Forensic Science Failures
Corrupt forensic scientists and imperfect forensic methods continue to put many innocent people behind bars

This piece is the seventh in a series on defects in our criminal justice system that lead to wrongful convictions. The first piece introduces these seven critical breakdowns in the rules of constitutional criminal procedure, the second examines the problem of eyewitness misidentification, the third shows how the Supreme Court is perpetuating the problem of false confessions, the fourth examines prosecutorial misconduct, the fifth looks at overly favorable treatment for prosecutors, and the sixth discusses the importance of good counsel.
Since the advent of the DNA revolution more than three decades ago, it’s been easy for us to think of science as an unequivocal boon for our legal system, ensuring that both victims and the accused get the justice they deserve. But is science—and the way it’s used in trials—really irrefutable?
DNA testing has proved to be the gold standard of criminal forensic truth-telling. Yet we now know that junk science—the misapplication of science caused by either dishonest forensic scientists or flawed forensic methods—has infected courtrooms around the country, resulting in countless wrongful convictions.
The U.S. Constitution guarantees every criminal defendant a fundamentally fair trial. But the Supreme Court has refused to interpret these clauses in ways that would aid those who might be wrongfully incarcerated due to unreliable scientific testimony and to prevent future outrages by keeping bad science out of criminal trials and good science in. While many state courts and legislatures around the country have been doing their part to fix our broken system, our highest court has, for the most part, chosen to stay idle or even make things worse. The Supreme Court has no more solemn duty than protecting the integrity of criminal trials. It needs to change course.
The Tragic Case of Janice May … and Lloyd Miller
On a late November afternoon in 1955, third-grader Janice Elizabeth May was doing the dishes in her family’s Canton, Illinois, home. Her parents had left her in the care of her two teenage brothers for the day. The boys went to the local playground for a few minutes, figuring Janice would be fine. But when they came back to check on her, she’d disappeared.
The boys didn’t think much of her absence at first, but after about an hour, they began to wonder. While searching the neighborhood near the family home, the younger of the two heard noises beneath an abandoned car. He fetched his older brother, who looked under the car and discovered the broken, unconscious body of their little sister.
Janice never woke up. She had been raped, beaten and dragged under the car where, likely believing her dead, her killer hoped she’d remain undiscovered long enough for him to escape undetected. The torture Janice endured was horrendous almost beyond description. Whoever did this was precisely the type of criminal for whom states pass death penalty laws.
Lloyd Eldon Miller Jr. was an unhappy man. His marriage had just fallen apart: The 29-year-old Miller was living on his own, working as a cabdriver in Canton, and his wife wanted financial support. A few hours after Janice’s murder, he gave a woman named Betty Baldwin a ride and, after dropping her off, fled town rather than answer his wife's summons to court. Baldwin told the police that, sitting alone with her in the cab, Miller had confessed to the murder. Lloyd Miller was now a wanted man.
Newspapers and radio stations quickly told the public to be on the lookout. Miller panicked and was now even more determined to flee. A bus station manager in Danville, Illinois, recognized him and contacted police. Miller was arrested and brought back to Canton.
These are the facts of Miller’s case according to the Supreme Court of Illinois, which upheld Miller’s death penalty conviction for the murder of Janice May. However, the court’s story masks a wilderness of mendacity, corruption and cunning psychological manipulation. The state’s star witness, Betty Baldwin, was a mentally erratic serial liar who later admitted to perjuring herself in this case—and that’s only the beginning. Lloyd Miller was framed.
There were no eyewitnesses to Janice’s murder. The state’s strongest piece of evidence was Miller’s confession to police, which was almost certainly coerced. A mentally broken Miller was subjected to a 52-hour interrogation, during which he admitted that he’d taken off and discarded a blood-soaked pair of men’s underwear that was found a few blocks from the crime scene.
At the beginning of Miller’s trial, his attorney requested the opportunity to examine all the forensic evidence the prosecution planned to present, including the underwear. The judge denied the request. The prosecution put its forensic expert, Forrest Litterly, a chemist with the state Bureau of Crime Identification, on the stand. He confirmed that Miller’s underwear was soaked with blood that was not his own and consistent with the victim’s, thereby strengthening the state’s case. Miller was sentenced to die in the electric chair.
