BREAKING NEWz you can UzE...

NEW IAI PHOTO CONTEST  This year, the IAI invites you to submit photographs for entry into a photo contest to be held in conjunction with the poster presentations at the conference in Las Vegas.  The IAI has also put a section on the web page regarding guidelines for both the posters and the photos.  Content on the photos are open this year, and depending on response, categories may be implemented next year.  Dennis Honeycutt, coordinator for the session, says "this being the first time for this contest, we just wanted to see how it would go, therefore we left it open; we are looking for a wide variety of photos."  Up to two photos may be entered, and they must be a minimum size of 14 x 16 and maximum of 20 x 20, and must be mounted on foam board with no
borders.

I asked Dennis specifically about the content of the photos, since I have a creative mind and I am proficient in Adobe Photoshop.  He insists that any
photo will be taken this year, and would like to see a lot of photos in keeping with the spirit of the educational conference.  I plan on having a little bit of fun with one of my entries, but I also expect to see case photos, fingerprint comparisons, unique patterns, nature shots, and anything else you can think of.  Get in on the fun!!  Contact Dennis Honeycutt at: Dhoneycutt@mail.jus.state.nc.us.

It's BidNow Week!  As you know, clpex.com auctions off a unique fingerprint-related item on Ebay starting the first Monday of each month.  Congratulations to the lucky winner of the ACE fingerprint kit last month.  They stole it for under $5!!  My wife wasn't very happy because I paid more than that for it, but I explained it's all good fun, and gets you guys involved.  Up for auction this week is "Classification and Uses of Finger Prints" by none other than Sir Edward Richard Henry.  And as usual, all auctions start off at one penny with no "reserve."  I'm sure you won't let this one go for so little!!  :)

Daubert Workshop  A 3-day Daubert workshop is planned for June. The American Board of Forensic Document Examiners is hosting the workshop. It will have discipline specific sessions for both QD and LP examiners and general sessions combining both disciplines' general Daubert issues. Attendance is limited to 150, so get your reservations in and the word out quickly. Check out their announcement and registration form.

Simon Cole and Michael Lynch Receive 1 Year NSF Grant  The Cornell Chronicle reports that $144,000 has been awarded to the pair to conduct a year long comparison study on the science and sociology involved in DNA and fingerprints  Some interesting comments from Lynch are given in this article, such as "whether fingerprinting is science or not is beside the point.  The question is, is it good evidence?"  Perhaps the joint effort will temper the apparent stark differences of opinion between the researchers regarding fingerprint examination.

            Although recent court challenges to the scientific validity of fingerprint evidence have upset many examiners, there are a number of reasons why we should not panic. First, any examiner who has been in court knows that an aggressive defense attorney will attack the evidence no matter how good it is. Fingerprints are usually the most damning evidence, so a good defense attorney will focus the attack there – against you, the fingerprint examiner, and your evidence. This is nothing new.

            Second, over forty challenges have resulted in complete acceptance of fingerprint evidence in court. None have yet resulted in actual restrictions. A few of the challenges have been full pre-trial Daubert hearings, but most have been lesser challenges made during the trials. Some have even been made in Frye states where Daubert does not apply and where they have been inappropriately disguised as Frye challenges. Of all of the challenges reported, only one has resulted in a ruling potentially restricting fingerprint testimony. In a preliminary decision, Judge Pollak ruled that a fingerprint examiner may testify and present charted enlargements, but may not give a final conclusion. Judge Pollak held a full hearing on the matter last week, but has yet to issue a final decision. There were negative and incomplete newspaper stories, not to mention rumors flying around on the internet, but Ed German, who was there and gives a more accurate report on his website, remains cautiously optimistic. However, even if Judge Pollak allows his initial decision to stand, that single unfavorable finding would be but a blip on the screen when compared to the over forty favorable decisions.

