What is enough.

[Neville]

Sir Francis Galton established that fingerprints are UNIQUE, persistant and classifiable as quote by Reznicek, Ruth and Schilens in the latest Journal of Forensic Identification. Dr de Puit in the same journal (vol. 60) in the conclusion of his article about transparency stated that in the future it cannot be acceptable to state that we are able to exclude all other possible sources, does this not appear to be a contradiction?

Are we to now believe that fingerprints are no longer unique? Sir Frances perhaps got it wrong? I hope not, I hope that what Dr de Puit is suggesting is that our identification may be false (maybe a lie) we are wrong or we are misguided some how. Now that I could live with, I am sure you could also.

I realise that I am putting my self out on a limb and I am probably cutting off the wrong end and yes I maybe using a chainsaw, but I just think something is going very wrong here. Surely uniqueness is one of the best thing we have, we are able to state without doubt that this is unique, I am unique, there will be and has never been another me, in the future I am sure that DNA will able to make the same unique identification.

Or maybe we are saying that there is not enough detail to be able to say this identification is unique, now that is some else and I can agree with this. So what is enough?

[David Johnson]

Are we to now believe that fingerprints are no longer unique?
Well fingerprints, as in the skin, may be unique but it is the ablity to say that a small distorted latent print was made by a certian individual and that no other person on the planet could have left a latent print that looks similar enough to fool an examiner is the part I think that is in question.

Sir Frances perhaps got it wrong?
He only used a very simple model to come up with a statistic that a particular pattern of minutiae would repeat. He was working with inked prints and thought that inked prints of all ten fingers could be reliably used to identify people.

Surely uniqueness is one of the best thing we have...
But we also dont need it either. DNA doesnt have it and they do just fine in the courtroom.

So what is enough?

Well if you want to say that I have individualized this latent print to the exclusion of all others on the planet then you better have a lot.

If you want to back off your absolute statement a bit then you probably dont need as much. That is where our training and experience come in. There are also tools in the works that maybe able to give us a number like DNA to support our conclusion that a particular person was the source of a latent print.

If you had even less then you could testify to an inconclusive.

It is the absolute statements that have been made in the past that are getting us into trouble.

[Neville]

Hi David

Thanks for the reply. I very much appreciate what you have said.

You said 'It is the absolute statements that have been made in the past that are getting us into trouble."
Is that really the case, I think it is more that experts are making idents that are of poor quality. Or lack sufficient knowledge to be giving evidence in the first place, though that is not the issue I am dealing with here. Was the FBI in trouble because they made an absolute statement or because the image quality was poor? Yes they did make an absolute statement but it would have made no difference if they had said there is a 100,000 or 10,000,000 to one chance of this being his print to the Spanish Police. The FBI later stated that it was because it was a poor quality print and that was the reason for the miss ID.

As for the Scotich debarcle it certainly appears to me to be a quality issue and once again, Yes they did make an absolute statement but it would have made no difference if they had said there is a 100,000 or 10,000,000 to one chance of this being her print.

'Sir Frances perhaps got it wrong?
He only used a very simple model to come up with a statistic that a particular pattern of minutiae would repeat. ' (Of 64 million to one).

Yes he did but what he conclude was correct and has been shown to be so by later statistical models ie, with 12 points of identity the ratio was found to be ten million million to one and it was stated at the time that this was also a simple model. So how many million million to one do we need? And how do we objectively decide which Minutiae are used in this new tool, as we all know AFIS is not good at selecting Minutiae. In Australasia we once agreed that 12 clear points or 16 poor quality points were sufficent. Many years ago.

[Charles Parker]

Neville---I think this quote from the recent FBI paper sums it up pretty good.

In practice, examiners do not compare friction ridge skin directly; they compare two-dimensional impressions made by the friction ridge skin. This limitation of the science introduces two additional considerations to the practice of friction ridge examination: whether an impression accurately transfers the unique features of the friction ridge skin and what amount of information must be present in an impression to achieve individuality.

