Peter Neufeld is a co-founder of the Innocence Project. Keith Findley is a professor at the University of Wisconsin Law School. Dean Strang is a criminal defense lawyer and law professor at Loyola University Chicago. Findley and Strang are also co-founders of the Center for Integrity in Forensic Sciences.

A new study from the University of Washington concludes that, over nearly 40 years, medical examiners and coroners undercounted killings by U.S. police by more than half. During that time, these officials missed or covered up more than 17,000 police killings between 1980 and 2018.

The study — published in the Lancet, a respected medical journal — raises important questions about the objectivity and reliability of physicians who work as government medical examiners and coroners. The findings add to an accumulating list of grave concerns about racial bias in forensic pathology and policing, adding evidence that police kill Black people at a rate 3.5 times greater than White people, and that medical officials underreported Black deaths at a higher rate than White deaths.

Although many forensic pathologists carefully follow scientific principles, the frequency of police killings overlooked or incorrectly diagnosed means the problem is systemic. Yet, because most medical examiners are specialized doctors (although some coroners are not), Americans may assume that they are the gold standard in forensic evidence.

One systemic feature that appears to produce error, however, is that medical examiners long have been close allies of police and prosecutors — frequently partisans, not neutrals. Many allow the police, but not others, to observe autopsies and to influence critical steps in death investigation. They often talk freely to prosecutors, but only grudgingly — if at all — to defense lawyers. In a 2011 survey, 22 percent of medical examiners and coroners reported pressure from government officials to change the cause or manner of death on a certificate.

Medical examiners claim, in essence, the dubious right to diagnose (or rule out) not a medical condition, but a crime. Worse, many react fiercely to questions about their objectivity, biases or ethics. They deny — irrationally and unscientifically — that they are subject to universal human cognitive biases at all.

Like other physicians, medical examiners often rely appropriately on patient medical history to assist in diagnosis. But they also routinely use nonmedical background information that can produce serious contextual bias. They often accept police suspicions, prior criminal records, police theories of criminal liability, tropes about drug or alcohol use, police concerns about their own culpability, and race in assessing cause and manner of death. The absence of rigorous science leads to false negatives — such as when the pathologist ignores the role of excessive police restraints — and to false positives, such as charging a civilian with homicide when the death was accidental (or natural).

Another recent study in the Journal of Forensic Sciences found that, in actual case work, pathologists who examined 10 years of children’s death certificates in Nevada were much more likely to declare a child’s death a homicide rather than an accident when the child was Black. Corroborating these findings, the researchers presented pathologists with a hypothetical postmortem examination of a young child. The pathologists were many times more likely to find the child’s death to be a homicide than an accident, based on identical medical evidence, if told that the child was Black and the caretaker was the mother’s boyfriend than when told that the child was White and the caretaker was the child’s grandmother.

Responding to the study, vocal forensic pathologists sought to bury the research and the researchers. They first demanded that the journal retract the peer-reviewed article, then asked the journal’s publisher to censor it, and finally filed ethics complaints against the forensic pathologists and neuroscientist who co-authored the article. The efforts failed, but the effect was chilling.

This reaction was telling. In a letter lashing out against the study, 74 pathologists wrote, “Manner determination [e.g., homicide, suicide, accident, natural or undetermined] is not a ‘scientific’ determination. It is a cultural determination that places a death in a social context for the purpose of public health statistics.” They elaborated that “there is no ‘right’ answer in many manner determinations” and “the goal is consistency rather than some nonexistent criteria for correctness.”

Such subjective determinations are especially vulnerable to cognitive and, in particular, racial biases. The shocking likelihood that medical examiners and coroners have missed (whether deliberately or through unexamined biases) more than half of police killings over almost four decades is a sharp reminder of these tendencies and their dangers.

Medical examiners and other doctors have an important role to play in making and explaining objective physical findings that laypeople would miss or misunderstand. When constrained by scientific rigor and objectivity, their courtroom contributions are valuable. But as a group, their eagerness to embrace contextual bias that distorts reliability and their willingness to act as partisans aiding the prosecution, raise real concerns. We now can point to about 17,000 more people who would second these concerns, if they could.

This study is posted on Moodle under "Class 8" heading.

This document is a pdf and is posted on Moodle under "Class 8."

Please go to our Moodle page and under "Class 8" you will find the prompt and submission folder for Writing Reflection #8.

This is an excerpt from an excellent four part series on bitemark evidence. You can access the full story at this link.

Excerpt from Part 3 of 4:

In 2007 Mary and Peter Bush, a married couple who head up a team of researchers at the State University of New York at Buffalo, began a project to do what no one had done in the three decades — conduct tests to see whether there’s any scientific validity to the bite mark evidence presented in courts across the United States.

The Bushes sought to test the two main underlying premises of bite mark matching — that human dentition is unique and that human skin can record and preserve bite marks in a way that allows for analysts to reliably match the marks to a suspect’s teeth. The Bush team was the first to apply sophisticated statistical modeling to both questions. It was also the first to perform such tests using dental molds with human cadavers. Previous tests had used animal skins.

When they first set out on the project, the Bushes received preliminary support from some people in the bite mark analyst community. “Franklin Wright was the ABFO president at the time,” says Mary Bush. “He visited our lab, and then put up a message praising our work on the ABFO website.” They also received a small grant from the ASFO, the discipline’s non-accrediting advocacy and research organization.

“There was a lot enthusiasm at the outset,” says Fabricant. “I think some analysts were excited about the possibility of getting some scientific validation for their field.”

But when the Bushes began to come back with results that called the entire discipline into question, that support quickly dried up.

The Bushes’ research found no scientific basis for the premise that human dentition is unique. They also found no support for the premise that human skin is capable of recording and preserving bite marks in a useful way. The evidence all pointed to what critics such as Bowers had always suspected: Bite mark matching is entirely subjective. The Bushes’ first article appeared in the January 2009 issue of the Journal of Forensic Sciences. The couple have since published a dozen more, all in peer-reviewed journals.

Outside of ABFO and their supporters, the Bushes’ research has been lauded. “I think there’s a chance that because of the Bushes’ research, five years from now we aren’t going to be talking about bite mark evidence anymore,” says Risinger. “It’s that good. Their data is solid. Their methodology is solid. And it’s conclusive.”

Other legal scholars and experts on law and scientific evidence interviewed for this article shared Risinger’s praise for the Bushes’ research but were less optimistic about its implications, in part because the criminal justice system so far hasn’t recognized the significance of their work.

But from a scientific standpoint, the Bushes’ research was a direct and severe blow to the credibility of bite mark analysis. At least initially, it threatened to send the entire field the way of voice print matching and bullet lead analysis, both of which have now been discredited. And so when defense attorneys began asking the couple to testify in court, the bite mark analysts fought back with a nasty campaign to undermine the Bushes’ credibility. In a letter to the editor of the Journal of Forensic Sciences, seven bite mark specialists joined up to attack the Bushes in unusually harsh terms for a professional journal. When that letter was rejected for publication, five of the same analysts wrote another. That, too, was rejected. A toned-down but still cutting third letter was finally published.

In the unpublished letter dated November 2012, the authors — all bite mark analysts who hold or have held positions within ABFO — declared it “outrageous that any of these authors would go into courts of law and give sworn testimony citing this research as the basis for conclusions or opinions related to actual bite mark casework, especially considering that no independent research has validated or confirmed their methods or findings.”

Of course, critics would say this was a bit of rhetorical jujitsu — that the last clause could describe exactly what bite mark analysts have been doing for 35 years. For emphasis they added, “This violates important principles of both science and justice.” In the other letter, the authors referred to the Bushes’ testimony in an Ohio case, which was based upon their research, as “influenced by bias” and “reprehensible and inexcusable.”

The primary criticism of the Bushes’ research is that they used vice clamps to make direct bites into cadavers that were stationary through the entire process. This is quite a different scenario than the way a bite would be administered during an attack. During an assault, the victim would probably be pulling away, causing the teeth to drag across the skin. For the Bush tests, the clamp they used to make the bites moved only up and down. A human jaw also moves side to side. A biter might also twist his head or grind his teeth. A live body will also fight the bite at the source to prevent infection, causing bruising, clotting and various other defenses that would alter the appearance of the bite.

“We acknowledge that our lab tests are different from how bites are made in the real world,” says Mary Bush. “But to the extent that our tests differed, they should have made for better preserved samples.”

In other words, the tests that the Bushes conducted made for cleaner, clearer bites that could be easily analyzed. If they were in error, they were in error to the benefit of the claims of bite mark analysts. And they still found no evidence to support the field’s two basic principles.

“That’s exactly right,” says Risinger. “If there was any validity to bite mark analysis at all, these tests would have found it. They gave the field the benefit of the doubt. The evidence just wasn’t there. Their data is very, very strong.”

To argue that the Bushes’ experiments should be disregarded because they weren’t able to replicate real-world bites is also an implicit acknowledgment that real-world bites aren’t replicable in a lab, and therefore aren’t testable. You won’t find many people volunteering to allow someone else to violently bite them for the purposes of lab research. Even if you could, a volunteer won’t react the same way to a bite that an unwitting recipient might.

The Bushes’ research not only failed to find any scientific support for bite mark matching, but it also exposed the fact that for four decades the bite mark community neglected to conduct or pursue any testing of its own. It put the ABFO and its members on the defensive. The bite mark analysts responded by intensifying their attacks on the couple and making the attacks more personal.

At the February 2014 AAFS conference in Seattle, the ABFO hosted a dinner for its members. The keynote speaker was Melissa Mourges, an assistant district attorney in Manhattan, one of the most outspoken defenders of bite mark matching in law enforcement.

Mourges already had a high profile. The combative, media-savvy prosecutor was part of the prosecution team featured in the HBO documentary “Sex Crimes Unit,” which followed the similarly named section of the Manhattan DA’s office, the oldest of its kind in the country. Mourges herself founded a cold-case team within that unit. At the 2012 AAFS conference she spoke on a panel called “How to Write Bestselling Novels and Screenplays in Your Spare Time: Tips From the Pros.” At this year’s conference, she’ll be on a panel that’s titled “Bitemarks From the Emergency Room to the Courtroom: The Importance of the Expert in Forensic Odontology.” She’ll be co-presenting with Franklin Wright, the former ABFO president who initially supported the Bushes’ research.

Mourges was also the lead prosecutor in State v. Dean, a New York City murder case in which the defense challenged the validity of the state’s bite mark testimony. In 2013, Manhattan state Supreme Court Judge Maxwell Wiley held a hearing on the scientific validity of bite mark evidence. Mary Bush testified about the couple’s research for the defense. It was the first (and so far the only) such hearing since the NAS report was released, and both sides of the bite mark debate watched with anticipation. In September 2014, Wiley ruled for the prosecution, once again allowing bite mark evidence to be used at trial. (I’ll have more on the Dean case in part four of the series.) Mourges’s talk at the ABFO dinner was basically a victory lap.

There’s no transcript of Mourges’s speech, but those in attendance say it was basically a no-holds-barred attack on Mary Bush. Cynthia Brzozowski has been practicing dentistry in Long Island for 28 years and sits on the ABFO Board of Directors. She practices the widely accepted form of forensic dentistry that uses dental records to identify human remains, but she doesn’t do bite mark matching, and she won’t testify in bite mark cases. Brzozowski was at the dinner in Seattle and says she still can’t believe what she heard from Mourges.

“Her tone was demeaning,” Brzozowski says. “It would be one thing if she had just come out and presented the facts of the case, but this was personal vitriol against the Bushes because of their research.”

According to Brzozowski, Mourges even went after Mary Bush’s physical appearance. “At one point, she put up an unflattering photo of Mary Bush on the overhead. I don’t know where she got it, or if it had been altered. Mary Bush is not an unattractive person. But it was unnecessary. You could hear gasps in the audience. It was clear that she had chosen the least flattering image she could find. Then she said, ‘And she looks better here than she does in person.’ It was mean. I had to turn my back. I was mortified.”

Other ABFO members — including two other members of the board of directors — also complained, to both the ABFO and the AAFS. The complainants described Mourges’s attack on Bush as “malicious,” “bullying” and “degrading.” According to accounts of those in attendance, other members were also upset by Mourges’s remarks but didn’t file formal complaints for fear of professional retaliation.

A few weeks later, Loomis sent an e-mail to the ABFO Board of Directors to address the complaints. Loomis defended Mourges and her presentation. He described the dinner as a “convivial affair” where members can socialize, have a libation and “be entertained” by the invited speaker. He argued that “anyone who understands litigation” should not have been unsettled by the talk and described the presentation as “sarcastic, serious, and even light-hearted.” He stood by the decision of his predecessor, Greg Golden, to invite Mourges, calling it “a good decision,” adding, “I apologize to those who were offended. However, I do not apologize for the message.”

“‘Bullying’ is exactly what it is,” says Peter Bush. “We’re scientists. We’re used to collegial disagreement. But we had no idea our research would inspire this kind of anger.”

. . .

Addendum: After this post was published, 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.

Excerpt from Part 4 of 4:

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.”

Texas science commission is first in the U.S. to recommend moratorium on bite mark evidence

By Brandi Grissom, Trail Blazers Blog, Feb. 12. 2016

Texas on Friday became the first state in the nation to recommend a ban on the use of bite mark analysis in criminal cases, a decision that could prompt change in courtrooms nationwide.

The Texas Forensic Science Commission recommended courts institute a moratorium on the use of bite marks until additional scientific research is done to confirm its validity. The decision comes after Steven Mark Chaney, a Dallas man, was freed from prison last fall when the court agreed that bite mark evidence used to convict him of murder was bogus. Courts nationwide have used such evidence for decades to identify suspects in murders, sexual assaults, child abuse and other violent crimes. In recent years, though, critics have posed serious questions about whether bite marks can be used accurately pinpoint the perpetrator of a crime. Texas, which has become a national leader in forensic science developments, becomes the first jurisdiction to advocate abandoning it as evidence in court until scientific criteria for its use is created.

“It shouldn’t be brought into court until those standards are done,” said Richard Alpert, an assistant criminal district attorney for the Tarrant County District Attorney’s Office and a member of the Texas Forensic Science Commission.

The New York-based Innocence Project asked the commission last fall to investigate the use of bite mark analysis as evidence. The Innocence Project and other critics of the practice argue that bite mark analysis has no scientific basis and shouldn’t be used in courtrooms.

Chris Fabricant, director of strategic litigation for the Innocence Project called the decision “incredibly significant.”

“The Texas Forensic Science Commission has taken a giant step in purging unscientific and unreliable bite mark evidence from court rooms nationwide,” he said in a prepared statement.

Members of the American Board of Forensic Odontology, which oversees the practice of bite mark analysis, have fiercely defended the validity of bite mark analysis. The dentists acknowledge some mistakes with analysis in the past but contend that when used appropriately, such evidence can be a useful investigative tool.

Bite marks, they say, can be particularly useful as evidence in child abuse cases, where the patterns left on a child’s body can help get a young one out of a dangerous household.

“What we hope for is to avoid unintended consequences,” said dentist David Senn, a clinical assistant professor at the University of Texas Health Science Center at San Antonio Dental School and member of the ABFO.

Chaney, whose case prompted the commission’s decision Friday, was freed from prison in October after Dallas County prosecutors agreed that the bite mark evidence that led to his conviction for a 1987 murder was invalid. Dr. Jim Hales told a Dallas County jury that there was a “1-to-a million” chance that someone other than Chaney left a bite mark on the arm of John Sweek, who had been stabbed to death.

Chaney’s case was hardly the first to involve bite mark evidence, though. The practice seems to date back to a 1954 robbery case in West Texas. A grocery store owner arrived one December morning to find someone had ransacked his shop and stolen two bottles of whiskey and 13 silver dollars. The culprit had left behind a mess of cold cuts and cheese on the meat counter.

Police found James Doyles’ teeth marks in a piece of cheese. They forced Doyle — who was already in jail for public intoxication — to bite into another piece of cheese.

A dentist concluded the teeth marks matched, and Doyle was sentenced to two years for burglary.

In recent years, a number of scientific studies, including a 2009 report from the National Academy of Sciences, have found that there is little to no scientific basis for conclusions that match a person to a set of apparent bite marks.

At least one study found that dentists were largely unable to agree on which injuries were the result of a bite. Other experts have raised questions about whether skin is an accurate template on which to assess bite marks and about how the passage of time affects the reliability of the patterns that teeth leave.

The Associated Press reported in 2013 that at least 24 people in the U.S., including two in Texas, had been exonerated in cases in which bite mark evidence played a central role in the conviction.

Even the ABFO, which certifies dentists who analyze bite marks, has decided the evidence can’t be used to draw the kind of strong conclusions that Hale made during Chaney’s trial.

“There’s no question that was improper,” Senn told commission members Thursday.

The commissioners left open the possibility that bite mark analysis could receive approval in the future if additional research and testing was performed that improved the reliability of it as evidence.

Senn told a panel of commission members on Thursday that certified ABFO dentists in Texas – there are nine of them – would work with state officials to keep them informed of additional research. Those dentists, he said, would also provide the commission with a list of the Texas cases in which they provided bite mark testimony.

That will help the commission and prosecutors who are undertaking the daunting task of identifying cases in which bite mark evidence may have led to a wrongful conviction. There is no central database of cases in which such testimony was crucial to a conviction, which has left the commission and prosecutors to search through legal cases on the web and comb through reams of trial transcripts.

It is unclear just how many cases involving bite mark evidence are pending in Texas courts. It is most commonly used in cases related to child abuse. A recommendation from the commission does not require judges to prohibit such evidence, but it is likely to influence their decisions about whether bite marks should be allowed as evidence.

The motion seeks to vacate the death sentence for a man who was convicted in Mississippi on the basis of bite mark testimony. (The motion is a PDF so is posted on Moodle.)

Feel free to skim the facts (and note there are facts about sexual assault). Be sure to read (a) the Table of Contents at page 2 to understand the structure of the argument; (b) the portion of the legal argument on pages 42-63; and (c) the conclusion on page 75.

One of the cases mentioned in The Washington Post’s series on bite mark evidence is the New York case of People v. Dean.  This memo is from the District Attorney’s office to the judge in that case. (Because this is a PDF document it is posted on Moodle.

Posted on Moodle.

Researchers recently learned that Immigration and Customs Enforcement used facial recognition on millions of driver’s license photographs without the license-holders’ knowledge, the latest revelation about governments employing the technology in ways that threaten civil liberties.

But the surveillance potential of facial recognition — its ability to create a “perpetual lineup” — isn’t the only cause for concern. The technological frontiers being explored by questionable researchers and unscrupulous start-ups recall the discredited pseudosciences of physiognomy and phrenology, which purport to use facial structure and head shape to assess character and mental capacity.

Artificial intelligence and modern computing are giving new life and a veneer of objectivity to these debunked theories, which were once used to legitimize slavery and perpetuate Nazi race “science.” Those who wish to spread essentialist theories of racial hierarchy are paying attention. In one blog, for example, a contemporary white nationalist claimed that “physiognomy is real” and “needs to come back as a legitimate field of scientific inquiry.”

More broadly, new applications of facial recognition — not just in academic research, but also in commercial products that try to guess emotions from facial expressions — echo the same biological essentialism behind physiognomy. Apparently, we still haven’t learned that faces do not contain some deeper truth about the people they belong to.

Composite photographs, new and old

One of the pioneers of 19th-century facial analysis, Francis Galton, was a prominent British eugenicist. He superimposed images of men convicted of crimes, attempting to find through “pictorial statistics” the essence of the criminal face.

Galton was disappointed with the results: He was unable to discern a criminal “type” from his composite photographs. This is because physiognomy is junk science — criminality is written neither in one’s genes nor on one’s face. He also tried to use composite portraits to determine the ideal “type” of each race, and his research was cited by Hans F.K. Günther, a Nazi eugenicist who wrote a book that was required reading in German schools during the Third Reich.

Galton’s tools and ideas have proved surprisingly durable, and modern researchers are again contemplating whether criminality can be read from one’s face. In a much-contested 2016 paper, researchers at a Chinese university claimed they had trained an algorithm to distinguish criminal from noncriminal portraits, and that “lip curvature, eye inner corner distance, and the so-called nose-mouth angle” could help tell them apart. The paper includes “average faces” of criminals and noncriminals reminiscent of Galton’s composite portraits.

The paper echoes many of the fallacies in Galton’s research: that people convicted of crimes are representative of those who commit them (the justice system exhibits profound bias), that the concept of inborn “criminality” is sound (life circumstances drastically shape one’s likelihood of committing a crime) and that facial appearance is a reliable predictor of character.

It’s true that humans tend to agree on what a threatening face looks like. But Alexander Todorov, a psychologist at Princeton, writes in his book “Face Value” that the relationship between a face and our sense that it is threatening (or friendly) is “between appearance and impressions, not between appearance and character.” The temptation to think we can read something deeper from these visual stereotypes is misguided — but persistent.

In 2017, the Stanford professor Michal Kosinski was an author of a study claiming to have invented an A.I. “gaydar” that could, when presented with pictures of gay and straight men, determine which ones were gay with 81 percent accuracy. (He told The Guardian that facial recognition might be used in the future to predict I.Q. as well.)

Please go to our Moodle Page and under "Class 10" you will find the prompt and submission folder for Writing Reflection #10.

Attorneys and Law Firms

OPINION OF THE COURT

Chief Judge DiFIORE.

False confessions elicited during custodial interrogations do exist. In People v. Bedessie, 19 N.Y.3d 147, 161, 947 N.Y.S.2d 357, 970 N.E.2d 380 (2012), we recognized that the phenomenon of false confessions during custodial interrogation is common knowledge, and we opined that expert psychological testimony relevant to the defendant and the custodial interrogation at issue could be admissible “to educate a jury about those factors of personality and situation that the relevant scientific community considers to be associated with false confessions.” The admissibility and limits of the expert's testimony lie primarily in the sound discretion of the trial judge, and the expert may not render an opinion as to the truthfulness or falsity of the confession. The primary issue presented in this case is whether the trial court, in denying defendant's Bedessie application, erred in precluding the testimony *480 of defendant's proffered expert witness on false confessions after holding Frye and Huntley hearings. On this record, the trial court's determination was not an abuse of discretion as a matter of law.

I.

Defendant was charged with having committed two elevator robberies over the course of three days. During the first incident on February 23, 2010, HT was robbed at knifepoint in an elevator of a building in the Queensbridge Housing Complex. Two days later, a second individual, EY, was robbed by a man displaying what appeared to be a box cutter, in an elevator of the same housing complex. EY first encountered the robber, whom she described as a six-foot-tall, light-skinned black male in his 30s, carrying an umbrella, when he asked her to light his cigarette as she exited a local grocery store at about 3:00 p.m. on the afternoon of February 25. As EY walked back to her nearby apartment, the same man followed her into her building. ***239 The two entered the elevator and, when the doors closed, the man physically attacked the victim. The robbery, including a face-to-face struggle between the victim and her assailant that lasted more than two minutes, was captured on the building's **1032 video surveillance system. Although the perpetrator's face is not clearly visible to the viewer of the video, the film depicts a tall man wearing a hooded, dark-colored coat and hat, carrying a large black umbrella with a curved handle.

EY's assailant fled with her electronic benefits transfer (EBT) card. Surveillance footage from a nearby deli, recorded at approximately 4:00 p.m., depicts a tall man similar in build and appearance to the man who committed the robbery—wearing the same type of clothing and carrying a large umbrella—attempt to use the stolen EBT card a short time after the robbery. The card was declined and the man then attempted to use it in the ATM. The surveillance footage from the well-lit store shows portions of the man's face from multiple angles. By interviewing witnesses and matching photographs with the surveillance footage, the police identified defendant as a suspect.

On March 1, 2010, defendant was arrested on the fifth floor of a building in the Queensbridge Housing Complex for possession of crack cocaine and was transported to the local precinct. There, he was interviewed about the robberies that took place on February 23 and 25. The following day, defendant made the *481 two statements that are in issue here. The first statement, elicited during the morning hours, was handwritten by defendant. Without providing any detail, defendant admitted to having committed robberies, claimed to have been under the influence of drugs and indicated he wanted to help the police.¹ The second statement was elicited in the afternoon, after defendant had been identified in lineups by both victims. The statement prepared and typewritten by the detective assigned to the case summarized defendant's detailed oral statement, in which he admitted following the victim (EY), who was carrying groceries, and robbing her of her EBT card after a struggle in the elevator. Defendant again claimed to have been “messed up on drugs” and that he wanted to help the police.² He signed the statement on the second page, which contained no factual allegations.

Defendant was indicted for two counts of robbery in the first degree.³ Prior to trial, defendant served a CPL 250.10 notice of intent to introduce psychiatric evidence for ***240 the purpose of demonstrating that he was suffering from psychiatric conditions that “adversely affected the voluntariness and reliability of the interrogations conducted.” In support of his position, defendant submitted medical records of his history of a seizure disorder, depression, and schizophrenia **1033 , as well as a forensic psychological report prepared by Sanford L. Drob, Ph.D., who conducted a clinical evaluation of defendant. Dr. Drob characterized defendant's *482 intellectual function to be within the “borderline range,” with an IQ of 78. He concluded that defendant had several mental health issues, including possible paranoid ideation and substance abuse issues. Dr. Drob opined that the combination of attendant factors, including defendant's mental illness and cognitive deficits, “could make him vulnerable to suggestion in a custodial setting.”

Huntley Hearing

Defendant moved to suppress the two noticed statements. At the Huntley hearing, Detective Grinder and defendant testified to contradictory narratives of the circumstances surrounding the custodial interrogation.⁴ Each version is summarized here, as is necessary to evaluate the trial court's Bedessie determination.

Detective Grinder testified that defendant was arrested at about 2:20 p.m. on March 1. While defendant was held in the precinct's interrogation room, one of his hands was handcuffed to the wall. Grinder, who arrived at the precinct about two hours after the arrest, advised defendant of his Miranda rights later that same evening at about 6:30 p.m. Defendant signed and initialed the Miranda card, waiving his rights, but did not make any admissions on March 1. When told his arrest was made in connection with robberies committed in the Queensbridge Housing Complex, defendant became visibly agitated and denied any involvement. The detective, who intended to conduct lineups the next day, ended the interview. Upon defendant's request, Detective Grinder traveled to defendant's friend's home, where he retrieved four bottles of medication prescribed to defendant. The detective returned to the precinct and vouchered the medications. He could not recall whether defendant took the medication. Shortly thereafter, at about 11:50 p.m., defendant was transported to central booking for lodging that night.

On the morning of March 2, at approximately 9:30 a.m., Detective Grinder and another detective transported defendant from central booking back to the precinct. Defendant was given breakfast—coffee and a bagel. Defendant's demeanor was now calm, and he agreed to speak with detectives. Miranda warnings *483 were not reissued, nor was the interview recorded. At about 10:00 a.m., the detectives began discussing the charges and informed defendant that he would be required to participate in lineup identification procedures. Detective Grinder supplied defendant with pen and paper and left him alone in the room where he handwrote the first statement.

Two lineups, each with six participants, were conducted on March 2 at 12:30 p.m. and both victims identified defendant. Shortly after the lineups were conducted, and while in the interrogation room with the two detectives at about 1:00 p.m., defendant asked the detectives if he had been identified by the victims. The detectives informed him that he had been identified by both victims and asked if he wanted to make any further statements. Defendant then admitted that he had committed ***241 four robberies, including the two charged here.⁵ He asked Grinder to write **1034 out the confessions for him, as he was not a good writer, and proceeded to give the second statement orally. Grinder, who did not take any notes, later typed up the confession, which defendant signed on the blank, second page. Detective Grinder testified that he spoke with defendant on-and-off “the whole day” on March 2 and that he provided defendant a meal from Burger King. He further testified that defendant did not request any medical assistance and did not appear to be experiencing any type of drug withdrawal. At about 8:30 p.m. that night, defendant was returned to central booking.

Testifying on his own behalf, defendant gave a very different version of the interrogation events. As to his background, he testified that he had low intelligence, suffered from seizures, and had a history of schizophrenia, depression and substance abuse. He had been prescribed various medications for those conditions but the last time he took his medications was at about 8:00 p.m. the night before his arrest (February 28). On the morning of March 1, defendant ingested heroin and crack cocaine. After his arrest that afternoon, he was feeling paranoid and scared and, later that day, at some time between 4:00 and *484 5:00 p.m., while he was handcuffed to the wall in the precinct interrogation room, he had a seizure and urinated on himself. He asked Detective Grinder for his medication and Grinder told him that he would not get the medication or medical treatment unless defendant cooperated. Further, according to defendant's testimony, Grinder threatened that if he asked to go to the hospital again, Grinder would make sure defendant did not go before a judge for four or five days. Grinder then left defendant to go to an outside location to retrieve defendant's medication at approximately 9:00 p.m. When he returned, Grinder placed the medication at the end of the table in the room, out of defendant's reach. Defendant claimed he received the medication the next day (March 2), but only after he gave the police the handwritten statement, which he denied was truthful. Defendant testified that he only wrote the statement to appease the detective because he was deprived of his medication and food and was scared that he would have another seizure.

On cross-examination, defendant alleged that, on the night of March 1, between 7:00 and 8:00 p.m., while in the interrogation room, Detective Grinder hit him in the head four or five times. Later that night, when he was brought back to central booking for lodging, he saw medical personnel but, other than providing basic history, he did not mention anything about his mental health or seizures because he did not feel that he could tell them what was going on. Defendant, when shown the exhibit of the typewritten, signed confession, repeatedly denied that he made those admissions or that he had ever “seen or heard th[o]se statements.” He testified that, after the lineups, he simply signed the blank second page. Notwithstanding his concession that it was his signature and initials on the Miranda card, he claimed that he did not receive Miranda warnings until about 6:00 p.m. on March 2, after he signed the second ***242 page of the typewritten statement, which contained no factual allegations.

The hearing court credited defendant's testimony only to the extent it did not conflict with Grinder's and denied the motion to suppress the statements.

Frye Hearing

Citing to People v. Bedessie, defendant moved to admit the testimony of Allison Redlich, Ph.D., as an expert witness “to educate the jury” on factors that were generally accepted in the *485 relevant scientific **1035 community as associated with false confessions. Defendant attached a 12-page report from the highly credentialed Dr. Redlich based on her review of the Huntley hearing testimony, police reports, and Dr. Drob's report. In order to resolve the conflicting witness accounts of the custodial interrogation in her report, Dr. Redlich credited defendant's account, which she found “more telling.” To demonstrate general acceptance in the field of false confession research, attached to the motion were several publications, including some that were co-authored by Dr. Redlich.⁶

The trial court ordered a Frye hearing to address the admissibility and scope of the proposed testimony.⁷ At the hearing, Dr. Redlich was found to be an expert in the field of false confession studies. As stated in her report, and mirroring the expert testimony proffered in Bedessie, Dr. Redlich set forth the three types of false confessions: voluntary (not coerced—could be offered to protect another or attain notoriety), coerced compliant (where the suspect's will is overborne) and internalized (through deceptive interrogation techniques, the suspect comes to believe he or she is guilty). She also set forth the paradigm of a series of dispositional and situational factors that have been recognized as contributing to the risk of false confessions. Relevant to the facts of this case, Dr. Redlich identified three dispositional factors that defendant displayed, based on his medical reports: mental illness, intellectual disability and substance abuse. As to situational factors, Dr. Redlich found that defendant was in custody and questioned intermittently for over 24 hours, that his statements evidenced minimization by the reference to his drug abuse, and that the statements admitting to the crimes charged did not provide any information that was not already known to the police. Related to these situational factors she focused primarily on the Reid technique, a nine-step method of police interrogation based on psychological principles, which she opined was widely used. As part of the Reid technique, the suspect is isolated and confronted with the interrogator's alleged knowledge of his *486 guilt. The interrogator then engages in tactics like presentation of false evidence, minimizing the suspect's responsibility for the crime, or theme development, making the suspect more compliant in offering a confession that mitigates his role. Dr. Redlich opined that these techniques could produce a false confession when employed on innocent individuals—especially where other dispositional or situational risk factors were present.

On cross-examination, Dr. Redlich conceded that she was unaware of the type of training, if any, members of the New York ***243 City Police Department received in interrogation techniques, but that, even if they were not specifically trained in the Reid technique, “all interrogations ... pretty much follow the same model.” She also testified that, in her role as a research psychologist, she analyzed videos of actual interrogations, placed reliance on self-reporting from subjects and reviewed legal cases. When queried as to laboratory studies in her field, she identified a study in which her team interviewed defendants in the criminal justice system, including inmates in the Santa Clara County Jail who **1036 had confessed to crimes—whether truly or falsely, based on their self-reporting—about their experiences with interrogation. The subjects all had mental illness and 86% of them “had a known serious mental disorder.” They were paid for their cooperation and, without the presence of counsel, were asked whether they had ever confessed to or pleaded guilty to a crime they did not commit. No measure was taken to ascertain if their self-reports were true or false. She also testified to two additional studies—the “Alt Key Paradigm” and the “cheating paradigm”—that attempted to use deceptive tactics to induce false confessions in a laboratory-like environment. She conceded that the “Alt Key” study was flawed and no longer used and that the “cheating paradigm” was distinguishable from police interrogation. Dr. Redlich noted that laboratory studies in the field of false confessions are not akin to the analysis done in other scientific fields, such as DNA evidence, where the tests conducted pursuant to an accepted methodology can be replicated to achieve the same results. Laboratory studies in the field of false confessions are likewise distinct from studies done in the field of misidentification, as there are practical difficulties and ethical considerations in recreating the circumstance of a real-life custodial interrogation in a laboratory setting. She further acknowledged that, as to the laboratory studies in her field, there are issues of *487 external validity and the ability to translate the results to the criminal justice system.

Following the hearing, the court denied defendant's request to call Dr. Redlich as a witness at trial, finding that, based on the evidence presented, defendant failed to meet his burden of establishing that Dr. Redlich's testimony in the area of false confessions was “readily acceptable in the scientific community.” The court deemed the showing that the phenomenon of false confessions exists and that interrogation tactics may increase the likelihood of same insufficient to sustain this burden. The court found Dr. Redlich's testimony “in many respects unpersuasive” and noted her failure to establish a known or potential rate of error and her lack of personal knowledge of the circumstances of defendant's confessions. The court also referenced Dr. Redlich's “strong” reliance on the three laboratory studies to form her opinions, noting that the “Alt Key” study had been discredited and that the “cheating study” was inapplicable. Significantly, the court observed that defendant failed to establish that his statement was induced by any of the factors outlined in Bedessie. However, recognizing that defendant may possess the dispositional characteristics identified in Bedessie, the court left open the possibility that a different expert could testify—one who had personal knowledge of defendant's medical history and the circumstances of the case that may have led to a false confession.

At the next appearance, the court explained that it did not dispute that “the science may be there with respect to false confessions,” but that it did not think Dr. Redlich was the proper witness to educate the jury about the phenomenon, given, in part, her testimony that some of the studies ***244 on which she relied had since been disavowed. The court clarified that its decision was based on its determination that Dr. Redlich lacked the necessary expertise, rather than the fact that she had not interviewed defendant. Defense counsel then advised the court that he did not intend to call Dr. Drob, the psychiatric examiner whose report on defendant's dispositional factors was relied upon by Dr. Redlich, because defendant was competent to testify at trial as to his own mental conditions.

At trial, defendant's testimony as to the circumstances of his custodial interrogation **1037 was consistent with the testimony he *488 gave at the Huntley hearing.⁸ He testified that his first handwritten statement, which contained no detail linking him to any particular robbery, was coerced—given by him in an attempt to escape the interrogation and get his medication. He testified that he never made the second detailed, typewritten statement and maintained it was wholly manufactured by Detective Grinder. He further claimed all the statements were elicited before he was advised of his Miranda rights.⁹

The victim, EY, testified at trial and identified defendant as the robber. During cross-examination, defendant elicited inconsistencies between the description she provided to police and his physical appearance—in particular, he was six feet, four inches tall as opposed to six feet, and was significantly older than her description of the perpetrator. Defendant also testified that he was missing 12 teeth in 2010, a circumstance that was not noted by the victim. The surveillance footage from the elevator and the deli were admitted into evidence and played for the jury.

As required by constitutional law in cases where the voluntariness of a statement is placed in issue at the trial (see Jackson v. Denno, 378 U.S. 368, 84 S.Ct. 1774, 12 L.Ed.2d 908 [1964]; CPL 60.45; CPL 710.70; People v. Graham, 55 N.Y.2d 144, 147, 447 N.Y.S.2d 918, 432 N.E.2d 790 [1982]), the jury was instructed that, before they could consider defendant's statement as evidence, they had to find the People proved beyond a reasonable doubt that it was voluntarily made. This instruction required the jury to find that defendant received, understood, and waived his Miranda rights, or the statement must be disregarded. The jury was further instructed that they must determine whether any statement was obtained by “improper conduct or undue pressure,” factoring in such dispositional factors as defendant's intelligence and his physical and mental *489 condition, as well as situational considerations such as the length of time defendant was questioned and the treatment he received from the police during that time. The court also gave the jury an expanded eyewitness identification charge, including a charge on cross-race effect (see People v. Boone, 30 N.Y.3d 521, 69 N.Y.S.3d 215, 91 N.E.3d 1194 [2017]).

***245 The jury convicted defendant of robbery in the first degree. Defendant then pleaded guilty to an additional count of first-degree robbery to resolve the count of the indictment involving the robbery of HT, which had previously been severed. The Appellate Division affirmed, holding that it was a provident exercise of discretion for Supreme Court to deny defendant's motion to present expert witness testimony on the phenomenon of false confessions because defendant failed to demonstrate the proposed testimony was relevant to the circumstances of his case (166 A.D.3d 660, 661, 87 N.Y.S.3d 31 [2d Dept. 2018]). A **1038 Judge of this Court granted defendant leave to appeal (33 N.Y.3d 980, 101 N.Y.S.3d 245, 124 N.E.3d 734 [2019]).

II.

The admissibility and scope of expert testimony are subject to the discretion of the trial court (see People v. Lee, 96 N.Y.2d 157, 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 [2001]), limiting our scope of review to whether the determination to exclude the proffered expert testimony was an abuse of that discretion as a matter of law. Here, the court was required to determine, under Frye, whether the proposed expert opinion testimony was based on principles and methodologies generally accepted within the relevant scientific community. In addition, even where based on reliable principles and methods, an expert's opinion may be precluded if it presents “too great an analytical gap between the data and the opinion proffered” (Cornell v. 360 W. 51st St. Realty, LLC, 22 N.Y.3d 762, 781, 986 N.Y.S.2d 389, 9 N.E.3d 884 [2014] [citation and quotation marks omitted]).

It is well-settled that the inquiry under Frye “ ‘is separate and distinct from the admissibility question applied to all evidence—whether there is a proper foundation—to determine whether the accepted methods were appropriately employed in a particular case’ ” (People v. Brooks, 31 N.Y.3d 939, 941, 73 N.Y.S.3d 110, 96 N.E.3d 206 [2018], quoting Parker v. Mobil Oil Corp., 7 N.Y.3d 434, 447, 824 N.Y.S.2d 584, 857 N.E.2d 1114 [2006]). Further, particularly “in the social science arena, we have measured the reliability of novel hypotheses and theories—not just methodologies—against the Frye standard” ( *490 Cornell, 22 N.Y.3d at 781, 986 N.Y.S.2d 389, 9 N.E.3d 884). And, under Bedessie, in addressing the phenomenon of false confessions, the trial court must also determine whether the same proffered testimony was “relevant to the defendant and interrogation before the court” (Bedessie, 19 N.Y.3d at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380). “ ‘It is for the trial court in the first instance to determine when jurors are able to draw conclusions from the evidence based on their day-to-day experience, their common observation and their knowledge, and when they would be benefited by the specialized knowledge of an expert witness’ ” (Lee, 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63, quoting People v. Cronin, 60 N.Y.2d 430, 433, 470 N.Y.S.2d 110, 458 N.E.2d 351 [1983]). We have otherwise characterized this determination as an evaluation of whether the testimony “ ‘would aid a lay jury in reaching a verdict’ ” (96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63, quoting People v. Taylor, 75 N.Y.2d 277, 288, 552 N.Y.S.2d 883, 552 N.E.2d 131 [1990]).

