to show that Willingham wanted to kill his children and that he stood by while they died their gruesome death. It was, in effect, a portrait of the defendant as an evil and abnormal person — the kind that does horrific things for no particular reason. Of course in modern-day Texas a man like that isn’t a witch, he’s a monster or a demon, but same difference.

It was a banal little bush-league witch trial, but not a purely metaphorical one — there’s a real undercurrent of superstition. Grann describes how the lead prosecutor, John Jackson, gave the faith-based forensics a bible-belt twist:

During his closing arguments, Jackson said that the puddle configurations and pour patterns were Willingham’s inadvertent “confession,” burned into the floor. Showing a Bible that had been salvaged from the fire, Jackson paraphrased the words of Jesus from the Gospel of Matthew: “Whomsoever shall harm one of my children, it’s better for a millstone to be hung around his neck and for him to be cast in the sea.”

On Nightline last fall, Jackson was more specific about what Willingham burned into the floor — it was, he said, “perhaps a pentagram kind of a figure that some people associate with devil worship.” In a letter to the local paper a few weeks earlier, Jackson pointed out that Willingham refused the “opportunity” he was offered to “eliminate himself as a suspect by polygraph examination.” A more fitting way for him to prove his innocence, which wasn’t in any meaningful way presumed, would have been to offer him a millstone and a body of water. He would have either sunk to his death or, if he was really a devil-worshiping monster, maybe he would have floated. That would have been the same kind of justice he got, just quicker and more humane. But the general sentiment was that Willingham had already failed his life-or-death test when he didn’t go back into the burning house for his kids.

Inside Todd Willingham's house after it burned
The front of Todd Willingham's house after it burned

Inside and outside the house after the fire (pictures from the Texas State Fire Marshal’s office, by way of the Dallas Morning News).

I’ll get back to the trial, but I want to stick with the inspectors for a while. Unless their critics are completely misrepresenting the science of fire, their gross scientific misconceptions are undeniable. That’s gotten plenty of coverage already, so I won’t get into the specifics (here’s an overview and some links, though). The poor understanding of fire was apparently fairly typical of inspectors back when Willingham was tried. The culture within the field must have been good at explaining away challenging information, unfortunately at the expense of the people who came under investigation. I suspect that an attitude like that is one thing that makes junk science so pernicious and hard to stamp out.

As the lead inspector in the Willingham case shows, it’s just the attitude for witch hunting. A local firefighter, Doug Fogg, did the initial inspection. Based on what he saw, a specialist named Manual Vasquez was sent in from the Texas State Fire Marshal’s Office. Craig Beyler’s bottom line on Vasquez is that “[his] opinions are nothing more than a collection of personal beliefs that have nothing to do with science-based fire investigation.” Fogg had pretty much the same beliefs, but arson investigation wasn’t his main gig, so it’s not so alarming that he relied uncritically on the conventional wisdom he’d been taught. Vasquez’s personal beliefs about fire had been reinforced by over a thousand investigations. They were articles of faith and all indications are that his close-minded self-assurance made him quite persuasive.

For the science-minded reviewers Vasquez’s testimony raises one red flag after another, and not just because of all the misconceptions about fire. Early in his testimony, as he was describing his experience and qualifications, he said that almost every fire he’d investigated had been arson. The expert reviewers have been uniformly incredulous of the claim — at the time his peers were finding about half of the questionable fires in Texas to be arson. Although he made it clear that most any burned-out building whispered “arson” in his ear, Vasquez’s conclusions were never seriously challenged in court.

Later in the testimony Vasquez presents himself as a kind of clairvoyant and the red flags go up again. A line of Beyler’s that’s gotten a lot of play describes Vasquez’s attitude as “hardly consistent with a scientific mindset” but “more characteristic of mystics or psychics.” The Lentini report criticizes the same passage as a dangerously deceptive performance of expertise in the courtroom (my emphasis):

Mr. Vasquez indicates that he understands the nature of expert testimony: that of interpreting fire artifacts for the jury. At page 244 [of the trial transcript], he states:

“The fire tells the story. I am just the interpreter. I am looking at the fire, and I am interpreting the fire. That is what I know. That is what I do best. And the fire does not lie. It tells me the truth.”

