Comparison results

[Stephany Louk-Denney]

Are there any individual(s) or agencies that report "cannot eliminate" as a conclusion? An example would be a latent suitable for elimination purposes only. There are similarties with the latent and the known print; however, there just isn't enough ridge detail visible in the latent to call an identification. The tenprint card is of good quality. Is it more appropriate to call it inconclusive or would stating the individual cannot be eliminated be more concise/accurate?

Please let me know if you do or do not use "cannot eliminate" and reasons for doing so.

Thank you,
Stephany

[Gerald Clough]

There's really nothing holy about any specific brief conclusion terminology. There may be practical issues of answering questions about why you use some term that's not in common use or tacitly endorsed by a body such as SWGFAST. BUT - any conclusion has to have a precise meaning. Individualization is obvious and clear, even considering squabbling about subtle difference from Identification. Exclusion is equally obvious and clear, so long as it's clear that it refers to excluding only the portion or portions of source skin available from the record. Inconclusive, then, becomes simply the inability to make one of the other findings. What's significant about Inconclusive is that it does not make any statement relative to the source at all, which is what allows it to combine with the other two to cover every possible outcome, without overlap.

So what does Cannot Eliminate mean? In the absence of a widely published meaning (as exists for the conventional conclusions), is it loaded with an implication of positive but less than conclusive meaning? It seems loaded with the implication that, if only there was more information in the latent, it would identify the source to which it was compared in the reported examination. It is awfully dang close to "Consistent With," a term that's caused no end of trouble in the forensic disciplines. Are you going to testify to a Cannot Eliminate?' If asked what it means, are you going to say, "It means nothing." Or are you going to assign it some factual weight? And what is that weight? If you say it has some degree of value in finding fact, how do you argue the validity of that claim? I would not like to have to try. I fear I would end up right back admitting that, so far as valid meaning, it meant nothing. I'm pretty sure the next question would be how often I report meaningless conclusions. It could get much, much nastier.

Anything that has valid positive associative power is something appropriate to add weight in something like an affidavit for arrest. If you can't back up the validity, what purpose does reporting serve? And I could say Cannot Eliminate about every non-ident that wasn't a valid Exclusion; even a smudge can't be eliminated. So what's the difference between Inconclusive and Cannot Eliminate. Only the invalid association of the latent with the offered potential source. We have quite enough problems with the less than concrete threshold for Individualization. It seems to me that Cannot Eliminate has a threshold ranging from a vague mark that might be a fingerprint to just short of the examiner's threshold for Individualization. Cannot Eliminate sounds like a coy way of saying Cannot Exclude and has the same lack of meaning and the same suggestion of reaching for a conclusion that satisfies a bias toward a selected target.

Now it's true that DNA reports commonly begin with X cannot be excluded at the source, but it always goes on to explain the calculated likelihood of duplication, often in numbers that lend it extraordinary weight. We have no such valid numerical argument. Would that we did. But almost certainly not in my lifetime.

There is sometimes an impulse to signal investigators that they may well be onto the right actor. I personally think that's where you have to decide if you're an analyst or an investigator. You can be both at different times, but it requires considerable mental discipline. The very first thing I was told in criminal analysis school was that you will corrupt the meaning of data if you can't maintain strict separation of the two mindsets and stay in the analyst mindset which simply has no interest in anything but the data, meaning no professional interest in who done it. (They were right. Half the group couldn't properly use the data because they couldn't do it.) And if offered to investigators, does it represent a useful analytical service to them or a potentially serious misdirection?

[16_Points]

This is an interesting question, and I've been lucky to hear/participate in substantial debate regarding the terminology used to render conclusions. The key point, previously mentioned, is what does it mean?

ID = Same source
Exclude = NOT the same source
Inconclusive = ????????

"Cannot exclude" is at least little bit more specific. But we can do better. Are there any minutiae in agreement? (Externally verifiable phenomena for those that don't like talking about those evil "points"). If so, from where (finger, palm, phalanx joint, etc.)? Do you have other supporting data (pattern type, ridge flow, presence of focal points such as a core and/or delta, or other features that substantiate the orientation and anatomical region)?