The friction ridge skin is a three-dimensional, pliable surface. The information present on the friction ridge skin can be affected by, or lost during, the translation from three dimensions to two. In this translation, any number of factors, such as pressure or the amount of substance being transferred, may affect the quantity and/or quality of the information contained in the resulting impression. Despite the effects of these various factors on the appearance of friction ridge impressions, controlled recordings from the friction ridge skin have shown that the information contained in friction ridge impressions does translate reliably as a true and accurate representation of what appears on the friction ridge skin (FBI Laboratory manuscript in preparation). Even though examiners are trained to recognize the factors that may affect the translation of information (Ashbaugh 1999), further study into all of the issues affecting the quality of a latent print, and the effect of quality on sufficiency, is warranted.

[Gerald Clough]

I think you have to view Galton's perspective and use of language in the light of his goal and to what end he was proposing the use of fingerprints. "Unique" is not really bad language; it's just neither very precise nor is it in any way something that you can substantiate. Can you imagine any other way you could reasonably expect Galton to put it? If you follow the discussions here, it's easily apparent how much development has to go into a statement about why fingerprints are powerful identifiers. And even when you satisfy most of it by qualifying it in terms of discriminating among impressions, using only what can be discriminated by human observation, the threshold for discrimination remains undefined except as some vague community standard. We can hardly fault Galton for not developing the statement into some form of "sufficient" "uniqueness." ("Sufficient uniqueness" is roughly equivalent to "partially pregnant." ) It's not contradictory, then, to state that we cannot, based solely on human observation, exclude all other sources. That's just a fact, both logically and from the point of view of vague threshold, as well as inability to test validity.

Uniqueness has been perhaps the single most pernicious aspect of fingerprint identification, the enemy within. When claimed, it immediately attracts the criticism of anyone professionally dedicated to testable validity. And it has been the fatal flaw in testimony, when the court is called upon to strictly judge the validity of argument, rather than insert their own interpretation. Look at the judge's analysis of the decision in Maryland. You can't make an absolute statement and just assume it will be taken to say what you really mean. The judge wasn't goofy. The argument was flawed. I understand Galton keeping it simple. If you're trying to convey the concept of identification as a new concept, you can get terribly bogged down in terminology if you try to qualify everything to a nicety.

The apparent contradiction is not one that operates to the genuine detriment of identification. But it can be seen as such when the two realms of scientific certainty and expert opinion in law are not clearly delineated. Scientific certainty is not and never has been the standard in law. An expert may form an opinion with extremely high confidence and state that opinion as a conclusion without it being subject to a test of absolute certainty. Unless, of course, if it can be proven to a certainty to be incorrect. Fingerprint identification has one of the highest confidences among all expert opinion. It has that striking record, when impressed features are correctly interpreted and the generally accepted community standard applied, of never having been shown to misidentify. (A lot of if's in there, and a somewhat circular argument, but true nevertheless.)

You can never prove absolute uniqueness, either of a fingerprint or an entire organism. You could not do that even if you could actually observe every possible variable, because you can never compare to every organism that has ever been or ever will be. All you can do is say that what you can observe has been such that a great deal of observation suggests you can very confidently, but not absolutely, discriminate among individuals. Galton can be read as making the accurate statement that he had done a great deal of observation and that he believed that among a very large set of individuals, a fingerprint viewed to the interpretive limits of impressions and discriminated by observing some sufficient amount of detail, it would match one and only one individual in that large set. While not sustainable as a scientifically valid statement, it was functionally accurate in the realm where the concept was being applied. The second statement that identification cannot be factually absolute assumes an incomparably larger set of individuals than what Galton was talking about, and is provably correct. But it doesn't in any way change the functional meaning of Galton's statement. It does, however, draw the line between scientific certainty and high confidence opinion, which is what we always have to remember when we frame arguments. We've not done a very good job of making that distinction clear. Much of the reason we haven't fared worse in the realm where we operate is because judges have seen past our bumbling responses to the real issues in law.

