CLPEX Chat Board

Hello,

I am a new Latent Print examiner-in-training (with 2.5 years previous experience in Tenprints) and was wondering what other agencies do after an ID is made.

Scenario: 4 suspects (A,B,C,D) and their tenprint cards, and a latent print. An identification is made to A, before B, C, and D are compared.
Do you stop comparing or continue?

There have been a lot of changes in definitions and uses of individualization/identification/association within the last decade or so and I am curious to if/how other agencies have adapted in response to these changes.

In another thread, someone (sorry, I can't remember which thread or who), stated they stop, but then talked about (again sorry based off my recollection, so not word for word) about not using 'exclusion' unless prints are actually compared, because to exclude, a comparison has to be made (so 'not compared' instead of 'exclusion'). But what would be said if, as a witness, the question, "If you can't exclude unless you compared, why didn't you compare the other suspect's tenprints?, You are showing bias against my client." or something similar is asked?



Thank You,
LEC

In todays climate of confusion over what you can't say during testimony, I can understand your question. I am not being sarcastic in my answer but a little frustration may show, please excuse that.

If I lose my i-phone and begin a search for it. I look until I find it. I don't keep looking after I find it.

To take your question to the next level. What if there is one subject submitted for comparison and you identify the print. Do you do an AFIS search to confirm that another print in the database doesn't match your latent.

If you get an AFIS hit in the local database. Do you initiate a statewide search after the local hit? An FBI search?

Where do we stop letting theory cripple our ability to complete cases.

Just my opinion

LEC--welcome to the board and welcome to the profession.

I will echo what Bill said but add one thing that another clpex user (ER) and I have discussed in detail; both in person and I think on this forum. When it comes to an AFIS hit, it is okay to keep looking a few more down IN THE EVENT THAT a different record is associated with that individual. For example, if this person had a criminal record and a civil record (because of a license or something), their fingerprints may be kept separately in AFIS and you may get more than one 'hit' in AFIS because of this. So, it's always a good thing to keep looking just to be sure. It certainly doesn't mean that you should expect to find someone else, completely different, to have the same minutiae.

Another caveat to your question is how do you record/report your observations when doing manual comparisons? If you have the four suspects, and looked at A, B, and C before ID'ing to D, do you report that you excluded A, B, & C and made the ID to D or do you just report that you made the ID to D? I think your notes should record exactly what you did (excluded A, B, & C and ID'd to D) but your report should only include what is likely what the customer is concerned about; the ID to Suspect D. Your notes will contain what you did in the event that it comes up later (if it would come up at all).

Bill Schade - Thank you for your answer. I'm here to learn, and value your opinion. I've brought up your point about the i-phone to colleagues myself (using a different object) and understand that.

continuing with

josher89 - Thank you as well. when working in Tenprints, we would get up to 3 candidates in AFIS (tenprint to tenprint). We would look at all candidates, for the reasons you stated; Database mergers, better matchers, "lights out" not catching bad quality tenprints, etc., but not because we were trying to find different people, but to consolidate records if someone had accidently had multiple SID numbers or if there were 'shell' records from before we had AFIS.

In notes we list everything. In the reports, we do as you said and list the item/latent number and the person IDed.

That's come up on here before, under the topic of continuing to compare AFIS candidates after you ID one on the list.

http://clpex.com/phpBB/viewtopic.php?f=2&t=1843

Boyd Baumgartner - Thank you for the link, I'll read over it.

Side Note: I want to thank you for the online latent print course you offered. I took it and it was extremely helpful in starting my career in fingerprints; first in Tenprints, and then in transitioning to latent prints.

Awesome! Glad I could be of help. Keep the questions coming, this site is a great resource.

Hi LEC,

Good discussion and Bill has a very important point: at what point does theory make things unnecessarily difficult and impractical? That's a philosophical discussion I'm not going to tackle here, but at the end of the day, it's a valuable consideration.

What I'd like to do is make you think a little about Bill's scenario and tweak it a bit to what I think is a closer representation of what we do in latent prints: making decisions under some level of uncertainty...perhaps it's trivial miniscule uncertainty (Back to Bill's practicality point), perhaps it's larger and measurable, and not so easily dismissed. At the moment, we can't quantify it, which makes this a more challenging discussion.

Suppose in Bill's scenario, it's YOU that is searching for BILL's phone. Not BILL (who will have no uncertainty whatsoever when he finds his phone). But say you are tasked with finding Bill's phone, which you have never seen before, and you have to do so from photographs. Maybe the photographs are very clear and show serial numbers and other identifying distinct characteristics. Maybe Bill has a pink polka dot, monogrammed, Hello Kitty case, autographed by Jeb Bush and that's all clearly in the photograph making it an easy search and no uncertainty when you find something matching that description.

On the other hand, suppose the photos are dark, blurry, maybe even "distorted", or something is blocking a portion of the phone so you only have a "partial" view of it. Moreover, there isn't anything distinctive about the phone, it's very generic. And to make matters worse, you are searching not Bill's house with a small number of phones and high chance of finding it there to begin with, but instead have to go to the iphone warehouse with 1000's of used phones. Perhaps you find something that looks A LOT like it...but is missing one feature from the photo, but has all the other features. Would you keep looking then? At what point does your uncertainty vanish to a point where you are sure no other phone will bear those characteristics and you are sure that you are not making an error when you declare "I found it!!! I can stop looking".