As we recognized in Bedessie in 2012, awareness of the phenomenon of false confessions has evolved to the point of “common knowledge, if not conventional wisdom” (19 N.Y.3d at 156, 947 N.Y.S.2d 357, 970 N.E.2d 380).¹⁰ This, however, does ***246 not mean that expert testimony on the theories behind the reasons for false confessions is rendered unnecessary. Certain aspects of the scientific study of the phenomenon might well be outside the ken of the typical juror (see 19 N.Y.3d at 156, 947 N.Y.S.2d 357, 970 N.E.2d 380; Lee, 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63). The proffered psychological expert testimony here is, in certain respects, similar to the expert psychological testimony on rape trauma syndrome addressed in **1039 People v. Taylor, 75 N.Y.2d 277, 552 N.Y.S.2d 883, 552 N.E.2d 131. That is, the testimony is not addressed to a particular scientific technology or procedure that, when properly performed, will generate results generally accepted as reliable in the scientific community and admitted into evidence to demonstrate an evidentiary fact (see e.g. People v. Wesley, 83 N.Y.2d 417, 422, 611 N.Y.S.2d 97, 633 N.E.2d 451 [1994]). Rather, it is meant to be used as an informational tool to educate the jury on the causal connection between relevant factors and false confessions outside their ken, and to do so without opining on the particular facts of the case. In Taylor, and the companion case of Banks, for example, we held that the expert testimony was admissible to dispel common misperceptions regarding recognized patterns of posttraumatic behavior on the part of rape victims (Taylor), but not to prove that the victim's behavior was aligned with a common perception *491 that a rape occurred (Banks). Indeed, the latter is akin to an improper bolstering of credibility. Here, the proffered testimony would not have been admissible for the purpose of establishing that a false confession occurred, but to educate the jury about the science of the association between psychological risk factors occurring in a particular custodial interrogation and the making of a false confession, in order to address common misconceptions about a person making a false admission of criminal conduct.

On this record, the trial court did not abuse its discretion in finding that the proffered testimony would not have aided the jury. Although Dr. Redlich is an impressively credentialed researcher, properly qualified by the trial court as an expert in her field, the trial court found that her testimony at the Frye hearing revealed her difficulty in linking her research on the possible causes of false confessions to the case at hand. Despite her review of the witnesses’ testimony at the Huntley hearing, she did not explain how her testimony was at all relevant to the circumstances presented by defendant's interrogation, even by crediting defendant's account of the events over Detective Grinder's.¹¹ For instance, defendant flatly denied ever making the second, more detailed, confession—so, expert testimony regarding dispositional and situational factors that create a risk of a false confession has no relevance to the oral or written version of that statement. Moreover, defendant maintained that the first handwritten statement was the product of outright coercion—including a physical assault the night before and the deprivation of food and medicine—rather than resulting from psychological coercion of police interrogation that creates the risk of false confession, consistent with a recondite theory of which Dr. Redlich would have testified. There is a difference between the classically, inherently coercive interrogation that ***247 produces an involuntary confession—an issue that the jury is well-equipped to understand (see e.g. Blackburn v. Alabama, 361 U.S. 199, 206, 80 S.Ct. 274, 4 L.Ed.2d 242 [1960]; People v. Anderson, 42 N.Y.2d 35, 37-38, 396 N.Y.S.2d 625, 364 N.E.2d 1318 [1977])—and the phenomenon of false confessions involving the interplay of situational and dispositional factors that produce a coercive compliant false confession from an innocent suspect, an occurrence that the jury may find counterintuitive.

*492 Dr. Redlich's written report and her hearing testimony covered a broad spectrum of situational factors, some of which required no explanation to a jury, and others of which had no relevance to defendant or the totality of the circumstances of his interrogation. Dr. Redlich did not anchor **1040 her testimony to a possible version of the custodial interrogation as set forth in the Huntley hearing testimony. Rather, she broadly stated that the basic tool of psychological tactics inherent in an interrogation is employed in all human interactions to obtain compliance, a statement entirely consistent with the commonsense ability of the jury to discern the effects of the interaction between defendant and the police, absent some extraordinary factor. She then focused heavily on the nine-step Reid technique, which she presumed was used in the case. As the opinion in Miranda makes unequivocal with its detailed exploration of the Reid technique, our law has long recognized these specific stratagems may compel the suspect to “merely confirm[ ] the preconceived story the police seek to have him describe” (384 U.S. 436, 455, 86 S.Ct. 1602, 16 L.Ed.2d 694 [1966]). At bottom, actual evidence that the Reid technique was used in the interrogation is a necessary predicate for any opinion to be probative evidence on that subject. Dr. Redlich also emphasized situational factors that did not exist in this case, such as sleep deprivation, lying to innocent suspects, presenting false evidence to convince the suspect it is no use to deny culpability, and contamination—where the police, and not the defendant, are the source of the information set forth in the confession.¹² Here, the trial court was well within its province to determine that, since neither witness's account at the Huntley hearing presented circumstances *493 supporting the conclusion that the police used these outlined psychological tactics, any expert testimony on that score would be speculative and delivered in a factual vacuum. Notably, no evidence elicited at trial changed this equation to provide any basis for a renewal of the proffer.

Dr. Redlich's report stated that the failure of the police to record the interrogation ***248 resulted in her uncertainty as to situational factors that may have been present. Video of a past event can certainly enhance the ability of any trier of fact to determine what occurred and video of a custodial interrogation is a boon to the truth. However, as we have recognized, “the neglect to record is not a factor or circumstance that might induce a false confession” (Bedessie, 19 N.Y.3d at 158, 947 N.Y.S.2d 357, 970 N.E.2d 380).¹³ With **1041 respect to defendant's isolation, Dr. Redlich believed it was “quite hard to discern” the length of the interrogation, asserting that defendant was under custodial arrest for more than 24 hours and that the majority (80%) of proven false confessions had interrogations lasting 6 or more hours. Despite the witness's conflation of the period of arrest and the period of interrogation, under either scenario posited by the police or defendant, defendant was concededly noncompliant the night of his arrest, was then lodged overnight at central booking for eight hours and was not the subject of continuous interrogation while at the precinct on March 2; nor did he experience the associated risk factor of sleep deprivation.

As to certain dispositional risk factors particular to defendant, Dr. Redlich identified cognitive impairment, mental illness and substance abuse.¹⁴ The defense chose to not call the CPL 250.10-noticed psychiatric examiner as a witness. Instead, defendant himself testified at trial to the presence of these dispositional factors and was subject to cross-examination, “display[ing] no sign” that he was particularly compliant, that he lacked understanding of the circumstances of the interrogation *494 or that he was impacted by the above dispositional factors (see Bedessie, 19 N.Y.3d at 159, 947 N.Y.S.2d 357, 970 N.E.2d 380). In any event, defendant's substance abuse was not linked to the circumstances of the confession, as there was no evidence that he was either intoxicated or in withdrawal at the time he was questioned. Nor did defendant's testimony provide any indication that he was psychologically tricked or induced to believe he had committed the robberies and admitted same. As to the factor of mental illness, Dr. Redlich testified that there was “some evidence” that there was a link between depression or anxiety and susceptibility to false confessions but then conceded that the “evidence is not entirely clear on that.” She then went on to suggest that “anybody can be rendered vulnerable in the police interrogation situation” and an individual did not need to have any identified dispositional characteristic in order to render a false confession.

The thrust of Dr. Redlich's testimony was that the presence of dispositional factors may be causally linked to false confessions because they may make a vulnerable individual more susceptible to a wide variety of situational factors in the inherently coercive setting of the interrogation. Relying on the research in her field, the witness noted there is no prevalence ***249 rate in the millions of cases where a confession occurred, but approximated that 15 to 20% of the 1,300 documented wrongful conviction cases involved a false confession. In addition, a study in the “white paper,” analyzing 125 cases of proven false confession cases in the United States between 1971 and 2002, found that 63% involved defendants who were under the age of 25 (see 34 Law Hum Behav at 5), a dispositional factor of substantial import. Yet, the witness did not account for defendant's age (51) or his criminal history in relaying these percentages, despite the need at the hearing to demonstrate causative factors in false confession cases she studied. Underscored by nisi prius, her own testimony expressed significant uncertainty as to the applicability of laboratory-like studies to real-life custodial interrogation and was not particularly probative of what actually occurred **1042 in this case. Although Dr. Redlich's testimony may have been informative, the speculative nature of the testimony and the lack of relevance to the particular circumstances of defendant's interrogation presented the risk that the jury might have been confused or misled, rather than aided, by the testimony (see e.g. People v. Cefaro, 23 N.Y.2d 283, 288, 296 N.Y.S.2d 345, 244 N.E.2d 42 [1968] [the jury should not be “encouraged to make a determination *495 in a factual vacuum, i.e., without evidentiary basis whatsoever”]; United States v. Redlightning, 624 F.3d 1090 [9th Cir. 2010], cert denied 563 U.S. 1026, 131 S.Ct. 2944, 180 L.Ed.2d 234 [2011] [expert testimony on false confession generally should not be permitted in the absence of evidence that interrogation techniques likely to extract a false confession were employed]).

The trial court properly recognized that proving false confessions occur, is not the equivalent of accepting Dr. Redlich's broad, unmoored testimony on the science of false confessions and given her difficulty in explaining the external validity of the studies, her imprecise testimony was insufficient to sustain the burden of establishing general acceptance of the psychological principles she was advocating.¹⁵ Contrary to the dissent's suggestion that the Frye hearing curbs the court's discretionary power in allowing an expert to provide educational background to help the jury better understand the risk factors associated with a psychological phenomenon (see dissenting op. at 513, 162 N.Y.S.3d at 262, 182 N.E.3d at 1054–55), as we held in Bedessie, the trial court may not abdicate its responsibility to determine the relevancy of the proffered educational testimony to the particulars of the individual case, to wit, defendant and the interrogation before the court. Further our precedent is clear that a Frye hearing is not the end of the court's determination and the admissibility of social science evidence is a question for the trial court.

We do not equivocate as to the existence of the phenomenon of false confessions or the “evils” that can result from the “interrogation atmosphere” (Miranda, 384 U.S. at 456, 86 S.Ct. 1602). Nonetheless, the scientific principles involve more complexity than the general conclusion that false confessions do occur, and the expert is supposed to articulate those principles so a jury can apply the information ***250 to the actual evidence in the case—not merely speculate in the absence of that evidence. We therefore hold that there is no abuse of discretion when the trial court disallows expert psychological testimony as to false confessions when it is not relevant to the circumstances of the custodial interrogation in the case at hand.

*496 

III.

The remaining issue for our review is whether it was an abuse of discretion for the trial court to deny defendant's motion to admit expert testimony in the area of eyewitness identification. Prior to trial, defendant sought to admit the testimony of Nancy Franklin, Ph.D., to explain factors affecting the reliability of eyewitness identification, as both robberies involved single-witness identifications, made after a brief encounter with an armed stranger of a different race. The court granted the use of an identification **1043 expert in the severed trial relating to the robbery against HT but denied the request in the trial of the robbery here at issue, because the identification by EY was corroborated by defendant's statements as well as the surveillance footage.¹⁶

There was no abuse of discretion in the court's determination. The evidence against defendant at trial did not “turn[ ] solely on the accuracy of the witness[’s] identification” but was corroborated by the surveillance video and defendant's statements (see LeGrand, 8 N.Y.3d at 457, 835 N.Y.S.2d 523, 867 N.E.2d 374). In light of our determination, it is unnecessary to reach defendant's remaining argument.

Accordingly, the order of the Appellate Division should be affirmed.

DISSENT:

RIVERA, J. (dissenting).

At issue on appeal is whether the denial of defendant's motion to present expert testimony at trial on false confessions or on cross-racial identifications violated his right to present a defense relating to his innocence. *497 The majority misapprehends our precedent and the role of the court in determining the admissibility of expert testimony. The proffered false-confession testimony satisfied the standards for admissibility under both People v. Bedessie, 19 N.Y.3d 147, 947 N.Y.S.2d 357, 970 N.E.2d 380 (2012) and Frye v. United States, 293 F. 1013 (D.C. Cir. 1923) because defendant established, through an extensive record, that the science of false ***251 confessions was generally accepted in the relevant scientific community.¹ The lower courts—and the majority—should have ended the analysis there, rather than focusing on questions of foundation, the fit between the proffered testimony and the facts of the case, and the methodologies used by social sciences researchers—none of which are relevant at a Frye hearing. Given our Court's recognition that false confessions occur and that expert testimony on this phenomenon would aid a jury in assessing the voluntariness of a defendant's incriminating statements, the majority **1044 unjustifiedly delays our inevitable determination that there is general acceptance within the scientific community that there are situational and dispositional factors that lead to false confessions.

The trial court also abused its discretion in denying defendant's request to present expert testimony on eyewitness identifications. The testimony satisfied the test for admission under People v. LeGrand, 8 N.Y.3d 449, 835 N.Y.S.2d 523, 867 N.E.2d 374 (2007) because it would have aided the jury in reaching a verdict, given that there were three factors present in defendant's case which the scientific community considers affect the accuracy of eyewitness identifications. The majority's holding that the confession and the video footage corroborated the victim's identification strays from our holding in LeGrand and gives courts license to make findings of fact before evidence has been tested through the adversarial process.

Since defendant asserted his innocence of the crime and presented evidence of how his face and build did not match the victim's description, preclusion of the experts’ testimony was not harmless. Defendant is entitled to a new trial where the jury can assess the evidence and his defense with the aid of *498 the false confession and misidentification expert evidence. I dissent.

I.

Defendant Howard Powell was charged with first-degree robbery for separate robberies of complainants HT and EY. The facts underlying the robberies are not disputed, only whether defendant was the perpetrator of the crimes. Defendant's confessions were central to the prosecution's case, and in turn, the circumstances of the interrogation under which those confessions were extracted were central to defendant's claim of innocence. Defendant moved to suppress his statements to the police and the victims’ pretrial identifications of defendant. The majority's description of the conflicting accounts of the interrogation provided by the investigating detectives and defendant at the hearing lacks sufficient context and detail. I write to clarify and highlight those parts of the record that shed proper light on defendant's legal challenges.

According to Detective Grinder, defendant was arrested at 2:00 p.m. on March 1, 2010, for unrelated crimes of trespass and drug possession; at the time, he was in possession of crack cocaine and a “crack pipe.” Another detective testified that, around 6:30 p.m., defendant was taken to an interrogation room at the precinct, where his arm was handcuffed to the wall. ***252 According to the detective, defendant was “aggravated” and “angry,” his “eyes were crossed,” the “tone of his voice was loud,” and he was cursing. Defendant waived his Miranda rights, and when asked about a recent spate of neighborhood robberies, defendant denied any involvement.

At around midnight, Detective Grinder transported defendant to Queens Central Booking. Shortly beforehand, the detective retrieved prescription medication that defendant told the officers he needed. The detective could not recall whether defendant informed him why he needed the medication, and he was unable to remember whether he allowed defendant to take any of it. Other testimony established that an officer must fill out a form when a prescription is given to a person in custody. Detective Grinder did not fill out the form and denied knowledge of this policy.

Defendant was returned to the precinct at 9:30 a.m. the following morning and placed in the interrogation room, handcuffed. Defendant was calm and, according to Detective Grinder, he appeared “somewhat jovial.” Defendant was given *499 a bagel and coffee. The detective informed defendant **1045 of “some of the charges that were being leveled against him” and that he would be put in a lineup. Defendant explained that he had been “very messed up on drugs,” but when he was not forthcoming about the robberies, he was left alone in the interrogation room with pen and paper. Defendant wrote the following statement at around 10:00 a.m.:

“Howard Powell, 

“Was using drugs for two weeks and was doing really bad. I did a few robberiey [sic] and I am very sorry please forgive me I want to help N.Y.P.D. but only you can help me with this, was really MESSED UP on DRUgS [sic] and I am sorry if Ive [sic] hurt any one.”

In separate lineups, EY and HT identified defendant. At approximately 1:00 p.m., Detective Grinder told defendant that both complainants had identified him. Defendant then confessed to the robberies of EY and HT as well as another two robberies, neither of which the police could verify. According to the detective, when asked to write a statement, defendant explained that “he didn't ... have the capacity or he wasn't a good writer,” so the detective typed up two separate police reports, one relating to each robbery, describing defendant's accounts.² The entirety of the robbery descriptions are contained on the first page of each respective report. The EY report states, in relevant part, that defendant followed an Asian female:

“from the 40 side of 10 Street into her building. She was carrying groceries. I went into the building and then into the elevator with her where I robbed her. This one gave me a fight for my money and tried to push me away. I kept pushing her but I didn't try to hurt her. I usually keep a razor knife in my pocket but didn't think I had time to get it out. The only thing I got from her was an EBT card with no numbers on it.”

The HT report states that defendant also followed a Latina:

*500 “from 12 or 10 street into a building. I tried not to be too close to her and when I saw what building she was going to I called to her and asked her to hold the door for me. I got into the elevator with her and asked her what floor she was ***253 going to, and pressed five when she answered. I robbed her with a steak knife by the time we reached the third floor and at the fifth floor I told her to ride it back down and I ran out the stairs.”

The final paragraph of each report is identical and substantially similar to defendant's handwritten statement:

“I was using drugs for two weeks and was doing really bad. I did a few robberies and I am very sorry. Please forgive me. I want to help the NYPD but only you can help me with this. I was really messed up on drugs and I am sorry if I've hurt anyone.”

Defendant's signature is on the second, blank page of these forms.

Defendant was provided some fast food that afternoon. At 8:30 p.m., defendant was transported back to Queens Central Booking. The detective testified that he knew defendant had been in possession of drugs, but at no time after defendant was picked up did defendant say he was experiencing withdrawal symptoms or ask to go to the hospital.

**1046 Defendant testified that he did not voluntarily confess to the crimes and described an interrogation that materially differed from the detectives’ version. Defendant stated that he had been using crack cocaine and heroin on the day he was arrested. At the precinct, he was scared, depressed, and paranoid, and asked for his medicine, which he had not taken in over a day. While in the interrogation room, handcuffed to a “pole,” defendant “went into convulsions.” He had a seizure, urinated on himself, and came to consciousness partially on the floor, partially in a chair. When Detective Grinder discovered defendant, he hit him on the head for urinating on the floor and hit him on the head an additional four or five times later that evening. After the seizure, defendant repeated his request for his medication. Defendant explained to the detective that he takes medication and needed it because he “wasn't feeling well.” He also testified that he was afraid he might die if he suffered another seizure. In response, the detective told defendant that *501 “he wouldn't help me unless I helped him .... He said if you cooperate I will think about going to get your medication for you.” When defendant asked to be taken to the hospital, the detective told him that if he went to the hospital, he would not see a judge for four to five days. Although defendant “couldn't think straight,” and he went to the hospital after every previous seizure he had had, he desisted. At around 9:00 p.m., the detective retrieved defendant's medicine and an unsoiled pair of trousers. Upon returning to the precinct, the detective placed the medicine at the end of the table but never gave it to defendant. Defendant testified that he refused to confess because he “really didn't know what kind of statement to give them” and he “was scared to give them a statement to something that I didn't do.” Defendant was transferred to Queens Central Booking, where officers accompanied him to his visit with the medical personnel there. The officers told defendant that if he told the medical personnel that he had had a seizure or that anything else was wrong, he would be brought to the hospital by the officers, who would have to work overtime, and that he “was going to pay for it.” Defendant did not eat at Central Booking because he missed dinner and slept through breakfast.

By the second day of interrogation, it had been over 24 hours since defendant had taken his medication, and he began to feel “very, very uncomfortable.” Defendant described feeling anxious and hungry and that he was experiencing a panic attack. He was worried he would have another ***254 seizure and finally capitulated because he “had to do something to get [his] medication.” Defendant clarified his physical and mental state, testifying, “I wanted my medication so bad I just did anything,” and “just wanted to go home because I know I didn't do anything.” The detective told defendant “if you work with us, we [sic] get you up out of here.” Then the detective told defendant what to write in a statement. Defendant supplied some of the text because he wanted to give the detective “something that he would be pleased with,” without expressly incriminating himself in the two robberies that the detective was investigating. It was only after he confessed that he received his medication and food.

Defendant further testified that the detectives never provided him with Miranda warnings on either day of his questioning. He signed the Miranda form after 6:00 p.m. on the second day *502 only because he had not eaten, he was stressed, and he was without his medication. Defendant provided a statement shortly before he signed the Miranda warnings and testified, “I wrote that [statement] because I was **1047 afraid, I wanted to eat, I wanted my medication.”³

After hearing these conflicting accounts, the court granted defendant's motion in part to the extent of suppressing the typed statement containing defendant's confession to the HT robbery for lack of proper notice, and otherwise denied the motion, including defendant's request to suppress HT's identifications of defendant.

II.

A. The Frye Hearing and the Court's Rejection of Defendant's False Confession Expert Testimony

Defendant's theory of the case was that he was innocent and that the detective coerced a confession by withholding food, medication, and medical care, and used interrogation techniques particularly effective on people with psychiatric and intellectual disabilities. Therefore, he sought to introduce expert testimony at trial on false confessions, specifically on the voluntariness and reliability of confessions provided under certain scenarios present in his case.

First, defendant moved to present psychiatric evidence, including defendant's medical records and the report of a forensic psychologist concerning defendant's psychiatric diagnoses and substance abuse history, and the psychologist's clinical evaluation of defendant. The report recounts defendant's long history of psychiatric treatment, both in adulthood and as a child while in custody in various institutional settings. *503 Likewise, defendant had a history of severe substance abuse that began in early adolescence and continued throughout his adult life. The psychologist also administered a battery of tests that revealed defendant's significant deficits in intellectual and cognitive functioning. The ***255 causes of these deficits—which may have been a function of a lifelong learning disability, defendant's history of substance abuse, potential brain injuries, mental health issues, or a combination of one or more of these factors—could not be determined. However, the psychologist concluded that defendant's chronic mental health issues, poor processing speed, deficits in reality testing, and impaired thinking were “factors that could make him vulnerable to suggestion in a custodial setting.”

Second, defendant moved to introduce the testimony of Allison Redlich, Ph.D. as an expert on false confessions.⁴ Based on Dr. Redlich's review of the hearing testimony, the psychologist's report, police reports, and the scientific literature, she was prepared to opine on the dispositional and situational factors that the relevant scientific community had associated with false confessions. In her report, submitted as part of defendant's motion, Dr. Redlich explained that dispositional factors relate to the “disposition” of the purported confessor, including considerations such as **1048 mental health conditions, impairments, and substance abuse. Situational factors concern the “situational” circumstances of the interrogation itself, such as physical isolation, the length of the interrogation, the use of deceptive tactics, and promises of leniency. In addition to this report, which discussed the science of false confessions generally and in relation to defendant's case specifically, and Dr. Redlich's 22-page curriculum vitae, defendant also submitted in support of his motion the psychologist's report; minutes from the detective's hearing testimony about his interrogation of defendant; defendant's hearing testimony; the three inculpatory statements attributed to defendant; two police reports containing the complainants’ initial accounts of the robberies; a police voucher listing defendant's medications, dated the day after defendant signed the confessions; the police logbook for the day defendant was arrested; and a selection of nine publications as evidence of the scientific consensus supporting the proffered testimony, including the brief for amicus curiae American *504 Psychological Association (“APA”) in Warney v. State, 16 N.Y.3d 428, 922 N.Y.S.2d 865, 947 N.E.2d 639 (2011).^5,6

***256 Initially, the court granted defendant's request to introduce this expert testimony with the proviso that any limitations on the scope of such testimony would be determined at trial. Thereafter, a different judge assigned to the case ruled that a Frye hearing was required to “establish the applicability, parameter, and relevance of this [expert] testimony.”

At the Frye hearing, Dr. Redlich testified that she was a tenured professor and Executive Director of the Hindelang Center at the University at Albany's School of Criminal Justice. Over the course of 15 years studying police interrogations, she had authored numerous peer-reviewed articles, literature reviews, and book chapters, presented over 100 lectures on the subject, and, to her knowledge, “review[ed] all of the literature” of the field. Dr. Redlich sat on the editorial boards of Law and Human Behavior; Psychology, Public Policy, and Law; Behavioral Sciences of Terrorism and Political Aggression: Journal of the Society for Terrorism Research; and the Albany Law Review for its annual **1049 issue, Miscarriages of Justice. She was the principal investigator on two surveys and the co-principal researcher on a study sponsored by the Federal Bureau of *505 Investigation's High Value Detainee Interrogation Group, which “studies the effectiveness of interrogation approaches and techniques” and commissions studies to fill research gaps (High-Value Detainee Interrogation Group, FBI, https://www.fbi.gov/about/leadership-and-structure/national-security-branch/high-value-detainee-interrogation-group [last accessed Oct. 19, 2021]). One study required that she review, code, and analyze video-recorded police interrogations from police departments across the United States. She also was the recipient of research grants from the MacArthur Foundation, the National Institute of Justice, and the National Science Foundation. She also worked for six years at a private research firm studying the intersection of mental health and the criminal justice system. She had been qualified as an expert in all seven previous cases where she was called as a witness, and she had previously been found competent to testify in a New York local criminal court, before the defendant took a plea deal. She was precluded from testifying in some other cases because those courts had determined that, at that time—approximately ten years prior to the hearing—“the science [was] not sufficiently proven.”

According to Dr. Redlich, 15-20% of proven wrongful convictions involve a false confession. She explained that the number was likely higher because most verified false confessions are recorded after DNA exonerations, which comprise only ten percent of cases and are mainly related to murder and rape trials. On cross-examination, she acknowledged that she did not know the number of false confessions in the “entire population of confessions.”

Dr. Redlich explained that, as relevant here, a “coerced compliant confession” is the result of a person's will being overborne in the interrogation room either from “a variety of techniques or because of [the person's] dispositional characteristics.” In coerced compliant confession cases, people give false confessions “in order to escape the situation” of interrogation. She explained that people with mental health issues or “cognitive limitations” are more likely to give a false confession because they are more vulnerable and less capable of withstanding “psychologically oriented interrogation techniques.” Dr. ***257 Redlich testified that people with mental illness and cognitive disabilities “could be more prone to confusion, they have concentration problems,” and are less able to weigh short- and long-term benefits. She admitted, however, that the evidence was “not entirely clear” with respect to whether any one mental *506 health issue makes a false confession more likely, but explained that evidence suggests that depression and anxiety are associated with false confessions. Dr. Redlich also discussed the steps of the Reid Technique, a psychologically-based interrogation method, including theme development, implied promises of leniency, and minimization, and outlined factors common to confirmed cases of false confession, including prolonged interrogations, no new evidence resulting from the confession, and the use of false promises. She testified that the Reid Technique is the foundation of modern interrogation techniques and that “people who study interrogation techniques are very familiar with” it. Dr. Redlich explained that the Reid Technique marked a turning point in the 1960s, when the “third degree” and physical means of interrogation were “outlawed” and it became necessary for law enforcement to develop new methods of interrogation.⁷ Dr. Redlich explained that **1050 other interrogation methods are “all very similar to the [Reid] technique” because “[t]hey rely on psychological techniques. They rely on principles of social influence.” On cross-examination, she confirmed that she did not know whether the New York City Police Department trained its officers on the Reid Technique specifically. However, Dr. Redlich reiterated that, based upon her own research and her knowledge of the literature, “[a]ll interrogation methods are like [Reid]” because “[a]ll interrogations are psychologically oriented using principles of ... psychological influence or social influence.”

Dr. Redlich also testified regarding research she had conducted. In one study, she interviewed inmates at the Santa Clara Jail, some of whom had histories of mental health issues, *507 about their experiences with police interrogations. Some inmates self-reported that they had falsely confessed to crimes. Although on cross-examination she acknowledged that she had not attempted to corroborate these self-reported false confessions, she also testified that self-reporting is a generally accepted social research methodology. She additionally testified about two other studies, each representing a paradigm for laboratory research in the field. In the Alt Key Study, researchers instructed test subjects not to touch the alt key on a keyboard, or else the computer would crash. Then, without touching the key, the computer would crash. One hundred percent of test subjects ***258 confessed to touching the alt key when accused of doing so—“kind of akin to presenting false evidence or lying to suspects.” The Cheating Paradigm, Dr. Redlich testified, entails a group of students engaged in independent study. A student who is in on the study suggests to another student that they cheat together. Forty-one percent of subjects who refused the suggestion nonetheless signed a false confession.

On cross-examination, Dr. Redlich explained that both studies represented the early stages of laboratory research in the field, and that they had faced some criticism regarding their methodologies and their applicability to criminal cases. Although the Alt Key Study was no longer employed, the Cheating Paradigm was still used in the field. Dr. Redlich noted that, in general, “there are limitations to every single scientific method that you employ,” including laboratory studies, which sometimes have issues with external validity, “but that's why you employ many different methods.”

Apart from her individual research, Dr. Redlich was a co-author of an American Psychology-Law Society (“AP-LS”) White Paper on false confessions research.⁸ Dr. **1051 Redlich explained that “in the 45-year history of the American Psychology-Law Society there have been two white paper scientific consensus papers,” and that the one that she had co-authored was the second. Dr. Redlich explained that

“in order for ... white papers to come about, there has to be a sufficient body of evidence that has accrued *508 and amassed[.] ... [I]n my mind[,] to general acceptance, it really decides the issues because these scientific white papers represent the American Psychology-Law Society. They don't represent the author's opinions. They have to be sponsored by the organization. They have to agree in the first place that there is a sufficient body [of evidence]. Then once it's written, they have to approve it. The entire organization has the chance to view it and weigh in on it, and ... then it goes through a peer review process again, and it will be published.”

Dr. Redlich testified that she was a researcher and not a clinician, and that she had never personally examined defendant. She explained that her testimony about dispositional factors was based on defendant's documented mental-health diagnoses and the psychologist's examination. The purpose of her testimony was to inform the jury about the general science of false confessions and to educate the jury about the dispositional and situational risk factors that create an increased risk of false confession. Dr. Redlich noted that there had been “several studies looking at the weight of confession[s] in jurors’ minds,” which found that “confessions are very weighty.” She noted that researchers and practitioners often call confessions “the gold standard of evidence.” She also testified that “false confessions are very counterintuitive” because it is “very hard for someone to understand that ... people would admit to committing a crime that they didn't commit.” She pointed to research finding that “very few people will admit that they would ever give a false confession themselves.”

The Assistant District Attorney did not call an expert or introduce evidence at the ***259 hearing. Nevertheless, in a post-hearing submission, the Assistant District Attorney argued that Dr. Redlich's proposed expert testimony was neither relevant to defendant's case under Bedessie nor beyond the ken of the average juror. The Assistant District Attorney further argued that Dr. Redlich's proposed testimony was not generally accepted in the relevant scientific community and that it would be within the court's discretion to preclude this testimony in its entirety.

The court denied that branch of the motion which was to permit Dr. Redlich to testify(see 53 Misc. 3d 171, 38 N.Y.S.3d 374 [Sup. Ct., Queens County 2014]). Although the court found that Dr. Redlich was an expert in the field of false confessions, the court *509 concluded that she did not establish “that expert testimony regarding false confessions is readily acceptable in the scientific community” (id. at 178, 38 N.Y.S.3d 374). Also, Dr. Redlich “did not convince th[e] court that an expert's testimony on false confessions is scientifically reliable” (id.). The court held that “[i]t is insufficient under Frye that researchers agree that the phenomenon of false confessions exist and that interrogation tactics will likely increase the risk of law enforcement obtaining a false confession” (id.) According to the court, “researchers ... differ on their opinions as to why and as to which tactics present a danger of obtaining these false confessions” (id.). Moreover, citing Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), the court concluded that Dr. Redlich “failed to establish that **1052 her expertise is generally accepted in the scientific community” and whether “there was a known or potential rate of error in her methods of research” (53 Misc. 3d at 179, 38 N.Y.S.3d 374). The court also faulted Dr. Redlich for her lack of “personal knowledge of the circumstances under which this particular defendant confessed” and noted that there was no proof that the police used the Reid Technique (id.).⁹ Moreover, Dr. Redlich failed to convince the court that the Technique induces or produces false confessions (see id.). The court clarified that counsel could call a qualified expert with personal knowledge of the defendant and the circumstances of the interrogation that may have affected whether defendant falsely confessed (see id. at 180, 38 N.Y.S.3d 374).

The day before trial for the EY robbery,¹⁰ again on the topic of the false confession expert testimony, the same judge stated that, although Dr. Redlich could not assure him “that science properly does exist that could be an aid to the jury,” counsel could call “a psychiatrist or some psychologist properly qualified to testify about [defendant's] medical conditions, and how *510 that could lead to a false confession in this case.” Counsel stated he could not find a psychologist who could present both the clinical evaluation and the research-based testimony as required by the court.

***260 B. The Court's Rejection of Defendant's Witness Identification Expert, the Trial, and the Plea

In addition to false-confession expert testimony, defendant also sought to present testimony by an expert on factors affecting witness identification. Defendant argued that the testimony was central to his misidentification defense because the robber and victims were strangers and of a different race, the robber was armed, and the identification was in part based on the highly stressful interactions during the robbery. Defendant proffered the testimony of Nancy Franklin, Ph.D.,¹¹ to opine on the factors affecting the reliability of eyewitness memory identification, including, among other topics, general principles of how memory works, the effect of high stress on memory of faces, weapon focus, and cross-race effect or own-race bias. In **1053 the alternative, defendant sought a Frye hearing. The court denied the motion on the ground that the testimony was unnecessary because surveillance videos and defendant's statement to the police corroborated the victim's identification.

At trial, EY described the robbery, testified to her pretrial identification of defendant, and identified him in court as the robber. EY also testified that she did not report the assailant's facial features when she called 911 after the robbery. The Assistant District Attorney presented surveillance video from the elevator where the robbery occurred, as well as from the deli, where a person had attempted to use EY's stolen EBT card. In the deli video, the alleged robber is wearing a driving cap, which, combined with the camera angled downward from above and the poor image quality, obscures a clear view of the person's face. In the elevator footage, the person has their coat hood *511 pulled over their head, which nearly totally obstructs the view of their face. Over defendant's objection, the court admitted the elevator video into evidence and admitted the deli video even though the court acknowledged that the video images do not clearly depict the robber's face—“That's correct, you can't actually see the face of the person.”¹² The deli owner confirmed the events depicted in the video and testified that the person who attempted to use the EBT card was a regular customer, but he did not identify the defendant as that person in court.

The detective's testimony was similar to that presented at the pre-trial hearing. Defendant's handwritten statement and the typed police report containing the confession to the EY robbery were admitted into evidence.

Defendant testified on his own behalf, denying that he had voluntarily and knowingly confessed to the crimes. In further support of his innocence and misidentification defense, he stated that he had 12 ***261 missing teeth—including three front teeth—just as he did the day of the crime. The record further reveals that at the time of his arrest, defendant was 51 years-old, six-foot four-inches tall, and weighed 200 pounds. Defendant also presented his medical records and evidence from his optometrist that he wore glasses and would have difficulty reading without them. On cross-examination, he admitted that during the interrogation he denied his involvement in two other robberies for which he was not charged.

As requested by defendant, the court instructed the jury that it could consider the confessions if the jury found them to be voluntary beyond a reasonable doubt. The court also gave an expanded eyewitness charge, including a cross-racial identification instruction (see CJI2d[NY] Statements [Admissions, Confessions]—Custodial Statements; CJI2d[NY] Statements [Admissions, Confessions]—Traditional Involuntariness; CJI2d[NY] Identification—One Witness]). During deliberations, the jury requested to see the videos. Thereafter, the jury returned a guilty verdict on the sole count of first-degree robbery.

Subsequently, defendant pleaded guilty to the severed count of first-degree robbery of HT, with a promise from the court that the sentences on the robbery counts would run concurrently. *512 The court and the prosecutor also agreed that the plea would be vacated if the trial conviction were reversed. The court sentenced defendant as a persistent felony offender to **1054 concurrent prison terms of 25 years to life on the trial conviction and 20 years to life on the plea.

The Appellate Division affirmed both judgments, concluding that “defendant failed to establish that his proffered expert testimony was relevant to the specific circumstances of this case” (166 A.D.3d 660, 661, 87 N.Y.S.3d 31 [2d Dept. 2018]). The court did not reach defendant's remaining claims. As I discuss, it was reversible error to deny defendant's requests to present expert testimony on false confessions and eyewitness misidentification.

III.

Under the standards set forth in Bedessie and Frye, defendant should have been allowed to present Dr. Redlich's expert testimony on factors associated with false confessions, which he alleged were present in his case. Contrary to the District Attorney's arguments and the majority's conclusion that Dr. Redlich would not have provided useful testimony for the jury, her proffer established that the phenomenon of false confessions was generally accepted in the relevant scientific community, as were the dispositional and situational factors associated with false confessions. Additionally, the court should have permitted expert testimony on factors impacting the accuracy of eyewitness identification. The victim's identification was not sufficiently corroborated: defendant's confession was contested and could not serve as corroboration, and the video evidence did not clearly show the alleged robber. The court abused its discretion as a matter of law in precluding the testimony of the proffered experts. Because the case turned on eyewitness testimony and defendant's confession, the preclusion was not harmless as it denied his ability to present a defense. Defendant was entitled to have the jury determine the validity of the confessions and the reliability of the eyewitness identification with the assistance of expert testimony. Therefore, I would reverse the Appellate Division's affirmance of the trial conviction. And because, as defendant argues, and the District Attorney concedes, the court informed defendant that the plea would be vacated if the trial conviction was reversed, ***262 I would also reverse the Appellate Division's affirmance of the plea conviction.

*513 A. False Confession Expert Testimony

The admissibility of expert testimony is a matter of discretion entrusted to the court (see De Long v. Erie County, 60 N.Y.2d 296, 307, 469 N.Y.S.2d 611, 457 N.E.2d 717 [1983], citing Selkowitz v. Nassau County, 45 N.Y.2d 97, 102, 408 N.Y.S.2d 10, 379 N.E.2d 1140 [1978]). Thus, a court's decision to preclude expert testimony after a Frye hearing is reviewed for abuse of discretion (see LeGrand, 8 N.Y.3d at 452, 835 N.Y.S.2d 523, 867 N.E.2d 374). “Expert testimony is admissible if the analysis involved is beyond the ken of the typical juror and the results would be relevant to an issue in the case” (People v. Allweiss, 48 N.Y.2d 40, 50, 421 N.Y.S.2d 341, 396 N.E.2d 735 [1979]). “[I]n recognition that expert testimony ... may involve novel scientific theories and techniques, a trial court may need to determine whether the proffered expert testimony is generally accepted by the relevant scientific community” (People v. Lee, 96 N.Y.2d 157, 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 [2001]). New York has adopted the test under Frye v. United States, 293 F. 1013, which asks “whether the accepted techniques, when properly performed, generate results accepted as reliable within the scientific community generally” (Parker v. Mobil Oil Corp., 7 N.Y.3d 434, 446, 824 N.Y.S.2d 584, 857 N.E.2d 1114 [2006], quoting People v. Wesley, 83 N.Y.2d 417, 422, 611 N.Y.S.2d 97, 633 N.E.2d 451 [1994]). “While the Frye test turns on acceptance by the relevant scientific community, we have never insisted that the particular procedure be ‘unanimously indorsed’ by scientists rather than **1055 ‘generally acceptable as reliable’ ” (Cornell v. 360 W. 51st St. Realty, 22 N.Y.3d 762, 780, 986 N.Y.S.2d 389, 9 N.E.3d 884 [2014], quoting Wesley, 83 N.Y.2d at 423, 611 N.Y.S.2d 97, 633 N.E.2d 451).

