This is part one in a four-part series. The rest of the series will be posted next week.

Before he left the courtroom, Gerard Richardson made his mother a promise. “I told her that one day she’d see me walk out of that building a free man,” he says.

Her response nearly broke him. “She said, ‘Gerard, I’ll be dead by then.’”

Richardson, then 30, had just been convicted for the murder of 19-year-old Monica Reyes, whose half-naked body was found in a roadside ditch in Bernards Township, N.J. The year was 1995, and Richardson had just been sentenced to 30 years in prison.

There were only two pieces of evidence implicating him. One was a statement from Reyes’s boyfriend, who claimed to have heard Richardson threaten to kill her. But that statement was made only after police had shown the boyfriend the second piece of evidence: a finding from a forensic odontologist that a bite mark found on Reyes’s body was a match to Richardson’s teeth. Dr. Ira Titunik, the bite mark expert for the prosecution, would later tell jurors there was “no question in my mind” that Richardson had bitten Reyes.

“I thought it was crazy,” Richardson says. “There was no way it was possible. The FBI looked at hairs, fibers, blood, everything the police found at the crime scene. None of it came from me. Just this bite mark.”

Despite his certainty, Titunik’s analysis consisted of little more than a one-page report identifying Richardson as the culprit. “There were only two things on that sheet. It said there was a bite mark on the victim, and that I had made it.”

In fact, when questioned at trial about his methodology — about why he was able to single out Richardson as the biter — Titunik relied on a more detailed report offered by Richardson’s own expert witness, Norman Sperber, also a bite mark analyst. But Sperber, also going off his own report, told jurors there was no way the bite mark could have been left by Gerard Richardson. Two witnesses who called themselves experts relied on the same report and came to diametrically opposing conclusions.

“I thought that was the very definition of reasonable doubt,” Richardson says. “The only physical evidence against me was Dr. Titunik’s testimony. But my own expert was just as qualified as he was and was saying the very opposite. And they were both using the same report. How could that not be reasonable doubt?”

In 1998, three years after Richardson’s conviction, a handyman named Edmund Burke was arrested for the brutal murder of a 75-year-old woman near her home in Massachusetts. Burke became a suspect after a police dog led officers to the house where Burke lived with his mother, about a quarter-mile from the crime scene. As with Richardson, the only physical evidence against Burke was the testimony of a bite mark expert who claimed to match Burke’s teeth to marks found on the victim. That expert, Dr. Lowell Levine, hadn’t actually examined the body, only photos of it. Nevertheless, he claimed he could match a dental mold of Burke’s teeth to the bite marks in the photos with “a reasonable degree of scientific certainty.” Prosecutors then turned to a second odontologist to verify Levine’s match: Ira Titunik. He did.

Burke was held in jail for 41 days. He was released when DNA testing on saliva recovered from the bite mark excluded him. Five years later, DNA from the saliva around bite mark hit a match in the FBI’s Combined Index DNA System (CODIS). The match was to a man already convicted of murder. Prosecutor William Keating, who took office after Burke’s arrest, told the Chicago Tribune he had “no question” that Burke was innocent.

Gerard Richardson wasn’t quite as lucky as Edmund Burke. Though saliva was also collected from the bite mark left on Monica Reyes, the DNA technology available at the time wasn’t able to produce a profile. It would take nearly two decades for the technology to improve to the point where enough useful DNA could be extracted from the small amount of saliva. When that finally happened, it showed that Richardson wasn’t Reyes’s killer.

Richardson was finally released in December 2013. He says that when his attorneys told him he could be released, he was elated and relieved. But he also didn’t want his vindication to be validated in a stack of papers.

“I wanted it done in a courtroom,” he says. “I wanted to hear it from a judge. I wanted my family and friends to hear a judge declare my innocence.”

Most important, he says, he wanted to keep his promise to his mother. “She saw me walk out of the courtroom. A free man.”


