Skip navigation
The Habeas Citebook: Prosecutorial Misconduct - Header
× You have 2 more free articles available this month. Subscribe today.

Faulty Forensics and Wrongful Convictions

The Innocence Project, a national litigation and public policy organization dedicated to exonerating innocent prisoners, focuses on DNA testing and excludes most other types of exonerations from its database. As of August 2020, the Innocence Project database showed 380 people who had been exonerated by DNA testing, 173 (45%) of whom had “invalidated or improper forensic science” as a “contributing cause.”

The National Registry of Exonerations (“NRE”), a database managed by the Newkirk Center for Science and Society at the University of California at Irvine, the University of Michigan Law School, and Michigan State University College of Law, is more inclusive in its definition of “wrongful convictions” but still requires a “post-conviction, pre-exoneration re-examination of the evidence” showing innocence for an exoneration to be included in the database. It lists 2,657 exonerations occurring between 1989 and August 2020, with 24% of its exonerations showing “False or Misleading Forensic Evidence” as a contributing factor.

However, there are discrepancies between the two databases with respect to what percentage of wrongful convictions are based upon faulty forensics. The discrepancies are likely caused by confusion resulting from the two databases using different terms for the factors contributing to a wrongful conviction, the fact that several factors are often present in an individual case, and the fact that a police crime lab technician committing perjury by claiming to have results from a test never performed may be listed as “official misconduct” instead of “unvalidated or improper forensic science” in the NRE database.

Notably, both databases exclude thousands of convictions that were demonstrably the result of faulty forensics. These include the crime lab scandals in which thousands of cases were tainted by the behavior of crime lab technicians. [CLN, Jan. 2018, p. l]. For instance, hundreds of Massachusetts drug cases were thrown out because state crime lab technician Annie Dookhan falsified results of drug tests she did not perform. Thousands of drug tests were tainted by Massachusetts state crime lab technician Sonja Farak, who stole drugs – including the standards used by all of the lab technicians to test against to prove a drug was present – and replaced them with baking soda and other household chemicals. 

Texas Department of Public Safety crime lab technician Jonathan Salvador fabricated drug tests, tainting at least 5,000 cases in 30 counties. Crime lab technicians in California and Montana stole drugs, tainting an unknown number of cases. An Oregon crime lab technician stole drugs and gave false testimony in thousands of cases between 2007 and 2015.

The St. Paul Police Crime Lab in Minnesota had to be shut down in 2013 after it was discovered that incompetence and contamination tainted 1,700 drug cases. The Houston Police Department’s Crime Lab’s DNA Division had to close after a news report by television station KHOU, using an independent DNA lab to test samples, showed that police lab technicians routinely misinterpreted the results of DNA testing.

Apparently, none of the thousands of wrongful convictions due to the aforementioned crime lab scandals are counted in the Innocence Project or NRE databases. Yet they are clearly wrongful convictions caused by faulty forensics. Thus, the number of wrongful convictions attributable, at least in part, to faulty forensics is significantly greater than either the Innocence Project or the NRE’s figures when taking a broader view of wrongful convictions than the methodologies utilized by those organizations.

Another category of wrongful convictions caused by faulty forensics came to light in Houston, Texas, when Kim Ogg, the then-newly elect Harris County district attorney, insisted the crime lab re-test substances the police had confiscated as illegal drugs even when the defendant had already pleaded guilty. They found 136 cases where a defendant pleaded guilty to drug possession when the substance in question was not even an illegal drug. Many of the cases were felonies, but the innocent defendants were willing to plead guilty to avoid waiting many months in jail for the crime lab to test the substance and potentially many more months to receive a trial.

The Harris County cases are also not in the foregoing databases despite undeniable proof of innocence. Further, presumably every case resulted from faulty forensics – cheap field drug tests being used by the Houston police at that time that had a high error rate for false positives. Thus, faulty forensics can and often do result in wrongful conviction without a trial taking place. Since those same field drug tests were used by police departments around the country, one can only speculate on how many thousands of innocent people pleaded guilty after being confronted with the dual specter of a “scientific” field drug test and coercive pre-trial incarceration. 

