ACE-V1- IS IT SCIENTIFICally reliable and accurate?

THE FOUNDATION

Current literature, training, and court cases cite ACE-V (Analysis, Comparison, Evaluation, and Verification) as being a scientific method in the comparison and identification of friction ridge impressions.  Pat Wertheim attempts to make the comparison methodology fit the steps of the scientific method.2 He mimics the first step of the scientific method which states the question or problem as the examination of the unknown (analysis), or whose fingerprint is it?  He states the second step of developing the hypothesis is that it has or has not been made by the same person as the inked prints.  David Ashbaugh also attempts to make the methodology fit the portions of the scientific method which state the question or problem, and develop the hypothesis.  He believes these are presupposed in friction ridge uniqueness and permanence. He suggests that ACE-V is the scientific method, as it tests the hypothesis by the comparison and verification process of each comparison conducted.3

 This foundation for individualization is reliant upon the prevalent belief that the friction ridge detail is unique from person to person, even down to the single ridge unit.4 This is strikingly similar to the snowflake paradigm5 that is discussed in the editorial “Nature never repeats itself.”6

 This argument may be true only in theory with a friction ridge as a three-dimensional object and at a cellular level, however it is limited in foundational proof based upon the ability of the examiner to detect that uniqueness.  The paradigm of single unit friction ridge uniqueness ceases to exist once the three-dimensional aspect is transferred during the deposition of a latent fingerprint to a two-dimensional image.  The presupposition that Ashbaugh relates to in the scientific method has never been subjected to scientific testing.  Since the transfer is so relevant to the hypothesis, the ability of the various levels of biological uniqueness to transfer must be scientifically tested to determine what can be reliably used in the application of ACE-V.  The combination of deposition matrix, distortion, and effects of the substrate during the transference process removes the probability of a reliable duplication of single ridge unit uniqueness.  The effects of the processing techniques utilized to visualize the impression must also be considered to determine how detrimental those may be to the fingerprint paradigm of single ridge unit uniqueness. 

 

THE APPLICATION

ACE-V is the description of a process used in comparing two different things and having someone else agree with the result of the comparison.  It is a process, which will guarantee precision of application, not accuracy of the conclusion.  In reversing his decision in Plaza, et al, Judge Pollak found that the Daubert factor of testing remained both pertinent and unsatisfied.  He noted that scientific tests of ACE-V would clearly aid in measuring ACE-V’s reliability.7 In order to ensure that ACE-V will guarantee accuracy of conclusion, scientific testing prior to application of the ACE-V process must validate the premise of the comparison hypotheses (V-ACE-V).

 Validation of the premise that a conclusion is drawn from is part of the scientific method that must be established prior to the application of the process.  The process is neither the scientific proof nor the scientific method.  The reliance of a fingerprint paradigm based upon a three-dimensional aspect applied to a distorted two-dimensional transfer, without validating the ability for that uniqueness to transfer, will leave one wondering how scientific any conclusion about the transfer impression can be. 

 Friction skin growing and the snowflake paradigm does not equate to individuality.  It is inductive logic, which is: “A principal of reason to a conclusion about all members of a class from examination of only a few members of the class; broadly, reasoning from the particular to the general.” 8 Inductive reasoning can be problematic in the accuracy of conclusions reached.  A number of analogies of untested subjective applications, such as two persons who are unaware of their color blindness, can have repeatable outcomes following ACE and Verify their perceptive conclusions inaccurately.  Accuracy is directly proportionate to the amount of testing that the hypothesis is subjected.  Constant modification must take place until consistency is reached between the stated or modified hypothesis and the testing results.  Once this consistency is established, then the stated hypothesis becomes a reliable theory.