If you searched the mark against a set of known prints and found agreement (but not enough to identify), then that should be recorded, at a minimum, in your notes (as it reflects what you actually did during your examination). I would suggest that it also be included in the written report (again because it reflects what you actually found during your exam). If we do not have a good way of talking about the "associative strength" of sub-threshold comparisons, that is not, itself, a reason to withold information from the court. Rather it is an opportunity for experts to be frank about the limitations of fingerprint identification. Further, it is an opportunity to begin the process of establishing how we should talk about these "inconclusive" comparisons.

In these cases, be clear that you have not identified the subject and that there are other potential donors in the population even though you don't know how many. After all, that's the truth.

[Amy Hart]

I interpret the question to mean going a little farther back than comparison - I think it relates more to suitability definitions. Here, a suitable print is one that, given the proper set of standards, can be identified. Since the suitability call is made before looking at any standards, we occasionally run into scenarios where the identification cannot be made. At that point, we make a distinction between suitable for comparison and suitable for identification. A print that cannot be excluded (but also cannot be identified) will then likely become suitable for comparison but not suitable for identification.

Since it is a subtle difference, whose meaning requires additional explanation, we usually only include that information in our notes. On occasion, we do report a not suitable for identification conclusion. I've never received a phone call regarding one of these reports, but then, 1) I have only sent out one or two in 15 years, and 2) I'm fairly certain no one really reads my reports anyway.

[josher89]

If something can be reported at "cannot eliminate", does that really mean anything?

I use that phrase, albeit not very often, when there are certain features in agreement but nowhere near enough to identify.

How about this; if you had a smudge with no minutiae present but you could tell that it was a left slant loop, and the suspect had five left slant loops, you could theoretically say that you cannot eliminate the suspect's left thumb, index, middle, ring, and little fingers...right? But does that really mean anything? Not really. How many people have left slant loops? You couldn't eliminate a ton of people!!!

I do use the phrase, however, I try to use it only when I have some minutiae in argreement with none not in agreement, and not enough to identify.

[Patrick Warrick]

When I read "cannot eliminate", I take that as the comparison is close but there isn't enough to exclude. That it is leaning towards exclusion. From reading some of the previous posts, some people could possibly take that as the opposite, that it is leaning towards identification but there just isn't enough. Huh? Now imagine an investigator, or even better...a lawyer, reading and interpreting that sentence!

What a perfect example of why it should not be used. Isn't it our job to be objective and give clear and unambiguous opinions. Reporting "Inconclusive" means just that...you cannot reach a decisive opinion of "Identification" or "Exclusion". I agree with Gerald, what does "cannot eliminate" really mean? Granted, having documentation in your contemporaneous notes of any consistent or inconsistent features between impressions is warranted and encouraged. But using terms that may invoke inuendos of I'm pretty sure it's him or It's close, but I just can't call it in your report without any real data (i.e., statistics) to support those statements are just seeds for disaster.

Unfortunately it seems that there may come a day in our discipline where conditional or variable conclusions, supported by feature limiting statistical models, would be acceptable for reports....but that's a whole other topic.

[Gerald Clough]

I'm fairly certain no one really reads my reports anyway.

I would caution that you can't expect that situation to continue. In the last few months, I've been shown a number of exhaustive subpoenas and discovery motions. Many or all of these are following models being actively promulgated by defense professional organizations. By exhaustive, I mean everything used or generated by the examination, a lot of other data relating to every examination by the examiner, and the entire disciplinary history of the unit. Some of these address the possible limited nature of the usual reports by demanding specifics about the basis for decisions made during examination, whether formally reported or not. I can't know how many of the defense attorneys themselves are able to interpret what they get in response, but with courts beginning to admonish defense attorneys to present experts on the issue, I have to assume that's the intent of the discovery demands or that it reflects the work of more attorneys becoming more knowledgeable.

