Chapter 17: Forensic Odontology and Human Identification: Teeth, Dental Records, and Identifying the Dead

"The teeth are the hardest and most enduring tissues of the body; they outlast the flesh, and often the bone, and so they outlast the questions we most need them to answer." — constructed line, in the practitioner voice of this book, summarizing the working premise of dental identification.

Overview

A body comes out of a house fire so badly burned that no one can recognize the face, no fingertips survive to print, and the soft tissue that would carry a routine identification is simply gone. The family is certain who it is. The insurance company is not — there is a large policy, and "we cannot be sure that is really him" is worth a great deal of money to the right person. Someone has to answer the oldest forensic question of all, the one that precedes every other: who is this? Not who killed him — that comes later, and depends entirely on this. Just: whose body is on the table.

For most of human history that question, once the face was gone, had no scientific answer. What changed it was a quiet, durable fact of anatomy: teeth are the most indestructible tissue the body owns. Enamel is the hardest substance the human body makes; it resists heat, decay, water, and time long after skin, muscle, and even much of the skeleton have failed. And teeth are individually informative — a lifetime of cavities filled, crowns placed, molars extracted, roots canaled, and bite quirks written into a mouth produces a configuration that, recorded in a dentist's file, can be matched to remains. That is the real, validated work of forensic odontology: not the bite-mark "matching" that this book has already named as junk science (Chapter 16), but the sober identification of the dead from their dental records — the discipline that tells a grieving family, or a skeptical court, yes, it is him, on grounds that hold up.

This chapter is about that distinction as much as about teeth. Forensic odontology is the field that sits, more starkly than any other, on both ends of the validity spectrum at once: rigorous and defensible when it identifies a body, discredited and dangerous when it claims one person's teeth made one mark in skin. We will spend most of our time on the valid half — how antemortem and postmortem records are made and compared, how a charting comparison reaches its verdict, how the method behaves in a single fire death and in a disaster that kills hundreds — and we will return, at the end, to why the same specialty produced both a humanitarian science and a wrongful-conviction machine.

In this chapter, you will learn to:

• Explain why teeth survive insults that destroy every other identifying tissue, and why that durability is the source of the method's value.
• Distinguish antemortem from postmortem dental records and describe how each is collected and what defeats it.
• Read a dental chart, name the notation systems, and follow a comparison to its honest conclusion.
• Place dental identification on the validity spectrum and describe how it works in fire and mass-disaster settings.
• Separate the discipline's valid identification work from its discredited bite-mark work, and state in court-ready language what an odontologist may and may not claim.

Learning Paths

🔎 Investigator/CSI: Your job at a scene with unrecognizable remains is to preserve the jaws and recognize that dental ID is on the table — §17.1 and §17.4. Mishandling charred remains (the jaw is fragile after fire) can destroy the one identification route left. 🧪 Lab analyst: You may never do odontology yourself, but you hand off to and receive from it. Weight §17.2 and §17.3 — how the postmortem record is generated and compared — and note where the comparison's subjectivity lives. ⚖️ Law/courtroom: The whole point for you is §17.5: the same expert specialty produces testimony at opposite ends of the validity spectrum. Know which kind you are hearing, and the verbs each is entitled to. 👥 General reader/juror: This is the chapter that explains how the burned, the decomposed, and the unrecognized are given their names back — and why "the dentist said it's a match" means something very different for a body than for a bite.

17.1 Why teeth: durability and the individual dental history

Begin with the question the evidence answers, because it reorders everything that follows. A medicolegal death investigation cannot proceed until the decedent is identified — you cannot determine cause and manner of one person's death (Chapter 11) and notify a family, settle an estate, file a life-insurance claim, or charge a suspect with his murder until you know whose body it is. When a face is intact and a wallet is in the pocket, identification is trivial and no one thinks about it. Forensic odontology exists for the cases where it is not trivial: the burned, the decomposed, the skeletonized, the dismembered, the long-submerged — the bodies that have lost every easy answer.

