Opinions Needed

[Forensic Scientist]

I have been struggling with this comparison and I would like some feedback from the community. Please take a look at it and let me know what you think. Sorry that it has to be in two posts. Thanks.

latent.jpg

[Forensic Scientist]

img138c.jpg

[Pat A. Wertheim]

Cute. Real cute.

For those like me who aren't too computer savvy, you can right click on the picture and save it to a file on your computer. Then open the images in photoshop and do your charting.

I would be interested in the whole story. Email me at foridents@aol.com if you'd like.

Cheers,
Pat

[Heather Baxter]

I'm confused. Is there a hidden joke here?

[WRoughead]

There may well be Heather.

But, I'm not qualified to form an opinion, though I've studied the above carefully.

[Pat A. Wertheim]

No joke here. This is a latent & inked combination that presents a conundrum.

My serious suggestion for latent print experts is this: Download the two prints, transform them to approximate same scale, tile them, and plot your points to see what you come up with. Try and discern if there are any discrepancies and why. This is a great exercise. I wonder if Alice Maceo, who has done more work on distortion than anyone else I know, could offer a thought or two.

There are some questions that come to mind:
1) What was the surface?
2) Was there anything unusual about the matrix?
3) Is the inked print from a named suspect or victim? Or was this an AFIS candidate?
4) Were there any adjacent prints (simultaneous)?
5) Were there any other identifications to the same person?
6) Is this a real case work dilemma or (please excuse my caution, Forensic Scientist) is this a trick question?

In my humble opinion, Forensic Scientist has posed one of the best discussion topics for true experts I've seen here in a long, long time. I'm hoping Forensic Scientist will provide us with some details of the case.

[L.J.Steele]

5) Were there any other identifications to the same person?

Wouldn't knowing the answer to this question create a risk of a confirmation bias?

[Pat A. Wertheim]

Wouldn't knowing the answer to this question create a risk of a confirmation bias?

True, that's possible. I'm just trying to get a little background. And consider this: One of the criteria under which the FBI now does blind verification is in single conclusion cases. I'm just wondering about the circumstances surrounding this comparison. Was this a single conclusion, like Mayfield? Was it an AFIS hit, like Mayfield? I've got my thoughts, but I wonder if there is more information available from "Forensic Scientist."

[Forensic Scientist]

In fact this is not a joke. I am struggling with how to explain the seemingly extra ridge that appears in this print. This print was a supreglued on a the curved surface of a glass vase. It was photographed under white light. I did not post any prior information because I wanted honest opinion. I believe this to be an identification but am struggling to put into words how this extra ridge is there. Sorry about the sizing issues, I wasn't sure how to post them for viewing.

[Gerald Clough]

It is indeed a good exercise. Really a good vehicle. It is an example that is different from a situation with features that lead immediately to the understanding that a particular distortion occurred, one in which there was never any question of a distortion effect and in which recognition of the distortion revealed the real features of the source or at least explained the observable anomaly. My question here has to do with intently seeking some distortion explanation that would justify the tentative conclusion of Identification. In this sort of issue, we can consider two sorts of distortion effects. One is a class of effects that present features that, if undistorted impressions, would represent ridge behavior that no one has ever observed, absent trauma, in natural skin. The other is a class of effects that present features that represent ridge behavior known to occur more or less commonly. It could be said that ALL latent impressions contain the second class of distortions, since, as impressions, they are arguably always distortions of reality. You can think of this as you could photographs, some of which, by accident or by Photoshop, exhibit something that has never been seen in nature, and some of which exhibit misrepresentations of reality, but are within what does occur. And photographs are all distortions, being two-dimensional, etc.

We have all seen photographs in which the distortion presented a perfectly natural-appearing feature that was only discovered to be distorted because the actual entity was available and could be observed to determine reality. Knowing that the photo does not depict reality, we propose explanations. Now, confronted with a photograph that, for one reason or another, depicts something in a way that would be significant but for a feature that might or might not be the result of distortion, we can (1) explain some way in which the "spoiling" feature could be merely a distortion effect and argue that it is or (2) admit to both possibilities, that it depicts undistorted reality and is therefore significant -or- it is undistorted and therefore not significant. A whole array of features might be said to be possible distortions. But the fact that one presumption leads to significance and the other presumption does not does not lend weight to the likelihood of either presumption being correct.