This Court's Frye jurisprudence is clear that the only matter for consideration at a Frye hearing is the general acceptance of the science presented. Critically, “[t]he Frye inquiry is separate and distinct from the admissibility question applied to all evidence—whether there is a proper foundation—to determine whether the accepted methods were appropriately employed in a particular case” (Parker, 7 N.Y.3d at 447, 824 N.Y.S.2d 584, 857 N.E.2d 1114, citing Wesley, 83 N.Y.2d at 422, 611 N.Y.S.2d 97, 633 N.E.2d 451, and People v. Wernick, 89 N.Y.2d 111, 115-116, 651 N.Y.S.2d 392, 674 N.E.2d 322 [1996]). The Court has explained that “matters going to trial foundation or the weight of the evidence” are “not properly addressed in the pretrial Frye proceeding” (Wesley, 83 N.Y.2d at 426, 611 N.Y.S.2d 97, 633 N.E.2d 451). Rather, questions pertaining to foundation should be addressed by a motion in limine (see People v. Brooks, 31 N.Y.3d 939, 941, 73 N.Y.S.3d 110, 96 N.E.3d 206 [2018]), and questions of weight and credibility are reserved for the factfinder. Likewise, whether “the scientific analysis must ‘fit’ the facts of the case” is a distinct question from general acceptance (1 Kenneth S. Broun et al., McCormick on Evidence § 203.3 [2020]; see also David H. Kaye et al., The New Wigmore: Expert Evidence § 8.3.1 [c] [3] [2021]).

*514 Here, the Frye court inappropriately relied on the United States Supreme Court's decision in Daubert, 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469, which announced the federal standard. Frye’s general acceptance test is different from the multi-factor validity and reliability standard of Daubert, adopted in rule 702 of the Federal Rules of Evidence. In Daubert, the Supreme Court announced a test of “scientific validity,” which focuses on the reliability and relevancy of evidence (see ***263 Expert Evidence §§ 8.3.1 [b], 8.3.2). Daubert explained that the federal approach is intended to be more flexible than the Frye standard and places more responsibility on the judge (see 509 U.S. at 588-589, 113 S.Ct. 2786). Under the federal test, general acceptance in the scientific community is but one factor for a court to consider. More pointedly, Daubert’s “emphasis is different from Frye’s and its impact depends on how the concepts of validity and fit are applied” (Expert Evidence § 8.3.2 [footnote omitted]). The Supreme Court clarified Daubert’s holding as to the “fit” of testimony in Gen. Elec. Co. v. Joiner, 522 U.S. 136, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997). In rejecting respondent's argument that, under Daubert, “the ‘focus, of course, must be solely on principles and methodology, not on the conclusions that they generate,’ ” the Court explained that “conclusions and methodology are not entirely distinct from one another” (id. at 146, 118 S.Ct. 512, quoting Daubert, 509 U.S. at 595, 113 S.Ct. 2786). Thus, under the federal standard, in determining whether expert testimony should be admitted, “[a] court may conclude that there is simply too great an analytical gap between the data and the opinion proffered” (id.).¹³ **1056 In contrast, under Frye, a court determines only whether the scientific community has generally accepted the results as reliable, not the appropriateness and soundness of the scientific methodology. “The general acceptance test operate[s] as a surrogate for reliability” (Daniel J. Capra, The Daubert Puzzle, 32 Ga L Rev 699, 703 [1998]), while questions of “fit” might go to the foundation and scope of testimony (see Brooks, 31 N.Y.3d at 941, 73 N.Y.S.3d 110, 96 N.E.3d 206; Wesley, 83 N.Y.2d at 426, 611 N.Y.S.2d 97, 633 N.E.2d 451) or to the weight of the evidence, questions New York entrusts to the trier of fact.

*515 The majority, like the Frye court below, misconstrues the applicable legal standards (see majority op. at 494–495, 162 N.Y.S.3d at 248–50, 182 N.E.3d at 1041–42, citing United States v. Redlightning, 624 F.3d 1090 [9th Cir. 2010] [“The district court identified the correct legal standard for determining the admissibility of expert testimony: Federal Rule of Evidence 702 and the Supreme Court's decision in Daubert, 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469”], cert denied 563 US 1026 [2011]).¹⁴ Instead of determining solely whether Dr. Redlich's testimony and the additional documentary evidence established acceptance within the scientific community of the dispositional and situational factors that lead to false confessions, the majority assesses matters of foundation and fit, which would be appropriate under Daubert, but not under our Frye standard. The majority also goes further, and like ***264 the Frye court, usurps the jury's role by weighing the evidence and assessing whether Dr. Redlich was credible.

“False confessions that precipitate a wrongful conviction manifestly harm the defendant, the crime victim, society and the criminal justice system” (Bedessie, 19 N.Y.3d at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380). In Bedessie, this Court noted that the phenomenon of false confessions has “moved from the realm of startling hypothesis into that of common knowledge, if not conventional wisdom” (id. at 156, 947 N.Y.S.2d 357, 970 N.E.2d 380), and thus held that, “in a proper case expert testimony on ... false confessions should be admitted” (id. at 149, 947 N.Y.S.2d 357, 970 N.E.2d 380). The Court recognized not only that false confessions occur, but that there was research purporting to show situational and dispositional factors associated with false confessions, and that those situational and dispositional factors are beyond the ken of the average juror (see id. at 159, 947 N.Y.S.2d 357, 970 N.E.2d 380). However, the Court concluded that none of the factors proffered by the expert were present in the defendant's case (see id. at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380). Thus, no Frye hearing was required and it was not an abuse of discretion to preclude the testimony (see id.). Nevertheless, the Court emphasized that “there is no doubt that experts in such disciplines as psychiatry and psychology or the social sciences may offer valuable testimony to educate a jury about those factors of personality and situation that the relevant scientific *516 community considers to be associated with false confessions” (id. at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380). In contrast, as the majority recognizes, “the expert may not testify as to whether a particular **1057 defendant's confession was or was not reliable, [and] the expert's proffer must be relevant to the defendant and interrogation before the court” (id.; accord majority op. at 479, 162 N.Y.S.3d at 238, 182 N.E.3d at 1031 [“(T)he expert may not render an opinion as to the truthfulness or falsity of the confession”]). Thus, under Bedessie, a court should permit expert testimony on the dispositional and situational factors generally accepted within the scientific community as associated with false confessions where the factors are suggested by the circumstances of the defendant's case (see Bedessie, 19 N.Y.3d at 159-161, 947 N.Y.S.2d 357, 970 N.E.2d 380).¹⁵

The majority does not seriously dispute that the field of false confessions research had gained general acceptance in the relevant scientific community at the time of the Frye hearing. Instead, the majority concludes that Dr. Redlich's testimony would not have assisted the jury in reaching a verdict based on its evaluation of the voluntariness and reliability of defendant's confession because Dr. Redlich had “difficulty in linking her research on the possible causes of false confessions to the case at hand” (majority op. at 491, 162 N.Y.S.3d at 246, 182 N.E.3d at 1039).¹⁶ In doing so, ***265 the majority conflates the “beyond the ken” analysis with Frye’s focus on general acceptance. The majority faults Dr. Redlich for “not explain[ing] how her testimony was at all relevant to the circumstances presented by defendant's interrogation” (id. at 18; see also id. at 24 [“(T)he scientific principles involve more complexity than the general conclusion that false confessions do occur, and the expert is supposed to articulate those principles so a jury can apply the information to the actual evidence in the case”]). Dr. Redlich can hardly be at fault here; indeed, she attempted to testify as to the facts of defendants’ *517 case, but was repeatedly stopped from doing so by the prosecution's persistent objections, most of which were properly sustained by the Frye court (see Bedessie, 19 N.Y.3d at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380; majority op. at 479, 162 N.Y.S.3d at 238, 182 N.E.3d at 1031). In any event, it was not Dr. Redlich's role at the Frye hearing to apply the science to the facts of the case; “matters going to trial foundation or the weight of the evidence” are “not properly addressed in the pretrial Frye proceeding” (Wesley, 83 N.Y.2d at 426, 611 N.Y.S.2d 97, 633 N.E.2d 451).¹⁷

**1058 The majority also implies that the jurors in this case had sufficient knowledge of false confessions such that Dr. Redlich's testimony was unnecessary. The majority points to voir dire in this case, and notes that most jurors “were generally familiar with” the concept of false confessions, including one juror who cited the Central Park Five case (majority op. at 490 n. 10, 162 N.Y.S.3d at 245–46 n. 10, 182 N.E.3d at 1038 n. 10). As Dr. Redlich testified, confessions are the “gold standard,” the strongest pieces of evidence that a prosecutor can present at trial.¹⁸ As she also ***266 noted, even the average juror who is generally aware of false confessions does not understand the situational *518 or dispositional factors that might lead to one (see Henkel et al. at 560-563, 570; see also Mindthoff et al. at 442 [noting that notwithstanding increasing awareness of false **1059 confessions, jurors still view confessions as strong evidence of guilt and “still generally believe that they themselves are relatively unlikely to falsely confess”]). Indeed, in the Central Park Five case, the supervising prosecutor—who certainly has general knowledge of false confessions—continues to maintain that the prosecution was not improper and that the confessions were not coerced (see Linda Fairstein, Netflix's False Story of the Central Park Five, Wall Street J, June 11, 2019, § A at 19, available at https://www.wsj.com/articles/netflixs-false-story-of-the-central-park-five-11560207823). If even an experienced prosecutor adheres to the reliability of a confession in the face of evidence to the contrary, we cannot expect that the average juror will comprehend the dynamic factors that lead to false confessions.

That expectation is borne out by research, which confirms that even potential jurors who know that false confessions occur believe that having expert testimony would be helpful to their deliberative process (see e.g. Mindthoff et al. at 440 *519 [“78.3% of the overall (study) sample agreed (that) ... having an expert testify about how and why false confessions occur would be useful for making a verdict decision in a disputed confession case”]). Here, Dr. Redlich would have aided the jury by testifying to dispositional and situational factors that the scientific community had found were associated with false confessions. Her expert opinion as to those factors and their general acceptance was not speculative but based on her scholastic and professional bona fides, including her tenured professorship and extensive publishing of book chapters and in peer-reviewed journals, as well as her participation in high-profile studies and professional organizations and her comprehensive knowledge ***267 of the scientific literature in the field of false confessions and precipitating factors.

That testimony is precisely the evidence the Bedessie Court anticipated would be admissible. First, the Frye court concluded that Dr. Redlich was an expert in the field of false confession studies. Second, as established by her testimony and the scientific studies admitted at the hearing, there was agreement in the scientific community on the phenomenon of false confessions as well as the dispositional and situational factors associated with false confessions (see e.g. Redlich et al., Comparing True and False Confessions; Redlich et al., Self-Reported False Confessions; Costanzo et al.). Indeed, Dr. Redlich referenced the AP-LS White Paper, which she co-authored, and which demonstrated that an overwhelming scientific consensus existed on the topic sufficient to warrant the AP-LS’ imprimatur; it was only the second white paper in the 45-year history of the AP-LS.

Dr. Redlich's testimony would have assisted the jury in determining whether to believe defendant's version of the interrogation and whether defendant falsely confessed based on the circumstances as he described them. Dr. Redlich discussed modern psychologically-based interrogation methods based on the Reid Technique, including, as relevant here, the role of theme development, implied promises of leniency, and minimization. The Frye court's conclusion that this testimony was irrelevant because Dr. Redlich could not testify that the interrogators or even the NYPD employed the Reid Technique, and because the court was not convinced that the Reid Technique produced false confessions, is unpersuasive. Dr. Redlich testified to interrogation tactics that defendant claimed he was subjected to, including withholding of food and medicine, *520 lengthy interrogation, and promises that Detective Grinder would “help” defendant if defendant “helped” him. Whether those tactics were labeled the Reid Technique is **1060 irrelevant. This myopic focus on the Reid Technique also ignores Dr. Redlich's testimony that there was a general consensus that all modern forms of interrogation are based, at least in part, on the Reid Technique and similar methods.

Moreover, the White Paper discloses that there is a “strong consensus” that individuals with cognitive impairments or psychological disorders are particularly susceptible to false confession under pressure (White Paper at 30; see also Redlich et al., Self-Reported False Confessions [discussing false confessions among people with mental health problems]; Redlich et al., Comparing True and False Confessions [comparing true and false confessions among people with serious mental illness]). The White Paper also concludes that a voluminous body of research supports that people make choices to “maximize their well-being given the constraints they face” (White Paper at 15). With respect to responses to interrogation, the cost-benefit analysis tends to favor outcomes that are immediate rather than delayed (see id.).

The majority calls Dr. Redlich's testimony “speculative,” “broad, unmoored,” and “imprecise” (majority op. at 494–495, 162 N.Y.S.3d at 248–50, 182 N.E.3d at 1041–42 ) because she also testified to various other situational factors, including methods consistent with the Reid Technique, such as sleep deprivation and the use of false evidence, that were not present in this case (see id. at 492, 162 N.Y.S.3d at 247, 182 N.E.3d at 1039–40). The majority misunderstands Dr. Redlich's testimony and the purpose of the Frye hearing. Dr. Redlich's testimony was “broad” because she was providing the court with sufficient ***268 information to evaluate whether the science of false confessions is generally accepted in the scientific community. Dr. Redlich was not suggesting that all of the factors she discussed were present in defendant's case, nor was she suggesting that she would testify as to those factors before a jury. That Dr. Redlich was overinclusive points to her thoroughness and professional acumen, not to her alleged inability to remain “moored” to the topic at hand or “precise” in her analysis. Moreover, that some of the situational factors Dr. Redlich referred to were not present in defendant's case goes to the scope of the testimony, not acceptance within the scientific community of those factors. The proper action *521 here was for the court to limit the testimony to the factors present in the case.¹⁹

The majority concludes that, because there was an overnight break in questioning and the written confession occurred less than four hours after his return to the precinct on the second day, the interrogation was not lengthy or coercive (see id. at 20-21). The majority again misunderstands the distinction between fact and law, jury and judge. It is beyond the ken of the jury to understand that, even under those conditions, the interrogation could be coercive for someone with defendant's mental and physical conditions. That Dr. Redlich described uninterrupted interrogations somewhat longer in length again goes to the weight of the evidence, not whether the situational factor of the duration of the interrogation is recognized by the scientific community. Indeed, after weighing the evidence, the jury might have disagreed with Dr. Redlich's expert opinion that even an interrogation that has been broken up into discrete periods can prove coercive, just as the jury might have disagreed with Dr. Redlich that Detective Grinder used methods similar to the Reid Technique. But, then again, the jury might have found Dr. Redlich's testimony persuasive. At a minimum, the testimony would have been helpful to the jury in assessing the evidence. Put another way, it is firmly within the **1061 jury's province to evaluate the evidence—in this case, the conflicting accounts of defendant and the police, the written confession, and the typewritten confession—and expert testimony should be admitted when it will aid the jury in that role. As the Court has repeatedly emphasized, the admissibility of expert testimony is determined first at a Frye hearing. Questions of relevance and foundation should be resolved separately, generally by pretrial motions in limine or through objections at trial (see Brooks, 31 N.Y.3d at 941, 73 N.Y.S.3d 110, 96 N.E.3d 206; Parker, 7 N.Y.3d at 447, 824 N.Y.S.2d 584, 857 N.E.2d 1114; Wesley, 83 N.Y.2d at 426, 611 N.Y.S.2d 97, 633 N.E.2d 451).²⁰

*522 The majority unpersuasively reasons that defendant's confessions to another two unconfirmed robberies undermined Dr. Redlich's expert conclusion that defendant's ***269 case exhibited a “classic element associated with false confession cases,” specifically that defendant did not provide “any ‘new information unbeknownst to the police’ ” (majority op. at 483 n. 5, 162 N.Y.S.3d at 241 n. 5, 182 N.E.3d at 1033 n. 5). Even if that were true, determining that this fact undermines the expert's conclusion requires evaluating and weighing the evidence, which is the province of the jury, not the Frye court. However, it can hardly be said that defendant's confession to two apparently non-existent robberies is “new information.” Indeed, such a “confession” points towards all of the confessions being false.

The majority also contends that Dr. Redlich's opinions were unreliable because she was “imprecise” in setting out the factors that were relevant to defendant's case (id. at 23). The record belies this argument. It is uncontested that defendant has an intellectual impairment and that such cognitive impairments were generally accepted as relevant dispositional factors within the relevant scientific community (see e.g. White Paper at 20-21; Kassin & Gudjonsson at 53 [“People who are intellectually impaired are also disproportionately represented in databases of actual false confessions”]).²¹

**1062 Further, and contrary to the majority's assertion, Dr. Redlich did not testify that it is “not entirely clear” whether mental illness is a dispositional factor (majority op. at 494, 162 N.Y.S.3d at 248–49, 182 N.E.3d at 1041–42). She expressly stated that mental illness, in general, is a well-known factor *523 because people with mental illness are “more prone to confusion” and “have concentration problems” (see e.g. Redlich et al., Comparing True and False Confessions at 5 [“Mental impairment is a commonly recognized risk factor for true and false confessions”]). The White Paper referenced by Dr. Redlich and submitted in support of defendant's request to present this expert testimony holds the same view (see White Paper at 21-22). Dr. Redlich also candidly acknowledged that the “evidence is not entirely clear” as to the effects of specific mental illnesses but that there is “some evidence that people with depression and symptoms of anxiety” are more likely to falsely confess (see e.g. Gisli H. Gudjonsson et al., Custodial Interrogation, False Confession and Individual Differences: A National Study Among Icelandic Youth, 41 Personality & Individual Differences 49, 55-57 [2006]). Dr. Redlich's alleged equivocation as to mental illness should have been grist for cross-examination at trial. Indeed, such a cross-examination might have effectively undermined the weight of her opinion with ***270 the jury. But it could not serve as a basis to conclude that she was not qualified to testify on the matter. Again, the majority substitutes its own weight and credibility determinations for what should have been a determination by the finders of fact.

The majority also points to additional factors that Dr. Redlich did not explicitly address in her testimony and that might have undermined defendant's claim that his confession was coerced, including that the bulk of false confessions discussed in one study cited in the White Paper were by people who were 25 years old or younger and that defendant had a prior criminal history (see majority op. at 494, 162 N.Y.S.3d at 248–49, 182 N.E.3d at 1041–42, citing White Paper at 5). It was not defendant's burden to address each and every dispositional factor that could conceivably affect the likelihood of a false confession. As with the mental illness factor, the prosecution could have cross-examined Dr. Redlich, or proffered the testimony of a rebuttal expert, to explain why defendant's prior criminal history and age undermined his claim of a false confession.

Dr. Redlich made clear that research methodologies used in the field—such as self-reporting studies—are generally accepted. That there is “no prevalence rate” for false confessions, as the majority notes (id.), is irrelevant. Not every field of scientific inquiry can produce prevalence rates with certainty, nor is a “prevalence rate” necessary to determine scientific validity (see APA Dictionary of Psychology, validity [https://dictionary.apa.org/validity] *524 [“Validity has multiple forms, depending on the research question and on the particular type of inference being made”]). Likewise, the Frye court's conclusion that social sciences methodologies lack an error rate might be relevant under Daubert (see e.g. United States v. Begay, 497 F. Supp. 3d 1025, 1076-1077 [D.N.M. 2020] [finding, under Daubert standard, that false confession research has an “unacceptably high rate of error”]), but not under Frye.

If all this were not enough to lay bare the deficiencies in the majority's analysis, the assertion that Dr. Redlich “expressed significant uncertainty as to the applicability of laboratory-like studies to real-life custodial interrogation” (majority op. at 494, 162 N.Y.S.3d at 249, 182 N.E.3d at 1041) reveals a further misunderstanding **1063 of the Frye standard. Presumably, the majority is referring to the Frye court's conclusion that Dr. Redlich's Alt Key and Cheating Paradigm studies were unpersuasive. The limitations on these earlier studies did not undermine the evidence presented on dispositional and situational factors generally recognized by the scientific community. The early studies are an example of the evolution of the science, not evidence that the identified factors have been rejected as factors present in false confessions. Indeed, they are paradigms in laboratory research, not the culmination of the current scientific literature. Moreover, as Dr. Redlich testified, while laboratory studies might not be generalizable to the field in all circumstances, researchers engage a variety of methods to ensure that they can produce generalizable results (see generally Alan Bryman, Social Research Methods 41 [4th ed 2012] [“(C)hoices of research strategy, design, or method have to be dovetailed with the specific research question being investigated”]). In any case, the Court has emphasized that some disagreement in the relevant scientific community does not itself support preclusion of the expert testimony (see Cornell, 22 N.Y.3d at 780, 986 N.Y.S.2d 389, 9 N.E.3d 884; Wesley, 83 N.Y.2d at 423, 611 N.Y.S.2d 97, 633 N.E.2d 451; People v. Middleton, 54 N.Y.2d 42, 49, 444 N.Y.S.2d 581, 429 N.E.2d 100 [1981]).

***271 In sum, the evidence at the Frye hearing established Dr. Redlich's undisputed credentials and standing in the field of false confessions research, including on situational and dispositional factors. The evidence also established that, at the time of the hearing, the phenomenon of false confessions was generally accepted in the relevant scientific community; there was scientific consensus in the literature on the situational and dispositional factors that increase the likelihood of false confession; the general public did not fully understand the factors *525 that result in a false confession; and several of those factors were present in defendant's case. The Frye court abused its discretion in denying the motion to present such expert testimony to the jury, and the Appellate Division erroneously concluded that the testimony was not relevant to the circumstances of defendant's case. The majority takes a different approach, decoupling and cabining each of the situational and dispositional factors and explaining why each, on its own, might not be good enough evidence of a false confession. That is the job of the jury, not this Court.

Preclusion of Dr. Redlich's testimony was not harmless beyond a reasonable doubt (see People v. Foster-Bey, 35 N.Y.3d 959, 961, 124 N.Y.S.3d 591, 147 N.E.3d 1129 [2020] [applying harmless error in the Frye context]). The evidence was not overwhelming. Indeed, defendant presented strong evidence of his innocence based on the differences in his age and his physical and facial appearance compared to the victim's description of the robber. He also presented medical evidence of his mental and physical condition. The confession was critical to the prosecution, and powerful evidence of defendant's guilt, even if, and perhaps especially if, the jury doubted the accuracy of the eyewitness identification. Defendant disputed the detective's description of the interrogation and defendant's statements. Thus, the expert testimony would have assisted the jury in reaching a verdict by providing specialized knowledge on false confessions and the factors associated with that phenomenon as they applied to defendant's case. Under the circumstances, and by any measure, it cannot be said that the error was harmless (see generally People v. Crimmins, 36 N.Y.2d 230, 237-238, 367 N.Y.S.2d 213, 326 N.E.2d 787 [1975]; People v. Rouse, 34 N.Y.3d 269, 281, 117 N.Y.S.3d 634, 140 N.E.3d 957 [2019]).

**1064 B. Eyewitness Identification Expert Testimony

The trial court also erred in denying defendant's request to present expert testimony on cross-racial eyewitness identification. In LeGrand, this Court held that “it is clear that expert testimony regarding the factors that affect the accuracy of eyewitness identifications, in the appropriate case, may be admissible in the exercise of a court's discretion. Moreover, there are cases in which it would be an abuse of a court's discretion to exclude expert testimony on the reliability of eyewitness identifications” (8 N.Y.3d at 456, 835 N.Y.S.2d 523, 867 N.E.2d 374).

Recently, in People v. Boone, we reaffirmed the fact that “[m]istaken eyewitness identifications are the single greatest *526 cause of wrongful convictions in this country, responsible for more wrongful convictions than all other causes combined” (30 N.Y.3d 521, 527, 69 N.Y.S.3d 215, 91 N.E.3d 1194 [2017] [cleaned up]). We explained that the “cross-race effect” contributes to misidentifications to a troubling degree:

“Social scientists have found that the likelihood of misidentification is higher when an identification is cross-racial. Generally, people have significantly greater difficulty accurately identifying ***272 members of other races than members of their own race. According to a meta-analysis of 39 psychological studies of the phenomenon, participants were ‘1.56 times more likely to falsely identify a novel other-race face when compared with performance on own-race faces’ ” (id. at 528, 69 N.Y.S.3d 215, 91 N.E.3d 1194, quoting Christian A. Meissner & John C. Brigham, Thirty Years of Investigating the Own-Race Bias in Memory for Faces: A Meta-Analytic Review, 7 Psych Pub Pol'y & L 3, 15 [2001]).

The Court has announced a broad concern regarding “the potential for misidentification when a person observes an assailant—usually a stranger—for the first time in a highly stressful environment” (People v. Muhammad, 17 N.Y.3d 532, 546, 935 N.Y.S.2d 526, 959 N.E.2d 463 [2011], citing People v. Santiago, 17 N.Y.3d 661, 934 N.Y.S.2d 746, 958 N.E.2d 874 [2011] [decided on the same day]; see also People v. Abney, 13 N.Y.3d 251, 268-269, 889 N.Y.S.2d 890, 918 N.E.2d 486 [2009] [distinguishing Abney’s companion case, Allen, in which the “defendant was not a stranger” to the robbery victims]). The cross-race effect and an eyewitness’ familiarity with a suspect, as well as the degree of event stress, are all factors that have been identified by this Court as being counterintuitive to a jury's understanding of human behavior and memory (see e.g. Santiago, 17 N.Y.3d at 672, 934 N.Y.S.2d 746, 958 N.E.2d 874; Abney, 13 N.Y.3d at 268, 889 N.Y.S.2d 890, 918 N.E.2d 486; Lee, 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63; c.f. e.g. Norstrand, 35 Misc. 3d 367, 939 N.Y.S.2d 261; People v. Williams, 14 Misc. 3d 571, 830 N.Y.S.2d 452 [Sup. Ct., Kings County 2006]; People v. Radcliffe, 196 Misc. 2d 381, 764 N.Y.S.2d 773 [Sup. Ct. 2003], affd 23 A.D.3d 301, 808 N.Y.S.2d 22 [1st Dept. 2005]). They are also factors generally recognized within the scientific community as potentially affecting eyewitness identification accuracy, which are not well understood by the general public, and therefore the proper subject of expert testimony (see e.g. Jules Epstein, The Great Engine that Couldn't: Science, Mistaken Identity, and the Limits of Cross-Examination, 36 Stetson L Rev 727 [2007]; Richard S. Schmechel et al., *527 Beyond the Ken? Testing Jurors’ Understanding of Eyewitness Reliability, 46 Jurimetrics J 177 [2006]; Charles A. Morgan el al., Accuracy of Eyewitness Memory for Persons Encountered During Exposure to Highly Intense Stress, 27 Intl J L & Psychiatry 265 [2004]; Peter N. **1065 Shapiro & Steven Penrod, Meta-Analysis of Facial Identification Studies, 100 Psych Bull 139 [1986]). Defendant's proffer clearly met the standard articulated in LeGrand. The identification was by a stranger, EY, who was of a different race than defendant and based, in part, on observations of the robber made under stressful conditions involving a weapon.

The trial court denied defendant's request, not because the expert testimony was unreliable or jury instructions would suffice, but because, in the court's assessment, the surveillance video and defendant's confessions corroborated the identification. In LeGrand the Court held that “the admissibility of such evidence would ... depend upon the existence of sufficient corroborating evidence to link defendant to the crime. In the event that sufficient corroborating evidence is found to exist, an exercise of discretion excluding eyewitness expert testimony would not be fatal to a jury verdict convicting defendant” (8 N.Y.3d at 459, 835 N.Y.S.2d 523, 867 N.E.2d 374). Assuming that the expert evidence is otherwise admissible, “it is an abuse of discretion to deny a motion for expert testimony on eyewitness identifications in a case that depends solely on the accuracy of eyewitness testimony if there ***273 is no corroborating evidence connecting the defendant to the commission of the charged crime” (Muhammad, 17 N.Y.3d at 545-546, 935 N.Y.S.2d 526, 959 N.E.2d 463).

Here, the trial court manifestly abused its discretion as a matter of law. Contrary to the majority's conclusion, the surveillance video, while “well-lit” (majority op. at 480, 162 N.Y.S.3d at 238, 182 N.E.3d at 1031), does not clearly depict the robber's face or other dispositive identification characteristics.

To the extent that the District Attorney contends that the jurors were able to compare the video images to their in-person observations of defendant as part of their fact-finding role, this argument ignores the potential effect of cross-racial identification on the jury's deliberations. In other words, like an eyewitness, jurors are subject to the same difficulties in discerning physical characteristics of a person of a different race (see CJI2d[NY] Implicit Bias—Final Instructions [“We all develop and hold unconscious views on many subjects. Some of those unconscious views may come from stereotypes and attitudes about people or groups of people that may impact on a person's *528 thinking and decision-making without that person even knowing it”]).²²

Nor can the defendant's confessions corroborate the eyewitness identification where defendant disputes the voluntariness of those confessions. Indeed, the CPL instructs that “a written or oral confession, admission, or other statement made by a defendant with respect to [the defendant's] participation ... in the offense charged, may not be received in evidence against [the defendant] in a criminal proceeding if such statement was involuntarily made” (CPL 60.45 [1]). And, as I have discussed (see supra at 41), the error was not harmless (see Crimmins, 36 N.Y.2d at 241-242, 367 N.Y.S.2d 213, 326 N.E.2d 787).

The error here is an example of how the case law has strayed from the justification for the corroboration prong of LeGrand. **1066 The majority has missed an opportunity to clarify the law and instead leads us further down a wrong path. Arguably, corroboration is an appropriate factor where there has been a trial and the identification evidence has gone through “the traditional truth-testing devices of the adversary process” (Harris v. New York, 401 U.S. 222, 225, 91 S.Ct. 643, 28 L.Ed.2d 1 [1971]). As the Court clarified in People v. McCullogh:

“The decision to admit or exclude expert testimony concerning factors that affect the reliability of eyewitness identifications rests within the sound discretion of the trial court. When the motion is considered during the People's case-in-chief, the trial court performs this function by weighing the request to introduce such testimony ‘against other relevant factors, such as the centrality of the identification issue and the existence of corroborating evidence’ ” (27 N.Y.3d 1158, 1161, 37 N.Y.S.3d 214, 58 N.E.3d 386 [2016] [citation omitted], quoting Lee, 96 N.Y.2d at 163, 726 N.Y.S.2d 361, 750 N.E.2d 63; see also LeGrand, 8 N.Y.3d at 459, 835 N.Y.S.2d 523, 867 N.E.2d 374).

“Courts reviewing such a determination simply examine *529 whether the trial court ***274 abused its discretion in applying the ‘standard balancing test of prejudice versus probative value’ ” (id. at 1161, 37 N.Y.S.3d 214, 58 N.E.3d 386, quoting People v. Powell, 27 N.Y.3d 523, 531, 35 N.Y.S.3d 675, 55 N.E.3d 435 [2016]). But on a pretrial motion, as here, a court cannot consider whether the identification is corroborated by proffers of hotly-contested evidence that turns on credibility issues and fact determinations because those matters may only be resolved by the jury after the evidence is admitted at trial. Indeed, because there is no evidence for the factfinder to weigh until trial, what appears a proffer of strong evidence of guilt may wither once subjected to cross-examination.²³

The Innocence Project in its amicus brief persuasively argues that research indicates that the notion of corroboration “rests upon an unscientific assumption” that corroborating evidence itself is not subject to what is known as “forensic confirmation bias” (see e.g. Karl Ask et al., The ‘Elasticity’ of Criminal Evidence: A Moderator of Investigator Bias, 22 Applied Cognitive Psych 1245 [2008]). Like all people, forensic analysts exhibit cognitive biases, including confirmation bias, whereby they credit evidence that supports their conclusions, and discount evidence that is **1067 inconsistent with their preconceived beliefs (see e.g. Saul M. Kassin et al., The Forensic Confirmation Bias: Problems, Perspectives, and Proposed Solutions, 2 J Applied Rsch Memory & Cognition 42, 45-48 [2013]; Itiel E. Dror et al., Context Management Toolbox: A Linear Sequential Unmasking [LSU] Approach for Minimizing Cognitive Bias in Forensic Decision Making, 60 J Forensic Sci 1111 [2015] [outlining *530 taxonomy of cognitive biases that affect forensic examiners]; Daniel C. Murrie et al., Perceptions and Estimates of Error Rates in Forensic Science: A Survey of Forensic Analysts, 302 Forensic Sci Intl 109887 [2019] [discussing forensic analysts’ wrongful perceptions of low false positive rates]). Amicus notes that these cognitive biases also infect how courts interpret evidence, creating “an ‘investigative echo chamber’ ” or “bias snowball effect” whereby misinterpreted evidence compounds and affects how courts evaluate corroboration (Kassin et al., The Forensic Confirmation Bias, at 45, 46). In this appeal, for example, if the trial court had assumed that defendant's confession was not false, it would have been predisposed to view the surveillance footage as corroborative ***275 of the eyewitness identification, and likewise for the converse. Thus, even after the adversarial process, a court might be susceptible to the same biases in evaluating whether evidence is corroborative under LeGrand and McCullough.²⁴

C. Defendant's Convictions Must be Reversed

For the reasons I have discussed, the Appellate Division order affirming defendant's conviction, upon a jury verdict, of first-degree robbery should be reversed and a new trial ordered. The order affirming the conviction upon defendant's plea to first-degree robbery of the second victim should also be reversed and the case remitted for further proceedings on the indictment. Prior to entering his plea, the court informed defendant that the sentences would run concurrently and that the court and the People agreed that the plea judgment would be “reversed and restored to its original state” if the trial judgment were reversed. Given the court's representations, the appeal from the guilty plea count is contingent on resolution of the appeal from the trial judgment. As such, because the trial *531 conviction should be reversed, the plea conviction cannot stand (see People v. Williams, 17 N.Y.3d 834, 836, 930 N.Y.S.2d 530, 954 N.E.2d 1155 [2011], citing People v. Fuggazzatto, 62 N.Y.2d 862, 863, 477 N.Y.S.2d 619, 466 N.E.2d 159 [1984]).

IV.

Defendant sought to establish his innocence with the aid of two highly credentialed and respected researchers, who would have informed the jury how various factors that have gained general acceptance **1068 in their respective fields might have contributed to a false confession and a misidentification, thus aiding the jury in its assessment of the evidence. In my view, the court erred in concluding that the science on false confessions was not generally accepted and that the identification was corroborated by the confession and the videos. The error in denying defendant's requests to call both experts was not harmless. We are long past equivocating about the reality of false confessions and eyewitness misidentification, giving a wink and a nod to science all the while affirming convictions where experts are precluded from testifying. It's time to put this research before those who hold a person's liberty in their hands so that they may judge for themselves.

Judges Garcia, Singas and Cannataro concur. Judge Rivera dissents in an opinion, in which Judges Fahey and Wilson concur.

Order affirmed.

This article was prepared by a public defender as a tool to help other defenders challenge false confessions. It describes the Reid Technique, summarizes the science of false confessions, and spells out a complete defense strategy, including introducing expert testimony. Please read pages 1-15 (up to “theory of the case”) and feel free to skim or skip pages 15-19.

This document is a PDF so it is posted on Moodle under this class heading.

ARGUMENT

POINT I
TRIAL COURT'S REFUSAL TO ALLOW EXPERT TESTIMONY ABOUT CHARACTERISTICS OF DAN EVANS DEFENDANT AND SITUATIONAL FACTORS DURING THE INTERROGATION AND THEIR RELATION TO POTENTIAL FALSE CONFESSIONS VIOLATED EVAN'S RIGHT TO A FAIR TRIAL, AND REQUIRES REVERSAL

Dan Evans, who having had almost no contact with the criminal justice system for the entirety of his life, suddenly - in one afternoon -- confessed to two separate crimes that were committed three years apart. Given the interrogation techniques involved and Dan Evans' individual psychological make-up, there is strong basis to believe that the confessions were unreliable or false.

In their interrogation of Evans, the police employed a variety of techniques that scientific research has shown to be highly correlated with eliciting false confessions. Moreover, the police were not interrogating an individual of normal intelligence and a well-adjusted personality. Rather, Evans has severe intellectual and cognitive deficiencies that place him on the borderline of intellectual functioning. Further, extensive psychological testing shows that his personality is highly suggestible and compliant. Well-established scientific research indicates that when a person with Evans' psychological make-up is exposed to the *32 interrogation techniques, which the police freely admit to having used in this case, then any resulting confessions cannot considered reliable.

Unfortunately, the jury was not able to evaluate Evans statements in light of these findings. The trial judge decided that expert testimony on false confessions would not be admitted. As a result, the jury had no way of knowing that the techniques the police used, in light of Evans' intellectual and psychological capacities, were of vital significance. Instead, the jury was continuously exposed to the prosecutor's scientifically unsupportable idea that a false confession was not possible in this case and that the confessions were reliable. Had expert testimony been available to the jury - as required by law - Evans may very well have been found not guilty of the charged crimes, especially since there was very little evidence against him but for his confessions.

A. False confessions experts are crucial in trials in which certain specific indicia of false confessions are present

In People v. Bedessie, the Court of Appeals ruled that expert testimony may be used in trials to educate jurors about false confessions. The court ruled: “[T]here is no doubt that experts in such disciplines as psychiatry and psychology or the social sciences may offer valuable testimony to educate a jury about those factors of personality and situation that the relevant scientific community considers to be associated with false confessions.” 19 NY3d 147, 161 (2012). Specifically, expert testimony would be useful to illuminate the “conditions or characteristics of an *33 interrogation which might induce someone to confess falsely to a crime” and to explain that, “certain types of defendants are more likely to be coerced into giving a false confession - e.g., individuals who are highly compliant or intellectually impaired or suffer from a diagnosable psychiatric disorder, or who are for some other reason psychologically or mentally fragile.” 19 NY3d at 159. The Court further explained that, “just because jurors know about [false confessions] in their experience, doesn't mean they know about psychological studies about reliability of the evidence.” 19 NY3d at 156.

