CLPEX Chat Board

Posted: **Mon Oct 29, 2007 8:44 am**

Two steps forward, one step back.

Judge bars use of partial prints in murder trial

Bryan Keith Rose (pictured) faces a possible death sentence if convicted in the Jan. 5, 2006, death of Warren T. Fleming. (Handout photo / January 10, 2006)

By Jennifer McMenamin | Sun Reporter
October 23, 2007

A Baltimore County judge has ruled that fingerprint evidence, a mainstay of forensics for nearly a century, is not reliable enough to be used against a homicide defendant facing a possible death sentence - a finding that national experts described yesterday as unprecedented and potentially far-reaching.

Baltimore County Circuit Judge Susan M. Souder's order bars prosecutors from using at trial the partial fingerprints lifted from the Mercedes of a Security Square Mall merchant who was fatally shot last year during an attempted carjacking at the shopping center. Prosecutors say the fingerprints - as well as those found in a stolen Dodge Intrepid in which witnesses said the shooter fled the mall parking lot - link a 23-year-old Baltimore man to the killing.

In her ruling, Souder outlined the long history of fingerprinting as a crime-solving tool but says that such history "does not by itself support the decision to admit it." In explaining her reasoning in a 32-page decision, the judge leaned heavily on the case of an Oregon lawyer mistakenly linked through fingerprint analysis to the 2004 Madrid train bombings.

Related links
Warren T. Fleming Photo

With defendant Bryan Keith Rose scheduled to go to trial today in Towson, prosecutors and defense attorneys in the capital case declined to comment yesterday on the judge's ruling.

But others who have researched the issue and litigated cases involving fingerprint evidence said the decision - if it stands up on appeal - could have implications that reach even beyond the use of fingerprint evidence in criminal courts.

"The repercussions are terrifically broad," said David L. Faigman, a professor at the University of California's Hastings College of the Law and an editor of Modern Scientific Evidence: The Law and Science of Expert Testimony.

"Fingerprints, before DNA, were always considered the gold standard of forensic science, and it's turning out that there's a lot more tin in that field than gold," he said. "The public needs to understand that. This judge is declaring, not to mix my metaphors, that the emperor has no clothes.

"There is a lot of forensic science that is considered second to fingerprinting," Faigman added, mentioning firearms and toolmark analysis, hair identification, bite pattern analysis and evidence used in arson investigations as examples. "If fingerprinting turns out to not be so good, people could start questioning that science as well."

The technology has come under scrutiny in recent years.

Stephan Cowans, a Boston man who spent six years in prison for the shooting of a police sergeant, was released in 2004 after the discovery that the fingerprint used to convict him was not his.

That same year, the FBI mistakenly linked Brandon Mayfield, an Oregon lawyer, to a fingerprint lifted off a plastic bag of explosive detonators found in Madrid after commuter train bombings there killed 191 people. Two weeks after Mayfield's arrest, Spanish investigators traced the fingerprint to an Algerian man.

The U.S. Justice Department issued a formal apology last year to Mayfield and awarded him $2 million.

Souder, the Baltimore County judge, referred repeatedly in her opinion to that case, as well as a March 2006 report from the Justice Department's internal investigators on the FBI's handling of the matter.

In the Mayfield case, three FBI fingerprint examiners and an independent court-appointed fingerprint analyst determined that the fingerprint on the bag of detonators belonged to the Oregon attorney.

"Up to that point, [the government] had maintained that if you have a competent examiner, the technique of fingerprinting can't produce a misidentification. Mayfield exposed that as a fallacy," said Robert Epstein, an assistant federal defender in Philadelphia who in 1998 was among the first lawyers to challenge the reliability of latent fingerprint identifications.

In the Baltimore County murder case, defense attorneys challenged the admissibility of fingerprint evidence that linked Rose to the killing Jan. 5, 2006, of Warren T. Fleming, the owner of a Cingular Wireless store at Security Square Mall.

Rose was arrested 13 days after the shooting after police received a call saying the "ringleader" of the attempted carjacking was a man called "Sticky," a nickname that was recognized as Rose's, according to court records.

