Evidence in Minn. v. HULL

[Les Bush]

The latest Weekly Detail covers Glenns involvement with the HULL case and his explanation of the reasons he chose a certain theory as to how fingerprint evidence was to be presented. The critical point Glenn makes is the differentiation between 'identify' and 'individualize" with his position being that 'research' does not support the claim of exclusive source. He clarifies his argument that latents could be produced from two sources that are similar enough that an examiner may not be able to exclude either source. In any court case it has always been a tactic by defence counsel to propose hypothetical situations in order to progress a certain strategy. The common response is usually to negate the proposition that it doesnt relate to the actual case circumstances and therefore shouldnt be considered. What Glenn is tinkering with here is a home goal. SWGFAST have defined very clearly that we deal with factual outcomes based on our professional judgement, there is no support for probable or likely results. By avoiding the term 'individualize' in preference to 'identify' Glenn is opening the door that our conclusions are not definitive. In the Mitchell case the main arguments in support of fingerprint evidence were its reliability when conducted by the safe assessment of an expert. Stepping back from that position requires a considerable amount of debate and argument and these things usually begin with a court case. Cheers from oz. Les

[Dogma]

From Les: The critical point Glenn makes is the differentiation between 'identify' and 'individualize" with his position being that 'research' does not support the claim of exclusive source. He clarifies his argument that latents could be produced from two sources that are similar enough that an examiner may not be able to exclude either source.

Glenn’s point that two fingerprint impressions, though from different sources, could appear similar enough to “fool” an examiner into the belief that they in fact share the same source, should not figure into his reporting process. Allowing for the conclusion that scientifically an absolute statement can not be made is one thing. The inability of an examiner to discern two different sources is not a part of that. Its like going into court and reporting the results of your chemical analysis and then issuing the caveat that the equipment you used was unreliable.


From Glenn (Detail): “…we actually did not use the word "individualization" because it is clearly defined by SWGFAST as "determination that two friction ridge impressions originated from the same source to the EXCLUSION OF ALL OTHERS". We used the term "Identified" (which is still allowable by SWGFAST)….”

It is disingenuous for Glenn to state that they used the term "Identified" because "individualization" means “to the exclusion of all others” according to SWGFAST. The fact is that as it now stands, and stood at the time of his testimony, the two terms are synonymous according to SWGFAST.


Fingerprint comparison conclusions have a number of conflicting and confusing aspects:

Examiner confidence: No two people have ever been shown to share the exact same arrangement of ridge detail. Research has demonstrated the unique aspect of friction ridge formations. A claim of individuality is therefore, intellectually possible and proper.

Scientific conclusion: Individualization infers an absolute exclusion of all other sources which can not be claimed unless all other sources have been examined.

Legal requirements: The standards for admissibility of an expert’s conclusions may vary from court to court and jurist to jurist. Different courts (jurisdictions) have arrived at different conclusions using the same criteria: Daubert, Frye, Mack, etc. (perhaps a failed methodology?)

Our challenge is to find a way for these diverse aspects to “play well together.”

[Gerald Clough]

All in all, I think this is has been a productive exercise in presenting latent print comparison without the traditional statements that are logically offensive. We need not fear admission that there is a possibility that some particular latent, sufficient for identification, might be mistaken as being from another source. That's where our observation that, being trained and experienced and in touch with the discipline, we know that our latent is such a sufficient impression that no other source has been shown to produce an impression that would mistaken as having the same source supports our belief that the possibility is extremely remote. It becomes a matter of presenting the evidence and demonstrating that it has considerable power to identify and that, because there is no other contrary evidence of such high credibility that it would cause one to believe that we have found one of those extremely unlikely cases.

I see this kind of like E=mc^2. Has been of great value for many years, no contradicting credible observations - but only proven this year. Except that I don't think we're ever going to prove anything absolute to answer challenges to print identification. We will, however, refine its power and very likely extend its value through probability. I predict great gnashing of teeth when we've adjusted our arguments and the critics find that fingerprints have essentially the same power with juries as before. Even more one day, if probabilities begin to have meaning, and they still have power.

