Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 53
Ballistic Imaging 3 Firearms Identification and the Use of Ballistics Evidence This chapter provides a key context for the next part of the report by examining the nature of ballistics evidence itself and the ways it has been analyzed and used over the past century. Since the primary focus in this study is the ability of computer-based systems to discern unique characteristics in ballistics evidence on the basis of images of different sorts, it is useful to first consider how uniqueness and reproducibility of characteristic marks have been discerned through the particular media of the comparison microscope and the human eye. We briefly describe the general nature of toolmarks as evidence (Section 3–A) and then outline the basic history and theory of firearms identification (3–B). Section 3–C summarizes the literature on the uniqueness, reproducibility, and permanence of marks as detected by traditional methodology. The circumstances of firing a weapon, the manufacture of firearms and ammunition, and measures taken (or not taken) by firearms users can all pose various complications to the identification and linking of ballistics evidence; we review some of these issues in Section 3–D. Section 3–E offers key commentary on the scope of the committee’s work in the context of the material in this chapter. The final section (3–F) briefly describes the general role of imaging and photography in firearms identification, as a prelude to the next section of the report. 3–A TOOLMARKS AS EVIDENCE Various branches of forensic science—dealing with the analysis of evidence as diverse as handwriting samples and soil or mineral samples—are
OCR for page 54
Ballistic Imaging increasingly compared with one of their sister branches: the analysis of deoxyribonucleic acid (DNA) markers and identification based on those markers. As we note in Section 1–C, the term “ballistic fingerprinting” has also come into common (albeit somewhat inaccurate) usage to describe some features of firearms identification, suggesting a comparison with fingerprint evidence. Firearms toolmark evidence differs from DNA and fingerprint evidence due to their basic point of reference: the former links to a particular firearm while the latter two link to a particular person. Links between pieces of ballistics evidence can point to a common gun from which exhibits were fired, but not necessarily to the same person pulling the trigger. A potential match suggested by a national reference ballistic image database could suggest a link from a piece of crime scene evidence to an original firearm point of sale, but that link is at least doubly indirect: the link is only to the location of the transaction and not immediately to the firearm’s purchaser, and subsequent identification of the purchaser does not necessarily mean that the purchaser still possesses the gun (or fired the shot in a crime). However, it is important to consider that—alone, absent any other evidence or knowledge of circumstances—even person-specific fingerprint and DNA evidence is necessarily one step removed and indirect. It is possible for fingerprint or DNA evidence to be present and retrievable at crime scenes without its source person having been the crime’s perpetrator. Forensic evidence can be used—in combination with other investigative findings—to develop a theory of what transpired at the scene and who may have committed the crime, but the link to any specific person from the ballistics evidence alone is necessarily indirect. Toolmark evidence and DNA evidence are markedly different in another crucial respect, which is the subjectivity inherent in the analysis. Firearms identification ultimately comes down to a subjective assessment—specifically, a subjective probability statement (although practitioners often render these as absolute statements). Firearms examiners observe concrete, objective phenomena, but—as Thornton and Peterson (2002:24–25) observe, “there is an incredible amount of difficulty attached to the development of a statistical basis for evidence evaluation” in forensic science fields like firearms examination: Behind every opinion rendered by a forensic scientist there is a statistical basis. We may not know what that basis is, and we may have no feasible means of developing an understanding of that basis, but it is futile to deny that one exists…. The most common and coherent theory of forensic identification is that where there is a high degree of variation among attributes (of toolmark striations, writing, friction ridges on skin, and so on), then where a “match” is observed the probability that the match is
OCR for page 55