Of the methods that survive into those hard cases, teeth are the most reliable, and the reason is physical. Enamel, the outer covering of the crown, is roughly 96% mineral — the hardest tissue in the human body, harder than bone. Dentin beneath it and cementum on the root are softer but still highly mineralized. Because teeth are mostly mineral and are partly shielded inside the bony sockets of the jaws, they endure conditions that destroy everything used for routine identification:

• Fire. Soft tissue chars and is consumed; fingerprints are lost early; even bone calcines and fragments. Teeth survive remarkably high temperatures, especially when protected by the lips, tongue, and jaw. Dental restorations — fillings, crowns, gold work — survive too, and many have melting points well above the temperatures of an ordinary structure fire.
• Decomposition. The bacterial and enzymatic processes that liquefy soft tissue (Chapter 11) do not touch enamel. Long after a face is unreadable, the dentition is intact.
• Water and time. Submerged and skeletonized remains lose almost every soft-tissue identifier; the teeth remain.

🔬 At the Bench Why "the most enduring tissue" is not the same as "indestructible." After intense or prolonged fire, teeth become extremely friable — they survive, but they turn brittle and can shatter at a touch. A jaw recovered from a fire scene may hold a complete, identifiable dentition that crumbles if it is handled like ordinary bone. This is why recovery of burned remains is done slowly, often with the jaws stabilized in place, sometimes encased or photographed extensively before anything is moved. The durability that makes dental identification possible is conditional on careful recovery; an impatient excavation can destroy the only identification the body had left. The lesson runs straight back to Chapter 2: the case is won or lost at the scene.

Durability alone would only make teeth a surviving tissue, not an identifying one. What makes them identifying is the second fact: a human dentition accumulates an individual dental history. Consider what is written into one adult mouth over a lifetime: which of the (typically) 32 teeth are present, missing, or impacted; which have been filled, and with what material, and on which surfaces; which carry crowns, bridges, implants, or root-canal therapy; the rotations, spacings, and crowding of the teeth; orthodontic history; the shape of restorations a particular dentist happened to cut. No two mouths share an identical complete history, and — crucially for forensics — much of that history is documented in dental records, X-rays, and charts as it is made.

This is what separates dental identification from the pattern-comparison disciplines that fill the rest of Part III. A fingerprint examiner compares a latent to a known print taken from the suspect. A firearms examiner compares marks to a test-fired bullet. The odontologist compares the body's teeth to the records of that living person's dental care — records created over years, by clinicians with no stake in any future case, for the ordinary purpose of treating a patient. The comparison is between a body and a paper trail, and the strength of any identification depends entirely on the quality and existence of that trail (§17.2).

🔍 Check Your Understanding 1. Name two reasons teeth are the identifying tissue of choice for a badly burned body, when fingerprints and face are gone. 2. Why does "teeth are the most durable tissue" still require careful, slow recovery at a fire scene? (What happens to a tooth after intense heat?)

It is worth naming the limit at the outset, because the rest of the chapter depends on it. Dental identification is a comparison method, and it can only be as good as the antemortem record it compares against. A body with a mouth full of distinctive dental work and a complete set of recent dental X-rays on file can be identified to a near-certainty. A body with perfect, untreated teeth and no dentist of record may be unidentifiable by odontology altogether — there is nothing documented to compare. The method's power is not a property of teeth alone; it is a property of teeth plus a record, and the record is the part that fails.

17.2 Antemortem and postmortem dental records

Every dental identification is, at bottom, a comparison of two data sets: what the person's mouth looked like before death, as documented in life, and what the body's mouth looks like now, as documented at the morgue. The forensic terms for the two are exact, and they are this chapter's load-bearing vocabulary.