In this case, defense counsel sought to introduce testimony from Dr. Maria Hartwig, an expert on false confessions, to educate the jury on the phenomenon of false confessions and the factors that scientific research has found to be associated with false confessions. Scientific research demonstrates that when certain police practices are used, particularly on uniquely vulnerable suspects, what can result is a “coerced compliant false confession” in which a suspect adopts a false narrative of the crime consistent with what the suspect believes the interrogating police officer wants to hear so that the suspect can end the stressful situation they find themselves in. Christine S. Scott-Hayward, Explaining Juvenile False Confessions: Adolescent Development and Police Interrogation, 31 Law & Psychol Rev 53, 55 [2007]. Some of the specific police practices associated with “coerced compliant false confessions” are: an investigating officer that believes in *34 the suspects guilt prior to an interrogation; isolation of the suspect for a lengthy interrogation; using deception during the interrogation; suggesting to the suspect that they committed the crime accidentally; and implying that some leniency or lack of culpability will result from cooperating with the investigator. Id.; Brandon L. Garrett, The Substance of False Confessions, 62 Stan L Rev 1051, 1063-64 [2010]; Richard A. Leo, Why Interrogation Contamination Occurs, 11 Ohio St J Crim L 193, 211 [2013]. These practices are particularly likely to cause false confessions from suspects who have low intelligence, issues with cognitive functioning, suggestive and compliant personality, and who suffer from certain personality disorders. Christine S. Scott-Hayward, Explaining Juvenile False Confessions: Adolescent Development and Police Interrogation, 31 Law & Psychol Rev 53, 56 [2007]; Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 22 [2010], Saul Kassin, a leading researcher on false confessions, explains: “There is a strong consensus among psychologists, legal scholars, and practitioners that juveniles and individuals with cognitive impairments or psychological disorders are particularly susceptible to false confession under pressure.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 22, 34 [2010].

*35 The need for expert testimony in cases in which these indicia for false confessions is present is particularly acute because the procedures police employ that elicit false confessions are often standard techniques that are, “entirely permissible under the U.S. Constitution and recommended by police training on modern psychological interrogation techniques.” Brandon L. Garrett, The Substance of False Confessions, 62 Stan L Rev 1051, 1063 [2010]; Frances E. Chapman, Coerced Internalized False Confessions and Police Interrogations: The Power of Coercion, 37 Law & Psychol Rev 159, 164-65 [2013]. Therefore, rather than resulting from some kind of police abuse, poor training, or bad-faith, false confessions in these situations arise out of the proper application of the various techniques that are standard practice in American policing. Richard A. Leo, Why Interrogation Contamination Occurs, 11 Ohio St J Crim L 193, 197-98 [2013]; Frances E. Chapman, Coerced Internalized False Confessions and Police Interrogations: The Power of Coercion, 37 Law & Psychol Rev 159, 164-65 [2013].

It is important to note that this research does not suggest that anytime a police officer uses standard police techniques a false confession will occur. Rather, this research reveals that in situations in which many of these techniques converge, particularly in relation to a suspect with a uniquely vulnerable *36 psychological make-up, resulting statements are difficult to deem reliable.5 In the present case, then, expert testimony was essential to helping a jury connect the unique factors present in Evans' interrogation with scientific research that links those factors with false and unreliable confessions.6

B. Denying the false confessions expert was abuse of discretion because specific factors present in this case made such testimony essential

The specific indicia of false confessions that are present in this case are: (1) a defendant with severe intellectual and cognitive deficiencies and a personality that is uniquely suggestible; (2) a lead detective who believes the defendant to be *37 guilty prior to initiating the interrogation and with knowledge of all the key facts needed to contaminate the confession; (3) the employment of the standard police interrogation techniques of confrontation, minimization, deception, and possibility for leniency with cooperation; and (4) an abnormally long, isolating interrogation.

1. Evans intelligence, cognitive abilities, and personality are in line with those that make individuals uniquely vulnerable to false confessions

As explained above, courts find that vulnerability of a defendant is a crucial factor that makes false confessions more likely. In this case, psychologist Dr. Sanford Drob performed an extensive psychological evaluation of Evans and found that, “Mr. Evans exhibits traits that would render him vulnerable to producing a false confession. These traits include borderline intellectual functioning, cognitive, social and emotional immaturity, severe deficits in reality testing and deficits in the capacity to understand the actions and intentions of others, deficits in his capacity to cope with interpersonal stress, anxiety, depression, dependency, passivity and a desire to please others, and a concomitant tendency to rely on others for direction and support.” (Drob Report, 13).

With respect to intelligence, Evans was found to have an IQ of 72, which places him in the low-end of borderline intellectual functioning. (Drob Report, 8, 12). Additional testing of Evans' basic cognitive/neuropsychological processes yielded a performance in the lowest 0.4% of the population with notable deficits in *38 basic language functions, constructional capacity, learning, and memory. (Drob Report, 8). Standard tests of suggestibility and compliance showed Evans to be highly suggestive, willing to provide answers to questions for which he has no information, and a tendency to comply with requests and obey instructions even when they are against his wishes. (Drob Report, 12).

Tests of Evans' personality revealed depressive, dependent, avoidant, schizoid and negative personality traits. These tests indicated strong dependency strivings, desire to lean on others, disillusionment with those he depends on, low self-esteem, tendency to retreat into depression and fantasy, distress over personal failures, sense of humiliation, feelings of unworthiness, uselessness and guilt, and evidence that Evans is timid, shy, anxious and sensitive to rejection. (Drob Report, 10). Moreover, these tests showed that Evans suffers from several personality disorders and syndromes, including dependent personality disorder, general anxiety disorder, and posttraumatic stress disorder. (Drob Report, 10). Rorschach testing of Evans showed significant impairment in reality testing and information processing that was so severe that it potentially predisposes him to psychosis. This can lead to impaired judgment and a failure to anticipate the consequences of his behavior. This testing also suggests that Evans' thinking can become odd and peculiar, especially when he is trying to understand people and their actions. (Drop Report, 11).

*39 Evans vulnerable, dependent and compliant personality was apparent early in his interaction with the police. While hospitalized for the gunshot wound that arose from the 2009 shooting, Evans was visited by Det. Melino who interviewed him about the shooting. During the interview he cried twice, explaining that he was having problems with his girlfriend who was scared and wanted to move out of town, with their child, because of the shooting. (PTH, 14). Specifically, Evans wanted Det, Melino to speak to his girlfriend and “to convince her not to move, you know, to stay supporting him.” (PTH, 14-15). Further, he wanted Det. Melino to speak to his doctor, “because he was upset that he's been there for, he said, three days.” (PTH, 17). It seems that Evans immediately and strangely believed that Det. Melino was a confidante of his who would help him with his personal problems - something that would be easier to understand if one imagined Evans as a child.

Importantly, the intellectual, cognitive, and personality deficiencies that Evans suffers from have all been linked to false confessions. Christine S. Scott-Hayward, Explaining Juvenile False Confessions: Adolescent Development and Police Interrogation, 31 Law & Psychol Rev 53, 56 [2007]; Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 22 [2010]. However, without expert testimony stating this, the jury had no way of knowing that these risk factors were present or relevant.

*40 2. Detective Mooney's single-minded approach to Evans' involvement with the 2006 shooting is an independent risk factor in producing a false confession

Standard police training instructs detectives and officers to, “interrogate only those suspects whose culpability they ‘establish’ on the basis of their initial investigation.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 6 [2010]. As a result, once the interrogation begins its purpose is not to, “discern the truth, determine if the suspect committed the crime, or evaluate his or her denials... the single-minded purpose of interrogation is to elicit incriminating statements, admissions, and perhaps a full confession in an effort to secure the conviction of offenders.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 6 [2010].
However, researchers have discovered that when investigators begin the interrogation process presuming the guilt of the suspect, “this presumption creates expectancies that guide the investigator's behavior” Brian Cutler et. al., Expert Testimony on Interrogation and False Confession, 82 UMKC L Rev 589, 610-11 [2014]. Specifically, it introduces confirmation biases and expectancy effects that predisposes them, “to ask confirmatory questions, use persuasive tactics, and seek confessions.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 6 [2010], citing Hill, Memon, & *41 McGeorge, 2008; Kassin, Goldstein, & Savitsky, 2003). With these biases in place, the investigator, “discounts and manages the suspect's denials, increases the suspect's anxiety, and, in the case of an innocent suspect, fulfills the prophecy by securing a false confession from the suspect.” Brian Cutler et. al., Expert Testimony on Interrogation and False Confession, 82 UMKC L Rev 589, 610-11 [2014]. This phenomenon has been demonstrated in research as well as through case studies of known false confessions. Id. As a result, “Investigator bias,' whereby officers focus on one suspect because they are convinced he or she is guilty,” is an independent risk factor increasing the likelihood of false confessions. Frances E. Chapman, Coerced Internalized False Confessions and Police Interrogations: The Power of Coercion, 37 Law & Psychol Rev 159, 164-65 [2013].

In this case, there is clear evidence that investigators approached Evans' interrogation with a belief in his guilt and a single-minded focus to illicit a confession. . . . All this information suggests that Det. Mooney had developed a belief that Evans was guilty of the 2006 homicide and that after sizing Evans up in the hospital following the 2009 shooting, he developed a plan to arrest Evans for the 2009 shooting so that he could then interrogate him about the 2006 homicide.

3. The police deployment of standard techniques of maximization and minimization, deception, and implied offers of leniency in the context of an isolating and lengthy interrogation are unique indicia of false confessions

The standard police technique utilized by the two officers who interrogated Evans - Det. Ragolia and Det. Mooney - include the following: long and isolating interrogation; use of deception; use of maximization and minimization, and use of implied offers of leniency. These are all techniques that scientific research has revealed to be linked with coerced and unreliable confessions. Specifically, scientific research demonstrates that the cumulative stress that these techniques create leads to a situation where the immediate gain of complying with an officer's *44 wish for a confession outweighs the uncertain long-term consequences of false confession. Christine S. Scott-Hayward, Explaining Juvenile False Confessions: Adolescent Development and Police Interrogation, 31 Law & Psychol Rev 53, 55 [2007], Brandon L. Garrett, The Substance of False Confessions, 62 Stan L Rev 1051, 1063-64 [2010].

a. Isolation and Interrogation Length

Two primary risk factors associated with false confessions are the length of the confession and the isolating manner in which they occur. Empirical studies have shown that most interrogations last around one hour, and that most proven false confessions arose out of interrogations lasting six hours or longer. Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 16 [2010]; Steven A. Drizin & Richard A. Leo, The Problem of False Confessions in the Post-DNA World, 82 NCL Rev 891, 948 [2004]. 

Further, the leading training manual on law enforcement questioning suggests that custodial interrogations should not last for more than four hours, acknowledging that lengthier interrogations are potentially coercive and increase the likelihood of eliciting involuntary compliance from a suspect who is seeking to escape the pressures of extended confinement and accusatory questioning. Fred Inbau, John Reid, Joseph Buckley and Brian Jayne, Criminal Interrogation and Confessions, 4th Ed., Aspen Publishers, p. 206 (2001)). Separately, the technique *45 of isolating suspects in small, private rooms during their interrogations enhances the coercive effects of standard police interrogation techniques and are an independent risk factor in producing false confessions. Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 6-7 [2010].

In this case, Evans' interrogation began at 11:00 a.m. and he was questioned through the day until almost midnight (PHT. 186-7). According to Det. Ragolia, Evans likely made his first incriminating statement sometime between 1:00 pm and 2:00 pm, roughly two to three hours after Evans was placed in the interrogation room. (TT 821, 828, 922). The time is uncertain because according to Det. Ragolia, Evans initial statement indicated that he had nothing to do with the shooting. (TT 814). Then, according to Det. Ragolia, he left the room - saw Det. Mooney and then reentered the room and requested another statement from Evans, which this time contained incriminating content. (TT 820-21). However, the time at which this occurred is not clear. According to Det. Ragolia, he Mirandized Evans and then asked for this second statement at 12:45 pm. However, according to Det. Mooney, he didn't arrive to the station until 1:00 pm and was told around 1:30 pm or 1:45 pm that Evans was now making an incriminating statement. (TT 922-24). Nevertheless, the statement that Evans gave did not implicate him as the responsible party in the crime. Therefore, at around 3:45 pm *46 (TT 925, 937), Det. Mooney initiated another interrogation that lasted until 5:00 or 5:30 pm (TT 831), which yielded a more incriminating statement. So Evans did not provide a statement that was used as his confession until approximately six hours after he was placed in the interrogation room. It's important to note, however, that Evans did not experience this statement as a confession. Based on conversations he had with Det. Ragolia and Det. Mooney, he was under the impression that the statement he made did not incriminate him and that he was free to go home. It wasn't until 7:45 pm, over eight hours after Evans was brought to the interrogation room, that Det. Mooney began questioning Evans about the 2006 homicide (PHT. 169). Further, it is important to note that in the entire time that he was interrogated, Evans did not eat any food except for one bite of a brownie. (PTH 74-75).

That Evans was interrogated for such an extraordinary length of time, and, further, that he was isolated in an interrogation room with barely anything to eat for that entire time, is strongly associated with false confessions. However, without expert testimony on these risk factors, the jury had no way of knowing their significance.

b. Maximization and Minimization

Maximization (also known as confrontation) and minimization is a push-pull process whereby investigators will alternate between confronting a suspect with *47 their absolute certainty of the suspect's guilt and suggesting to the suspect that perhaps there are circumstances that would mitigate or minimize their guilt. The confrontation thrusts “the suspect into despair” while the minimization “makes a confession seem like an expedient means of escape. Interrogators are thus trained to suggest to suspects that their actions were spontaneous, accidental, provoked, peer-pressured, drug-induced, or otherwise justifiable by external factors.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law 8 Hum Behav 3, 18 [2010]. Brian Cutler explains: “While maximization techniques provide the pressure to confess, minimization techniques ease the way toward confession.” Brian Cutler et. al., Expert Testimony on Interrogation and False Confession, 82 UMKC L Rev 589, 608-09 [2014].

In this case, Det. Mooney freely admits that he employed the maximization and minimization techniques prior to any statement by Evans regarding the 2006 homicide:

I explained to him that you are going to get charged with murder as a result of this investigation. I don't know exactly what happened because I wasn't there, but I don't know whether you did it on purpose or by accident. I don't know. I wasn't there. Only way I'm gonna find out is if you tell me the truth about what occurred. I explained to him what mitigating meant. There were - perhaps there were circumstances beyond his control that caused this to happen. (TT, 960).

Mooney further explains: “We had a little bit more of an extended conversation where I again explained to him, I used the word ‘mitigating’ when I *48 was talking to him, and I explained to him what I meant about that. Once I thought that he understood what I was talking about, I said, ‘Just tell me what happened,’ and then he began to give a narrative about what happened.” (PHT, 178-79). Mooney is clear that before he asked Evans for his narrative, he told Evans, “I don't know whether you shot her on purpose or it was accidental. The only way that I'd find out is if you were to tell the truth about it.” (PHT, 231-32). Thus, Mooney suggested to Evans that if it was accidental, the version Evans ultimately adopted, it would not be so bad for him.

Although the interrogating officer in this case freely admits to using the interrogation technique of maximization and minimization, the jury was not allowed to know that scientific research has demonstrated that this technique significantly increases the possibility of a false confession.

c. Deception

The use of deception and presentation of false evidence in interrogations is a particular police technique that research has shown leads innocent people to confess to prohibited acts. Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 17 [2010]. False confessions expert Saul Kassin explains: “Basic psychological research reveals that misinformation renders people vulnerable to manipulation... Scientific evidence for human malleability in the face of misinformation is broad and pervasive. The forensic literature on confessions reinforces and extends this classic point, indicating that *49 presentations of false evidence can lead people to confess to crimes they did not commit.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 17 [2010].

Courts have recognized that such techniques, “particularly when applied to those of immature minds, may cause, with a substantial degree of probability, false confessions.” Singletary v Fischer, 365 F Supp 2d 328, 336-37 (EDNY 2005). See also People v Thomas, 22 NY3d 629, 639 (2014)(detailing how combination of police tactics can result in psychological coercion and render incriminating statements wholly unreliable).

There can be no doubt that that Det. Mooney used this technique as part of his interrogation with Evans. Mooney explained: “Commonly in the course of an interrogation and it's an accepted practice that detectives are allowed to use what's refer[red] to as a ruse in order to have the subject of the interrogation think that I know more than I actually do about what happened.” The purpose of this ruse, Mooney explained, is “to gain cooperation from the subject of the interrogation.” (TT 1064), In this case, Det. Mooney told Evans that several witnesses, when shown a photo array, had identified Evans as the shooter in the 2006 homicide although this was entirely untrue. Further, Mooney told Evans that these people would be coming to view him in a physical lineup the following morning. (TT 956; *50 PHT. 234). Det. Mooney reinforced this ruse by continuously telling Evans that he was certain of Evans' guilt. (TT 960).

Det. Ragolia suggested that he used a similar technique when questioning Evans about the 2009 shooting when he testified, that “I explained to him that we had witnesses that say that he was one of the shooters,” even though all the police had was a single child witness who identified Evans under classic suggestive circumstances. (PTH, 63).

Although both detectives used ruses in their interrogations, the jury was not allowed to find out that such ruses - particularly when utilized against an individual like Evans - are highly correlated with false confessions.

d. Implied Offers of Leniency

Social science researchers understand the technique of minimization to be particularly influential on inducing false confessions because it implies, without any explicit statement, that if a suspect provides a confession they will receive some kind of leniency. Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 18 [2010] Empirical research suggests that techniques that impliedly suggest leniency in exchange for a confession are a primary cause of false confessions. Richard A. Leo et. al., Bringing Reliability Back in: False Confessions and Legal Safeguards in the Twenty-First Century, 2006 Wis L Rev 479, 517 [2006].

*51 In this case, in addition to Det. Mooney's minimizing suggestion and associated implication of leniency - that Evans' involvement was accidental - Det. Mooney also created the impression that he was willing to provide Evans with leniency in exchange for his services as an informant. According to Det. Mooney, early in the interrogation Evans had expressed to Det. Ragolia that he wanted to be an informant. (PHT. 221-22). Mooney further explains: “One of his lines of response to me was that he was interested, he had seen on television where people sometimes would have a problem that get arrested become informants for the police... He thought that he could perhaps provide us with a lot of information on various subjects.” (PHT. 171-72). In response to Evans' offer to be an informant, Mooney told Evans: “Without you telling me the truth about things, there's no point even discussing being an informant at this point.” (PHT. 172). The statement suggests that if Evans provides a narrative in which he is the cause of the 2006 homicide, then Mooney would consider making him an informant with its implied benefits. Given that Mooney told Evans that he firmly believed Evans was guilty and fortified this stated belief with lies about the strength of the case, Evans was trapped into thinking that admitting the crime was the only way to avail himself of the opportunity to be an informant.

Det. Mooney took further action to make it seem as if he was taking Evans' offer to be an informant seriously by showing him - in direct response to his *52 request - photographs of two men who were involved in an unrelated homicide. (PHT. 224-25). Mooney explained that these were not the only photographs he showed Evans during the interrogation, and in fact, that he brought additional photos of unrelated cases for Evans to look at. (PHT. 174-75). Mooney explains: “He had already stated that he was interested in being an informant. These are a lot of open cases. Philip Diaz through all of the other non-fatal shootings. If he could have provided closure on some of those, I certainly would have tried to get that information from him.” (PHT. 239).

It is also important to note that it was at this point in the interrogation that Det. Mooney, for the first time, explained to Evans that he was under arrest and that he was not free to go home. (PHT. 170). Very significantly, it is also at this point, that Det. Mooney admits to having employed the minimization and maximization techniques described above - and it is at this point that Evans asked Det. Mooney if he could be an informant like the ones from TV. Given that Evans has borderline intellectual functioning, his personality is uniquely suggestible, and he is prone to misinterpret the intentions of others, it is understandable how Evans would be so compliant and influenced by these minimization and maximization techniques and why he might actually believe that Det. Mooney was now *53 employing him as an informant.8 For example, while Det. Mooney wrote out a statement for Evans admitting his culpability, he noticed that Mr. Evans, “was looking at those papers that were on the table. The photographs as well as the summaries that were attached to some of them.” (PHT. 186). Given the combination of techniques used, it is very likely that Evans believed that he was being used as an informant and that by stating that the shootings were accidental he would be helping the police while absolving himself of responsibility.9

4. The need for a false confessions expert in this trial was particularly acute as Evans behavior suggests that he was, in fact, producing a false confession that was being actively contaminated by the interrogating officers.

a. Evans' did not understand the severity of his situation and believed that he would be allowed to go home after providing statements that included the details interrogating officers' asked for.

. . . . Det. Mooney explained that even after Evans had given a recorded statement in which he claimed to be responsible for the 2009 shooting, Evans believed that he would be allowed to go home, “to put his baby to bed because his grandmother was baby-sitting.” (PHT. 169). . . . Even the trial judge noted that Evans' did not seem to understand the situation he was in following his initial confession, noting, “I guess it's still in his mind. He didn't think he was in custody, I guess because he's not experienced in the criminal justice system. I'm not exactly sure.” (PTH, 284-85).

Evans' reaction - and his lack of understanding of the severity of his situation - is consistent with innocent individuals with his psychological make-up experiencing the kind of interrogation that he did. Saul Kassin explains: “People with low IQ and high [suggestibility] scores are most vulnerable. They are substantially more likely to think they'll be able to go home after a confession and that they do not need legal advice if they are innocent.” Saul M. Kassin et. al., Police-Induced Confessions: Risk Factors and Recommendations, 34 Law & Hum Behav 3, 21 [2010]. Evans functions intellectually and emotionally much closer to *55 the level of a child. As a result, it would be easy for him to believe that if he simply told the police what they wanted to hear he would not suffer any consequences for it - particularly when the police reinforce this idea by telling him repeatedly that if he says that the shooting was accidental then they would understand, implying that he would not be in trouble.

The fact that even the trial court was perplexed by Evans' reaction to his admission of guilt is a clear indication that expert testimony on false confessions was essential for a jury. Evans' reaction is consistent with false confessions, but one would not think so if relying exclusively on common sense. Expert assistance, then, was necessary to explain how the specific factors present in Evans' interrogation - including his mental make-up - cause false confessions.

b. Based on police testimony Evans' behavior is entirely consistent with a false confession contaminated by police.

False confessions often appear authentic because they contain details consistent with commission of the crime. Richard A. Leo, Why Interrogation Contamination Occurs, 11 Ohio St J Crim L 193, 195-96 [2013]. However, these details can enter a suspect's narrative through a process of contamination where, “Police may, intentionally or not, prompt the suspect on he v the crime happened so that the suspect can then parrot back an accurate-sounding narrative.” Brandon L. Garrett, The Substance of False Confessions, 62 Stan L Rev 1051, 1053 [2010]. Leading false confessions expert Richard Leo explains:

*56 [T]he psychological design of American interrogation methods virtually dictate the contamination once detectives have selected an innocent target for questioning. American interrogation is a laboratory of confirmation bias. Contaminated confessions are the product of what we might call the confirmation bias effect of the assumptions, goals and techniques of guilt-presumptive, accusatory police interrogation. Richard A. Leo, Why Interrogation Contamination Occurs, 11 Ohio St J Crim L 193,211 [2013].

Importantly, however, police contamination of a given suspect's narrative is typically inadvertent and, “investigators do not appear to typically realize when they are contaminating confessions.” Id. Research indicates that even in cases in which it is known that confessions were contaminated, as the accused were subsequently exonerated by DNA evidence, detectives subjectively believed themselves to be assiduously avoiding contaminating the suspects' narratives. Brandon L. Garrett, The Substance of False Confessions, 62 Stan L Rev 1051, 1057 [2010]; Deborah Davis & Richard A. Leo, To Walk in Their Shoes: The Problem of Missing, Misunderstood, and Misrepresented Context in Judging Criminal Confessions, 46 New Eng L Rev 737, 767 [2012]. Detective Jim Trainum recounts a personal experience that demonstrates how unknowing contamination of a suspect's narrative can occur: “We had unintentionally fed her almost the entire case over a several hour period. And Kimberly - the woman - she would guess. She would guess a lot. And sometimes the guesses were right. And we wouldn't see the ones that weren't because ‘She was being evasive.’ ‘She was protecting someone.’ So that's how we wrote that off... But fortunately, we had accidentally let the tape continue to run, and we captured the whole thing on *57 video. But if we hadn't had that video, we never would have been able to go back years later and catch our mistakes.” As told to the Mid-Atlantic Innocence Project's Rachel Cicurel. Available at: http://www.exonerate.org/in-their-words/james-trainum (Last visited August 15, 2014).

In this case the entire interrogation consists of the interrogating detectives specifically asking Evans to change details in his statements so that they fit with the detectives beliefs of what occurred. Det. Mooney was clear that after reading Evans first statement - written by Det. Ragolia - that he wanted to further interrogate Evans because, “there were discrepancies between what he said and what we knew based on witness accounts.” (TT 925, 937, PHT. 145-46). Det. Mooney confronted Evans, telling him that he knew that he was lying about certain details in the written statement. (TT 830). According to Det. Mooney: “When I tried to explain to him what some of the things were that - thought were not true, he seemed to me very easily acquiesced to the idea.” (TT 927-28). The interrogation continued with Det. Mooney supplying Evans with the details he thought were true, and Evans agreeing to include them in his statement, [examples omitted]

These examples demonstrate that the narratives that are attributed to Evans came in direct response to police suggestion. The interrogating detectives are clear that they asked Evans to change details that did not conform to the evidence they had gathered and, in response, Evans complied. In addition, this interrogation featured detectives continuously lying about the case against Evans, implying that if he admits guilt he will be allowed to go home because the events were not his fault, and further insinuating that no harm would come his way because they would be using his services as an informant. Use of these techniques combined with the fact that Evans' mental functioning is closer to that of a compliant child makes it clear that his statements are not reliable. Many if not all details he supplied came from the officers. The lack of a videotape of the interrogation makes it impossible to know the true extent of contamination that occurred - but even based on the *60 detectives' own testimony, it is clear that his narratives were coached and suggested.

c. Based on the presence of so many indicia of false confessions, and evidence that the interrogation actually did induce a suggestible, compliant, confused state in the defendant, the defendant's statements cannot be considered reliable.

It is not in question that Evans is mentally handicapped and has a highly suggestible and compliant personality. It is not in question that Evans was held in an isolated interrogation room for over twelve hours as part of the interrogation that yielded his confessions. It is also not in question that detectives used techniques such as maximization and minimization, ruses, and implied offers of leniency in their interrogation. Further, it is not in question the detectives supplied Evans most of the details that he included in his statements. That all these factors appear together is hugely significant and results in statements that cannot be considered reliable. “Employing standard interrogation methods on mentally handicapped suspects and conducting lengthy interrogations... appear to play a major part in precipitating untrustworthy confessions. When either of these practices are employed, there is a very substantial likelihood that suspects will provide the answers sought by their interrogators, regardless of their own beliefs in the truth of those answers.” Welsh S. White, What Is an Involuntary Confession Now?, 50 Rutgers L Rev 2001, 2042-43 [1998].

. . . . It is only after Evans is exposed to a panoply of police techniques that are known to induce false confessions that Evans makes an admission connected to the 2006 homicide. Given that Evans is functioning at the mental level of a child, with no clear understanding of the severity of his situation, with a uniquely compliant and suggestible personality it is very likely, then, that he was simply providing Det. Mooney with the statement that Det. Mooney wanted to hear so that Evans could go home to his grandmother and baby. At minimum, however, there is no way to understand this statement as reliable given the number of indicia of false confessions that are present.

5. Lack of false confessions expert altered the entire trial by presenting the jury with a skewed world in which social science research about false confessions does not exist.

a. A lack of testimony from a false confessions expert was used by the prosecution to introduce various theories and ideas about false confessions that have no basis in fact.

Scientific research is clear that certain factors strongly increase the likelihood of false confessions. Although many of these factors are present in this case, Evans was not allowed to introduce any of this research as evidence. However, the prosecutor, particularly in their closing argument, introduced various theories about false confessions that have no basis in any scientific research, but rely exclusively on conventional wisdom that is not accurate in this case. These *64 theories stood as uncontested thus creating the impression that the prosecutor's arguments were somehow credible.

i. Length of interrogation
It is well established that the length of an interrogation is a crucial risk factor in false confessions. Nevertheless, because no expert was allowed to testify on this subject, the prosecutor was able to state in her closing summation: “In all, the whole thing took from the minute he gets to the precinct until the end of the second video is twelve hours. Certainly not any unnecessary delay and nothing that would overcome somebody's will.” (TT 1497). The idea that being held in an interrogation room for twelve hours is not a factor that increases the likelihood of false confessions is simply false. As explained above, average interrogations last one hour, and interrogations that last over six hours are highly correlated with false confessions.

ii. Individual characteristics
The prosecutor also challenged scientifically established knowledge about the role that intellectual and cognitive functioning and personality have on the likelihood of a given suspect providing a false confession. While research cited above - including by the Court of Appeals in Bedessie - establishes that individuals with low intellectual and cognitive functioning and suggestible personalities are uniquely vulnerable to providing false confessions, the prosecutor *65 states the following in her summation: “None of these traits would ever suggest that he would falsely confess, if anything, they help to explain why he went to kill a man who he felt had humiliated him, fire indiscriminately at him down a crowded street and then truthfully confessing to it.” (TT 1507-08). The prosecutor explored this theme again suggesting that rather than making defendant more vulnerable to false confession, his low intelligence, “explains why he committed these two crimes in broad daylight with lots of people looking and why he thought he could talk his way out of them and didn't refuse to talk to the police.” (TT 1500-01).

When defense counsel attempted to introduce evidence that such traits do suggest that a person might falsely confess, as an overwhelming amount of scientific research demonstrates, the trial judge refused. (TT 1349). As a result, the jury was presented with a world in which low intelligence actually decreases the likelihood of a false confession - a world exactly opposite to the one we live in.

iii. Minimization and Maximization Techniques

The prosecutor also distorted the significance of the techniques of minimization and maximization, which Det. Mooney freely admitted to having used in his interrogation of Evans, and which extensive research links to the production of false confessions. This research explains that, for example, when an interrogator insists that they know a suspect is guilty, this event induces incredible *66 stress in the suspect, makes them feel “trapped” and “hopeless” irrespective of whether or not they are guilty. Brian Cutler et. al., Expert Testimony on Interrogation and False Confession, 82 UMKC L Rev 589, 608-09 [2014]. Nevertheless, the prosecutor takes the scientifically unsupportable position that, “[c]onfronting [Evans] with the fact that witnesses have identified him is not psychological coercion. No[where] near what it would take an innocent person no matter what their IQ to confess to shooting a gun in a crowded basketball court.” (TT, 1472-73). The prosecutor goes on to say, contrary to expert findings, that only a guilty person would experience such maximization techniques as coercive:
An innocent person being told we're going to put you in a lineup, witnesses from that homicide will see you, is going to say, Great, put me in a lineup, I'm not the right guy, you'll see there is no way I'm going to get picked out. An innocent person wouldn't hear that he was about to be put in a lineup and then confess to a murder.
(TT 1499).

. . . . Of course, these characterizations by the prosecutor demonstrate why expert testimony was so essential in this case. The prosecution is presenting theories *68 about how innocent and guilty people act in this kind of situation. Scientific research demonstrates that these theories are wrong. However, because this scientific research was excluded, the jury was only exposed to those inaccurate accounts, without any corrective information.

b. The prosecutor was allowed to mischaracterize the significance of the videotaped statements while actual expertise on the videotaping was excluded

The prosecutor repeatedly suggests that the videotape of Evans statement shows that he is relaxed, which means there was no coercion in the interrogation. (TT, 1495-96). In cross-examination of Dr. Drob, the prosecutor was allowed to ask: “Isn't it true according to the literature that the videotape is the best evidence of an individual at the time that we're talking about, if we're talking about his competency at that time the video is the best evidence of that, is it not?” (TT 1302). Dr. Drob responded: “It's certainly very good evidence because it's a film of the individual at the time you're considering, yes.” (TT 1302). The prosecutor then proceeded to use Dr. Drob's testimony to suggest that Evans was not anxious in the video and, by implication, was not coerced.

The reality is, unfortunately, that these kinds of ‘final-hour’ tapings, which are commonly made at the end of lengthy interrogations, are not useful measures of what actually occurred during an interrogation. In fact, the highest courts in at least seven states have limited admissibility of recorded statements in such *69 situations and have condemned such final-hour taping of interrogations.  . . .  .

Nevertheless, during the redirect of Dr. Drob, defense counsel was not allowed to ask Dr. Drob about the scientific research related to such final hour tapings: . . . . the videotape is not a valuable measure of Evans likelihood of having confessed falsely. But, the jury was not allowed to be exposed to this reality.

c. Cumulative effect of judge's rulings was to create a fantasy world where scientific evidence about false confessions doesn't exist and that the only source of information about factors that contribute to false confessions was the prosecutor's inaccurate theories

Extensive scientific research suggests that certain factors substantially increase the likelihood of false confessions. Many of these factors were present in this trial. However, the scientific evidence that links these factors with false confessions was systematically excluded. Consequently, although the prosecution's case was based almost entirely on Evans' confessions, Evans was not allowed to illuminate to the jury that various indicia of false confessions were present in his interrogation.

As a result, the prosecutor could credibly state the following in summation:
*72 There is absolutely no shred of evidence presented at this trial, by either side, that the defendant falsely confessed. No one ever said it... There is absolutely no evidence whatsoever that the defendant was coerced to make this statement or any evidence that he had gotten his information from anywhere other than his own memory of the events... You can only base your verdict on evidence and here there was nothing presented to you that in any way suggest a false confession.
(TT 1500)

In other words: the prosecutor was able to capitalize on the judge's erroneous ruling's to create the false impression that nothing in Evans' case suggested that a false confession may have occurred. Yet, extensive evidence does suggest, given the unique factors present in this case, that a false confession was likely.

Defense counsel objected to the prosecutor's summation, arguing: “No one presented evidence that the confession was false. I was prevented from doing that so this reopens the issue about a false confession expert and her exploiting the fact that I wasn't able to put on a false confession expert, so I object to that. I'm just amplifying my grounds and that is also a basis for a mistrial.” (TT 1514). The trial court's only response was “Okeydokey.” (1514).

There is no way that, in light of this situation, Dan Evans received a fair trial. This case featured numerous indicia of false confessions that courts have found require explanation by a false confessions expert to a jury. The jury was denied access to this vitally important scientific research and, thus, was presented *73 with a world in which no such research existed. As a result, the prosecutor was able to blatantly lie about the causes of false confessions and the likely behavior of those subject to the kind of techniques employed by the police in this case. With no opportunity to effectively rebut these erroneous claims, Evans opportunity for a meaningful defense and his right to a fair trial was destroyed.

This is “the first study to provide evidence suggesting that full episodic false memories of committing crime can be generated in a controlled experimental setting.” When reading, consider: would you confess?

Note: You only need to read the highlighted portions (there’s a lot of highlighting at the beginning and less at the end). Beacuse this is a PDF it is posted on Moodle under this class heading.

Please go to our Moodle Page and under "Class 11" you will find the prompt and submission folder for Writing Reflection #11.

Attorneys and Law Firms

[This article is available online here and is pasted in below.]

They say you are the company you keep. Should that apply tenfold for the family you’re born into? Law enforcement certainly thinks so. But if guilt by association still strikes a chord, we should all be concerned with familial DNA searching, because it is guilt by genetic association.

Imagine learning you are the target of a police investigation for a violent crime, not based on suspicion, but because a close relative was convicted of a crime and you share DNA. Familial searching scans the convicted offender database not for full matches, but for close non-matches to crime scene DNA, hoping someone in the database is related to the person of interest. From there, anyone in their family tree is a possible suspect, and many innocent people may wind up targeted by police.

Once you become a viable target (meaning, not an infant or in jail at the time of the crime), the cops may nab your DNA profile from a discarded soda can while you’re not looking. Hopefully, the DNA will establish your innocence (though consider Lukis Anderson, whose DNA was recovered from the fingernails of a murder victim but luckily had an airtight alibi), and hopefully they won’t store it forevermore (though consider the unregulated New York City database, which stores thousands with little oversight).

Then again, the investigation could take a more perilous turn. The police question your neighbor or boss. What if you come to find out your baby brother was convicted of a crime? Or that you have a baby brother? Or your dad isn’t your biological father?

Proponents claim familial searching is a “no harm, no foul” situation and only generates potential leads. If there’s no more evidence connecting a person to a crime, they say, that’s that. Only the guilty should be worried, they say. But “generating leads” is another way of saying they get to poke around with zero probable cause or accountability. And the harms are real.

Not all New Yorkers experience the harms of the criminal justice system the same way. Communities of color, in particular, have known for years the damage done by invasive surveillance and over-policing. Because familial searching utilizes the convicted offender database (including broken windows and quality-of-life violators), the method promises to extend that disgraceful legacy.

This month, an appellate court struck down the use of familial DNA searching in New York, finding the unelected members of the Commission on Forensic Sciences overstepped their authority when they took it upon themselves to approve it in 2017. The court ruled that the decision to use this controversial technique is a matter of policy only legislators can make.

The court also acknowledged the disparate impact this method will inevitably have on communities of color: “The likelihood of being targeted because someone is a person of color (likely a male) is increased, a consideration that can only be evaluated as a matter of social policy, not science.”

Indeed. By trolling the database for potential connections to law-abiding relatives, this technique endorses the rationale that those related to criminals are more likely themselves criminals. What’s worse, the NYPD makes no secret that it agrees. Emanuel Katranakis, the NYPD’s chief of forensics, repeatedly cites a statistic that almost half of all prison inmates have a close relative who had also been in prison, arguing that “this supports the belief in and bolsters the efficacy of familial searching to produce valuable leads.” It turns out this thinking is not just dangerous but scarcely effective. As the appellate court notes, the Department of Justice reports that successful prosecutions based on this technique are rare.

All New Yorkers want to prioritize safety, which requires striking a balance with privacy. But we must remain vigilant to strike that balance the same way for all New Yorkers. Albany lawmakers have considered familial searching over the years, but repeatedly left it on the legislative cutting room floor. It seems like the perfect place for this inequitable policy.

Lewis is staff attorney with the DNA Unit at The Legal Aid Society.

[The article is available online here and is pasted in below.]

If you think clandestine DNA dragnets and secret databases are things of science fiction, I have some upsetting news.

From the same law enforcement playbook that brought you stop-and-frisk comes the latest form of racial profiling and policing: the knock-and-spit. We’ve learned that NYPD officers are willing to knock on doors to take New Yorkers’ DNA, whether by “consent” or at the precinct by offering people a cigarette or drink to collect their spit. And once again, as the city’s primary public defenders, we see this latest law enforcement sweep is happening mainly to New Yorkers of color.

Take as an extreme example the police investigation of the Howard Beach jogger case. Before they identified a suspect, the NYPD collected well over 500 DNA profiles from men in the East New York area. Imagine police knocking on doors, in uniform, with a cheek swab in hand, asking residents to prove they didn’t kill the jogger in the nearby park.

They were willing to do it in East New York. Do you think this would happen on Park Ave.? In Park Slope? Not likely.

But things get worse from there. For those people excluded from the jogger case, the Office of the Chief Medical Examiner, the city’s crime lab, permanently keeps those profiles in their databank and routinely compares profiles to all city crimes.

As of 2017, the OCME had more than 64,000 profiles, stored regardless of whether they were identified as a suspect or cleared. Based on the lab’s claim that state laws are ambiguous and city laws are silent, the lab feels justified to invent their own rules about how this information is used.

And so, those 500 people who “chose” to cooperate with police may never know that they, and possibly their close relatives via familial searching, will be routinely compared to DNA samples from every rape, murder and violent crime in town. In this respect, they will be treated just like someone convicted of a crime.

Maybe none of this bothers you. A common refrain from law enforcement goes, if you don’t commit crimes, you needn’t worry. But innocent people have legitimate cause for concern because DNA has unparalleled power to wrongly accuse. Genetic testing can now reveal a profile based on a few cells, and we shed hundreds of thousands of cells a day. Your DNA can show up in places you’ve never been.