Every examiner and unit manager will have to decide what they want to report. One of those considerations is the balance between trying to milk every possible bit of information out of a case and how various forms of conclusion and narrative can be supported and viewed within the larger view of the examination case. One of the realities is that the days of the official examiner being the only analyst to provide results are, to some degree, closing. (Not too fast, because trial courts are always short of money for experts.) How should you view your responsibility to the substantive results when there is more you perhaps could say but that was vulnerable to critique as vague and of questionable factual meaning, and that could use used to imply that the entire results should carry little weight?

If the legal hurdles that a convicted offender has to scale to get issues involving questionable forensics weren't so high, I'd worry more (from the state's perspective) about things getting into court that imply things that just aren't technically valid and having them cause trouble on appeal. That being the case, you have to settle for yourself how much you're going to report beyond the conventions of the usual conclusions, knowing that what you report can be used in ways that you may not endorse. I guess if you observe agreement short of Individualization, and that leads you to form an opinion that there is some likelihood less than certainty that you found the source, you can report it. It's probably true that the agreement you found would indeed reveal some particular likelihood, if the field had developed to the point of that kind of knowledge so you had some idea of what that likelihood was. But as the state of the art is today, someone other than you will be representing the weight to be given to your information.

There are other products of an examination, other observations you might report. L1 agreement, for instance. At least you know how to weight that in terms of patterns appearing in the population. Maybe you could report apparent agreement in the size of the fingers. Or the density of the ridges. I know few would pretend that those have much real weight, but I would also say that a genuine expert is not bound to agree with other experts and may legitimately state their own professional opinion, even if other experts would argue against it.

I think a lot of this comes down to how you view your role. If you are committed as the state's expert to develop every bit of information that could be interpreted to the state's benefit, you certainly would be doing that by explicit reporting of a "partial" identification. A defense expert would report similarly on anything that could be interpreted to their benefit.

16-points:

Inconclusive = ????????

"Cannot exclude" is at least little bit more specific.

I don't quite see that. I see Inconclusive as very precise, meaning that there are two possible conventional conclusions that can be relied upon to express beliefs in absolute facts, either an identified source or a portion of skin that cannot be the source, and that Inconclusive states neither of those conclusions are possible. I would accept that "Can neither exclude nor individualize" means the same thing as Inconclusive, but it seems to me that "Cannot exclude" is less specific, that it must be heavily qualified for it to have clear meaning. It cannot stand alone without taking on the implication of belief in some likelihood that the print to which the latent was compared was that of the true source. All three conventional conclusions can stand alone with their full meaning complete and correct. I know you weren't advocating for Cannot Exclude, per se. But I think it's incorrect to consider it at all specific.

[16_Points]

16-points:

Inconclusive = ????????

"Cannot exclude" is at least little bit more specific.

I don't quite see that. I see Inconclusive as very precise, meaning that there are two possible conventional conclusions that can be relied upon to express beliefs in absolute facts, either an identified source or a portion of skin that cannot be the source, and that Inconclusive states neither of those conclusions are possible. I would accept that "Can neither exclude nor individualize" means the same thing as Inconclusive, but it seems to me that "Cannot exclude" is less specific, that it must be heavily qualified for it to have clear meaning. It cannot stand alone without taking on the implication of belief in some likelihood that the print to which the latent was compared was that of the true source. All three conventional conclusions can stand alone with their full meaning complete and correct. I know you weren't advocating for Cannot Exclude, per se. But I think it's incorrect to consider it at all specific.

To say that something is more specific is not meant to imply that is specific. It's a relative term in this case. Inconclusive essentially means the comparison resulted in no conlcusion. Or that you conducted your comparison and determined nothing. But this isn't necessarily true. You may have eight minutiae in agreement insufficient to identify (might that actually be your conclusion?).

Pat Warrick helped illustrate my earlier point (Thank you, Pat). He took "Inconclusive" to mean leaning towards an exclusion but not quite there. Whereas I was using it to refer to the opposite. So I would submit that "Inconclusive" essentially means nothing.