Antemortem records ("ante-mortem," before death) are the dental records generated during the person's life — by their dentist, orthodontist, oral surgeon, or military or institutional dental service. They are the known in the comparison, the dental equivalent of the reference sample from Chapter 3. A complete antemortem file may contain:

• Radiographs (X-rays) — bitewings, periapicals, panoramic films, or modern digital equivalents and cone-beam CT. These are the single most valuable antemortem material, because they show roots, bone, restorations, and unerupted teeth that no chart fully captures, and because they can be directly overlaid against postmortem films (§17.3).
• Dental charts — the clinician's notation of which teeth are present, missing, filled, crowned, or otherwise treated, surface by surface (§17.3).
• Treatment notes and billing records — narrative descriptions of procedures and their dates, which can reconstruct a dental history even when films are missing.
• Study models, photographs, and impressions — physical or digital records of tooth position and shape, increasingly common with orthodontic and implant work.

Postmortem records ("post-mortem," after death) are the records the forensic odontologist creates from the body — the questioned side of the comparison. The odontologist examines the dentition (sometimes requiring resection of the jaws when access or rigor demands it), charts every tooth and restoration present, and takes postmortem radiographs using the same projections as common antemortem films, precisely so the two can be compared like to like.

🔬 At the Bench The single most important procedural rule in dental identification is one of sequence and independence, and it is a direct application of this book's bias theme. Wherever practical, the postmortem examination should be charted before the antemortem records are studied in detail, or by an examiner who has not yet absorbed what the antemortem file "should" show. The reason is the same one that governs every comparison discipline (Chapter 31): if you chart the body already knowing that the presumed decedent had, say, a gold crown on the upper-left first molar, you are primed to see a gold crown there, to resolve ambiguity in its favor, and to underweight discrepancies. Charting the body first, on its own terms, then comparing, keeps the postmortem record honest. Not every office enforces this, which is itself a finding.

Where do antemortem records actually come from in a real case? This is the unglamorous engine of the whole method, and it is where most identifications are made or lost. Once investigators have a presumptive identity — a missing-person report, a wallet, a vehicle registration, a family's belief — they request that person's dental records from their dentist of record. This is the rate-limiting step. The records may be:

• Available and complete, with recent films — the best case, and the one that yields a confident identification.
• Available but thin — a chart with no films, or films a decade old that predate later dental work.
• Unavailable — the person had no regular dentist, the dentist has retired and purged old files, the records were never digitized, or the person is genuinely unknown and there is no presumptive identity to request records for.

🧠 Cognitive-Bias Watch Notice the structural trap built into how antemortem records are obtained. Dental identification almost always begins with a presumed identity — the investigation already believes it knows who the body is, and goes looking for that person's records to confirm it. That is reasonable; you cannot request records from "everyone." But it means the odontologist usually works knowing the answer they are expected to produce, which is precisely the condition under which confirmation bias operates (Chapter 31). The safeguard is methodological: the conclusion must rest on concordant features that could have excluded the identity and didn't, not on a global impression of similarity. A responsible odontologist asks, "what would I have to see to reject this identification?" — and looks for it.

There is one more category worth flagging because it complicates real cases: dental work performed after the last antemortem record. People keep going to the dentist. If the most recent film on file is from three years ago and the decedent had two teeth extracted and a bridge placed since, the body and the record will disagree on those teeth — not because it is the wrong body, but because the record is stale. A naive examiner could read those discrepancies as an exclusion. This is why the distinction between an explainable discrepancy (later dental work, in a biologically plausible direction — a tooth that was present can later be missing; a tooth that was missing cannot reappear) and an unexplainable discrepancy (a tooth recorded as extracted that is now present, intact) is central to honest charting (§17.3). The direction of change is a logical check the method leans on heavily.

17.3 Dental charting and comparison

Dental charting is the systematic notation of the state of every tooth — present, missing, restored, and how — in a standardized format that lets two examiners (and two records made years apart) describe the same mouth the same way. Without a shared charting convention, an antemortem chart from one dentist and a postmortem chart from an odontologist could not be compared at all. Three numbering systems dominate, and you should recognize them because mismatched systems are a real source of error.

THREE WAYS TO NUMBER THE SAME MOUTH (upper teeth, patient's right → left)

  Universal (US):   1  2  3  4  5  6  7  8  | 9 10 11 12 13 14 15 16
  FDI (two-digit):  18 17 16 15 14 13 12 11 |21 22 23 24 25 26 27 28
  Palmer (quadrant + number, with a bracket denoting which quadrant)

  Tooth "3" (Universal) = "16" (FDI) = upper-right first molar.
  Same tooth, three labels. A chart that doesn't state its system is a chart waiting to be misread.

