This is the last part of a four-part series. The first three parts can be found here, here and here.

The 2009 National Academy of Sciences report that was highly critical of the way forensics is used in the courtroom was entitled “A Path Forward.” The words expressed the hope of the report’s authors that it would serve as a catalyst to spur scientific testing of forensic specialties, more vigorous policing of what expert witnesses say on the stand and the development of uniform standards and procedures, all pointing toward an ultimate goal of preventing more wrongful convictions caused by unsupported expert testimony.

Reform, of course, is a long process, but in the field of bite mark matching — which again was the forensics specialty the NAS report singled out for some of its harshest criticism — the “path forward” looks to be obstructed. That’s probably because with bite mark matching, the debate isn’t just about adopting better standards or practices, but also about whether the field should exist at all.

“Most people in forensic odontology are practicing dentists, or academics. They don’t make their living doing bite mark analysis,” says Michael Saks, an Arizona State University law professor who studies the role of science in criminal law. “They do it on the side. Many of these cases involve sex crimes and crimes against children. So they see themselves as avenging angels. They’re protecting the weak. They’re putting away the bad guys. Then along come critics like Michael Bowers or the Bushes, calling their good work into question. You can see why they’d be angry, even though Bowers and the Bushes are right.”

Perhaps that’s why courts and prosecutors have been so reluctant to acknowledge the field’s shortcomings as well. Since the NAS report was released, there have been several court challenges to the validity of bite mark evidence. So far, every challenge has been struck down.

In 2011, for example, a Pennsylvania judge upheld the 1994 conviction of John Kunco, who had been convicted of rape due in part to the testimony of bite mark analysts Michael N. Sobel and Thomas J. David. (David is a previous president of the American Board of Forensic Odontology.) In his closing argument, the prosecutor emphasized the importance of the testimony:

[T]here’s no way, no way on this earth, for Mr. Kunco to explain how his tooth marks got on Donna Seaman’s shoulder unless you accept the fact that he’s the one who attacked and brutalized Mrs. Seaman. That’s the only explanation, ladies and gentlemen. That’s why the evidence is better than fingerprints or hair samples … [T]he bite mark on Donna Seaman’s shoulder was as good as a fingerprint. And I submit to you it was that, ladies and gentlemen, for all intents and purposes. Ladies and gentlemen, I’d submit to you that John Kunco should have just signed his name on Donna Seaman’s back, because the bite mark on Donna Seaman’s shoulder belongs to John Kunco.

The alleged bite marks on the victim’s shoulder weren’t actually examined by Sobel and David until five months after the rape, a length of time long enough for most wounds to heal. In a 1994 article for the Journal of Forensic Sciences, Sobel and David explained they were able to “recapture” a bite after so much time had passed. They wrote that they employed a technique using ultraviolet light to find, isolate and photograph the mark. They then used the photograph to match the marks to Kunco. The article included a footnote to cite the bite mark analyst who had pioneered the technique. That bite mark analyst: the discredited Michael West.

(Thomas David is also quoted at length in Melissa Mourges’s brief in the Dean case, discussed below.)

To win a new trial after conviction, an inmate must show that he or she has discovered new evidence, that the new evidence was not discoverable at the time of trial and that if the evidence had been available, the jury would probably have acquitted. The inmate must also file his or her petition within a year of when the new evidence was discovered or should have been discovered. Kunco’s petition hinged on the NAS report and its findings on bite mark evidence. In denying Kunco’s petition for a new trial, Judge Rita Donovan Hathaway acknowledged that there are problems with bite mark analysis, but she found that the NAS report wasn’t new evidence. Rather, it was based on older research for which Kunco had already missed his deadline to file.

Hathaway’s ruling may have been correct on the law, but it underscores just how difficult it can be to get a conviction based on bad science overturned. Many, many defendants in fact had challenged bite mark evidence based on the prior research and criticisms Hathaway ruled that Kunco should have discovered earlier. They, too, were denied. At this point, even the ABFO may disclaim Michael West. But his legacy in bite mark analysis continues to keep people in prison.