At a pretrial hearing in May, prosecutors argued that fingerprint evidence has been accepted by the courts and relied upon for nearly 100 years. Defense attorneys countered that there is no similar history of subjecting the evidence to scientific review.

"The state is correct that fingerprint evidence has been used in criminal cases for almost a century," Souder, the judge, wrote in her decision. "While that fact is worthy of consideration, it does not prove reliability. For many centuries, perhaps for millennia, humans thought that the earth was flat."

She criticized the common method of fingerprinting as overly subjective and lacking in standards. She discounted the proficiency tests that the state's expert witness testified about. And she characterized as "neither credible nor persuasive" testimony that fingerprinting is an infallible methodology.

Souder acknowledged that the crime lab technicians' conclusions that Rose's fingerprints match those found on the cars "appear to be the heart of the state's case."

Issued late Friday afternoon, the judge's decision has already attracted significant attention within the world of forensic sciences. The decision was included in yesterday's edition of The Detail, an e-mail newsletter distributed by fingerprint examiners.

jennifer.mcmenamin @baltsun.com

More articles

Copyright © 2007, The Baltimore Sun

http://www.baltimoresun.com/news/local/ ... 0011.story

Posted: **Mon Oct 29, 2007 8:45 am**

Oh, and here's her statement.

http://www.baltimoresun.com/news/local/ ... 58.acrobat

And the original thread from Red.Sox.Fan.

http://clpex.com/phpBB/viewtopic.php?t=735

Posted: **Tue Oct 30, 2007 11:11 am**

Posted: **Tue Oct 30, 2007 2:03 pm**

sharon cook wrote:When Meagher says that errors made by human beings are not a reflection of the reliability of the methodology, she says he's "not credible." If I add 2 and 2 and get 5, does that destroy the entire science of mathematics?

I've criticized this analogy here before. That might apply to analysis of very clear latents to very clear exemplars, but that's not the question. Here the problem is more akin to having a group of accountants trying to add up figures in a ledger written in water soluble ink that's been sitting in the rain for a few days. As your texts say, every known impression is slighty different. The examiner has to make judgment calls about whether a difference is explicable, and doesn't preclude a match, or is inexplicable and indicates a non-match (I'm not going to get into the discrepancy/dissimilarity jargon debate.) Once one has to make judgment calls about blurry, sometimes ambiguous data, then the human being is part of the methology. You can't seperate the two out as cleanly as FBI wants to.

If the community said that there is a potential for human error, but there are very few known cases of error by properly trained personnel. There are likely some additional undiscovered cases of error, but no reason to believe this is a large number. Thus, the error rate given the huge numbers of identifications made every day can be confidently said to be very, very low. Far less than 1%. Probably less than one in a couple million. (I think Pat had some off-the-cuff guestimates at the Chesepeake Bay IAI talk.)

Then you can go on to talk about why no error occured in this specific case. You'd likely gain credibility with the courts and jurors, without ceding any useful ground to defense counsel.

Posted: **Tue Oct 30, 2007 6:02 pm**

Hello Ms. Steele,
A few points:

I don’t think that there is a fingerprint examiner out there who would mind stating in court that the error rate for fingerprint identification is extremely low (but not zero). I will give you that. However, I still have nothing to base that statement on. Just as I don’t have any “scientific” evidence that will allow me to claim a zero error rate, the fact is I don’t have any “scientific” evidence that will allow me to quantify an error rate in any amount. And probably never will. Unlike some technically sophisticated testing processes that use equipment manufactured to an industry or regulated standard, all of the comparative analyses involved in fingerprint examination take place in the brain of the examiner. This brain function does not necessarily inject subjectivity. Studies have shown that people trained to conduct comparative analyses actually “think” or process data differently than others. But, to date, no one has figured out how to calculate an error rate based on the function of the human brain. This is why we have separated ACE-V methodology from the examiner. Because ACE-V describes a methodology but does not actually process data (it is not a mathematical formula applied to a data set), it can be said to have a theoretical error rate of zero. Examiners, however, as a group, can not make that same claim. I believe this is what Mr. Meagher means when he testifies that the methodology has a zero error rate. However, since the individualization of an unknown fingerprint impression to a known source can not be accomplished without the methodology and the processor (examiner) working together, it is probably fairer to assign an error rate to the practice as a whole. Maybe all you lawyers can get together at some convention of the ABA and issue a policy statement that says as long as fingerprint experts testify to an error rate for the individualization of fingerprints of something more than zero, they get a pass on that issue. You may even help us out by suggesting a value or two that we can choose from.

