C.E. Credit. Bitemark Analysis: The Legal vs Scientific Battle for Justice

ABSTRACT

The forensic odontology discipline of bitemark analysis has been under scrutiny for a number of years. The significant issues associated with the practice were outlined in the 2009 National Academy of Sciences Report, the 2016 Texas Forensic Science Commission report, the 2018 President’s Commission on Science and Technology, and most recently in a 2022 draft report released by the National Institute of Standards and Technology. Current scientific study does not support the underlying premises of bitemark analysis; that the human dentition is unique and that the unique features transfer to human skin. As such, a number of wrongful convictions have resulted. Some of those convicted have spent over thirty years behind bars for a crime they did not commit, while others were on death row. This review article examines the emergence of bitemark analysis to the legal system, the research work accomplished by the authors of this review, how the legal, forensic odontology, and scientific communities have responded, and the possible path forward.

Continuing Education Credit Available: The practice worksheet is available as supplemental material for this article: https://doi.org/10.1080/19424396.2023.2191391.

A CDA Continuing Education quiz is online for this article: https://www.cdapresents360.com/learn/catalog/view/20.

KEYWORDS:

• Bitemark
• bitemark research
• reliability
• distortion
• wrongful conviction
• exoneration
• forensic odontology

Background

On April 8^th, 2016, Keith Harward walked out of prison as a free man.¹ He had spent 33 years behind bars, narrowly escaping death row, for a crime he did not commit. The main evidence used to convict him: a bitemark on the victim’s skin.

Mr. Harward is not alone. Twenty-six people have been wrongfully convicted based on bitemark evidence.² Unlike Mr. Harward, some were sentenced to death. One such individual was Eddie Lee Howard who spent twenty-six years on Mississippi Death row before being exonerated.^3,4

The dental forensic discipline of bitemark analysis has a dubious and contentious past.⁵ The blame can be levied in the fact that it is grounded in absolutely no scientific studies, or even any scientific fact. That people can lose life and liberty due to unsubstantiated evidence is something that nightmares are derived from. Case in point, Mr. Harward watched helplessly as a forensic dental expert explained to the jury how his teeth made the bite, and with such convincing detail that even his own family doubted his innocence.⁶ For Mr. Harward, DNA evidence identified the real perpetrator more than three decades later.

The premise of bitemark analysis rests on two main assumptions: That the arrangement of the human dentition is unique, and these unique features transfer to the skin. However, there are no scientific studies to support these premises, and the research that has been published demonstrates bitemark evidence is subject to serious flaws and startling misinterpretation.^5,7,8 This highlights and underscores the fact that when evidence is used in a court of law without any scientific backing as to its validity, serious errors will result. To date, no US court has excluded bitemark evidence.⁹

In the courtroom, there are two additional significant problems that arise with bitemark evidence. First, this type of evidence is very prejudicial to the jury. Not only will the expert demonstrate to the jury a comparison between the tracings of the suspect’s dentition and a photograph of the wound, the jury will also be privy to both, and they themselves can align the wound to the outlines of the teeth. Of course, the teeth ‘fit’ reasonably well. They always do. It’s hard for the jury to not be convinced at this point. The question is; how many other people could have ‘fit’ the bite. This, the jury does not know. No one could have known this without scientific study. The problem is that, based on scientific work completed thus far, this number could be significantly large, and those that fit best, are not usually those from the actual biter.^10–12

Second, bitemark evidence creates the necessary conditions for a death penalty case. Bitemarks typically occur during violent altercations and in cases of abuse. Introduction of this evidence implies the heinousness and aggravated circumstances that would worsen the seriousness of the crime. The defendant’s life could now be on the line in states that have the death penalty.

How is the legal battle of evidence different from that of science? Bitemark evidence made its way into a courtroom based on precedence. Several landmark cases made this possible. One of the first was People vs Marx in 1974.⁹ This case was significant in that it determined the methods of comparison were not novel and were scientifically established. Thus, photography, dental models, tracings, etc. were all established techniques. What they failed to consider was if there was any scientific basis for the comparison.

The next case, People vs Milone in 1976 added the caveat; bitemark evidence was admissible because it follows from the logical extension of “each person’s dentition is unique based on victim ID.⁹ ” Victim identification however is quite different. For victim identification, the combinations of present, missing, and restored of potential thirty-two teeth are analyzed and compared. Also, the evaluation is most often a direct comparison of dental radiographs. In bitemark analysis only the incisal edge of the six anterior maxillary and mandibular teeth are what typically leave the mark.

By 1978, bitemarks attained general acceptance by the relevant scientific community in California.⁹

In 1979, People vs Bundy gave it significant momentum. Ted Bundy was a notorious serial killer, and it was a bitemark on one of his victims that contributed to his arrest and subsequent execution.