[RL Tavernaro]

Perhaps I qualify as something of a 'dinosaur' in the profession, since I have been involved with fingerprint identification/individualization since the early 1970's. I have seen, heard and participated in many debates regarding terminology, practices, and so-called strategies or approaches during the past 35+ years. In my experience, both testifying, giving presentations both in and outside of the criminal justice system, and teaching at the college level (where I am now presented with hundreds of new and fresh examples of members of our potential juror pool), I have found that simplifying is often the most effective method of communication. This applies especially in the courtroom, where your time is limited and largely out of your control, and interaction with arguably the most important members of your audience (the jurors) is prohibited.

Specifically, where we deviated from typically strategies were the following issues:

1) ... "zero methodology/all errors are up to examiner/and you can't hold me accountable for everyone else's error rate" is neither a helpful answer for the court nor an informative one. Forget if it's true or not...it just doesn't help the court (in our opinion). ...".

I believe you are on to something here, although not necessarily related or limited to the immediate content. The entire purpose for our testimony should be to 'help' the court in it's understanding and decision making. Testimony should first & foremost be directed to the jury, although it may be helpful to keep in mind the advisability of creating a record for potential critical appellant review.

2) We did not commit... the "Examiner Fallacy". ... "BECAUSE friction ridge skin is unique, it directly follows that when I have found 'enough ridge detail in sequence to individualize' it is that person and could be no one else (and there is zero chance I have made an error since I followed the method)." We took what we believed to be a more defensible position: "We found X features in agreement. The more features you have in agreement, the less likely chance of reproduction in another individual".

As stated previously, I have been around awhile. I do not recall ever having testified to a factual infallibility of fingerprint identification/individualization, or a "...zero chance I have made an error...". This includes testimony given in the beginning of my career based on training I received at that time, along with my own sensibilities regarding logic and defensible positions. Although I have seen testimony and transcripts along these lines, I have never felt comfortable with it.

3) Therefore, because of #2, we actually did not use the word "individualization" because it is clearly defined by SWGFAST as "determination that two friction ridge impressions originated from the same source to the EXCLUSION OF ALL OTHERS". We used the term "Identified" (which is still allowable by SWGFAST) and interpreted it in the following manner:

"We have identified the latent prints as having originated from the defendant. We did not compare these latent prints to every single human on the planet (past, present or future) therefore we cannot dismiss the theoretical possibility that someone out there may be able to produce a similar latent print. We don't believe that someone has IDENTICAL friction ridge skin to the defendant, but we can't dismiss the possibility that someone may have an area of friction ridge skin that may be similar enough to produce a similar looking latent, and that given the quantity, quality and specificity of the ridge detail, a human examiner applying the ACE-V method may not necessarily be able to exclude it as having come from the other source."

And finally, back to my concept of simplifying. My college classes consist of a relatively broad cross-section of our general population (USA). They range in age from 18 years to 60+ years old, and across a wide spectrum of economic, cultural, and even geographic areas. They all have at least a high school or equivalent education, and some have some prior college experience. I believe they give a fair representation of a major portion of our potential juror pool. They virtually all recognize the concept of fingerprint 'identification' on the first day of class, prior to contamination by my instruction (I teach both 'identification' & 'individualization' as presented in SWGFAST. On that first day of class however, fingerprint 'individualization' confuses most of them.

Explanations concerning theoretical possibilities of 'error' may be necessary in certain situations for rebuttal, however it could work to a disadvantage if attempts were made to incorporate it unnecessarily in primary testimony. In my opinion, it could well serve to complicate things, when to simplify would be the better option.
Regards, RLT

[Les Bush]

Hi again,

Having had the opportunity to read the transcripts of Glenns evidence the following is offered in response to his position that "you can't prove individualisation". In the late 1980's I had a similar discussion with a fellow trainee who I regarded as intellectually sound and scientifically astute. His dilema was that if all persons have not been fingerprinted and compared then there is a chance of duplication. Mathematically that position meets the requirements of probability but it doesnt incorporate fingerprint knowledge or the conditions under which fingerprint examinations are conducted. Also there is no probability model existing that incorporates all the variables associated with the development of fingerprint patterns and the transfer onto a substrate. In the absence of a mathematical solution the fingerprint community relies upon the following conditions for its belief in the individualising power of fingerprints. Friction ridge identification science is supported by the known biological development of friction ridge detail and its patterns. This same biological process continues to apply to every human across both palmar and plantar areas ( with the exception of abnormalities). The extensive scrutiny of the friction ridge detail on the distal phalanges of hands has substantially proven that duplication of patterns of friction ridge details sufficient for individualisation does not occur on two humans. The principles of the biological uniqueness and permanence of friction ridge detail meet the requirements for scientific falsifiability but as yet they have not been proven to be untrue. As David Ashbaugh states on page 9 of his text "The level of knowledge required to function as a forensic identification specialist today is far greater than only a few years ago." ( 1999). Cheers from oz. Les