Faulty Forensic Methods

A landmark report by the National Academy of Sciences (“NAS”) published in 2009 rocked the world of forensic sciences by showing that many of the routinely accepted forensic “pattern matching” methods, such as fingerprint comparison and ballistics, lack any scientific underpinning while some methods such as bite-mark comparison and hair comparison were simply unscientific in that it was impossible to “match” bite marks or hairs to individuals. In 2013, the National Institute of Standards and Technology issued a startling report on crime labs revealing a 68% error rate in DNA evidence among the 108 labs tested nationwide.

In December 2015, the National Commission on Forensic Sciences (“NCFS”) called for prosecutors to only use evidence processed by accredited labs. In 2016, the President’s Council of Advisors on Science and Technology (“PCAST”) issued a report finding many of the problems noted in the 2009 NAS report had been unaddressed. It contained recommendations to improve forensic science. However, prosecutors fought back. [PLN, March 2019, p. 31].

As then-U.S. Attorney General Loretta Lynch stated, “when used properly, forensic science evidence helps juries identify the guilty and clear the innocent. We appreciate [PCAST’s] contribution to the field of scientific inquiry, [but] the department will not be adopting the recommendations related to the admissibility for forensic science evidence.”

The problem intensified during the Trump era. In 2017, then-U.S. Attorney General Jeff Sessions disbanded the National Commission on Forensic Science – a partnership of independent forensic scientists, evidence examiners, lawyers, and law enforcement personnel formed in 2013 to improve the study and use of forensic science – and “replaced” it with the Forensic Science Working Group within the U.S. Department of Justice led by a prosecutor and with the very limited mission of developing a “uniform language” for scientific expert testimony in federal cases. 

Faulty Forensics Still Admissible
in Criminal Cases

Despite the pushback by prosecutors, in February 2016 the Texas Forensic Science Commission proposed a temporary moratorium on the use of bite-mark comparison analysis as evidence in criminal trials, noting that its validity has not been scientifically established. In doing so, Texas became the first official governmental entity to propose such a moratorium. The suspension came years after bite-mark evidence became the poster child for what is wrong with pattern comparison analysis in the NAS and PCAST reports.

The validity of hair-comparison evidence is similarly questionable. The NAS and PCAST reports both noted that there has been no scientific validation of the concept that a hair could be matched to a specific individual and this was likely not possible since each individual has hairs that vary in structural characteristics and color. The NAS report referred to forensic hair comparison as “highly unreliable,” and in 2015, the FBI admitted that its own examiners presented flawed microscopic hair comparison testimony an astonishing 95% of the time over the previous two decades. Nonetheless, no criminal court has held that bite-mark or hair-comparison evidence was inadmissible as of 2017.

Criminal courts continue to rely on this and other debunked forensic techniques that would not be admissible in a civil court proceeding. Why? The fault lies in the divide between civil cases and criminal cases in American courts and the influence of repeat litigators.

Unequal Pretrial Discovery

Civil litigants have enormous pretrial discovery rights, including admission requests, interrogatories, depositions, mandatory disclosure, and disclosure of potential expert witness testimony. With few exceptions, criminal defendants have no such discovery rights. Their discovery rights are limited to the requirements set forth in Brady v. Maryland, 373 U.S. 83 (1963), that prosecutors disclose any evidence favorable to the defense.

One obvious problem with Brady is that it depends upon the prosecutor to determine what is favorable to the defense. Another is that prosecutors face no real penalty for withholding evidence favorable to the defense, though California makes it a felony. At most, an appellate court might reverse the criminal conviction. But even that is not assured, as demonstrated by Turner v. United States, 137 S. Ct. 1885 (2017). In Turner, the prosecution withheld witness statements that pointed to another perpetrator in a 1984 robbery and murder. The Turner defendants had steadfastly maintained their innocence, and the prosecution admitted that the statements in question were favorable and had not been given to the defense. Instead, the Government maintained “the withheld evidence was not material.” The Supreme Court agreed, ruling that failure to disclose the statements did not run afoul of Brady. Turner serves as a textbook example of a criminal defendant’s limited right to pretrial discovery.