As to the last question about what is enough. This is an excellent example of mixing the two realms. There is never enough to make a statement of absolute uniqueness. There very often is enough to form the expert opinion of who made the mark. How much is enough to form the opinion? It's whatever it takes for each individual expert to form an opinion. The only reason for resorting to an expert opinion is that expert judgment must be applied. We don't need an astronomer to testify to the phase of the moon. A reputable almanac can do that. (Although I'll bet you could nit pick that to death, too, since it's never "full"; there's always a bit in shadow.) I think we have done a rather poor job of training examiners in that distinction and that our lack has led many to make insupportable factual statements, because they do not understand what their conclusion means and think they have to defend an absolute factual certainty that doesn't exist.

[16_Points]

Sir Francis Galton established that fingerprints are UNIQUE, persistant and classifiable as quote by Reznicek, Ruth and Schilens in the latest Journal of Forensic Identification.

Are we to now believe that fingerprints are no longer unique? Sir Frances perhaps got it wrong? I hope not, I hope that what Dr de Puit is suggesting is that our identification may be false (maybe a lie) we are wrong or we are misguided some how. Now that I could live with, I am sure you could also.

I realise that I am putting my self out on a limb and I am probably cutting off the wrong end and yes I maybe using a chainsaw, but I just think something is going very wrong here. Surely uniqueness is one of the best thing we have, we are able to state without doubt that this is unique, I am unique, there will be and has never been another me, in the future I am sure that DNA will able to make the same unique identification.

Or maybe we are saying that there is not enough detail to be able to say this identification is unique, now that is some else and I can agree with this. So what is enough?

UNIQUENESS is dead . . . (that's from me not Nietzsche). Bury it, mourn it, move on. Why you ask? Because uniqueness is irrelevant to the utility of friction ridge identification. Uniqueness does not imply sufficiency. If it did, then all friction ridge skin impressions would be sufficient for identification. Clearly this is not the case. Furthermore, the concept of a friction ridge impression "containing sufficient uniqueness to individualize" is simply absurd. It is unique or it is not. As such, I do not see what the concept of uniqueness contributes to the identification process. Rather the crux of successful identification is sufficiency, and uniqueness simply cannot help us answer questions of sufficency. It's dead weight; get rid of it.

I would further propose the alternate proposition that a friction ridge impression can be sufficient without being unique. What's the difference? One can no longer profess to the court that the ID is 100% certain with a zero error rate. For some reason, people don't want to give this up.

[kevin]

uniqueness is irrelevant to the utility of friction ridge identification

I tend to think it is kind of important, again, unless you are performing an examination in some parts of Scotland in which case it is required to be neither unique, nor similar.

(doesn't this come down to semantics and how its presented???)

[16_Points]

uniqueness is irrelevant to the utility of friction ridge identification

I tend to think it is kind of important, again, unless you are performing an examination in some parts of Scotland in which case it is required to be neither unique, nor similar.

(doesn't this come down to semantics and how its presented???)

10-47 (negative). What does uniqueness do? It sounds nice but doesn't solve any issues. Friction ridge skin doesn't need to be unique. It just needs to show enough variability in friction ridge arrangements to be useful for identification (like DNA).

[Pat A. Wertheim]

Dear Kevin & 16_Points,

Two definitions come into play, and failure to carefully distinguish either of them dooms us to an inability to resolve the question.

First definition: FINGERPRINT. We must differentiate between the friction ridge skin and the impression it leaves behind. The skin is unique, if for no other reason that at the elemental level its cells contain DNA molecules. While the uniqueness of that skin (DNA aside) is crucial to our science, the truth of the matter is that we do not compare skin. (Spare me the rare exception when we actually do compare skin.) We compare impression to impression. Let's call the skin "skin," the latent impression a "mark," and the inked impression a "print." Just for the sake of being arbitrary, the agreement to use these terms to differentiate between the three things simplifies the discussion.