It seems to me that the hazard in trying to discover a potential distortion that would validate a conclusion based upon presumably undistorted features is that, to do so, essentially declares that if some quantity of consistent features is present, any other inconsistencies must be distortions, because they could be. And that is an untenable position. It may well be that we may strongly suspect a source but cannot declare a valid conclusion, on account of an inconsistency that cannot be strongly argued to be a distortion without bolstering the argument by taking suspicion as fact. Any subsequent distortion explanation in the impressions that are the subject of this thread must consider if they are revelations made from experience that unquestionable reveals the nature of the distortion and rules out accurate impression or if they are "could-be" explanations that cannot be certain and therefore no resolution.

[Michele]

I’ll commit.

It’s nice to be able to explain why a dissimilarity occurred but I’m a firm believer that things should only be explained if there are visual clues to support your explanation. If there aren’t visual clues then I think it’s better to say, “I don’t know what’s going on in this area but it doesn’t matter because I have enough information to make the identification regardless” (if in fact this is the case). I know that some people think that the inability to explain something is a sign of incompetence, obviously I don’t feel this way.

You have to decide if the consistency is overwhelming enough to for you to feel comfortable discounting the weight of the dissimilarity. (Obviously, I’m not a follower of the one discrepancy rule)

In this case, I don’t think the reason behind the dissimilarity matters because there’s plenty of other information to support an ID regardless of this dissimilarity. I’d make the ID even if I didn’t think I could explain what happened, but I think I can. What we are seeing isn’t ridges, we are viewing residue that was left by ridges. When there is extreme pressure, multiple impressions, or some sort of twisting then the residue looks like a single ridge when in fact it’s multiple ridges (that’s why they looks so thick). There’s subtle signs that the ridges aren’t continuous, like little jogs (that’s a technical term) where the ridges shift slightly. There’s also signs that the ridges are more inverted in some areas than others (which is a clear sign of distortion).

[Terry A. Smith]

Greetings everyone.

First, I'd love to see this at highest possible resolution but, as that's unlikely, I'll just give my thoughts.

While I agree, or my gut agrees, that like Michele I'm seeing distortion effects -probably associated with pressure and the inherent little movements which occur when a finger contacts a surface (particularly a convex surface like a vase) under any pressure great enough to flatten out ridges- I respectfully can't agree with Michele's assertion that it "doesn't really matter". At a time in our history as examiners when we're under fire, for what is perceived as the maverick ability to make conclusions of absolute certainty despite the lack of documented error rates or a statistical probability model, I think it might matter a lot. Yes there is a nice volume of corresponding friction ridge detail between the questioned and the suspect exemplar. Yes, clarity in the impression is far from ideal. Yes, I want to agree that Forensic Scientist has found the donor source of this latent. But does "I think I can explain it" make the grade these days?

The impression presented poses a problem and the problem requires some sort of resolution. The examiner should consider conducting a series of test impressions on a similar item, in an attempt to replicate the distortion effect observed. If the tenprint of the suspect in the case reveals that he/she had particularly robust ridges, or many incipient ridges, I'd find a suitable test subject with similar physical features, even similar patterns if possible. If the deposit appeared to be composed of something other than normal sweat, I'd try to replicate the matrix as best I could. In the end, if the distortion in question can be justified through testing, then the appearance of the distortion in the latent print is resolved with a reasonable explanation, and the full value of those corresponding ridge features can shine through. If the distortion cannot be duplicated in a similar fashion, then the possibility exists (however slight) that the distortion is indeed a discrepancy and an (absolute) conclusion of identification would be inappropriate.

I know as well as the rest of you how many cases back up when this type of extra effort is required. But in latents, we get what we get and we can't abuse our use of "tolerance for apparent dissimilarity" between prints. Look at it this way, you'll appreciate the next "drive-by ident" even more if you put in the extra time on the tough ones.