In the 2012 case Coronado v. Texas, a state appeals court upheld bite mark evidence on the grounds that forensic odontology is a “soft science” and thus does not need to be subject to a more rigorous analysis. Under a 1998 state appeals court decision, “soft” sciences are admissible if they come from a credible field and if the expert is practicing the principles of that field. In determining that bite mark analysis is a credible field, the court cites the 1990 state appeals court decision upholding the conviction of David Wayne Spence. As noted previously in this series, Spence was convicted primarily due to bite testimony from Homer Campbell, a forensic odontologist who had participated in another wrongful conviction and had given preposterous probability statistics to the Arizona Supreme Court. There were also significant doubts about Spence’s conviction. He was executed in 1998. In upholding that conviction, the Texas appeals court upheld bite mark evidence, in part because “our research has not yet led us to a reported case where bite mark evidence has been ruled not to be admissible evidence.”

Here again a court upheld bite mark evidence in large part because it has always done so in the past. And it has always done so in the past because other courts had done so before that. As previously noted in this series, many of those precedent-setting cases were supposed to be limited in scope, were misinterpreted by later courts or actually involved suspects who were later exonerated. These opinions aren’t scientific analysis so much as a jurisprudential version of the childhood game of Telephone.

The California Supreme Court then took things to new heights of absurdity in the case of William Richards. In 1997, Richards was convicted of killing his wife, Pamela, due in large part to testimony from bite mark analyst Norman Sperber. He had looked at an autopsy photo of Pamela Richards’s body and found a mark he thought was a bite. Sperber testified that a gap in the alleged bite was a match to William Richards’s unusual dentition. More than a decade later, Sperber recanted his testimony, calling the gap a flaw in the photo. He added that he no longer even thought the bite was made by a human. Four other forensic odontologists said that the photo did not offer enough detail to provide a match to William Richards.

This was still not enough for the California Supreme Court to overturn Richards’s conviction. In what the publication California Lawyer called “the worst opinion of the year,” the majority ruled in 2012 that once he was in post-conviction, Richards had to prove that the evidence against him was false, not merely overstated. The bite mark analysts who advocated for him after his conviction could not rule him out as the source of the bite (if it was a bite at all); they could only say that the photo from which Sperber originally drew his conclusions was too blurry to prove that Richards was a match.

In other words, a bite mark analyst making grand claims from a blurry photo was good enough to convict Richards, but other analysts — including that same analyst who helped convict him — stating after the fact that the photo was inconclusive was not enough to free him. (At this week’s American Academy of Forensic Sciences conference in Orlando, Sperber, who kept a man in prison for more than a decade before changing his mind, received three lifetime achievement awards, one from the AAFS, one from the ABFO and one from the American Society of Forensic Odontology. The latter is a new award called the Norman D. Sperber Award for Forensic Dental Excellence.)

The most significant challenge to bite mark evidence since the NAS report was released came in State v. Dean, the New York case mentioned in part three of this series. In 2013, attorneys for defendant Clarence Dean challenged the prosecution’s plan to use bite mark evidence against their client. Manhattan state Supreme Court Judge Maxwell Wiley granted a hearing to assess the validity of bite mark matching. It was the first such hearing since the NAS report was published, and both sides of the bite mark debate watched closely. Mary Bush testified for the defense, as did Karen Kafadar, chair of the statistics department at the University of Virginia and a member of the National Institute of Standards and Technology’s Forensic Science Standards Board.

The prosecutor in that case was Manhattan assistant district attorney Melissa Mourges, an aggressive 30-year prosecutor with a high profile. Mourges was featured in a 2011 HBO documentary and holds the title of chief of the District Attorney’s Forensic Science/Cold Case Unit in what is arguably the most influential DA’s office in the country. So her advocacy for bite mark matching is significant.

As reported in part three, Mourges has not only defended bite mark evidence but also seems to be on a campaign to denigrate its critics, going so far as to heckle scientific researchers Mary and Peter Bush at a panel, and then to personally attack Mary Bush during a dinner talk at a forensics conference. Her bite mark brief in the Dean case compared bite mark evidence critic Michael Bowers to the notorious bite mark charlatan Michael West. It was a particularly egregious comparison because Bowers had helped expose West back when he was still embraced by the ABFO.

In her brief, Mourges first encouraged Wiley to embrace the “soft science” approach to bite mark analysis used by the Texas court in Coronado. Conveniently, doing so would allow bite mark specialists to testify to jurors as experts with almost no scrutiny of their claims at all.

