"Yes. And it is your knowledge, training, experience and the body of accepted knowledge in your field that provides you the ability to "rule in" (or postulate) plausible hypotheses based on the data. However, there are times when hypotheses are formed that are outside the accepted practice's knowledge base - that is when new or novel discoveries are made. The phlogiston "theory" had plenty of support and evidence for years until someone formed a hypothesis counter to the prevailing approach - a hypothesis based on data which became "evidence" when evaluated against the hypothesis."

There can be a discussion of the general methodology of the SM. For example, identification of the sequential steps of one iteration of the SM. Then, there is the methodology of applying those steps to actual data to arrive at a reliable determination. This is more of a discussion of the general methodology. My interest is in the formal application of the SM to fire investigation. How does one implement the SM in a reliable manner. Can I produce a written detailed explanation that would allow a new investigator to reach reliable determinations in fire investigation using the SM?

"Yes. The pathologists narrowed down the pool of possible solution hypotheses by eliminating certain causes (based on data) from the large pool of possible causes of death. The lack of data suggesting a gunshot wound becomes evidence for the hypothesis "This patient died of a gunshot wound." The evidence refutes the hypothesis and therefore, the hypothesis having failed, data is reviewed against other hypotheses. (Postulating the "gunshot" hypothesis does not make it "invalid" because there is no evidence on which to formulate the hypothesis. It still has value in the overall process.) The pathologist (or microbiologist or whoever is considering the problem") Can, based on their knowledge, training, experience, research, etc. form hypothesis...If it wasn't a gunshot, maybe it was viral, bacterial, parasite, toxin....then compare data to those hypotheses. That's where the gene sequencing comes in. Eliminate the human, then look for something anomalous...NO known virus sequences - eliminate virus...etc."

"Eliminating or no longer considering a hypothesis based on a lack of evidence. Not the Process of Elimination.

You lost me at the second sentence ... "The lack of data suggesting a gunshot wound becomes evidence for the hypothesis "This patient died of a gunshot wound." The evidence refutes the hypothesis ..."
A lack of data becomes evidence? A lack of data becomes evidence for the hypothesis "This patient died of a gunshot would?" Sorry, but I am not seeing how this is a reliable application of the SM.

"... forms and eliminates hypotheses so quickly, they may not even consciously realize it."

Some might consider some steps of the SM as trivial because of this (e.g., first two steps) I do not disagree. My goal is to better describe how to explicitly apply the SM to fire investigation to reach a reliable determination. It is my opinion that NFPA 921 does not do this very well. I made the argument and revised the origin chapter in earlier versions of 921 because if you gave the chapter to someone who is new to this process, I do not believe they would be able to reliably determine the area of origin. I believe this is true of the methodology chapter in 921.

Douglas J. Carpenter, MScFPE, CFEI, PE, FSFPE
Vice President & Principal Engineer
Combustion Science & Engineering, Inc.
8940 Old Annapolis Road, Suite L
Columbia, MD 21045
(410) 884-3266
(410) 884-3267 (fax)
www.csefire.com