Unfortunately for Mr. Willingham, while the fire may not have “lied,” Mr. Vasquez misinterpreted what it was telling him. Such willingness to offer “expert” testimony, while lacking the knowledge to present accurate information to the jury, may excuse Mr. Vasquez’s many serious errors. The judicial system that allows such testimony to be presented, however, is clearly flawed and in need of reform.

The fire told Vasquez an awful lot. Not just that it was started in three places with a liquid accelerant but that it was Willingham who started it and that he did so in order to kill his children. The fire told Vasquez that it wasn’t the cause of Willingham’s injuries so he concluded they must have been self-inflicted — Beyler calls the diagnosis “remarkable” and concludes that “Vasquez seems to be wholly without any realistic understanding of fires and how fire injuries are created.” Because it was so “aggressive,” the fire even told Vasquez that it “was not a planned fire. It was a spur-of-the-moment fire.” That, for Beyler, is a degree of “mysticism” that’s “beyond belief in the context of fire investigation as an applied science.” It seems to me that’s putting it pretty tactfully.

As Vasquez was uncovering the tell-tale signs of arson he was also unmasking the arsonist. Grann describes the moment when Vasquez’s quasi-mystical certainty encountered Willingham’s confused and defensive attempts to explain himself. The inspector knew what he knew and had no qualms about dismissing Willingham as a liar. It’s a moment that epitomizes how ignorant and unprofessional the whole investigation was.

During the interrogation, Vasquez let Fogg take the lead. Finally, Vasquez turned to Willingham and asked a seemingly random question: had he put on shoes before he fled the house?

“No, sir,” Willingham replied.

A map of the house was on a table between the men, and Vasquez pointed to it. “You walked out this way?” he said.

Willingham said yes.

Vasquez was now convinced that Willingham had killed his children. If the floor had been soaked with a liquid accelerant and the fire had burned low, as the evidence suggested, Willingham could not have run out of the house the way he had described without badly burning his feet. A medical report indicated that his feet had been unscathed.

Willingham insisted that, when he left the house, the fire was still around the top of the walls and not on the floor. “I didn’t have to jump through any flames,” he said. Vasquez believed that this was impossible, and that Willingham had lit the fire as he was retreating—first, torching the children’s room, then the hallway, and then, from the porch, the front door.

This might be a good thing to remember if you’re ever tempted to explain yourself to the authorities without an attorney. In court, the story must have weighed heavily against Willingham. It left the jury with a memorable image — Willingham’s unscathed feet — and it also burnished Vasquez’s cleverness and his authority. To the Lentini committee, though, the simplistic assumptions behind his little trap were sadly familiar.

Fire investigators who reach false conclusions, then hear descriptions of events from fire survivors that do not comport with their conclusions, frequently have testified that only the killer or the arsonist has a motive to lie. The undersigned investigators, having been involved in cases of fires misattributed to arson, are familiar with this phenomenon. Mr. Vasquez first formed the conclusion that the fire was intentionally set. Then he was allowed to tell the jury:

“I’ve talked to the occupant of this house and I let him talk and he told me a story of pure fabrication.”

Mr. Vasquez’s only basis for reaching that conclusion was his own misinterpretation of the meaning of the fire artifacts that he observed.

Willingham did lie, there’s no doubt about it. Not about getting down the hall in bare feet — that was Vasquez’s fabrication. But shortly before he was killed Willingham admitted that he hadn’t really gone into the children’s bedroom as he claimed. It’s the kind of lie an ordinary man might tell out of fear or pride, knowing he’d been scrutinized and judged from the moment he was spotted outside of his burning house. As far as I’m concerned that’s a lot more plausible than the outrageous lies he supposedly told to hide his monstrous guilt. It was a foolish lie, of course, and I’m sure it cost him. A decent advocate might have helped him get his story straight and saved him from himself.

