___________________________________
Last week, we looked at an internet article on Faking Fingerprints.  If anyone tried this, I would be interested in how it worked out.

This week, we keep abreast of latent print related information on the internet by looking at another interesting public website article.  This material was not altered in any way from the public posting at the link below as of yesterday, June 19.  The article by Simon Cole and William Thompson is from the online version of The Champion, newsletter of the National Association for Criminal Defense Lawyers, reprinted without permission for educational purposes only, so please secure permission before re-distributing this article.

http://www.nacdl.org/public.nsf/01c1e7698280d20385256d0b00789923/3419bfeee2f5c72c85256ff600528500?OpenDocument
copied June 19, 2005

___________________________________
Forensics
by William C. Thompson; Simon A. Cole

April 2005, Page 42

Forensics
By William C. Thompson; Simon A. Cole

Lessons From The Brandon Mayfield Case

Fingerprint examiners have always claimed that their craft is infallible. They routinely testify that a fingerprint match is a positive identification. If the defendant “matches” a latent print from a crime scene, for example, fingerprint experts claim to know with absolute certainty that the latent print was left by the defendant to the exclusion of all other human beings on the planet. They deny that it is even possible for a trained examiner who follows proper procedures to reach the wrong conclusion.1

This testimony might lead one to conclude that there has never been a documented case in which a qualified fingerprint examiner made an erroneous identification — but that conclusion would be wrong. False fingerprint identifications have been known since the 1920s.2 Errors have also occurred on proficiency tests.3 But the fingerprint community has attributed these errors to incompetent examiners, thus allowing latent print examiners and prosecutors to continue claiming that “the method itself” remains infallible.4

It is unclear whether the examiners who made errors were really less competent than their peers — typically the only evidence of their incompetence was the very error that incompetence purportedly explained. We suspect that the fingerprinting community has attributed errors to incompetence, rather than to the inherent unreliability of their craft, in order to maintain the pretense of infallibility in the face of irrefutable evidence that misidentifications have occurred.

Even this flawed pretense of infallibility suffered a serious blow last Spring when the FBI matched a latent print from the March 11 Madrid terrorist bombing to Brandon Mayfield, an Oregon attorney. Mayfield was a Muslim convert whose wife was born in Egypt. He was held for two weeks as a material witness. Fortunately for him, the Spanish national police found another suspect, an Algerian named Ouhnane Daoud, and found that he “matched” the latent print better than Mayfield. Mayfield was released with a rare FBI apology and is now suing the government.5

The incompetence theory won’t fly in the Mayfield case because a highly experienced and respected FBI examiner made the error. Two additional FBI examiners (one retired) also compared the two prints and (erroneously) confirmed the first examiner’s judgment. Finally, a highly respected independent expert retained by the court also corroborated the erroneous identification.

If the Mayfield error was simply a matter of incompetence, then we must accept that three top examiners from a lab that claims to be a world leader in the field, as well as a leading independent examiner, were all incompetent. Obviously the fingerprint community needs a new explanation this time.

Three possible explanations have been offered. First, an FBI press release suggested that the error arose from a poor-quality digital image.6 But that theory was a nonstarter. Digital images are widely used for fingerprint comparisons.7 None of the FBI examiners expressed any reservations about the image quality.

A second theory being discussed among forensic scientists was that FBI examiners had been influenced by Mayfield’s background. The FBI examiners may have thought he was a likely suspect due to his conversion to Islam, his Egyptian wife, his military training, and his having represented (in a child custody case) one of the “Portland Seven,” a group of Muslim men convicted of terrorist conspiracy. Consequently, they may have been predisposed to find a match and may have accepted too readily that they had. The problem with this theory, however, is that it suggests that the FBI examiners had allowed their “scientific” judgment to be biased by collateral evidence — i.e., evidence not pertinent to the scientific question. This type of bias is a particularly sensitive issue for the FBI laboratory.

In 1997, the Justice Department’s Office of Inspector General conducted an investigation of alleged scientific misconduct by the FBI laboratory’s explosives unit and issued a scathing report.8 Among the most serious charges was that FBI examiners had relied on collateral evidence when making key “scientific” determinations.

In the first World Trade Center bombing in 1993, for example, FBI examiners had relied on the fact that the suspects had access to urea nitrate to reach the conclusion that urea nitrate had been used to make the bomb. The inspector general rightly condemned this biased, circular reasoning as scientific misconduct. The FBI laboratory management agreed and pledged to take steps to ensure that the problem never happened again. The embarrassment surrounding this history makes allegations of bias based on collateral evidence a particularly sore subject for the FBI. An admission by the FBI that its examiners were biased by Mayfield’s “suspicious” background would be more than an admission of error in one case — it would be an admission of management failure and evidence of persistent and recalcitrant scientific misconduct. It is not surprising, then, that the FBI strenuously denied that its examiners had been informed of Mayfield’s background.9

This raised the third theory: that the Mayfield match was just a coincidence — that Mayfield just happened to have a fingerprint similar enough to the print left in Madrid to be mistaken for it. This theory seems plausible. Because the Madrid bombing was an important case, the latent print undoubtedly was circulated to law enforcement agencies throughout the world and compared against many fingerprint databases. Indeed, given the relatively recent development of technology for easily circulating digital images, the Madrid latent might well have been searched against more fingerprints than any other latent print in the history of fingerprinting. Perhaps if you search a latent print widely enough, particularly one containing a limited amount of friction ridge detail, eventually you will find someone who has a fingerprint similar enough to be mistakenly matched to it.