Consider Lukis Anderson, whose DNA was discovered under a murdered billionaire’s fingernails. Luckily, Anderson’s attorney discovered he was dead drunk in the hospital at the time and couldn’t possibly be the killer. How did his DNA get there? The authorities figure the same paramedics treated each man that night and inadvertently transferred Anderson’s DNA to the dead man. Anderson was facing a death sentence for seven months before he was cleared in 2013.

You might also take a cue from the police themselves. Under their labor contract with the city, rank-and-file officers don’t give the lab their DNA, which means the lab can’t easily rule out possible crime-scene contamination. This means that the officers knocking on doors ask people to volunteer to do what they won’t.

Basic privacy is another genuine worry. We see every day how our personal information, once set loose upon the internet, can never be recaptured. The past few years have heralded incredible expansions of the uses of DNA. Now imagine what will be possible in 30 years.

The mayor proudly considers the demise of stop-and-frisk a shining example of the city’s equality in law enforcement. It has been replaced by something arguably worse. We must stop the profiling, illegal storing and rogue searching that target people of color all over again.

Lewis is staff attorney with the DNA Unit at The Legal Aid Society.

To learn more about the demographic composition of DNA databases, we undertook two complementary investigations. First, we submitted requests to every jurisdiction seeking information pursuant to the local freedom of information act or sunshine law.64 Second, we endeavored to “reverseengineer” the database on our own. . . . Before turning to those results, however, we add a few caveats. First, our effort to ascertain the racial and ethnic composition of DNA databases
admittedly relies on a socio-cultural practice (e.g., ascertaining “race”) that genetics in many ways itself belies. Recognizing this limitation, which we discuss further in Part III.B, we nonetheless think it valuable to explore database composition in light of the enduring socio-cultural significance of these categories. Second, there is wide variation in how categories like race and ethnicity are experienced, perceived, and recorded. Such variation presents several problems. Most importantly, the lack of care in how these categories are used or assigned may undermine the ultimate integrity of the data and the labels used may not always map perfectly either within or across groups. These and other areas of contention can lead to confusing or difficult to reconcile data that obscures rather than illuminates the questions at hand, and for this reason
we have elected to streamline and simplify, even at the expense of important nuance. 

. . .

Although white people constitute over three-quarters (77%) of the combined population of the disclosing states, they are only just over a third (39%) of the DNA database. Conversely, people of color make up almost a half of the national DNA database (47%), even though those groups make up only a little over a third (39%) of the population. As expected, persons in the Black, Hispanic, and Native American communities bore the brunt of DNA collection practices.

Our estimation data paints a similar, though still more complicated picture. We show that although white people make up 62% of the US population, they make up only 49% of the DNA database. In comparison, although black people make up only 13.26% of the US population, they make up 34.47% of the DNA database. In other words, DNA has been collected
from black persons at two and a half times the rate of white persons and from Native Americans at one and a half times more than the rate for white persons. Indeed, although people of color bear a disproportionate burden of DNA collection and storage, the burden is particularly concentrated on the black population. We estimate that 2.26% of the black population has their DNA collected per year, whereas only 1.21% of all non-white persons have their DNA collected within that same year.

Please go to our Moodle Page and under "Class 12" you will find the prompt and submission folder for Writing Reflection # 12.

Introduction

Serious concerns pervade the use of forensic evidence in criminal cases. Despite efforts by a dedicated coterie of scholars who have long endeavored to expose fraudulent forensic methods,¹ it perhaps took the DNA-exoneration cases to finally bring the breadth and depth of the problems to light.² Much of the critical attention has centered upon what I have elsewhere termed “first-generation forensic techniques”--methods such as ballistics, handwriting analysis, and tool or bite mark analyses.³

Although the judicial and executive branches have been slow to respond, some incremental changes have finally started to occur.⁴ Scrutiny has revealed that the methodological foundations that purport to legitimate many *490 longstanding methods either do not exist or are woefully inadequate.⁵ While renewed efforts to scientifically validate some techniques-- such as fingerprint typing⁶--may prove effective, other methods have already fallen into disrepute. For instance, after a national expert panel discredited the forensic technique of bullet lead analysis, the Federal Bureau of Investigation (FBI) issued a statement that it would no longer conduct such testing.⁷ Similarly, courts have also recently called into question other long-accepted forensic techniques.⁸

Often omitted from critiques of forensic methods, however, is a robust discussion of nuclear DNA typing. If anything, DNA typing is typically held out as the pinnacle of “good” forensic evidence, in that it exemplifies the kind of scientific rigor that first-generation techniques lack.⁹ After all, DNA analysis emerged from scientific processes, and it is a testable, reproducible, and falsifiable technique.¹⁰ DNA analysts even testify to their findings with expressions of statistical probabilities rather than arbitrary and unsupportable statements of certain identity.¹¹

Without question, this praise is well-deserved. DNA typing represents a marked advance beyond the shamanistic “sciences” of the first generation. Yet the seeming corollary--that DNA typing is therefore an exercise in purely objective, indisputable science--does not hold true. This is not to suggest that DNA has no basis in objective science, or even that it is as subjective as other forensic techniques; comparing most first-generation methods to DNA typing is like comparing astrology to neuroscience. Nevertheless, not unlike neuroscience, the fact that DNA typing is scientifically grounded does not mean that there are not plenty of things that we still do not understand about it, *491 and plenty of instances in which the best conclusions we can draw are nonetheless tentative ones.

Many people know this, but a surprising number of laypersons, and even lawyers, do not.¹² To be clear, I am not saying that DNA typing done poorly entails an exercise of subjective judgment. Rather, DNA typing--done perfectly and precisely according to protocol--still often entails making discretionary calls and choices. But just because DNA typing is not wholly objective does not mean that it is wholly indeterminate--it simply means that it may be more like meteorology than mathematics.

Why is it important to understand this discretionary aspect of DNA typing? Because the perception that DNA typing is an exercise in purely objective testing and observation obscures many of the difficult issues raised by this forensic method. If DNA typing is viewed as a practice that is as objective as algebra, then the only concerns it presents are those of human error. While human error is a serious problem--ranging from innocent laboratory mistakes to more pernicious or structural deficiencies¹³--eliminating human error (even assuming that were possible) would not alter the fundamental fact that most forensic cases will involve some exercise of discretion. And acknowledging this fact is a critical step in any appraisal of the proper scope and application of forensic genetic testing.

The purpose of this Essay is neither to provide a comprehensive explanation of the mechanics of forensic DNA typing nor to offer a checklist for attorneys or judges seeking to test the reliability of such evidence. Other *492 sources have already achieved both of these objectives quite successfully.¹⁴ Indeed, this Essay will not consider the many ways in which DNA typing can go wrong--such as contamination of samples, intentionally malfeasant analysts, or statistical misrepresentations. Nor is it an essay that deals with cutting-edge issues in forensic typing, such as new methods for obtaining results from incredibly small amounts of DNA, or continuing debates about the optimal way of calculating probabilities in certain cases.

Rather, this Essay is about DNA typing done absolutely correctly. It is intended for the forensic science outsider--the person perhaps casually acquainted with forensic techniques--who may even suspect that bite, tool, or handwriting analyses are faulty, but assumes that DNA typing is science in its purest, most objective form. It provides an answer to questions like: why do DNA analysts often demand a suspect’s sample before reaching definitive conclusions about what profiles are contained in the crime scene sample, or why should genetic attribution statements that may be made with a high degree of confidence nevertheless not be confused with those made with absolute certainty?

This Essay proceeds as follows: Part I provides a superficial overview of nuclear DNA typing as the basis for the remainder of the discussion. Part II presents, in as clear and simple terms as possible, a sampling of the kinds of discretionary decisions that analysts often confront when interpreting crime scene samples. Part III wraps up with remarks about current disputes in forensic DNA typing, and how recognition of its inherent subjectivity might inform and illuminate these debates.

1. Fundamentals of DNA Typing

To understand the nature of the subjective decision making required by forensic DNA typing, it is first necessary to have some general notion of how it works. At the outset, it is worth spending a moment to resolve what, for many readers, may be a nagging preliminary contradiction. That is, some may wonder why criticisms of forensic DNA typing used to inculpate suspects should not likewise call into question the legitimacy of DNA typing used to *493 exculpate people. After all, DNA typing has grabbed headlines for its use to exonerate wrongfully convicted persons, and thus the suggestion that it might have shortcomings understandably causes alarm. However, the use of DNA typing to inculpate a person--by which I mean to say that a suspect is the likely source of a sample--fundamentally differs from its use to exculpate.¹⁵ The simplest analogy is to blood typing. Imagine a murder scene at which police find a blood sample certain to belong to the killer. Crime scene technicians test the blood sample and show that it is type O⁺. Later, the police find and draw blood from two suspects. One suspect is type AB; the other is type O⁺. We can, with unreserved confidence, say that the first suspect is not the killer, but regarding the second suspect, we can only say that she is included within the class of people that includes the killer. The probability that she is the actual killer turns on how many other people have that blood type, along with any other evidence that we might be able to adduce.

DNA typing works in a similar fashion. When a genetic profile is generated, it is far easier to determine with confidence those individuals from whom the sample could not have come than to identify with certainty the individual to whom the sample absolutely belongs. The difficulty is in determining the parameters of inclusion--how to define the characteristics and size of the class of persons to whom it may belong. To provide another analogy, one can imagine that a witness glimpses a six-digit license plate and detects some symbols, but not others, which can be written as follows: “??2 3?6” or “??2 3? 8.” That makes a suspect out of all those with plates of “XX2 3X6” and “XX2 3X8” (using “X” to represent the variable), but it does not call into question the certainty with which we can exclude all those with plates of “XX8 4X9.”

Next, it is important to note that the general labels “DNA typing” or “forensic DNA” comprise a variety of techniques and methodologies. Most generally, forensic DNA typing currently consists largely of two different forms--nuclear DNA and mitochondrial DNA (mtDNA). As the name suggests, nuclear DNA typing looks at the strands of DNA found in the nucleus of cells-- specifically those pieces of DNA found on a person’s chromosomes.¹⁶ However, mtDNA typing examines DNA found in the *494 mitochondria-- organelles in the cell outside of the nucleus.¹⁷ The greatest advantage of mtDNA is that it is present in copious quantities and, therefore, may be more resilient to cell degradation.¹⁸ However, because mtDNA typing is both more time-intensive and produces less specific results (in terms of individuating persons), it is typically used only when nuclear DNA typing is not possible.¹⁹ For this reason, and for reasons of clarity, this Essay focuses only on nuclear DNA typing; however, it is worth noting that mtDNA also entails a large number of subjective determinations.²⁰

Lastly, even within the practice of nuclear DNA typing, many different individuation techniques have emerged. When DNA typing first appeared in 1986, a method known as restriction fragment length polymorphism (RFLP) was used to measure the length of certain parts of the DNA strand known as variable number tandem repeats (VNTRs).²¹ Ultimately, the technique that has prevailed and reflects contemporary practice is known as short tandem repeat multiplexing (STR), and thus it is that technique which this Essay discusses.²² The attractiveness of this particular method stems largely from its efficacy; it offers a high degree of discriminatory power at a rapid speed and relatively low cost.²³

1. Basics of STR Typing

How does STR typing work? I have explained that nuclear DNA typing looks at pieces of the genome found on chromosomes within the nucleus of the cell. The whole human genome is incredibly long--if unraveled it would stretch to more than six feet.²⁴ Although examining the entire genome would be an excellent means of individuating persons, it would take a very long time and expend a great deal of resources. It would also be largely unnecessary: *495 over 99% of the genomes of two human beings are identical.²⁵ In fact, human DNA is roughly 98% identical to that of a chimpanzee.²⁶

It is on some of these regions of difference among people, then, that DNA typing focuses. Specifically, STR typing looks to particular regions of the genome where certain known sequences of the four DNA base pairs (GATC) repeat themselves, and then measures how many times those repeats occur.²⁷ These repetitive sections are particularly useful because, presently, they have no known function. That is why they are sometimes called “junk” DNA--because unlike sections (called genes) of the DNA strand that “code” for something and have a purpose (such as determining hair color), these genes are “non-coding.”²⁸

In light of the standard set by the FBI when it put together the national database of genetic profiles, known as CODIS (Combined DNA Index System),²⁹ most crime laboratories look at thirteen different places (or “loci,” the plural of “locus”) where these repeats occur.³⁰ To make matters somewhat more complicated, at each place there are two possibilities for the number of repeats. The reason for this is that individuals have two sets of chromosomes--one from their father and one from their mother--and the number of repeats they have in a particular section might be different on each chromosome.³¹ In the end, a genetic profile is expressed as a list of twenty-six numbers--two numbers for each of the thirteen places.³² Those numbers represent the repeats present at the thirteen specific sites on both chromosomes.³³

For example, looking at a particular locus on the gene strand, an analyst might observe six repeats on one chromosome, and fifteen on another. The *496 person would therefore be a 6, 15 at that locus.³⁴ The repeats at a particular locus are called the alleles.³⁵ Thus, in the example, the person would have a 6 allele and a 15 allele. All twenty-six alleles together--the two alleles from each of the thirteen loci--constitute the forensic DNA profile.³⁶

By crude analogy, you can imagine a DNA profile as a composition of the size measurements of a person. If we measure precisely, we might find a person-- let’s call her Jean--has a 34-inch bust, a 28-inch waist, 38-inch hips, and a 34-inch inseam. We could then think of Jean’s profile as 34x28x38x34. By comparison, Jean’s friend Dana might have a profile of 28x28x32x26. If we had a big pile of clothing, and we knew both Jean’s and Dana’s profiles, then we could probably separate the clothes accordingly.

Similarly, DNA typing allows us to take the measurements of the individual in two respects at thirteen different places and then compare them to the measurements of another individual. Like the example given above, the range of “sizes” at each spot is not limitless--there is a finite degree of possibilities.³⁷ Of course, the variation in body sizes is both less diverse and less precise than the variation in genetic material, but the general idea is the same.³⁸

1. Crime Scene Sample Processing

Where, then, does the subjectivity come in? Shouldn’t a person either possess or not possess a particular allele? Why might such information be the subject of dispute? The simplest answer would be that all of these assumptions are true: testing conducted under idyllic conditions often eliminates a significant degree of the necessary interpretive discretion. But drawing blood *497 for the purpose of DNA testing allows an analyst to exercise a great deal of control that is lacking in the typical forensic context. In a clinical setting, the sample derives from a single source, and it can be quantified and pristinely preserved to optimize the results. And, if anything should go wrong, another sample can easily be taken and tested again.

But forensic DNA testing rarely occurs in such idyllic conditions. Crime scene DNA samples do not come from a single source obtained in immaculate conditions; they are messy assortments of multiple unknown persons, often collected in the most difficult conditions.³⁹ The samples can be of poor quality due to exposure to heat, light, moisture, or other degrading elements. They can be of minimal or insufficient quantity, especially as investigators push DNA testing to its limits and seek profiles from a few cells retrieved from cigarette butts, envelopes, or soda cans.⁴⁰ And most importantly, forensic samples often constitute a mixture of multiple persons, such that it is not clear whose profile is whose, or even how many profiles are in the sample at all. All of these factors make DNA testing in the forensic context far more subjective than simply reporting test results, as the following section elaborates more carefully.

When a crime scene sample arrives at a laboratory, an analyst must check to see if the item contains any usable biological material--most typically blood, semen, or saliva.⁴¹ Once detected, the DNA must be “extracted” from the item, a delicate process that creates a high risk of sample contamination or further degradation.⁴² The analyst then measures the amount of DNA present to select an optimal quantity for testing, and to ensure that the sample is human DNA.⁴³

Once the sample is ready, the first step is to amplify the genetic material so that testing it is easier. Again, to use a crude illustration, imagine you saw something that looked like a termite in your house. You want to know if it was a termite or just an ant, whether there are more termites, and if so, where they are. But it is very hard to figure that out from observing just one termite. If you saw one hundred termites, then you could not only more readily conclude that, in fact, you have a termite problem, but it would also be easier to *498 determine the specific type of termite and where they have nested. DNA amplification, through a process known as polymerase chain reaction (PCR), takes the DNA strand, cuts it, and then creates identical copies of only the part of the DNA strand important for testing.⁴⁴ Thus, it not only eliminates the irrelevant parts of the genetic strand, but it also amplifies, through replication, the relevant bit so it is easier to measure its characteristics.⁴⁵

The actual testing and measuring of the DNA sample is conducted using kits made by commercial companies and a process called electrophoresis.⁴⁶ Detailed explanation of this process is not necessary for the purposes of this Essay, and even a brief description can be surprisingly complex.⁴⁷ What matters for our purposes is that the results of this process are interpreted by computer software that separates out the relevant pieces of information. In this phase, DNA testing appears to the analyst not as a list of numbers, but rather as a kind of graph with peaks and valleys.⁴⁸ Different computer software then assigns values to those peaks and valleys, based on a template of information.⁴⁹ For a single-source sample--that is, a sample known to contain the DNA of only one person like that collected in clinical conditions from prisoners or parolees--the graphical representation is typically a fairly clean and easily interpreted chart, such as Figure 1.

*499 Figure 1

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

When the DNA of more than one person is involved, this entire process can become a bit more complicated. To return to the clothing analogy above, imagine that both Jean and Dana take their clothes to the same dry cleaner. Unfortunately, the cleaner is having a bad day and instead of putting the clothes in the “to clean” bag, he accidentally throws them into the garbage. Fortunately, he realizes his mistake in time and grabs the bag before it leaves the shop. Also fortunately, the cleaner knows the relative sizes of Dana and Jean, and that both of them brought in five different items: a skirt, a blouse, slacks, a jacket, and a sweater.

In trying to straighten things out, the cleaner might first sort through the pile--separating out the trash from the clothes, and then separating out skirts from pants and so on. The cleaner hopes, at this point, to have one of each--and to only have to figure out which belongs to Dana and which to Jean. Assume that the cleaner finishes separating and identifying the relevant clothes, but the cleaner still does not know to whom each article of clothing belongs. The next step he might undertake is to assess the relative sizes of each of the items in each category--distinguishing the larger from the smaller skirt, blouse, slacks, jacket, and sweater.

But our cleaner still is not finished. He has not yet reached his goal of ultimate attribution. He has separated the clothes from the garbage, and even *500 the types of clothes from one another. He has figured out which is the larger and which is the smaller of each kind of item. But recall that he knows both Jean and Dana’s relative sizes. Now he must take what he knows and make inferences about to whom each piece of clothing belongs. For instance, he should be able to surmise that the longer pants belong to Jean and the shorter ones to Dana. Or the tighter fitting blouse is likely to fit only Dana, and thus the looser one is probably Jean’s. And in most cases, he will be right.

This somewhat silly example of the bumbling cleaner mirrors the process of DNA typing. In the first stage, a machine and computer software work together to amplify and then observe the relevant characteristics of the pieces of genetic material passing through the machine.⁵⁰ This is akin to the cleaner looking into the trash bag and distinguishing between both clothes and trash, and between types of clothes, like skirts and pants. In the second stage, the software assigns values to those characteristics based on an internal sizing standard that tells it how to quantify what it detected.⁵¹ This is akin to the cleaner looking at the clothing and determining that the inseam on one pair of pants is about 28? and on another it is about 34?.

The last part--where the cleaner determines that the shorter pants go with Dana and the longer pants with Jean--is done not by a machine, but by the forensic analyst. And although as a general rule such determinations can be made with confidence, they are nonetheless fraught with the potential for erroneous inferences.⁵² The cleaner, for instance, might be wrong in assuming that just because Jean is taller she prefers longer pants. Perhaps the shorter pants belong to Jean, and Dana, in turn, buys her pants long so that she can roll them up.

Still more complicated, imagine that the cleaner accidentally overlooked a third pair of pants that were in the trash bag. Suppose further that this overlooked pair in fact belonged to Jean, and so not only was the cleaner wrong in assuming that the longer pants were Jean’s, but also that there were only two people’s clothing mixed up in the pile. Or maybe he looked carefully, but somehow only one pair of pants was found in the bag--he now knows that he has lost one set of pants, but he is not sure whose they were or how they got lost.

*501 These examples of the mixed-up cleaner help to illustrate how this act of sorting and labeling may in fact be more difficult than it seems at first blush. In much the same way, forensic DNA typing churns out an unsorted mass of information, which must be deciphered by an analyst, and which is not always amenable to only one interpretation. This leads us to the necessary subjectivity of forensic DNA typing.

1. Subjective Determinations in DNA Typing

The job of the DNA analyst, like the mixed-up cleaner, often involves taking the information processed by the computer and attributing it meaning. As in the case of the cleaner, this process relies largely on reasoning abilities, processes of elimination, subjective judgment calls, and inferences; it is not a mathematically certain, objective enterprise. If it was, we would not need DNA analysts at all because there would be no need for interpretation of DNA results. This is not to say that interpretation involves unbounded discretion--a DNA analyst works within a range of assumptions and knowledge that forms the basis of the inferences and conclusions drawn. But DNA interpretation is a discretionary act--more like stepping outside and predicting the afternoon weather than like reciting multiplication tables.

The remainder of this Part gives brief overviews of the various kinds of subjective determinations that an analyst may make in interpreting DNA typing results. In providing the examples below, I want to make clear that I have focused on ordinary, run-of-the-mill issues that arise in forensic DNA typing. I have specifically avoided discussion, for instance, of contamination--either at the crime scene or in the laboratory--even though it constitutes a significant problem.⁵³ I have also avoided discussion of transfer--the phenomenon by which DNA may appear to be present in a place where it was not directly deposited.⁵⁴ For example, my DNA may show up on a towel simply because I shook hands with a person who later grabbed that towel himself.⁵⁵ And lastly, I am not discussing any ongoing methodological *502 disputes or vanguard technologies, such as “low copy number” (LCN) DNA typing.⁵⁶

Rather, my focus in this Part is on forensic samples collected in typical crime scene conditions, and the subjective determinations that often go into interpreting them. Figure 1 illustrated the typical clarity of a single-source forensic DNA result. But for a forensic sample collected from more chaotic conditions and with an unknown number of contributors, the results can be a bit more difficult to interpret. Consider, for instance, the image reproduced below as Figure 2.

Figure 2

TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE

At each locus--identified by the boxed-off information at the top of different sections and typically abbreviated to the first few letters or numbers (for instance, locus “vWA” on the top row)--there are multiple identifiable peaks representing the alleles. For instance, locus vWA shows at least four identifiable peaks with clearly observable alleles--16, 17, 18, and 19. The amount of the genetic material (the quantity, not the length) is reflected by the numbers off to the right. At locus vWA, there is a large amount of material for the 16, 18, and 19 alleles, and much less for the 17 allele. The differences in *503 quantity can be significant for typing purposes; a large amount of material may reflect that one person has two of the same allele at that locus or that multiple persons share the same allele. On the contrary, very little material may indicate any number of things--for instance, it could mean nothing; it could mean that the peak is “spurious” (not a real indication of genetic material); or it could mean simply that fewer persons in the mixture possess that allele than others. That is why, for instance, there is an unlabeled peak to the left of the 16 allele.

This is the essential difficulty in interpreting forensic DNA samples-- determining what to count and what not to count, while also recognizing that the logical inference is not necessarily the only or correct inference. When analysts are given the known suspect’s profile--as opposed to being asked what profiles are possible, given the results they have generated--the risk of erroneous attribution becomes heightened. An analyst may unwittingly fall prey to confirmation bias--seeing in the results what she expects to see, rather than what may or may not be there.⁵⁷ The paragraphs below give additional examples of the ways in which even the most conscientious forensic analyst may make the kind of subjective calls that risk an erroneous interpretation of DNA test results.

1. Deciding Which Peaks Should Count: Peak-Height Thresholds

As explained in the previous section, the first thing an analyst must do is determine what information matters. The peaks and valleys of a forensic DNA sample are not always indisputably clear, but instead require interpretation that labels some as legitimate and others as spurious. The height of the peak (literally, how tall it looks) reflects the amount of genetic material measured at that allele. The amount of measured material can be a function of many different things, some of which are covered in the following sections. Regardless, analysts typically aim for a peak height within an optimal range to ensure that what is observed is genuine genetic material. For instance, the FBI Protocol allows only peaks over 200 relative fluorescence units (RFU) to “be considered conclusive for match purposes,” though it recommends interpretation of any peak over 50 RFU, including for exculpatory purposes.⁵⁸ *504 The makers of commercial equipment for DNA testing recommend a minimum of 150 RFU.⁵⁹ Other laboratories allow thresholds as low as 50 RFU for inculpation.⁶⁰

The importance of peak heights in interpreting DNA results is several-fold. If the threshold is set too low, then many small peaks that do not represent true alleles, but appear for other reasons,⁶¹ will be treated as representing genetic material. Without minimum thresholds, an analyst might think that a peak belongs to the suspect, when in fact it is not from the suspect. Conversely, a very large quantity of genetic material raises risks of a spillover effect, wherein additional peaks that are relics of the excess of material, rather than a reflection of a true allele, are created.⁶² A very large peak might also (correctly or erroneously) suggest that one person has two copies of the same allele at that locus (recall that each locus has genetic material from the father and mother), or that multiple persons share that allele.⁶³ Almost always, a forensic profile does not contain only perfectly formed peaks that reflect true genetic alleles; rather, before making any conclusions, an analyst must first determine which peaks should factor into consideration at all.

1. Choosing How to Interpret Peaks of Different Sizes: Peak-Height Imbalance

When interpreting which peaks are significant, the analyst is typically guided by some rules of thumb. For instance, a genetic profile should have relatively balanced material across the loci. A pristine, single-source sample collected in clinical conditions will often reflect relatively consistent peak heights, with the primary exception being loci where the individual has two copies of the same allele at that locus (known as homozygosity).⁶⁴ In those places, the peak height should be roughly twice the size of the other peaks because twice as much genetic material was observed for that allele.⁶⁵

*505 But perfect balance can be difficult to achieve with crime scene samples. Such samples, as previously noted, often arrive at the crime laboratory in imperfect condition, both in terms of quantity and quality.⁶⁶ For a variety of reasons, some of which are explored below, peak heights can vary dramatically.⁶⁷ As a result, an analyst often must interpret that variation and decide what, if any, significance it carries.

For instance, suppose the analyst knows that the defendant is homozygotic at a locus. The analyst might then interpret a very large peak as evidence of that homozygosity, and dismiss any additional peaks at that locus as spurious. Even if the peak is not extremely high, the analyst might still claim that it represents a “true” allele that matches the defendant, while at the same time dismissing another peak for various reasons. Such determinations would not necessarily be wrong, but they also would not necessarily be right. The key is that these decisions involve the exercise of (hopefully reasoned) discretion, and no analyst can state without reservation which peaks will always, and which will never, count as “true.”

1. Ignoring Some Peaks Altogether: Artifacts and Stochastic Effects

What are some of the reasons, then, that an analyst might ignore or dismiss a peak--even a peak that is taller (and thus suggests a significant quantity of material) than other peaks in the same profile? There are numerous responses, but I will only discuss a few. Notably, at times an analyst may not have any explanation for the presence of a peak, but often the presence of a particular peak may be explained by one of several recognizable phenomena. These phenomena are typically lumped under the label “artifacts”--because they create spurious peaks rather than true alleles, or “stochastic effects”--meaning that they occur as a result of a random variable.⁶⁸ Such effects have very expressive names like stutter,⁶⁹ pull-up or bleed through,⁷⁰ spikes,⁷¹ and blobs.⁷²

*506 Some artifacts are “reproducible,” meaning that repeat testing should generate identical artifacts, while others are not.⁷³ The occurrence of artifacts, however, requires human interpretation of DNA testing results in order to separate true peaks from spurious peaks. Again, to be clear, a good laboratory will have standards and protocols for undertaking this effort,⁷⁴ but even such standards are apt to be mere guidelines--not precise, unalterable rules. Ultimately, the decision to dismiss a peak as an artifact, or label it a true allele, is a discretionary decision left to the analyst.

1. Assigning Meaning to Missing Peaks: Allelic Dropout

At times, the difficulty in interpreting DNA evidence is not the presence of extraneous peaks, but the absence of an expected peak or allele. One way to interpret such an absence would be to identify it as an exclusion; that is, if the suspect has a particular profile that matches the crime scene sample in all but one place, it might simply be assumed that the suspect is then excluded as the source. Commonly, however, analysts will explain the absence of an anticipated or expected allele as a function of “allelic dropout.”⁷⁵

Dropout often occurs when the DNA is not properly amplified so that it can be detected.⁷⁶ This commonly results from degraded or low quantity DNA samples, where some material fails to amplify.⁷⁷ A similar result can also occur as a result of genetic mutations that cause an allele not to amplify properly,⁷⁸ creating what is commonly referred to as a null allele.⁷⁹

The fact that an allele may simply fail to appear--for any number of reasons--obviously causes problems in interpreting DNA typing results. A *507 missing allele could mean any number of things; for example, it could represent grounds for excluding a suspect as a possible source because the suspect, unlike the sample, has a typeable allele at that locus. Or it could simply generate ambiguity in the possible interpretations of the results because it is not known with certainty why the allele did not appear. What matters is that the analyst reading the crime scene profile has the opportunity to ascribe significance to the missing material. It is like evidence that the perpetrator lives at 55 Jones Street; if a suspect in fact lives at 55 Jones Court, but matches the evidence in every other way, we might not be too worried. But if that address is the only evidence, we might not feel comfortable concluding it was just a trivial mistake.

1. Labeling Peaks as Belonging to Particular Profiles: Mixture Deconvolution

In the area of mixture deconvolution, forensic DNA analysts have another major opportunity for interpretive discretion.⁸⁰ This is simply a fancy way of saying that crime scene samples often contain the DNA of multiple persons, and analysts frequently attempt to reach conclusions about the number and characteristics of the various contributors.⁸¹ Recall Figure 2, which contained the DNA typing results of a fictive crime scene stain. The interpreted data does not arrive labeled with the number of contributors or their particular profiles; rather, the technology simply analyzes all the DNA present and identifies the results indiscriminately. It is up to the analyst to try to separate out and ascribe relevant profiles, much like our bumbling cleaner had to do with the mixed up clothing.

Because deciphering mixtures constitutes such an important part of forensic DNA typing, many efforts are underway to improve the available technology.⁸² One recent, significant advance is the capacity to separate male DNA from female DNA, and then focus on amplifying only the male fragment. This technique, known as Y-STR typing, is particularly helpful in rape cases, in which samples are commonly a mixture of DNA (frequently, a male *508 perpetrator and a female victim).⁸³ However, because this form of typing only examines half of the quantity of material as a full DNA profile (i.e., just the male fraction), it is less discriminating and therefore less determinate than a full profile.⁸⁴

As with most aspects of DNA interpretation, the discretion exercised by analysts is not limitless. There are principles and guidelines that instruct analysts on how to separate out a mixture, assign the number of contributors, and perhaps even tentatively identify major or minor contributor profiles.⁸⁵ The FBI Protocol, for instance, permits an analyst to label a major and minor contributor in a mixed specimen sample “if there is a distinct contrast in peak intensities between the alleles, and the alleles contributing the largest peak height values satisfy the conditions of a single source specimen.”⁸⁶

The problem with mixtures is that they exacerbate all of the above issues regarding the interpretation of stutter, pull-ups, spikes, blobs, or dropout. Not only may it be more difficult to distinguish true from false peaks, but even in the cleanest of profiles, contributors can readily mask one another’s contributions.⁸⁷ Studies have shown that roughly 3% of three-person mixtures are easily misidentified as two-person mixtures, and that over 70% of four-person mixtures would be wrongly labeled two- or three-person mixtures.⁸⁸ And even where the number of contributors is known, there may be an extraordinarily large number of ways of interpreting the evident genetic material to create possible individual profiles.⁸⁹

*509 III. Policy Implications

Although the primary purpose of this Essay is to highlight the subjectivity inherent in even the most favorable conditions of forensic DNA typing, it seems only fitting to conclude with some brief meditations on how acknowledging the existence of this discretion might affect policymaking in the area. Forensic DNA testing has developed rapidly and expansively in the past twenty years, and all evidence suggests that it will continue its rapid growth. This is a good thing: DNA typing offers a powerful means of identifying suspects and bolstering the evidence in a large number of otherwise difficult cases. At the same time, difficult questions about the proper use of forensic DNA presently confront our society. Legal and ethical controversies swarm around the proper scope for collecting DNA samples, searching DNA databases, and investigating and prosecuting cases based on the results of DNA typing.

With regard to the collection of DNA samples, legislatures and courts continue to confront concerns about the limits on collecting DNA from various classes of persons, ranging from convicted felons and misdemeanants to simple arrestees. A recent bill signed into law by President Bush, authorizing a committee to set guidelines regarding the preservation of blood samples taken from newborns, has spurred concerns about the quiet creation of a national DNA database.⁹⁰ Complaints have also arisen with regard to law enforcement’s use of informal means of collecting DNA--whether through DNA dragnets in which large swaths of a community are asked to volunteer a sample, surreptitious collection by police of discarded or abandoned items from a suspect, or even outright trickery.⁹¹

With regard to searches of DNA databases, there are two primary debates about the permissible breadth of such searches. The first concerns what are commonly called cold hits--these arise when a database is trawled for a match to a crime scene sample. Cold hits increasingly generate convictions in cases that are many years old and in which little or no additional evidence is adduced. Moreover, there is not yet a consensus as to how to calculate match *510 probabilities in such cases.⁹² Nevertheless, no court has ruled that a conviction cannot stand on a cold hit alone.⁹³

Secondly, and more recently, localities have pushed to use “low stringency” or “familial” search techniques in DNA databases. Such queries seek not the exact profile of a suspect, but rather a similar profile that is likely to correspond to a relative.⁹⁴ Such familial searches purport to help narrow the class of suspects in cases in which the actual perpetrator is not in the database, but relatives or family members are. Apart from the scientific legitimacy of this approach, the legal repercussions of such searches have yet to be fully explored, including whether a familial match should suffice for probable cause, or even reasonable suspicion, to engage in additional intrusions against either the family member or the suspect himself or herself.

It is beyond the scope of this Essay to outline all the arguments in favor of or against certain approaches to DNA collection, databasing, and investigation, much less to offer any specific policy prescriptions. Rather, the modest point made here is that a clear-eyed understanding of the inherent subjectivity and discretion involved in forensic DNA typing is essential to any debate. All too often, the conversation about DNA methods seems to presume an almost mathematical rigor, as though a crime scene DNA test result was as indisputable as the product of two times four.

In contrast, an appreciation of the subjectivity required in forensic typing might enhance the quality of deliberation on all of these topics. It might also enable us to see greater parallels between scientific evidence and other, more traditional forms of evidence, like information from eyewitnesses or informants. For example, we have an intuition in the criminal justice system about the quanta of evidence required before the police may undertake certain actions; searching a home typically requires a warrant and probable cause, and investigatory detentions typically demand at least reasonable, articulable suspicion. In giving content to those terms, we in turn have a notion that *511 certain types of information are more reliable than others--for instance, a known informant’s report of criminal activity is more reliable than that of an anonymous tipster.⁹⁵

The same kinds of questions arise in the context of genetic typing, but in a far less familiar guise. How many loci are necessary to constitute enough probative evidence to justify a search warrant, an arrest, or a conviction? What confidence level must we have in the accuracy of the claimed genetic profile? When are the profiles extracted from a mixture or found in a familial search ones in which we can rely on with reasonable confidence, versus conclusions that give us no more than a hunch or a line along which to pursue further investigation? All too often, those discussing legal regulations simply assume that the results of DNA testing will be indisputably certain and complete, and overlook the possibility that there may be shades of gray or areas of dispute. In short, DNA typing is not an exercise in pure, objective observation. Some DNA typing results are more probative than others.

Similarly, the specter of erroneous attribution, for reasons beyond issues of human error, should loom largely in the minds of policymakers as they approve expansive forms of DNA databasing and searching. In allowing investigations to proceed on the basis of partial matches--such as trawls and familial searches in DNA databases--we must acknowledge the very real concern that we expose certain individuals to the risk of arrest or intrusive investigation on the basis of coincidental, and only coincidental, matches. This is particularly true given that many efforts to expand collection and investigation policies uniquely affect politically disadvantaged racial, ethnic, or socioeconomic groups. In a criminal justice system indisputably fractured upon such demographic lines, the determination to permit familial searches, for instance, disproportionately exposes particular segments of society to intrusion or harassment. When predominantly white, affluent legislators cite scientific certainty as the basis for promoting and adopting intrusive policies shouldered predominantly by minority, low-income individuals, there is cause for reflection and concern. Either our confidence in such forensic techniques should be so great that we should not bristle at universal application, or we should recognize that the fallibility and subjectivity of such methods spark doubts that should dampen the fever for limitless expansion.

*512 Conclusion

DNA typing is a powerful, useful, and revolutionary forensic technique. Many of the benefits of its development have already been realized in the closing of otherwise impossible-to-solve cases and in the exoneration of wrongly convicted individuals. Yet despite its formidable power, it is not a perfect and purely objective science. Good inculpatory DNA methods nonetheless entail significant exercises of discretion on the part of forensic analysts, and whenever there is discretion there is always the heightened risk of error or mistake. As we enter this new era of scientific proof, it behooves us not only to remember that we must be vigilant in revisiting our settled expectations of forensic proof so as to ensure their continued viability, but also that we must display the proper degree of humility and restraint in expressing our confidence in these powerful, but still evolving, methods.

Article available here and below.

Pets at a crime scene may be helpful in gathering key evidence but rarely are they considered for their role in human DNA transfer.

This research considers cats both as receptors and vectors for DNA of a person of interest—key evidence in criminal investigations.

In collaboration with the Victoria Police Forensic Services Department, forensic science researchers Heidi Monkman and Dr. Mariya Goray, from the College of Science and Engineering at Flinders, collected human DNA from 20 pet cats from multiple households.

Detectable levels of DNA were found in 80% of the samples and interpretable profiles that could be linked to a person of interest were generated in 70% of the cats tested.

"Collection of human DNA needs to become very important in crime scene investigations, but there is a lack of data on companion animals such as cats and dogs in their relationship to human DNA transfer," says Monkman.

"These companion animals can be highly relevant in assessing the presence and activities of the inhabitants of the household, or any recent visitors to the scene."

An experienced crime scene investigator Dr. Goray, an expert in DNA transfer, says this data can be very relevant when interpreting forensic DNA results obtained from a crime scene that includes pets.

"This type of data can help us understand the meaning of the DNA results obtained, especially if there is a match to a person of interest.

"Are these DNA finding a result of a criminal activity or could they have been transferred and deposited at the scene via a pet?"

"Further research is required on the transfer, persistence and prevalence of human DNA to and from cats and other pet animals and the influences animal behavioral habits, the DNA shedder status of the owners and many other relevant factors," the researchers say.

To this point, further collaborative work on cats and dogs is currently underway at the Flinders University forensic laboratory.

The article, "Is There Human DNA on Cats?," has been published in Forensic Science International: Genetic Supplement Series.

THE PEOPLE OF THE STATE OF CALIFORNIA DO ENACT AS FOLLOWS:

SECTION 1.

Section 1156.5 is added to the Evidence Code, to read:

1156.5.

(a) Evidence that a person suffered or experienced excited delirium shall not be admitted in any civil action.

