Chapter 35 — Key Takeaways

A scannable one-page card. For the full argument and the worked figures, see index.md.

The core claims

• Same discipline, new scale. DVI is not new methods — it is the same evidentiary discipline (fingerprints, dental, DNA) applied when the dead overwhelm the routine, one-body-at-a-time system. What changes is scale, and scale changes everything about how the work is organized and checked.
• A mass fatality is defined by the system being overwhelmed, not by a body count. Ten deaths can be a mass fatality in a small county and a routine week in a large one.
• Three conditions make the DVI problem hard: scale (forces standardization — the system is the method, not heroic individual attention); commingling (remains of several people mixed; "one set of remains = one person" fails, so fragments must be re-associated); and condition (damage destroys whichever identifier it touches, so no single identifier works for every victim).
• The Interpol DVI phases impose order: (1) scene → (2) postmortem → (3) antemortem → (4) reconciliation → (5) debriefing. The framework's deepest discipline is the wall between PM and AM: they are built independently and not compared until Phase 4, so the dead are described on their own terms before anyone fits them to an expected identity — pure bias discipline (Chapter 31).
• Three primary identifiers — fingerprints, dental, DNA — each can identify on its own because each can individualize and their failure modes differ; DVI runs all three in parallel. Secondary identifiers (clothing, jewelry, tattoos, build) guide and corroborate but must never confirm alone.
• DNA does two things in a disaster that nothing else can: identify tiny, degraded, burned fragments, and re-associate commingled remains. It is often indirect — kinship matching against relatives (weaker and more interpretive than a direct match).
• Reconciliation judges every match against the candidate pool. A single dental or kinship match is weaker in a pool of hundreds (a quiet prosecutor's fallacy, Chapter 9). The safeguard is convergence across independent identifiers, confirmed by an Identification Board. Exclusions are the cleaner result.
• The same discipline serves human rights. Mass-grave forensics pursues a humanitarian goal (the missing) and a judicial one (atrocity prosecution) — which can be in tension. Trauma can show how a victim died; it cannot, from the bone alone, name who ordered it (Chapter 12, unchanged).
• DVI is, above all, a humanitarian act. Its conservatism (leave a body unidentified rather than guess) is a duty to families; a misidentification is a double catastrophe. The science can give a family the most important fact — whether their person is dead and which remains are theirs — but cannot always succeed, and must be honest about that.

The method-validity verdict (under disaster conditions)

Where they sit: all three primaries can reach the individual end of the class→individual arrow (Chapter 1); secondary identifiers are mostly class characteristics that narrow but don't confirm. Confirmation is reserved for the primaries, ideally by convergence.

What you can honestly say on the stand / to a family

• An established identification: "This individual is identified as [name] on the basis of [a direct DNA match with a random match probability of one in X / a dental identification with concordant features and no unexplainable discrepancies / a fingerprint identification], corroborated by [secondary identifier]. The identification was confirmed by the identification board."
• A kinship identification: "The DNA from these remains is consistent with the proposed relationship to [relatives] with a likelihood ratio of X; this strongly supports that these are the remains of [name]."
• An open case: "We have not yet been able to identify these remains to the standard we require, and we will continue testing. I would rather tell you we are not sure than risk returning the wrong person."
• What you must NOT say: confirmation of identity from personal effects or visual recognition alone; a kinship match stated as flat certainty without its statistic; or, in a mass grave, that trauma "proves" who ordered a killing.

The cold-case line (a teaching aside, not new evidence)

The single-victim identification of Marcus Diallo (Chapter 17) is the mass-fatality process collapsed to its easiest case: no commingling, an antemortem record that already existed and pointed at one named person, and a candidate pool of effectively one (so the pool-size/prosecutor's-fallacy worry never arose). This chapter adds no new facts about his death or any suspect — (teaching aside; no new suspect data). The identification stands exactly where Chapter 17 left it.

Key terms (one line each)

• Disaster victim identification (DVI) — the organized, standardized process for identifying many dead from a mass-fatality event and returning them to families.
• Mass fatality — an event producing more dead than the local system can handle with routine resources (defined by being overwhelmed, not by a number).
• Commingled remains — remains of two or more individuals mixed together, which must be re-associated to individuals before/alongside identification.
• Interpol DVI phases — the five-phase framework (scene, PM, AM, reconciliation, debriefing) that keeps PM and AM data separate until reconciliation.
• Antemortem/postmortem reconciliation — the case-by-case comparison of the PM record (from bodies) against the AM record (on the missing) to establish who is who.

The themes this chapter advanced

• Exclusion over proof — exclusions empty the reconciliation matrix faster and more confidently than matches fill it; the trauma reading in a mass grave establishes how, not who.
• The validity spectrum — the three primaries individualize (and differ in failure mode); secondary identifiers narrow but cannot confirm; kinship DNA is weaker than direct DNA — all the same yardstick.
• (Also advanced: cognitive bias — the PM/AM wall is bias discipline made into international procedure, §35.2; and the CSI effect / honesty — real DVI is slow, conservative, and sometimes fails, the opposite of the televised instant identification.)