PRINCIPLES OF FRICTION RIDGE ANALYSIS
Although Francis Galton was not the first person to propose the use of fingerprints for identification, he was the first to study them scientifically, thereby laying the foundation for their use in criminal cases, biometrics, and anthropology. Galton, a dilettante who studied a wide variety of disciplines including anthropology, genetics, geology, and statistics, was influenced by his cousin Charles Darwin and collaborated with Karl Pearson. Fingerprints, the first scientific text on the subject, was published in 1892 by Macmillan; Galton went on to publish additional works on fingerprints in 1893 (Decipherment of Blurred Fingerprints) and Fingerprint Directories (1895). In his 1888 paper for the Royal Institution, Galton estimated the probability of two persons having the same fingerprint and studied the heritability and racial differences in fingerprints. Galton’s work on fingerprints summarized common patterns in fingerprints and devised a classification system that is still used to this day. The method of identifying criminals by their finger prints had been introduced in the 1860s by William Herschel in India, but their potential use in forensic work was first proposed by Dr Henry Faulds in 1880 . The concept of uniqueness is typically associated with the philosopher Gottfried Wilhelm Leibniz, who stated, “For there are never in nature two beings that are perfectly alike and in which it is not possible to find a difference that is internal or is founded on an intrinsic denomination” . While it is one thing to understand all people and things are separate in space and time, it is quite another to prove this supposition. Galton was the first to attempt to calculate the likelihood of finding two friction ridge patterns that are the same. Numerous researchers have recalculated this probability over the years by various calculations based on differing assumptions (see Table 1). But they all indicate that the probability of any one particular fingerprint is somewhere between 0.000000954 and 1.2 × 10−80 (0.0 with 78 zeros and 12)—all very small numbers indeed. Technically, even infinitesimal probabilities such as these are still probabilities and do not represent true uniqueness (which would be a probability of 1 in ∞), but the values are such that latent fingerprints, with sufficient minutiae, can be considered practically unique in many cases. The values in Table 19.2 also demonstrate the importance of finding as many points of comparison as possible; more similarities—with no significant differences—lead to a lesser probability of a coincidental match (false positive). No standard model for random match probabilities has been adopted for forensic fingerprint casework, but that day is soon coming.
Table 1 Comparison of Probability of a Particular Fingerprint Configuration Using Different Published Models for 36 Minutiae and 12 Minutiae (Matches Involve Full, Not Partial, Matches)
Under low-power magnification, friction ridge patterns are studied for the kind, number, and location of various ridge characteristics, or minutiae. As with many other types of forensic evidence, it is not merely the presence or absence of minutiae that make a print unique: It is the presence, kind, number, and, especially, arrange ment of those characteristics that create a one-of-a-kind pattern. When two or more prints are compared, a careful point-by-point study is needed to determine whether enough of the significant minutiae in the known print are present in the questioned print, with no relevant differences. This comparison process is demonstrated graphically in Figure 1. Figure 19.5 shows the comparison of two fairly complete prints; in reality, the majority of prints that are identified, resolved, and compared are partial prints, representing only a portion of the complete print pattern. A friction ridge print scientist must then determine whether a partial print is suitable for comparison, that is, if the print has the necessary and sufficient information to allow a proper comparison. A partial print, or even a complete print for that matter, may be identifiable as such but be smudged, too grainy, or too small for the scientist to make an accurate and unbiased comparison. Often this is the crucial step in a friction ridge print examination that is dependent on the scientist’s experience, visual acuity, and judgment.
FIGURE 1 The presence, kind, number, and arrangement of minutiae create the pattern used in a fingerprint comparison. The points are studied side by side with a magnifying lens.
One of the ongoing debates among forensic scientists is how many points of com parison are necessary and sufficient to reach a conclusion of identification. For years, many agencies had a point-counting standard that dictated how many points of comparison were required before a positive conclusion could be reached. The num ber of points varied from 8 to 16 to even 20 in some agencies. A concern with point counting, however, is that no scientific or statistical basis has been established that would indicate that 8 is not enough, but 10 might be, or 16, or more. The frequency of individual kinds of minutiae (deltas, bifurcations, crossings, etc.) in any population is not known and so begs the question of a numerical standard’s significance. Another concern is the question of the threshold limit: If an agency requires 10 points for an identification, what if 9 very clear points are found? Could a tenth point possibly be found or must an exclusion be made? What if it’s a small partial print and nine very clear points are all that are found? At an agency with a threshold of eight, this would be a match, and the problem is obvious. Many, if not most, agencies have now adopted a “no-point” standard, summarized in 1973 by the IAI, a professional association for forensic scientists involved in identification techniques, as “no valid basis exists for requiring a predetermined minimum number of friction ridge characteristics, which must be present in two impressions in order to establish positive identification” . The threshold then becomes one of a sufficient number of characteristics necessary to make a conclusion of identification, however many that might be. A scientist’s experience and judgment become central to the process of a quality examination; this judgment, ultimately, is derived from proper and comprehensive training coupled with a mentoring process of practical experience. This does not absolve scientists of the obligation to be able to articulate the points of comparison, their significance, and why they lead them to a conclusion of identification. Two or more experts may disagree, but they need to be able to offer cogent arguments as to why and how they reached their different conclusions.
