Case Study 17.2 — The Two Faces in One Courtroom: Bite-Mark Testimony in *Florida v. Bundy* and the Collapse of Its Foundation

DataField.Dev

Case Study 17.2 — The Two Faces in One Courtroom: Bite-Mark Testimony in Florida v. Bundy and the Collapse of Its Foundation

Sourcing and tone. This case study draws on the public record of a widely documented U.S. trial (Florida, 1979) and on the subsequent, publicly reported reassessment of bite-mark comparison by standards bodies and by one of the case's own experts. It is used to teach a single methodological point central to this chapter: that the same specialty — forensic odontology — produces valid identification work and a discredited comparison claim, and that the difference lies in the question and the substrate, not in the practitioner's credentials. We treat the victims soberly and confine ourselves to documented, public facts. The defendant was convicted (and the conviction stands); this study is not a claim of innocence but an examination of one kind of evidence whose scientific foundation has since been rejected, and of why that matters for every future case.

Background

In January 1978, two young women, Margaret Bowman and Lisa Levy, were murdered and others attacked in the Chi Omega sorority house at Florida State University. Ted Bundy was tried for these crimes in 1979. Among the evidence the prosecution presented was a bite mark found on one victim, which a forensic odontologist compared to impressions of Bundy's teeth.

The case is one of the most famous in the history of forensic odontology — and it is famous for the wrong reason, which is exactly why it belongs in this chapter. It is routinely cited as a courtroom triumph of "dental evidence." But the dental evidence that convicted Bundy was not the valid kind this chapter has spent its length defending (identifying a body from dental records). It was bite-mark comparison — the discredited application, owned and dissected in Chapter 16 — presented at the height of the era when courts and juries trusted it.

The forensic evidence

The bite-mark testimony followed the pattern §17.5 describes and Chapter 16 debunks:

The mark and the impressions. A bite mark was documented on one victim, photographed, and preserved. Impressions of Bundy's dentition were obtained, and the odontologist, Dr. Richard Souviron, compared the suspect's teeth to the injury.
The demonstration. The comparison was presented to the jury in a now-iconic form: an enlarged photograph of the bite mark, overlaid with an enlarged image of the suspect's teeth, with the irregularities of Bundy's dentition argued to correspond to features of the mark. The testimony asserted, in substance, that the suspect's teeth matched the bite to a high degree of dental certainty.
The persuasive power. To a 1979 jury, this was compelling, visual, expert testimony, and it contributed to the conviction. It also helped cement bite-mark comparison's reputation, in the public mind and in courtrooms for years afterward, as a legitimate forensic identification tool.

⚠️ Junk-Science Alert Notice what the jury could not see. The same expert authority that could validly identify a fire victim from dental records — a claim resting on stable teeth and a documented history (§17.1–17.3) — was here deployed to make a different kind of claim: that a particular set of teeth made a particular mark in human skin. As §17.5 lays out, that second claim rests on an unvalidated premise (that a dentition is unique in the features a bite transfers, and that skin faithfully records them), and on a substrate — skin — that stretches, swells, bruises diffusely, and changes over hours. The credential did not travel with the validity. A board-certified odontologist made the claim, so it sounded like science of the same kind that identifies the dead. It was not.

What the evidence did — and didn't — establish

Here the case demands unusual care, because the lesson is not "an innocent man was convicted." Bundy's guilt was established by a broad body of evidence and is not in serious dispute, and his conviction stands. The point is narrower and, for this book, more important: the bite-mark testimony was presented as a valid scientific identification, and the scientific foundation for that kind of testimony has since collapsed. That a confidently asserted method happened to point at a genuinely guilty man does not make the method valid — it makes this a case where an unvalidated technique was lucky in its target, which is the most dangerous way for junk science to build its reputation.