The pretrial phase of a case, whether criminal or civil, is when most cases are resolved. Obviously, the limited pretrial discovery in a criminal case puts the defendant at an extreme disadvantage compared to a civil litigant, especially when it comes to pretrial determination of the validity of forensic methods because the lack of pretrial discovery means the criminal defense attorney does not necessarily even know what kind of forensic evidence the prosecutor intends to introduce, much less how to challenge it.

The Standards for Forensic Evidence

Historically, for forensic evidence to be admissible in court, the party seeking its admission had to show that it was generally accepted in the particular field to which it belongs. Frye v. United States, 293 F. 1013 (DC Cir. 1923). This was superseded in Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993), in which the Supreme Court ruled that Federal Rule of Evidence 702 controls and requires trial courts to admit evidence that is deemed reliable under federal rules (general acceptance of principle underlying scientific evidence not necessary precondition to admissibility) and will assist the factfinder in understanding the evidence or determining a fact in issue. This effectively made judges, who typically have little or no scientific training, the gatekeepers of the admissibility of evidence allegedly based on scientific principles.

Rule 702 applies equally to federal civil and criminal litigants, so what can explain the fact that the same judges who conducted rigorous examinations of well-accepted scientific evidence in civil cases continue to admit or uphold the use of invalid “scientific evidence” such as bite-mark comparison long after it was debunked in the NAS report? The question applies equally to most of the states, whose rules of evidence are patterned after the federal rules and apply equally to civil and criminal trials.

The difference cannot be fully explained by judicial bias favoring the prosecution, although that surely exists, the lack of scientific acumen of judges or lawyers, or the difference in complexity of the issues. Judges and lawyers act competently in civil case Daubert hearings to determine the validity of scientific methods, and those methods are often more complex than the methods presented in criminal cases.

The answer to this mystery lies in who the repeat litigator is in each type of case. In criminal cases, the prosecutor is the repeat litigator while in civil cases, it is large corporations that repeatedly appear in court. Just as a tort plaintiff is unlikely to sue over the same issue again, the criminal defendant is unlikely to return to court with the same charges in the near future. Thus, the civil tort plaintiff and the criminal defendant are both interested in short-term gains – the outcome of a single trial – whereas corporate civil defendants and criminal prosecutors can focus on long-term goals. This could be as simple as forum shopping – waiting until a case is in front of a sympathetic judge, then using a favorable ruling to establish a precedent to be used in future cases. Likewise, repeat litigators soon discover which expert witnesses testify most persuasively and lock them in for repeat appearances while one-time litigators may have to hire an expert without knowing how persuasive the expert’s testimony will be.  

A corporate civil defendant facing an unfavorable precedent can still demand a Daubert hearing, which is required by the rules. By contrast, judges in criminal cases frequently bypass the Daubert hearing and admit forensic evidence based solely upon precedent, and the precedent is overwhelmingly favorable to the admission of the evidence. That is why bitemark evidence continues to be admitted even after it has been proven to have caused over two dozen wrongful convictions in which the defendant was later exonerated by DNA testing.

Even worse, some courts rely on case law, both precedential and persuasive, allowing admission of bitemark evidence purportedly because such evidence is generally accepted in the scientific community such that it can be judicially noticed (when a court declares a fact presented as true without a formal presentation of evidence). However, there are instances where the case cited to as justification for the admission of bitemark evidence involves a defendant who is subsequently exonerated as a result of DNA testing.

This is what happened when the West Virginia Supreme Court of Appeals took judicial notice of the “general acceptance” of bitemark evidence in State v. Armstrong, 369 S.E.2d 870 (W.Va. 1988), which relied on State v. Stinson, 397 N.W.2d 136 (Wisc. 1986), in making its decision. Armstrong became the controlling or persuasive precedent for numerous subsequent cases, but Robert Lee Stinson was exonerated by DNA testing in 2009. Thus, bad precedent begets bad precedent in criminal cases and leads to the continued admission of unvalidated scientific evidence, i.e., junk science.