But if the Mayfield match was a simple coincidence, what are the chances that the unlucky individual would turn out to be a Muslim-convert with a connection to Muslim fundamentalists and not, say, a Presbyterian policeman from Peoria? More importantly, the coincidence theory undermines the long-held claim of fingerprint experts that coincidental fingerprint matches are impossible (at least when the match is declared by a competent examiner). Is the FBI ready to abandon that claim?

Ironically, it was only two years ago that a federal district court for the Eastern District of Pennsylvania, in a closely watched case, relied on the FBI’s assertions that fingerprints are essentially unique and that FBI fingerprint examiners had never made an error in casework when holding that the FBI’s fingerprint identification method had passed muster under the Daubert standard.10 Flimsy at the time, the foundation for fingerprint admissibility now looks even shakier.11

After apologizing to Mayfield, the FBI promised a review by “an international panel of fingerprint experts.” We viewed this as a positive development. The best way to learn from errors is to allow an independent group to conduct an open and public investigation. We were heartened that the FBI appeared ready to do exactly that.

What emerged from this investigation, however, was not exactly what we had expected. In November the Journal of Forensic Identification published a “synopsis” of the conclusions reached by the International Review Committee.12 The first curious aspect of this “synopsis” is that it was written by an FBI employee, Robert Stacey, of the bureau’s Quality Assurance and Training Unit, rather than the international review committee itself. By analogy, imagine a bank audit in which the conclusions are written by one of the bank’s managers rather than the auditors themselves.

According to Stacey, he relied on written reports prepared by individual members of the international committee. When we contacted a member of that committee personally, and asked if we could see copies of their individual reports, we encountered another dubious circumstance. The committee member told us the FBI had required them to sign a “non-disclosure agreement” that prevented them from releasing their conclusions to the public. So much for an open and public investigation.

Perhaps because of Mayfield’s impending lawsuit,13 Stacey’s report avoids the key issue of what the FBI’s latent print unit knew about Mayfield’s background, and when they knew it. Instead, the report blames “the inherent pressure of working an extremely high-profile case” for the Mayfield error. The conclusion that latent print examiners are more error prone in high-profile cases is disturbing, but not altogether implausible. In a recent study, one of us (Cole) found that more than 70 percent of exposed cases of fingerprint error occurred in serious cases, such as murder. Whether this means that more errors occur in these cases or simply that more errors are exposed in these cases is unclear.

Stacey’s report does not say how the committee was able to determine that it was pressures associated with a high-profile case that caused the error. Did they uncover evidence that the FBI examiners were pressured? Or did they simply reach this conclusion merely because this particular error happened to occur in a high-profile case? An adequate report of the incident would answer these questions; Stacey’s report does not.

Interestingly, the report also blames “confirmation bias,” which is the human tendency to see what one expects and desires to see when evaluating ambiguous evidence. This is potentially a very important conclusion. Academic commentators have criticized forensic scientists for failing to adopt adequate procedures to prevent confirmation bias.14 This report may be the first admission by the organized forensic science community that confirmation bias is a problem.

The report also concedes another point made by forensic science critics: that a second examiner who is asked to “verify” the results of a first examiner should do so blind — i.e., without knowing what the first examiner decided. According to the report, that was part of the problem in the Mayfield case. “[B]ecause the initial examiner was a highly respected supervisor with many years of experience,” the report states, “it was concluded that subsequent examinations were incomplete and inaccurate. To disagree was not an expected response.”

This conclusion undermines another favorite argument of the fingerprinting community: that any possible problem with a fingerprint identification can be cleared up by the simple expedient of having another expert check the results of the first. It is encouraging that the report endorses blind verification, which would be a significant improvement over current practices.

Finally, the report recommends better documentation of the procedures followed for fingerprint identification. This is another important step. One of the problems in understanding the reliability of fingerprint identification is that much of the process occurs inside a latent print examiner’s head. Because latent print examiners produce almost no documentation of what they see, it is very difficult to reconstruct — after the fact — what might have led an examiner to conclude that a finger is the source of a print when, in fact, it is not. All we know is that the examiner reached the wrong conclusion; it is very difficult, perhaps impossible, to determine why.15

Unfortunately, the Stacey report endorses blind review and better documentation only for “designated cases,” which we gather means high-profile cases. This limitation makes little sense unless forensic scientists really believe (and where is the evidence?) that there is a greater risk of error only in high-profile cases. We think these improvements should be adopted in all cases.

A high-profile error like the Mayfield case provides an important opportunity to learn about the limitations of forensic science and find ways to improve it — if the errors can be examined in an open and objective manner. If we are really to begin to understand what causes fingerprint errors, the FBI needs to be more forthcoming. One member of the International Review Committee has told us the FBI will “probably” soon publish the committee members’ original reports on its Web site. We trust and hope that the FBI will do just that. And we hope that future investigations will be more open and transparent than this one.

