Invented Evidence

Richard Jonathan O. Taduran, Ph.D.  |  6 August 2026

In 1998, Jimmie “Chris” Duncan was sentenced to death for the murder of his girlfriend’s 23-month-old daughter, Haley Oliveaux. The prosecution relied heavily on marks on the child’s body that a forensic dentist identified as human bites matching Duncan’s teeth.

For nearly three decades, those marks kept him on death row.

But there was a video the jury never saw.

During the postmortem examination, forensic dentist Michael West was recorded pressing and scraping a dental mold of Duncan’s teeth directly against the child’s skin. Experts who later examined the footage concluded the procedure may have altered existing injuries or created the very marks presented as evidence. West called it a “direct comparison” technique. Courts and independent experts called it scientifically indefensible, leading to the eventual overturning of Duncan’s conviction.

The disturbing question is not merely whether an expert made a mistake, but whether the evidence used to condemn a man was created during the examination itself.

Developing Evidence vs. Creating It

Forensic experts rarely examine evidence without touching or altering it. Some material must be developed, enhanced, collected, or tested before it can be observed.

Fingerprint powder reveals latent prints. Chemical reagents detect blood. Dental stone creates footwear casts. Biological material is swabbed or consumed during DNA testing. These accepted procedures aim to reveal, preserve, or measure something that already exists.

The critical distinction: an examiner’s intervention must never create the feature being interpreted.

Any procedure must be scientifically justified and thoroughly documented. Original conditions must be recorded before testing begins, materials and techniques identified, and post-examination changes clearly distinguished from pre-existing conditions. This is essential when procedures are destructive or non-repeatable, allowing future experts to verify whether results came from the crime or the examination.

In Duncan’s case, pressing a dental mold into the victim’s skin crossed this line. If that contact altered the tissue or created new impressions, subsequent comparisons became circular. The dental model appeared to match marks that it had helped produce.

An examiner’s job is to uncover what the crime left behind—never to invent it.

The Persistent Failure of Bite-Mark Identification

Even without physical manipulation, bite-mark identification rests on weak scientific foundations.

The method assumes human dentitions are sufficiently unique to identify an individual and that skin accurately preserves dental arrangements. It further assumes examiners can consistently recognize and match bite marks to a suspect.

Skin, however, is a poor recording surface. It stretches, compresses, swells, heals, and decomposes. Injury shapes shift depending on body position, bite force, movement, and time elapsed before examination.

These fundamental flaws were highlighted in landmark reviews by the National Academy of Sciences and the President’s Council of Advisors on Science and Technology, both of which found bite-mark analysis lacking scientific validity. Examiners frequently disagree on whether a mark was even caused by teeth. Yet in court, bite-mark conclusions were long presented with the certainty often associated with fingerprints or DNA.

The consequences are clear in wrongful convictions built on bite-mark testimony.

Keith Allen Harward was convicted of murder and rape based on bite-mark testimony supported by multiple odontologists. DNA testing later excluded Harward and identified the true perpetrator. Harward was exonerated in 2016 after serving 33 years.

Robert Lee Stinson served over 23 years for murder after forensic dentists testified with extraordinary confidence that bite marks matched his teeth. DNA testing later exonerated him and implicated another person.

These were not isolated mistakes. Multiple practitioners applied accepted field methods and reached conclusions that were objectively wrong.

Forensic odontology remains vital for identifying human remains through dental records. But that legitimate application does not validate using distorted skin injuries to identify a suspect. A valid scientific method requires proven reliability and clear limitations. Bite-mark analysis failed both tests.

The Human Cost of Unreliable Evidence

Duncan spent 27 years on death row because of invented evidence, while a flawed methodology was framed as objective science.

An overturned conviction cannot return those decades. It cannot restore lost relationships, health, or an adulthood lived outside prison walls. Release ends confinement, but it does not erase the damage of condemnation.

This is why forensic science demands strict accountability. Experts hold immense influence because judges and juries perceive their findings as impartial truth. When evidence is poorly preserved, physically altered, or interpreted through unvalidated methods, that authority turns error into a verdict.

Forensic practitioners do not merely handle specimens and lab reports. Their conclusions dictate whether another person retains freedom, reputation, or life itself. When forensics is done badly, evidence does not merely fail—it destroys innocent lives.