Once precedent is set, it is hard to overcome. Courts are content to allow the evidence in and let rigorous cross-examination bring out any fault in the ‘science’. This leaves it to the lawyers, who may not have a scientific background to adequately cross-examine. However, it is not scientific background that can win a case, it is the ability to be convinced that may sway the courtroom, regardless of science.

In science, it is hypothesis-driven research and repetition of studies that establish validity. To begin scientific study into bitemark evidence, it would only seem logical to first test the transfer of patterns of teeth to skin. Does it transfer in a reliable manner? If the pattern of the dentition does not reliably transfer, then it doesn’t matter if the dentition is unique or not.

Review of Research

Examining the transfer of the dentition to skin is the initial approach we took in our studies at the University of New York at Buffalo School of Dental Medicine.

Study Example One

Methods and Materials

For our first study, we made twenty-three bites on un-embalmed cadavers, all with the same set of teeth and the same opening diameter.¹³ Metric measurements of the biting dentition as well as the bitemarks were completed and compared. The overall intercanine, mesial to distal, and angle of rotation distortion were calculated. We then photographed the bitten body part in varying positions to examine the effect of movement.¹³ It stands to reason that in an altercation a body part may be bitten in one position, but examination of the mark is conducted while the body part is in another. For example, if an arm is raised in a defensive move but later the wound is examined with the arm at the victim’s side.

Results

Of the twenty-three bites made, none were measurably identical to each other, nor the biting dentition, and in a number of cases, dramatic distortion was noted. For the mesial to distal length of each tooth, the range of distortion was −29% to +5% for a total of 34%. The intercanine range (arch width) was −27% to+ 24% for a total of 51%, and the range of angulation of teeth was −81% to+ 80% for a total of 161% (Figure 1). With regard to changing a body position, again dramatic distortion was seen due to postural changes (Figure 2).

Figure 1. All of these bites were created with the same set of teeth. Outlines of the biting dentition are displayed above and below the bite such that the maxillary arch corresponds with the maxillary bite and the mandibular arch corresponds with the mandibular bite. Note distortion among the bites.

Figure 2. The bite is outlined in black. The set of teeth that created the bite is illustrated inside the bite. Notice the distortion that results when the body part is moved.

Observations

Distortion results mostly due to the biomechanical properties of the skin. Skin stretches, and it stretches unevenly. The result: the alignment of the biting dentition does not transfer to skin in a reliable manner.¹³

Study Example Two

Methods and Materials

In subsequent studies, we employed a statistical analysis called geometric morphometric analysis, which can quantify shape change variation between large datasets.^11,12,14 It has been validated in multiple scientific areas of study, particularly evolutionary biology where it is used to track the change in species shape over time. Here, it was used to evaluate the distortion across multiple bitemarks along with the set of teeth that made them. In these studies, we again made multiple bites all with the same set of teeth, same opening diameter and same pressure. We compared the bites to each other, to the set of teeth that created them and then against a database of dentitions.

Results

The bites did not transfer to the skin in a reliable fashion and the set of teeth that best matched the bitemark was not that of the biting dentition.

Observations

Again, we determined distortion results mostly due to the biomechanical properties of the skin and the alignment of the biting dentition does not transfer to skin in a reliable manner.

Study Example Three

Methods and Materials

Though the pattern of the dentition did not transfer reliably to skin we thought it would be interesting to examine the second premise of bitemark analysis: Is the human dentition unique. For this, we explored the uniqueness of the dentition, as it pertains to bitemark analysis. For this study we measured the alignment pattern of the six anterior maxillary and mandibular teeth of 1100 sets of digital 3D dental models and compared the models against one another to see how many ‘matched’ to another.¹⁰ A match was defined as being indistinguishable at a given measurement resolution. To establish this, we placed 12 landmarks on each arch in 3D space which defined the relative position and alignment of the teeth. Using a repeat measures test (measuring the same specimen multiple times), we determined that the accuracy and repeatability of landmark placement was high, with a resolution of 0.12 mm per landmark. This became our measurement resolution and defined criteria for a ‘match’.

We then applied a distortion factor equating to 25%, 50% and 75% to simulate the distortion we saw in our previous studies with bitemarks on skin (Table 1).

Table 1. The number of matches can exceed the number of specimens (n = 1100) because many of the individuals had more than one match in the population.

