Art Forensics #5:
Forensics and the Fingerprints of the Artist
by
Dr. John Daab CFE, CFC, AFC, RI, for Fine Art Registry®
Introduction
Fingerprint match evidence is recognized by the legal system as one of the strongest elements in establishing the guilt or innocence of a person charged with a crime. If a person's print is on a piece of real evidence such as a gun, knife, or the steering wheel of a getaway car, it is generally accepted that the person was involved in the crime. The foundation for the forensic science of fingerprint analysis comes from its longevity of use without oppositional findings questioning its reliability. One hundred years without being proven false is accepted by the court system as the smoking gun in proving guilt or innocence. It is further bolstered by the belief that no two fingerprints are alike, and if a match is found, logic propels the conclusion that it could not be the case that someone else has a similar print and could be the guilty party. It has been noted that even identical twins do not have matching prints.
Fingerprints and Fine Art Authorship
In a recent case reported by Fine Art Registry® it was noted that the authorship and authenticity of an alleged Jackson Pollock painting was established by a fingerprint purportedly belonging to the artist, found on the painting. The investigator reporting the find was a self styled fingerprint "expert" who forgot one very important point: although there was a fingerprint on the painting there was no exemplar print with which to compare the alleged print. In any comparisons to determine authenticity it is absolutely necessary to have an authentic work or sample to use to determine if the questioned sample matches. The argument proposed by the self styled investigator was that the print on the paint can in the vicinity matched the print on the work and as such this established that Pollock was the author. Unfortunately the print on the can failed the chain of custody test. There was no evidence that this paint can print was that of Pollock. In point, it could have been anyone's print. It was later revealed by a police department fingerprint expert with over 25 years experience that the print on the painting was actually forged. Thus, not only was the work of questionable authenticity, but the forensic evidence itself was fake. This failure of correct analysis by a self styled expert comes at a time when fingerprint analysis is being challenged by not only the scientific community but by a spate of failed matches by various government fingerprint units.
Case Studies of Fingerprint Problems
The Los Angeles Times reported in 2008 that the Los Angeles Police Department convicted a slew of individuals improperly due to faulty fingerprinting conclusions. An internal memo indicated that some of LAPD fingerprint supervisors and technicians failed to properly carry out their responsibilities resulting in the possibility that over the years many individuals have been jailed as a result of improper fingerprint identification.
Brandon Mayfield was incarcerated for two weeks by FBI investigators who alleged that his fingerprints were on a bag of detonators found at the Madrid bombing in Spain in March of 2004. FBI agent Richard Werder stated that the print match was 100% positive. As it turned out the match was totally negative causing the FBI to retort that the "100% match" was caused by poor digital image used to match. Thus, the move by the stalwart investigation agency of the US government was to jail first, and then do the analysis later.
In 2005 Jeremy Brian Jones was arrested on misdemeanor charges. By using an alias he was able to bypass the fingerprint system. The FBI system did not pick up via fingerprints that Jones was a wanted sex offender. Jones went on to murder some people after his release. In order to understand how egregious mistakes happen in fingerprint matching it is necessary to understand the process of fingerprint identification.
The Forensics of Fingerprint Evidence Gathering: Making the Match
Fingerprints consist of "rolled" and latent. Rolled prints are those obtained by a technician taking each fingertip, pressing it onto an ink pad and rolling it over a given paper to secure the fingerprint. These are also known as the exemplars or authentic fingerprints of a given individual. Such prints are rolled from side to side to secure the full imprint of the fingertip. Latent fingerprints are prints secured from crime scenes, weapons, or any item which can be touched by hand and capable of recording that touch. Biro went to the alleged Pollock painting to determine if a print existed; Werder went the bag of detonators to secure prints. The latent print may be given relief by using various chemicals to develop an image. The image is then digitized or photographed and inputted into a file for future comparisons or identification.
The matching takes place by taking the latent print and comparing it to the rolled or exemplar print (see - below for sample prints and their matching details). The nature of a given fingerprint is that it has ridges, whorls, arches, loops and minutiae. A given print has between 30-50 points of comparison which are categorized into three levels. Level one looks at the gross details and the flow of the ridges. Level two examines the details of the ridges such as forks and islands. Level three identifies even more detail of the pores, shape of the ridge, and width. Identification of match occurs when level one matches the exemplar and either level two or three provide further matches. The full match in its objective process is the Analysis, Comparison, and Evaluation (ACE) of the prints. The subjective phase occurs when the examiner or fingerprint examiner at the Verification stage verifies that there is a match. The protocol demands that one or two experts confirm the match via the ACE-V paradigm.