Ironically, one of the rationales used for replacing the Frye “general acceptance” standard with the Daubert standard was that some “knowledge,” especially that which has been accepted for a long time, can creep into general acceptance without careful examination of its scientific reliability or validity. The NAS and PCAST reports showed that not only bitemark comparison and hair comparison lack scientific validation but also other widely accepted pattern matching methods, such as comparing fingerprints, groove marks on bullets, tool marks, fabrics, and footwear imprints have also never been scientifically validated.

Yet, instead of engendering Daubert challenges to those methods that might spur forensic scientists into finally validating their methodology, the reports resulted in opposition from prosecutors and politicians who believe they, not scientists, should determine whether a scientific method is valid. This is reminiscent of some politicians’ attitude toward the COVID-19 pandemic – ignoring the science of epidemiology and believing they can bend the outcome to the desired one simply by asserting their political will. Just as that attitude led to the worst possible response to the pandemic – widespread infection and needless deaths – it has led to the worst possible outcome for forensic science, with debunked and unvalidated methods continuing to be used by prosecutors and accepted as evidence in courts to wrongly convict and imprison or execute the factually innocent. 

An obvious solution to this problem is to permit criminal defendants greater pre-trial discovery. Ideally, this would be open-file discovery in which the defense is allowed to view the prosecutor’s entire file in a criminal case. Six states have open-file discovery laws, but they vary in the scope of what “open-file” means from “all evidence in the prosecutor’s file” in North Carolina to “exculpatory evidence only” in West Virginia. Full disclosure of the entire prosecutor’s file is the more desirable rule as it “do[es] not rely on the ethical judgment of a prosecutor involved in a fiercely competitive adversary trial process to determine what is exculpatory.”

Criminal defendants should also be given the right to a Daubert hearing when challenging the admissibility of forensic evidence. This could complicate criminal prosecutions and require the state to pay for experts for indigent criminal defendants, but the monetary impact would not be prohibitive since the vast majority of criminal defendants enter into a plea bargain – pleading guilty in exchange for a lighter sentence – and never go to trial. This would provide a desperately needed check on the introduction of unsound or unvalidated forensics. 

Routine Daubert hearings in criminal cases might also reduce the “CSI-effect,” a documented phenomenon in which people who watch television crime shows that present forensic scientists as infallible and the science they deploy as able to quickly determine truth and identify the guilty, that can and often does taint judges and juries. The CSI-like TV shows influence judges and juries to accept as ironclad the word of any forensic scientist presented by the prosecution while dismissing any issues raise by the defense.

The “CSI-effect” is magnified by overconfident forensic scientists who show neither doubt nor reservation when testifying and have been known to bolster their conclusions with fictitious statistics or overstate the strength of the evidence.

“A lot of the problem with forensic testimony is that the diagnosticity is overstated,” according to Professor Barbara O’Brien of the Michigan State University School of Law and author of an NRE report on exonerations in 2018. For instance, a hair sample at the crime scene that resembles a suspect’s hair “gets dressed up with this scientific certainty that isn’t justified,” resulting in wrongful convictions.

The Promise and Disappointment of Ex Parte Chaney

On December 19, 2018, the Texas Court of Criminal Appeals (“TCCA”) delivered its opinion in Ex parte Chaney, 563 S.W.3d 239 (Tex. Crim. App. 2018). Steven Mark Chaney had been convicted of the 1987 murder of a drug dealer to whom he owed a debt. His prosecution involved a litany of faulty forensics, including a bogus presence-of-blood test, a useless shoe sole pattern comparison, a lie about a thumbprint, and the centerpiece of the State’s case – a faulty bitemark comparison match. Despite the fact that Chaney had nine alibi witnesses who could account for his movements the entire day of the murder, on the strength of the forensics and the testimony of a waffling and self-interested witness testifying that Chaney said he had “cleared up [his debt] with John,” Chaney was convicted and sentenced to life in prison.

Chaney’s shoe had been seized by a detective because the detective thought the sole pattern matched a shoe print from the crime scene, and there were tiny brown spots on it. The problem with that was the sole pattern was used on around 50%-80% of all tennis shoes sold at the time. The detective took the shoe to a forensic blood technician who determined that the spots were not blood. Undaunted, the detective took the shoe to another blood technician, who later testified that the test showed blood, but there was too little present to know if it was human – yet the response time to the test reagent in the test indicated to her that it was human blood. Of course, the prosecutor never told Chaney about the first test showing no presence of blood at all.

