"All data is evidence. The question becomes is that evidence relevant to the question you are attempting to answer. One uses their experience, education, and training to determine the relevance. The process of elimination is what is used to determine what is relevant. I agree that evidence is needed to formulate a hypothesis. What I have found is that many do not explain in their reports what hypotheses were tested. The reports fail to show where the investigator did seek alternative hypothesis. In most cases, there are more than one hypothesis to be considered."

All data is not evidence. All evidence is based on data, but not all data is evidence. If all data is evidence, then the "for sale" sign would be relevant and reliable in the context of an arson fire.

Evidence is data that is assessed to be relevant and reliable in a specific context.

There is a significant different between "considering a hypothesis" and "formulating a hypothesis." Different steps of the SM.

"I disagree that the POE is only a reliable methodology under the SM for certain conditions. Under all conditions, the process of elimination is used from the very beginning when the analysis of the data is being conducted. Maybe you can explain in what condition it cannot be used. Every scientist using the scientific method uses the process of elimination. Why would one think that fire investigation is so special that it differs from other scientific endeavors?"

It can not be used for an open system where the known outcomes are not finite and not known before the investigation. It is reliable for a multiple choice exam and for a finite number of uniquely colored marbles in a closed box.

"If one does not have all the available evidence should that person be rendering any opinion. All I was saying is that the process of elimination should not have been removed from the document."

What if that investigators does not know that they do not have all the available evidence?

"I was not suggesting there needs to be a consensus. What I was saying is that when a fire investigator reaches an opinion in most cases it is that person’s opinion only. However, in most cases, the cause being undetermined is presented as if there was insufficient evidence for anyone to reach a different conclusion."

Opinion based on the application of the SM. Are you saying that a person's opinion is subject?

"When it comes to the use of the process of elimination the fire investigation profession is no different than any other profession. We, as a profession, need to get out of the idea fire investigators are something special when compared to other professions. I believe the fire investigation profession does use the scientific method. The problem is, just as it is here, is that we cannot agree within our own profession as to the proper application of the scientific method. I will never say a person is ignorant because that person does not agree with my possession. I expect that went multiple persons look and evaluate the same evidence there will be a difference of opinion as to the meaning of that evidence. That does not mean one person is ignorant as compared to the others."

The difference is that your argument is based on the premise that the process of elimination is a reliable methodology for fire investigation. Remove that premise and the FI community is not different. The FI community is different if they are ignorant of how the reliably apply the SM to a specific set of data.

You can disagree on determinations, but the basis for that disagreement needs to be transparent and not subjective. I had a case where I testified and the difference in determinations was based on a difference of opinions as to the condition of an opening. The opposing expert determined that it was closed and I determined it was open. I testified that if the opening was covered, then the opposing expert's determination was the most reliable. If the opening was not covered , then my determination was the most reliable. My determination was based on available data that supported my position. His determination did not use this data. Reasonable men can disagree, but the methodology and application must be transparent.

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