Just six hours before his scheduled electrocution, Miller was granted a stay of execution by a federal judge. The judge allowed Miller to have his own forensic expert, James Martin, test the bloody underwear. But Martin did not find the victim’s blood on the underwear. In fact, he found no blood at all. Only one staining material was present, and nearly the entire pair of underwear was covered with it: paint.
So how do we account for Litterly’s testimony? Even if we give him the extreme benefit of the doubt and assert, as the state did in its appellate arguments, that blood co-existed with paint on the underwear at the time of the trial but had evaporated, that wouldn’t explain why Litterly failed to report the massive splatters of paint to the jury. Miller’s case raised an alarming question: How many defendants were being imprisoned, or even executed, due to forensic scientists like Forrest Litterly?
Fortunately, the U.S. Supreme Court ultimately overturned Miller’s conviction in 1967 by finding that the prosecutor knowingly deceived the jury when presenting the paint on the underwear as blood. But while the court was focused on the misconduct of the prosecutor, Litterly's testimony suggested a potentially widespread problem in criminal forensics that also needed to be addressed. The scope of the problem wouldn’t be known until the DNA revolution in criminal forensics about 25 years later.
Still a Huge Problem
DNA testing has proved so reliable that it has become a kind of answer key to check the work of government forensic witnesses at trial. It is the gold standard by which the reliability of all other scientific techniques and experts can be measured. However, junk science remains a problem. There are more Forrest Litterlys lying, faking evidence and bungling testing in state crime labs than we’d expect.
Junk science continues to put innocent defendants in prison around the country. Since the start of the DNA revolution, at least one man, Cameron Todd Willingham, has been wrongly executed because of it. Using since-discredited fire science, Willingham was convicted of setting the fire that killed his three children. Since 1989, the National Registry of Exonerations has recorded more than 3,300 exonerations—and nearly 800 of these wrongful convictions were the result of the admission of junk science at trial. Of the first 241 inmates the Innocence Project exonerated using DNA testing, more than half (53%) had been convicted on the strength of “misapplied forensic science.” And this data only accounts for those fortunate few able to find competent—often extraordinary and self-sacrificing—post-conviction attorneys willing to take these cases. The actual number of convictions based on junk science is almost certainly much higher.
These cases can be divided into three groups: 1) those involving incompetent or deceitful scientists; 2) those using forensic techniques that are unreliable, even when used in good faith; and 3) cases in which more reliable, and potentially exonerating, forensic testing becomes available post-conviction but the state will not allow testing. In all three types of cases, the Supreme Court has been holding back progress by showing an unwillingness to interpret the constitution’s due process clauses, which guarantee fundamental fairness in criminal trials, in ways that allow defendants to use the best available scientific evidence to show their innocence.
In the history of the modern innocence movement, there is perhaps no greater villain than Fred Zain. Throughout the 1980s, Zain was director of the Serology Department for the West Virginia Department of Public Safety—West Virginia’s go-to chemist for blood-type testing. He always seemed to deliver for prosecutors, and his testimony led to hundreds of convictions. However, while slamming the prison door on many defendants in the 1980s, Zain never could have foreseen that eventually his work would one day be held up to the kind of scrutiny DNA testing can provide.
One of Zain’s victims, Glen Woodall, would become the Innocence Project’s first exoneree in 1992. Zain had falsified blood test results in Woodall’s case and, to ensure a conviction, secretly revised the findings of his hair-matching tests so that, instead of a pubic hair from an unknown person, they read that the hair found in the rape victim’s car came from Woodall’s beard. Woodall’s DNA exoneration sparked an investigation into Zain’s other cases, and the findings shook West Virginia’s criminal justice system to its core: While an exact count is impossible, Zain’s misconduct might have caused as many as 180 wrongful convictions. Zain’s case doesn’t just teach us how much damage a corrupt police scientist can cause; it also teaches us how easy it is for such forensic scientists to go undetected, and how unprotected criminal defendants are from them.
As Dean Jane Moriarty of the Thomas R. Kline School of Law at Duquesne University has written, “Zain, unfortunately, is not unique.” It’s truly startling to see how many examples of unethical forensic scientists there have been. Oklahoma City Police Department chemist Joyce Gilchrist engaged in decades of serial fraud and abuse, regularly framing defendants such as Curtis McCarty, who spent two decades on death row for a crime of which he was ultimately exonerated.