The Universal Numbering System (teeth 1–32, starting at the upper-right third molar) is standard in the United States. The FDI World Dental Federation system (two-digit: first digit = quadrant, second = tooth) is the international standard. The Palmer notation uses a quadrant bracket plus a number. The practical forensic point is blunt: the same tooth carries different numbers in different systems, so an antemortem chart in one notation compared against a postmortem chart in another is an error waiting to happen. Confirming the notation system is step one of any comparison.

Onto that numbered framework, the chart records the condition of each tooth: present and sound; missing (and whether extracted, congenitally absent, or lost postmortem); and every restoration — amalgam or composite fillings (noted by which of the tooth's surfaces they cover), crowns, bridges, implants, root canals, and so on. The surfaces themselves have a notation (mesial, distal, occlusal, buccal/facial, lingual), so a filling can be recorded precisely as, say, "MOD amalgam" — covering the mesial, occlusal, and distal surfaces. It is this surface-level, tooth-by-tooth specificity that gives the comparison its discriminating power: it is not "this person had fillings" but "this person had a three-surface amalgam here, a crown there, this molar extracted, this one root-canaled," a configuration of features whose joint occurrence in a random other person becomes vanishingly unlikely as the features accumulate.

🔬 Read the Evidence

text FIGURE 17.1 — "Two charts of one upper jaw" [constructed teaching example] THE ITEM An antemortem dental chart (from a dentist's file, 14 months before death) and a postmortem chart of the same upper arch made at the morgue, plus a pair of bitewing radiographs, antemortem and postmortem, of the upper-right quadrant. THE CONTEXT A presumptive identity exists; the dentist's records were requested and received. The postmortem arch was charted first, independently, before the AM chart was studied. WHAT IT SHOWS Concordant: upper-right first molar (Universal #3) carries an MOD amalgam of matching outline on both films; #14 is crowned in both; #1 and #16 (third molars) absent in both. The radiographs overlay closely on root morphology and the trabecular bone pattern. One discrepancy: #5, present in the AM chart, is absent in the body. WHAT IT DOESN'T The charts alone don't reveal *why* #5 is gone. The match of one MOD amalgam, by itself, is a class-level feature shared by many people; strength comes from the *combination*. THE INFERENCE The #5 discrepancy is EXPLAINABLE — a tooth present 14 months ago can be extracted since; the socket shows healed bone consistent with an extraction after the AM film. The concordant restorations, missing third molars, and overlaying root/bone anatomy together are a strong, concordant identification: this is very probably the presumed person. THE LESSON Direction matters: present→missing is explainable, missing→present is not. And no single feature identifies; the configuration does. State the strength as the features earn it.

How does a comparison reach a verdict? The American Board of Forensic Odontology has published guidelines that sort the conclusion into a small number of categories. The exact wording has been revised over the years, but the logical structure is what matters and is stable:

• Identification (positive / established): the antemortem and postmortem data show concordant features of sufficient quantity and specificity, with no unexplainable discrepancies, such that the examiner concludes the records are from the same individual.
• Possible / consistent identification: the features are concordant and there are no unexplainable discrepancies, but the available data are insufficient to establish identity to a confident level (e.g., a sparse record, few restorations, no films).
• Insufficient evidence: the available records do not permit any conclusion — too little antemortem data, or postmortem remains too fragmentary.
• Exclusion: the antemortem and postmortem data are irreconcilably inconsistent — an unexplainable discrepancy establishes that the records are not from the same person.

🔍 Check Your Understanding 1. The body's upper-right first molar is intact; the antemortem record says that tooth was extracted two years ago. Is this an explainable or an unexplainable discrepancy, and what does it support? 2. Why is matching a single common restoration (one ordinary filling) weak, while matching a dozen specific features across the mouth is strong? (Think back to the class-vs-individual logic of Chapter 1.)

