Another Exoneration Casts Doubt on Bite Mark Forensics

The Innocence Project has secured yet another exoneration, this time of a man who served 19 years in prison for a murder he didn’t commit. The primary basis for that conviction was testimony from an “expert” on bite mark analysis, something my friend Radley Balko has shown to be empty pseudoscience.

Somerset County Prosecutor Geoffrey Soriano filed a motion today dismissing the indictment against Gerard Richardson, fully exonerating him of murder for which he served 19 years. Richardson’s exoneration comes nearly two months after Somerset County Superior Court Judge Julie M. Marino overturned the conviction and ordered his release from prison.

Richardson was convicted of the February 1994 murder of Monica Reyes based largely on the testimony of a forensic dentist who claimed that a bite mark on the victim’s body matched to Richardson. He always maintained his innocence and was eventually granted the right to submit a swab recovered from the bite mark to DNA testing, but testing was inconclusive. His attorneys eventually sought help from the Innocence Project, which took over his representation. The remaining evidence was submitted once again for testing, and the lab was able to detect a complete male DNA profile from the evidence that excluded Richardson.

Because of FBI regulations, officials in New Jersey have been barred from entering the DNA profile from the bite mark into the CODIS DNA database, which contains over ten million profiles of convicted offenders and could identify the person responsible for the 1994 murder. The true perpetrators have been identified in nearly half of the 311 DNA exonerations.

Those regulations need to be changed. Exonerating those wrongfully convicted is only half the battle for justice; the other half is finding the real perpetrator if that is possible.

"Also, a pardon essentially removes the ability to take the Fifth (it's impossible to self-incriminate ..."

Looks Like Flynn Has Flipped on ..."
"In the other side of the screen, it looked so easy."

Looks Like Flynn Has Flipped on ..."
"That... doesn't make me feel any better. Of course the state of the nation is ..."

Looks Like Flynn Has Flipped on ..."
"Also, you’re replying on the wrong thread on the wrong blog, Boris. Our conversation is ..."

Looks Like Flynn Has Flipped on ..."

Browse Our Archives

Follow Us!

What Are Your Thoughts?leave a comment
  • sqlrob

    I can understand not putting Richardson’s DNA in, but the perps? Huh?

  • rq

    Wait, he discredited bite mark analysis, or a particular specialist who was indulging in a very bad practice of bite marks? (I found this and this, which seem to point to the latter. A similar issue rose up with Charles Smith and more-or-less manufactured (at the very kindest, wrongly interpreted) pathologic evidence, but that didn’t discredit forensic pathology as a whole…)

    Also, I was under the impression that any unidentified profiles went into the database – that’s sort of the point of it, isn’t it?

  • Nathair

    Repeatedly demonstrating that this type of “evidence” has and does completely fail leading directly to wrongful convictions is “casting doubt”? How uncharacteristically gentle of you.

  • dysomniak, darwinian socialist

    Last year NOVA did an episode called Forensics on Trial, which can be streamed from PBS and various pay services. It’s not just bite marks, the entire field of forensic “science” is a mess – even down to fingerprints.

  • cthulhusminion

    One problem with large DNA databases is the chance for false matches. 5 million to one odds are great for excluding someone but with a database of 10 million…./

  • Erk12

    @4 DNA evidence can only ever truly exclude someone, not absolutely include someone. E.g.) if there are 10 DNA makers per profile (I’m pulling a number out of my ass), and a subject shares 9/10 with the perpetrator, that one other marker is the conclusive evidence that the subject is not the perp. It’s impossible that they are the same person if that one marker is different. However, if a subject and a perp match 10/10, it’s always possible, however unlikely, that if you had an 11th marker, that one would be different.

  • eoraptor

    “…half the battle…”

    I dunno. Seems like about 1/3 the battle at best. First there’s the exoneration, then nabbing the right person, then somehow compensating the falsely convicted for all the damage done to their lives. In some cases, there’s even a fourth prong of doing something for the family for the victim (of the original crime).

    As for compensating the falsely accused, in all too many cases that compensation is nothing but a pat on the back and well-wishes.

  • sc_770d159609e0f8deaa72849e3731a29d

    Exonerating those wrongfully convicted is only half the battle for justice; the other half is finding the real perpetrator if that is possible.

    Not convicting those who are not guilty to begin with is also pretty important. How far do aspects of the US court system- the election of judges and prosecutors in some cases, for example- make getting convictions more important than establishing the truth? how often are defendants who are not wealthy unable to properly establish their innocence because they can’t afford to hire lawyers who will look at the prosecution’s case properly?

  • colnago80

    Re ERK12 @ #5

    According to the statistician who couldn’t add, Prof. Bruce Weir of NSU, the odds of a 4 loci RFLP id of someone’s DNA other then O. J. Simpson on the sock was 6 billion to 1. That’s pretty definitive.

  • rq


    I’m not sure where he pulled those numbers, but that’s a bit too definitive.