[Gerald Clough]

It occurs to me that something like this could easily become a demonstration that there's really no bright line between absolute conclusions and the identifications that become issues that will be decided by the fact-finders at trial. If we presume that an examiner in such a case produces an explanation for how the dissimilarity might have been the product of a distortion, there is an obvious counter-argument that it is no distortion. That counter-argument has appeal, because it is Oakum's choice as the simplest answer. Any argument that the quantity of similarities should go to the likelihood of distortion gets into the issue of if the already undefined quantity necessary to conclude identification is the same quantity required to presume a dissimilarity to be an artifact. The plain sufficiency argument is tenuous enough. Extending it as a factor in determining the true or artificial nature of a dissimilarity is even shakier.

In a way, this gets right to the heart of validation and the answers to proposals that lack of information that did not impress in blank areas could invalidate conclusions of identification made from clearly impressed area, solely on vaguely developed thresholds. Were a case like this to be tried and experts to present opposing views of validity of the identification, it becomes a test of the degree of doubt or confidence assigned by the finder of fact. An issue of, yes, probability. I am quite convinced that the evolution of latent print examination is in the direction of probabilistic conclusion. Probability arguments, whether non-numerical arguments of explanations and experience in which the logic and credibility of the experts is the issue - or numerical arguments in which any fact-finding issues fall back on the validation studies that produced the numbers or how the likelihood conclusions interact with the other evidence in the case. Which of those becomes common first depends upon how fast vigorous expert defense advocacy evolves and how successfully research produces credible probabilities. I would not begin to guess which would move faster. Research, by its very nature, is not predictable, and the other is largely a matter of legal evolution that depends on astutely crafted argument in the appeals track and the general inertia of court attitudes. But the players are all in place and have available the basics of what they need to move it along. Like it or not, that's where it's going. And examiners had better think carefully about how their conclusions will stand up to a fully competent opposing expert before a jury that is perfectly aware of who you work for and that is bound to award the toss-ups to the defendant. And before you get too comfortable with the local status quo of little or no expert defense advocacy, remember that defense counsel's decision on whether or not to seek and employ an expert can be reexamined years later, in ineffective assistance of counsel and actual innocence claims.

As to this case, it may well be one of those where you "know" but can't say.

[Michele]

Terry,

I hope my comments didn’t sound as if I nonchalantly ignored the extra ridge in light of doing a thorough analysis. That was not the case. If I had to explain this to a jury, I’m sure my presentation would have multiple layers showing different signs of distortion. I would also present other examples of images that showed the visual data I’m seeing here, to show they are common signs of distortion. That’s just wasn’t possible on this chat board.

My comment “it doesn’t really matter” probably seemed flippant as well, which is not the case. Over the years, I’ve seen examiners feel like they must explain everything. This feeling is so strong that when something can’t be explained then they make something up that sounds reasonable, even if there’s no visual signs to support their beliefs. My point was that it’s always best if we can explain everything but we need to recognize that this isn’t always possible and if we can’t explain something, we should say so instead of making something up. Look at Detail 254. There are two inked prints where one has an extra ending ridge. This has been discussed on this chat board several times and I’ve heard several explanations for the extra ridge. Some explanations are plausible and others are impossible. I don’t think there’s consensus for what happened but even if we can’t explain what happened, there is still plenty of information to make an identification.

So, with the images in detail 254, there are 4 possibilities, 1) someone might say that if they can’t explain it them they must exclude 2) someone can explain away the extra ridge so they feel comfortable making the ID (even if their explanation is unsupported) 3) they can say that “ the possibility exists (however slight) that the distortion is indeed a discrepancy and an (absolute) conclusion of identification would be inappropriate” or 4) they can acknowledge that we can’t always determine what happened but there’s still plenty of information to make an identification. In my opinion, option 1 and 2 are wrong, option 3 may be appropriate in some conditions and inappropriate in others (depending on how much agreement exists), and option 4 is also appropriate in some situations (depending on how much agreement exists). Discounting that number 4 is an option forces examiners to explain things when sometimes they can’t.

[Neville]

Hi Guys
I was once told by a Photographer that a picture is worth a thousand words.

I also can see that this image is of very low quality, an image of 142kb is hardly a great starting point. Perhaps we are falling in to the same trap that the FBI fell into.