And that leads me to the second definition: UNIQUE. The argument into whether a "fingerprint" is unique misses the point. I think we can all accept that at the cellular level, skin is unique. The issue we should be addressing is whether the uniqueness in a particular mark or print is discernable or detectable, and whether it can be successfully compared to identify or exclude. To argue whether the mark or print is "unique" misses the point. A worthless smear (mark) with no ridge detail is "unique" in that it could not have been deposited in its exact detail except by the exact arrangement of skin cells with the exact composition of residue on the exact surface where it is deposited with the exact pressure and direction of touch and slippage that occurred. But in spite of the uniqueness of a mark, our ability to identify or exclude that mark depends on whether the uniqueness in it is discernable and comparable, not whether it exists.

Distinguishing between skin, mark, and print aids the discussion. Getting past the idea of uniqueness to a determination of whether that uniqueness can be discerned and compared to a level of sufficiency between a mark and a print is what allows us to make an identification.

[kevin]

Getting past the idea of uniqueness to a determination of whether that uniqueness can be discerned and compared to a level of sufficiency between a mark and a print is what allows us to make an identification.

Thanks Pat - I couldn't have said it better....the old forest through the trees.

The whole modus ponens (in a roundabout fashion) of fingerprints is that we compare fingerprints to make an identification (et al) and thereby establish an identity to go along with a mark. We can establish identity because prints are identifiable to an individual. Like I said before, uniqueness and how it applies to our presentation in the context of the legal system v scientific fact v practical application is always going to be a sematic arguement that evolves with the legal system and the science itself. If words like 'unique' become some kind of taboo buzz word because it doesn't conform to a narrow definition we are applying to an adjacent science (DNA) then we might as well close up the whole shop and and concede two prints can be made by the same individual. I think it has always been the work of trained, dedicated examiners that keeps us out of that abyss.

Somebody post that picture of the short bus again - I need a lift out of here!!

[Gerald Clough]

Aside from those who cling to the illusion of the existence of anything that is provably unique, among ourselves we understand what we mean by "uniqueness." It's a semantically convenient way of talking about the quantity of observable features that contribute to discriminating among prospective sources. But the term is loaded by historical false, certainly unprovable, presumption that any impression sufficiently detailed to identify was truly unique. And the discipline does intersect science, where the "unique" has a specific meaning and where "uniqueness" is a meaningless. And it more than intersects law. Law is the reason for being. We tend to shy away from saying what we really mean, because if we frankly state is has to do with rarity, likelihood, degree of confidence, or probability, we feel we are vulnerable on account of being unable to quantify those qualities on demand.

The legal landscape has changed to one in which conclusions are attacked with science (or lack of), which means if we let the discipline be drawn into a test of scientific validity, we leave it up to the court to independently apply insight to interpret poor representations of the discipline with favorable opinions. But you can't expect the court to do that every time. Eventually, the courts are likely to either believe we really mean what we keep saying or that, if we can't figure out how to say it well, maybe we shouldn't be saying anything. It's all too easy to carry internal vernacular over into testimony, and the cross-implicating terms, unique, uniqueness, and 100% certainty, make it hard to focus the presentation where it belongs, on general reliability, subject to adversarial argument, rather than pretense, real or perceived, to absolute certainty that can only be justified by proof, which does not exist.

I think we have to think about two things. One is constructing a rational and straightforward representation of why latent print examination produces reliable and useful information to legal fact-finders. That does not require resort to much science. The biological knowledge is interesting, but it doesn't contribute to a reliability argument, except as some explanation that's consistent. What matters is that, when there is known error, it is detectable and explainable. Instead of something like Mayfield being ammunition for an attack on reliability, we should think about making revealed error part of our argument. Latent print examiners discovered the error. Not attorneys, not circumstances. Regardless of the fact that latent print examiners made the error, latent print examiners revealed it. A defense examiner could have revealed it and demonstrated it to the satisfaction of any lay person, had it reached trial. No undetectable error made through applying accepted process and standards of judgment has been found, and where there's a difference of opinion, the matter can be placed before a fact-finder by demonstration. Adversarial expertise is applicable. That's what courts properly demand of expert evidence. In short, we have to keep the issue one of plain demonstrated reliability and be prepared to make the distinction between scientific certainty and appropriate reliability.