 Friction ridge skin growth and single ridge unit uniqueness do not equate to all remarkable features of the ridge being repeatable upon transference.  The ability for ridge detail to transfer accurately needs to be tested under all aspects of the transfer and the development of latent fingerprints.  It is not scientifically sufficient to compare one inked impression to another to form an opinion of detail transfer that is reliable and accurate.  Scientific testing is not a simple linear process.  Since individualization is occurring on a less than perfect, chance latent impression, testing must be done accounting for all aspects of the effects upon the transfer. The ability for friction ridge details known as level one (ridge flow) and level two (ridge path) to reliably persist during transfer is well established.  Pressure, distortion, substrate, debris, and development methodologies affect the persistence of the fine ridge features known as level three (ridge shape) detail.  It is not sufficient testing to find that there is agreement of ridge shape between one impression against another impression.  Testing the persistence of level three detail needs to be shown in thousands of transferred and visualized impressions of the same area of friction ridge skin.  Once the transferred persistence of level three friction ridge detail is established, it can be deduced as being a reliable theory for individuality.  It is deductive logic, which is: “The process of reasoning in which a conclusion follows necessarily from the stated premises; inference by reasoning from the general to the specific.” 9

 

THE CHALLENGE

Recent legal (Daubert) challenges to fingerprints have raised the issue of unreliability, due to the foundational basis for individualization not being tested.  Individualization by fingerprint comparison, which will affect life and liberty, must be founded on more than a subjective belief that it is biologically impossible for two persons to have the same fingerprints.  This biological premise is specifically lacking adequate testing of the ability to transfer, as well as the ability to perceive level three detail reliably. It relies entirely on the empirical observation noted by practitioners.

 Citing over 100 years of friction ridge observations as validation, one must also include the acceptance of the objective scientific testing and the quantitative validation that occurred during that time.  Individualization by fingerprints must be founded on the basis of it being biologically and scientifically impossible for an impression to have been made by more than one person.  Scientifically, there are no absolutes.  However, the probabilities established by the various statistical studies that different persons can share similar characteristics in the same sequential relationship can be so remote that they may be ignored. 

 Statistical studies of the random placement of Galton (level two) ridge characteristics determined reliable thresholds to establish individuality.10 These studies have shown over the years that they are viable as statistical thresholds for conclusions of individualization.  Most countries throughout the world have established a standard for individualization, which takes these statistical thresholds into account.

 Consistent definition and description of what counts as an applicable characteristic needs to be established.  Automated Fingerprint Identification Systems (AFIS) utilize ending ridges and bifurcations.  A number of friction ridge characteristics in agreement in an x- y coordinate relationship do not alone equate to a conclusion of individualization.  Although there may be no validity to require a particular number of matching ridge characteristics be present for individualization, there certainly is a utility in doing so.11  Practitioners augment those thresholds during the evaluative stage of the comparison process.  They must determine if other ridge features are present and whether they are consistent, within tolerance, to validate the statistical model reliability threshold.  At this point one can deduce individuality. 

 

THE VALIDATION

These statistical models are constantly being tested with AFIS systems that are used throughout the world.  Although the earlier statistical models of random placement of ridge detail do not take into account all features of a fingerprint, each latent search being conducted on the AFIS system is attempting to disprove the theory of fingerprint individuality.   

In the latent searchable ten print databases of just a portion of the United States, there are approximately 8 million subjects representing approximately 80 million fingers.  At least 1000 latent searches are conducted each day.  This equates to 38.8 - 80 billion searches a day, or 14.162 - 29.2 trillion searches a year based upon a search of either a single pattern or a total pattern reference. Multiply these figures by the AFIS systems throughout the world and the number is staggering.

 As a result of these automated searches, the resulting similarities of ridge characteristic placement found between two different persons are consistent with those mentioned by Dr. Edmund Locard around 1914.12  These threshold levels were formed by sound reasoning, and have been used as a vigorous guide by many countries in establishing individualization standards.  They have been performing amazingly good throughout the years, and they have been validated through scientific testing utilizing the power of fingerprint automation in combination with the empirical observations of fingerprint practitioners.  Fingerprint Whorld reported in their October 1996 issue the finding of seven similar Galton details in a sequential relationship that were not from a common source.

 Found on an AFIS search, the below images have eight details in sequence that are not from a common source.  This example is numerically the most level two characteristics found in agreement, as well as being the closest in tolerance of spatial relationship that this author is aware of that are not from the same source.