Two honest cautions belong here, both tied to the book's themes. First, the conclusion is, at its core, an examiner's judgment. "Sufficient concordant features" and "no unexplainable discrepancies" are not bright-line thresholds; reasonable odontologists can weigh the same charts and films and reach "identification" versus "possible." Radiographic overlay — superimposing antemortem and postmortem films to see whether root shapes, sinus outlines, and trabecular bone patterns coincide — is the most objective single technique available, and good identifications lean on it heavily, but the final call still involves expert interpretation. That places even the valid half of odontology not at the rigorously-quantified end of the spectrum where good DNA sits, but in the better-grounded middle: a method with a sound logical basis and real discriminating power, whose error is dominated by human factors — charting mistakes, missed discrepancies, and confirmation bias — rather than by any flaw in the underlying anatomy.

Second, the exclusion is the cleaner result, exactly as Chapter 1 promised for forensic science generally. A single unexplainable discrepancy — a tooth confidently recorded as crowned that is, in the body, plainly sound and never treated — can exclude an identity with more confidence than any number of concordances can establish one, because concordance is shared by an unknown number of similar mouths while a genuine irreconcilable difference can come from only one place: the wrong body. Odontology's surest voice, like the rest of forensic science's, is the one that says not this person.

17.4 Identification in fire and mass-disaster contexts

Dental identification earns its place in forensic science most clearly in the two settings where almost nothing else works: the fire death, where soft tissue and prints are destroyed but the protected dentition survives, and the mass-fatality disaster, where the dead are many, commingled, and often fragmented. These are the situations the discipline was built for.

The single fire death. Return to the body that opened this chapter. A structure fire has consumed the soft tissue, taken the fingertips, and left the face unreadable. What the fire has usually not destroyed is the dentition, because the lips, tongue, and closing of the jaws shield the teeth, and because dental restorations endure heat that obliterates skin and char bone. So the recovery team stabilizes and removes the jaws with great care (recall §17.1 — burned teeth are friable and can shatter), the odontologist charts and radiographs the dentition, the presumed decedent's antemortem records are requested, and the comparison proceeds exactly as in §17.3. In a very large fraction of otherwise-unidentifiable fire deaths, the teeth are the identification.

🔬 At the Bench A fire scene imposes a specific recovery discipline that the CSI of Chapter 2 must know. Because the post-fire dentition is fragile, the cranium and jaws are documented in place, the surrounding debris is excavated rather than swept (small restorations, crowns, and tooth fragments are easily lost in ash), and the jaws may be supported or even consolidated before lifting. Loose teeth and restorations recovered from the debris are collected — a gold crown found in the ash three feet from the body still belongs to the identification. The temptation to recover quickly, especially with a family or a press presence on scene, is exactly the pressure that destroys the dental identification. Slow is correct.

The mass-fatality disaster. When an aircraft crashes, a building collapses, a fire sweeps a crowded structure, or a natural disaster kills dozens to thousands, identification becomes a different kind of problem — one of scale, condition, and commingling — and it is governed by a formal international framework that this book treats fully in Chapter 35 under the name disaster victim identification (DVI). We preview it here because dental identification is one of its three primary pillars.

Internationally recognized DVI practice (codified in the Interpol DVI framework, Chapter 35) accepts three primary identifiers — the categories of evidence considered reliable enough to establish identity on their own: fingerprints, dental records, and DNA. Everything else — clothing, jewelry, tattoos, personal effects, physical characteristics — is a secondary identifier that can support but generally cannot, alone, confirm an identification. Dental identification holds its place among the primaries for reasons that are now familiar and that make it especially valuable in disasters:

• Survivability. In fires, explosions, and high-energy impacts that destroy fingertips and degrade DNA, teeth and restorations frequently survive.
• Speed and cost. A trained odontologist can produce a comparison quickly and cheaply relative to DNA, which (outside of rapid-DNA settings, Chapter 29) can mean weeks of laboratory work per sample and is sensitive to the very degradation a disaster causes.
• Availability of antemortem data. In populations with regular dental care, antemortem records exist for a large share of victims — though, importantly, not all, which is the method's structural ceiling.