(b) A party or witness may describe the factual circumstances surrounding the case, including a person’s demeanor, conduct, and physical and mental condition at issue, including, but not limited to, a person’s state of agitation, excitability, paranoia, extreme aggression, physical violence, and apparent immunity to pain, but shall not describe or diagnose such demeanor, conduct, or condition by use of the term excited delirium, or attribute such demeanor, conduct, or physical and mental condition to that term.

(c) For the purposes of this section, “excited delirium” means a term used to describe a person’s state of agitation, excitability, paranoia, extreme aggression, physical violence, and apparent immunity to pain that is not listed in the most current version of the Diagnostic and Statistical Manual of Mental Disorders, or for which the court finds there is insufficient scientific evidence or diagnostic criteria to be recognized as a medical condition. Excited delirium also includes excited delirium syndrome, excited delirium, hyperactive delirium, agitated delirium, and exhaustive mania.

SEC. 2.

Chapter 3.5 (commencing with Section 24400) is added to Division 20 of the Health and Safety Code, to read:

CHAPTER  3.5. Excited Delirium

24400.

 For the purposes of this chapter, “excited delirium” means a term used to describe a person’s state of agitation, excitability, paranoia, extreme aggression, physical violence, and apparent immunity to pain that is not listed in the most current version of the Diagnostic and Statistical Manual of Mental Disorders, or for which the court finds there is insufficient scientific evidence or diagnostic criteria to be recognized as a medical condition. Excited delirium also includes excited delirium syndrome, excited delirium, hyperactive delirium, agitated delirium, and exhaustive mania.

24401.

(a) Excited delirium shall not be recognized as a valid medical diagnosis or cause of death in this state.

(b) A state or local government entity, or employee or contractor of a state or local government entity, shall not document, testify to, or otherwise use in any official capacity or communication excited delirium as a recognized medical diagnosis or cause of death.

(c) A coroner, medical examiner, physician, or physician assistant shall not state on the certificate of death, or in any report, that the cause of death was excited delirium. The term excited delirium and terms inclusive of excited delirium defined in Section 24400 shall not be listed anywhere on the death certificate.

24402.

 A peace officer shall not use the term excited delirium to describe an individual in an incident report completed by a peace officer. A peace officer may describe the characteristics of an individual’s conduct, but shall not generally describe the individual’s demeanor, conduct, or physical and mental condition at issue as excited delirium.

24403.

 Pursuant to Section 1156.5 of the Evidence Code, evidence that a person suffered or experienced excited delirium is inadmissible in any civil action. A party or witness may describe the factual circumstances surrounding the case, including a person’s demeanor, conduct, and physical and mental condition at issue, including, but not limited to, a person’s state of agitation, excitability, paranoia, extreme aggression, physical violence, and apparent immunity to pain, but shall not describe or diagnose such demeanor, conduct, or condition by use of the term excited delirium, or attribute such demeanor, conduct, or physical and mental condition to that term.

2020 WL 7232303 (Ga.Super.) (Trial Order)

Superior Court of Georgia.

Ware County

STATE of Georgia,

v.

Sheila DENTON, Defendant.

No. 04-R-330.

February 7, 2020.

Order Granting Extraordinary Motion for New Trial

Dwayne H. Gillis, Judge.

*1 Before the Court is Sheila Denton’s Extraordinary Motion for New Trial. The Court has heard evidence, argument, and reviewed the briefs from Mrs. Denton and the State. It is noted that Mrs. Denton’s presence was waived for the hearing on the motion.

Upon conviction for the murder of Eugene Garner, the Defendant filed a motion for new trial, which was denied by the trial court. A direct appeal followed which was affirmed, Denton v. State. 286 Ga. 494: 689 S.E. 2d 322 (2010).

The issues before the Court surround the legitimacy of forensic bite mark evidence. Specific to Mrs. Denton’s case is whether advancements in scientific knowledge regarding bite mark evidence are new evidence warranting the relief sought. The advancements at issue are changes to the ABFO (American Board of Forensic Odontology) guidelines which occurred in 2016. The undisputed evidence at present is that the advancements, scientific understanding and ABFO guidelines would compel a different expert opinion if the case were tried today. Thusly, the seminal question before the Court is whether a change in the expert opinion testimony of a key trial witness, resulting from changes in scientific knowledge and understanding as expressed and contained in the ABFO guidelines is newly discovered evidence.

The Defendant in this case, Sheila Denton, was convicted of murder and sentenced to life in prison in 2006. The evidence against her at trial included testimony by forensic dentist Thomas David, who opined that a bite mark on Denton was probably caused by the victim’s teeth and that a probable bite mark on the victim was probably caused by Denton. (T. 145-80.)¹ At the time of Denton’s trial, such testimony was accepted as the product of a credible forensic science discipline.

On November 8,2017, Denton filed an extraordinary motion for new trial, alleging that recent developments in the understanding of the limitations of bite mark analysis and comparison demonstrate that the testimony today would no longer be inculpatory. An evidentiary hearing was held before this Court on May 29,2018. The Court then invited post-hearing briefing.

Having carefully considered the evidence that was presented at the 2006 trial, the 2018 hearing, and the parties’ briefs, for the reasons that follow, Denton’s motion is GRANTED.

1. The Legal Standard

To prevail on her extraordinary motion for new trial, Denton must satisfy the six-factor test elucidated in Timber lake v. State, 246 Ga. 488 (1980). Specifically, she must satisfy this Court “(1) that the evidence has come to [her] knowledge since the trial; (2) that it was not owing to the want of due diligence that [she] did not acquire it sooner; (3) that it is so material that it would probably produce a different verdict; (4) that it is not cumulative only; (5) that the affidavit of the witness himself should be procured or its absence accounted for; and (6) that a new trial will not be granted if the only effect of the evidence will be to impeach the credit of a witness.” Id. at 491.

*2 In granting the extraordinary motion, this Court is cognizant that motions for new trial filed pursuant to O.C.G.A. § 5-5-41(b) are an extraordinary remedy. Mitchum v. State, No. S19A0554, 2019 WL 4924049, at *2 (Ga. Oct. 7, 2019).

1. Findings of Fact

At the hearing, Denton presented the live testimony of three forensic dentists-Dr. Thomas David, Dr. Cynthia Brzozowski, and Dr. Adam Freeman. All three experts are Board-Certified Diplomates of the American Board of Forensic Odontology (“ABFO”). The ABFO is the country’s only board-certifying entity for forensic dentists. (E.M.N.T. 42, 143,198.) Their testimony unanimously established that Denton is entitled to a new trial.

Dr. David, who testified against Denton at her trial, admitted that if he testified at Denton’s trial today, his testimony would be different, following advances in the scientific understanding of the limitations of bite mark evidence that are reflected in the recent change in ABFO Guidelines. (E.M.N.T. 126,129-30.) Dr. Brzozowski and Dr. Freeman agreed with that assessment. (E.M.N.T. 165-66,238.) In addition to these live witnesses, Denton also filed, in advance of the hearing, two affidavits, co-authored by a total of five Board-Certified ABFO Diplomates. The first affidavit was co-authored by Dr. Brzozowski, Dr. Anthony Cardoza, and Dr. James Wood; the other was co-authored by Dr. Freeman and Dr. Iain Pretty. Each of the experts agreed that the trial testimony would not be incriminating today based on advances in scientific understanding.

This consensus follows the 2016 change in ABFO Guidelines, which were described as “very significant” (E.M.N.T. 161), and a “seminal moment in the ABFO and a quantum leap” (E.M.N.T. 235). Thus, this Court was presented with unanimous expert opinion that the bite mark evidence presented at trial against Denton would not be incriminating today. The Court found the experts and evidence Denton presented, which challenged the reliability of the bite mark evidence presented at Denton’s trial based on new scientific developments, to be credible.

1. The Evidence at Trial

On May 21,2004, the body of Eugene Garner was discovered inside of his home at 1638 ABC Avenue, in Waycross, Georgia. (T. 71, 76,270.) Observation of the scene indicated that a struggle had occurred. (T. 88, 115,129, 374.) The autopsy revealed that the cause of Garner’s death was “blunt force head trauma in conjunction with manual strangulation.” (T. 311.)

Denton was arrested for giving police officers a false name during a stop that occurred around midnight on the day Garner’s body was found. (T. 360, 366.) She was subsequently charged for his murder. At trial, forensic dentist Dr. David testified that an injury found on Denton was a bite mark “to a reasonable scientific certainty,” and an injury found on Garner was “probabl[y]” a bite mark. (T. 154.) Dr. David further opined, “to a reasonable degree of scientific certainty,” that “Denton was the probable source of the bite mark on Mr. Garner and that Mr. Garner was the probable source of the bite mark on Denton.” (T. 159,167.)

Aside from the bite mark evidence, the State presented little other incriminating evidence. An analyst from the Georgia Bureau of Investigation’s Division of Forensic Sciences testified concerning “hairs that were identified as recovered from the victim’s hand” at the scene. (T. 183.) She compared those hairs to “known head hairs” from both the decedent and Denton. (T. 183.) The analyst testified that “the questioned hair in the victim’s hand matched his own head hair.” (T. 187-88.) No hair matching Denton was found at the scene.

*3 The State presented a witness, Sharon Jones, who was an admitted drug addict and crack cocaine dealer with a criminal record. Jones testified that at approximately 4:30 a.m. one day in May of 2004, Denton came to a known crack house where Jones was guarding about $500 worth of crack cocaine. (T. 336-38, 352.) Jones admitted on the stand that she had been smoking a lot of crack cocaine that night, such that days would bleed into one another, making her unable to really say what else had happened that evening. (T. 346-47.) Jones claimed that Denton had cuts on her arms and legs, and Denton told Jones that she “just killed the man that stay in front of Walley’s.” (T. 336-38.) Jones testified that Denton gave no other information about the murder. (T. 338.) Jones testified that she did not believe Denton’s claim that she had killed someone, and instead Jones thought that Denton had been scratched after falling into some bushes. (T. 338-39, 348.) According to Jones, she then sold Denton some crack cocaine. (T. 339.)

Jones testified that when she initially spoke with police, she did not tell the questioning officers any information incriminating Denton. Instead, Jones’s account of Denton’s alleged admission came only after an interrogation during which two police officers “kept saying I did it.” (T. 342.) The interrogation, as retold by Jones at trial, then went as follows:

1. They tried to say - well, [Officer] Larry Hill said he had my fingerprints at the house and he had a tape, videotape with me there, which I knew wasn’t true.
2. Okay. He said you were involved?
3. He said he had my fingerprints at that house.

1. Right.

1. And a videotape with me being there.

1. Did he tell you he was going to put you in jail if you didn’t make a statement to him about it?

1. Uh-huh (positive response).

1. And he showed you a blank tape and told you he had a videotape of you at Mr. Garner’s house involved in something?

1. Uh-huh (positive response).

1. And he was going to put you in jail?

1. Yes, sir.

(T. 344-45.) Jones’s trial testimony concluded with her admitting that she lied to the police three different times. (T. 354.) None of the other witnesses at trial provided testimony linking Denton to the crime.

1. Dr. Cynthia Brzozowski and Dr. Adam Freeman

At the May 2018 evidentiary hearing, Denton called three experts, including Dr. Brzozowski and Dr. Freeman. The Court finds that Dr. Brzozowski and Dr. Freeman were forthcoming and credible in their testimony.

Dr. Brzozowski received her doctor of medical dentistry from the University of Boston Dental School in 1986, completed a residency program at the Veteran’s Administration the following year, has been a fellow with the American Academy of Forensic Sciences since 1994, has been a member of the American Society of Forensic Odontology since 1992, and has served on the board of American Board of Forensic Odontology for three terms. (E.M.N.T. 142-44.)

Dr. Freeman graduated from dental school at Columbia University, completed a fellowship at the University of Texas, is a consultant for the Connecticut Medical Examiner, and has consulted with the FBI. (E.M.N.T. 196-97.) He has served as president, vice-president, and secretary of the ABFO and has chaired the bite mark and metrology committees. (E.M.N.T. 198.)

Drs. Brzozowski and Freeman testified that they came to testify before this Court out of what they termed their “ethical” and “civic” duty. (E.M.N.T. 149,203.) They did so without being paid for their work, and each suffered a financial loss to be away from their respective practices to testify. (E.M.N.T. 149, 203.) They both testified that at one point in their careers in forensic dentistry they believed bite mark evidence could be probative in court. But advances in understanding, recent scientific studies, the lack of proficiency testing, and the growing number of wrongful convictions attributable to bite mark evidence led them to see the error of their past beliefs. (E.M.N.T. 149-50,246-47.) Their testimony was devoid of self-interest bias.

*4 Dr. Brzozowski candidly admitted that when she became a member of the ABFO in 2006, she believed that bite mark analysis was based on valid science; however, she no longer does. Dr. Brzozowski testified:

I think there are several contributing factors that have shaped my opinion over the last decade. The first being that we have no scientific studies to validate accuracy and reliability of bite mark methods. We actually have studies that show the unreliability of bite mark methods, but we have none to show the reliability and the accuracy.

Two, there have been three independent scientific panels who have come to the same conclusion, that being the Texas Forensic Science Commission, the PCAST and the National Academy of Sciences.

Three, the failure of the ABFO to initiate and conduct such studies to validate their methods. And lastly, the increasing number of wrongful convictions that have been largely unaddressed by the ABFO.

(E.M.N.T. 149-50.) She then testified that at the time of Denton’s trial, individualization and probabilistic testimony² from forensic dentists was supported by the ABFO. (E.M.N.T. 153.) In 2016, however, in response to the factors that Dr. Brzozowski referenced in her testimony, the range of permissible associations changed. (E.M.N.T. 153.)

Dr. Brzozowski testified that at a threshold level, when determining the nature of the injury, “probable bite mark”-the characterization Dr. David provided at trial for the injury to Garner-is no longer a permitted conclusion by the ABFO Guidelines. (E.M.N.T. 157-58.) Dr. David’s remaining trial testimony in this case was inconsistent with today’s scientific understanding of bite marks as well. (E.M.N.T. 165.) “He testified that Ms. Denton was the probable biter of Mr. Garner and that Mr. Garner was the probable biter of Ms. Denton. And today we would not testify as to a probabilistic association.” (E.M.N.T. 165.) Rather, the strongest conclusion that could be offered today would be “that one could not exclude that individual as being the source.” (E.M.N.T. 165.)

Dr. Brzozowski further testified that she reviewed photographs of the evidence in this case and determined that there was “insufficient information to determine that either of these injuries are human bite marks.” (E.M.N.T. 170.) Accordingly, Dr. Brzozowski would not make any comparisons between the injuries and any dentition under today’s Guidelines. (E.M.N.T. 173.)

The Court also credits Dr. Brzozowski based on her demeanor. She testified that this testimony was not easy for her to give, telling the Court that she took “pride in being a diplomate of the ABFO ... invested 12 years in [the] organization. But the failure of the ABFO to address the impact of scientifically invalidated bite mark methods on people’s lives and liberty causes [her] great concern.” (E.M.N.T. 177.) The Court notes that this statement was consistent with Dr. Brzozowski’s demeanor in court. It was clear to the Court that it was difficult for Dr. Brzozowski to testify and that she felt a professional obligation to do so. The Court finds her testimony credible.

*5 Dr. Freeman corroborated Dr. Brzozowski’s testimony. Dr. Freeman testified that he became a member of the ABFO in 2009, and eventually served as the president of the organization in 2016. (E.M.N.T. 198.) When he first began studying bite marks in 2003, he learned about individualization and probabilistic testimony, and accepted such testimony at the time as based on valid science. (E.M.N.T. 204-05.) However, Dr. Freeman no longer holds that belief today. (E.M.N.T. 237.) Dr. Freeman testified that his belief changed, in large part, after the results of a study he conducted, along with Dr. Pretty, called the Construct Validity Test. The Construct Validity Test was designed to answer the following threshold question in bite mark analysis: whether Board certified odontologists could reliably determine if an injury is a bite mark or not. (E.M.N.T. 217-19.) He testified that the study revealed that the answer is no: “there was a wide spread disagreement about what constituted a bite mark.” (E.M.N.T. 214.) Dr. Freeman testified that, in part because of his Construct Validity Study, the ABFO Guidelines changed to reject probabilistic testimony. (E.M.N.T. 235.) He classified this change as a “seminal moment” and a “quantum leap.” (E.M.N.T. 235.)

Dr. Freeman testified that he reviewed the bite mark evidence in Denton’s case, as well as Dr. David’s testimony. (E.M.N.T. 237-38, 242.) He further testified that-under the current ABFO Guidelines-Dr. David’s trial testimony concerning probability would be “a misrepresentation of what the underlying science supports.” (E.M.N.T. 238.) Dr. Freeman testified that it was his expert opinion that neither mark in Denton’s case constituted a bite mark. (E.M.N.T. 243-46.) Dr. Freeman presented credibly and answered questions in a clear and unbiased manner. His tone and demeanor indicated that he had genuine concern about errors in the use of bite mark evidence and that he felt a professional obligation to inform the Court about those concerns. The Court credits Dr. Freeman’s expert testimony as reliable, informative, and devoid of self-interest.

1. Dr. Thomas David

This Court found Dr. David to be less forthcoming, with a self-interest that became notable to the Court. Dr. David testified that he wanted to “control the narrative” in this case. (E.M.N.T. 113.) He sometimes failed to answer questions directly. (See, e.g., E.M.N.T. 59-63 (testifying that giving indirect answers to questions provides the witness with “wiggle room”).) Moreover though, Dr. David noted the only thing the [new] guidelines really changed was the terminology used in his analysis [EMNT. 127]. The Court finds this statement incredulous in light of the evidence produced in this case. For instance, Dr. David’s trial testimony that an injury to the victim was a probable bite mark (EMNT. 70-71, 87) and that the Defendant was the probable source of the injury to the victim to a reasonable degree of scientific certainty (emphasis added) (EMNT. 72-73,77), yet under the new guidelines he could not even conclude that that injury was a bite mark, much less to a comparison (EMNT. 129). The defense testimony and evidence in this case is profound and leads one to conclude that bite mark evidence is no longer a valid subject of scientific inquiry unless it can exonerate one or exclude a particular person. To describe the change in the ABFO stance on bite mark evidence as one involving mere “terminology” is at best disingenuous. The Court credits Dr. David’s testimony only to the extent it is corroborated by the other credible evidence and expert testimony.

On the most critical questions in the case, Dr. David agreed with the other experts who testified at the hearing. Specifically, when questioned by the Court and Denton, Dr. David admitted that today his testimony would be different than the testimony he gave at Denton’s trial. (E.M.N.T. 126.) Dr. David admitted that, today, he would not even compare the marks on the decedent with the teeth of Denton, because it is uncertain as to what caused the injury at all:

Q: In this case when you testified, you testified that the mark on Mr. Garner was a probable bite mark.

A: Correct.

Q: Under today’s guidelines - applying today’s guidelines, you wouldn’t do a comparison.

*6 A: That is correct, under the current guidelines.

(E.M.N.T. 129.)

The Court finds that Dr. David’s testimony supports Denton’s position-that the ABFO Guidelines changed since the time of the trial in this case, and if testifying today, Dr. David would follow the Guidelines i.e., his testimony today would be different than his original trial testimony.

1. Application of the Facts to the Law
2. The Evidence Came to Denton’s Knowledge Since the Trial.

This Court finds that the recent changes in the field of bite mark analysis, which led to the 2016 change in the ABFO Guidelines, and the corresponding change of the trial testimony in Denton’s case, have come to Denton’s knowledge since trial. In 2016, for the first time in the forty-year history of the ABFO, the organization prohibited probabilistic or individualization testimony in the field of bite mark analysis and comparison. This change, which occurred after Denton’s trial, means that the evidence offered at trial would be altogether different today.

The evidence at trial was that Denton was the “probable biter” of the “probable bite mark” on Garner and that Garner was the “probable biter” of the “bite mark” on Denton. (T. 154, 159,167.) At trial, Dr. David rendered those opinions with a reasonable degree of scientific certainty. Now, however, Denton’s experts and, ultimately, Dr. David, agree that, under the new Guidelines, the following new expert testimony would be presented at a trial today:

1) Dr. David would no longer compare the injury to Garner to any suspected dentition;

2) Dr. David would no longer testify that the injury to Garner was a bite mark;

3) Dr. David would no longer testify that Denton was the “probable biter”;

4) The best Dr. David could say is that Garner “could not be excluded” as the source of the injury to Denton.

(See, e.g., E.M.N.T. 75-78.) Today, therefore, the most that could be offered, consistent with the best practices reflected in the ABFO Guidelines, is that any person’s given dentition cannot be excluded as one of innumerable possible dentitions to have potentially inflicted a given injury. Such an opinion will seldom, if ever, be probative of one having inflicted a particular bite mark, nor shall it likely be of any aid to a jury in reaching a decision. The future admissibility of such evidence is dubious at best.

The changes to the ABFO Guidelines constitute newly discovered evidence, as courts across the country have found. See, e.g., Ex Parte Chaney, 563 S.W.3d 239,260 (Tex. Ct. Crim. App. 2018) (“[T]he 2016 ABFO Manual has completely invalidated any population statistics, regardless of whether the population is open or closed ... the Manual no longer allows examiners to give opinions to a ‘reasonable degree of dental certainty.”’); State v. Hill, 125 N.E.3d 158,168 (Ohio Ct. App. 2018) (the “trial court’s own statements acknowledge[ed] that the [discrediting of] bite mark evidence is newly discovered evidence”); Commonwealth v. Kunco, 173 A.3d 817, 821-22 (Pa. Super. Ct. 2017) (citing post-conviction testimony by three ABFO Diplomates that acknowledged that the change in the ABFO Guidelines represents a change in scientific understanding of the limitations of bite mark evidence, and that it is impossible to know how many people might be associated with an alleged bite mark).

*7 The Texas Court of Criminal Appeals recently set aside a conviction that had been based on bite mark evidence. Ex Parte Chaney, 563 S.W.3d 239 (Tex. Ct. Crim. App. 2018). The Texas court reasoned that “the body of scientific knowledge underlying the field of bite mark comparisons evolved in a way that discredits almost all the probabilistic bite mark evidence at trial.” Id. at 257. In reaching the conclusion that there was new, material evidence before the court, the Chaney court cited to the National Academy of Sciences Report, the new ABFO Guidelines, and studies and affidavits by forensic dentists similar to those that Denton has cited to this Court.³ And the State of Texas conceded that the new “‘bite mark evidence, which once appeared proof positive of... Chaney’s guilt, no longer proves anything.”’ Id. at 258. Likewise here-the new bite mark testimony no longer proves anything.

At the hearing, and in its post-hearing brief, the State advanced two counterarguments. The Court is not persuaded by either argument.

First, the State argued that Dr. David’s testimony today would be unchanged. However, the State also conceded in its post-hearing brief, that “under the new ABFO Guidelines [Dr. David’s] opinion about the injuries on the defendant would change from the [decedent] being the ‘probable biter,’ to the [decedent] ‘could not be excluded’ as the source of the bite mark.” (State’s Br. at 17.) In addition, the prosecution agreed that “the probable bite mark [Dr. David] found on the victim would be considered inconclusive under the new guidelines, and [Dr. David] should not do a comparison on that injury unless he could explain his reasons for deviating from the new guidelines.” (State’s Br. at 17.) These concessions undermine the prosecution’s position that the testimony would be unchanged. Therefore, the Court credits the testimony of the expert witnesses who testified at the hearing (including Dr. David, who was also the State’s witness at Denton’s trial), who agreed that, under the new Guidelines, the dental testimony at trial today would be fundamentally different. This constitutes new evidence.

Second, the prosecution argued that the change in the ABFO Guidelines, which led to the change in the opinion Dr. David would render in this case and the opinion of Denton’s forensic dentists, does not constitute “new” evidence. (State’s Br. at 16.) The prosecution claimed that “[c]hanges or updates in scientific guidelines that are used by experts, especially when these changes are not to be applied retroactively, are not new evidence.” (State’s Br. at 16.)⁴ The Court does not find the prosecution’s argument persuasive. The prosecution cited no authority in support of this contention, and the Court has found none. Rather, new scientific Guidelines that create a consensus among prosecution and defense forensic experts that the forensic evidence is no longer inculpatory, is a prime example of “new” evidence. Moreover, a focus on scientific policies, guidelines, techniques, and procedures are not key evidence upon which convictions are based, it is the “expert opinion” evidence generated from the application of such policies, guidelines, techniques, and procedures. This record is clear that the expert opinion evidence has changed. Additionally, it is clear that the ABFO guidelines changed due to the realization that bite mark evidence is inherently unreliable.

1. It Was Not Owing to the Want of Due Diligence that Denton Did Not Acquire the New Evidence Sooner.

*8 The Court finds that Denton acted with due diligence in filing her extraordinary motion for new trial for a number of reasons. First, she filed in 2017, following the change in the ABFO Guidelines, which occurred in 2016. The ABFO has been in existence for more than forty years, and that change marked the very first time in its history that probabilistic and individualization testimony, like that offered by Dr. David at trial, was disallowed. By filing as soon as she did after that “seminal” change, Denton acted with due diligence.

Second, Denton filed her motion in this case prior to numerous decisions that have since granted relief based on the change in the ABFO Guidelines and new understanding regarding the limitations of bite mark evidence. See, e.g., Order on Defendant’s Motion to Stay Execution of Sentence, Commonwealth v. Gary Cifizzari, No. 8385-CR-4051, at 2 (Super. Ct. Worcester Co. July 12, 2019); State v. Roden, 437 P.3d 1203 (Or. 2019); Ex Parte Chaney, 563 S.W.3d 239 (Tex. Ct. Crim. App. 2018).

Third, the Court is aware of the many, and increasing, exonerations of people wrongfully convicted based on bite mark evidence of the sort offered against Denton in this case. See Innocence Project, https://www.innocenceproject.org/wp-content/uploads/2019/01/Description-of-bite-mark-exonerations-and-statistical-analysis_UPDATED-01.28.19.pdf (last visited Nov. 12,2019). As far as this Court can tell, none of those exonerations are from Georgia, so she is among the first in the state to raise and litigate this claim.

Fourth, the Court notes that since the hearing in this case, an additional exoneration occurred as well. On July 12, 2019, Gary Cifizzari was released from prison pending further proceedings pursuant to a motion for new trial he filed challenging his 1984 murder conviction. See “Cifizzari gets out of prison; seeks new trial in “79 Milford slaying,” Telegram and Gazette (July 12, 2019), available at https://www.telegram.com/news/20190712/cifîzzari-gets-out-of-prison-seeks-new-trial-in-79-milford-slaying. The same Dr. Freeman and Dr. Pretty who offered affidavits and testimony in this case also offered affidavits in Cifizzari’s case. In granting Cifizzari’s release, the judge cited the forensic odontologists who are “condemning bite mark comparison evidence as there is no scientific basis for forensic odontologists to proffer testimony concluding that a suspect is ‘the biter’ to the exclusion of all others.” Order on Defendant’s Motion to Stay Execution of Sentence, Commonwealth v. Gary Cifizzari, No. 8385-CR-4051, at 2 (Super. Ct. Worcester Co. July 12, 2019) (emphasis added).

In light of all the above, the Court finds Denton exercised due diligence in filing an extraordinary motion for new trial in 2017, following the significant 2016 change in the ABFO Guidelines.

1. The New Evidence Is So Material that It Would Probably Produce a Different Verdict.

The Court has reviewed the entire trial record in this case in light of the new evidence. After that review, the Court finds the materiality prong of Timberlake satisfied because the new evidence is so material that it would probably produce a different verdict.

First, the prosecution conceded at trial that without the bite mark evidence, Denton would have been acquitted. In his closing argument, the prosecutor argued that if not for the testimony that the decedent inflicted a bite mark on Denton, “I’d have to admit that’s reasonable doubt.” (T. 434.) As discussed above, the new evidence provided by the experts before this Court established that the bite mark evidence used at trial is now known to be unsupported by science. Rather than purportedly tying Denton to the crime of which she was convicted, the undisputed forensic evidence today would not do so. Dr. David’s trial testimony-consistent with science at the time it was given but inconsistent with science today-was powerful evidence that the prosecutor relied on in urging the jury to convict. Today, the evidence would not support the conclusions that the prosecution relied on to secure a conviction. Without such testimony, the Court agrees with the prosecutor’s statement at trial: that Denton probably would be acquitted today.

*9 In its post-hearing brief, the State argues that the new evidence is not material. In making this argument, the State relies on Sharon Jones’s trial testimony, including her statements to the police. The Court does not find this argument persuasive.

Jones’s claim about Denton’s alleged admission came after she herself was accused of murder during an interrogation in which two police officers, according to Jones, “kept saying I did it.” (T. 342.) Jones did not originally implicate Denton, she only implicated Denton after being pressured by the police, she admitted to having been under the influence of drugs around the time of the offense, and she admitted, under oath, that she had lied to the police. (T. 336-38, 346-47,352, 354.)

Significantly, the prosecutor did not rely on Jones’s statement in his closing argument to the jury. Instead, the prosecutor essentially undermined the statement’s credibility during closing argument. He stated:

Sharon Shontae Jones ... [tjhat’s who she’s smoking crack with. I guess, we could have checked the churches, but I guess we couldn’t find anybody there, so we had to get the ones who actually were with her that morning. It would have been easier to go to a church and find people who could come and testify to these things, but they’re not out smoking crack cocaine in the early morning hours.

(T. 437.) The prosecution informed the jury that it could not pick its witnesses, but had to live with the witnesses it had, despite their incredibility. Because ofthat, the prosecution explicitly asked the jury to convict based on the bite mark evidence, saying, “If that’s all we had, I’d have to admit that’s reasonable doubt-----But then we’ve got also the bite mark on Sheila Denton.” (T. 434.) Employing that logic, and applying the new ABFO Guidelines, today Denton probably would be acquitted.

The State presented no other forensic or physical evidence connecting Denton to the crime. There was no DNA evidence presented. While hair was found in the victim’s hand, the State’s trial expert determined that the hair belonged to the victim; Denton’s hair was not found on the scene. (T. 183,187-88.) The only other evidence against Denton was that she ran from the police and gave a false name when arrested. These facts do not alter the Court’s materiality analysis, however, because they provide no independent affirmative evidence that Denton committed a murder. The materiality prong is satisfied.⁵

In addition, as recently as March of 2019, a court, with the concession of the State, found that bite mark evidence should not have been admitted in a child murder case, and reversed the conviction. State v. Roden, 437 P.3d 1203 (Or. 2019), involved a murder and child abuse case in which testimony indicated that three children suffered various adult bite marks to several different body parts. Id. at 1208. On appeal, the prosecution conceded that admission of the bite mark testimony was erroneous. And the Oregon court agreed:

*10 [T]he state concedes that the trial court erred in admitting that evidence ... The state’s concession is appropriate ... The state’s odontologist claimed that bite mark identification had undergone testing and peer review and had an error rate of “zero,” but he admitted that he could not cite a single peer-reviewed study testing and validating the technique, and he was unable to support his assertion regarding error rates. Meanwhile, defendant cited studies highlighting concerns within the scientific community regarding the high rate of error and lack of objective, standardized results in bite mark analysis and identification.

Roden, 296 Or. App. at 615-16 (internal citations omitted). Accordingly, the court reversed convictions for felony murder, murder by abuse, and first-degree manslaughter. Id. at 1213.⁶

The recent opinion in Ex parte Chaney is also instructive as to the question of whether the verdict today probably would be different. Last year, the Texas Court of Criminal Appeals found the following:

[The] 2016 ABFO Manual has completely invalidated any population statistics, regardless of whether the population is open or closed ... The body of scientific knowledge underlying the field of bite mark comparisons has evolved since his trial in a way that contradicts the scientific evidence relied on by the State at trial. New peer-reviewed studies discredit nearly all the testimony given by [the ABFO-certified dentists] about the mark on [the victim’s] left forearm and [Petitioner] being a “match.”

Chaney, 563 S.W.3d at 260-61. At Chaney’s trial, the prosecutor, in closing, argued:

But, most of all, we have the bite mark. I wouldn’t ask you to convict just based on the testimony of the tennis shoes, of the statements Chaney made to Westphalen, or the statements he made to Curtis Hilton. But, by golly, I’m going to ask you to convict on that dental testimony.

Id. at 262. That closing argument bears resemblance to the closing argument the prosecutor made at Denton’s trial, namely, “If that’s all we had, I’d have to admit that’s reasonable doubt. ... But then we’ve got also the bite mark on Sheila Denton.” (T. 434.) In each case, there was other evidence-including potentially inculpatory statements-that arguably tied the defendant to the crime, but in each case the “linchpin” of the case was the bite mark evidence. See id. at 262. Thus, the Chaney court granted relief, finding that the prosecution’s case would have been “incredibly weakened” had the “newly available scientific evidence been presented at his trial.” Id.

Just like in Chaney, at Denton’s trial the prosecution argued the necessity of the bite mark evidence to a conviction. Just like in Chaney, the prosecution’s case would have been incredibly weakened by the newly available scientific evidence. Accordingly, just like in Chaney, the outcome of trial today probably would be different, and Denton is entitled to a new trial.

1. The New Evidence Is Not Cumulative Only.

The Court finds that the new evidence is not cumulative and further notes that the prosecution has not argued otherwise.

1. Affidavits and Live Testimony Were Presented.

The Court finds that Denton submitted two affidavits and presented live testimony in support of her request for a new trial, thereby satisfying this Timberlake factor.

1. The Effect of the New Evidence Is Not Only To Impeach the Credibility of a Witness.

*11 At trial, Dr. David testified that “Denton was the probable source of the probable bite mark” on Eugene Garner and that “Mr. Garner was the probable source of the bite mark on Denton.” (E.M.N.T. 72.) Dr. David further testified that he held those beliefs “to a reasonable degree of scientific certainty.” (E.M.N.T. 72-73.) The new evidence does not simply impeach those conclusions; rather, the new evidence demonstrates that Dr. David would not draw those conclusions, nor testify to those conclusions today. Instead, Dr. David would follow the ABFO Guidelines. As a result, he would not do a comparison between Denton’s dentition and the injury to Garner at all, because he would not conclude that the injury to Garner was a bite mark in the first place. (E.M.N.T. 129.) As to the mark on Denton, the most Dr. David would offer is that any person’s dentition, including Garner’s, could not be excluded as one of an unquantifiable number of possible dentitions that could have inflicted the injury. (E.M.N.T. 77.)

The following exchanges occurred at the hearing during the examination of Dr. David, and are determinative of this prong:

Q: Under today’s guidelines - applying today’s guidelines, you wouldn’t do a comparison [to the injury on Mr. Garner]?

A: That is correct, under the current guidelines.

(E.M.N.T. 129.)

Q: Today, consistent with the guidelines, the strongest conclusion you could come to would be that Mr. Garner could not be excluded as having made the bite mark [on Sheila Denton], correct?

A: That he couldn’t be - yes, correct.

Q: That would be the strongest association you could make under today’s guidelines, correct?

A: Under the current guidelines, that is correct.

(E.M.N.T. 77.)

The other experts presented at the hearing reiterated these inarguable conclusions in straightforward testimony. For example, with respect to the injury to Mr. Garner, Dr. Brzozowski testified:

Q: What if you determine that the injury is a probable bite mark?

A: You stop. No further analysis. No comparison is done.

Q: Is that even a permitted conclusion today, probable bite mark?

A: No, it’s not.

(E.M.N.T. 157-58.)

In other words, the testimony at trial was that Denton was the probable source of the probable bite mark on Garner, an opinion Dr. David held to a reasonable degree of scientific certainty. Today, all that could be said of the injury is that it is “inconclusive” as to how it was inflicted; no comparison would be done to any dentition.⁷ Rather than a “bite mark,” the injury to Garner, today, would be called just that, an “injury.” That is not impeachment, it is substantively different expert testimony that would not incriminate Denton.

Dr. Freeman analogized to another feature-comparison discipline at the hearing:

[W]e should all have some level of agreement as to whether or not the injury we are looking at, the threshold, that that injury is a bite mark. And as an example I would give, fingerprint examiners often will disagree about who has made that fingerprint, who attributed that fingerprint. They don’t disagree over whether or not it is a fingerprint.

And if... we can’t establish step one, that something is a bite mark, if you’re wrong about that step then surely any conclusion you make attributing a person to making that bite mark would be faulty.

(E.M.N.T.209,216.)

Likewise, an opinion that someone is the “probable biter” is dramatically different from an opinion that someone is in a class of an innumerable number of people who potentially could have inflicted the injury. With respect to the injury to Denton, Dr. Brzozowski testified:

Q: What would be the strongest conclusion that could be offered consistent with the guidelines today?

A: The strongest conclusion was that one could not exclude that individual as being the source.

*12 Q: You said you can not exclude as opposed to probable biter, which would be different conclusions?

A: Yes.

Q: Is that a matter of semantics or is that a difference, a significant difference?

A: Well, it is a significant difference because can not exclude may mean that there might be thousands of other people that could possibly have inflicted the injury.

(E.M.N.T. 165-66.) Dr. Freeman characterized it as follows:

Making a statement like someone is the probable biter [the testimony at trial], that is significant, versus saying I can not exclude this person as having made this bite mark [the testimony now]. That indicates that the population that could have made the bite mark is very wide spread.

(E.M.N.T. 235.)

In other words, the expert testimony at trial, to a reasonable degree of scientific certainty, was that Garner was the probable source of the bite mark on Denton. Today, even assuming that the injury is a bite mark (a conclusion not supported by the expert testimony before this Court), the most that could be said is that Garner could be one of innumerable people whose dentition could have inflicted that injury. That is not impeachment; it is qualitatively different testimony that no longer implicates Denton to the exclusion of myriad suspects.

Conclusion

When evidence which is deemed to be “scientific” or “expert”, it transcends the normal understanding of the lay juror. Courts generally admit scientific evidence and the expert opinions based thereon, if the Court affirmatively determines, “whether the technique has gained general acceptance in the scientific community which recognizes it,” Harper v. State, 249 Ga 519,524-525 (1982). The Court in Harper went on to state, “An evaluation of whether the principle has gained acceptance will often be transmitted to the trial court by members of the appropriate scientific community testifying as expert witnesses at trial.” Id. at 525. What then should the Court do when the experts testify that the scientific principle has lost acceptance in the appropriate scientific community? In the present case, the loss of acceptance within the field of forensic odontology is not disputed. The Harper test is still the law under the new evidence code. Winters v. State, 305 Ga. 229 (2019). The object of the Court’s inquiry in Harper was to determine “whether a scientific principle or technique is a phenomenon that may be verified with such certainty that it is competent evidence in a court of law...” Id. At 524. It stands to reason then, that the Harper test shall serve to determine that a scientific principle or technique has become a phenomenon that can not (emphasis added) be verified with the requisite certainty and that it is incompetent evidence in a court of law.

The ABFO’s recognition of the inadequacies of bite mark evidence is profound and undisputed. We now know the bite mark evidence presented in Denton’s trial was not competent evidence under Harper. The State’s attempt to cast this case as involving mere changes in scientific guidelines misses the point. The changes in ABFO guidelines are clearly intended to represent the state of scientific understanding in the field. These guidelines even limit the conclusions that an ABFO diplomat may testify to, presumably due to the inadequacies of bite mark evidence.

