Fingerprint Experts Shouldn’t Make Claims They Can’t Support

So, last year, I wrote an article entitled “Fingerprint Evidence Not Always Accurate“, and the article discussed a few issues regarding the scientific reliability of fingerprint evidence.

Basically, fingerprint comparisons are a statistical science due to the probabilistic nature of such comparisons (isn’t most science, at its core, probabilistic? The behavior of matter, on the quantum level, is probabilistic). For example, when a fingerprint examiner claims that two prints “match”, the general public just assumes that the two prints are identical. But this is an incorrect statement. Actually, the expert fingerprint examiner has merely “matched” a few unique points on the two prints, and from those matches he/she just presumes that the rest of the fingerprint must match up as well. That’s hardly an exact match, or an exact science.

When a fingerprint expert claims that the defendant’s fingerprint matches a print at the crime scene, the expert has merely identified patterns and placements of features like loops and ridges– leading to the claim that no two prints on Earth would have 4 loops and 5 ridges in said locations. Is that so? Well, the odd thing is, no large scale scientific study has ever been done to create any sort of foundation to support such statistical claims. Indeed, it may be that several hundred, or thousand, or whatever, people may have those loops and ridges in those locations on their finger. We just don’t know, no one has bothered to study it. But I can assure you, every prosecutor in Florida is ready to tell a jury the “fact” that the fingerprint at a crime scene IS from the defendant’s finger. That the fingerprint IS a MATCH. Wow! No two prints are the same, right?

Alright, so the reason we’re discussing fingerprints for the tenth time is because I found an excellent court order addressing the so-called “science” of fingerprint evidence. Miami-Dade Judge Milton Hirsch wrote an order a few months ago restricting what fingerprint experts are allowed to tell a jury in his courtroom. (State v. Borrego, Miami-Dade Case No. F12-101, October 25, 2012) It is an excellent, well written opinion, so read it if you get a chance. Here’s what prompted the order: Borrego’s criminal defense attorney filed a motion in limine requesting that the prosecutor’s fingerprint expert be banned from claiming that the defendant is “the one and only source of the print”, or from using misleading words like “match”. Judge Hirsch agreed, pointing out a few problems with fingerprint analysis along the way.

The order states, correctly, that “the proposition that no two persons can have the same fingerprint is one for which there exists much empirical, but no theoretical, support.” Remember that fingerprint comparisons are just statistical claims, and Judge Hirsch reasons that if the “average human fingerprint is said to have between 75 and 175 ridge characteristics”, then, “[c]an different people have as many as 74, or even 174, such characteristics in common? Can they have 73? Or 72? Or 71?” p. 12. Again, with no research or scientific studies cataloging fingerprint characteristics–how can anyone give an opinion on such evidence?

Judge Hirsch’s order digs deep into the history of fingerprints, noting that “[a]s early as 1920, Scotland Yard imposed the requirement that there must be a minimum of 16 points of identity between the latent fingerprint and the same fingerprint for a fingerprint examiner to determine that they came from the same source. U.S. fingerprint examiners of the same period commonly cited eight or 12 points of identity as being the requisite minimum. No scientific or other basis was offered for these numbers. If a given sample fingerprint had 175 ridge characteristics–the upper limit of the average–and 16 point of comparison are found with a partial latent impression, then even an examiner following Scotland Yard’s conservative protocol would be entitled to declare a match; this, despite the remaining 159 ridge characteristics as to which no comparison could be made.” p. 13. [internal citations omitted] Gee, almost a hundred years, and we’ve gone from requiring 16 points of comparison down to 8 – 12. Is that progress?

The crux of the problem here is the effect so-called ‘experts’ have on a jury’s opinion. The criminal defense attorney defending Borrego simply requested that the state’s expert not claim the two fingerprints “match”. After all, the ‘expert’ only matched 8 to 12 ridges between the two prints. Its’ the ‘conclusion’ that is of concern here. More accurately, a naked conclusion that is not supported by scientific evidence. Judge Hirsch eloquently puts the issue as such:

“If the jury were merely offered a blow-up of the latent impression and a blow-up of the defendant’s exemplar and invited to make its own comparison, no problem would arise. Even if a witness went so far as to direct the jury’s attention to certain loops, or whorls, or arches in the latent impression and the corresponding features in the defendant’s exemplar, no problem would arise. No Frye-testing is needed for this. No Frye issue is raised by this. The jury is simply being shown two evidentiary artifacts, each presumably properly authenticated, and asked to reach its own opinions and conclusions as to their similarity. The problem arises when a witness is vested in the raiment of an expert, and permitted to express his opinions and conclusions.” p. 13.

Finally, Judge Hirsch recognizes two important, but negative, aspects of the fingerprinting community. First, “fingerprint identification is advocated and practiced almost exclusively by employees of the law-enforcement and prosecutorial authorities”. It’s a scary thought. As such, a community of so-called experts so tightly controlled by law enforcement should be subject to some level of scrutiny before their conclusions are placed before a jury. And, second, “fingerprint identification is based principally upon the examiner’s experience, not upon an independent, pre-existing body of scientific theory.” p. 15.

So, what’s admissible according to Judge Hirsch? Testimony like “I direct the jury’s attention to the arch appearing here, and the loop appearing here”. p. 16.

What’s not admissible? “I have concluded that this fingerprint matches that of the defendant to the exclusion of all other fingerprints in the history of the world.” p. 16.