🔬 Read the Evidence

text FIGURE 17.2 — "The reconciliation board" [constructed teaching example, after general DVI practice] THE ITEM A simplified DVI reconciliation matrix: rows are postmortem cases (bodies/fragments, PM-001 …), columns are antemortem files (reported-missing persons, AM-001 …); each cell holds the status of a dental comparison between that body and that missing person. THE CONTEXT A mass-fatality event with ~40 presumed victims; remains are commingled and incomplete. Postmortem dental exams and reported-missing antemortem records are entered separately, then matched. (Fingerprint and DNA matrices run in parallel.) WHAT IT SHOWS Most cells are blank (no plausible match). A handful show strong dental concordance; one row (PM-014) excludes three candidate AM files outright on unexplainable discrepancies and concords strongly with a fourth. WHAT IT DOESN'T A single dental concordance in a large population is *less* discriminating than the same concordance in a two-person problem — with 40 candidates, common restoration patterns can coincide. The board narrows; it does not, by one method alone, finalize an ID. THE INFERENCE PM-014 is dentally identified as the fourth AM individual — strong on its own, and confirmed when the parallel DNA matrix agrees. Reconciliation across *multiple* primary identifiers is how DVI reaches confident IDs at scale. THE LESSON At scale, the prior pool is large; a method's discriminating power must be judged against the number of candidates, and convergence of independent identifiers is the safeguard.

That last figure carries a statistical point worth making explicit, because it is a quiet form of the prosecutor's fallacy (Chapter 9) lying in wait. A dental concordance that would be highly persuasive when you are deciding between two possible identities is weaker when you are searching a pool of hundreds, because in a large enough pool, mouths with broadly similar restoration patterns exist by chance. The strength of a dental identification is not an absolute property of the teeth; it depends on the size of the candidate pool it is drawn from. This is why mature DVI does not rest a contested identification on dental concordance alone in a large event — it seeks convergence across the independent primary identifiers, the same convergence logic that governs the whole cold case and that the capstone (Chapter 39) is built on.

There is a humane dimension here that the clinical language can hide, and it is worth stating plainly. Behind every row of a reconciliation matrix is a family waiting to bury someone. Forensic identification — dental, fingerprint, DNA — is, in the disaster setting, as much a humanitarian act as an investigative one: it is the act of returning the dead to the people who loved them, and of refusing to hand back the wrong body. Chapter 35 develops that responsibility in full.

17.5 The two faces of odontology: valid ID versus invalid bite marks

We come to the reckoning this chapter has been pointing toward, and to the reason forensic odontology is the cleanest single illustration of this book's second theme — not all forensic methods are equally valid — because here the same specialty, practiced by the same credentialed experts, produces work at opposite ends of the validity spectrum depending only on the question asked. Hold the two side by side.

Dental identification asks: whose body is this? It compares a documented antemortem dental record to a postmortem examination of the dentition, reaches a conclusion that can be excluded by a single unexplainable discrepancy, leans on objective radiographic overlay, and rests on a sound logical foundation — a lifetime of documented, individually specific dental history. It is not flawless (its errors are human: charting mistakes, missed discrepancies, bias when the answer is "known"), and it is not as rigorously quantified as good DNA. But it is valid: it does, reliably, what it claims to do.

Bite-mark comparison asks something entirely different: did this person's teeth make this mark? — typically a bruise or laceration in human skin. This book formally introduced bite-mark analysis, and named it as discredited, in Chapter 16; here we can say precisely why the contrast with dental identification is so total, because the differences are not matters of degree but of kind:

The decisive difference is the substrate and the premise. Dental identification compares a stable, mineralized dentition to a record of that same dentition. Bite-mark comparison asks whether a set of teeth made a mark in human skin — a material that stretches, distorts, swells, bruises diffusely, heals and changes over hours and days, and registers a bite differently depending on the body region, the victim's movement, and time. The premise that (a) every person's dentition is unique in the features a bite transfers, and (b) skin faithfully records those features well enough to match them back to one mouth, has never been validated, and study after study has shown examiners disagreeing even on whether an injury is a bite, let alone whose. The 2009 NAS report and the 2016 PCAST report (Chapter 6) treated bite-mark comparison as a paradigm of a method lacking the foundational validity its courtroom claims require.