[Gerald Clough]

The extensive scrutiny of the friction ridge detail on the distal phalanges of hands has substantially proven that duplication of patterns of friction ridge details sufficient for individualisation does not occur on two humans. The principles of the biological uniqueness and permanence of friction ridge detail meet the requirements for scientific falsifiability but as yet they have not been proven to be untrue. As David Ashbaugh states on page 9 of his text "The level of knowledge required to function as a forensic identification specialist today is far greater than only a few years ago." ( 1999). Cheers from oz. Les

To play Lucifer's Lawyer, I respond with perspective on how we may be useful.

"Substantially proven" is very far from "proven." For instance, any number or mathematical theories have been or long were proven for most cases but are placed, as they must be, in the Unproven category. Now, I think realistically we have to recognize that those mathematical theories are, for the most part, believed to be provable, that one day some clever fellow will absolutely prove or absolutely disprove. The basis for believing they are provable is largely a feeling that to the mathematician it seems right. In the fingerprint problem, the theory is not only unproven, but is unprovable and grounded in circular dependence on the term "individualization," essentially saying, "It is possible to individualize because there's always some point where we assume we can individualize." As you point out, we have no good probability model of the factors being considered in fingerprint examination. If we can't model the thing we depend upon, we really can't claim "substantial proof" or provability. We either play on the scientific field, or we play on the practical field. "Proof" is a home run in the scientific ball game. Useful application of convincing reliability is the equivalent value in the practical game.

We are further looking to get slammed whenever we talk about "not proven to be untrue." That, in fact is one way of saying a theory is unproven and is often a statement of blind faith on account of presumed unprovability. The four-color map theorem had not, until recently, been "proven to be untrue." In the scientific realm, that is exactly the same as saying it had not been proven, and no one would bother to point out is had not been proven to be untrue. That is the ultimate destination for well stated theories. You hope to one day show one or the other. The four-color theorem could not, therefore, until proven true or false, be considered mathematically true and could not be credibly used to extend knowledge by taking it as true. But map makers could go blithely along buying only four colors for their production with considerable confidence that they were right.

I think we need to walk away from grasping for validity based on presumed provability. Unlike mathematical theories an expert might reasonably presume to be provable, I don't think we can ever prove uniqueness. We always fall back into qualifications like "sufficient" and "substantially," and we certainly don't pretend to be able to divine all the conditions of deposition by observation of the impression.

We can, though, make a solid argument that latent print examination can provide valuable and very persuasive evidence, in fact one of the most powerful of physical evidences, for exactly those reasons cited in trying to establish "proof," the considerable knowledge of how friction ridge skin forms, the very considerable history of observation, and the ability to demonstrate the process and conclusions. I think we tend to fear backing away from absolute conclusions because we fear what will happen when latent print evidence is the sole evidence. I don't think we have any valid fear there. But we should recognize that, in such cases, we will be closely examined, and the evidence will have to meet a high standard. That is exactly as it should be. Since we cannot absolutely eliminate all chance of a source different from the defendant, the latent evidence is inescapably subject to being judged on its likelihood of having correctly identified the source. In simple terms, that can run from "could be" to sufficient to accept as "beyond any reasonable doubt."

We don't like that. We would like it to be absolute, but that is neither true nor necessary. And it's becoming increasingly clear that trying to present it as a kind of "good as" scientific absolute can appear dishonest and invites attack. The last quoted line provides modern perspective. We know enough, we have enough solid argument on our side, that we can present evidence of great power, and we can turn the attacks around, showing that it's they, not we, who are desperately trying to defy reason. Look back to the four-color theory. If there happened to be a legal case in which the issue turned on whether or not every map could be colored with no more than four colors, prior to the published proof, the theory could not have been offered as absolute truth. But the practical motivation for considering the problem and the vast experience that no map maker has ever had to run out and buy a fifth color, a mathematician's explanation of why it's likely to be at least practically applicable, as well as graphical demonstrations of a variety of aspects of the problem, and the testimony of expert map makers would have made it terribly power evidence that failing to get that fifth color was not negligent. It is quite impossible to make a substantial attack that argues the evidence should be ignored or that it does not offer considerable value to a finder of fact.