            Third, we need to consider the expert witnesses used by defense attorneys to challenge fingerprint evidence. David Stoney (a respected scientist) has appeared in a couple of cases to testify to the lack of a statistical basis for fingerprint identification, but he has also testified to positive fingerprint identifications, which dilutes his effectiveness as a Daubert witness against fingerprints. Allan Bayle has testified as a Daubert defense witness only once. In spite of the fact that he testified that the FBI’s proficiency test is too easy, more significantly he testified that fingerprint identification is a valid science and if the correct methodology is followed (ACE-V), the conclusions will be reliable. Although a defense witness, he doubtless did us far more good than harm.

The two critical witnesses most often used by the defense are James Starrs and Simon Cole. Don’t let their claims or their publications fool you – neither Starrs nor Cole are scientists, and certainly neither is a fingerprint expert. Starrs is a law professor and Cole is a historian. Both are easy to discredit on the witness stand, as shown by the fact that no court has ever recognized either as a fingerprint expert.

            As an aside here, I might mention that I am listed three times in the index of Simon Cole’s book, Suspect Identities, and in all three cases he makes massive errors. He gets his definitions confused, his facts terribly awry, and even names of people completely wrong. And it is not only in the references to me that he gets things wrong – the whole book is chocked full of inaccuracies that are easy to expose. If his research is that shoddy, how can one place any confidence in his conclusions?

            But to return to the Daubert issues, one of the main points the defense attorneys are pressing is that of “error rate.” They claim that erroneous identifications occur and can readily be cited, and yet we have no calculated error rate. Therefore, they say, all identifications are suspect. However, according to Stephen Meagher (Section Chief, Latent Fingerprint Section, FBI) and Paul Sarmousakis (Assistant US Attorney in Philadelphia), and as we have seen in past Details, the US Supreme Court, in its decision in Daubert v Merrill Dow Pharmaceutical, intended consideration of “error rate” to apply to the error rate inherent in a science or the methodology itself, not the error rate of the individual practitioner. So when we testify that the error rate is “zero,” what we mean is that no two people ever have had or ever will have the same fingerprint and that ACE-V, when applied correctly, will consistently produce reliable conclusions.

            Not all fields of science can make that claim. Some tests and methodologies, although considered valid science, have significant error rates. Take, for example, tests for cancer. There are false positive and false negative results in many of those tests, regardless of the skill of the administering doctor. That is the type of error rate the Supreme Court intended for lower courts to consider in evaluating the reliability of a scientific methodology. Judge Joyner (Mitchell case in Philadelphia) clearly grasped the intent of the Supreme Court when he took notice that the error rate of the science and methodology of fingerprint identification is zero. Judge Joyner acknowledged that mistakes may be made by individual examiners and he pointed out to the defense that they always have the right to retain their own expert to review a case and testify to mistakes make by the prosecution’s experts. That is the proper way to deal with erroneous identifications – not to simply throw out all fingerprint evidence because occasional mistakes might be made by individual practitioners. For an analogy in a “live” debate with Simon Cole for the BBC Radio program “World Service” last summer, I pointed out to the moderator that if she were unable to balance her checkbook one month, it would not be due to a breakdown in the science of mathematics. Likewise, if a fingerprint examiner makes a mistake, that does not invalidate the science of fingerprint identification.

In Daubert defense arguments, however, there are enough germs of truth to merit our consideration. First of all, one of the main points the attorneys are pressing is that we lack a standard for identification. In a narrow sense that is true, at least to the degree that we lack a clearly articulated standard for identification. For most examiners, explaining the standard is sort of like trying to define “pornography.” I think it was Jesse Helms who said, “I may not be able to define it, but I know it when I see it.” For most fingerprint examiners, the same is true of a fingerprint identification. You may not be able to explain clearly the standard by which you make an identification, but you know an identification when you see one. The problem is not that we lack a standard, but that we do not yet have widespread agreement upon a concise, clearly worded statement.

