William Haughey walked out of Wallkill Correctional Facility on May 9, after serving eight years and four months of a ten-year prison term. But it would be another two weeks before he felt any real sense of freedom.
On May 23, Haughey, who had been convicted of setting a fire in a Putnam County tavern eight years prior, was officially exonerated of the crime. As a court order declared, “It appears that the defendant has spent years in prison for helping to put out a fire, not start one.”
In recent decades, advances in forensics science have exposed faulty methodology in fire investigations and punctured some of the most commonly accepted assumptions about what was considered arson evidence
Burn marks on the floor, for instance, were previously thought to indicate the use of accelerants such as gasoline, kerosene or diesel fuel. However, it is now believed that flashover, the point at which an entire room reaches ignition temperature and becomes engulfed in flames, can cause such marks.
Similarly, a pattern of cracks in a window, known as “crazed glass,” was thought to be the product of rapid heating and a sign of arson. It has since been shown that “crazed glass” is a result of rapid cooling, such as when a firefighter’s hose hits a hot window.
In Haughey’s case, the arson determination was made after it was believed that electrical causes of the fire had been ruled out. However, the initial investigation was deemed “fundamentally flawed” in a recent court order. Subsequent fire investigators who examined the case post-trial ruled the cause of the fire to be undetermined.
In the past, fire investigators often relied on a so-called “negative corpus” approach when determining fires to be incendiary in origin. Negative corpus refers to a process by which potential causes of a fire are eliminated until only one cause remains.
However, that approach has come under scrutiny.
“You should not have a lot of confidence in fire investigation,” says John Lentini, an arson expert and principal investigator at Scientific Fire Analysis in Islamorada, Florida.
“You should especially not have a lot of confidence if the arson determination is based on the elimination of all accidental causes and not on finding affirmative evidence of arson.”
In 1992, the National Fire Protection Association (NFPA) issued its Guide for Fire and Explosion Investigations, a resource for science-based investigation and analysis. Updated versions of the guide have been issued every three to four years since. In recent editions, the NFPA has discouraged the use of negative corpus methodology, stating “it is improper to opine a specific ignition source that has no evidence to support it even though all other hypothesized sources were eliminated.”
While the NFPA guide has become a widely adopted resource, the methodology used in past fire investigations remains questionable.
Perhaps most notable is the case of Cameron Todd Willingham, who was convicted and sentenced to death for the 1991 arson murder of his three young children. Days before Willingham’s scheduled execution, renowned arson expert Gerald Hurst issued a report stating that most of the fire marshal’s conclusions in the Willingham case “would be considered invalid in light of current knowledge.”
The state of Texas executed Willingham in 2004.
Earlier this year, Lentini of Scientific Fire Analysis testified for the defense in a strikingly similar case involving Daniel Dougherty. Dougherty, who was convicted in the arson murder of his two children, was awarded a new trial after it was determined he received ineffective assistance of counsel at his original trial in 2000.
In April, Dougherty was convicted of the crime a second time. After the verdict was announced, Dougherty shouted at jurors, “You got it wrong!”
Despite the bad science used in past fire investigations, overturned convictions remain relatively infrequent in arson cases. While convictions that rely on DNA evidence can be overturned with a lab result, arson convictions often rely on human judgment in the form of expert opinion, which can, in turn, lead to human folly.
“It’s a whole lot easier if you’ve got biological evidence than if you’ve got two experts-one of them saying this isn’t arson and another one saying yes, it is,” says Lentini.
Haughey was assisted in his exoneration by The Jeffrey Deskovic Foundation For Justice, a non-profit organization that investigates both DNA and non-DNA cases of individuals believed to be innocent.
“Wrongful convictions aren’t rare,” says Deskovic, who was exonerated of a rape and murder after spending 16 years in prison, and started his foundation with money he received in a settlement from that case.
“I wish that they were. The rarity is that the exonerations happen.”
Particularly, it would seem, in arson cases. Of the 1,825 exonerations listed by The National Registry of Exonerations, only 44 involve crimes of arson.
Efforts are now underway to enhance fire investigators’ understanding of ventilation’s impact on fires and post-flashover fire behavior. The U.S. Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) regularly hosts training sessions that include field exercises during which investigators are asked to determine the origin of a fire set by trainers.
At a 2005 training seminar the ATF held in Las Vegas, 53 students were asked to determine the quadrant of origin for two identical fires set in different locations. Only three correctly identified the quadrant of origin for the first fire. The exercise was repeated with the second fire; once again only three students, though a different three from the first set of students, correctly determined its quadrant of origin.
“If you get the origin wrong, you’re going to get the cause wrong,” says Lentini. “That still happens fairly frequently.”
Haughey has left Putnam County since his release and now resides with family in Florida. He has no other immediate travel plans.
“This is my vacation,” says Haughey. “Anything away from jail is a vacation.”
Brooke L. Williams is a freelance writer based in New York City. She reports on the criminal justice system with an emphasis on those wrongfully convicted. Follow her on Tiwtter at @williamslbrooke