Finally, we hope that fingerprint examiners will rethink their longstanding claim that competent examiners cannot make errors. Always wrong, this claim is now absurd. Like emperors caught without clothes, fingerprint examiners must now decide whether to acknowledge reality or continue their brazen pretense. And if they continue, courts will need to decide whether this unseemly conduct should be allowed.

Notes

1.See, Simon A. Cole, Grandfathering Evidence: Fingerprint Admissibility Rulings from Jennings to Llera Plaza and Back Again, 41 Am.Crim.L.Rev. 1189, 1196-97 (2004); Pat A. Wertheim, Scientific Comparison and Identification of Fingerprint Evidence, 16 The Print 1, 6 (2000).; Mary Beeton, The Fingerprint Controversy, 20 Issues in Sci. & Tech. 9, 10 (2004).

2.See, James E. Starrs, More Saltimbancos on the Loose? Fingerprint Experts Caught in a Whorld of Error, 12 The Scientific Sleuthing Newsletter 1, Spring 1988.; Robert Epstein, Fingerprints Meet Daubert: The Myth of Fingerprint “Science” is Revealed, 75 So.Cal.L.Rev. 605 (2002); Simon Cole, Suspect Identities: A History of Fingerprinting and Criminal Identification (2001).

3. See, Joseph L. Peterson & Penelope N. Markham, Crime Laboratory Proficiency Testing Results, 1978-1991, II: Resolving Questions of Common Origin, 40 J. Forensic Sci. 1009 (1995). Lyn Haber & Ralph N. Haber, Error Rates for Human Fingerprint Examiners. In Ratha and Bolle, Eds. Automatic Fingerprint Recognition Systems (2003), 349.

4.Simon A. Cole, Witnessing Identification: Latent Fingerprint Evidence and Expert Knowledge, 28 Social Studies of Science 687 (1998).

5. See, E.E. (Bo) Edwards, They may come for you next… The Champion (July 2004), 4.

6. Federal Bureau of Investigation, Statement on Brandon Mayfield Case, Press release (May 24, 2004)

7.See Michael Cherry, Edward J. Imwinkelried & Larry Meyer, Does the Use of Digital Techniques by Law Enforcement Authorities Create a Risk of Miscarriages of Justice? The Champion (December 2004).

8.Office of the Inspector General, U.S. Dep’t of Justice, The FBI Laboratory: An Investigation Into Laboratory Practices And Alleged Misconduct In Explosives-Related and Other Cases (1997).

9. Sarah Kershaw & Eric Lichtblau, Spain Had Doubts Before U.S. Held Lawyer in Madrid Blasts, New York Times, May 26, 2004.

10. United States v. Llera Plaza, 188 F.Supp. 2d 549, 566 (E. D. Pa. 2002).

11. Cole, Grandfathering Evidence at 1258.

12. Robert B. Stacey, A Report on the Erroneous Fingerprint Individualization in the Madrid Train Bombing Case. 54 J.Forensic Identification 706 (2004).

13.Rukmini Callimachi, Celebrity Lawyer to Represent Former Kansan in Suing FBI Associated Press (Aug. 28, 2004).

14.See Risinger et al., The Daubert/Kumho Implications of Observer Effects in Forensic Science: Hidden Problems of Expectation and Suggestion, 90 Cal.L.Rev. 1 (2002).

15.Simon A. Cole, More than Zero: Accounting for Error in Latent Print (Fingerprint) Identification 95 J. Crim. L. & Crim, (forthcoming May 2005).

National Association of Criminal Defense Lawyers (NACDL)
1150 18th St., NW, Suite 950, Washington, DC 20036
(202) 872-8600 • Fax (202) 872-8690 • assist@nacdl.org

http://www.nacdl.org/public.nsf/01c1e7698280d20385256d0b00789923/3419bfeee2f5c72c85256ff600528500?OpenDocument
copied June 19, 2005
______________________________________________________________________
Remember, the message board is always open: (http://www.clpex.com/phpBB/viewforum.php?f=2).  For more formal latent print discussions, visit onin.com: (http://www.onin.com)

______________________________________________________________________
UPDATES ON CLPEX.com

No major updates on the website this week.
______________________________________________________________________
Feel free to pass The Detail along to other examiners in weeks other than this one.  This is a free newsletter FOR latent print examiners, BY latent print examiners. There are no copyrights on The Detail, and the website is open for all to visit.

If you have not yet signed up to receive the Weekly Detail in YOUR e-mail inbox, go ahead and join the list now so you don't miss out!  (To join this free e-mail newsletter, send a blank e-mail to: theweeklydetail-subscribe@topica.email-publisher.com)  Members may unsubscribe at any time.  If you have difficulties with the sign-up process or have been inadvertently removed from the list, e-mail me personally at kaseywertheim@aol.com and I will try to work things out.

Until next Monday morning, don't work too hard or too little.

Have a GREAT week!