The failure of DNA forensic testing: a response to Hill

Hill's response to my paper is essentially a manifesto on the infallibility of science and the scientific method of inquiry and a defense of the right for it not to be questioned. My standpoint was that all processes during the 2009 Australian bushfire disaster should have been open to public scrutiny, including the forensic process. My paper was written from a public policy perspective and was based on a systematic analysis of information that was in the public domain. The “open and transparent” processes to which Hill referred severally were “Coronial Identification Boards” and the “Commission of Enquiry”. The former were private and internal processes that were not open to the general public. The latter public inquiry's stated terms of reference included matters of causation, planning, preparation, management and emergency services responses, but did not investigate “the objectives and undertakings of the DVI process” as Hill implied. His main issue seems to be what he regards as my ignorance of the internal workings of the forensic process, and therefore the inappropriateness of my commentary. He also criticized my use of terminology, particularly in relation to Hilgartner's (2007) model used for the analysis of public disasters. Furthermore, Hill interprets my paper as a criticism of the work he and his colleagues do. I will respond to these issues in the order in which they were made.

Science and the scientific method

In relation to Hill's first point about the Indonesian tsunami, I questioned Fitrasanti and Syukriani's (2009) interpretation of events and the conclusion they reached. I argued that religious and cultural beliefs of the people were stronger than, and therefore prioritized over, their belief in science. In other words, the strong religious imperative to bury the dead in a timely manner, together with the physical recognition of a family member victim, overrode their need to await the results of scientific tests. I contrasted this with the cultural trust in science that Australians have, which supersedes cultural and religious beliefs as well as confidence in their own meaning-making capacity. I did not say that forensic identification should not be used, although I did imply the importance of physical recognition as one crucial form of evidence used to identify deceased persons, something that Hill seems to dismiss entirely. While scientific validation is important, it is by no means the only way we know things. My paper specifically questioned the status and limits of scientific knowledge in the particular setting of the 2009 Victorian bushfire disaster where the facts of death were known to kin. Knowledge of the location of the victim at home defending their property, as dictated by the “stay and defend” policy and confirmed by real-time telephone contact, was not second-hand knowledge or hearsay; it was crucial information that could have directed the necessary confirmatory scientific tests. I did not suggest that this information be used instead of forensic tests, but proposed that it could have informed the process rather than be used only when identification had “eluded scientific tests” (Coroner, quoted in Bachelard 2009).

Internal inconsistency about what evidence was used and what was not

Hill's suggestion that “the faith of the community in the rigors of scientific investigation would have been lost” is problematic on two counts. In the first instance he seems to be conflating the rigor of the scientific method (or process) with scientific testing. A rigorous scientific process would systematically take account of all available evidence, including the use of circumstantial as well as empirical and scientific evidence. Instead, he seems to want the public to maintain a blind faith in the clinical science of popular fiction rather than have access to the internal workings of the forensic laboratory. In other words, inside the “black box” of science (Latour 1987) resides a world of uncertainty, hypothesis testing and interpretation that stands in stark contrast to the black and white certainty of science as understood by the lay person. The second related point is that Hill himself claims (and then refutes later on) that circumstantial evidence was ultimately taken into account but neither he nor the coroner name which type of evidence was used and, importantly, which was not. For example in relation to the first case study, Hill states that “all of the information supplied by [the mother] – circumstantial, geo-location, e-communication, dental information, DNA sample – was used in the identification of her son.” Two paragraphs later he refutes this statement claiming that “‘circumstantial evidence in the form of geo-location of bodies and 'witness accounts’ was not used – standards and protocols could not be compromised” (my emphasis). The point of interest here, as well as the internal inconsistency, is: what was used? We are not privy to the internal machinations of the “black box” so we simply do not know. In the absence of this, my reading is that, of the “primary” methods (“fingerprint, dental or DNA”), the dental ante-mortem matching looks like the best contender, given the intensity of the fires. Of the “secondary” methods (“age, sex, ethnicity, scars, medical intervention, clothing, jewelry and documentation”), the anthropological evidence but not material evidence may also have been available for use. The secondary, circumstantial evidence specifically listed in the disaster victim identification (DVI) protocol would not have been readily available because of incineration. My contention was that the strong objective evidence of identity that was available should have been used. Location, together with real-time e-witness accounts, was crucial information that could have directed the identification process along with the scientific information from dental and bone data. But I am only guessing because Hill did not enlighten us on this point.