Mourges next argued that if the court must do an analysis of the validity of bite mark testimony, it do so on the narrowest grounds possible. When it comes to assessing the validity of scientific evidence, New York still goes by the older Frye standard, which states that evidence must be “generally accepted” by the relevant scientific community. The question then becomes: What is the relevant scientific community?

In her brief, Mourges urged Wiley to limit that community to analysts who “have actually done real-world cases.” In other words, when assessing whether bite mark matching is generally accepted within the scientific community, Mourges says the only relevant “community” is other bite mark analysts.

Saks offers a metaphor to illustrate what Mourges is asking. “Imagine if the court were trying to assess the scientific validity of astrology. She’s saying that in doing so, the court should only consult with other astrologers,” he says. ”She’s saying the court shouldn’t consult with astronomers or cosmologists or astrophysicists. Only astrologers. It’s preposterous.”

Saks, who submitted a brief in the case on behalf of Dean, also offers a real-world example: the now-discredited forensic field of voiceprint identification. The FBI had used voiceprinting in criminal cases in the 1970s but discontinued the practice after an NAS report found no scientific support for the idea that an expert could definitively match a recording of a human voice to the person who said it.

“If you look at the Frye hearings on voiceprint identification, when judges limited the relevant scientific community to other voiceprint analysts, they upheld the testimony every time,” Saks said. “When they defined the relevant scientific community more broadly, they rejected it every time. It really is all about how you define it.”

In urging Wiley to only consider other bite mark analysts, Mourges also casts aspersions on the scientists, academics and legal advocates urging forensics reform. She writes:

The make-up of the relevant scientific community is and should be those who have the knowledge, training and experience in bitemark analysis and who have actually done real world cases. We enter a looking-glass world when the defense urges that the Court ignore the opinions of working men and women who make up the ranks of board-certified forensic odontologists, who respond to emergency rooms and morgues, who retrieve, preserve, analyze and compare evidence, who make the reports and who stand by their reasoned opinions under oath. The defense would instead have this Court rely on the opinions of statisticians, law professors and other academics who do not and could not do the work in question.

Of course, one needn’t practice astrology or palm reading to know that they aren’t grounded in science. And if police and prosecutors were to consult with either in a case, we wouldn’t dismiss critics of either practice by pointing out that the critics themselves have never read a palm or charted a horoscope.

Mourges also attempts to both discredit the NAS report and claim that it isn’t actually all that critical of bite mark analysis. For example, she laments that the report was written by scientists and academics, not bite mark analysts themselves. This, again, was entirely the point. The purpose of the NAS report was to research the scientific validity of entire fields. If it were written by active practitioners within those fields, every field of forensics would have been deemed valid, authoritative and scientifically sound.

Mourges also misstates and mischaracterizes what the report actually says. She writes in one part of her brief that “the NAS report does not state that forensic odontology as a field should be discredited.” That’s true. But bite mark matching is only one part of forensic odontology. The other part, the use of dental records to identify human remains, is widely accepted. What the report makes abundantly clear is that there is zero scientific research to support bite mark analysis in the manner it is widely practiced and used in courtrooms.

In another portion of the brief, Mourges selectively quotes part of the the report, cutting out some critical language. She writes:

When Dr. Kafadar and her NAS committee created the NAS report, they wrote a summary assessment of forensic odontology. In it they said that “the majority of forensic odontologists are satisfied that bite marks can demonstrate sufficient detail or positive identification …

That ellipsis is important, as is the word that comes before the quote. Here’s the passage quoted in full:

Although the majority of forensic odontologists are satisfied that bite marks can demonstrate sufficient detail for positive identification, no scientific studies support this assessment, and no large population studies have been conducted. In numerous instances, experts diverge widely in their evaluations of the same bite mark evidence, which has led to questioning of the value and scientific objectivity of such evidence.

Bite mark testimony has been criticized basically on the same grounds as testimony by questioned document examiners and microscopic hair examiners. The committee received no evidence of an existing scientific basis for identifying an individual to the exclusion of all others.

The report only acknowledges the near consensus within the community of bite mark analysts for the purpose of criticizing them. Mourges’s selective quotation implies that the report says the relevant scientific community accepts bite mark matching. The full passage reveals that the report is essentially pointing out just the opposite: The insular community of bite mark analysts may believe in what they do, but the larger scientific community is far more skeptical.