The field of forensics has reached an important moment. In 2009, the National Academy of Sciences published a congressionally commissioned report on the state of forensic science in the courtroom. The report was highly critical of a wide range of forensic specialties, from fingerprints to hair and fiber analysis to blood spatter analysis. It found that many of the claims forensic analysts have been making in courtrooms for decades lacked any scientific foundation to back them up. Yet judges and juries have taken and continue to take those claims as foolproof science, often because the experts themselves frame them that way.

The report was particularly critical of an area of forensics loosely known as pattern matching. That area encompasses a group of largely subjective specialties in which an analyst looks at two pieces of evidence, such carpet fibers, hair fibers or marks made by tools, and simply declares based on his or her experience and expertise whether the two are a match.

Bite mark analysis is also part of this group. But even within the pattern matching disciplines, the NAS report singled out bite mark matching for some especially harsh criticism. The report found “no evidence of an existing scientific basis for identifying an individual to the exclusion of all others.” The problem is that this is precisely what bite mark analysts do — and what they have been doing for decades.

Bite mark matching has been around since the 1970s. Generally speaking, bite mark analysts look at indentations found in human skin thought to be caused by human teeth. They first confirm the marks are actually a human bite. They then compare those marks to plaster molds taken of the teeth of one or more suspects. In some cases, analysts will make a mold of the bite itself. In others, they’ll perform their analysis from a photograph of the bite, sometimes with the aid of software like Adobe Photoshop.

But this is just a generalization of the practice. It’s difficult to lay out the standard procedure of a bite mark analysis because it varies from analyst to analyst. Professional organizations claim to have established best practices and procedures, but there is little to no self-policing within the field. The discipline has been evolving on the fly, not usually in response to scientific research but ahead of it, often as a defensive move after the exonerations of people convicted by the testimony of its practitioners.

Bite mark analysis is most commonly used in criminal rape, murder and child abuse cases. It also sometimes comes into play in child custody disputes. The field has always had its critics. As with other flawed forensic specialties, those critics found some vindication in the 1990s when DNA testing started uncovering wrongful convictions won primarily on bite mark testimony. According to the Innocence Project, 24 people — including Gerard Richardson — have been exonerated after they were either convicted or arrested because of the assertions of a bite mark analyst. (A 25th, Douglas Prade, was initially cleared in Ohio, but that decision was overturned by an appeals court last March.) Chris Fabricant, a director of special litigation for the Innocence Project who specializes in bite mark evidence, estimates that there are still hundreds of people in prison today due to bite mark testimony, including at least 15 awaiting execution.

The NAS report generated a lot of media coverage. It spawned congressional hearings, think pieces, and legal conferences. It also spurred the White House to turn to the classic Washington response to a pressing problem: the blue ribbon panel. The National Commission on Forensic Science has been given the mission to “enhance the practice and improve the reliability of forensic science” and to “promote scientific validity” in the courtroom. The NAS report also elicited promises from large forensic professional organizations to establish best practices, take ethics more seriously and do a better job of policing their members.

There is no better example of the pitfalls of allowing junk science into the criminal justice system than bite mark analysis: The field has helped convict a disturbingly high number of people later proven to be innocent. The National Academy of Sciences found it to be lacking of any basis in science. And its members have a poor track record of policing themselves. Yet this particular forensic discipline is not only still going strong; it may be as strong as ever.

The community of bite mark analysts wields considerable influence. The field’s foremost advocacy group and certifying organization, the American Board of Forensic Odontology (ABFO), is aggressive, dogged and holds a lot of sway within the American Academy of Forensic Sciences (AAFS), one of America’s largest professional forensics organizations. And while bite mark matching and the ABFO may be looked upon with some suspicion by other forensic specialists (and certainly by the scientific community), the AAFS is generally seen as an innocuous, credible umbrella organization. As one advocate for forensic reform put it in an interview for this series, “The ABFO needs AAFS for legitimacy. The AAFS doesn’t really need the ABFO.”