⚠️ Junk-Science Alert The danger is precisely that the credential is the same. A board-certified forensic odontologist who correctly identifies a fire victim from dental records on Monday may take the stand on Tuesday and testify that a suspect's teeth "match" a bruise on a victim's arm "to a reasonable degree of dental certainty" — and the jury hears one continuous expert authority, with no way to know that the first claim is well-founded and the second is not. The credential does not travel with the validity. What an expert is qualified to do in one task does not certify what they claim in another. This is the single most important thing a juror, attorney, or judge can carry out of this chapter: ask not "is this witness a qualified odontologist?" but "which question is this odontologist answering, and is that question one the method can validly answer?"

⚖️ In the Courtroom The two faces produce two utterly different sets of permissible verbs. For a dental identification, an odontologist can defensibly testify: "The antemortem and postmortem dental records exhibit concordant features and no unexplainable discrepancies; in my opinion they are from the same individual." That is a claim the method supports. For a bite mark, the honest ceiling has collapsed over the past two decades: many odontologists and standards bodies now hold that an examiner may, at most, say a suspect cannot be excluded as a possible biter, or that an injury is consistent with a bite — and a growing number say the comparison should not establish identity at all. An expert who testifies that a defendant's teeth made a particular mark, to the exclusion of others, is making an individualization claim (Chapter 1) that the science cannot back. On cross, the question that exposes it is the one from Chapter 1: what is the measured error rate of this comparison, and where is the study that established it? For dental identification there is a coherent answer about concordance, discrepancy, and overlay; for bite-mark individualization there is not.

The exonerations make the contrast concrete and human. Innocent people have been convicted — some sentenced to death — on bite-mark testimony later shown to be worthless, and subsequently freed, often by the very method that anchors the valid end of forensic science: DNA. The book examines those bite-mark exonerations in Chapter 16, where the method is owned. The point to carry from this chapter is the structural one: a single discipline can be both a humanitarian science and a wrongful-conviction engine, and the difference between them is not the expertise of the practitioner but the validity of the specific claim. Forensic odontology, looked at honestly, is the second theme of this book made flesh — proof that "is the witness an expert?" is the wrong question, and "is the claim valid?" is the right one.

🔍 Check Your Understanding 1. A witness is "a board-certified forensic odontologist." Why does that credential, by itself, tell you nothing about whether their testimony is valid? 2. Name the single biggest reason bite-mark comparison fails where dental identification succeeds. (Hint: what is each one compared against, and what is the substrate?)

🗂️ The Case File

Carrow County — confirming whose body it is. A theory has been circulating around the Mill Creek case, and it is the kind of theory a large insurance policy tends to generate: that the badly burned body in the cabin is not Marcus Diallo at all — that Diallo staged his own death, planted someone else's body, and means to collect or vanish. It is not a crazy theory on its face. The fire destroyed the soft tissue and the face; routine visual identification was never possible; and there is real money riding on the answer. Before the investigation can ask who killed Marcus Diallo, it has to answer the prior question this whole chapter is about: is the dead man actually Marcus Diallo?

Here is where odontology does its real, valid work. Diallo had a regular dentist, and his antemortem dental records — a chart and a set of bitewing and panoramic radiographs from within the last two years — are obtained. At the morgue, the forensic odontologist charts the burned body's dentition independently first (the postmortem record), then compares. The concordance is strong and specific: matching restorations on multiple teeth, a distinctive crown, a pattern of missing teeth, and — most tellingly — antemortem and postmortem radiographs whose root morphology, sinus outline, and trabecular bone pattern overlay closely. There are no unexplainable discrepancies. The conclusion is a positive dental identification: the burned body is Marcus Diallo.