One common tactic that shows up in Mourges’s brief and has also shown up in defenses of bite mark analysis across multiple forums — court opinions, forensic odontology journals and public debates — is a sort of meticulous recounting of the care and precision into which bite mark analysts collect and preserve evidence as well as the scientific-sounding nomenclature used by the field’s practitioners. Mourges devotes more than 10 pages to laying out the procedures, methods and jargon of bite mark matching.

In any field of forensics it’s of course important that evidence be carefully handled, properly preserved and guarded against contamination. But to go back to the astrology metaphor, even the most careful, conscientious, detail-oriented astrologer . . . is still practicing astrology. If the field of bite mark analysis cannot guarantee reliable and predictable conclusions from multiple practitioners looking at the same piece of evidence, if it cannot produce a margin for error, if its central premises cannot be proved with testing, then it doesn’t matter how pristine the bite mark specimens are when they’re analyzed or what the mean number of syllables may be in each word of a bite mark analyst’s report.

But ultimately, Mourges was effective. In September 2013, Wiley rejected the defense challenge to bite mark evidence in the Dean case. He never provided a written explanation for his ruling. In an e-mail, Joan Vollero, director of communications for the Manhattan District Attorney’s Office, wrote of the ruling: “Following the months-long Frye hearing, Judge Wiley denied the defendant’s motion to preclude the bite mark evidence, finding that the field of bite mark analysis and comparison comports with New York State law.”

Moving on, but without looking back

Generally speaking, since the NAS report came out, the courts have treated other pattern-matching disciplines in the same way they’ve treated bite mark matching — they haven’t really factored in the NAS report at all. There have been some exceptions, but by and large even with the exceptions, the courts have merely limited the degree to which an analyst can declare a “match.” That is, a court may rule that an expert witness can say a bite mark or hair fiber is consistent with the defendant, but they can’t say it could only have come from the defendant.

In some cases, the courts (and defenders of the forensic disciplines under fire) have noted that the NAS report itself doesn’t explicitly call on the courts to invalidate any field of forensics. That’s true, but that wasn’t the report’s intent. The intent was to draw attention to the lack of scientific research to support what many forensic analysts have been claiming in court — its purpose was to review the science, not to change existing law. The fact that the NAS didn’t explicitly tell the courts to invalidate fields such as bite mark analysis doesn’t mean that the NAS report was validating them. Nor does it mean that judges weren’t to take the report into consideration when conducting analyses on admissibility.

But Mourges made that very argument in her Dean brief. Other prosecutors have made it as well. To support it, they often cite congressional testimony given by Harry T. Edwards, a federal judge with the U.S. Court of Appeals for the D.C. Circuit and the chairman of the NAS committee that wrote the report. In a speech at a Yale conference on technology and law, Edwards thoroughly refuted this argument.

I recently had an opportunity to read several briefs filed by various U.S. Attorneys’ offices in which my name has been invoked in support of the Government’s assertion that the Committee’s findings should not be taken into account in judicial assessments of the admissibility of certain forensic evidence . . .

This is a blatant misstatement of the truth. I have never said that the Committee’s Report is “not intended to affect the admissibility of forensic evidence . . . To the degree that I have commented on the effect of the Report on admissibility determinations, I have said something quite close to the opposite of what these briefs assert.

What Edwards did say was that judges will continue to follow the law — that they’ll continue to use the Daubert and Frye analyses. His point was that he hoped the NAS report would inform those analyses.

[T]here is a critical difference between saying that judges will continue to apply existing legal standards . . . and saying that the Report should have no effect on how judges apply those standards. I most certainly never said, or even suggested, that judges should not take into account the new information provided by the Report in assessing the validity and reliability of forensic evidence while making admissibility determinations. Claims to the contrary are without basis in fact and utterly absurd.

That speech was in 2010. Mourges filed her brief in Dean in 2013.

There are at least a few hopeful signs that some policymakers are taking notice of the effects of bad forensics. After the California Supreme Court ruling in the Richards case, the state’s legislature passed a law that makes it easier for inmates to challenge convictions based on bad science. William Richards is now mounting another challenge to his conviction under the new law.

Texas also recently passed a “junk science” law, mostly in response to the faux-arson science used to convict and execute Cameron Todd Willingham. And a federal judge in Wisconsin recently issued a well-informed opinion striking down a conviction based on handwriting analysis.