The fingerprint was indeed Chaney’s, but he frequented his drug dealer’s home. So it wasn’t necessarily evidence of guilt that his fingerprint was present, especially since it’s impossible to determine forensically when a fingerprint was left.

Two forensic odontologists testified to the effect that the bitemarks made at the time of the murder were a “perfect match” to the biting surfaces of Chaney’s teeth to a “reasonable degree of dental certainty,” and there was a “one to a million chance” that they had not been made by Chaney. The forensic odontologist who testified at trial that the bitemarks were inflicted at the time of death and that there was a one to a million chance that someone other than Chaney made them changed his opinion after the trial, concluding instead that the bitemarks were in fact made two-to-three days before the time of death.

Chaney filed a state post-conviction petition for a writ of habeas corpus under a new Texas law that allowed prisoners to raise a claim of changes in forensic sciences or the opinions of forensic experts who testified at their trial. In Chaney’s case, it was both, as forensic odontology had recognized the uncertainty in the “science,” and the prosecution’s lead expert witness changed his opinion on matters to which he testified at trial.

The habeas trial court found that forensic odontology, as it relates to bitemark comparisons, had changed since Chaney’s conviction and that it was now accepted by the American Board of Forensic Odontology (“ABFO”) that a bitemark could not be matched to an individual.

The habeas court concluded that Chaney was entitled to relief as the bitemark evidence was central to the prosecution’s case even to the point that the prosecutor asked the jury to convict based upon the bitemark evidence during closing arguments.

The trial court also found that the testimony about the “one to a million chance” and that the bitemark was made at the time of the murder were false evidence. It also found that the prosecutor improperly withheld exculpatory evidence about the first presence-of-blood test. The court recommended that relief be granted on the basis of actual innocence.

The TCCA adopted the trial court’s findings and noted that new DNA testing excluded Chaney from every item tested, including hairs found in the hand of a victim. The State even admitted to finding a “vast amount of evidence” supporting Chaney’s actual innocence claim during a post-conviction investigation. The TCCA concluded Chaney had proven he was actually innocent and set aside the conviction. What the TCCA failed to do was instruct that forensic odontology evidence should not be admitted as evidence in a criminal prosecution. Thus, the hope that the Chaney Court would become the first court to ban bitemark analysis came to naught.

“People continue to be exonerated, and we continue to see the courts let it in,” said Duke University Law School professor Brandon Garrett.

People Still Being Convicted
by Discredited Forensics

In 2018, two Ohio men, one a former police sergeant, lost the appeals of their murder convictions that were obtained with the help of bitemark evidence. Blood-spatter evidence was used during murder prosecutions in Seattle, New Mexico in 2018, and North Carolina in 2017 despite the NAS report characterizing it as having “enormous” uncertainties. Shoe print comparison evidence was used to convict men in Tennessee and Pennsylvania in 2018, despite government researchers’ conclusion that it could not be used to generate a match to a particular shoe or person.

“We are going backward,” said Virginia criminal defense attorney Betty Layne DesPortes, a former president of the American Academy of Forensic Sciences, a professional organization of attorneys and forensic examiners. “We need to get past the trust-me phase and get us to the show-us phase, because that’s what science is supposed to be – based not on faith, but on results.”

The Real Cost of Faulty Forensics

Faulty forensics cost society in many ways. It removes the innocent from society while leaving the guilty free to victimize more people. It undermines our collective faith in the criminal justice system when, years later, defendants are exonerated – sometimes posthumously. It can even lessen our trust in science, undermining one of the pillars of Western civilization. But the people who pay the steepest price for faulty forensics are those who are incarcerated – often for decades – or executed for crimes they did not commit.  

Exoneration is not the same as being restored or getting your life back. Wrongful incarceration leaves scars physically, mentally, emotionally, and financially. Even states that offer generous compensation cannot remove those scars.

Worse, exoneration is hardly assured. The best estimates are that the known felony exonerations represent less than 5% of the number of wrongful convictions.