In 2012, Massachusetts forensic chemist Annie Dookhan was caught forging records and routinely falsifying lab reports. A Massachusetts Supreme Judicial Court Justice dismissed more than 20,000 criminal convictions tainted by her wrongdoing. (Netflix recently released a documentary about the havoc she wreaked across her state’s criminal justice system.) Houston Crime Lab chemist Jonathan Salvador tainted many of the nearly 5,000 drug cases he worked on between 2006 and 2012. In El Paso, Texas Department of Public Safety forensic analyst Ana Romero serially reported fraudulent blood tests in 2013 and 2014, resulting in 22 overturned convictions. Sadly, the list goes on.
Bad Forensic Scientists, Bad Forensic Methods
Recognizing this widespread problem in forensic science, the National Academy of Sciences commissioned what would become the most important report ever written on the use of forensic science in our criminal justice system. But the 2009 report made clear that the ethics and qualifications of the state scientists weren’t the only problem. Many of the forensic techniques themselves, hitherto thought reliable, were more error-prone than we’d realized. Ten years after the report’s publication, Judge Harry Edwards, who led the team that wrote the report, explained: “With the exception of nuclear DNA analysis . . . no forensic method has been rigorously shown to have the capacity to consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual or source.”
Most forensic methods rely on connecting evidence from a crime scene (such as blood, semen or hair) to something known to belong to the suspect—for example, comparing the handwriting on a ransom note against that in a suspect’s diary. This type of matching is known as “forensic individualization.” The problem with forensic individualization techniques is that, as opposed to DNA testing and fingerprinting, they can never truly identify just one person to the exclusion of everyone else. Even though all of us have unique fingerprints, it still takes a set of fallible human eyes to compare two sets of prints.
The result of forensic individualization has been the rise of now-dreaded phrases like “similar to,” “consistent with” and “do not exclude”—phrases that can mean very little, yet are still seized on by prosecutors as proof and steer juries toward wrongful convictions. Worse, even with advanced computer technology reducing human error, these techniques remain highly subjective and only as reliable as the scientist doing the matching. Many have error rates so high that they’ve been challenged as a pseudoscience. For instance, forensic odontology (or bite mark evidence) burst on the scene in heroic fashion when prosecutors used it to help convict world famous serial killer Ted Bundy during a televised trial in 1979. However, Bundy actually happened to be guilty. Several subsequent DNA exonerations have put bite mark matching under a dark cloud of suspicion.
Courting Disaster
And yet, the Supreme Court has done little to ameliorate the problem of forensic science failures. The Bill of Rights and the due process clauses of the Fifth and 14th Amendments guarantee fair criminal trials. Therefore, the court ought to interpret these clauses such that when state forensic experts testify, criminal defendants are permitted have an expert of their own independently test the evidence and, when appropriate, rebut the state expert’s testimony. This is precisely how James Martin, who discovered that the “blood” used to convict Lloyd Miller was actually paint, saved Miller’s life. But that’s not how the court has chosen to interpret these clauses to date—and countless innocents are currently languishing in prison because such life-saving scientific tests aren’t constitutionally required.
The Supreme Court seemed to be starting down the right path on this issue with its 1985 decision in Ake v. Oklahoma. In 1980, Glen Burton Ake was found guilty of a crime eerily reminiscent of that described in Truman Capote’s “In Cold Blood.” He and an accomplice invaded the home of an Oklahoma pastor, murdered the pastor and his wife, and brutally attacked their children. During Ake’s arraignment, the judge removed Ake from the courtroom and ordered a psychiatric examination due to his odd behavior. Doctors diagnosed Ake with severe paranoid schizophrenia. Since he unquestionably killed the victims, his legal team planned an insanity defense.
However, because his court-ordered psychiatrists only examined him to determine his current mental state rather than his mental state at the time of the killings, and he couldn’t afford to hire an expert of his own, Ake was effectively denied the opportunity to put on what might have been a winning defense. He was convicted and sentenced to death.