The reassessment is now on the public record and is striking precisely because it comes partly from inside the case. Dr. Souviron himself has been reported, decades later, as acknowledging that he would testify differently today given how the field's understanding has changed. More broadly, the 2009 NAS report and the 2016 PCAST report (Chapter 6) treated bite-mark comparison as a paradigm of a feature-comparison method lacking established foundational validity, and standards bodies have sharply curtailed what an examiner may claim — many now holding that bite-mark analysis should not be used to assert that a specific person made a mark, and in some formulations should not establish identity at all. The exonerations driven by bite-mark testimony in other cases (the wrongful convictions examined in Chapter 16, e.g., the well-documented cases of Ray Krone and Roy Brown) are where the method's body count is counted; here, the lesson is the reputation such a famous "success" lent a method that could not back its central claim.

Contrast all of this with the valid face of the same specialty, demonstrated in Case Study 17.1 and throughout this chapter. A dental identification compares a body to a documented record of that same body's dentition, can be excluded by a single unexplainable discrepancy, leans on objective radiographic overlay, and rests on a sound logical foundation. A bite-mark comparison compares a wound in skin to a suspect's teeth under an unvalidated premise, with no comparable objective anchor. Same experts; opposite ends of the validity spectrum (§17.5).

⚖️ In the Courtroom The two faces produce two different sets of permissible verbs (§17.5). For a dental identification, an odontologist may defensibly say the antemortem and postmortem records are concordant with no unexplainable discrepancies and, in their opinion, from the same individual. For a bite mark, the honest ceiling has collapsed: at most, modern, cautious testimony may say a suspect cannot be excluded or that an injury is consistent with a bite — and a growing position holds the comparison should not establish identity at all. The cross-examination question from Chapter 1 is the solvent: 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 — and a famous conviction does not supply one.

The lesson

Three lessons, all central to this chapter:

A method is not validated by hitting a guilty target. Bite-mark testimony helped convict a man who was, by other evidence, genuinely guilty — and that very fact lent the method an unearned credibility that traveled into later cases where the target was innocent. Validity is a property of the method's foundation and error rate, not of any single outcome. This is the most counterintuitive and important point in the study.
The credential does not certify the claim. The same board-certified odontologist can make a valid identification claim and an invalid bite-mark claim, and a jury hears one continuous expert voice. The right question is never "is the witness qualified?" but "is this particular claim one the method can validly support?" (§17.5). Dental identification can support its claim; bite-mark individualization cannot.
The contrast is the validity spectrum. Forensic odontology is the cleanest single illustration of this book's second theme because the same specialty sits at both ends of the spectrum at once. The difference between the humanitarian science of Case Study 17.1 and the discredited comparison here is not expertise — it is the question asked and the substrate compared against: a documented dental record versus a bruise in elastic, healing skin.

Discussion questions

The study insists the lesson is not "an innocent man was convicted." Restate, in your own words, what the lesson is, and explain why "the method happened to point at a guilty person" is presented as dangerous rather than reassuring.
Using §17.5's comparison table, list the differences between dental identification and bite-mark comparison along the dimensions of question, substrate, objective anchor, and uniqueness assumption. Which single difference does the chapter call decisive, and why?
Dr. Souviron is reported to have said he would testify differently today. Using Chapter 5 (Daubert) and Chapter 6 (foundational validity), explain what changed — not about the case, but about the field's understanding of the method — that would justify different testimony on the same facts.
A juror hears "a board-certified forensic odontologist" testify and treats all of the testimony as equally scientific. Using the Junk-Science Alert in this study and in §17.5, explain why the shared credential is precisely what makes this dangerous, and what a juror should ask instead.
Compare the role of forensic odontology in (a) Case Study 17.1 (the Noronic identifications), (b) this case (bite marks), and (c) the cold case (confirming the body is Marcus Diallo). In which is the discipline operating validly, and what feature of each — the question, the substrate, the record — sets that?
Ethics tie-in (Chapter 31, previewed). Imagine an examiner who has already identified many bodies validly by dental records, now asked to compare a bite mark with the prosecution's preferred answer known to them. Explain how confidence built on the valid work could bleed into overconfidence on the invalid task, and what safeguard (context management, knowing the limits of the claim) should prevent it.