What this does and does not establish. It establishes, on valid grounds, the identity of the decedent — and in doing so it kills the "someone else's body" fraud theory that the insurance question raised. That is not a small result; an entire alternative narrative is now closed. What it does not establish is anything about how Diallo died, when, or who was responsible — dental identification answers "who is this body," full stop. It does not speak to the homicide finding from the autopsy (Chapter 11), the arson question (Chapter 22), or any suspect. It also rests, honestly, on the antemortem records being genuinely Diallo's and correctly filed — a chain that the investigation verifies but that a careful analyst keeps in view. Running status: the victim is confirmed to be Marcus Diallo; the "faked-death / substituted-body" theory is excluded; the questions of manner, cause, and responsibility remain open. (Log the identification, the records' provenance, and the "no unexplainable discrepancies" finding in your Appendix I workbook.)

Conclusion

Forensic odontology answers the first question any death investigation must answer — who is this? — in the hardest cases, the ones where fire, decomposition, water, or time have destroyed every easier identifier. It can do so because teeth are the body's most enduring tissue and because a lifetime of dental care writes an individually specific, documented history into the mouth. The method is a disciplined comparison of an antemortem record to a postmortem examination, expressed through standardized dental charting, reaching a verdict — identification, possible, insufficient, or exclusion — that is anchored, at its most objective, by radiographic overlay. It is genuinely valid: not as rigorously quantified as good DNA, its errors dominated by human factors (stale records, charting mistakes, and the confirmation bias that comes from usually knowing the answer in advance), but reliable for the question it answers. And its surest result, true to the whole book, is the clean exclusion a single unexplainable discrepancy can deliver.

The chapter's deeper lesson is the one that makes odontology the sharpest illustration of the validity spectrum in the book: the same specialty, the same experts, produce valid identification work and discredited bite-mark work, and the difference between justice and wrongful conviction lies not in the credential but in which question is being answered. "Is the witness a qualified odontologist?" is the wrong question. "Is the claim one the method can validly support?" is the right one.

In the cold case, that valid work has now confirmed that the burned body is Marcus Diallo and closed the faked-death theory for good. The next chapter turns from the body to the paper it left behind — the charred insurance and partnership documents in the cabin (Chapter 18) — where questioned-document examination will ask whether the policies that make this death profitable were themselves altered.

Key Terms

• Forensic odontology — the application of dental science to legal questions; centrally the identification of human remains from dental records (valid), as distinguished from bite-mark comparison (discredited).
• Dental identification — establishing the identity of a deceased person by comparing antemortem dental records to a postmortem examination of the dentition.
• Antemortem records — dental records created during a person's life (radiographs, charts, treatment notes, models); the known side of a dental-identification comparison.
• Postmortem records — the dental examination, charting, and radiographs the forensic odontologist creates from the body; the questioned side of the comparison.
• Dental charting — the standardized, tooth-by-tooth, surface-by-surface notation of the dentition (present/missing/restored) that allows two records to be compared like to like.

Spaced Review

1. Explain why a single unexplainable dental discrepancy (a tooth recorded as extracted that is present in the body) can exclude an identity more confidently than a dozen concordant restorations can establish one. (§17.3, and the exclusion logic of §1.6)
2. The cold case used dental records to confirm the burned body is Marcus Diallo. What exactly does that identification establish, and what does it not establish about how he died? (§17.4, The Case File)
3. Validity-spectrum question: Dental identification and bite-mark comparison are practiced by the same experts, yet sit at opposite ends of the NAS/PCAST validity spectrum. Name the single feature most responsible for the gap, and state where each method sits. (§17.5, §6)
4. In Chapter 11, the autopsy concluded Diallo was dead before the fire. How does this chapter's finding (a confirmed identity) relate to that one — does either depend on the other? (§17.4; Chapter 11)
5. A latent-print examiner (Chapter 14) and an odontologist both compare a "questioned" item to a "known." For dental identification, what plays the role of the known reference, and why is its source (a treating dentist with no stake in the case) relevant to its reliability? (§17.1–17.2; Chapter 3 reference samples)