Bite mark comparison rested on two premises that research never established. In February 2016 a state oversight body did something no institution in this series had managed: it stopped the method, ordered identification of every affected conviction, and opened a path to relief. It was the same commission that had just been prevented from assigning responsibility in an execution case.
Bite mark comparison assumes human dentition is unique and that its features transfer accurately to skin. Neither has been established. In February 2016 the Texas Forensic Science Commission unanimously recommended a moratorium and ordered a review of every Texas conviction using the technique. That combination of discontinuation, case identification, and a path to relief is the only complete institutional response documented anywhere in this series.
How did bite mark comparison get into courtrooms in the first place?
Through the same route as the disciplines in Parts II and III. It was admitted because practitioners in the field accepted it, and the field consisted of the practitioners. A California appellate decision in the mid 1970s is generally treated as the seminal admission case.
Is bite mark evidence banned now?
No. The Texas commission recommended a moratorium and lacks power to impose one. Recommendations from scientific bodies do not bind courts. Admissibility decisions remain with individual judges in individual cases.
How many convictions are involved nationally?
Reporting around the 2016 Texas decision described at least two dozen people convicted or charged on bite mark evidence who were later exonerated. That counts cases somebody found and litigated. It is not a count of cases affected.
Why does a discipline this narrow matter?
Because its narrowness is the point. Reviewing every bite mark conviction in the country would be a manageable task, and the failure to do it nationally cannot be explained by scale.
What I Notice First About This One
I have spent four installments describing institutions that did not do their jobs, and I want to be careful here, because this installment breaks the pattern and I nearly missed it.
The reflex when writing a series like this is to find the failure in every case. Bite mark comparison is the weakest discipline examined so far, weaker than hair, weaker than arson doctrine, and the temptation is to treat it as the worst chapter. On the science, it is. On the institutional response, it is the best thing that happens anywhere in this series.
In February 2016 a state body did all three of the things Part II said nobody ever does. It stopped the method. It ordered someone to go find the affected cases. And it built a route toward relief for the people in them.
What makes it worth an entire installment is which body did it. It was the Texas Forensic Science Commission, four and a half years after the sequence described in Part IV left it unable to say whether anyone had been negligent in a case that ended with an execution. The commission that lost its accountability function used the function it still had, and used it better than any comparable institution in the country.
Two Premises, Neither Established
Bite mark comparison asks an examiner to look at a patterned injury on skin and determine which set of teeth produced it. The entire discipline rests on two assumptions stacked on top of each other.
The first is that human dentition is unique at the individual level, distinctive enough that one person’s bite can be distinguished from every other person’s. The second is that this uniqueness transfers to skin, so that the pattern left in tissue preserves enough individual detail to support a comparison.
Neither assumption was established before the discipline entered courtrooms, and research has since put both in serious doubt. Skin is not a recording medium. It is elastic, it moves, it swells, it bruises unevenly, and the same set of teeth applied to the same person twice will not leave the same mark. Whatever individuality exists in a mouth, the question is what survives the transfer, and the answer appears to be very little.
Hair comparison and arson doctrine at least began as attempts to read physical evidence, and failed at the step where similarity was converted into identification. Bite mark comparison fails one step earlier. There was never a foundation for the lag to run against. The method did not become invalid over time. It was never valid.
What the Reviews Found
Four separate scientific bodies have now examined the discipline and reached compatible conclusions. The 2009 National Academy of Sciences report identified the weakness of its foundations. The Texas Forensic Science Commission examined it in 2016. The President’s Council of Advisors on Science and Technology examined it the same year. The National Institute of Standards and Technology addressed it again in a 2022 draft report.
The PCAST findings are the most specific and the most damaging. Few studies had been conducted, and none of them were the appropriately designed black box studies that establish whether a subjective method works. In the studies that did exist, observed false positive rates were typically above ten percent and sometimes far above. Several used closed set designs likely to understate the true rate. PCAST concluded that examiners not only cannot identify the source of a bite mark with reasonable accuracy, they cannot consistently agree on whether an injury is a human bite mark at all.