Glenn Payne’s 1990 conviction for child sexual abuse was vacated on January 25, 2018. The two-year-old alleged victim, a neighbor, was discovered sleeping in the cold outside the house in torn and stained clothing. Her underwear was missing.

The previous week, Payne had accompanied his mother, who was babysitting the child, to the child’s house. Police searched him and discovered a hair. Another hair was found on a tablecloth the child had been using as a blanket, which was found in a nearby field.

During Payne’s trial, a laboratory analyst testified that there was a 1 in 2,700 chance that the hair recovered from Payne came from someone other than the child and a 1 in 48 chance that the hair from the tablecloth came from someone other than Payne. Additionally, a second-year pediatric resident testified that he found evidence of anal penetration. Payne was convicted and sentenced to 27 years’ imprisonment. He was paroled in 2005.

After Payne was released, the FBI determined that microscopic hair analysis could not be used to “match” two hairs, the hair examiner recanted, the doctor determined that he had mistaken a case of anal strep for a sexual assault, and the signs of the alleged sexual assault disappeared after treatment with an antibiotic. Further, the child, now a young woman, repeatedly denied that anybody had hurt her, but said she went outside looking for her cat and dog.

Because there was no other course of action open to an innocent person who has completed his sentence, the Northern California Innocence Project, which had been assisting Payne, approached the Santa Clara County District Attorney’s conviction integrity unit. They filed a joint motion to vacate the conviction, which was granted on January 25, 2018.

Other Cases

• In 1990, Richard Beranek was convicted of the 1987 rape of a 28-year-old Wisconsin woman and sentenced to 243 years in prison. Despite having six alibi witnesses that he was in North Dakota when the crime occurred, he was convicted based on the victim’s confident identification of him as the perpetrator in a photo lineup two years after the crime and bogus hair comparison testimony.

After the assault, the hair was found in the victim’s underwear along with some semen. A forensic hair analyst testified that it was Beranek’s. DNA testing performed in 2017 showed that neither the hair nor semen came from Beranek. Testing of men’s underwear found at the crime scene revealed DNA that did not belong to Beranek. Based on this, a judge granted him a new trial.

The prosecutor threatened to retry Beranek based solely on the victim’s identification, but dismissed the charges in 2018 after the results from the DNA testing of the men’s underwear came back. [This case also implicates another issue that leads to an alarming number of wrongful convictions, viz., eyewitness misidentification. See cover story for December 2018 issue of CLN.]

• Kentucky residents Jeffery Clark and Keith Hardin were convicted of murdering Hardin’s girlfriend after hairs found in her hand were allegedly “matched” to Hardin. Other hairs found in her hand did not “match” Clark or Hardin. Police ignored a woman’s report that another man admitted murdering the woman during an argument and physical altercation, instead concentrating on Clark and Hardin with an implausible theory of Satanic sacrifice. Clark and Hardin were convicted of murder and sentenced to life.

With the assistance of Innocence Projects in New York and Kentucky, after a four-year court battle, DNA testing proved that the hair “matched” to Hardin at trial belonged to neither Clark nor Hardin nor the victim. Further, the lead detective in the case, who had falsely testified that Hardin admitted performing animal sacrifices and desiring to perform human sacrifices, had falsely testified that a different defendant confessed to a crime in an unrelated case, and had erased the recording of the defendant’s actual statement, a crime for which he was later indicted.

All of this led a judge to grant Clark and Hardin a new trial in 2016. The prosecutor appealed, then indicted them on new charges of kidnapping and perjury. The perjury charges were based on Clark and Hardin “admitting” the crime before the parole board in an effort to be granted parole. The Kentucky Supreme Court upheld the granting of a new trial, and the judge dismissed the new charges in January 2018. The murder charges were dismissed the next month.

• Elmer Daniels was convicted of a 1980 Delaware child sex abuse charge. His first-degree rape conviction was based on the faulty testimony of an FBI hair analyst. A 15-year-old witness, who may have had sex with the 15-year-old victim, pointed police to then-18-year-old Daniels whom he said he recognized because they were in the same middle-school homeroom – a claim later proven to be false. Daniels was exonerated in 2018 after the FBI admitted its analyst’s testimony “exceeded the limits of science.”

