CLPEX Chat Board

2,600 cases possibly affected in latest police lab scandal

Highlights from the article

The letter to the defense lawyers didn’t provide much detail about the problem with Palacio, saying only that “the performance issues of Marco Palacio consisted of clerical errors, failure to identify prints of value and the mislabeling of print cards.” Washington declined to elaborate.

and

Palacio was apparently well known in the public defender’s office for his certainty in his work, and Wesley produced the audio of a 2013 pretrial deposition of Palacio by an assistant public defender. The audio shows that Palacio said in a burglary case, “Almost 15 years of working with fingerprints, I would say I have not gotten anything wrong yet. Everything I do, I compare many times before I submit it. My findings will check out. I have not gotten any mistakes … ”

The defender asked: “So you are saying you have never made a mistake in 15 years?”

Palacio: “Doing fingerprints, never.”

Palacio was also asked what oversight or verification was used on analyzing fingerprints. “When we have a beautiful print,” he said, “with 25 points or more, we don’t bother with the verification process.”

I realize that news tends to be heavy on the sensationalism and light on the facts, but at face value I have to say that clerical errors and failure to label prints of value seem pretty small time on the list of things from which one would be removed from case work. It's a point I brought up on a different thread regarding likelihood ratios, but bears repeating. What happens when the algorithm that determine level of 'associations' gets 'better' (whatever that means) and things that used to have a low level of association now have a higher level of association and prove useful to a prosecution or investigation in some regard where they previously did not? Is this not the same thing as the examiner was charged with, namely not calling something of value that was? Obviously there's a fair amount of equivocation on my part going on, but I think it's an interesting ethical question.

Also noteworthy is the focus on claims of certainty in accuracy. 'Mistakes' in fingerprint examinations are not limited to misidentifications, and I'm intrigued how their SOP defines the standard of a 'beautiful' print... Missed identifications are equally 'mistakes'. Black box data shows 7.5% missed identification rate, and evencase specific AFIS demonstrated that mistakes are more likely to occur on an exclusion. I've heard people who don't verify exclusions say they have not missed identifications. But logically what they are actually saying is 'No one found any evidence of the thing they didn't look for', which call me Sherlock, is not the same thing as not making a mistake.

Lets not forget that the response "I have never made a mistake in fingerprint examination" was a standard answer for testimony and training classes drilled that into practitioners from day one.

Bill, if you have some of that old training that you might be able to scan in with such statements, I'd love to see it!

Here's something from 1987 that seems to indicate there were also other trains of thought out there. Obviously, before my time, but interesting to see the attitudes of juries.

I have alot of old training material going back to the 1970's but I'm not sure I can produce anything that addresses the specific question "have you ever made a mistake?"

But I stand by my statement that admitting to the possibility of a mistake (erroneous identification)in court during fingerprint testimony was the kiss of death to your career and was not condoned by the discipline "back in the day"

I'm not saying I doubt that happened, just that the historian in me loves old documents. Between the Internet Archive and retired examiners that leave their old material for the benefit of the office, I've found a treasure trove of things. I think it gives context to periods of time and helps explain some of the shifts in attitudes.

Bill Schade wrote: ↑Mon Mar 06, 2017 9:38 am. . .admitting to the possibility of a mistake (erroneous identification)in court during fingerprint testimony was the kiss of death to your career and was not condoned by the discipline "back in the day"

Not entirely true for all latent print examiners who worked before Mayfield or the NAS report. I was taught, and always testified, that in any field of human endeavor there exists the possibility for human error. But in my career I have never been found to have made an erroneous identification and in the case at trial, I am totally confident in my conclusion.

Point taken

I guess I'm still clinging to the notion that the only "error" in fingerprints was an erroneous Identification.

I know better than that now, but old values are hard to shake

Boyd Baumgartner wrote: ↑Mon Mar 06, 2017 7:47 am Bill, if you have some of that old training that you might be able to scan in with such statements, I'd love to see it!

Here's something from 1987 that seems to indicate there were also other trains of thought out there. Obviously, before my time, but interesting to see the attitudes of juries.

Expert-Witness-1987.jpg

"The fingerprint expert is unique among forensic specialists. Because fingerprint science is objective and exact, conclusions reached by fingerprint experts are absolute and final." Advances in Fingerprint Technology, 2nd Ed., page 390!