With the generous support of the JEHT Foundation, the Center has launched an Arson Screening Project, designed to:
- Survey and analyze the range of “Arson” convictions that in fact involved accidental fires,
- Forward meritorious cases to a blue-ribbon Fire Science panel, led by Center Advisory Board member John Lentini,
- Report to scientists and practitioners the depth and extent of the problem and develop targeted remedies.
A series of notorious arson convictions—for example, the case of Cameron Todd Willingham, who was executed by the state of Texas in 2004—has been analyzed in light of modern scientific findings in the chemistry, physics and thermodynamics of fire.
The results of this inquiry are startling: they reveal that at least a generation of American “arson” cases has been investigated by officials poorly trained in fire science who innocently relied on obsolete professional dogmas that attribute to meaningless features of fire scenes the infallible power to reveal “set” fires. The endemic use of this erroneous folk-science of fire scene indicators such as the presence of “spalling” or “crazed glass” has generated hundreds of arson findings, which have triggered hundreds of prosecutions, supported by hundreds of instances of compelling but completely mistaken “expert” testimony, that easily convinced hundreds of juries that fires that were in fact accidental were criminally set. As one panel of experts put it, “The significant lack of understanding of the behavior of fire…can and does result in significant misinterpretations of fire evidence, unreliable determinations, and serious miscarriages of justice with respect to the crime of arson.” Arson Review Committee, Report on the Peer Review of the Expert Testimony in the Cases of Texas v. Willingham and Texas v. Willis, at 40.
A 2002 compilation by the Bureau of Justice Statistics (a study that surveyed only half of the states) counted over 5,000 prisoners serving arson sentences. At least a dozen arson convictions have generated death sentences. There is no question that many of these arson convictions and sentences rest on false science: that many “arsons” were accidents. The Innocence Project, with which the Center for Modern Forensic Practice has consulted closely in developing this initiative, reports that it already has a backlog of arson cases deserving scrutiny. (The Innocence Project restricts its own direct representation efforts to cases in which biological evidence can provide a conclusive answer, and biological evidence does not figure in arson investigations. The Innocence Project cannot undertake the review of arson cases, but it would refer the many inquiries it receives to the proposed Project.) There is no question that in America’s highly localized system of criminal justice bad science still mesmerizes officials and jurors and sends innocent men and women to prison. Arson findings are still routinely based on criteria that one expert describes as “A hodgepodge of old wives’ tales.”
The mistaken “arson” cases represent a distinctive class of challenges (and opportunities) to criminal practice born of the collision between advancing science and legal system tradition. In one important respect this class of challenges contrasts with the more familiar class of cases developed by advances in DNA technology. The DNA findings that undercut existing convictions simultaneously delivered a new, final answer: the convicted inmate didn’t do the crime. When new science proves that an arson conviction rests on false science, it does not categorically answer the central “arson/not arson” question; it states only that the prior answer was derived by invalid means. Answering the ultimate question—and the new waves of similar questions that will be raised by scientists studying other forensic science applications such as bullet composition and pattern evidence—requires an exhaustive inquiry and the dedication of substantial criminal justice system resources.
Both justice in individual cases, and accuracy in system-wide reform of classes of cases in which existing assumptions have been undermined by modern forensic science findings will depend on developing a new approach to applying science retrospectively to convictions. The Arson Screening Project will explore how that challenge can be met most effectively, while contributing to just results in individual cases and disseminating its findings broadly throughout the criminal justice and allied communities.
The challenge is one of mobilizing and applying new (but existing and accepted) science: the pace of integration of this science into practice has been glacial. A striking feature of the arson problem is that a novel understanding of the science of fire origin and behavior does not need to be discovered; a consensus on a new science firmly based in physics, chemistry and thermodynamics has been available since the publication by the National Fire Protection Association of its landmark NFP 921 in 1992. A related consensus view of appropriate fire scene evidence gathering practices has been available since a Technical Working Group of the National Institute of Justice published Fire and Arson Scene Evidence: A Guide for Public Safety Personnel in 1990.
The Arson Screening Project aims to provide the missing scientifically and legally critical “front door” to the review process. This is a crucial function. Review by a panel of qualified experts can identify those cases in which false science was employed. A cadre of volunteer experts, organized as The Arson Review Group, is now available to review suitable cases under the leadership of John Lentini. This team of experts, however, is drawn from a small pool, and represents a terribly scarce resource. This fact presents an enormous challenge, since arson cases (unlike, for example, DNA cases) cannot be resolved conclusively by a bare laboratory analysis. Authoritative review is extremely time-consuming. This challenge is compounded by the fact that all arson convictions that include mistaken expert scientific testimony are not unreliable. It would be a fatal waste to exhaust an expert scientific review panel’s limited capacity on cases where the science was bad but the defendant is irrefutably guilty, when there are so many meritorious cases awaiting analysis. As things now stand, there are only two alternatives: allowing dozens (or hundreds) of convictions based on false science to stand, or drowning a handful of volunteer experts under a tidal wave of cases, many of which are without merit.
The arson cases present a different but important challenge at the end of the case Review process. Final Arson Review Group investigations that expose unreliable convictions have generated headlines in local jurisdictions, but those headlines have provided only episodic impetus for reform. There has been very little sustained momentum toward reform derived from the exposure of individual wrongful arson convictions, because the arson modernization movement lacks the central focal point that the Innocence Project has been able to provide in DNA exonerations in eyewitness misidentification, false confession, or laboratory error cases.
If You Think You Know of a Bad-Science Wrongful Arson Conviction
The Arson Screening Project will not provide legal representation. It will, however, offer an assessment of the science used to gain a conviction which can be employed in trying to persuade prosecutors, lawyers, and courts that a given case is worthy of review. The Project is not designed to assess “whodunit” cases; it is designed to offer a scientific analysis of whether the crime of arson was committed, not to identify arsonist or rule out a defendant in case where a fire was set.
If you believe you or a client have a case that would benefit from authoritative review, please:
- Review the Center’s Arson Screening Application For Evidence Review Form. Please do not send materials to the Project until you hear from us. We will ask you to submit the whole record of the case in one package once the case is accepted for review. Remember, the Project is not providing legal representation, and although we will treat submissions carefully we cannot offer the protections of an attorney-client privilege to our correspondents.
- If you believe that you have a case that should be reviewed, please complete the application form and send it to us.
- We will contact you, and either notify you that the case will be reviewed once certain requested materials are collected, or that we cannot offer review. (If a case is turned down, we will supply the reasons, so that corrections if any can be made.)