But these instances have been few and far between, especially in the courts. Michael Saks says it all goes back to asking judges to be the gatekeepers of science. He suggests a sort of national forensics panel that would evaluate new and existing forensic specialties and decide which have sufficient scientific support to be allowed in the courtroom. “We need to move outside the courts,” he says. “Look at these forensic areas that even the government now admits have been discredited. Bullet lead composition, voice print analysis, and so on. The courts had been letting this stuff in for years. It took declarations from the scientific community to put at an end to it. What does that tell us? It tells us that these decision shouldn’t be made by judges.” Edwards seems to agree. In a speech last year, he cautioned that “Judicial review, by itself, will not cure the infirmities of the forensic community.”

While the courts have been slow to embrace the NAS conclusions, there are some indications that the ABFO is at least aware of the heightened public scrutiny. The organization now discourages members from using terms such as “scientific certainty,” “the only person in the world” or Michael West’s trademark phrase “indeed, and without a doubt.” Last year, the ABFO issued a “decision tree,” essentially a flow chart for bite mark analysts, and encouraged its members to use phrases such as “included,” “excluded,” “not excluded” and “the probable biter.”

The problem is that the flow chart still provides no objective criteria for making those assessments. It’s still an entirely subjective process. It’s still an “eyeball test.”

Moreover, according to Brandon Garrett, a University of Virginia law professor who studies innocence cases, it’s far from clear that such changes in language have much of an effect on jurors. “What we’ve found in jury studies is the precise phrasing an expert witness uses doesn’t really matter,” Garrett says. “Whether they say something careful like ‘this is consistent with the suspect’ or  something more definitive like ‘this is a scientific match,’ all the jurors hear is an expert witness saying ‘this guy did it.’ ”

In our interview, ABFO President Peter Loomis also said that the ABFO no longer recommends that analysts claim they can match a biter to the exclusion of everyone else on the planet. Instead, it recommends only making positive identifications in what they call “closed populations” — that is, the police or prosecutor give the analyst a list of suspects, and the analyst then determines who is the “probable biter.”

“It’s a meaningless distinction,” says Chris Fabricant, director of strategic litigation for the Innocence Project.  “It all depends on who is defining the closed population. The Kennedy Brewer and Levon Brooks cases were both closed populations. And they still identified and convicted the wrong men.”

What troubles critics such as Fabricant most is that all this talk about moving forward cavalierly glosses over what has already happened. Even if you believe the current promises from the forensics communities that things are better now, if you don’t change the structural failures that allowed bad science to convict innocent people in the first place, it’s almost certain to happen again.

But there’s also plenty of reason to question those assurances that things are better now. “The ABFO just dismisses these innocence cases as rogue examiners, or artifacts from a bygone era,” Fabricant says. “But they did immeasurable damage, not just to human lives, but to our jurisprudence. Where is the accounting for that? Where is the acknowledgment? Where is the reckoning?”

With the exception of West, the ABFO has never suspended or disciplined one of its members, even when their analysis contributed to a wrongful arrest or conviction. Several who have participated in such injustices are today outspoken advocates or hold leadership positions within the organization. For example, in 1998, bite mark matching by Franklin Wright helped convict Ohio police officer Douglass Prade of killing his wife. But in 2010, DNA testing on saliva taken from the bite mark excluded Prade. An Ohio judge gave Prade a new trial and released him before an appeals court overruled her and ordered Prade back to prison. Today, Wright serves on several ABFO committees, including the ethics, bite mark evidence and proficiency testing committees.

And the larger forensics community isn’t exactly showing bite mark analysts the door. The absurd AAFS ethics hearing on Michael Bowers is a pretty good indication of that. (Note: After this series began on Friday, the AAFS board of directors voted to dismiss the ethics complaint against Bowers, overriding the recommendation of the organization’s ethics committee. Bowers says the cost of his legal defense topped $100,000.)

The theme of this year’s AAFS conference is “Celebrating the Forensic Science Family,” which feels like a plea for unity in a field under criticism. The event features at least eight panels focusing on bite mark evidence, plus the annual “Bitemark Breakfast,” with remarks by ABFO Vice President Adam Freeman and Jeffrey Ashton, prosecutor in the Casey Anthony case. (In conjunction with the conference, for $700 the ABFO is also offering a one-day course in bite mark analysis. Completion of the course will get you one credit in bite mark analysis toward qualification to take the group’s certification exam.)