Regarding your comment which addresses the quality or clarity issue, again if ACE-V is properly followed then the latent will have been assessed for its individualizing characteristics before the suspect print is ever seen. In this way, there is no bias toward any known source. Now, might the examiner change his opinion about some of these characteristics once he sees the known exemplar? Of course. The same way you may be stumped by a clue in a crossword puzzle, the actual word becomes clear as the adjoining blocks are filled in. It may be that the word you first thought was called for was right or maybe it was wrong. But it is legitimate to change a conclusion when the change is based on new data which was not part of the original analysis. Sort of like getting a new trial when evidence unavailable at the first trial is revealed.

Perhaps another thing you lawyers could agree on at that ABA funfest: since most AFIS systems require that latents entered meet a minimum quality/quantity standard (certainly way above your “rained on ink” analogy) perhaps you’ll grant us another pass when individualizations are made to a known source suggested by an AFIS, which so far as I know, has been shown to be uninfluenced by the biases that afflict us mere humans. Maybe you could agree to stipulate to an individualization if your client showed up in the top ten of the AFIS candidate list and eliminate the corrupting influence of human examiners altogether. I mean, statistically and therefore scientifically, what are the chances that not only would AFIS pick your client as the source of a completely chance impression, but your client would have had to have been where the crime was committed and there at the time the crime was committed. Lump in any other evidence and the odds are staggering. Scientifically speaking, of course. It would sure help with our overtime budgets.

Posted: **Wed Oct 31, 2007 8:38 am**

Dogma wrote:I don’t think that there is a fingerprint examiner out there who would mind stating in court that the error rate for fingerprint identification is extremely low (but not zero).

I think you'll get some disagreement on that on this board alone.

Dogma wrote:Just as I don’t have any “scientific” evidence that will allow me to claim a zero error rate, the fact is I don’t have any “scientific” evidence that will allow me to quantify an error rate in any amount. And probably never will).

I recall an exchange with Pat W. in this forum a few months back about a theoretical framework for a test of all IAI certified examiners. I'd agree there are significant issues in coming up with a test that the examiner community and the critics would feel confident about.

I recall some promising research on analyzing DNA from the fatty acids left in the print itself. If that panned out, we might be able to find an error rate by testing a sufficiently large sample size of fingerprint matches against DNA matches from the latent.

Dogma wrote:This brain function does not necessarily inject subjectivity. Studies have shown that people trained to conduct comparative analyses actually “think” or process data differently than others.

I've seen some of those presentations as well. But I think Busey and the other folks involved would not support an assertion that any difference in recognition changes the extent to which the examiner's judgment is subject to subjective and subconcious influences. As I understand it, Busey was arguing that the examiner's brain uses the same mechanism for fingerprint analysis that it uses for facial recognition. (As seen from rotated images and measuring the N170 gap). We know there's problems with facial recognition -- the DNA exoneration cases are full of mis-IDs. (Yes, the analogy between prints and eyewitness ID has its own problems...)

Dogma wrote:This is why we have separated ACE-V methodology from the examiner. Because ACE-V describes a methodology but does not actually process data (it is not a mathematical formula applied to a data set), it can be said to have a theoretical error rate of zero.

And the problem I see is circular. Any mistake by an examiner, no matter how well trained, is explained as an error by the individual, not by the method. To exaggerate, one could claim that astrology or phrenology (personality analysis by the bumps on one's head) have a method error rate of zero -- any mistakes are solely the fault of people not applying the method correctly. Here, the individual examiner's judgment is a fundamental part of the method.

