When Science is Not So Scientific: The World of Forensics

From fingerprints to DNA profiles, the fallibility of forensics continues to plague our justice system.

Reading Time: 5 minutes

Cover Image
By Emily Young-Squire

Edgar Marx was a suspect in the murder of his landlady, Lovey Benovsky, in February 1974. A key piece of evidence was a bite mark left on Benovksy’s nose. Despite claims from the California Court of Appeals that there was “no established science of identifying persons from bite marks,” dentists used an impression of Marx’s teeth to prove him guilty. Combined with Marx’s previous inconsistent statements, the evidence persuaded the jury to find Marx guilty of voluntary manslaughter.

People v. Marx became a reference during future convictions utilizing bite-mark analysis, and the practice found its way into the case of Robert Lee Stinson in 1985. The 21-year-old was charged with the murder and rape of 63-year-old Ione Cychosz. The only available evidence was bite marks left on the victim. After analyzing the indentations, forensic odontologist Dr. L. Thomas Johnson concluded that the perpetrator was missing an upper lateral incisor. Stinson had a missing central incisor, but the difference was overlooked by investigators.

To prove Stinson’s guilt, Johnson held a model of Stinson’s teeth against a mold of a bite mark found on Cychosz and disregarded how three different dental models also fit the impression. Judge Genine Jeske even claimed that “there were adequate standards and controls in the area of forensic odontology” despite the lack of scientific research to back the statement. The jury found Stinson guilty of first-degree murder. He was 44 years old when he was exonerated, and the real killer, Moses Price Jr., was caught. Price had committed a series of robberies, raped another woman, and killed a man while Stinson was serving a life sentence.

Stinson’s case is not unique. According to the Innocence Project, an organization that uses DNA testing to exonerate the wrongfully convicted, at least 26 people convicted by bite-mark evidence were later acquitted. Bite-mark evidence remains admissible in most states, but many forensic odontologists have regarded the evidence as unreliable.

A more widely employed technique, microscopic hair analysis, determined the fate of 18-year-old Kirk Odom. The D.C. resident was accused of raping a woman at gunpoint in February 1981. A few strands of his hair were collected to see if they matched those found on the victim’s nightgown, and an FBI microscopist determined them to be a match. Despite the uncertainty, prosecutors claimed that the microscopist was rarely mistaken in the thousands of cases he’d worked on, and convinced the jury to find Odom guilty of rape, sodomy, armed robbery, and burglary. Odom served his 21-year sentence before DNA evidence brought the truth out. The real perpetrator had already committed another crime and was not prosecuted due to the statute of limitations, which places restrictions on how long after an alleged crime legal action can be taken.

With the breakthrough of DNA analysis, the reliability of hair analysis was called into question. A 2002 study found that approximately 11 percent of the time, microscopically indistinguishable hair strands contain different mitochondrial DNA. Such a significant discrepancy has profound implications. According to the National Association of Criminal Defense Lawyers, of over 300 people exonerated by DNA evidence, nearly a quarter were wrongfully convicted when using hair analysis. In 2012, the FBI announced an internal investigation to reexamine 21,000 cases involving hair analysis.

Even fingerprint analysis, largely regarded as an accurate technique, can be flawed. In 2004, Brandon Mayfield was accused of bombing commuter trains in Madrid when fingerprints found on a bag containing detonators were photographed and sent to the FBI. After running them through an Integrated Automated Fingerprint Identification System, the prints matched those of Mayfield, who was arrested for two weeks before Spanish authorities found new information that proved Mayfield innocent.

Additionally, a 2011 study found that the chances of a fingerprint false positive identification was 0.1 percent. While this seems like a small probability, it is still concerning given the large number of cases that use fingerprint analysis. Since 1920, there have been 22 known cases of fingerprint analysis errors. However, future advancements may be able to improve the accuracy of this method. A 2018 Nature report details how biological components found on prints, like proteins and lipids, can provide insight into the age of the print and help investigators better recreate crime scenes, which could potentially eliminate suspects and decrease the opportunity of false convictions.

Though DNA testing is the most accurate forensics method to date, this technique has also encountered problems. Lydia Fairchild found herself the victim of the method’s limitations in 2002. When Fairchild and Jamie Townsend, the father of their two children, applied for welfare, DNA tests taken to confirm their parenthood indicated that Fairchild was not the biological mother of her children. The couple was suspected of welfare fraud and risked having their children taken. With the pair unable to explain the findings and Lydia pregnant with their third child, the court decided that an observer could witness the birth and test Fairchild and her newborn after. The subsequent test showed that Fairchild’s DNA did not match that of her child, suggesting an anomaly in Fairchild’s genetic makeup.

Fairchild’s attorney came across a paper in the New England Journal of Medicine that detailed another case, involving Karen Keegan, a mother whose DNA did not match that of her children. Keegan had chimerism. In this condition, an individual has more than one set of DNA due to a migration of genetic material between expecting mothers and their fetuses, blood transfusions, organ transplants, or in the case of Keegan and Fairchild, the fusion of twins very early in embryonic development.

This medical phenomenon challenges the notion that DNA analysis is infallible. Though there are only around 100 documented cases of chimerism, the number of chimeras is unknown. And while Fairchild did not lose custody of her children, other chimeric parents may have. In contrast, chimeric criminals who leave one set of DNA at a crime scene and another for a test can be proven innocent when guilty. Though DNA evidence has freed many of the wrongfully convicted, there is a chance that a chimeric convict could be exonerated with the aid of their inner twin when they are actually guilty.

Though these are unlikely scenarios given the rarity of the fusion of twins resulting in a single individual, they should nonetheless be addressed when conducting DNA analysis. In paternity cases, for example, analyzing DNA taken from sperm cells rather than swabbed cheek cells minimizes the chances of a father’s profile not matching that of his child.

Despite the rapid progressions in forensic science in the past few decades, forensic techniques remain subject to error. Since the lives of millions are changed by our use of forensic technology, we must be cautious when equating flawed science with truth.