Fire Origin & Cause Investigations: The Expert and the Document Record

By John Mahoney · June 2026 · 9 min read

A fire loss is almost entirely reconstructed after the fact, from a debris field, a stack of reports, and the memory of people who were there. By the time a subrogation, product-liability, or defense lawyer opens the file, the building is gone and the central factual questions — where did this fire start, and what made it start — can only be answered through an expert investigation of what remains. The fire origin-and-cause (O&C) expert is the witness who builds that answer, and the entire case stands or falls on whether the investigation was done to a defensible standard and documented well enough to survive scrutiny. This guide walks through what the O&C investigator does, the NFPA 921 framework that governs the work, the document record the opinion is built on, and the spoliation and reliability attacks that decide whether the opinion holds — for litigators on either side of a fire-loss file.

Disclaimer: This article is for informational purposes only and does not constitute legal advice. Fire investigation standards, the admissibility framework, and the controlling case law vary by jurisdiction and by the facts of each loss. Always verify the applicable standard and review the actual investigative record before relying on any generalization here.

What the Origin-and-Cause Investigator Does

The O&C investigator's job is to answer two distinct questions in sequence. First, the origin: the specific location where the fire began. Second, the cause: the ignition sequence — the heat source, the first fuel ignited, and the circumstance that brought them together. These are separate findings, and a weakness in the first undermines the second. An investigator cannot reliably opine on cause without first having pinned down origin, because the cause analysis is an examination of what was present at the area of origin.

The work is part forensic science and part disciplined reasoning. Investigators read burn patterns, char and calcination depth, heat and smoke movement, and the condition of materials to reason backward from the most-damaged areas toward the point where the fire started — while accounting for the fact that the area of origin is not always the most heavily burned, because ventilation and fuel load can shift the damage. Once origin is established, the investigator examines the potential ignition sources in that area — electrical wiring and devices, appliances, heating equipment, smoking materials, open flame, or an intentional act — and works to confirm one while excluding the others.

The NFPA 921 Scientific-Method Framework

NFPA 921, Guide for Fire and Explosion Investigations, is the dominant reference for how a fire investigation should be conducted, and it frames the work as an application of the scientific method rather than intuition or pattern-matching from experience alone. Courts and opposing experts treat departures from it as a serious problem. The framework runs in a recognizable sequence:

• Recognize the need and define the problem — establish that an investigation is warranted and what it must determine.
• Collect data — the scene examination, photographs, witness accounts, physical evidence, and records, gathered before forming conclusions.
• Analyze the data — interpret the collected facts based on the investigator's knowledge, training, and experience.
• Develop hypotheses — formulate explanations for the origin and cause that fit the data.
• Test each hypothesis — subject each explanation to deductive reasoning and, where possible, empirical testing; a hypothesis that cannot withstand testing must be discarded.
• Select a final hypothesis — reach a conclusion only if one explanation survives testing; otherwise the cause should be classified as undetermined.

Two features of this framework drive the courtroom fights. First, the willingness to return a verdict of undetermined is a feature, not a failure — an investigator who forces a conclusion the data does not support has departed from the method. Second, the framework requires the investigator to guard against expectation bias and to avoid reasoning that fits the evidence to a predetermined answer. An opinion that skipped hypothesis testing, or that never seriously considered the alternatives, is exposed precisely because NFPA 921 demanded that it do so.

The Document and Evidence Record

The opinion is only as good as the record beneath it, and that record is unusually layered in a fire case. A thorough review traces each conclusion back to the documents that support it:

• The fire-department incident report. The responding agency's account of conditions on arrival, where flames and smoke were showing, suppression actions, and any initial cause determination. This is often the earliest contemporaneous record and a frequent source of friction when a later private investigator reaches a different conclusion.
• The investigator's origin-and-cause report. The core document — the narrative of the scene examination, the basis for the origin finding, the ignition-source analysis, the hypotheses considered and excluded, and the final classification.
• Scene photographs and diagrams. The visual record of burn patterns, the area of origin, the condition of suspected ignition sources, and the post-fire layout. The photographs are how an opposing expert re-examines a scene that no longer exists.
• Witness and occupant statements. Accounts of what was seen, heard, and smelled before and during the fire — first observations, alarm activation, recent electrical or appliance problems, and timeline anchors.
• Electrical and appliance evidence. Recovered wiring, breakers, receptacles, and the suspect appliance or component — the physical artifacts at the center of most product-defect theories.
• Laboratory and failure-analysis reports. Bench examination of recovered components, accelerant-detection testing of fire-debris samples, and the engineering failure analysis that ties (or fails to tie) a defect to the ignition.
• Weather, utility, and monitoring data. Lightning and wind records, utility load and outage logs, and alarm or monitoring panel histories that corroborate or contradict the proposed sequence and timeline.

These sources rarely all point the same direction, and the discrepancies are where the case lives. A first-arriving firefighter who reported flames at one corner, an O&C report that locates origin at another, and a lab report that cannot confirm the proposed ignition source together describe a contested case — and the lawyer who has read all three against each other is the one who can frame it.

The Origin → Cause → Responsibility Chain

Fire litigation is a chain, and every link has to hold. Origin establishes where to look; cause establishes the ignition mechanism at that location; and only then does the case reach responsibility — the legal theory connecting the cause to a defendant. A subrogation or product case typically needs the cause to resolve to a specific component or act: a defective appliance, faulty wiring, improperly installed equipment, or negligent conduct. A defendant breaks the chain by attacking any single link: the origin is wrong, so the cause analysis examined the wrong area; the cause is unproven, so the ignition source is speculation; or even granting cause, the proposed defect or breach is not what ignited the first fuel.

