by Tony Cafe

Reproduced from "Firepoint" magazine - Journal of Australian Fire Investigators - September 1998.

Some time ago I received an email from a 34 year old American law student called Terri Strickland who had been charged with the murder of her 4 year old son after a fire at her home. She was on bail awaiting trial and she asked me if I could help her because she had no money to hire a fire expert for her defence. She believed that the fire was accidental and that the fire investigator was wrong in his claim that it was a deliberately lit fire. I offered to help her because I know from experience that there is a greater probability of a fire investigator getting it wrong than there is for a woman to murder her child by fire.

One of the main reasons why fire investigators get it wrong is because some of the information and education they have received over the years is not based on tested scientific methodology. The investigator when confronted with the extensive literature on fire investigation must therefore determine which of this information is unreliable which is often a difficult task.

Consider the claims made in the literature 10-20 years ago that collapsed bed springs, concrete spalling and crazed glass indicate a hot fire and therefore an accelerant. In recent years there have been many articles published which debunk these theories. Think about all the fires during the intervening years where investigators relied on these indicators and said there was an accelerant present without positive laboratory proof and imagine the damage they have done to people's lives. What was their testimony in court - "Yes your honour, I read it in the fire investigation literature and so it must be true."

Consider the theory made many years ago that the level of copper oxide in an arc bead could be used to determine whether the arc caused the fire or was a consequence of the fire. Recent literature suggests that making conclusions from such tests is simplistic because the level of copper oxide in an arc bead can rise or fall during a fire. I often wonder how qualified scientists who conducted these tests for so many years could do so without asking themselves the fundamental question - how stable is copper oxide during a fire?

Even to this day I often receive reports by other fire investigators who state that a floor burn through indicates an accelerant even though they have not even taken a sample for laboratory analysis. In most of these fires the roof had collapsed and continued to combust at floor level. Did any of these investigators consider that floors will eventually burn through during a fire with or without the use of an accelerant?

Why is it that the information that fire investigators rely upon is often later proven to be wrong? It is probably because the people who publish the original information are poor scientists or they are not scientists at all and the work is not scientifically scrutinised. Scientific research is based on a methodology which firstly makes a full review of the scientific literature, then the research work is conducted in carefully controlled tests and then the results are scientifically scrutinised before publication. The results are then published in refereed journals or presented at conferences where peers can question the work or make comment.

There is little information produced by scientists in local magazines such as Firepoint which leaves it up to the other fire investigators to contribute. One possible reason for this is because scientists aren't encouraged to contribute. There are too few scientists asked to speak at our conferences which makes the conferences very unappealing for other scientists to attend. In the case of electrical engineers who are frequently called from the universities to investigate fires, not one has ever been invited to speak at any of our recent conferences which is a pity because this is one area of fire investigation where we really need expertise. Of course scientists should not wait to be asked to contribute, they should want to contribute to journals and conferences. They do however seem to be reluctant, possibly because as one scientist warned me before writing an article some time ago - "It would be giving up the secrets." We all know that he was implying that if all people had this information then the work for qualified experts could dry up. Never underestimate self interest in discussions such as these.

Fire investigation is a branch of forensic science and so the scientific methodology outlined above should apply to fire investigation literature, but it isn't. Using a strict scientific methodology for fire investigation literature would mean that only scientists with tertiary qualifications are allowed to publish research and opinions and the implications of this is that only scientists are allowed to investigate fires. This topic has always been a very prickly one here in Australia and I remember a conference many years ago where a scientist was literally howled down after asking a prominent barrister what formal qualifications should a fire investigator have. Recently in the USA a huge row has erupted on the internet bulletin boards concerning the Daubert issue where the qualifications of fire investigators has been questioned. In the end, the judgement of who should investigate fires is up to the government and the courts and also our own organisation because of its role in collectively representing fire investigators in Australia. There are many sides to the story and this magazine is an excellent way of putting forward your opinions.

The second reason why fire investigators get it wrong is that they believe that fire investigation is an art. Art is about creativity and fire investigators are not at fire scenes to ponder creativity but are there to physically find and interpret the evidence which will indicate the cause of the fire. Fire investigators who believe in art are often found at the fire scene, or worse in court, talking about the fire as if they were actually present during the fire. They are using their imagination rather than the part of their brain which controls logic and reasoning.

So why do some fire investigators and more importantly their educators believe that there is an art to fire investigation? Its probably because they have no scientific qualifications and they need to justify to themselves and to others of their presence at a fire scene or in their role as an educator. They see themselves as having a sixth sense which makes them an extraordinary human being. The most vital sense of all is common sense and the investigator needs to couple this with the ability to observe, reason and deduce.

Having said what I have so far it would seem that scientists, with their fantastic abilities, make great fire investigators. However this is not always the case. I frequently come across fire reports written by qualified scientists who believe that floor burn throughs solely indicate accelerants. Another problem with scientists is that they never seem to agree on anything. The reasons why scientists never seem to agree is that they rely too much on opinion evidence, and there are the hired guns who show a lack of ethical consideration.

Opinion evidence will always be a part of fire investigation especially when opposing barristers produce their experts in court. Ways of avoiding expert's dissenting on opinion evidence is to provide as much irrefutable evidence as possible. For example, laboratory proof of an accelerant is always a good indicator of a deliberately lit fire. Also, lots of good photographs in a report should be used in a systematic way to prove every point the investigator wishes to make. The investigator should avoid making detailed claims of how the fire progressed unless they actually witnessed the fire or spoke to eyewitnesses or they are experienced fire brigade investigators who have witnessed many similar fires in progress and so feel confident in giving an opinion.

Ethics are constantly talked about in the fire investigation literature and should be uppermost in our minds when preparing our reports so that they are unbiased and independent. The IAAI has its own ethical code which does not seem to bear any resemblance to those used by scientists. For example the line "I will regard my fellow investigators with the same standards as I hold for myself" has nothing to do with science and resembles something used by the RSL or as one prominent USA investigator puts it, "the IAAI is becoming a church." Compare the IAAI code of ethics (Firepoint, December 1995) with those of the Californian Association of Criminalistics (Firepoint, March 1997). The reason why the IAAI code does not have anything to do with science is because it should not be the expert's character which is under scrutiny but it should be the expert's evidence and their reasoning.

Finally I will get back to Terri Strickland who wrote to me recently after spending 2 months in jail and 18 months on bail for a crime she was found totally innocent of. She is going to finish her law studies and then open a legal reference centre to help people and prevent them from going through the same awful mess she went through. Her hopes are "That maybe one day the "experts" will stop doing what they are doing and ruining peoples lives." I believe we can help her by being more scientific in our approach and by leaving the art to the artists.

To read more on the Terri Strickland case, see