*13 An analysis or determination that a given scientific principle or technique is no longer generally accepted in the scientific community, will by definition most often involve a post-trial review. With the current understanding of bite mark evidence in the field of forensic odontology, the strongest testimony that can be rendered, in favor of the prosecution, is that the defendant cannot be excluded as the biter. Such testimony will likely be deemed highly prejudicial and of little probative value in the future. With the future of such expert opinion and testimony in serious doubt, the question arises as to whether a Harper analysis can and should be allowed retroactively post-trial, as newly discovered evidence. The Court has answered that question in the affirmative. However, another issue is whether due process would dictate that the Court undertake a reverse Harper type analysis where prior scientific evidence, principles, techniques and testimony are essentially de-ligitimized. This analysis may and should occur without the constraints of Drane v. State, and our extraordinary motion for new trial body of law. Proven unreliable scientific evidence should never serve as the basis of a conviction and should be dealt with by the Courts if and when it is found. Applying such an analysis to the facts of this case, it is uncontroverted that bite mark analysis and testimony as existed at the time of Denton’s conviction has been proven to be unreliable and not generally accepted within the scientific community of forensic odontology. Therefore, the Court grants the Defendant’s motion pursuant to Drane and a reverse Harper analysis.

WHEREFORE, Denton’s Extraordinary Motion for New Trial is hereby GRANTED. The Defendant shall be remanded to the custody of the Sheriff of Ware County, when this order becomes final.

Date: 2-7-20

<<signature>>

Dwayne H. Gillis, Chief Judge

Superior Court, Ware County

Waycross Judicial Circuit

Please go to our Moodle Page and under "Class 13" you will find the prompt and submission folder for Writing Reflection #13.

Trial by Fire

Did Texas execute an innocent man?

By David Grann

August 31, 2009

Cameron Todd Willingham in his cell on death row, in 1994. He insisted upon his innocence in the deaths of his children and refused an offer to plead guilty in return for a life sentence. Photograph By Ken Light

The fire moved quickly through the house, a one-story wood-frame structure in a working-class neighborhood of Corsicana, in northeast Texas. Flames spread along the walls, bursting through doorways, blistering paint and tiles and furniture. Smoke pressed against the ceiling, then banked downward, seeping into each room and through crevices in the windows, staining the morning sky.

Buffie Barbee, who was eleven years old and lived two houses down, was playing in her back yard when she smelled the smoke. She ran inside and told her mother, Diane, and they hurried up the street; that’s when they saw the smoldering house and Cameron Todd Willingham standing on the front porch, wearing only a pair of jeans, his chest blackened with soot, his hair and eyelids singed. He was screaming, “My babies are burning up!” His children—Karmon and Kameron, who were one-year-old twin girls, and two-year-old Amber—were trapped inside.

Willingham told the Barbees to call the Fire Department, and while Diane raced down the street to get help he found a stick and broke the children’s bedroom window. Fire lashed through the hole. He broke another window; flames burst through it, too, and he retreated into the yard, kneeling in front of the house. A neighbor later told police that Willingham intermittently cried, “My babies!” then fell silent, as if he had “blocked the fire out of his mind.”

Diane Barbee, returning to the scene, could feel intense heat radiating off the house. Moments later, the five windows of the children’s room exploded and flames “blew out,” as Barbee put it. Within minutes, the first firemen had arrived, and Willingham approached them, shouting that his children were in their bedroom, where the flames were thickest. A fireman sent word over his radio for rescue teams to “step on it.”

More men showed up, uncoiling hoses and aiming water at the blaze. One fireman, who had an air tank strapped to his back and a mask covering his face, slipped through a window but was hit by water from a hose and had to retreat. He then charged through the front door, into a swirl of smoke and fire. Heading down the main corridor, he reached the kitchen, where he saw a refrigerator blocking the back door.

Todd Willingham, looking on, appeared to grow more hysterical, and a police chaplain named George Monaghan led him to the back of a fire truck and tried to calm him down. Willingham explained that his wife, Stacy, had gone out earlier that morning, and that he had been jolted from sleep by Amber screaming, “Daddy! Daddy!”

“My little girl was trying to wake me up and tell me about the fire,” he said, adding, “I couldn’t get my babies out.”

While he was talking, a fireman emerged from the house, cradling Amber. As she was given C.P.R., Willingham, who was twenty-three years old and powerfully built, ran to see her, then suddenly headed toward the babies’ room. Monaghan and another man restrained him. “We had to wrestle with him and then handcuff him, for his and our protection,” Monaghan later told police. “I received a black eye.” One of the first firemen at the scene told investigators that, at an earlier point, he had also held Willingham back. “Based on what I saw on how the fire was burning, it would have been crazy for anyone to try and go into the house,” he said.

Willingham was taken to a hospital, where he was told that Amber—who had actually been found in the master bedroom—had died of smoke inhalation. Kameron and Karmon had been lying on the floor of the children’s bedroom, their bodies severely burned. According to the medical examiner, they, too, died from smoke inhalation.

News of the tragedy, which took place on December 23, 1991, spread through Corsicana. A small city fifty-five miles northeast of Waco, it had once been the center of Texas’s first oil boom, but many of the wells had since dried up, and more than a quarter of the city’s twenty thousand inhabitants had fallen into poverty. Several stores along the main street were shuttered, giving the place the feel of an abandoned outpost.

Willingham and his wife, who was twenty-two years old, had virtually no money. Stacy worked in her brother’s bar, called Some Other Place, and Willingham, an unemployed auto mechanic, had been caring for the kids. The community took up a collection to help the Willinghams pay for funeral arrangements.

Fire investigators, meanwhile, tried to determine the cause of the blaze. (Willingham gave authorities permission to search the house: “I know we might not ever know all the answers, but I’d just like to know why my babies were taken from me.”) Douglas Fogg, who was then the assistant fire chief in Corsicana, conducted the initial inspection. He was tall, with a crew cut, and his voice was raspy from years of inhaling smoke from fires and cigarettes. He had grown up in Corsicana and, after graduating from high school, in 1963, he had joined the Navy, serving as a medic in Vietnam, where he was wounded on four occasions. He was awarded a Purple Heart each time. After he returned from Vietnam, he became a firefighter, and by the time of the Willingham blaze he had been battling fire—or what he calls “the beast”—for more than twenty years, and had become a certified arson investigator. “You learn that fire talks to you,” he told me.

He was soon joined on the case by one of the state’s leading arson sleuths, a deputy fire marshal named Manuel Vasquez, who has since died. Short, with a paunch, Vasquez had investigated more than twelve hundred fires. Arson investigators have always been considered a special breed of detective. In the 1991 movie “Backdraft,” a heroic arson investigator says of fire, “It breathes, it eats, and it hates. The only way to beat it is to think like it. To know that this flame will spread this way across the door and up across the ceiling.” Vasquez, who had previously worked in Army intelligence, had several maxims of his own. One was “Fire does not destroy evidence—it creates it.” Another was “The fire tells the story. I am just the interpreter.” He cultivated a Sherlock Holmes-like aura of invincibility. Once, he was asked under oath whether he had ever been mistaken in a case. “If I have, sir, I don’t know,” he responded. “It’s never been pointed out.”

Vasquez and Fogg visited the Willinghams’ house four days after the blaze. Following protocol, they moved from the least burned areas toward the most damaged ones. “It is a systematic method,” Vasquez later testified, adding, “I’m just collecting information. . . . I have not made any determination. I don’t have any preconceived idea.”

The men slowly toured the perimeter of the house, taking notes and photographs, like archeologists mapping out a ruin. Upon opening the back door, Vasquez observed that there was just enough space to squeeze past the refrigerator blocking the exit. The air smelled of burned rubber and melted wires; a damp ash covered the ground, sticking to their boots. In the kitchen, Vasquez and Fogg discerned only smoke and heat damage—a sign that the fire had not originated there—and so they pushed deeper into the nine-hundred-and-seventy-five-square-foot building. A central corridor led past a utility room and the master bedroom, then past a small living room, on the left, and the children’s bedroom, on the right, ending at the front door, which opened onto the porch. Vasquez tried to take in everything, a process that he compared to entering one’s mother-in-law’s house for the first time: “I have the same curiosity.”

In the utility room, he noticed on the wall pictures of skulls and what he later described as an image of “the Grim Reaper.” Then he turned into the master bedroom, where Amber’s body had been found. Most of the damage there was also from smoke and heat, suggesting that the fire had started farther down the hallway, and he headed that way, stepping over debris and ducking under insulation and wiring that hung down from the exposed ceiling.

As he and Fogg removed some of the clutter, they noticed deep charring along the base of the walls. Because gases become buoyant when heated, flames ordinarily burn upward. But Vasquez and Fogg observed that the fire had burned extremely low down, and that there were peculiar char patterns on the floor, shaped like puddles.

Vasquez’s mood darkened. He followed the “burn trailer”—the path etched by the fire—which led from the hallway into the children’s bedroom. Sunlight filtering through the broken windows illuminated more of the irregularly shaped char patterns. A flammable or combustible liquid doused on a floor will cause a fire to concentrate in these kinds of pockets, which is why investigators refer to them as “pour patterns” or “puddle configurations.”

The fire had burned through layers of carpeting and tile and plywood flooring. Moreover, the metal springs under the children’s beds had turned white—a sign that intense heat had radiated beneath them. Seeing that the floor had some of the deepest burns, Vasquez deduced that it had been hotter than the ceiling, which, given that heat rises, was, in his words, “not normal.”

Fogg examined a piece of glass from one of the broken windows. It contained a spiderweb-like pattern—what fire investigators call “crazed glass.” Forensic textbooks had long described the effect as a key indicator that a fire had burned “fast and hot,” meaning that it had been fuelled by a liquid accelerant, causing the glass to fracture.

The men looked again at what appeared to be a distinct burn trailer through the house: it went from the children’s bedroom into the corridor, then turned sharply to the right and proceeded out the front door. To the investigators’ surprise, even the wood under the door’s aluminum threshold was charred. On the concrete floor of the porch, just outside the front door, Vasquez and Fogg noticed another unusual thing: brown stains, which, they reported, were consistent with the presence of an accelerant.

The men scanned the walls for soot marks that resembled a “V.” When an object catches on fire, it creates such a pattern, as heat and smoke radiate outward; the bottom of the “V” can therefore point to where a fire began. In the Willingham house, there was a distinct “V” in the main corridor. Examining it and other burn patterns, Vasquez identified three places where fire had originated: in the hallway, in the children’s bedroom, and at the front door. Vasquez later testified that multiple origins pointed to one conclusion: the fire was “intentionally set by human hands.”

By now, both investigators had a clear vision of what had happened. Someone had poured liquid accelerant throughout the children’s room, even under their beds, then poured some more along the adjoining hallway and out the front door, creating a “fire barrier” that prevented anyone from escaping; similarly, a prosecutor later suggested, the refrigerator in the kitchen had been moved to block the back-door exit. The house, in short, had been deliberately transformed into a death trap.

The investigators collected samples of burned materials from the house and sent them to a laboratory that could detect the presence of a liquid accelerant. The lab’s chemist reported that one of the samples contained evidence of “mineral spirits,” a substance that is often found in charcoal-lighter fluid. The sample had been taken by the threshold of the front door.

The fire was now considered a triple homicide, and Todd Willingham—the only person, besides the victims, known to have been in the house at the time of the blaze—became the prime suspect.

Police and fire investigators canvassed the neighborhood, interviewing witnesses. Several, like Father Monaghan, initially portrayed Willingham as devastated by the fire. Yet, over time, an increasing number of witnesses offered damning statements. Diane Barbee said that she had not seen Willingham try to enter the house until after the authorities arrived, as if he were putting on a show. And when the children’s room exploded with flames, she added, he seemed more preoccupied with his car, which he moved down the driveway. Another neighbor reported that when Willingham cried out for his babies he “did not appear to be excited or concerned.” Even Father Monaghan wrote in a statement that, upon further reflection, “things were not as they seemed. I had the feeling that [Willingham] was in complete control.”

The police began to piece together a disturbing profile of Willingham. Born in Ardmore, Oklahoma, in 1968, he had been abandoned by his mother when he was a baby. His father, Gene, who had divorced his mother, eventually raised him with his stepmother, Eugenia. Gene, a former U.S. marine, worked in a salvage yard, and the family lived in a cramped house; at night, they could hear freight trains rattling past on a nearby track. Willingham, who had what the family called the “classic Willingham look”—a handsome face, thick black hair, and dark eyes—struggled in school, and as a teen-ager began to sniff paint. When he was seventeen, Oklahoma’s Department of Human Services evaluated him, and reported, “He likes ‘girls,’ music, fast cars, sharp trucks, swimming, and hunting, in that order.” Willingham dropped out of high school, and over time was arrested for, among other things, driving under the influence, stealing a bicycle, and shoplifting.

In 1988, he met Stacy, a senior in high school, who also came from a troubled background: when she was four years old, her stepfather had strangled her mother to death during a fight. Stacy and Willingham had a turbulent relationship. Willingham, who was unfaithful, drank too much Jack Daniel’s, and sometimes hit Stacy—even when she was pregnant. A neighbor said that he once heard Willingham yell at her, “Get up, bitch, and I’ll hit you again.”

On December 31st, the authorities brought Willingham in for questioning. Fogg and Vasquez were present for the interrogation, along with Jimmie Hensley, a police officer who was working his first arson case. Willingham said that Stacy had left the house around 9 a.m. to pick up a Christmas present for the kids, at the Salvation Army. “After she got out of the driveway, I heard the twins cry, so I got up and gave them a bottle,” he said. The children’s room had a safety gate across the doorway, which Amber could climb over but not the twins, and he and Stacy often let the twins nap on the floor after they drank their bottles. Amber was still in bed, Willingham said, so he went back into his room to sleep. “The next thing I remember is hearing ‘Daddy, Daddy,’ ” he recalled. “The house was already full of smoke.” He said that he got up, felt around the floor for a pair of pants, and put them on. He could no longer hear his daughter’s voice (“I heard that last ‘Daddy, Daddy’ and never heard her again”), and he hollered, “Oh God— Amber, get out of the house! Get out of the house!’ ”

He never sensed that Amber was in his room, he said. Perhaps she had already passed out by the time he stood up, or perhaps she came in after he left, through a second doorway, from the living room. He said that he went down the corridor and tried to reach the children’s bedroom. In the hallway, he said, “you couldn’t see nothing but black.” The air smelled the way it had when their microwave had blown up, three weeks earlier—like “wire and stuff like that.” He could hear sockets and light switches popping, and he crouched down, almost crawling. When he made it to the children’s bedroom, he said, he stood and his hair caught on fire. “Oh God, I never felt anything that hot before,” he said of the heat radiating out of the room.

After he patted out the fire on his hair, he said, he got down on the ground and groped in the dark. “I thought I found one of them once,” he said, “but it was a doll.” He couldn’t bear the heat any longer. “I felt myself passing out,” he said. Finally, he stumbled down the corridor and out the front door, trying to catch his breath. He saw Diane Barbee and yelled for her to call the Fire Department. After she left, he insisted, he tried without success to get back inside.

The investigators asked him if he had any idea how the fire had started. He said that he wasn’t sure, though it must have originated in the children’s room, since that was where he first saw flames; they were glowing like “bright lights.” He and Stacy used three space heaters to keep the house warm, and one of them was in the children’s room. “I taught Amber not to play with it,” he said, adding that she got “whuppings every once in a while for messing with it.” He said that he didn’t know if the heater, which had an internal flame, was turned on. (Vasquez later testified that when he had checked the heater, four days after the fire, it was in the “Off” position.) Willingham speculated that the fire might have been started by something electrical: he had heard all that popping and crackling.

When pressed whether someone might have a motive to hurt his family, he said that he couldn’t think of anyone that “cold-blooded.” He said of his children, “I just don’t understand why anybody would take them, you know? We had three of the most pretty babies anybody could have ever asked for.” He went on, “Me and Stacy’s been together for four years, but off and on we get into a fight and split up for a while and I think those babies is what brought us so close together . . . neither one of us . . . could live without them kids.” Thinking of Amber, he said, “To tell you the honest-to-God’s truth, I wish she hadn’t woke me up.”

During the interrogation, Vasquez let Fogg take the lead. Finally, Vasquez turned to Willingham and asked a seemingly random question: had he put on shoes before he fled the house?

“No, sir,” Willingham replied.

A map of the house was on a table between the men, and Vasquez pointed to it. “You walked out this way?” he said.

Willingham said yes.

Vasquez was now convinced that Willingham had killed his children. If the floor had been soaked with a liquid accelerant and the fire had burned low, as the evidence suggested, Willingham could not have run out of the house the way he had described without badly burning his feet. A medical report indicated that his feet had been unscathed.

Willingham insisted that, when he left the house, the fire was still around the top of the walls and not on the floor. “I didn’t have to jump through any flames,” he said. Vasquez believed that this was impossible, and that Willingham had lit the fire as he was retreating—first, torching the children’s room, then the hallway, and then, from the porch, the front door. Vasquez later said of Willingham, “He told me a story of pure fabrication. . . . He just talked and he talked and all he did was lie.”

Still, there was no clear motive. The children had life-insurance policies, but they amounted to only fifteen thousand dollars, and Stacy’s grandfather, who had paid for them, was listed as the primary beneficiary. Stacy told investigators that even though Willingham hit her he had never abused the children—“Our kids were spoiled rotten,” she said—and she did not believe that Willingham could have killed them.

Ultimately, the authorities concluded that Willingham was a man without a conscience whose serial crimes had climaxed, almost inexorably, in murder. John Jackson, who was then the assistant district attorney in Corsicana, was assigned to prosecute Willingham’s case. He later told the Dallas Morning News that he considered Willingham to be “an utterly sociopathic individual” who deemed his children “an impediment to his lifestyle.” Or, as the local district attorney, Pat Batchelor, put it, “The children were interfering with his beer drinking and dart throwing.”

On the night of January 8, 1992, two weeks after the fire, Willingham was riding in a car with Stacy when swat teams surrounded them, forcing them to the side of the road. “They pulled guns out like we had just robbed ten banks,” Stacy later recalled. “All we heard was ‘click, click.’ . . . Then they arrested him.”

Willingham was charged with murder. Because there were multiple victims, he was eligible for the death penalty, under Texas law. Unlike many other prosecutors in the state, Jackson, who had ambitions of becoming a judge, was personally opposed to capital punishment. “I don’t think it’s effective in deterring criminals,” he told me. “I just don’t think it works.” He also considered it wasteful: because of the expense of litigation and the appeals process, it costs, on average, $2.3 million to execute a prisoner in Texas—about three times the cost of incarcerating someone for forty years. Plus, Jackson said, “What’s the recourse if you make a mistake?” Yet his boss, Batchelor, believed that, as he once put it, “certain people who commit bad enough crimes give up the right to live,” and Jackson came to agree that the heinous nature of the crime in the Willingham case—“one of the worst in terms of body count” that he had ever tried—mandated death.

Willingham couldn’t afford to hire lawyers, and was assigned two by the state: David Martin, a former state trooper, and Robert Dunn, a local defense attorney who represented everyone from alleged murderers to spouses in divorce cases—a “Jack-of-all-trades,” as he calls himself. (“In a small town, you can’t say ‘I’m a so-and-so lawyer,’ because you’ll starve to death,” he told me.)

The Willingham family in the days before Christmas, 1991.

Not long after Willingham’s arrest, authorities received a message from a prison inmate named Johnny Webb, who was in the same jail as Willingham. Webb alleged that Willingham had confessed to him that he took “some kind of lighter fluid, squirting [it] around the walls and the floor, and set a fire.” The case against Willingham was considered airtight.

Even so, several of Stacy’s relatives—who, unlike her, believed that Willingham was guilty—told Jackson that they preferred to avoid the anguish of a trial. And so, shortly before jury selection, Jackson approached Willingham’s attorneys with an extraordinary offer: if their client pleaded guilty, the state would give him a life sentence. “I was really happy when I thought we might have a deal to avoid the death penalty,” Jackson recalls.

Willingham’s lawyers were equally pleased. They had little doubt that he had committed the murders and that, if the case went before a jury, he would be found guilty, and, subsequently, executed. “Everyone thinks defense lawyers must believe their clients are innocent, but that’s seldom true,” Martin told me. “Most of the time, they’re guilty as sin.” He added of Willingham, “All the evidence showed that he was one hundred per cent guilty. He poured accelerant all over the house and put lighter fluid under the kids’ beds.” It was, he said, “a classic arson case”: there were “puddle patterns all over the place—no disputing those.”

Martin and Dunn advised Willingham that he should accept the offer, but he refused. The lawyers asked his father and stepmother to speak to him. According to Eugenia, Martin showed them photographs of the burned children and said, “Look what your son did. You got to talk him into pleading, or he’s going to be executed.”

His parents went to see their son in jail. Though his father did not believe that he should plead guilty if he were innocent, his stepmother beseeched him to take the deal. “I just wanted to keep my boy alive,” she told me.

Willingham was implacable. “I ain’t gonna plead to something I didn’t do, especially killing my own kids,” he said. It was his final decision. Martin says, “I thought it was nuts at the time—and I think it’s nuts now.”

Willingham’s refusal to accept the deal confirmed the view of the prosecution, and even that of his defense lawyers, that he was an unrepentant killer.

In August, 1992, the trial commenced in the old stone courthouse in downtown Corsicana. Jackson and a team of prosecutors summoned a procession of witnesses, including Johnny Webb and the Barbees. The crux of the state’s case, though, remained the scientific evidence gathered by Vasquez and Fogg. On the stand, Vasquez detailed what he called more than “twenty indicators” of arson.

“Do you have an opinion as to who started the fire?” one of the prosecutors asked.

“Yes, sir,” Vasquez said. “Mr. Willingham.”

The prosecutor asked Vasquez what he thought Willingham’s intent was in lighting the fire. “To kill the little girls,” he said.

The defense had tried to find a fire expert to counter Vasquez and Fogg’s testimony, but the one they contacted concurred with the prosecution. Ultimately, the defense presented only one witness to the jury: the Willinghams’ babysitter, who said she could not believe that Willingham could have killed his children. (Dunn told me that Willingham had wanted to testify, but Martin and Dunn thought that he would make a bad witness.) The trial ended after two days.

During his closing arguments, Jackson said that the puddle configurations and pour patterns were Willingham’s inadvertent “confession,” burned into the floor. Showing a Bible that had been salvaged from the fire, Jackson paraphrased the words of Jesus from the Gospel of Matthew: “Whomsoever shall harm one of my children, it’s better for a millstone to be hung around his neck and for him to be cast in the sea.”

The jury was out for barely an hour before returning with a unanimous guilty verdict. As Vasquez put it, “The fire does not lie.”

II

When Elizabeth Gilbert approached the prison guard, on a spring day in 1999, and said Cameron Todd Willingham’s name, she was uncertain about what she was doing. A forty-seven-year-old French teacher and playwright from Houston, Gilbert was divorced with two children. She had never visited a prison before. Several weeks earlier, a friend, who worked at an organization that opposed the death penalty, had encouraged her to volunteer as a pen pal for an inmate on death row, and Gilbert had offered her name and address. Not long after, a short letter, written with unsteady penmanship, arrived from Willingham. “If you wish to write back, I would be honored to correspond with you,” he said. He also asked if she might visit him. Perhaps out of a writer’s curiosity, or perhaps because she didn’t feel quite herself (she had just been upset by news that her ex-husband was dying of cancer), she agreed. Now she was standing in front of the decrepit penitentiary in Huntsville, Texas—a place that inmates referred to as “the death pit.”

She filed past a razor-wire fence, a series of floodlights, and a checkpoint, where she was patted down, until she entered a small chamber. Only a few feet in front of her was a man convicted of multiple infanticide. He was wearing a white jumpsuit with “DR”—for death row—printed on the back, in large black letters. He had a tattoo of a serpent and a skull on his left biceps. He stood nearly six feet tall and was muscular, though his legs had atrophied after years of confinement.

A Plexiglas window separated Willingham from her; still, Gilbert, who had short brown hair and a bookish manner, stared at him uneasily. Willingham had once fought another prisoner who called him a “baby killer,” and since he had been incarcerated, seven years earlier, he had committed a series of disciplinary infractions that had periodically landed him in the segregation unit, which was known as “the dungeon.”

Willingham greeted her politely. He seemed grateful that she had come. After his conviction, Stacy had campaigned for his release. She wrote to Ann Richards, then the governor of Texas, saying, “I know him in ways that no one else does when it comes to our children. Therefore, I believe that there is no way he could have possibly committed this crime.” But within a year Stacy had filed for divorce, and Willingham had few visitors except for his parents, who drove from Oklahoma to see him once a month. “I really have no one outside my parents to remind me that I am a human being, not the animal the state professes I am,” he told Gilbert at one point.

He didn’t want to talk about death row. “Hell, I live here,” he later wrote her. “When I have a visit, I want to escape from here.” He asked her questions about her teaching and art. He expressed fear that, as a playwright, she might find him a “one-dimensional character,” and apologized for lacking social graces; he now had trouble separating the mores in prison from those of the outside world.

The aftermath of the fire on December 23, 1991. Photograph from Texas State Fire Marshal’s Office

 Photograph from Texas State Fire Marshal’s Office

When Gilbert asked him if he wanted something to eat or drink from the vending machines, he declined. “I hope I did not offend you by not accepting any snacks,” he later wrote her. “I didn’t want you to feel I was there just for something like that.”

She had been warned that prisoners often tried to con visitors. He appeared to realize this, subsequently telling her, “I am just a simple man. Nothing else. And to most other people a convicted killer looking for someone to manipulate.”

Their visit lasted for two hours, and afterward they continued to correspond. She was struck by his letters, which seemed introspective, and were not at all what she had expected. “I am a very honest person with my feelings,” he wrote her. “I will not bullshit you on how I feel or what I think.” He said that he used to be stoic, like his father. But, he added, “losing my three daughters . . . my home, wife and my life, you tend to wake up a little. I have learned to open myself.”

She agreed to visit him again, and when she returned, several weeks later, he was visibly moved. “Here I am this person who nobody on the outside is ever going to know as a human, who has lost so much, but still trying to hold on,” he wrote her afterward. “But you came back! I don’t think you will ever know of what importance that visit was in my existence.”

They kept exchanging letters, and she began asking him about the fire. He insisted that he was innocent and that, if someone had poured accelerant through the house and lit it, then the killer remained free. Gilbert wasn’t naïve—she assumed that he was guilty. She did not mind giving him solace, but she was not there to absolve him.

Still, she had become curious about the case, and one day that fall she drove down to the courthouse in Corsicana to review the trial records. Many people in the community remembered the tragedy, and a clerk expressed bewilderment that anyone would be interested in a man who had burned his children alive.

Gilbert took the files and sat down at a small table. As she examined the eyewitness accounts, she noticed several contradictions. Diane Barbee had reported that, before the authorities arrived at the fire, Willingham never tried to get back into the house—yet she had been absent for some time while calling the Fire Department. Meanwhile, her daughter Buffie had reported witnessing Willingham on the porch breaking a window, in an apparent effort to reach his children. And the firemen and police on the scene had described Willingham frantically trying to get into the house.

The witnesses’ testimony also grew more damning after authorities had concluded, in the beginning of January, 1992, that Willingham was likely guilty of murder. In Diane Barbee’s initial statement to authorities, she had portrayed Willingham as “hysterical,” and described the front of the house exploding. But on January 4th, after arson investigators began suspecting Willingham of murder, Barbee suggested that he could have gone back inside to rescue his children, for at the outset she had seen only “smoke coming from out of the front of the house”—smoke that was not “real thick.”

An even starker shift occurred with Father Monaghan’s testimony. In his first statement, he had depicted Willingham as a devastated father who had to be repeatedly restrained from risking his life. Yet, as investigators were preparing to arrest Willingham, he concluded that Willingham had been too emotional (“He seemed to have the type of distress that a woman who had given birth would have upon seeing her children die”); and he expressed a “gut feeling” that Willingham had “something to do with the setting of the fire.”

Dozens of studies have shown that witnesses’ memories of events often change when they are supplied with new contextual information. Itiel Dror, a cognitive psychologist who has done extensive research on eyewitness and expert testimony in criminal investigations, told me, “The mind is not a passive machine. Once you believe in something—once you expect something—it changes the way you perceive information and the way your memory recalls it.”

After Gilbert’s visit to the courthouse, she kept wondering about Willingham’s motive, and she pressed him on the matter. In response, he wrote, of the death of his children, “I do not talk about it much anymore and it is still a very powerfully emotional pain inside my being.” He admitted that he had been a “sorry-ass husband” who had hit Stacy—something he deeply regretted. But he said that he had loved his children and would never have hurt them. Fatherhood, he said, had changed him; he stopped being a hoodlum and “settled down” and “became a man.” Nearly three months before the fire, he and Stacy, who had never married, wed at a small ceremony in his home town of Ardmore. He said that the prosecution had seized upon incidents from his past and from the day of the fire to create a portrait of a “demon,” as Jackson, the prosecutor, referred to him. For instance, Willingham said, he had moved the car during the fire simply because he didn’t want it to explode by the house, further threatening the children.

Gilbert was unsure what to make of his story, and she began to approach people who were involved in the case, asking them questions. “My friends thought I was crazy,” Gilbert recalls. “I’d never done anything like this in my life.”

One morning, when Willingham’s parents came to visit him, Gilbert arranged to see them first, at a coffee shop near the prison. Gene, who was in his seventies, had the Willingham look, though his black hair had gray streaks and his dark eyes were magnified by glasses. Eugenia, who was in her fifties, with silvery hair, was as sweet and talkative as her husband was stern and reserved. The drive from Oklahoma to Texas took six hours, and they had woken at three in the morning; because they could not afford a motel, they would have to return home later that day. “I feel like a real burden to them,” Willingham had written Gilbert.

As Gene and Eugenia sipped coffee, they told Gilbert how grateful they were that someone had finally taken an interest in Todd’s case. Gene said that his son, though he had flaws, was no killer.

The evening before the fire, Eugenia said, she had spoken on the phone with Todd. She and Gene were planning on visiting two days later, on Christmas Eve, and Todd told her that he and Stacy and the kids had just picked up family photographs. “He said, ‘We got your pictures for Christmas,’ ” she recalled. “He put Amber on the phone, and she was tattling on one of the twins. Todd didn’t seem upset. If something was bothering him, I would have known.”

The aftermath of the fire on December 23, 1991. Photograph from Texas State Fire Marshal’s Office

 Photograph from Texas State Fire Marshal’s Office

Gene and Eugenia got up to go: they didn’t want to miss any of the four hours that were allotted for the visit with their son. Before they left, Gene said, “You’ll let us know if you find anything, won’t you?”

Over the next few weeks, Gilbert continued to track down sources. Many of them, including the Barbees, remained convinced that Willingham was guilty, but several of his friends and relatives had doubts. So did some people in law enforcement. Willingham’s former probation officer in Oklahoma, Polly Goodin, recently told me that Willingham had never demonstrated bizarre or sociopathic behavior. “He was probably one of my favorite kids,” she said. Even a former judge named Bebe Bridges—who had often stood, as she put it, on the “opposite side” of Willingham in the legal system, and who had sent him to jail for stealing—told me that she could not imagine him killing his children. “He was polite, and he seemed to care,” she said. “His convictions had been for dumb-kid stuff. Even the things stolen weren’t significant.” Several months before the fire, Willingham tracked Goodin down at her office, and proudly showed her photographs of Stacy and the kids. “He wanted Bebe and me to know he’d been doing good,” Goodin recalled.

Eventually, Gilbert returned to Corsicana to interview Stacy, who had agreed to meet at the bed-and-breakfast where Gilbert was staying. Stacy was slightly plump, with pale, round cheeks and feathered dark-blond hair; her bangs were held in place by gel, and her face was heavily made up. According to a tape recording of the conversation, Stacy said that nothing unusual had happened in the days before the fire. She and Willingham had not fought, and were preparing for the holiday. Though Vasquez, the arson expert, had recalled finding the space heater off, Stacy was sure that, at least on the day of the incident—a cool winter morning—it had been on. “I remember turning it down,” she recalled. “I always thought, Gosh, could Amber have put something in there?” Stacy added that, more than once, she had caught Amber “putting things too close to it.”

Willingham had often not treated her well, she recalled, and after his incarceration she had left him for a man who did. But she didn’t think that her former husband should be on death row. “I don’t think he did it,” she said, crying.

Though only the babysitter had appeared as a witness for the defense during the main trial, several family members, including Stacy, testified during the penalty phase, asking the jury to spare Willingham’s life. When Stacy was on the stand, Jackson grilled her about the “significance” of Willingham’s “very large tattoo of a skull, encircled by some kind of a serpent.”

“It’s just a tattoo,” Stacy responded.

“He just likes skulls and snakes. Is that what you’re saying?”

“No. He just had—he got a tattoo on him.”

The prosecution cited such evidence in asserting that Willingham fit the profile of a sociopath, and brought forth two medical experts to confirm the theory. Neither had met Willingham. One of them was Tim Gregory, a psychologist with a master’s degree in marriage and family issues, who had previously gone goose hunting with Jackson, and had not published any research in the field of sociopathic behavior. His practice was devoted to family counselling.

At one point, Jackson showed Gregory Exhibit No. 60—a photograph of an Iron Maiden poster that had hung in Willingham’s house—and asked the psychologist to interpret it. “This one is a picture of a skull, with a fist being punched through the skull,” Gregory said; the image displayed “violence” and “death.” Gregory looked at photographs of other music posters owned by Willingham. “There’s a hooded skull, with wings and a hatchet,” Gregory continued. “And all of these are in fire, depicting—it reminds me of something like Hell. And there’s a picture—a Led Zeppelin picture of a falling angel. . . . I see there’s an association many times with cultive-type of activities. A focus on death, dying. Many times individuals that have a lot of this type of art have interest in satanic-type activities.”

The other medical expert was James P. Grigson, a forensic psychiatrist. He testified so often for the prosecution in capital-punishment cases that he had become known as Dr. Death. (A Texas appellate judge once wrote that when Grigson appeared on the stand the defendant might as well “commence writing out his last will and testament.”) Grigson suggested that Willingham was an “extremely severe sociopath,” and that “no pill” or treatment could help him. Grigson had previously used nearly the same words in helping to secure a death sentence against Randall Dale Adams, who had been convicted of murdering a police officer, in 1977. After Adams, who had no prior criminal record, spent a dozen years on death row—and once came within seventy-two hours of being executed—new evidence emerged that absolved him, and he was released. In 1995, three years after Willingham’s trial, Grigson was expelled from the American Psychiatric Association for violating ethics. The association stated that Grigson had repeatedly arrived at a “psychiatric diagnosis without first having examined the individuals in question, and for indicating, while testifying in court as an expert witness, that he could predict with 100-per-cent certainty that the individuals would engage in future violent acts.”

After speaking to Stacy, Gilbert had one more person she wanted to interview: the jailhouse informant Johnny Webb, who was incarcerated in Iowa Park, Texas. She wrote to Webb, who said that she could see him, and they met in the prison visiting room. A man in his late twenties, he had pallid skin and a closely shaved head; his eyes were jumpy, and his entire body seemed to tremble. A reporter who once met him described him to me as “nervous as a cat around rocking chairs.” Webb had begun taking drugs when he was nine years old, and had been convicted of, among other things, car theft, selling marijuana, forgery, and robbery.

As Gilbert chatted with him, she thought that he seemed paranoid. During Willingham’s trial, Webb disclosed that he had been given a diagnosis of “post-traumatic stress disorder” after he was sexually assaulted in prison, in 1988, and that he often suffered from “mental impairment.” Under cross-examination, Webb testified that he had no recollection of a robbery that he had pleaded guilty to only months earlier.

Johnny Webb claimed that Willingham confessed to him in prison. Photograph by Alex Garcia / Chicago Tribune

 Photograph by Alex Garcia / Chicago Tribune

Webb repeated for her what he had said in court: he had passed by Willingham’s cell, and as they spoke through a food slot Willingham broke down and told him that he intentionally set the house on fire. Gilbert was dubious. It was hard to believe that Willingham, who had otherwise insisted on his innocence, had suddenly confessed to an inmate he barely knew. The conversation had purportedly taken place by a speaker system that allowed any of the guards to listen—an unlikely spot for an inmate to reveal a secret. What’s more, Webb alleged that Willingham had told him that Stacy had hurt one of the kids, and that the fire was set to cover up the crime. The autopsies, however, had revealed no bruises or signs of trauma on the children’s bodies.

Jailhouse informants, many of whom are seeking reduced time or special privileges, are notoriously unreliable. According to a 2004 study by the Center on Wrongful Convictions, at Northwestern University Law School, lying police and jailhouse informants are the leading cause of wrongful convictions in capital cases in the United States. At the time that Webb came forward against Willingham, he was facing charges of robbery and forgery. During Willingham’s trial, another inmate planned to testify that he had overheard Webb saying to another prisoner that he was hoping to “get time cut,” but the testimony was ruled inadmissible, because it was hearsay. Webb, who pleaded guilty to the robbery and forgery charges, received a sentence of fifteen years. Jackson, the prosecutor, told me that he generally considered Webb “an unreliable kind of guy,” but added, “I saw no real motive for him to make a statement like this if it wasn’t true. We didn’t cut him any slack.” In 1997, five years after Willingham’s trial, Jackson urged the Texas Board of Pardons and Paroles to grant Webb parole. “I asked them to cut him loose early,” Jackson told me. The reason, Jackson said, was that Webb had been targeted by the Aryan Brotherhood. The board granted Webb parole, but within months of his release he was caught with cocaine and returned to prison.

In March, 2000, several months after Gilbert’s visit, Webb unexpectedly sent Jackson a Motion to Recant Testimony, declaring, “Mr. Willingham is innocent of all charges.” But Willingham’s lawyer was not informed of this development, and soon afterward Webb, without explanation, recanted his recantation. When I recently asked Webb, who was released from prison two years ago, about the turnabout and why Willingham would have confessed to a virtual stranger, he said that he knew only what “the dude told me.” After I pressed him, he said, “It’s very possible I misunderstood what he said.” Since the trial, Webb has been given an additional diagnosis, bipolar disorder. “Being locked up in that little cell makes you kind of crazy,” he said. “My memory is in bits and pieces. I was on a lot of medication at the time. Everyone knew that.” He paused, then said, “The statute of limitations has run out on perjury, hasn’t it?”

Aside from the scientific evidence of arson, the case against Willingham did not stand up to scrutiny. Jackson, the prosecutor, said of Webb’s testimony, “You can take it or leave it.” Even the refrigerator’s placement by the back door of the house turned out to be innocuous; there were two refrigerators in the cramped kitchen, and one of them was by the back door. Jimmie Hensley, the police detective, and Douglas Fogg, the assistant fire chief, both of whom investigated the fire, told me recently that they had never believed that the fridge was part of the arson plot. “It didn’t have nothing to do with the fire,” Fogg said.

After months of investigating the case, Gilbert found that her faith in the prosecution was shaken. As she told me, “What if Todd really was innocent?”

III

In the summer of 1660, an Englishman named William Harrison vanished on a walk, near the village of Charingworth, in Gloucestershire. His bloodstained hat was soon discovered on the side of a local road. Police interrogated Harrison’s servant, John Perry, and eventually Perry gave a statement that his mother and his brother had killed Harrison for money. Perry, his mother, and his brother were hanged.

Two years later, Harrison reappeared. He insisted, fancifully, that he had been abducted by a band of criminals and sold into slavery. Whatever happened, one thing was indisputable: he had not been murdered by the Perrys.