Effect of Increased Distortion (decreasing measurement resolution)
	0%		25%		50%		75%
	Matches	Individuals	Matches	Individuals	Matches	Individuals	Matches	Individuals
Maxilla	763	396	9,660	826	39,854	1001	93,091	1,059
Mandible	75	83	1,658	451	9,510	759	29,308	931
Both	2	4	166	144	2,056	502	20,217	773

Results

According to this study, just a 25% distortion includes 75% of the sample population for the maxillary arch and 41% for the mandibular arch. In this dataset, it was noted that there was much more malalignment (and thus fewer matches) in the mandibular arch vs the maxillary arch. It is important to note our bitemark studies were completed on unembalmed cadaver skin which did not bruise. Thus, we were able to examine distortion on actual indentations of the teeth left on skin.

Observations

In an actual bitemark case, a bruise, not indentations of teeth, is what is typically examined. A bruise may create further uncertainty as it is diffuse and may not accurately reflect the tooth arrangement that created the wound.

In all we completed twelve studies.^10–20 The results: Bitemark transfer to skin is not reliable. Significant distortion occurs, and the distortion is not predictable. Many times, a set of teeth other than the biter is a better ‘fit’ to the bite. Within a population of 1100 people, with just 25% distortion, a significant number of the population could have created the bite. We should note and emphasize that our motivation in conducting all of this research was to understand the nature of bitemark evidence, as so few studies existed prior. Each study encouraged us to delve deeper as the results were alarming. By presenting at national meetings and publishing in peer-reviewed journals, we sought to inform the practitioners of the potential issues with the evidence and the possible liabilities of testifying at trial.

Response from the Legal and Forensic Odontology Community

There has been a strong move on the part of the American Board of Forensic Odontology to acknowledge that an individual cannot be identified from a bitemark, however at the time of this writing there is still a push to use this evidence to exclude an individual. But how do you know that you are not excluding a guilty person? Distortion works both ways. There does not appear to be a movement to end bitemark analysis altogether. Indeed, it appears that there is a desire to keep it in the toolbox for prosecutors and for dentists to testify in court as it is felt there is still a place for its use.

Many times, presentations or editorials,²¹ based on opinion, and not scientific, peer-reviewed studies, have lauded the successful outcomes of bitemark analysis. They describe individuals who were found to be guilty based on preponderance of evidence however, and the bitemark analysis may have been only a small part of the overall evidence against the individual. Some feel this is a confirmation that bitemark analysis has value. In reality, the investigators simply arrested the right person in that situation. Ted Bundy would be a perfect example of such a case. Successful use is not a confirmation that bitemarks are reliable. It is more an example of coincidence rather than consistency, reliability or validity.

Those advocating that bitemark analysis is useful and valid also often blame inexperience or incompetency on the part of the analyst for failures. This argument again is an attempt to keep the technique relevant. All too often bitemark cases in a court of law become a “battle of the experts”. These experts are often the “most respected” in the field. For example, in a 1978 case in Illinois, involving Karla Brown²² the “experts”, two for the prosecution and two for the defense were divided on everything from the evidentiary value, the actual recognition of a bitemark, as well as relating it to the suspect. The bitemark in this case wasn’t even recognized as such until two years after the murder when a forensic dentist reviewed crime scene photos. Shockingly, the bitemark analysis was completed after an exhumation, three years after the murder. This would certainly raise questions about the evidentiary value, if any existed at the time. Although there are many criminal cases where the presented evidence is questioned, in bitemarks it appears to be the norm. Despite the controversy the bitemark evidence was critical to the conviction in the Karla Brown murder.

The Harward case, mentioned above, was also another case of divided opinion. Initially two forensic odontologists agreed that a bitemark was present, but eliminated Harward as the biter. An additional expert was found that refuted the original findings. Three other odontologists were located that also implicated Harward as the biter. The original two analysts succumbed and also concluded that Harward was the biter. The fact that controversy existed with respect to the bitemark from the beginning should have made the evidence questionable and not sufficient to place an individual in prison for life.

It should be stressed that the errors in analysis are not the result of incompetence or worse, intentional perjury. The conducted research demonstrates mistakes can easily be made. Simply stated, skin does not capture the details of the dentition reliably and distortion plays a significant role. Scientific studies have revealed that other individuals could have certainly left similar marks or there is a probability that an individual could have left a mark on skin that is not representative of their dentition. Either of these scenarios could lead to a false conviction or the opposite, the exoneration of a guilty person. Neither of these scenarios or outcomes is acceptable. An expert in the Harward case stated after the exoneration that, “This case should persuade all my colleagues to agree with the need for more scientific research and investigation.” Maybe the tide is turning?

Unfortunately, any research that did not support the reliability of bitemarks was heavily criticized by those supporting bitemark analysis, maybe for not reaching the conclusions some had desired- that bitemarks still have merit as an evidentiary tool.