Let's got to Dror's test. We have five examiners -- each testified to a match in court. Assume they were adequately trained and applied ACE-V to that. Dror gives them the same data (same exemplar and latent), but with biasing information leading them to believe they are looking at the Mayfield mis-ID. Four of five change their minds. Presumably they didn't follow ACE-V the second time, but let's assume that instead they were being told they were being hired to testify as an expert in a lawsuit by Mayfield and needed to render an opinion they'd testify to in court. They are given the materials and told to treat it like normal coursework and apply ACE-V. Assuming that we got the same result -- four of five changed their minds about the match -- is that a failure of the method or of the practitioner?

Dogma wrote:Maybe all you lawyers can get together at some convention of the ABA and issue a policy statement that says as long as fingerprint experts testify to an error rate for the individualization of fingerprints of something more than zero, they get a pass on that issue.

Sadly, there isn't any such body that could herd cats and get lawyers on the same page. The best you might get is a position taken in a specific case in an organization's amicus brief.

I think, however, that if one explained that the rate for trained practitioner using ACE-V is not zero (some mistakes have happened) and explained that there is a very low number of known mistakes compared to the large number of identifications made daily, that the rate, while unknown and unknowable, is acceptably low, that would statisfy the courts and defang the criticism that "zero" can't be right.

Dogma wrote:Regarding your comment which addresses the quality or clarity issue, again if ACE-V is properly followed then the latent will have been assessed for its individualizing characteristics before the suspect print is ever seen. In this way, there is no bias toward any known source.

The problem here is that, as I understand ACE-V, there's no caution, no warning, that the examiner should try to avoid any baising information about the source of the exemplar. Nor, generally, does the examiner document his or her conclusions prior to seeing the exemplar so that if he or she changes an opinion about a feature, there's some way to know that the change occured.

Dogma wrote:Perhaps another thing you lawyers could agree on at that ABA funfest: since most AFIS systems require that latents entered meet a minimum quality/quantity standard (certainly way above your “rained on ink” analogy) perhaps you’ll grant us another pass when individualizations are made to a known source suggested by an AFIS.

Wasn't Mayfield suggested by AFIS? As I gather from this forum, there are huge issues with AFIS coming up with good candidates who examiners conclude aren't actual matches. The false positive rate for AFIS matches is way too large, and I gather the false negative rate is also unacceptable.

Dogma wrote:I mean, statistically and therefore scientifically, what are the chances that not only would AFIS pick your client as the source of a completely chance impression, but your client would have had to have been where the crime was committed and there at the time the crime was committed. Lump in any other evidence and the odds are staggering. Scientifically speaking, of course. It would sure help with our overtime budgets.

Again, ask Mayfield. Tho I suspect if the match had belonged to a middle-aged Soccer Mom in Kansas who was in the database from prior gov't service in a clerical position, the FBI examiners would not have proceeded to call it a match (which might miss some Soccer Moms with shady pasts). Other independently analyzed evidence is good -- and likely means an attorney is going to plea the case, not make a challenge (which should help with those budgets) -- the key there is that the conclusions need to be independent. If fingerprint lab guy has lunch with DNA lab guy and firearms ID lab guy, they get to talking about the same crime, and one of the guys tells the others they have a match to a suspect, then the other two's conclusions have a confirmation bais issue.

Posted: **Thu Nov 01, 2007 9:27 am**

Posted: **Thu Nov 01, 2007 8:01 pm**

This is, hopefully, an isolated incidence in which a mis-informed and poorly educated or highly biased judge has attempted, however well-meaning, to place any and all comparison evidence in jeopardy. If latent print comparisons, long proven to be a highly accurate and trusted form of forensic evidence, are eliminated from the courtroom, then the logical consequences are to also eliminate any and all subjective evidence in every criminal and civil trial. Fingerprints are unreliable, then so must ballistics, cartridge casings, bullets, shoeprints, tire prints, blood spatter, toolmarks, questioned documents, drug analysis, video surveillance, eye-witnesses, DNA and virtually every other form of evidence previously admitted in silent witness of criminal activity be banned from consideration in courts of law. This is a potential field day for the defense attorneys of the world. What is left now to the prosecution to defend the victims and the public of these horrible crimes? Absolutely nothing but claims and counter claims from defendent and suspect.