 

    

 The illustrated impressions are within tolerance for level two distortion, and have reached Locard’s statistical lower level of eight matching Galton characteristics, but lack the additional agreement of other ridge features Locard found necessary to individualize.13

 

The full images clearly display a number of dissimilarities. 

If only the cropped images were the evidence mark, could the analyst rationalize ridge features to be within tolerance?

 Locard also established If more than 12 concurring points are present and the fingerprint is sharp, the certainty of identity is beyond debate”.14 This level of reliability has been validated daily through the automated search of trillions upon trillions of fingerprints throughout the world.  The exponential probability of finding twelve characteristics above the illustrated eight characteristic non-matches and not being from the same source is mathematically impossible.  It has never been reported that two different persons, as the result of a computerized search, have ever been found to have anywhere near 12 Galton characteristics in agreement of level one and level two detail in the approximate relative position, with the identical characteristic type and ridge path direction.  Therefore, it can be said that it is scientifically and biologically impossible for two persons to have twelve or more Galton detail in sequence. 

 

 Anything less than this number the analyst should be prepared to defend with supporting ridge features that are clearly visible.  Once the premise of single ridge unit uniqueness for fingerprint individualization and the ability to transfer that uniqueness is scientifically validated, the application of ACE-V will be a scientific application. Individualization of fingerprints must be accomplished with a thorough understanding of both the biological uniqueness establishing the basis and the scientific validation of the theory of individuality.  Rejecting either is myopic and unscientific. 

 

 

John D. “Dusty” Clark

 


1 Ashbaugh, David, Quantitative--Qualitative Friction Ridge Analysis (Boca Raton, Florida:  CRC Press, 1999).

2 Wertheim, Pat A., “Scientific Comparison and Identification of Fingerprint Evidence”, Fingerprint Whorld, Vol. 26 Number 101, July 2000, pp 95-106

3 Ashbaugh, David, personal communication, January 28, 2002

4 Meagher, Steve: Regarding the testimony of Dr. William Babler;  U.S. vs  Byron Mitchell  presented at the Calif. Division,  IAI 2001Annual Training Seminar, Concord, Ca.

5 Thornton, John I. (Forensic Science Group, Department of Biomedical and Environmental Health Sciences, University of California, Berkeley, CA.), “The Snowflake Paradigm” (letter to the editor), Journal of Forensic Science 31 (2), April 1986, 399.

6 McRoberts, Alan L. Editorial, SCAFO “THE PRINT” Volume 12(5), September/October 1996, pp 1-3  http://www.scafo.org/library/120501.html.

7 U.S. v. Plaza, Acosta and Rodriguez, United states District Court for the Eastern District of Pennsylvania, Cr. No. 98-362-10,11,12, March 13, 2002, pp. 48.

8 The American Heritage Dictionary, New College Edition

9 Ibid.

10 Stoney, D.A. and Thornton, J.I., " A Critical Analysis of Quantitative Fingerprint Individuality Models, " Journal of Forensic Sciences, JFSCA, Vol.31, No. 4, Oct. 1986, pp.1187-1216.

11 Thornton, Dr. John - "Setting Standards In The Comparison and Identification" 84th Annual Training Conference of the California State Division of IAILaughlin, Nevada - May 9, 2000.    http://www.latent-prints.com/Thornton.htm.

12 Christophe Champod, Institut de Police Scientifique et de Criminiologie BCH/Universite de Lausanne,    Edmond Locard - Numerical Standards & "Probable" Identifications, Journal of Forensic Identification, 45 (2) 1995, pp136-155.

13 Ibid:  “If 8 to 12 concurring points are involved, then the case is borderline and the certainty of identity will depend on: a) the sharpness of the fingerprints; b) the rarity of its type; c) the presence of the center of the figure [core] and the triangle [delta] in the exploitable part of the print; d) the presence of pores; e) the perfect and obvious identity regarding the width of the papillary ridges and valleys, the direction of the lines, and the angular value of the bifurcations.”

14 Ibid.