Certainty Expressions and Expert Opinion in Fire Investigation

One of the most important parts of fire investigation isn’t just the science, its how we explain our conclusions. Courts listen closely to the language we use, and NFPA 921 makes it very clear that not all expressions of certainty are created equal. If we’re not careful, we risk turning solid work into testimony that sounds speculative or unreliable.

The guide tells us that expert opinion begins when a hypothesis is probable that is, more likely true than not. Once the evidence supports a conclusion past the 50% mark, we can say it was “probably caused by” a certain factor, and that qualifies as expert testimony. This is what NFPA 921 calls “reasonable fire investigation certainty,” and it is the level courts expect from investigators.

There are also times when the evidence doesn’t let us go that far. Sometimes two or more scenarios remain equally likely, or the data only shows that something could have happened but doesn’t prove it one way or the other. In those cases, we must call it possible. That still counts as expert opinion, but it shows the court that we understand the limits of the available evidence. Being transparent about those limits actually strengthens our credibility, because it demonstrates that we tested the data, considered alternatives, and didn’t oversell the findings.

Where investigators get into trouble is with the word suspected. NFPA 921 is blunt on this point: if a cause is only suspected, it is not an expert opinion. Suspected is speculation. It might help you when you’re thinking through scenarios, but it has no place in a report or in testimony.

There’s also the old habit of saying a conclusion is reached “to a reasonable degree of scientific certainty.” It sounds authoritative, but NFPA 921, along with the Department of Justice and the National Commission on Forensic Science, now warns against using it. The problem is that it has no agreed definition, and juries can interpret it as stronger than it really is. If a court insists on hearing that phrase, the best approach is to ask the judge to define it before using it. Otherwise, avoid it altogether.

In the end, the message is simple: probable and possible are the language of expert opinion. Suspected is not. Our job is to give the court clarity, not confusion, and the words we choose carry as much weight as the evidence itself. When we stick to NFPA 921’s certainty expressions, we keep our work defensible, our testimony credible, and our role as experts intact.