Something to think about.

g.

It's good to see some new blood here... Welcome aboard LEC!

I agree to some extent with Bill's shake on the profession, there is a practical application side to this business that shouldn't be ignored.
Identification at it's core is about belief; The stronger your belief that you have resolved the print the more confident your decision.
With a robust print the strength of your belief in an identification can be overwhelming. However, when we have analyzed a print that has put us close to our sufficiency threshold(s) we will need additional focus and scrutiny to secure our beliefs. In these cases we should do everything we can to test an identification hypothesis.

Shane

g.

I like your further explanation on the search for my phone. It makes an important distinction and shows why we need to move away from the binary decision of "ID" / "No ID".

So the conversation comes back to "Strength of Conclusions"

Whether its a numerical probability or a verbal scale, its the biggest shift in the discipline I have seen and one I still struggle with.

It's nice to see some discussion on the chat board again. Thanks to everyone who is willing to put their thoughts out there for comment

Bill

Thanks everyone for the continued responses.

The scenarios on trying to find the i-phone with a photograph reminded me of a situation that happened some years ago to a friend. This was back when cars had 2 keys. One to open the door and one to start the engine. He was returning to his parked car in a mall parking lot. He 'found it' and opened it, got in, and went to start the engine, but nothing happened. Looking around in the car, he realized there were things in it that were not his. come to find out, it was a different car parked a few spaces away - same make/model/year/color. He happened to encounter that extremely improbable statistical number that all of these factors plus the door key matching were identical. Had he stopped after opening the door, he would have been sure to 'ID' the car as his (such as dropping off items into the car in order to return to shopping). But going that extra step to start the engine he was able to see it as a "close non-match".

When the basic premise is: permanence + uniqueness = ability to ID, it seems easy. But then distortions, partials, lack of quantity/quality, etc. are added.

A 'small world' case with only 3 possible suspects that you have tenprints for vs. a 'large world' case with no suspects that you are searching through AFIS/FBI. Does that change my "ID and stop" based on the situation? And if so, then I have to find where the point between the stop and go is. And then what if my point is at X and one coworker-verifier's point is at B and another's is at Z. And if I add the verbal scale to it there will be debate on where each of those points should be.

A lot to think about.

-LEC

Going back to the original proposition about persons A, B, C, and D....

There is an additional slant on this topic that I believe is important. Exclusion decisions are very rarely probative. It may be extremely important in other forensic disciplines (e.g. DNA mixtures) to continue comparing and exclude the other people even after identifying the first one. However, in latent prints an exclusion normally doesn't provide any useful information. An ID indicates that a person touched a surface, but an exclusion doesn't indicate that a person didn't touch a surface. This limitation of exclusions could be perceived as bias, but it is just inherent in this field.

tl;dr It's not just the ID that causes comparisons to stop. It's also the (mostly) uselessness of the exclusion decision.

I'm going a little off-topic, but I thought I'd share a recent case I had where an exclusion would have been probative.

A man is found dead in a bloodstained hotel room with knife injuries. Before a PM can determine whether it was suicide or not bloody fingermarks are found and photographed.
I was called out to examine the marks. None of them are of value for identification, but one of them has value for exclusion (based on a clear whorl pattern and very limited L2 detail).
If I can exclude the victim it literally indicates another person with blood on their hands. As it turned out I couldn't exclude the victim, but it would have been valuable in the very early stages of the investigation to implicate the presence of another person at the scene between the time of the injury and the time of discovery of the body.

Hey Bill and g,

Better use an example other than the iPhone...there are always going to be these people.

https://www.youtube.com/watch?v=HxXbrnJ6l4A

Just thought I'd begin the day with a laugh!

g. wrote:Suppose in Bill's scenario, it's YOU that is searching for BILL's phone. Not BILL (who will have no uncertainty whatsoever when he finds his phone). But say you are tasked with finding Bill's phone, which you have never seen before, and you have to do so from photographs. Maybe the photographs are very clear and show serial numbers and other identifying distinct characteristics. Maybe Bill has a pink polka dot, monogrammed, Hello Kitty case, autographed by Jeb Bush and that's all clearly in the photograph making it an easy search and no uncertainty when you find something matching that description.

On the other hand, suppose the photos are dark, blurry, maybe even "distorted", or something is blocking a portion of the phone so you only have a "partial" view of it. Moreover, there isn't anything distinctive about the phone, it's very generic. And to make matters worse, you are searching not Bill's house with a small number of phones and high chance of finding it there to begin with, but instead have to go to the iphone warehouse with 1000's of used phones. Perhaps you find something that looks A LOT like it...but is missing one feature from the photo, but has all the other features. Would you keep looking then? At what point does your uncertainty vanish to a point where you are sure no other phone will bear those characteristics and you are sure that you are not making an error when you declare "I found it!!! I can stop looking".

Something to think about.

g.

Quite possibly the best fingerprint analogy I have ever read.