If Willingham had an advocate it might have saved the investigators from their own worst instincts, too. Willingham was the only person alive who could have seen what was going on inside the house during the early stages of the fire, so when Vasquez wrote Willingham off as a liar he wrote off one of his strongest reality checks. It was Vasquez’s job to be skeptical, of course. Healthy skepticism is a two-way street, though, and there’s no sign that Vasquez had any for the story the fire was telling him — if he did, he didn’t act on it in any useful way. Instead he decided, in effect, that the chances he and Fogg were wrong were less than the chances that their prime witness — a man they knew very little about — was a liar who woke up and for no apparent reason torched his house while his children were inside. My sense is that it’s much more likely for a run-of-the-mill expert to be wrong than for a typical or even a questionable witness to be an infanticidal monster. You can see what a great racket junk science is, though. Don’t let anyone challenge your precious theory just because they saw what happened! Call ‘em a liar, send ‘em to jail, get ‘em executed! That’ll take care of ‘em, huh?

Willingham’s story wasn’t the only serious challenge to the investigators’ theory of the fire. Laboratory tests found no traces of the all-important liquid accelerant anywhere but the front threshold, where there was an alternate explanation. The prosecutor if not the investigators was bothered by this. Jackson told Grann that “he ‘never did understand why they weren’t able to recover’ positive tests” from other places that were supposedly doused with lighter fluid. Apparently he was unwilling to draw the obvious conclusion, though. Also, Vasquez’s scenario was contradicted by the first eyewitnesses, who looked straight at the porch and through the front door and saw no flames. These are not small problems.

It’s also hard for me to believe these investigators had a good case because I haven’t come across a single word from them about how Willingham went about setting the fire and how it progressed from there. There was no way they could say exactly what happened, but if they had a robust theory they should have been able to say something concrete. There was an hour or so after Willingham wife left before neighbors saw him on his porch — plenty of time to douse some carpet with lighter fluid and light a few matches. But where did he get the lighter fluid? How much would it take to make the puddle shapes the inspectors found? What would happen when he lit those puddles and how long would it take to develop to the stage firefighters observed when they got there? When (if ever) would someone look at a house set on fire that way and say it was just “smouldering”? One inspector suggested that the entryway was set on fire in a way that’s “typically employed to impede firemen in their rescue attempts.” How did he know to do that?

Naturally, coverage and criticism of the case focusses on the fire inspectors but the police were busy, too. Beyler notes in passing that they interviewed about 40 people and collected “information about Willingham’s arrest history, his relationships with others, the dynamics of the household, and his past in general.” In other words, they were digging up dirt. Perhaps that was appropriate, but it seems they did little else.

The investigators didn’t tax their brains working on this case. They didn’t have to. All they needed was enough evidence to charge Willingham with a crime and then to make him seem guilty, which is a lot easier than proving that he committed the crime. They didn’t have to worry about much of a challenge from Willingham’s bargain-basement defense. The way they approached the case, in fact, less is more. It’s best not to get too real about the details — setting a fire, for instance, or waking up in a burning house — because that might get people thinking. In a witch trial, ignorance is strength.

{ 5 } Comments

  1. Dan Weber | July 4, 2010 at 12:01 | Permalink

    I think they just referenced this on Law & Order: SVU last night. Not by name, of course.

    ~   ~   ~

    Yeah, it was a rerun. I haven’t seen the episode, but there was a little buzz about it back in April.

  2. nick | July 6, 2010 at 02:57 | Permalink

    perhaps you would “enjoy” the documentaries on the west memphis 3 (named paradise lost; lilly lib used to have them). that was a witchhunt.

    ~   ~   ~

    Hi Nick. I just took a quick look on wikipedia, and I see what you mean. I’m not sure that anything beats the day care sex abuse/satanism trials — for me, those are the classic modern witch trials — but it seems like there’s plenty of competition.

  3. Chris Halkides | July 14, 2010 at 15:48 | Permalink

    I think that this is an excellent overview of the case. It gets to the heart of the story without getting lost in every forensic detail. Please allow me to share, however, one tidbit I found while researching the case. Todd Willingham told the investigators that they might find traces of cologne he sprinkled in the house because his children liked the smell. Recently, some, including at least one reporter, have interpreted his statement as meaning the cologne was the accelerant.