Read that second finding again, because it is the one that ends the argument. Before an examiner can compare a bite mark to a suspect, the examiner has to determine that the injury is a bite mark. If practitioners cannot agree on that threshold question, everything downstream is comparison of something that has not been identified.
The Clutch Justice course track covers reading forensic reports and testimony against the underlying research, including how to identify when a stated conclusion exceeds what a method can support.
Browse the CoursesThe Complaint That Started It
The Texas review did not begin with a commission deciding to examine a discipline. It began with a complaint filed on behalf of one person.
Steven Mark Chaney was convicted of the murder of John Sweek and served twenty-eight years. Bite mark testimony was central to the case against him. Following an investigation by the Dallas District Attorney’s Conviction Integrity Unit, he was released in October 2015. The Innocence Project then filed a complaint with the Texas Forensic Science Commission asking it to examine the discipline itself rather than only the case.
That distinction is the mechanism worth noticing. A conviction integrity unit resolves a case. A complaint that asks an oversight body to examine the underlying method converts one case into a review of a category. Very few post-conviction efforts are structured to do the second thing, and the ones that are require an institution capable of receiving the question.
February 2016
After a six month investigation, the commission concluded that bite mark comparison did not meet the standards of forensic science, and unanimously recommended a moratorium on its use in Texas criminal prosecutions until the technique is scientifically validated. Texas became the first state in the country to take that step.
The commission went further. It ordered a review of every Texas conviction in which the technique had been used, to be conducted by a panel of experts including forensic dentists working from trial transcripts. The American Board of Forensic Odontology agreed to help identify the convictions, and reported that nine of its members practicing in Texas would assist.
The commission did not close the door permanently. Its April 2016 report set out conditions under which bite mark comparison could return, beginning with criteria for determining when a patterned injury is a human bite mark. Commissioners placed the burden on the practitioners who defend the technique to produce research adequate to meet it.
Moratorium recommended unanimously, February 2016. Conditions for return specified in the April 2016 report.
Performed. This is the step institutions reliably manage, because it is internal and costs nothing.
Review ordered of every Texas conviction using the technique, with an expert panel working from transcripts and the professional body assisting in locating cases.
Performed. This is the step that does not happen anywhere else in this series.
A path opened, since an identified conviction can be litigated. Relief itself still requires counsel, a filing, and a court.
Partially performed. Identification is not vacatur, but it is the precondition without which vacatur is accidental.
Every other episode in this series ends at obligation one. Texas reached obligation two, and reaching obligation two is what converts a scientific finding into a list of human beings. A list can be worked. A finding cannot.
A California appellate decision is generally treated as the seminal case admitting bite mark identification testimony. The discipline enters American courtrooms carrying the vocabulary of odontology.
The NAS assessment of forensic science identifies the weakness of bite mark foundations, alongside its findings on hair comparison and other pattern disciplines.
After twenty-eight years and an investigation by the Dallas District Attorney’s Conviction Integrity Unit, Chaney is released. The Innocence Project files a complaint with the Texas Forensic Science Commission about the discipline itself.
After a six month investigation, the commission unanimously recommends a moratorium on bite mark comparison in Texas criminal prosecutions, the first such recommendation by a state body, and orders a review of every conviction where the technique was used.
The commission’s report sets out what would have to be established before bite mark comparison could be admitted again, beginning with criteria for determining when a patterned injury is a human bite mark.
The President’s Council of Advisors on Science and Technology finds bite mark analysis does not meet scientific standards for foundational validity and is far from doing so, citing false positive rates typically above ten percent and examiner disagreement on whether injuries are bite marks at all.
A National Institute of Standards and Technology draft report becomes the fourth major scientific body to find the central premises of bite mark analysis unsupported by research.