Of the 20-plus speakers panels specifically related to bite mark analysis, all but three include practitioners in or proponents of the field. One session in defense of bite mark matching will feature Melissa Mourges and Franklin Wright. Neither Mary nor Peter Bush will be speaking, nor will Michael Bowers.

More troubling is that the federal reform apparatus put in place in the wake of the NAS report may have already been captured by the bite mark practitioners. Last October, the National Institute of Standards and Technology (NIST) announced the members of the subcommittee working group that would be studying the scientific validity of forensic odontology. Ten of the 16 members are either practicing bite mark analysts or people who have openly advocated the practice.

The committee includes ABFO president Peter Loomis, ABFO vice president Adam Freeman and ABFO general counsel Haskell Pitluck. It also includes Franklin Wright and David Senn, who testified for the prosecution in the Dean case. The chair of the committee is Robert Barsley, the former ABFO president, former AAFS president and bite mark analyst who helped put an innocent man in prison in the 1990s. The committee does also include a few bite mark skeptics, including Mary Bush. But they’re vastly outnumbered. In fact, the committee includes all five authors of the vitriolic letter to the editor of the Journal of Forensic Sciences that castigated Bush.

After the announcement of the subcommittee, Loomis sent out a celebratory e-mail to the ABFO membership. “It is quite an honor for the ABFO to be so well represented in the Odontology Subcommittee,” Loomis wrote. “Nine (9) of its sixteen (16) members are ABFO Diplomates with Dr. Robert Barsley as its chairman. Even the ABFO legal advisor, the Honorable Judge Haskell Pitluck was appointed as a member . . . Congratulations to all of you!”

If the subcommittee collectively approves of bite mark matching, then the ABFO and its supporters will be able to say that the field survived scrutiny even by the government committee put together to address forensics in the criminal justice system. What was supposed to be a process to rid the courts of dubious expert testimony will have become an official government imprimatur for that very sort of testimony. If it was already difficult to get judges to understand the limits of bite mark matching, de facto approval from a subcommittee put together by a government agency as reputable as NIST will make it nearly impossible. That the subcommittee was stacked from the start probably won’t matter.

The move toward forensics reform was spurred by the revelations unveiled by DNA testing. The hard science of DNA analysis, which was born of the scientific method and extensively peer-reviewed, has shown time and time again that practitioners of the “soft sciences” of forensics were wrong, and have probably been wrong for decades. But DNA testing is only applicable in a small percentage of criminal cases, and the flaws in forensic analysis likely produce unjust outcomes just as often in non-DNA cases as they do in DNA cases. DNA testing was a wake-up call that the system is in need of repair. There probably won’t be another one. So it’s important that we learn the correct lessons, and that reform is done right.

Unfortunately, if bite mark matching is indicative of the larger reform process, the ultimate result of the wake-up call may end up being fairer, more just DNA cases — only because of DNA testing — but business as usual everywhere else. That not only calls into question the fairness and integrity of the criminal justice system, but also brings up the far more fundamental question of whether the system even aspires to be fair.

“We can’t let go of the past, because the past is still the present,” says Fabricant. “You still have people in prison because of bite mark analysis. Some are on death row. There has been no accounting for the damage done. It sounds nice to talk about the path forward. But it would be foolish to embark on a long journey forward without a rearview mirror.”


Addendum: After the third installment in this series, the office of Manhattan District Attorney Cyrus Vance sent the following statement:

Melissa Mourges is a veteran prosecutor and a nationally recognized leader in her field. As Chief of the Manhattan District Attorney’s Forensic Science/Cold Case Unit, she has solved dozens of cold case homicides, including two recently attributed to “Dating Game” serial killer Rodney Alcala. In addition to being a Fellow at the American Academy of Forensic Sciences, ADA Mourges has also served as co-chief of the DNA Cold Case Project, which uses DNA technology to investigate and prosecute unsolved sexual assaults. As part of that work, she pioneered the use of John Doe indictments to stop the clock on statutes of limitation and bring decades-old sexual assaults to trial. Her work and reputation are impeccable, and her record speaks for itself.