The other thing is to avoid, so far as is possible, allowing the argument to be drawn into science through the use of terminology that has its most precise definition found in science and logic. It's all too easy to allow convenient internal vernacular into testimony. If we're going to use "unique" and related terms in our own discussions, we should always be mindful that we are using them strictly with the understanding that they are not taken for their precise meanings.

I want to again point out that we can more effectively represent and defend the discipline if we don't allow improper argument to succeed by default. I think it's pretty clear that much of the criticism grows out of the fact that so much latent print evidence goes through the trial courts with no opposition and no critical independent review. We should not allow the discipline to be attacked on account of legal system faults. We should not be held to a test of scientific certainty simply because the defending litigators don't get expert review. If we allow the official examiners to be placed in the position of the ultimate and absolute authority as a matter of legal practice, we are placing them in a position that calls for a high standard of scientific validation. If you want to criticize the attorneys for not seeking expert help or the courts if they don't allow it, that's one thing, especially in a death penalty case where there is a consistent acknowledgment in opinions that the standard of evidence is higher. But we should not passively allow such a standard to be imposed to fix a problem we aren't responsible for.

[kevin]

I'm not arguing that FR identification does not intersect the law or that is does not intersect science, only that it does not intersect law and science in the same manner that DNA does, nor should it.

Terms like 'to exclusion of all others' , 'to all others on earth' -okay, that is a bit of a misleading statement and wholly unprovable but it was out there for a time and it is in the legal playbook so to speak. I for one have had that term thrown at me in classes but I do not use it in favor of saying 'one in the same source' or something along those lines. Either way I have to be prepared to answer why two different people do not have the same fingerprints or I will probably be out of a job - they are unique to each individual.

However, 'unique' is a term unto itself that i believe lends itself very well to FR analysis. Do I take it to mean what DNA terms it? Of course not. Should a jury member constantly have in the back of his or her mind, WWDNAD (What would DNA do!) when weiging my testimony on a FP analysis?? The definition of scientific fact and legal fact are often two separate paths (as we have discussed on this board before). I wholeheartedly agree with that arguement but I do not think we can't just apply it piecemeal to certain buzz words of one's testimony solely for the appeasement of another science.

I do not argue that the forensic field has evolved in such a way that we have to rethink some of the presentation and terminology but I find myself in the same position as Galton - what other word would I want to use besides 'unique' when telling a jury no two people's fingerprints are alike when it fits the bill (and then explaing that, no, I haven't compared everyone on earth's fingerprints)?? If I simply use another phrase is that still not the same thing? I'm certainly more comfortable qualifying a term like unique in this manner rather than saying; fingerprints aren't unique and do not have to be for us to make an identification. Who is to say a judge will not throw that arguement out on its ear faster than failing to adequately 'explain' an overstatement like 'to the exclusion of all others'? what are we basing our arguement on without uniqueness and permanence?

So how are we using the term unique?? As a FP community or a DNA community?? I remember a similar issue arising when looking at the definition of subjective and objective in legal context??

Webster gives a 'unique' a broader definition as well but it notes their commentors appears to be in somewhat of the same pickle as the FP community....

http://www.merriam-webster.com/netdict/unique

Many commentators have objected to the comparison or modification (as by somewhat or very) of unique, often asserting that a thing is either unique or it is not. Objections are based chiefly on the assumption that unique has but a single absolute sense, an assumption contradicted by information readily available in a dictionary.

Gerald - I'm afraid at this point I have alot more questions than answers on this topic but I always enjoy your responses and your perspective!

[Neville]

Very good Pat I like it.