The fear that an innocent person might be executed has long haunted jurors and lawyers and judges. During America’s Colonial period, dozens of crimes were punishable by death, including horse thievery, blasphemy, “man-stealing,” and highway robbery. After independence, the number of crimes eligible for the death penalty was gradually reduced, but doubts persisted over whether legal procedures were sufficient to prevent an innocent person from being executed. In 1868, John Stuart Mill made one of the most eloquent defenses of capital punishment, arguing that executing a murderer did not display a wanton disregard for life but, rather, proof of its value. “We show, on the contrary, most emphatically our regard for it by the adoption of a rule that he who violates that right in another forfeits it for himself,” he said. For Mill, there was one counterargument that carried weight—“that if by an error of justice an innocent person is put to death, the mistake can never be corrected.”

The modern legal system, with its lengthy appeals process and clemency boards, was widely assumed to protect the kind of “error of justice” that Mill feared. In 2000, while George W. Bush was governor of Texas, he said, “I know there are some in the country who don’t care for the death penalty, but . . . we’ve adequately answered innocence or guilt.” His top policy adviser on issues of criminal justice emphasized that there is “super due process to make sure that no innocent defendants are executed.”

In recent years, though, questions have mounted over whether the system is fail-safe. Since 1976, more than a hundred and thirty people on death row have been exonerated. DNA testing, which was developed in the eighties, saved seventeen of them, but the technique can be used only in rare instances. Barry Scheck, a co-founder of the Innocence Project, which has used DNA testing to exonerate prisoners, estimates that about eighty per cent of felonies do not involve biological evidence.

In 2000, after thirteen people on death row in Illinois were exonerated, George Ryan, who was then governor of the state, suspended the death penalty. Though he had been a longtime advocate of capital punishment, he declared that he could no longer support a system that has “come so close to the ultimate nightmare—the state’s taking of innocent life.” Former Supreme Court Justice Sandra Day O’Connor has said that the “execution of a legally and factually innocent person would be a constitutionally intolerable event.”

Such a case has become a kind of grisly Holy Grail among opponents of capital punishment. In his 2002 book “The Death Penalty,” Stuart Banner observes, “The prospect of killing an innocent person seemed to be the one thing that could cause people to rethink their support for capital punishment. Some who were not troubled by statistical arguments against the death penalty—claims about deterrence or racial disparities—were deeply troubled that such an extreme injustice might occur in an individual case.” Opponents of the death penalty have pointed to several questionable cases. In 1993, Ruben Cantu was executed in Texas for fatally shooting a man during a robbery. Years later, a second victim, who survived the shooting, told the Houston Chronicle that he had been pressured by police to identify Cantu as the gunman, even though he believed Cantu to be innocent. Sam Millsap, the district attorney in the case, who had once supported capital punishment (“I’m no wild-eyed, pointy-headed liberal”), said that he was disturbed by the thought that he had made a mistake.

In 1995, Larry Griffin was put to death in Missouri, for a drive-by shooting of a drug dealer. The case rested largely on the eyewitness testimony of a career criminal named Robert Fitzgerald, who had been an informant for prosecutors before and was in the witness-protection program. Fitzgerald maintained that he happened to be at the scene because his car had broken down. After Griffin’s execution, a probe sponsored by the N.A.A.C.P.’s Legal Defense and Educational Fund revealed that a man who had been wounded during the incident insisted that Griffin was not the shooter. Moreover, the first police officer at the scene disputed that Fitzgerald had witnessed the crime.

These cases, however, stopped short of offering irrefutable proof that a “legally and factually innocent person” was executed. In 2005, a St. Louis prosecutor, Jennifer Joyce, launched an investigation of the Griffin case, upon being presented with what she called “compelling” evidence of Griffin’s potential innocence. After two years of reviewing the evidence, and interviewing a new eyewitness, Joyce said that she and her team were convinced that the “right person was convicted.”

Supreme Court Justice Antonin Scalia, in 2006, voted with a majority to uphold the death penalty in a Kansas case. In his opinion, Scalia declared that, in the modern judicial system, there has not been “a single case—not one—in which it is clear that a person was executed for a crime he did not commit. If such an event had occurred in recent years, we would not have to hunt for it; the innocent’s name would be shouted from the rooftops.”

“My problems are simple,” Willingham wrote Gilbert in September, 1999. “Try to keep them from killing me at all costs. End of story.”

During his first years on death row, Willingham had pleaded with his lawyer, David Martin, to rescue him. “You can’t imagine what it’s like to be here, with people I have no business even being around,” he wrote.

For a while, Willingham shared a cell with Ricky Lee Green, a serial killer, who castrated and fatally stabbed his victims, including a sixteen-year-old boy. (Green was executed in 1997.) Another of Willingham’s cellmates, who had an I.Q. below seventy and the emotional development of an eight-year-old, was raped by an inmate. “You remember me telling you I had a new celly?” Willingham wrote in a letter to his parents. “The little retarded boy. . . . There was this guy here on the wing who is a shit sorry coward (who is the same one I got into it with a little over a month ago). Well, he raped [my cellmate] in the 3 row shower week before last.” Willingham said that he couldn’t believe that someone would “rape a boy who cannot even defend himself. Pretty damn low.”

Because Willingham was known as a “baby killer,” he was a target of attacks. “Prison is a rough place, and with a case like mine they never give you the benefit of a doubt,” he wrote his parents. After he tried to fight one prisoner who threatened him, Willingham told a friend that if he hadn’t stood up for himself several inmates would have “beaten me up or raped or”—his thought trailed off.

Over the years, Willingham’s letters home became increasingly despairing. “This is a hard place, and it makes a person hard inside,” he wrote. “I told myself that was one thing I did not want and that was for this place to make me bitter, but it is hard.” He went on, “They have [executed] at least one person every month I have been here. It is senseless and brutal. . . . You see, we are not living in here, we are only existing.” In 1996, he wrote, “I just been trying to figure out why after having a wife and 3 beautiful children that I loved my life has to end like this. And sometimes it just seems like it is not worth it all. . . . In the 3 1/2 years I been here I have never felt that my life was as worthless and desolate as it is now.” Since the fire, he wrote, he had the sense that his life was slowly being erased. He obsessively looked at photographs of his children and Stacy, which he stored in his cell. “So long ago, so far away,” he wrote in a poem. “Was everything truly there?”

Inmates on death row are housed in a prison within a prison, where there are no attempts at rehabilitation, and no educational or training programs. In 1999, after seven prisoners tried to escape from Huntsville, Willingham and four hundred and fifty-nine other inmates on death row were moved to a more secure facility, in Livingston, Texas. Willingham was held in isolation in a sixty-square-foot cell, twenty-three hours a day. He tried to distract himself by drawing—“amateur stuff,” as he put it—and writing poems. In a poem about his children, he wrote, “There is nothing more beautiful than you on this earth.” When Gilbert once suggested some possible revisions to his poems, he explained that he wrote them simply as expressions, however crude, of his feelings. “So to me to cut them up and try to improve on them just for creative-writing purposes would be to destroy what I was doing to start with,” he said.

Despite his efforts to occupy his thoughts, he wrote in his diary that his mind “deteriorates each passing day.” He stopped working out and gained weight. He questioned his faith: “No God who cared about his creation would abandon the innocent.” He seemed not to care if another inmate attacked him. “A person who is already dead inside does not fear” death, he wrote.

One by one, the people he knew in prison were escorted into the execution chamber. There was Clifton Russell, Jr., who, at the age of eighteen, stabbed and beat a man to death, and who said, in his last statement, “I thank my Father, God in Heaven, for the grace he has granted me—I am ready.” There was Jeffery Dean Motley, who kidnapped and fatally shot a woman, and who declared, in his final words, “I love you, Mom. Goodbye.” And there was John Fearance, who murdered his neighbor, and who turned to God in his last moments and said, “I hope He will forgive me for what I done.”

Willingham had grown close to some of his prison mates, even though he knew that they were guilty of brutal crimes. In March, 2000, Willingham’s friend Ponchai Wilkerson—a twenty-eight-year-old who had shot and killed a clerk during a jewelry heist—was executed. Afterward, Willingham wrote in his diary that he felt “an emptiness that has not been touched since my children were taken from me.” A year later, another friend who was about to be executed—“one of the few real people I have met here not caught up in the bravado of prison”—asked Willingham to make him a final drawing. “Man, I never thought drawing a simple Rose could be so emotionally hard,” Willingham wrote. “The hard part is knowing that this will be the last thing I can do for him.”

Another inmate, Ernest Ray Willis, had a case that was freakishly similar to Willingham’s. In 1987, Willis had been convicted of setting a fire, in West Texas, that killed two women. Willis told investigators that he had been sleeping on a friend’s living-room couch and woke up to a house full of smoke. He said that he tried to rouse one of the women, who was sleeping in another room, but the flames and smoke drove him back, and he ran out the front door before the house exploded with flames. Witnesses maintained that Willis had acted suspiciously; he moved his car out of the yard, and didn’t show “any emotion,” as one volunteer firefighter put it. Authorities also wondered how Willis could have escaped the house without burning his bare feet. Fire investigators found pour patterns, puddle configurations, and other signs of arson. The authorities could discern no motive for the crime, but concluded that Willis, who had no previous record of violence, was a sociopath—a “demon,” as the prosecutor put it. Willis was charged with capital murder and sentenced to death.

Willis had eventually obtained what Willingham called, enviously, a “bad-ass lawyer.” James Blank, a noted patent attorney in New York, was assigned Willis’s case as part of his firm’s pro-bono work. Convinced that Willis was innocent, Blank devoted more than a dozen years to the case, and his firm spent millions, on fire consultants, private investigators, forensic experts, and the like. Willingham, meanwhile, relied on David Martin, his court-appointed lawyer, and one of Martin’s colleagues to handle his appeals. Willingham often told his parents, “You don’t know what it’s like to have lawyers who won’t even believe you’re innocent.” Like many inmates on death row, Willingham eventually filed a claim of inadequate legal representation. (When I recently asked Martin about his representation of Willingham, he said, “There were no grounds for reversal, and the verdict was absolutely the right one.” He said of the case, “Shit, it’s incredible that anyone’s even thinking about it.”)

Willingham tried to study the law himself, reading books such as “Tact in Court, or How Lawyers Win: Containing Sketches of Cases Won by Skill, Wit, Art, Tact, Courage and Eloquence.” Still, he confessed to a friend, “The law is so complicated it is hard for me to understand.” In 1996, he obtained a new court-appointed lawyer, Walter Reaves, who told me that he was appalled by the quality of Willingham’s defense at trial and on appeal. Reaves prepared for him a state writ of habeas corpus, known as a Great Writ. In the byzantine appeals process of death-penalty cases, which frequently takes more than ten years, the writ is the most critical stage: a prisoner can introduce new evidence detailing such things as perjured testimony, unreliable medical experts, and bogus scientific findings. Yet most indigent inmates, like Willingham, who constitute the bulk of those on death row, lack the resources to track down new witnesses or dig up fresh evidence. They must depend on court-appointed lawyers, many of whom are “unqualified, irresponsible, or overburdened,” as a study by the Texas Defender Service, a nonprofit organization, put it. In 2000, a Dallas Morning News investigation revealed that roughly a quarter of the inmates condemned to death in Texas were represented by court-appointed attorneys who had, at some point in their careers, been “reprimanded, placed on probation, suspended or banned from practicing law by the State Bar.” Although Reaves was more competent, he had few resources to reinvestigate the case, and his writ introduced no new exculpatory evidence: nothing further about Webb, or the reliability of the eyewitness testimony, or the credibility of the medical experts. It focussed primarily on procedural questions, such as whether the trial court erred in its instructions to the jury.

The Texas Court of Criminal Appeals was known for upholding convictions even when overwhelming exculpatory evidence came to light. In 1997, DNA testing proved that sperm collected from a rape victim did not match Roy Criner, who had been sentenced to ninety-nine years for the crime. Two lower courts recommended that the verdict be overturned, but the Court of Criminal Appeals upheld it, arguing that Criner might have worn a condom or might not have ejaculated. Sharon Keller, who is now the presiding judge on the court, stated in a majority opinion, “The new evidence does not establish innocence.” In 2000, George W. Bush pardoned Criner. (Keller was recently charged with judicial misconduct, for refusing to keep open past five o’clock a clerk’s office in order to allow a last-minute petition from a man who was executed later that night.)

On October 31, 1997, the Court of Criminal Appeals denied Willingham’s writ. After Willingham filed another writ of habeas corpus, this time in federal court, he was granted a temporary stay. In a poem, Willingham wrote, “One more chance, one more strike / Another bullet dodged, another date escaped.”

Willingham was entering his final stage of appeals. As his anxieties mounted, he increasingly relied upon Gilbert to investigate his case and for emotional support. “She may never know what a change she brought into my life,” he wrote in his diary. “For the first time in many years she gave me a purpose, something to look forward to.”

As their friendship deepened, he asked her to promise him that she would never disappear without explanation. “I already have that in my life,” he told her.

Together, they pored over clues and testimony. Gilbert says that she would send Reaves leads to follow up, but although he was sympathetic, nothing seemed to come of them. In 2002, a federal district court of appeals denied Willingham’s writ without even a hearing. “Now I start the last leg of my journey,” Willingham wrote to Gilbert. “Got to get things in order.”

He appealed to the U.S. Supreme Court, but in December, 2003, he was notified that it had declined to hear his case. He soon received a court order announcing that “the Director of the Department of Criminal Justice at Huntsville, Texas, acting by and through the executioner designated by said Director . . . is hereby directed and commanded, at some hour after 6:00 p.m. on the 17th day of February, 2004, at the Department of Criminal Justice in Huntsville, Texas, to carry out this sentence of death by intravenous injection of a substance or substances in a lethal quantity sufficient to cause the death of said Cameron Todd Willingham.”

Willingham wrote a letter to his parents. “Are you sitting down?” he asked, before breaking the news. “I love you both so much,” he said.

His only remaining recourse was to appeal to the governor of Texas, Rick Perry, a Republican, for clemency. The process, considered the last gatekeeper to the executioner, has been called by the U.S. Supreme Court “the ‘fail safe’ in our criminal justice system.”

IV

One day in January, 2004, Dr. Gerald Hurst, an acclaimed scientist and fire investigator, received a file describing all the evidence of arson gathered in Willingham’s case. Gilbert had come across Hurst’s name and, along with one of Willingham’s relatives, had contacted him, seeking his help. After their pleas, Hurst had agreed to look at the case pro bono, and Reaves, Willingham’s lawyer, had sent him the relevant documents, in the hope that there were grounds for clemency.

Hurst opened the file in the basement of his house in Austin, which served as a laboratory and an office, and was cluttered with microscopes and diagrams of half-finished experiments. Hurst was nearly six and half feet tall, though his stooped shoulders made him seem considerably shorter, and he had a gaunt face that was partly shrouded by long gray hair. He was wearing his customary outfit: black shoes, black socks, a black T-shirt, and loose-fitting black pants supported by black suspenders. In his mouth was a wad of chewing tobacco.

A child prodigy who was raised by a sharecropper during the Great Depression, Hurst used to prowl junk yards, collecting magnets and copper wires in order to build radios and other contraptions. In the early sixties, he received a Ph.D. in chemistry from Cambridge University, where he started to experiment with fluorine and other explosive chemicals, and once detonated his lab. Later, he worked as the chief scientist on secret weapons programs for several American companies, designing rockets and deadly fire bombs—or what he calls “god-awful things.” He helped patent what has been described, with only slight exaggeration, as “the world’s most powerful nonnuclear explosive”: an Astrolite bomb. He experimented with toxins so lethal that a fraction of a drop would rot human flesh, and in his laboratory he often had to wear a pressurized moon suit; despite such precautions, exposure to chemicals likely caused his liver to fail, and in 1994 he required a transplant. Working on what he calls “the dark side of arson,” he retrofitted napalm bombs with Astrolite, and developed ways for covert operatives in Vietnam to create bombs from local materials, such as chicken manure and sugar. He also perfected a method for making an exploding T-shirt by nitrating its fibres.

His conscience eventually began pricking him. “One day, you wonder, What the hell am I doing?” he recalls. He left the defense industry, and went on to invent the Mylar balloon, an improved version of Liquid Paper, and Kinepak, a kind of explosive that reduces the risk of accidental detonation. Because of his extraordinary knowledge of fire and explosives, companies in civil litigation frequently sought his help in determining the cause of a blaze. By the nineties, Hurst had begun devoting significant time to criminal-arson cases, and, as he was exposed to the methods of local and state fire investigators, he was shocked by what he saw.

Many arson investigators, it turned out, had only a high-school education. In most states, in order to be certified, investigators had to take a forty-hour course on fire investigation, and pass a written exam. Often, the bulk of an investigator’s training came on the job, learning from “old-timers” in the field, who passed down a body of wisdom about the telltale signs of arson, even though a study in 1977 warned that there was nothing in “the scientific literature to substantiate their validity.”

In 1992, the National Fire Protection Association, which promotes fire prevention and safety, published its first scientifically based guidelines to arson investigation. Still, many arson investigators believed that what they did was more an art than a science—a blend of experience and intuition. In 1997, the International Association of Arson Investigators filed a legal brief arguing that arson sleuths should not be bound by a 1993 Supreme Court decision requiring experts who testified at trials to adhere to the scientific method. What arson sleuths did, the brief claimed, was “less scientific.” By 2000, after the courts had rejected such claims, arson investigators increasingly recognized the scientific method, but there remained great variance in the field, with many practitioners still relying on the unverified techniques that had been used for generations. “People investigated fire largely with a flat-earth approach,” Hurst told me. “It looks like arson—therefore, it’s arson.” He went on, “My view is you have to have a scientific basis. Otherwise, it’s no different than witch-hunting.”

In 1998, Hurst investigated the case of a woman from North Carolina named Terri Hinson, who was charged with setting a fire that killed her seventeen-month-old son, and faced the death penalty. Hurst ran a series of experiments re-creating the conditions of the fire, which suggested that it had not been arson, as the investigators had claimed; rather, it had started accidentally, from a faulty electrical wire in the attic. Because of this research, Hinson was freed. John Lentini, a fire expert and the author of a leading scientific textbook on arson, describes Hurst as “brilliant.” A Texas prosecutor once told the Chicago Tribune, of Hurst, “If he says it was an arson fire, then it was. If he says it wasn’t, then it wasn’t.”

Hurst’s patents yielded considerable royalties, and he could afford to work pro bono on an arson case for months, even years. But he received the files on Willingham’s case only a few weeks before Willingham was scheduled to be executed. As Hurst looked through the case records, a statement by Manuel Vasquez, the state deputy fire marshal, jumped out at him. Vasquez had testified that, of the roughly twelve hundred to fifteen hundred fires he had investigated, “most all of them” were arson. This was an oddly high estimate; the Texas State Fire Marshals Office typically found arson in only fifty per cent of its cases.

Hurst was also struck by Vasquez’s claim that the Willingham blaze had “burned fast and hot” because of a liquid accelerant. The notion that a flammable or combustible liquid caused flames to reach higher temperatures had been repeated in court by arson sleuths for decades. Yet the theory was nonsense: experiments have proved that wood and gasoline-fuelled fires burn at essentially the same temperature.

Vasquez and Fogg had cited as proof of arson the fact that the front door’s aluminum threshold had melted. “The only thing that can cause that to react is an accelerant,” Vasquez said. Hurst was incredulous. A natural-wood fire can reach temperatures as high as two thousand degrees Fahrenheit—far hotter than the melting point for aluminum alloys, which ranges from a thousand to twelve hundred degrees. And, like many other investigators, Vasquez and Fogg mistakenly assumed that wood charring beneath the aluminum threshold was evidence that, as Vasquez put it, “a liquid accelerant flowed underneath and burned.” Hurst had conducted myriad experiments showing that such charring was caused simply by the aluminum conducting so much heat. In fact, when liquid accelerant is poured under a threshold a fire will extinguish, because of a lack of oxygen. (Other scientists had reached the same conclusion.) “Liquid accelerants can no more burn under an aluminum threshold than can grease burn in a skillet even with a loose-fitting lid,” Hurst declared in his report on the Willingham case.

Hurst then examined Fogg and Vasquez’s claim that the “brown stains” on Willingham’s front porch were evidence of “liquid accelerant,” which had not had time to soak into the concrete. Hurst had previously performed a test in his garage, in which he poured charcoal-lighter fluid on the concrete floor, and lit it. When the fire went out, there were no brown stains, only smudges of soot. Hurst had run the same experiment many times, with different kinds of liquid accelerants, and the result was always the same. Brown stains were common in fires; they were usually composed of rust or gunk from charred debris that had mixed with water from fire hoses.

Another crucial piece of evidence implicating Willingham was the “crazed glass” that Vasquez had attributed to the rapid heating from a fire fuelled with liquid accelerant. Yet, in November of 1991, a team of fire investigators had inspected fifty houses in the hills of Oakland, California, which had been ravaged by brush fires. In a dozen houses, the investigators discovered crazed glass, even though a liquid accelerant had not been used. Most of these houses were on the outskirts of the blaze, where firefighters had shot streams of water; as the investigators later wrote in a published study, they theorized that the fracturing had been induced by rapid cooling, rather than by sudden heating—thermal shock had caused the glass to contract so quickly that it settled disjointedly. The investigators then tested this hypothesis in a laboratory. When they heated glass, nothing happened. But each time they applied water to the heated glass the intricate patterns appeared. Hurst had seen the same phenomenon when he had blowtorched and cooled glass during his research at Cambridge. In his report, Hurst wrote that Vasquez and Fogg’s notion of crazed glass was no more than an “old wives’ tale.”

Hurst then confronted some of the most devastating arson evidence against Willingham: the burn trailer, the pour patterns and puddle configurations, the V-shape and other burn marks indicating that the fire had multiple points of origin, the burning underneath the children’s beds. There was also the positive test for mineral spirits by the front door, and Willingham’s seemingly implausible story that he had run out of the house without burning his bare feet.

As Hurst read through more of the files, he noticed that Willingham and his neighbors had described the windows in the front of the house suddenly exploding and flames roaring forth. It was then that Hurst thought of the legendary Lime Street Fire, one of the most pivotal in the history of arson investigation.

On the evening of October 15, 1990, a thirty-five-year-old man named Gerald Wayne Lewis was found standing in front of his house on Lime Street, in Jacksonville, Florida, holding his three-year-old son. His two-story wood-frame home was engulfed in flames. By the time the fire had been extinguished, six people were dead, including Lewis’s wife. Lewis said that he had rescued his son but was unable to get to the others, who were upstairs.

When fire investigators examined the scene, they found the classic signs of arson: low burns along the walls and floors, pour patterns and puddle configurations, and a burn trailer running from the living room into the hallway. Lewis claimed that the fire had started accidentally, on a couch in the living room—his son had been playing with matches. But a V-shaped pattern by one of the doors suggested that the fire had originated elsewhere. Some witnesses told authorities that Lewis seemed too calm during the fire and had never tried to get help. According to the Los Angeles Times, Lewis had previously been arrested for abusing his wife, who had taken out a restraining order against him. After a chemist said that he had detected the presence of gasoline on Lewis’s clothing and shoes, a report by the sheriff’s office concluded, “The fire was started as a result of a petroleum product being poured on the front porch, foyer, living room, stairwell and second floor bedroom.” Lewis was arrested and charged with six counts of murder. He faced the death penalty.

Subsequent tests, however, revealed that the laboratory identification of gasoline was wrong. Moreover, a local news television camera had captured Lewis in a clearly agitated state at the scene of the fire, and investigators discovered that at one point he had jumped in front of a moving car, asking the driver to call the Fire Department.

Seeking to bolster their theory of the crime, prosecutors turned to John Lentini, the fire expert, and John DeHaan, another leading investigator and textbook author. Despite some of the weaknesses of the case, Lentini told me that, given the classic burn patterns and puddle configurations in the house, he was sure that Lewis had set the fire: “I was prepared to testify and send this guy to Old Sparky”—the electric chair.

To discover the truth, the investigators, with the backing of the prosecution, decided to conduct an elaborate experiment and re-create the fire scene. Local officials gave the investigators permission to use a condemned house next to Lewis’s home, which was about to be torn down. The two houses were virtually identical, and the investigators refurbished the condemned one with the same kind of carpeting, curtains, and furniture that had been in Lewis’s home. The scientists also wired the building with heat and gas sensors that could withstand fire. The cost of the experiment came to twenty thousand dollars. Without using liquid accelerant, Lentini and DeHaan set the couch in the living room on fire, expecting that the experiment would demonstrate that Lewis’s version of events was implausible.

The investigators watched as the fire quickly consumed the couch, sending upward a plume of smoke that hit the ceiling and spread outward, creating a thick layer of hot gases overhead—an efficient radiator of heat. Within three minutes, this cloud, absorbing more gases from the fire below, was banking down the walls and filling the living room. As the cloud approached the floor, its temperature rose, in some areas, to more than eleven hundred degrees Fahrenheit. Suddenly, the entire room exploded in flames, as the radiant heat ignited every piece of furniture, every curtain, every possible fuel source, even the carpeting. The windows shattered.

The fire had reached what is called “flashover”—the point at which radiant heat causes a fire in a room to become a room on fire. Arson investigators knew about the concept of flashover, but it was widely believed to take much longer to occur, especially without a liquid accelerant. From a single fuel source—a couch—the room had reached flashover in four and a half minutes.

Because all the furniture in the living room had ignited, the blaze went from a fuel-controlled fire to a ventilation-controlled fire—or what scientists call “post-flashover.” During post-flashover, the path of the fire depends on new sources of oxygen, from an open door or window. One of the fire investigators, who had been standing by an open door in the living room, escaped moments before the oxygen-starved fire roared out of the room into the hallway—a fireball that caused the corridor to go quickly into flashover as well, propelling the fire out the front door and onto the porch.

After the fire was extinguished, the investigators inspected the hallway and living room. On the floor were irregularly shaped burn patterns that perfectly resembled pour patterns and puddle configurations. It turned out that these classic signs of arson can also appear on their own, after flashover. With the naked eye, it is impossible to distinguish between the pour patterns and puddle configurations caused by an accelerant and those caused naturally by post-flashover. The only reliable way to tell the difference is to take samples from the burn patterns and test them in a laboratory for the presence of flammable or combustible liquids.

During the Lime Street experiment, other things happened that were supposed to occur only in a fire fuelled by liquid accelerant: charring along the base of the walls and doorways, and burning under furniture. There was also a V-shaped pattern by the living-room doorway, far from where the fire had started on the couch. In a small fire, a V-shaped burn mark may pinpoint where a fire began, but during post-flashover these patterns can occur repeatedly, when various objects ignite.

One of the investigators muttered that they had just helped prove the defense’s case. Given the reasonable doubt raised by the experiment, the charges against Lewis were soon dropped. The Lime Street experiment had demolished prevailing notions about fire behavior. Subsequent tests by scientists showed that, during post-flashover, burning under beds and furniture was common, entire doors were consumed, and aluminum thresholds melted.

John Lentini says of the Lime Street Fire, “This was my epiphany. I almost sent a man to die based on theories that were a load of crap.”

Hurst next examined a floor plan of Willingham’s house that Vasquez had drawn, which delineated all the purported pour patterns and puddle configurations. Because the windows had blown out of the children’s room, Hurst knew that the fire had reached flashover. With his finger, Hurst traced along Vasquez’s diagram the burn trailer that had gone from the children’s room, turned right in the hallway, and headed out the front door. John Jackson, the prosecutor, had told me that the path was so “bizarre” that it had to have been caused by a liquid accelerant. But Hurst concluded that it was a natural product of the dynamics of fire during post-flashover. Willingham had fled out the front door, and the fire simply followed the ventilation path, toward the opening. Similarly, when Willingham had broken the windows in the children’s room, flames had shot outward.

Hurst recalled that Vasquez and Fogg had considered it impossible for Willingham to have run down the burning hallway without scorching his bare feet. But if the pour patterns and puddle configurations were a result of a flashover, Hurst reasoned, then they were consonant with Willingham’s explanation of events. When Willingham exited his bedroom, the hallway was not yet on fire; the flames were contained within the children’s bedroom, where, along the ceiling, he saw the “bright lights.” Just as the investigator safely stood by the door in the Lime Street experiment seconds before flashover, Willingham could have stood close to the children’s room without being harmed. (Prior to the Lime Street case, fire investigators had generally assumed that carbon monoxide diffuses quickly through a house during a fire. In fact, up until flashover, levels of carbon monoxide can be remarkably low beneath and outside the thermal cloud.) By the time the Corsicana fire achieved flashover, Willingham had already fled outside and was in the front yard.

Vasquez had made a videotape of the fire scene, and Hurst looked at the footage of the burn trailer. Even after repeated viewings, he could not detect three points of origin, as Vasquez had. (Fogg recently told me that he also saw a continuous trailer and disagreed with Vasquez, but added that nobody from the prosecution or the defense ever asked him on the stand about his opinion on the subject.)

After Hurst had reviewed Fogg and Vasquez’s list of more than twenty arson indicators, he believed that only one had any potential validity: the positive test for mineral spirits by the threshold of the front door. But why had the fire investigators obtained a positive reading only in that location? According to Fogg and Vasquez’s theory of the crime, Willingham had poured accelerant throughout the children’s bedroom and down the hallway. Officials had tested extensively in these areas—including where all the pour patterns and puddle configurations were—and turned up nothing. Jackson told me that he “never did understand why they weren’t able to recover” positive tests in these parts.

Hurst found it hard to imagine Willingham pouring accelerant on the front porch, where neighbors could have seen him. Scanning the files for clues, Hurst noticed a photograph of the porch taken before the fire, which had been entered into evidence. Sitting on the tiny porch was a charcoal grill. The porch was where the family barbecued. Court testimony from witnesses confirmed that there had been a grill, along with a container of lighter fluid, and that both had burned when the fire roared onto the porch during post-flashover. By the time Vasquez inspected the house, the grill had been removed from the porch, during cleanup. Though he cited the container of lighter fluid in his report, he made no mention of the grill. At the trial, he insisted that he had never been told of the grill’s earlier placement. Other authorities were aware of the grill but did not see its relevance. Hurst, however, was convinced that he had solved the mystery: when firefighters had blasted the porch with water, they had likely spread charcoal-lighter fluid from the melted container.

Without having visited the fire scene, Hurst says, it was impossible to pinpoint the cause of the blaze. But, based on the evidence, he had little doubt that it was an accidental fire—one caused most likely by the space heater or faulty electrical wiring. It explained why there had never been a motive for the crime. Hurst concluded that there was no evidence of arson, and that a man who had already lost his three children and spent twelve years in jail was about to be executed based on “junk science.” Hurst wrote his report in such a rush that he didn’t pause to fix the typos.

V

“I am a realist and I will not live a fantasy,” Willingham once told Gilbert about the prospect of proving his innocence. But in February, 2004, he began to have hope. Hurst’s findings had helped to exonerate more than ten people. Hurst even reviewed the scientific evidence against Willingham’s friend Ernest Willis, who had been on death row for the strikingly similar arson charge. Hurst says, “It was like I was looking at the same case. Just change the names.” In his report on the Willis case, Hurst concluded that not “a single item of physical evidence . . . supports a finding of arson.” A second fire expert hired by Ori White, the new district attorney in Willis’s district, concurred. After seventeen years on death row, Willis was set free. “I don’t turn killers loose,” White said at the time. “If Willis was guilty, I’d be retrying him right now. And I’d use Hurst as my witness. He’s a brilliant scientist.” White noted how close the system had come to murdering an innocent man. “He did not get executed, and I thank God for that,” he said.

On February 13th, four days before Willingham was scheduled to be executed, he got a call from Reaves, his attorney. Reaves told him that the fifteen members of the Board of Pardons and Paroles, which reviews an application for clemency and had been sent Hurst’s report, had made their decision.

“What is it?” Willingham asked.

“I’m sorry,” Reaves said. “They denied your petition.”

The vote was unanimous. Reaves could not offer an explanation: the board deliberates in secret, and its members are not bound by any specific criteria. The board members did not even have to review Willingham’s materials, and usually don’t debate a case in person; rather, they cast their votes by fax—a process that has become known as “death by fax.” Between 1976 and 2004, when Willingham filed his petition, the State of Texas had approved only one application for clemency from a prisoner on death row. A Texas appellate judge has called the clemency system “a legal fiction.” Reaves said of the board members, “They never asked me to attend a hearing or answer any questions.”

The Innocence Project obtained, through the Freedom of Information Act, all the records from the governor’s office and the board pertaining to Hurst’s report. “The documents show that they received the report, but neither office has any record of anyone acknowledging it, taking note of its significance, responding to it, or calling any attention to it within the government,” Barry Scheck said. “The only reasonable conclusion is that the governor’s office and the Board of Pardons and Paroles ignored scientific evidence.”

LaFayette Collins, who was a member of the board at the time, told me of the process, “You don’t vote guilt or innocence. You don’t retry the trial. You just make sure everything is in order and there are no glaring errors.” He noted that although the rules allowed for a hearing to consider important new evidence, “in my time there had never been one called.” When I asked him why Hurst’s report didn’t constitute evidence of “glaring errors,” he said, “We get all kinds of reports, but we don’t have the mechanisms to vet them.” Alvin Shaw, another board member at the time, said that the case didn’t “ring a bell,” adding, angrily, “Why would I want to talk about it?” Hurst calls the board’s actions “unconscionable.”

Though Reaves told Willingham that there was still a chance that Governor Perry might grant a thirty-day stay, Willingham began to prepare his last will and testament. He had earlier written Stacy a letter apologizing for not being a better husband and thanking her for everything she had given him, especially their three daughters. “I still know Amber’s voice, her smile, her cool Dude saying and how she said: I wanna hold you! Still feel the touch of Karmon and Kameron’s hands on my face.” He said that he hoped that “some day, somehow the truth will be known and my name cleared.”

He asked Stacy if his tombstone could be erected next to their children’s graves. Stacy, who had for so long expressed belief in Willingham’s innocence, had recently taken her first look at the original court records and arson findings. Unaware of Hurst’s report, she had determined that Willingham was guilty. She denied him his wish, later telling a reporter, “He took my kids away from me.”

Gilbert felt as if she had failed Willingham. Even before his pleas for clemency were denied, she told him that all she could give him was her friendship. He told her that it was enough “to be a part of your life in some small way so that in my passing I can know I was at last able to have felt the heart of another who might remember me when I’m gone.” He added, “There is nothing to forgive you for.” He told her that he would need her to be present at his execution, to help him cope with “my fears, thoughts, and feelings.”

On February 17th, the day he was set to die, Willingham’s parents and several relatives gathered in the prison visiting room. Plexiglas still separated Willingham from them. “I wish I could touch and hold both of you,” Willingham had written to them earlier. “I always hugged Mom but I never hugged Pop much.”

As Willingham looked at the group, he kept asking where Gilbert was. Gilbert had recently been driving home from a store when another car ran a red light and smashed into her. Willingham used to tell her to stay in her kitchen for a day, without leaving, to comprehend what it was like to be confined in prison, but she had always found an excuse not to do it. Now she was paralyzed from the neck down.

While she was in an intensive-care unit, she had tried to get a message to Willingham, but apparently failed. Gilbert’s daughter later read her a letter that Willingham had sent her, telling her how much he had grown to love her. He had written a poem: “Do you want to see beauty—like you have never seen? / Then close your eyes, and open your mind, and come along with me.”

Gilbert, who spent years in physical rehabilitation, gradually regaining motion in her arms and upper body, says, “All that time, I thought I was saving Willingham, and I realized then that he was saving me, giving me the strength to get through this. I know I will one day walk again, and I know it is because Willingham showed me the kind of courage it takes to survive.”

Willingham had requested a final meal, and at 4 p.m. on the seventeenth he was served it: three barbecued pork ribs, two orders of onion rings, fried okra, three beef enchiladas with cheese, and two slices of lemon cream pie. He received word that Governor Perry had refused to grant him a stay. (A spokesperson for Perry says, “The Governor made his decision based on the facts of the case.”) Willingham’s mother and father began to cry. “Don’t be sad, Momma,” Willingham said. “In fifty-five minutes, I’m a free man. I’m going home to see my kids.” Earlier, he had confessed to his parents that there was one thing about the day of the fire he had lied about. He said that he had never actually crawled into the children’s room. “I just didn’t want people to think I was a coward,” he said. Hurst told me, “People who have never been in a fire don’t understand why those who survive often can’t rescue the victims. They have no concept of what a fire is like.”

The warden told Willingham that it was time. Willingham, refusing to assist the process, lay down; he was carried into a chamber eight feet wide and ten feet long. The walls were painted green, and in the center of the room, where an electric chair used to be, was a sheeted gurney. Several guards strapped Willingham down with leather belts, snapping buckles across his arms and legs and chest. A medical team then inserted intravenous tubes into his arms. Each official had a separate role in the process, so that no one person felt responsible for taking a life.

Willingham had asked that his parents and family not be present in the gallery during this process, but as he looked out he could see Stacy watching. The warden pushed a remote control, and sodium thiopental, a barbiturate, was pumped into Willingham’s body. Then came a second drug, pancuronium bromide, which paralyzes the diaphragm, making it impossible to breathe. Finally, a third drug, potassium chloride, filled his veins, until his heart stopped, at 6:20 p.m. On his death certificate, the cause was listed as “Homicide.”

After his death, his parents were allowed to touch his face for the first time in more than a decade. Later, at Willingham’s request, they cremated his body and secretly spread some of his ashes over his children’s graves. He had told his parents, “Please don’t ever stop fighting to vindicate me.”

In December, 2004, questions about the scientific evidence in the Willingham case began to surface. Maurice Possley and Steve Mills, of the Chicago Tribune, had published an investigative series on flaws in forensic science; upon learning of Hurst’s report, Possley and Mills asked three fire experts, including John Lentini, to examine the original investigation. The experts concurred with Hurst’s report. Nearly two years later, the Innocence Project commissioned Lentini and three other top fire investigators to conduct an independent review of the arson evidence in the Willingham case. The panel concluded that “each and every one” of the indicators of arson had been “scientifically proven to be invalid.”

In 2005, Texas established a government commission to investigate allegations of error and misconduct by forensic scientists. The first cases that are being reviewed by the commission are those of Willingham and Willis. In mid-August, the noted fire scientist Craig Beyler, who was hired by the commission, completed his investigation. In a scathing report, he concluded that investigators in the Willingham case had no scientific basis for claiming that the fire was arson, ignored evidence that contradicted their theory, had no comprehension of flashover and fire dynamics, relied on discredited folklore, and failed to eliminate potential accidental or alternative causes of the fire. He said that Vasquez’s approach seemed to deny “rational reasoning” and was more “characteristic of mystics or psychics.” What’s more, Beyler determined that the investigation violated, as he put it to me, “not only the standards of today but even of the time period.” The commission is reviewing his findings, and plans to release its own report next year. Some legal scholars believe that the commission may narrowly assess the reliability of the scientific evidence. There is a chance, however, that Texas could become the first state to acknowledge officially that, since the advent of the modern judicial system, it had carried out the “execution of a legally and factually innocent person.”

Just before Willingham received the lethal injection, he was asked if he had any last words. He said, “The only statement I want to make is that I am an innocent man convicted of a crime I did not commit. I have been persecuted for twelve years for something I did not do. From God’s dust I came and to dust I will return, so the Earth shall become my throne.” ♦

Published in the print edition of the September 7, 2009, issue.

David Grann, a staff writer at The New Yorker since 2003, is the author of, most recently, “The White Darkness,” and of “Killers of the Flower Moon: The Osage Murders and the Birth of the F.B.I.,” which won an Edgar Award and was a finalist for the National Book Award.

Please go to our Moodle Page and under "Class 11" you will find the prompt and submission folder for Writing Reflection #11.