• In 2014, Courtney Hayden called Corpus Christi, Texas, police to report she shot a man who had broken into her house and tried to rob her. Police theorized she shot the man when they fought over proceedings from a previous robbery they committed. A firearms examiner testified that the man was shot from three feet away. A medical examiner agreed that the shotgun was at least three feet away, and the prosecutor used this distance to argue the shooting was not self-defense.

Less than a month after Hayden was sentenced to 40 years in prison, the defense filed a motion for new trial because the prosecutor failed to reveal that the medical examiner initially determined that the man was shot at close range and only changed his estimate to three feet or more after the prosecutor told him a ballistics analyst reported the distance of three feet. The conviction was vacated on January 31, 2016, and the charge dismissed in 2018.

• Patrick Pursley was convicted of a 1993 Rockford, Illinois, murder based largely on ballistics evidence that “matched” the groove marks of a bullet and markings on a shell casing found at the crime scene to a gun he owned. He spent the next 23 years in prison proclaiming his innocence and fighting to get the evidence retested using the more advanced Integrated Ballistics Identification System (“IBIS”), even lobbying state lawmakers to pass legislation to permit the testing. The IBIS testing, which uses very high-resolution and multi-dimensional images to compare bullets and casings, showed Pursley’s gun could not be the murder weapon.

In March 2017, a state judge vacated Pursley’s conviction and released him on bail but allowed a retrial of the murder charges. Pursley, wisely it seems, opted to have that judge try the case without a jury. He was acquitted on January 16, 2019. Illinois remains the only state to allow post-conviction challenges based upon new IBIS evidence.

• Brad Jennings called police in Texas County, Missouri, to report having found his wife dead of a gunshot wound to the head just after midnight on December 24, 2006. The gun, which was registered to Jennings, was found under the body, and police and the coroner agreed that she committed suicide.

Months later, the wife’s sister called police and told them she believed Jennings murdered her sister. A police sergeant looked at crime scene photos and noticed a drop of blood on the wife’s right hand. Although he had little training in blood spatter analysis, he decided that, if she had shot herself, there should be more blood on her right hand. The case was reopened.

At Jennings’s 2009 murder trial, the sergeant testified that the wife’s blood, which was found on the bathrobe Jennings wore when he discovered her body, was caused by “blowback” from a nearby gunshot, indicating Jennings murdered his wife. He was convicted and sentenced to 25 years’ imprisonment.

Six years later, defense lawyers discovered that police had not shared the fact that no gun powder residue was found on the robe, undermining their theory that Jennings fired the shot while wearing the robe. That, combined with a highly-qualified blood spatter expert who concluded it was “clearly a suicide,” was enough to win Jennings exoneration in February 2018.


The cases summarized above are only a few of the most recent exonerations after faulty forensics helped put an innocent person in prison. Often there were other factors, such as coerced or false testimony, mistaken eyewitnesses, or other factors involved, but the faulty forensics put the nail in the innocent person’s coffin. Juries might believe an eyewitness is mistaken, but they do not believe that “scientific” evidence offered by a forensics expert is in error. Yet innocent people have been and continue to be imprisoned by faulty forensics even in fields believed to be virtually unassailable, such a fingerprint and DNA comparisons. Often, the fault is not in the science, but in the so-called expert.


Some forensic methods are powerful tools in determining truth in criminal cases. However, they are only as good as the forensic scientist or technician deploying them. Further, proper use of forensic methods – especially pattern matching methods – requires validation that, for the most part, is maddeningly lacking in the field.

Until those methods are scientifically validated, it will be necessary to grant defendants increased pretrial discovery, access to and possibly funding for expert witnesses, and a Daubert hearing to act as a check on the prosecution’s use of questionable science to persuade judges and juries to convict even the innocent. Unless there is reform and validation of methodology in the forensic sciences, any one of us is subject to wrongful prosecution, conviction, and imprisonment based upon the seemingly infallible and unassailable evidence of faulty forensics. 


As a digital subscriber to Criminal Legal News, you can access full text and downloads for this and other premium content.

Subscribe today

Already a subscriber? Login



Prisoner Education Guide side
Advertise Here 4th Ad
PLN Subscribe Now Ad