In its opinion overturning his conviction, the Supreme Court explained that, in cases like this one, a fair trial requires that the defendant get the assistance of a court-appointed, state-funded independent forensic expert because, without one, the jury might wrongly convict. But while Ake was a landmark victory for mentally ill defendants in capital cases, the court has since refused to extend the rule in this case to those who need non-psychiatric forensic experts to refute the state’s evidence. In cases in which only one expert testifies, and everyone else in the courtroom is a layman incapable of detecting errors, the expert often gets undue deference. It’s only when the other side gets the assistance of at least a neutral expert that the fight becomes fair.
In Caldwell v. Mississippi, another 1985 decision, the court, citing Ake in a footnote, refused to extend this reasoning to non-psychiatric experts and hasn't extended it since. In this case, the trial court judge denied Bobby Caldwell’s funding request for forensic experts in fingerprint analysis and ballistics, even though the prosecution would eventually convict and put him on death row, in part on the strength of the testimony of its ballistics expert. The Supreme Court was unimpressed with Caldwell’s argument that he was entitled to these funds and ignored it, instead deciding his case on other grounds.
While many lower courts around the country have extended the Ake ruling to provide defendants with scientific experts for issues other than mental illness, the Supreme Court, whose constitutional decisions are binding on every court in the land, hasn’t budged. The Texas Court of Criminal Appeals in Rey v. State, for instance, took a more enlightened approach when it ruled that the defendant was entitled to funds to hire a pathologist who might testify that the state’s expert had misidentified the victim’s cause of death in the case—that the death may have been by natural causes rather than strikes to the head.
The Texas court couldn’t understand why the Constitution would require defense access to psychiatric but not non-psychiatric experts when they can be equally important to a fair trial: “The question in each case must be not what field of expert knowledge is involved, but rather how important the scientific issue is in the case, and how much help a defense expert could have given.” The Supreme Court’s decision not to extend the rule in Ake is arbitrary and irrational, and it continues to infect our criminal justice system with wrongful convictions.
The Supreme Court has also repeatedly chosen to deny inmates the benefits of scientific evidence that may exonerate them. William Osborne’s case may be the most notorious and disheartening example. Osborne was convicted of kidnapping, raping and beating a woman to near-death in Anchorage, Alaska, in March 1993. Lee Ann Strickland of the Alaska State Crime Lab performed a “relatively inexact” form of DNA testing known as a DQ Alpha test on the semen found in the condom used during the rape. She explained that the DQ Alpha test neither confirmed nor exonerated Osborne as the perpetrator but was consistent with the prosecutor’s contention that Osborne was guilty. The defense could have sought a more precise DNA test that might have exonerated Osborne, but his attorney, believing him guilty, chose not to request it. Osborne was sentenced to 26 years in prison.
Sixteen years later, Osborne, proclaiming his innocence, asked the Supreme Court to rule that the due process clause of the 14th Amendment, which ensures fundamental fairness in criminal cases, required the state of Alaska to turn over the evidence in his case so he could test it. Alaska had nothing to lose. If the DNA matched, Osborne would remain behind bars. If not, justice would be served.
However, Alaska remained stubborn and refused to allow Osborne access to the evidence, even when Osborne offered to pay for the more-advanced DNA test himself. In a major blow to the innocence movement, the Supreme Court ruled 5-4 that defendants who have had a fair trial, like Osborne, are not entitled to access to the state’s evidence during post-conviction proceedings, even when testing that evidence may actually prove them innocent. Chief Justice Roberts’ opinion in District Attorney’s Office v. Osborne is a frustrating failure to appreciate the importance of accurate, reliable science as a tool for correcting less reliable evidence presented at trial and identifying wrongful convictions.
Osborne had been convicted, in part, as a result of a DNA test that not only didn’t positively identify him but may have misled the jury toward the wrong verdict. He merely asked the court to recognize that inmates in his situation, whose factual guilt or innocence can be determined by more-advanced scientific testing, are entitled to use evidence in the state’s possession for that purpose. By denying him and every other similarly situated inmate that right, the court interpreted the Constitution in a way that allows less reliable forensic evidence to control the final resolution of criminal cases, even when a superior method is readily available and might remedy a grave injustice.
Forensic science continues to evolve at a bracing pace and can now identify wrongdoers and, when allowed, rescue those who’ve been unjustly convicted in ways that were unimaginable only a few years ago. These developments have brought hope to innocent prison inmates around the nation. For many of them, the science is already on their side. Unfortunately, so far, the Supreme Court hasn’t been. It’s long past time for that to change.