It is virtually impossible to provide error rates for comparison evidence. Sure, we can create tests of known samples, created by committees using known examples. But we cannot ever know for sure when a latent print comparison in actual case work is truly a positive individualization. Education, training and vast amounts of experience, coupled with absolute integrity and honesty on the part of the examiner is the only solution, followed by objective and independent verification. If this fails, there is nothing left. All we can do is provide honest opinion based on scientific underpinnings....rules, alas, by which the defense need not abide.

Pushing this yet farther, perhaps it would be in the best interests of the public to demand an accounting of error rate from the prosecution, defense attorneys and judges themselves before allowing them into the courtroom. How many innocent persons have been convicted by Division III court under judge Bias? How many criminals have been excused due not to lack of evidence but due to legal technicalities irrelevent to the evidence itself? Can I, as an accused citizen, demand to be tried only in courts with a less than 1% error rate? How will we determine that error rate?

The answer, as with forensic examiners, is to trust the best and eliminate the rest. But who are the best, and who are the rest? If every examiner is required to forfeit his/her job upon a known, proven false individualization reported, perhaps that knowledge would lower the bar on pushing those borderline idents into evidence. The vast majority of idents are not borderline, yet we see the entire discipline being unreasonably attacked. Ah, that such standards could apply to the legal profession .....

Posted: **Fri Nov 02, 2007 11:16 am**

Posted: **Sat Nov 03, 2007 5:47 am**

sharon cook wrote:Analyze the DNA from latent print residue to find the error rate of fingerprint matches? Only a lawyer could come up with something so convoluted and unworkable.

I understand some of the difficulties, but done as a study it would provide a way to cross-check the print match itself, assuming it were done in a blind manner (DNA folks excluded from print folks results and vice-versa). Sure, there's going to be some inconclusives on both sides -- doesn't validate or invalidate either's results.

sharon cook wrote:Don't you think the evaluation of DNA results is a subjective process? You bet it is! And so is all other expert opinion testimony. And not a single other forensic science is being browbeaten about error rates like latent print analysis is.

Yes, I know DNA has problems too. And you folks aren't alone in being beat up. The toolmark and firearms folks have even more troubles with subjectivity issues and standards for a match. The arson guys have had all sorts of troubles in recent years. Various labs have had testing scandals. The FBI had its problems with bullet lead analysis. There's alot of scrutiny of forensics that just wasn't around a decade or so ago.

I think error rate may be a big red herring. Defending "zero" means that every time there is an error (and given the huge number of IDs done every year, there's going to be some now and then), one has to again explain why that one's human error not practioner error. Perhaps saying mistakes can happen, but are very rare, may be a response that works with the courts and juries.

The key question may be simply good training, good proficiency testing, and better verification and review protocols to catch mistakes, and good analysis when the mistakes happen to figure out why. (Avoiding the mystery of Cowans, where the underlying images still haven't been made public, and all the fighting over McKie where I gather there still hasn't been a good public inquiry.)

Posted: **Sat Nov 03, 2007 9:34 am**

Is it the issue of fingerprints themselves being challenged, or the examiners approach to comparing said fingerprints?

Isn't there really two seperate issues and the courts are just grouping them together into one? The reliability of fingerprint evidence or the infallibility of the individual examiner?

The concern people have with fingerprint comparison has me confused because in my opinion, fingerprints ARE or they ARE NOT a match. Not including the examinations that end in results of inconclusive, et. al, but this science seems pretty easy to me. It's not a pseudoscience, it's not magic, it's not an illusion. The proof is in the prints!

I just don't get it.

Posted: **Sat Nov 03, 2007 9:35 am**

Ack, double post, sorry.