The link most often overlooked is the jump from cause to responsibility. An investigator may credibly establish that a fire started at an electrical receptacle without being able to say whether the receptacle was defective, improperly installed, or simply overloaded by the occupant — three findings that point at three different defendants, or at no one. The O&C opinion frequently establishes less than the legal theory needs, and the gap is where cases are won and lost.

Spoliation and Chain-of-Custody Pitfalls

Fire cases are uniquely vulnerable to spoliation because the investigation itself is destructive and the evidence is fragile. Examining a scene means moving debris; testing a component can consume it; and a burned structure is frequently demolished, cleaned, or rebuilt before every interested party has had a chance to inspect it. The recurring failure is a party — often the insurer's investigator — who examines, photographs, and removes the suspect component, then disposes of the remainder of the scene before notifying the manufacturer or other potential defendants.

That sequence sets up a spoliation motion that can be case-dispositive. A defendant deprived of the chance to inspect the origin area or the suspect artifact will argue it cannot test the plaintiff's hypothesis — and courts have a range of sanctions available, up to exclusion of the evidence or an adverse inference. The defensive practice is well known: preserve the scene, give notice and an inspection opportunity to all known potential parties before destructive testing, and document an unbroken chain of custody for every piece of physical evidence from collection through laboratory examination. A gap in custody — an artifact whose handling between the scene and the lab cannot be accounted for — is its own line of attack on the reliability of whatever the lab found.

How the Opinion Is Attacked

Under NFPA 921 and the Daubert reliability framework, the cross-examination of an O&C expert follows a predictable set of lines, and screening for them early tells you how durable the opinion is:

• The negative corpus. Concluding a cause — classically that a fire was incendiary — by excluding all accidental causes rather than affirmatively proving the cause. NFPA 921 treats reasoning purely from the absence of other causes with caution, and an opinion that rests on a negative corpus without independent positive support is exposed.
• The untested hypothesis. A conclusion the investigator developed but never subjected to the hypothesis-testing step — the cause that fits the investigator's first impression but was never stress-tested against the alternatives the data also permits.
• Expectation bias. An investigation that started from a presumed answer — arson, or a particular manufacturer's product — and collected and read the evidence to confirm it.
• Origin error propagating into cause. If the area of origin is wrong, every downstream finding is examining the wrong location, and the cause opinion collapses with it.
• Departure from the standard. Skipped steps, undocumented reasoning, an inspection that destroyed evidence before others could examine it, or a leap to a definite cause where the data supported only undetermined.

The same record is screened by both sides for the same signals; the factors separating a strong opinion from a vulnerable one are largely symmetrical.

What strengthens the proponent's position

• An origin finding supported by converging evidence — burn patterns, witness first-observations, and damage progression all pointing to the same area.
• An affirmatively proven ignition source, confirmed by physical or laboratory evidence, rather than a cause reached only by exclusion.
• A documented hypothesis-testing step that names the alternatives considered and explains, with reference to the data, why each was excluded.
• A preserved scene and an unbroken chain of custody, with notice and inspection given to other potential parties before any destructive testing.

What strengthens the challenger

• An origin or cause that conflicts with the fire-department report or with credible first-observation testimony, unexplained by the investigator.
• A cause reached on a negative corpus, or a hypothesis the investigator never tested against the data's alternatives.
• Evidence consumed or a scene destroyed before the challenger had an inspection opportunity — the spoliation and chain-of-custody attack.
• An opinion that overstates what the science supports — a definite cause where the honest classification was undetermined.

Why a Verifiable, Page-Cited Review Matters

The decisive facts in a fire case are buried in a multi-source paper record — the incident report, a long O&C narrative, lab and failure-analysis reports, hundreds of scene photographs, witness statements, and the investigator's deposition — and the case turns on the discrepancies between them. The work of finding those discrepancies is exactly the work of reading every document against every other, and an AI assistant that summarizes the record only helps if you can trust and verify what it surfaces. That is why every finding from MedLegal AI cites the exact source page it came from: when the tool flags that the firefighter placed origin in a different corner than the O&C report, or that the lab could not confirm the proposed ignition source, you can click straight to the page and read it in context before you ever rely on it. In a field defined by the gap between what the data shows and what the opinion claims, a review you cannot trace back to the source is worse than no review at all. The point is not to replace the fire investigator or the failure-analysis engineer — it is to put a complete, page-cited map of their record in front of you before the deposition.

Bottom Line

A fire-loss case is reconstructed almost entirely from documents and physical artifacts, and the O&C expert is the witness who turns a debris field into an origin, a cause, and a path to responsibility. NFPA 921 supplies the scientific-method framework the opinion must follow — collect, analyze, hypothesize, test, and be willing to return undetermined — and the recurring attacks track its requirements: the negative corpus, the untested hypothesis, expectation bias, and origin error that propagates into cause. Layered on top are the field's signature vulnerabilities: a destructive investigation that invites spoliation and a fragile chain of custody.

For both sides the work is the same: pin down whether origin is genuinely supported, whether the cause was affirmatively proven or reached only by exclusion, whether the hypothesis-testing step was actually done, and whether the scene and evidence were preserved with notice to everyone who needed to inspect them — then test whether the cause finding actually reaches the defendant the legal theory requires. The merits should decide the case, so verify every generalization against the actual investigative record.

Questions? Contact us at [email protected] or (856) 979-6525