(Not all forensic dentistry is this controversial. The ABFO also represents practitioners who use dental records to identify human remains, a field that is well established and well regarded by both the scientific and forensic communities.)

The AAFS and its leadership have been active in the forensics reform movement that has arisen in the wake of the NAS report. The problem is that the AAFS has also been aiding bite mark analysts and their supporters, lending some of its legitimacy to the cause. That’s in part because the ABFO has been persistent and successful in promoting its own membership for leadership positions within AAFS. Four bite mark analysts have served as president of the AAFS, most recently in 2012. And so even as the influential AAFS and its leadership pay lip service to forensics reform and the problems addressed in the NAS report, the group is actively working to push bite mark analysis, for which — again — the NAS report reserved some of its harshest criticism. This narrative was confirmed by several prominent advocates for forensics reform, though it’s worth noting that most of these outspoken advocates were unwilling to criticize the AAFS for attribution. The organization is still that influential.

Much of the NAS report’s bite mark section was based on the research of Michael Bowers, a 65-year-old dentist, college professor and deputy medical examiner in Ventura County, Calif. Bowers has seen lots of cases like Richardson’s. He has personally assisted in seven exonerations of people convicted because of bite mark evidence. For about a quarter-century, Bowers has been basically trying to eradicate bite mark matching from the courtroom.

“I’ve watched over and over as these people take the witness stand and give testimony that isn’t just false and misleading, but that has put innocent people in prison,” Bowers says. “It’s such a corruption of justice. But for a long time people just didn’t want to hear about it.”

Bowers fought his battle from within as a member of the ABFO, where he once served on committees and held leadership positions. When that didn’t work, he resigned from the organization and began criticizing it from the outside. That’s when he became a target.

Two months after Gerard Richardson was released from prison, Peter Loomis, then president of the ABFO, filed an ethics complaint with the AAFS. It was the first time the president of the ABFO had ever filed such a complaint. But Loomis’s complaint wasn’t against Ira Titunik, the man whose testimony sealed Richardson’s conviction. Nor was it against any of the bite mark analysts who have contributed to other false arrests or convictions over the years.

Instead, Loomis’s complaint was against Michael Bowers. If successful, the complaint could get Bowers expelled from the AAFS and effectively destroy his credibility as an expert witness. It would remove an important critic from the courtroom. (I’ll have more on the specifics of the complaint in Part 3.) UPDATE (Monday): At its annual meeting, the AAFS’s board dismissed the complaint against Bowers.

The 2009 NAS report also expressed the need for more scientific research on the underlying assumptions of bite mark analysis — that human dentition is unique and that human skin is capable of registering bite marks in a way that makes them identifiable and distinguishable. Just as the NAS report was being published, a research team at the University of Buffalo led by the husband-and-wife team of Peter and Mary Bush was preparing to publish the results of a series of studies they had designed to probe exactly those questions. The results so far have been damning for the field of bite mark analysis. Bite mark analysts and their supporters have since subjected the Bushes to vicious, sometimes highly personal attacks.

This past fall the Justice Department announced the members of the National Forensic Science Commission subcommittee that will study the scientific validity of bite mark matching. This is part of the reform process put in motion by the NAS report. Incredibly, a majority of the members of the subcommittee are people who either practice or have openly defended the very sort of bite mark matching that the NAS report criticized. Robert Barsley, the chairman of the committee, is not only a practicing bite mark analyst, but his testimony also helped put an innocent man in prison for 17 years. 

Meanwhile, every time someone has challenged the science of bite mark matching in court since 2009, the court has ruled the other way. In short, the scientific community has declared that bite mark matching isn’t reliable and has no scientific foundation for its underlying premises, and that until and unless further testing indicates otherwise, it shouldn’t be used in the courtroom. And so far, the criminal justice system has said it doesn’t care. If bite mark matching is a bellwether issue for meaningful forensics reform, the prospects for meaningful reform appear to be dim.