Further, I don't see the problem with reporting an association between a mark and a known that is below the threshold for identification. Arguments that such testimony will mislead the jury are spurious at best. I've read numerous transcripts of examiners testifying, in court, to a 0% error rate for ACE-V. Now THAT is misleading a jury. Yet this practice is still acceptable and reporting a subthreshold association between a mark and a known is not. This seems to be a contradictory position.

It is a jury's duty to "weigh" out the facts of the matter. Their ability or inability to successfully complete that duty is mainly a refelection on the court. Though the expert does have a responsibility to educate and assist the jury in interpreting technical data. You are not helping a jury determine a matter of fact by withholding relevant findings from your examination. But do not assume that this means we should practice "drive by science" that is to say reguritating some canned phraseology and leaving the jury to figure it out on their own, not at all. The expert should be prepared to qualify his/her statements and provide detailed explanations, if necessary. When asked a question that you do not have a proper scientific basis, just say you don't know.

For example, if asked for a match probability for the eight minutiae found in agreement, you say, "I don't know. The statstical studies needed to calculate that type of probability have not been carried out."

What's wrong with that answer if it is, in fact, the truth?

[Gerald Clough]

To say that something is more specific is not meant to imply that is specific. It's a relative term in this case. Inconclusive essentially means the comparison resulted in no conlcusion. Or that you conducted your comparison and determined nothing. But this isn't necessarily true. You may have eight minutiae in agreement insufficient to identify (might that actually be your conclusion?).

Okay. I see the point now. Essentially, it's reporting the data, and the conclusion is that you found data. So this really becomes a broad question that I might put as, "Is it appropriate to report data that does not lead to a valid conclusion that identifies or excludes?"

There's a conundrum here that I have no ready answer for. I'm not sure at all how various examiners might educate a jury on the implications of a less than conclusive match. I suspect they may not get much chance to do that anyway, since witnesses are strictly constrained by the questions they are asked. How would such a thing be represented in testimony? Just pointing out the agreement with no further comment? I suspect that's where most prosecutors would leave it. And I suspect most defense attorneys would ask just what you anticipate, asking if there's any basis for associating the latent with the defendant. (They very well may not, given the common state of ignorance and the court's routine conservation of funds for extras, like defense expertise.) I agree that you'd have to say there was no basis for assigning a likelihood.

I suppose my answer is that, the less precise the conclusion, the more the examiner has to expound formally in the report to place those observations in proper perspective, or at least the perspective intended by the examiner, which among various examiners looking at the same case, might range from a great deal of weight to none at all. Maybe this sort of thing has not often been reported simply out of shrinking from presenting results even more difficult to explain than the ever-present undefined identification threshold. There's a certain tit-for-tat logic for reporting partial match. Defense counsel frequently claims, "My client's fingerprints were not found at the scene." Unless no prints were found, or all prints found excluded the defendant as the source, the claim that his fingerprints weren't found is in supportable as fact. Maybe they were found, just not of sufficient detail to identify.

After this exchange, I really don't have a firm position on this issue. I have some concerns, both about how the discipline is viewed and for how the information is weighted by the jury. I'm always bothered by something that has no clear limit. Do you report all agreement, no matter how little? Five details in agreement? Three? Two? Or only when the agreement reaches some other threshold of "just short of Individualization"? I guess a lot of the questions are exactly because this hasn't been much discussed as a presumed valid reporting issue. Of course, a lot of this also grows out of a poorly simplistic focus on specific "approved" conclusion words and no consensus on reporting beyond that. None of them mean anything specific to anyone outside the field without a lot of explanation.