Just one issue I would like to raise at this point. Did Galton use inked impressions or did he make skin to skin comparisons, I was under the impression he used inked impressions, therefore his use of the word unique becomes interesting. I would have to say his choice of this word is significant; he was a scientist with a use of the English language of his time. My oldest Dictionary is an Oxford reprinted in 1960 so it is 70 odd years out of date, but in it there is no other meaning to the word implied 'being the only one of its kind; having no like or equal or parallel.' However 15 years later the Webster dictionary adds the meaning 'unusual' a slight shift in the meaning in just 15 years not to mention a cultural shift. I am very sure that Galton knew exactly what he was saying. I am not trying to be a linguist but I do not for a moment believe the Galton's use of the word was an oversight or mistake.

It maybe that this word is a dead word in the fingerprint world of some but it is still there in our foundation documentation, I am raising this topic mainly to get some discussion going on it before we as a discipline go racing off into the future without appreciating the past, much of the future can be viewed by first appreciating the past.

Are we just creating little boxes in to which we place peoples views as it does not fit our belief, perhaps, I just think it would be a mistake to dismiss Galton's intellect.

Amazing Kevin just beat me to the punch, You obviously use more than one finger to type with. You have said what I should have.

[Gerald Clough]

Gerald - I'm afraid at this point I have alot more questions than answers on this topic but I always enjoy your responses and your perspective!

I don't know that there's an entirely satisfactory answer. I think we have two kinds of critics (in court and in public), with regard to "unique." One kind simply doesn't understand what we mean and is offended by what they perceive as an untruth. The other kind knows what we mean but either sees in it a potential vulnerability or just objects on pedantic grounds. And, of course, there are still examiners who mean actual factual exclusion of every individual who has existed or ever will exist. But common usage does count, as in the alternate dictionary meanings. I guess we just have to be prepared to explain it when it's attacked through the narrow meaning, and brief the attorneys who sponsor us as witnesses that it's an issue and that we have an answer. Moreover, we can turn an attempt to attack through the narrow view back on the attacker with a simple explanation of legitimate meaning, just as I think we can turn revealed error to our side in support of the idea that, if every examination result can't be presumed reliable, one can rely on the ability to detect error and argue it.

Don't argue from the other guy's interpretation, in either issue. I think that in many forensic issues, litigators depend on others to develop the objections and responses to objections, generally acquiring their arguments from tutorials and examples in other briefs. They tend to run on rails. If the rails are on the wrong track, we should be able to put things back where they belong. A very large part of expertise is how well you can testify. Not just explaining your findings, but also how well you can maintain the focus. I see some excellent analysts in a number of fields who are practically useless because, while they can nicely explain what they did and what the concluded and are unquestionably expert, the can't handle cross and recross.

[Neville]

Hi Gerald

Your second paragraph I totally agree with, I recall one expert witness giving DNA evidence who was a Doctor an Expert in their field, no doubt, but also a total moron in the box, the judge was asleep with his eyes open and both solicitors were struggling to stifle yawns. Then on the other side of the equation an expert in digital enhancement (V++) who we were told was going to be shredded in the witness box by defence, so half the fingerprint section turned up to watch including some UB inspectors, this guy nailed his brief and some, there was no wiggle room, I am sure the defence solicitors face changed colour while he listened, he stood up at the end and said no questions.

My old boss told me once that the Crown Prosecutor said the best expert witness he had ever heard in court was a particular fingerprint expert witness, this guy was brilliant, but no one knew that he used the witness box as practice for the stage he was acting the whole time. He told me it was the best place to practice "the loneliest stage in the world". Another very good expert witness (Handwriting/document) was very big in Toastmasters. So as you suggest it isn't so much the content as the delivery, we (meaning the public) are so easily sucked in by big noters, just look at politics.

Gerald, can you clear up one small issue please a yes or no answer will suffice. If you had every set of fingerprints of every one living today would you expect to find a small area of ridged skin repeated on another person (say 1/2 an inch square)?

I see no point in including every person who has or has not yet lived as they are less likely to leave a fingerprint at a scene of crime today.