(I know I'm using as an example a theory that has, in fact, now been proven for all cases. I just thought it worked well in the argument to take it as not yet proven.)

[Michele]

I don’t think this is anything new. For years many people within our discipline have questioned the contradictory statements about our discipline being a science and the idea that our conclusions are absolute and conclusive. Science doesn’t offer absolute proof but the term ‘individualization’ seems to be a statement of absolute proof.

I’ve seen fingerprint examiners handle this in a variety of ways. Some hang on to both ideas and testify this way. These people validate the truth of these ideas by seeing that it’s worked in previous court cases. I personally think that it’s held up in court because we’ve had very few good challenges to it. It’s probably worth pointing out that it hasn’t always held up in court but that’s a different conversation.

I’ve seen others handle these conflicting ideas differently. These people understand that science doesn’t offer absolute conclusions. They testify with the word ‘individualize’ but they do so with the understanding that their conclusion is a deduction. It’s not proof, it’s the most reasonable conclusion and the accuracy depends on the principles used, the thoroughness of the practitioner, and the rigor of the review process. I believe that these people use the word ‘individualization’ not as a statement of proof but because they understand the scientific research behind friction skin formation and it leads them to believe that the likelihood of it being anything different is so remote that they can use this term. Is it a leap of faith? It could be, but it’s based on scientific research and it’s a standard leap in science. Scientists in every field make certain statements that may sound like fact but they do so with the understanding that certain things aren’t proven but they are accepted as being true.

I think the testimony in the Hull case is similar to this idea but worded differently, choosing not to use the word individualize because it may overstate the weight of the conclusion.

Although I agree with the idea behind the testimony, I do have some questions about the manner that it was stated, maybe this is the confusion or controversy that Glenn mentioned? After reading the Detail and the transcripts, I still have questions.

From this weeks Detail: “… we actually did not use the word "individualization" because it is clearly defined by SWGFAST as "determination that two friction ridge impressions originated from the same source to the EXCLUSION OF ALL OTHERS". We used the term "Identified" (which is still allowable by SWGFAST) and interpreted it in the following manner: … ”

Doesn’t SWGFAST also clearly define the word Identification to have the same meaning as Individualize, (not as interpreted in the testimony or this weeks Detail)? Were the people testifying unaware of this definition or was it intentionally not stated? One attorney did ask about it but they asked if it was defined on a certain document, they didn’t directly ask how SWGFAST defined individualization or if it was defined in another document.

This weeks Detail states, “We certainly had a distinct interpretation of these concepts…..” but the testimony left me with a different feeling. I felt like the testimony implied that this interpretation is accepted by SWGFAST. Can someone say that they follow SWGFAST but then go by their own definition, not the one SWGFAST provides? Don’t we criticize people like Cole, the Haber’s, and Starrs for twisting the truth for their own benefit? Is the above statement, along with several others in the testimony, doing the same thing? Is this accurate information or is it trying to pass certain ideas off as being acceptable in our disciple when they are not? Are they following SWGFAST? Do they believe they are following SWGFAST? Or are they testifying that they follow SWGFAST to gain credibility?

On another topic, is the CTS test mentioned in the testimony under review by CTS? We take these tests but I don’t recall getting any notification about these results being reassessed. Does anyone know how many agencies requested a review of these answers?

As I said above, I agree with the idea behind the testimony that individualization isn’t proven. I just question the justification presented in court. Do others think this is good testimony that we should try to emulate?

[Les Bush]

Wow it looks like we are bringing out the big hitters with this one, as I suspected we are tinkering with a fundamental. "Individualisation seems to be a statement of absolute proof", and "I dont think we can ever prove uniqueness". The beginnings of fingerprint identification science had the likes of Galton to assess whether it was reasonably scientific to utilise the markings of fingerprints as a reliable means of separating one human from another. This ofcourse centred on the use of ten prints and by association followed with latent prints. Both type of prints rely upon the same fundamental principles that define our science. Over the past hundred years the practice of examining both ten prints and latents has been more pragmatic than scientific but even so the amount of empirical data that is recorded is overwhelming in support of the scientific principles. Classic example of this application is the data involved in the Peter Griffiths case. We are a science but what we are not is a pure science such as mathematics that is why there is no probability model. Can we model our science, yes we can, through the combination of quality, quantity of detail and its spatial and sequential arrangement. The Mayfield case another classic that has been resolved by both exclusion and individualisation. Our science has substantially proven itself as reliable through the repetition of examinations, the variety of conditions both tenprints and latents and the distribution across the globe where it is applied. The evidence is there to be used. Since this is a forensic forum we all understand the value of a physical match where it is irrefutable(absolute) that the conditions under which two parts separated can only be restored to one configuration, a three dimensional problem solved without any scientific modelling or probability. I enjoy the exercise of visualising the anatomical three dimensional aspects of source skin and hypothesising if there is a proveable relationship between it and a two dimensional reproduction. The source skin is the third element in our conclusion of individualisation and one which should be at the forefront of our minds when explaining how the exemplar and latent form a match. Cheers once again from oz. Les