Our research initially received acclaim from the pro-bitemark advocates, but soon our work was admitted in the courtroom for the defense and as a result the pro-bitemark advocates became our adversaries. We were openly criticized, bullied, attacked verbally and on social media, all in an attempt to discredit our work and keep bitemark evidence in the courtroom.²³ This continued until a study far more damning than any of ours was presented, then the attacks shifted in their direction. This study investigated reliability amongst bitemark examiners themselves. It asked the question: Could the experts reliably tell if a wound was a human bitemark? They failed miserably. In this study, there was no reliability on the most basic and important question. Thus, forget the comparison, forget the distortions, bitemark examiners could not even agree if a wound was a human bitemark at all.²⁴

Response from the Scientific Community and the Path Forward

Where the legal system failed to take ownership of nonscientific testimony, science would force the hand and science would prevail.

The first such instance was with the Texas Commission on Forensic Science. We joined with the authors of the aforementioned study²⁴ in testifying at a hearing before the Commission. Advocates for the use of bitemark analysis also testified. After a six-month investigation, it was determined that bitemarks ‘did not meet the standards of forensic science’ and the Commission called for a moratorium on bitemark analysis until further investigation could be completed.²⁵

Next, was the President’s (Obama) Council for Assistance in Science and Technology (PCAST). We testified under strict whistleblower protection hearing rules. Again, advocates for the use of bitemark analysis also testified, and the president’s council concluded: “The prospects of developing bitemark evidence into a scientifically valid method are low. We do not advise devoting significant resources to such efforts.”²⁶

In 2016 a paper was published which we coauthored with thirty-six others. Our fellow authors comprised the most prominent and respected legal scholars in the nation. To collaborate with and be recognized and supported by such a revered cadre of colleagues meant a tremendous amount to us and substantiated our approach.⁹

Further, The National Institute of Standards and Technology (NIST) has also reviewed the scientific foundations of bitemark analysis. In October 2022, NIST published its findings in a draft report, Bitemark Analysis: A NIST Scientific Foundation Review.²⁷ It should be noted that the authors of the draft will consider all comments submitted before publishing a final version of the report. Consistent with what we state in this article, the draft review finds that “forensic bitemark analysis lacks a sufficient scientific foundation because the key premises of the field are not supported by the data. First, human anterior dental patterns have not been shown to be unique at the individual level. Second, those patterns are not accurately transferred to human skin consistently. Third, it has not been shown that defining characteristics of those patterns can be accurately analyzed to exclude or not exclude individuals as the source of a bitemark.” The draft review involved a thorough review of the literature, and the authors stated “they examined every publicly available, English language scientific article available on this topic”. The authors also examined book chapters, conference presentations, professional standards and guidelines, and other material totaling more than four hundred publications. The published draft was open for public comment prior to the publication of this article and before publishing a final version of the report. The release of the final report will coincide with the publication of this article and should be a source of reference and critical analysis regarding the use of bitemarks in the legal system.

Even with all of this, bitemark evidence may still be admitted in courts of law. So, where does this leave us with Bitemark Analysis: Legal vs Scientific Battle for Justice? Together with the work of others in the legal and scientific communities, we hope that our efforts in science will go a long way in preventing wrongful convictions based on bitemark evidence in the cause of true justice.²⁸

Supplementary data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19424396.2023.2191391

Additional information

Notes on contributors

Mary A. Bush

Mary A. Bush, DDS, is an associate professor at SUNY at Buffalo School of Dental Medicine. She is past president of the American Society of Forensic Odontology and is a fellow of the American Academy of Forensic Sciences. She is on the Editorial Board for the Journal of Forensic Science, has published numerous articles, has contributed to various textbooks, and lectures widely on the topic of forensic odontology including an invited presentation at a congressional hearing on Capitol Hill.

Raymond G. Miller

Raymond G. Miller, DDS, is a Clinical Associate Professor of Oral Diagnostic Sciences at the State University of New York School of Dental Medicine. He has published numerous articles, contributed to various textbooks and lectures extensively on forensic odontology. He is a retired Lieutenant Colonel in the New York State Air National Guard 107^th Medical Squadron. Dr Miller is a Fellow in the American Academy of Forensic Sciences (AAFS), and past Section Chair of the Odontology section of the AAFS. He is a member of the American Society of Forensic Odontology and the Disaster Mortuary Operational Response Team, responding to mass fatality incidents nationally. Dr Miller led the dental team for victim identification in the local crash of Flight 3407.

Peter J. Bush

Peter J. Bush, BS, is director of the South Campus Instrument Center at the State University of New York School of Dental Medicine. He is a fellow of the American Academy of Forensic Sciences and is a member of the Research Committee for the American Society of Forensic Odontology. Mr. Bush has worked in many scientific fields including Forensic Odontology. He has published over 70 articles and his work is referenced in numerous sources including the NASA website.