    However, someone who read the police report told me that Mr. Willingham sprayed the cologne after the fire, not before. Yet when I mentioned this to another reporter, she said that maybe Willingham was using the cologne to cover up some other accelerant. As a chemist but not an expert in arson, it unlikely that such an attempt to mask an accelerant would work. I don’t know why Mr. Willingham sprayed with cologne, but maybe it was his private way of sending them off into the next world. It seems as if people still want to twist every little detail of the story against Mr. Willingham.

    ~   ~   ~

    Thanks, Chris.

    I’m confident that the cologne was spread after the fire. I don’t have the police reports, but the Corsicana fire chief summarizes some of them in his response to Beyler. This is on page 6, based on a conversation between a firefighter and Willingham at his burned-up house “several days after the fire”:

    Fireman Franks also said that after the fire, Mr. Willingham came back to his house and poured a large bottle of British Sterling cologne on the floor from the bathroom to the room in which the twins had died and said that if any more samples were taken from the floor, those samples would have cologne on them.

    I agree about how every detail is twisted. In fact, I believe it was during the same visit that Fireman Franks decided that Willingham cared more about his dart board than his children — an especially rotten piece of “evidence,” it seems to me. The way the cologne is brought up is typical of the case against Willingham — long on suspicion and short on proof. In the debates kicked up by Grann and Beyler it’s more of the same. I don’t think anyone captured the spirit like our old buddy RRH, in his comments on your blog but most especially on the Texas Death Penalty blog.

  4. wayne fontes | July 25, 2010 at 09:46 | Permalink

    One peculiar feature of the daycare sexual abuse/satanism trials is that it provides an opportunity for the it’s all about the children left and the law and order right to unite on an issue. It requires some real guts to stand up and support the alleged purp in these cases.

    HBO made an excellent movie about the McMartin day care trial. They still play it occasionally and I recommend it.

    You might try reading some of the essays Publicon has written at Chez Odysseus. While you might not agree with the content or conclusions I’ll think you would admire the effort he puts into his posts.

    ~   ~   ~

    Thanks for the tips.

  5. Greg W | July 30, 2010 at 14:42 | Permalink

    Maybe want to read this.

    I have read these reports, and in no way has the prosecution been debunked. Each individual piece of evidence was taken apart and rebutted out of context with the whole case. In fact, all that has changed is terminology of fire investigations. Even though Vasquez called it a “pour pattern” and it is now called an “irregular pattern” NFPA 921 says it may be called a pour pattern if accelerant is found.

    No, cologne would not come back as mineral spirits, which is what the lab found. Without a canine to point out evidence, selecting samples for the lab is often hit or miss, with known accelerants being missed.

    As for Stacy, she believes Willingham did it.

    ~   ~   ~

    Here’s the problem. I’m not a fire expert so all I can do is decide which experts are credible and which arguments are convincing. On one side in this case are the critics who write long reports that carefully explain each point with references to the literature. On the other side are the defensive reponses of people like Doug Fogg and Donald McMullan, which pick a few points here or there and add some ad hominem pot shots at the critics. They explain nothing and avoid all of the basic scientific issues.

    If I used astrology to “prove” something, the counterargument would be that astrology is bullshit. There’s no need to rebut each piece of astrological “evidence” (the alignment of the planets or whatever). What Beyler and the other “latter-day experts” have been critical of is the method and broad basis of the conclusions in the Willingham case. An effective rebuttal — one that I’d find effective, anyway — has to address the forest, not just a few trees. The critics, its true, make their forest tree by tree and a convincing rebuttal could do the same. If someone with roughly the credentials of Beyler, Lentini, or Hurst took the evidence and the criticism point by point and explained in credible detail things like the failure to find accelerant anywhere but by the door, that might be convincing.

    What I see instead these random issues like the cologne, which as far as I can tell is a red herring. The story about W pouring it sounds suspicious and that’s all there is to it. Would it be an effective accelerant? How much would it take to make the “pour patterns,” etc.? Would the normal battery of lab tests have found it? Was it detected in the lab tests? Why was it not brought up (apparently) in either Fogg’s or Vasquez’s reports? If he poured it after the fire, why should we care? etc. etc.

    I’ve read Stacy’s statement and it’s no more credible than the testimony of the jailhouse snitch at the trial. Grits for Breakfast took on the unpleasant task of calling Stacy a liar and makes a fine case.