(Images from Wikipedia.)
Theoretical Issues
That each fingerprint is unique and that there are no duplicates is an assumption not established or warranted by empirical investigation. Further, even though the courts believe that fingerprinting has existed for 100 years without a confirmation of non-uniqueness, it does not follow that a duplicate would not be found in 2015, 2020, etc. Let us remember that the sun used to revolve around the planets and that the horizon meant the end of the earth.
Practical Problems
There are 30-50 possible hits to determine a match between exemplar and latent. Depending upon the fingerprint agency and locale it would be acceptable to call a match with as little as 8 hits out of the 30 to 50 possible hits. Normally latent prints are not full or rolled out but parts of a print. The fingerprint examiner is going short to long to verify – he is taking a short amount and concluding that the short data matches the long (30-50) points. This, one would argue, is a gross assumption especially in the absence of any quantitative analysis. Corroboration takes place not in the blind but from one examiner to the other examiner in the same office. The rule is that the second or third examination be accomplished so that there is no connection between the first examiner and the others. Being in the same office and being paid by the same agency pollutes the results of the corroboration. Further, the fact that an examiner chose to work for a police organization adds an element of bias into the analysis even though examiners believe that they are acting objectively. After moving through ACE – the objective part of the match – it is accepted that the "V" (verification) phase is totally subjective or based on the experience of the examiner. This has been labeled as the art part of the process and as such is replete with serious and problematic baggage. In point, the legal system accepts the examiners call simply because he is an examiner. If the court accepts this, why not accept the fact that since John is a detective for 25 years and based on his experience he believes that "X" is guilty, then "X" is guilty?
Fingerprinting, Fine Art Authentication, Authorship and its Problematic Structures
Is it possible to claim that an artist's fingerprint on a work establishes that the artist authored the work? The answer is that merely because the print is on the work does not establish that the artist created the work. If somehow a match was established and not by present legal system processes, (keep in mind that engineers working on a project do not establish the standards of engineering, which are - set by outside professional and scientific processes) it would not establish that the author of the print created the work but merely touched the work. Fingerprints on a work merely establish that the fingers - which left them and the structure of the work are connected, nothing more and nothing less. But then there is an issue with to what extent are there prints from our artists to establish this connection. If fingerprinting is only 100 years old what happens to all the artists beyond the 100 year mark? Further, do we have a data bank of artists' fingerprints within the last 100 years to serve as exemplars to make the connection between the work and the artist? Probably not. One would conclude at this point, that since:
- Fingerprint analysis is not science but built on practitioner processes lacking standards
- Artist fingerprints are only good over the last 100 years
- The quantity of fingerprints available to demonstrate a connection between the artist and his work is almost negligible unless he or she was involved in a fingerprint requirement – crime, passport, etc.
- Fingerprints on a work do not establish that a work was created by the artist
It follows that the use of fingerprints to establish art authenticity or authorship is significantly problematic so as to engage the process minimally. If anything fingerprint identification on a work of art should be used to establish provenance considerations. If in fact we have an exemplar, and a fingerprint match on a work, it would be reasonable to conclude that a connection in the chain of custody exists between the artist and work. It does not establish that it is authentic or that it was from the hand of the artist. Given that other telltale factors exist such as style, materials used and so on, the fingerprint match may be added to the array to add greater weight to the attribution of the work.
Conclusion
The above analysis notes that the current process of fingerprinting has many problems in determining fine art authorship. Fingerprints alone do not establish authorship. Fingerprints merely establish a link between a material and an individual artist's exemplar prints. If, it is found that a work contains fingerprints matching the authenticated samples, and further analysis of style, brushstroke, subject matter, materials used, time frame, and cultural telltales, match comparative works of a given artist, - fingerprints then become the smoking gun of authorship. The aggregate of factors make the probative value or the preponderance of evidence, compelling to the point that denying authorship becomes unreasonable.
— by Dr. John Daab CFE, CFC, AFC, RI
| October 19, 2009
Discuss on FAR® Forum
|
Print
|
Fine Art News Feed
View all Articles