Why It Worked, and What It Did Not Fix
Three conditions made the Texas response possible, and it is worth naming them precisely, because they are the design specification for anywhere else.
First, an institution existed that could receive a complaint about a method rather than about a case. Second, the complaint was framed that way by people who understood the difference. Third, the professional body that certifies practitioners agreed to help identify affected convictions instead of defending the field. That third condition is unusual and should not be assumed anywhere else.
What it did not fix is also worth stating. The moratorium is a recommendation, and the commission has no power to impose one. Courts are not bound by scientific bodies. Identification of a conviction is not vacatur, and every person on that list still needs counsel, a filing, and a judge. And the review was confined to Texas, so the same limitation that ended the Willingham inquiry in Part IV, jurisdiction bounded by state lines and statutory dates, still bounds the remedy here.
The national picture is worse than the Texas picture by a wide margin. Reporting around the 2016 decision described the number of affected convictions nationally as manageable, in the range of a few hundred at most. A review of that size is not blocked by cost or by scale. It has not happened because no institution exists whose job it is, which is the same finding this series reached in Part II about a discipline used in tens of thousands of cases.
Michigan Cannot Do This
I want to be careful about what I am asserting here, because the honest statement is narrower than it looks and more useful.
I do not know whether there are bite mark convictions in Michigan. Neither does anyone else. There is no public record from which that question can be answered, and there is no institution with the authority to compel an answer.
Consider what Texas actually needed in order to act. It needed a standing body empowered to examine a forensic discipline. It needed authority to order a review across cases rather than within one. It needed a way to identify which convictions had used the technique, which in Texas meant an expert panel reading transcripts with assistance from the professional body.
Michigan has none of the three. The Task Force on Forensic Science recommended creating the first, and that body was not created. Without it there is no one to order the second. And the third, the ability to identify which convictions rest on which discipline, is exactly the case-level coding this series identified as missing in Part II and the register that Michigan’s own task force proposed and Michigan did not build.
Authority to review by discipline, not only by case. This is the specific power that made the Texas response possible and it is distinct from complaint jurisdiction over individual laboratories. Any Michigan enabling statute should grant the body express authority to open an inquiry into a method and to order identification of affected convictions as part of it.
A duty on certifying bodies to assist in identifying affected cases. In Texas the professional organization cooperated voluntarily. Voluntary cooperation is not a system. Certification carries privileges, and assisting in the identification of convictions that relied on a discipline the certifying body sponsored is a reasonable condition of holding them.
Why This Matters Beyond One Discipline
I said at the top that this installment breaks the pattern, and I want to end by being precise about what it proves, because it is easy to read as either more optimistic or more pessimistic than it is.
It proves the complete response is possible. Not theoretically possible, actually possible, performed by a real state body with a modest budget in six months, on a discipline whose practitioners were still actively defending it. Every argument that the full response cannot be done, that the caseload is unknowable, that the professional bodies will not cooperate, that oversight commissions are toothless, has a counterexample now.
It also proves the response depends entirely on the existence of an institution capable of receiving the question. Texas did not act because Texas is more committed to accuracy than Michigan. Texas acted because a complaint about a method had somewhere to go. Michigan’s version of that complaint has nowhere to be filed.
And there is one more thing in this record that I keep returning to. The commission that did this is the same commission that Part IV showed being systematically prevented from assigning responsibility in a case that ended in an execution. It kept the function it was allowed to keep, and then used that function to do more good than any comparable body in the country.
Read one way, that is encouraging. Read the way I read it, it is the sharpest possible illustration of what was taken. An institution capable of this was capable of the other thing too.
Bite mark comparison was stopped by an outside body over the objections of its practitioners. Part VI examines comparative bullet lead analysis, a discipline the FBI retired on its own initiative after a National Academy of Sciences review, and asks the question that retirement leaves open: what is owed to the people already convicted by a method an agency voluntarily abandons.
Continue Your Investigation
If this reporting raised more questions, use the Clutch Justice ecosystem to keep going.