I'm afraid, though, that I still see it as reporting data that we say requires expertise to conclude from and asking a lay jury to form their own conclusion from data that from which we cannot conclude. If that's appropriate use of the data in the form of the latent evidence, what is the point of expert conclusion, then? I suppose we could hand the jury all the physical evidence, guns, bullets, casings, fingerprints, tires and tire marks, and shoes and shoe marks, and show them how to use comparators and let them work out a conclusion for themselves. Then we could all go home and get some gardening done. Maybe examiners thinking about this should just consider, if they were on trial, would they care to have what they report presented to their jury. (Assuming, of course, that they'd be okay with being convicted by a valid ID. ) And, would they care to have the prosecutor in any case wave their report at the jury and declare as the very last words spoken to the jury, "Here's the report. The expert says they match."

[16_Points]

After this exchange, I really don't have a firm position on this issue. I have some concerns, both about how the discipline is viewed and for how the information is weighted by the jury. I'm always bothered by something that has no clear limit. Do you report all agreement, no matter how little? Five details in agreement? Three? Two? Or only when the agreement reaches some other threshold of "just short of Individualization"? I guess a lot of the questions are exactly because this hasn't been much discussed as a presumed valid reporting issue. Of course, a lot of this also grows out of a poorly simplistic focus on specific "approved" conclusion words and no consensus on reporting beyond that. None of them mean anything specific to anyone outside the field without a lot of explanation.

Excellent point (no pun intended). I would suggest that a "partial match" (if that's what we wish to call it) should only be reported if the mark is sufficient for Exclusion. However, I'm open to debate on this topic.

I'm afraid, though, that I still see it as reporting data that we say requires expertise to conclude from and asking a lay jury to form their own conclusion from data that from which we cannot conclude. If that's appropriate use of the data in the form of the latent evidence, what is the point of expert conclusion, then? I suppose we could hand the jury all the physical evidence, guns, bullets, casings, fingerprints, tires and tire marks, and shoes and shoe marks, and show them how to use comparators and let them work out a conclusion for themselves. Then we could all go home and get some gardening done. Maybe examiners thinking about this should just consider, if they were on trial, would they care to have what they report presented to their jury. (Assuming, of course, that they'd be okay with being convicted by a valid ID. ) And, would they care to have the prosecutor in any case wave their report at the jury and declare as the very last words spoken to the jury, "Here's the report. The expert says they match."

We are always asking the jury to form some type of conclusion. In one case, I report that mark "A" was identified to John Doe. In another case, I report that mark "A" has minutiae in agreement with John Doe insufficient to identify and that all other subjects were excluded as the source of mark "A".

In either case, you could argue that the jury is not really interested in whether or not John Doe is the source of the mark, but rather, "If John Doe is the source of the mark, how will that affect the jury's interpretation of his guilt or innocence." The answer may be that the source of the mark has no bearing on the jury's deliberation. In which case, the type of conclusion rendered is irrelevant. It may be that the mark is quite influential and that any association between the mark and the subject is significant (even if it is not deemed to be so by the strictures of established fingerprint conclusions). We may have no way of knowing, and it is not for the expert to decide what information the court/jury is entitled to. The role of any witness is to tell the whole truth, not just bits and pieces. Furthermore, an approach that rejects the detailed reporting of "inconclusive" results is inherently misleading.

Consider an expert who only testifies to IDs or exclusions. To a jury, it appears that the expert always knows the answer. "Yes, John Doe made this mark. No, he did not make that one." This is patently false and lacks credibility. It seems to imply that ACE-V always leads to an answer (it doesn't). You may argue that that is self evident, but, if so, where are the examples? Who knows because no one reports them?

[Stephany Louk-Denney]

Thank you to everyone who has responded so far. It's helpful to receive feedback from others in the field.

Okay. I see the point now. Essentially, it's reporting the data, and the conclusion is that you found data. So this really becomes a broad question that I might put as, "Is it appropriate to report data that does not lead to a valid conclusion that identifies or excludes?"

Yes, Gerald this is my main question. Should we report these findings and if so, how should we do so?

In another case, I report that mark "A" has minutiae in agreement with John Doe insufficient to identify and that all other subjects were excluded as the source of mark "A".

16_Points, do you write report this way? I'm interested to see if you do and how this holds up in court.