The next point of obfuscation in the forensic process was the use or non-use of DNA evidence. Given that these are time-consuming tests and the reason given (and accepted) for lengthy delays in identifying the bodies of the victims, it is important that we know whether DNA tests were used, and if so, what kind of test was used. The point is we do not know and it is insufficient to claim that the “faith of the community in the rigors of scientific investigation would have been lost” if full information were made public. It is in the public interest that the forensic process is transparent. Was, for example, the supply of DNA samples from grieving families necessary when other knowledge of certain death was available and could be confirmed by more traditional forensic methods? The point here is that the public have a firm trust in DNA testing and, without information to the contrary, are left to believe that this was how victims were identified.

The Victorian Bushfire incident as an “open disaster”

I agree with Hill that in the very initial stages this could be defined as an open disaster of the kind that he describes but my argument was that the DVI protocol needed to be flexible enough to readjust to facts as they became known. The reality was that the interim report into the fires clearly stated that 113 victims did die in their own private homes, 27 others outside defending their house, six in their garage and one in a shed on his own property (Teague et al. 2009), so it certainly was not “open” in the same way as were the other disasters (tsunami, Bali, etc.) to which Hill refers, as they were in public places and both residents and tourists were significantly relocated by the turbulent water or bomb blast. In other words, they were displaced and relatives would have had no certainty about where they were at the occurrence of the disaster or to where they had been relocated. From my understanding, fire does not displace people in the same way as do other disasters. The main critique I put forward was about the inflexibility of the protocol to adjust to the uniqueness of this, as a different kind of disaster. The exceptionality of this bushfire disaster was that there was a compelling reason for people to be where they were. The remains found on private property were most likely to be residents of that property. This was validated by information that was available to significant others in real-time communication. The difference in procedure can be illustrated by a single fire fatality in a suburban home. The traditional starting point of investigation would be to make a provisional assumption that the owner or usual occupant was the most likely victim. Questions would be asked of witnesses to strengthen or refute this hypothesis. Alongside this process, forensic material would be gathered and tested. This is quite the reverse of the process to which Hill refers where victims were relegated to the “unknown” category as a starting point. My point was, had the forensic process started with the geo-location in which bodies were found and the likely assumption that the victim was the owner or occupier of the property, then the process would have been quite different and perhaps not so time consuming and distressful for families whose loved one was relegated to a “missing” category.

Terminology: “missing” versus “unknown,” objectification and “inflammatory” terms

Hill took objection to what he claimed to be my confusion of the terms “missing” and “unknown.” I used the term “missing” in exactly the way he described and how the family members in the case studies did. I did not use the term “unknown” at all in relation to the bodies of victims (that is his term) but I did imply this distinction when I referred to the process of identification where the bodies had been disentangled from their social and environmental connections so that they could be examined objectively with “unknown” as a starting point. Hill took exception to my reference to the body of the victim as “an object of scrutiny.” I am unsure how to respond to this being “despicable” as the forensic scientific method is based on objective empiricism. Hill seemed to equate the idea of the body being an object of scrutiny or inquiry with some form of disrespect. My argument here was that bodies, as well as being the objects of scientific inquiry to be objectively read, were also subjectively linked within a network of social relations and to a particular place, something which was ignored until the objective processes failed.

In a similar manner, Hill refers to Hilgartner's (2007) model for analysis of public policy regarding disasters as “inflammatory” and which “smacks of conspiracy, manipulation and a hidden agenda on behalf of authorities.” It is surely the role of governments to manage calm and stability during times of disaster and to do this they need to “frame up a narrative” or tell an account that maintains public trust in their ability to deal with the situation by pre-empting panic and ensuring the restoration of calm and equilibrium. In so doing, some things are emphasized and others are managed and controlled where possible. We saw this recently in 2011 with the catastrophic floods and cyclone in Queensland where political leaders, in their response, framed up accounts that focused on the strength of the affected populace to rally in the face of adversity, as well as assuring funding to support those affected, promising to rebuild homes, towns and infrastructure. Such positive rhetoric was the dominant narrative of government leaders who were successful in downplaying the very real “doom and gloom” outcomes. In so doing, they effectively ensured that those who govern were seen to be strong and in control, thus averting public outcry and potential chaos.

In conclusion, I am personally disappointed by Hill's response in taking my paper as a criticism of the valuable work done by forensic teams in disasters across the world. That was never my intention. Forensic recovery is skilled, difficult and essential work. It is therefore of concern to me that, in his reading of my paper, the fundamental intent was completely lost, together with the opportunity to engage in interdisciplinary dialogue in a manner that might advance a broader understanding of the interaction of science, culture and social processes. I appreciate that some of the problems discussed here are essentially a matter of epistemology or competing understandings of the source of knowledge and how that translates into human action. It is regrettable that these differences mean that we are unable to read and understand different disciplinary stances in the interests of improving practices in relation to human disasters.