Forensics and Law in Focus shared an article from the CBC that touched upon the number of points of similarity in ballistic comparisons. Yawn. We’ve been through points of similarity in the fingerprint world, and the same issues exist in ballistics.
Ballistics
Whether or not you believe forensic science is an oxymoron, ballistics analysis seeks to determine if a bullet (or casing) fired from a particular gun. It does this by firing a second bullet from the same gun, comparing the markings on the two bullets (or casings), and seeing if they match.
But Sara Gordon has a problem with the way these matches are determined.
“She…says there are no set standards for the minimum number of similarities needed to match bullets.”
Well, I have a problem with Gordon’s problem.
Because we’ve seen this before.
Let’s look at fingerprints
The fingerprint and friction ridge world employs a similar method. Except that instead of looking at metal markings, a computerized automated fingerprint identification system (or a forensic examiner) looks at the friction ridges and searches for minutiae. These are the points at which the friction ridges end, or bifurcate into two ridges.
The (really simplified) theory is that if you take a constellation of minutiae from one fingerprint, and a constellation from another fingerprint, and the constellations are the same, then the fingerprints came from the same finger.
So how do you determine if the fingerprints are the same?
Sara Gordon would love one method.
Edmond Locard enters the chat
Dr. Edmond Locard was a French forensic examiner who was active in Lyon in the early 20th century. Among other things, Locard worked with Alphonse Bertillon (yeah, that guy) and established Lyon’s crime investigation laboratory.
He also created the tripartite rule in 1914 for minutiae comparisons. David Doglietto summarizes it as follows:
“If more than 12 concurring points are present and the fingerprint is sharp, then certainty of identity is beyond debate….
“If 8 to 12 points are involved, then the case is borderline….
“If a limited number of characteristic points are present, the fingerprint cannot provide certainty for an identification, but only a presumption for identification, and only a presumption proportional to the number of points available and their clarity.”
Sounds clear cut, doesn’t it? 14 points of similarity and you’re certain. 7 points, nope.
But think about it.
- What if you have 20 points of similarity (more than Locard’s 12), but have one glaring dissimilarity that cannot be explained by poor print quality? Locard or no Locard, that is NOT a match.
- And what if your crime scene (latent) print sample is very small and you only have 7 matching minutiae (fewer than Locard’s 8)? The late Clarence Phillips loved to show this very example. Locard or no Locard, that is a match.
Phillips enjoyed a long career in forensics with the Federal Bureau of Investigation, Printrak, Motorola, MorphoTrak…and the International Association of Identification (past President, 1986-87).
The IAI enters the chat
Before Phillips’ Presidency, and because of the various concerns about the Locard method, the IAI convened a group in 1970 to study the matter. On August 1, 1973, the IAI issued a statement (PDF) which read in part as follows:
“The International Association for Identification assembled in its 58th Annual Conference at Jack-son, Wyoming, this First Day of August, 1973, based upon a three-year study by its Standardization Committee, hereby states that no valid basis exists at this time for requiring that a pre-deter-mined minimum number of friction ridge charac. teristics must be present in two impressions in order to establish positive identification. The foregoing reference to friction ridge characteristics applies equally to fingerprints, palm prints, toe prints and sole prints of the human body.”
But the IAI’s determination had no effect on the laws in various countries. French and Italian courts require 16 matching points or the evidence is inadmissible.
In most other countries, present practice centers on a process (ACE-V) that admittedly has qualitative characteristics, but which is more accurate than the old 12-green 7-red system.
Back to ballistics
Returning to ballistics, if a minimum number of matching minutiae doesn’t work for friction ridges, why would it work for bullets and casings? The Death Penalty Information Center has published its concerns on the topic.
Nevertheless, consecutive matching striae is a thing.
“In three dimensional toolmarks when at least two different groups of at least three consecutive matching striae appear in the same relative position, or one group of six consecutive matching striae are in agreement in an evidence toolmark compared to a test toolmark. In two dimensional toolmarks when at least two groups of at least five consecutive matching striae appear in the same relative position, or one group of eight consecutive matching striae are in agreement in an evidence toolmark. For these criteria to apply, however, the possibility of subclass characteristics must be ruled out.”
The Association of Firearm and Toolmark Examiners (AFTE) permits a more qualitative analysis from its practitioners, “sufficient agreement”—a practice that concerns the Sara Gordons of the world.
“Modern practice adds a second layer: three-dimensional surface topography acquired by confocal or structured-light scanning, processed by the Congruent Matching Cells (CMC) algorithm to produce a quantitative comparison score that supplements the optical examination. The scoring output, when converted to a likelihood ratio under the Algorithmic Comparison Score (ACS) framework, meets the ENFSI evidential-weight reporting standard now required in several European jurisdictions.”