[L.J.Steele]

We are always asking the jury to form some type of conclusion. In one case, I report that mark "A" was identified to John Doe. In another case, I report that mark "A" has minutiae in agreement with John Doe insufficient to identify and that all other subjects were excluded as the source of mark "A".

You, the expert, are asking the jury to form a conclusion to accept you, your methods, and your results as credible evidence. The prosecutor is the one asking the jurors to reach conclusions about the significance of your testimony and evidence.

Once the possibility of other subjects is raised, that may get the expert into all sorts of interesting testimony -- who were those other subjects? how were those people included and why? If the answer is "all other subjects" included only police and crime scene personnel at the scene, that may lead to one set of closing arguments. If "all other subjects" includes law enforcment/crime scene personnel, and those with legtimate access to the object, and the defendant and half-a-dozen members of the gang defendant is associated with, that leads to different closing arguments (and perhaps into a thicket of issues about third-party culpability, prior misconduct evidence, and so on.)

This may be something to discuss with the lab's legal folks, perhaps in consultation with the prosecutors and your local public defenders.

[David Johnson]

L.J.Steele said

who were those other subjects? how were those people included and why?

It is most likely that you would be refering to the other subjects that you were asked to compare in the case...

I was asked to compare Jane Doe, John Doe and FNU LNU, ect. Jane Doe has detail in agreement, insufficient to identify, but I can exclude John Doe and FNU LNU. You could then go one to state why you can not exclude or identify the print to Jane Doe. Something wrong with the standards or is there just not enough information in the print to reach that conclusion.

[Pat A. Wertheim]

This thread seems to be dancing around the "possible/probable/likely" conclusion without actually addressing it. So let me play Devil's Advocate and ask the question, although I know some would crucify me for even suggesting there is an elephant in the room.

We currently use a three level scale: 1) Identification; 2) Inconclusive; and 3) Exclusion.

What is wrong with a 5 level scale: 1) Identification; 2) Probable match; 3) Inconclusive; 4) Probably not; and 5) Exclusion.

I am not a document examiner, but isn't a similar scale used in handwriting conclusions?

Or a 7 level scale: 1) Identification; 2) Strong probability; 3) likely; 4) Inconclusive; 5) not likely; 6) strong probability not; 7) Exclusion.

Is the 3 level scale really good science, or is it just dogma?

[Gerald Clough]

Excellent point (no pun intended). I would suggest that a "partial match" (if that's what we wish to call it) should only be reported if the mark is sufficient for Exclusion. However, I'm open to debate on this topic.

Okay. In effect saying there's enough there to exclude, were the record print actually not that of the proposed source. I guess that's a given. If there's not enough for exclusion, how could you know if it was specifically not that source? So with that, you're automatically in the middle ground between "could be" and "definitely is." And if you cant' get into that space, there's truly nothing to say about the latent and the record.

We are always asking the jury to form some type of conclusion. In one case, I report that mark "A" was identified to John Doe. In another case, I report that mark "A" has minutiae in agreement with John Doe insufficient to identify and that all other subjects were excluded as the source of mark "A".

Yes. That's a classic problem for judges, deciding if the jury can make appropriate use of some evidence and whether a particular expert witness will render the evidence "more probative than prejudicial." The most interesting area where this is evolving is in eyewitnesses, where the common question used to be whether general experts, like psychologists and those who study perception could assist the jury with judging eyewitness evidence. General experts are usually excluded. (Same reason Simon Cole and the like don't get to testify.) Federal circuits are split over whether it's error to exclude experts who specifically study eyewitnesses, as they do in such places as the UT-El Paso Eyewitness Lab. The trend appears to be that they should testify. Obviously, an examiner is an appropriate expert for a fingerprint issue. But it would require close and knowledgeable questioning to establish if the scope of the primary examiner's testimony genuinely made it possible for the jury to grasp the weight of the partial match. (Using that term for convenience, since I think we know what we mean.) And, of course, another examiner offering opinion on whether or not the partial match should even be presented would likely be proper and would make for an interesting pre-trial hearing. The core argument would be whether or not the agreement is sufficiently probative.