[Gerald Clough]

I have gone back and forth on the "Is it science?" issue. End the end, I think it well to remember that being "a science" doesn't require that it be a field in which everything must be known. Now it's true that no science has reached the limits of knowledge, but many have a lot of established knowledge in the bag. In other words, what's not known it considered to be out on the frontier, calling for very sophisticated study. We have a somewhat odd discipline in that regard. We began with some observations and some bits of research product and a number of frankly overconfident assumptions. And it developed in a substantially unchallenged environment. We've matured it beyond that position. And it's perhaps a measure of status that the challenges are reasoned and vigorous. We jumped from more or less scientific observation to fairly intense practical application in which a lot was accepted without particularly solid evidence. I think we're now making far more disciplined scientific inquiries, and it's refining our view of our own field. I really think we're just now exploring our frontiers.

What is gratifying, I think, is that most courts take a pragmatic view that there is much of value here and are not going to invalidate fingerprint evidence because we can't answer every question. They take the proper view of expertise as significantly helpful to the triers of fact.

You know, I've always questioned the idea that there are "pure" sciences. Mathematics creates constructs, largely to help explain the natural world. You can, of course, construct mathematics that have little apparent application, but we usually end up applying it or at least use it to further other mathematical tools that have more obvious applications. But it's the nature of the constructs that they must form a systematic structure. You get to make some guesses, but you have to explore them to the end of making them work in an orderly way. And I don't think it's "pure" by virtue of operating as abstracts or thought problems. That the arithmetics don't have to refer to a physical reality doesn't make them "pure."

And a great deal of practical application has been carried out by operating mathematical systems that, although orderly, don't really reflect the subtleties of the realities. They work within the application. You're okay, unless you and until you find that they don't work under some circumstance or at some level. When you find a situation in which your math doesn't work, you seek new math. And you do the same thing with applications. When you learn that some aspect of an application isn't supported by what you think you know, you qualify the application so that you don't cause an avoidable calamity through overconfidence. That's the nature of human affairs. You don't suspend your application just because you're pretty sure your model isn't quite what it might become.

I feel like we're making full use of applied knowledge when experts can debate conclusions. In that respect, I think we are lacking something. And that is the historical habits of a system in which our conclusions are not routinely examined and challenged. I think that when that's present, we grow and learn a good deal about our discipline and how we express things. There's some work being done in the direction of a more dynamic model of what is required for reliability in a given case. I think that's good. It means we're learning. And it means we are more and more moving into scientific inquiry conducted by experts. We should remember that new knowledge doesn't take a fast track into standards and guidelines. It's argued and re-examined and tested and even brought into court before it enters the catechism. And it's not often a pretty sight or very comfortable. Be glad you're in latent print examination. It's clean and neat and a league of gentlemen compared to such issues as Atkins mental retardation issues where the experts don't agree and where folks like James Flynn put their oars in the water and muddy it further.

[Les Bush]

I too agree that it is good when we are learning and anything newly learnt should be evaluated in how it affects the established knowledge and this is the purpose of peer review prior to publication. I'll keep this one short and a little off track from the actual thread. The comment that 'arithmetics dont have to refer to a physical reality doesnt make them "pure".' Yes our science does have to incorporate all the physical realities and variables of skin and its transfer. The arithmetic that avoids the reality test can be as simple as the following, 1+1=2, but conditions not included are that both entities must be infertile, or, both entities must remain whole during the calculation, such as if two sections of glass were being added and one became broken we still have the same two sections but the number of glass pieces has risen. Final comment is fingerprint science is a forensic science applied by experts who are aware of the conditions that vary examination outcomes. Cheers Les