Posted: **Sat Nov 03, 2007 2:55 pm**

LJ

I believe what Sharon was alluding to was, that there would be a high probability that to conduct a DNA analysis of a latent that had been identified would result in the destruction of the original latent evidence. This is something the defense attorney would undoubtedly jump on, in that there is no latent evidence remaining for them to scrutinize with their own expert, if they so wished. You could photograph the latent first before its destruction by the subsequent DNA analysis and just retain the photo as the only evidence. This just provides another string to the bow of the defense.

Before you can even consider this, research needs to be done to ascertain what developed latents remain as suitable candidates for DNA analysis, post latent processing. DNA is possible after latent development by black powder, but how do we remove the tape from the lift card to gain access to the latent? (Duh) - Thus destroying the lift card & latent.
What effects do DFO, GV, RTX, R6G etc, etc, have on the subsequent DNA analysis?

We know that sequential latent processing when correctly applied will not necessarily destroy the latent before the next process is utilized. We need to know the same if DNA is to be a consideration after sequential processing. I know, I know, we can do the DNA first – but we have not yet developed and identified the latent!!!

What about the latent in blood. If no development is needed, you can photograph it and maybe cut out and retain the latent. Will the analysis be able to differentiate between the victims DNA and the suspects, commingling in the blood? It depends on all the variables.

What if the latent is from a commonly handled item? If the DNA does not match the suspect does that mean that the latent has been misidentified? (Inwardly laughs to himself). I suppose some defense brief will proffer this in the future and some judge will accept the argument – so sad really, but it just could happen. What if multiple suspects handle the item and we obtain a DNA match from one and a latent from the other, do we have a misidentified latent and/or mismatched DNA?

The legal system will need to understand that what they request may well destroy the existing latent evidence. If they are willing to accept that, OK. But the forensic practitioner should not be beaten over the head if this has occurred when requested by the judicial system. The buck stops with the court's request.

It all boils down to being the old “chicken & egg” question, lumped with “you pay your money and make your choice,” plus a bit (tad) of “you cannot have it both ways.”

Posted: **Sat Nov 03, 2007 4:33 pm**

JP,

Different states have different admissibility requirements so in some states there may only be a single issue and in other states there may be several issues. And to make it even more confusing, different states use different admissibility requirements for different issues (so one state may use Frye, Daubert, Rule 702, and more).

Generally speaking…..
California is a Kelly/Frye state
Maryland is a Frye-Reed State
Minnesota is a Frye-Mack State
New Hampshire seems to be a Daubert State but in their recent case they also talk about Rule 702.
Oregon is a Daubert State
Washington is a Frye State
Federal courts usually follow Daubert and/or Rule 702
And other states have to meet other requirements.

Some States only need to show that their ‘discipline’ is generally accepted while other states have to show that the ‘science’ is reliable, their ‘methodology’ is reliable and/or generally accepted. I’m pretty sure that some requirements even say that the application of the technique has to be reliable.

It’s easy for us to jump to conclusions about whether any judge was right or wrong, but we might not know if we don’t know that states requirements.

Posted: **Sat Nov 03, 2007 5:09 pm**

Whatever height the legal bar is set in any state or federal district, each judge is a god in his or her own courtroom and can reset the bar on a whim. Judges come with biases and may make impulsive rulings. Even within the same courthouse, judges may rule differently when confronted with the same set of circumstances. Uniformity? Maybe not.

Even if we had the perfect science with a logical answer to every conceivable question, it would still be the task of the defense attorney to probe and try to find weaknesses in the witness, the evidence, the methodology, or the conclusion. A defense attorney who failed to cross examine the witnesses may later be found to have failed to adequately represent the client.

With a million judges in the US (anybody got a better number?), does anyone think that there aren't a few out there who sooner or later wouldn't throw fingerprints out? That's not to say that we shouldn't continue to try and improve our science, it's just to say that no how good we get, there will still be room for more improvement. And as long as there is room for improvement, there will be lawyers and judges only too eager to point that fact out.

Don't get discouraged. Just keep working, keep testifying, keep learning, keep improving. And do the research necessary to satisfy each new objection that comes along.

CLPEX Chat Board

Baltimore Judge declares Fingerprints not reliable.

Baltimore Judge declares Fingerprints not reliable.