Furthermore, an approach that rejects the detailed reporting of "inconclusive" results is inherently misleading.

That would be the legal question. Whether or not presenting the agreement would mislead the jury by implying the source or not. Of course, that's a trial issue, not really a reporting issue. It's hard to say how many prosecutors would use it. If they are knowledgeable, I think many would consider it dangerous. And that comes back to what the examiner is telling the prosecutor about what the agreement means and how it is vulnerable. I've long held that a true expert is an expert and may opine according to his beliefs, without being bound by conventions.

Consider an expert who only testifies to IDs or exclusions. To a jury, it appears that the expert always knows the answer. "Yes, John Doe made this mark. No, he did not make that one." This is patently false and lacks credibility. It seems to imply that ACE-V always leads to an answer (it doesn't). You may argue that that is self evident, but, if so, where are the examples? Who knows because no one reports them?

I largely agree. The worst example was the classic testimonial routine where the prosecutor asks if the examiner carried out analysis. "Yes, I did." Did you form an opinion? "I did." What was that opinion? "That the mark, Exhibit N, was made by the right index finger of the defendant." No more questions. I hold that, lacking true validation of conclusions, examiners can be represented as expert guessers. Guessing isn't evil. Your physician is an expert guesser. He's not concluding from a validated process leading to scientific certainty, but you bet your life on his guesses. And he's wrong a lot more often than examiners. We just can't afford to say he can't guess, because the need is urgent and unavoidable. And another physician can have a different opinion and often does. We just rarely know the ground truth, because patients get better for various reasons. (There's always autopsy, of course.)

I have to wonder if an expert (assuming competent defense counsel - a poor assumption) would even be allowed to testify to data from which he could not draw a fact conclusion. The fact of the agreement is not a conclusion about the source of the mark. It's just a reporting of observations. I think this would be a very real legal issue. It's not analogous to a firearms examiner reporting on caliber. That's a specific class conclusion, just as is reporting L1 agreement where no L2 details are clear, and we're reporting a specific class conclusion and can characterize the (very weakly associative) frequency in the population. Would an expert be allowed to testify to observations that cannot reliably produce a fact conclusion beyond the data itself? Never? Only when it rebuts something like, "Those aren't my client's fingerprints."? Lisa!!!

We're placed in a difficult position by historical reliance on the state's examiner as the only analyst. It shouldn't be that way. Ideally, we would be routinely challenged, or at least reviewed, by a defense expert. We should not be placed in the position of knowing there are other expert opinions that might well oppose our own and also knowing we probably have it within our power to seer that those opinions are not revealed. I think that's where those who imagine that making latent labs independent of police agencies will provide any real benefit are mistaken. I think, in fact, that it is likely to make things worse by implying that removing an imagined bias means the results will be accurate. When it's a matter of professional expert opinion that may be argued, the fact-finder doesn't have the full picture unless it's subject to argument.

Stephany:

Not answering for 16, but I think we're talking about reporting something like, "Eight minutiae were sufficiently clear to be reliably characterized. All eight were observed to be in agreement with minutiae of the same character in the record print, and in each print, all eight appeared in the same spacial relationship among themselves. In this case, those minutiae were not sufficient to conclude that the latent print was made by the same source, and the source of the record print could not be excluded as the source of the latent."

I don't think one could go on to say that more detail would certainly identify the source, because that can't be known. But I think one could accurately report that more detail would be needed to conclude whether the latent was made by the same source or if it was made by a different source. We can't know whether that would require some specific additional quantity, not without making a lot of presumptions about the nature of the prospective additional details.

Pat:

I'm kind of with you on the multi-level opinion thing. There are indeed other fields where experts do, from time to time, conclude "probably" or "probably not." Not if they can at all help it, if only because wishy-washy opinions often don't get them hired again. But the option is there.