The second germ of truth to the defense challenges has to do with the issue of testing and validation. While there is massive empirical proof of the validity of fingerprint evidence, we do not have a complete model on which to base statistical validation. Starting with Galton’s model, which was terribly flawed, and proceeding down to the most recent study, the FBI’s 50Kv50K study, all of the models have looked at “points,” almost exclusively just bifurcations and ending ridges. None have considered scars, creases, wrinkles, prominent sweat pores, incipient ridges, ridge shapes, or any of the myriad of other features that appear in prints. Whether you realize it consciously, your brain uses these features in any identification decision. To be accurate, then, a statistical model should include weighted values for these features. That would not be an easy model to design and to date no serious attempt has been made to include even scars or major creases, much less “Level 3” detail. A number of us were looking forward to the study proposed by the NIJ, but their withdrawal of the solicitation means more delay in the search for such a model.

There are some in our field who feel that such a study is a waste of time and money. After all, don’t we have enough evidence that fingerprints are permanent and unique? Why waste anything more on another study that can’t possibly add anything? Or so goes the argument. But the truth of the matter is that good science in any field is always being researched and advanced. More studies and better models can only help advance our science, as well.

A third germ of truth to the defense challenges is that some fingerprint examiners in the US receive inadequate training – not all, but some. Unfortunately, because there is no nationally mandated standard for training, some police agencies take the fastest and cheapest approach to selection and training of new fingerprint examiners. Even barbers have to go through mandated training and State licensing before they can cut hair. But fingerprint examiners with the power to present evidence in court resulting in long sentences, or even death sentences, face no such requirement. Other countries have nationally mandated training programs and competency testing for latent print examiners. In New Zealand and Australia, for example, fingerprint examiners go through five years of training and comprehensive testing before they begin case work and testimony on their own. In the UK, the training period can range from three to five years and is subject to several levels of review. But in the US, a “fingerprint expert” might have as little as 40 hours of training and no significant experience before his department assigns him to case work without oversight or review. The solution, of course, is not to throw out fingerprints altogether, but to mandate minimum training and experience requirements before a new employee examines cases without supervision. And the correct solution for the defense is to question the training and experience of the examiner in the courtroom, not challenge the science itself.

There is a fourth germ of truth in the defense arguments that many fingerprint experts are never tested for competency, nor are they checked for accuracy. The sad fact is that many police agencies still do not administer annual proficiency testing and have no measures in place for assuring that things are done correctly. The IAI and SWGFAST may recommend guidelines for the correct operation of a latent print section, but compliance with these guidelines is strictly voluntary. For some police agencies, unless they suffer a devastating civil lawsuit or at least a major embarrassment over sloppy work, they will continue to do things the way they always have, and that means inadequate training, no proficiency testing, and little or no quality assurance.

To sum up my assessment of these Daubert challenges, I might say that I feel strongly both ways. I believe we face no serious threat to the acceptance of fingerprint evidence in court. It is now and for over a century has been the most dynamic and absolute means of personal identification known to mankind. In spite the fact that defense attorneys claim fingerprint identifications are unreliable, the courts have consistently upheld the validity of what we do. But at the same time, defense attorneys are exposing some truths. We should continue to advance the science, not just rely on a century of success. We should improve training, quality assurance, and testing and validation more aggressively than we have in recent decades. Like other sciences, there is always room for improvement. We cannot rest on our laurels.

UPDATES on CLPEX.com this week...

Updated the CLPEX.com latent print examiner bookstore with sold and new books.

Changed the instructions for subscribing, un-subscribing, and changing your e-mail address regarding The Weekly Detail.  I recently received these instructions from the Topica team, who are always looking for ways to improve their customer service.  All someone now needs to do is send a blank e-mail to "theweeklydetail-subscribe@topica.com" (without the quotes) and a confirmation e-mail will be sent to the address from which that e-mail originated.  A simple reply to the confirmation e-mail finishes out the sign-up!  It's that easy.  Unsubscribe your old e-mail address by sending a blank e-mail (from your old account) to "theweeklydetail-unsubscribe@topica.com", follow through with the verification e-mail, and it's done!  To change e-mail addresses, simply do both steps, the first from your NEW address and the second from your OLD address.

The Detail Archives

Discuss This Issue

Subscribe to The Detail