Ballistic Imaging coincidental rather than reflecting a shared source will be very small… Forensic individualization sciences that lack actual data, which is most of them, have no choice but to either intuitively estimate those underlying probabilities and calculate the coincidental match probability from those subjective probabilities, or simply to assume the conclusion of a miniscule probability of a coincidental match (and in fact they do the latter). In the specific context of firearms and toolmark examination, derivation of an objective, statistical basis for rendering decisions is hampered by the fundamentally random nature of parts of the firing process. The exact same conditions—of ammunition, of wear and cleanliness of firearms parts, of burning of propellant particles and the resulting gas pressure, and so forth—do not necessarily apply for every shot from the same gun. Ultimately, as firearms identification is currently practiced, an examiner’s assessment of the quality and quantity of resulting toolmarks and the decision of what does or does not constitute a match comes down to a subjective determination based on intuition and experience. By comparison, DNA analysis is practically unique among forensic science specialties as having a strong objective basis for determination and as being amenable to formal probability statements. Thornton and Peterson (2002:Fig. 1) rank various forensic science subfields on a continuum of relative subjectivity. On the low end of that scale is DNA analysis, along with serology (blood type determination) and drug and narcotic identification. They identify firearms and toolmark identification as having relatively high subjectivity, on par with fiber identification. They identify blood spatter interpretation, voiceprint analysis, and bite-marks as a group of forensic science specialties just slightly more subjective than toolmark identification, and handwriting and hair identification as a cluster slightly more subjective yet. 3–B TRADITIONAL FIREARMS IDENTIFICATION Smith (2004:130) succinctly summarized the basic task of a firearms examiner in making an identification between pieces of evidence: Before a microscopic comparison begins, a foundation is built by measuring and comparing available class characteristics, such as General Rifling Characteristics (GRCs). These objective criteria are used to narrow the pool of candidates for determining a common source. Once an available foundation has been established, a common source often can be determined by evaluating individual microscopic marks of value using pattern recognition. In traditional firearms identification—part science and part art form, still carried out today using the same basic tools that gave rise to the field
OCR for page 56
Ballistic Imaging in the 1930s—the firearms examiner faces the formidable cognitive task of forming a mental pattern of identifying marks and features on bullet and cartridge case evidence. That pattern must then be matched to those from other exhibits.1 3–B.1 Individualization and Identification: Class, Subclass, and Individual Characteristics The label “firearms identification” is another instance in which language can be a bit elusive. Although “the terms ‘identification’ and ‘identity’ are used constantly by practitioners” of criminalistics, or the forensic analysis of evidence, Kirk (1963:236) argued that the usage is an “unfortunate failure of nomenclature.” Rather than identification (as that term is commonly understood), Kirk argues that “criminalistics is the science of individualization:” The criminalist does not attempt identification except as a prelude to his real function—that of individualizing. The real aim of all forensic science is to establish individuality, or to approach it as closely as the present state of the science allows…. What was actually done was not the identification of the fingerprint, but rather the individualization of a person as the one who left the fingerprint…. [Likewise,] if the firearms examiner said that the bullet was a Colt 45 A.C.P. but could not individualize the gun that fired it, his value would be relatively slight. Thornton and Peterson (2002:8, 9) further differentiate between the two terms: “Individualization is the process of placing an object in a unit category which consists of a single unit. Individualization implies uniqueness; identification, strictly speaking, does not require it.” They also note the frequent use in forensic science of “identification” when “individualization” is meant, but recognized that “it is a constraint imposed by our language”; relying on the term “individualize” would likely lead to public confusion. “It should be appreciated, however, that the process of identification means one thing to the forensic scientist, and another thing to the botanist or the zoologist.” (See also Champod [2000:1077] on “individualization” versus “identification.”) The phrasing of “individualization” as the act of associating an object 1 Further information on current practice in firearms identification and images connected with that work are available through resources at the Association of Firearm and Tool Mark Examiners (http://www.afte.org) and the Scientific Working Group for Firearms and Toolmarks (http://www.swggun.org). Additional information and images—and tools for simulating the use of the comparison microscope for comparing bullet and cartridge evidence—are accessible at http://www.firearmsid.com.
OCR for page 57
Ballistic Imaging with a category that contains only one unit is a useful one in addressing the basic approach of firearms examination. Firearms examiners typically give testimony to the effect that this particular gun, and not any other gun in the world, was used to fire particular shots. Yet even in the profession’s earliest days, Hatcher (1935:266) anticipated a common reply to this type of argument: The firearms witness may expect to be asked how he can be sure of his findings, in view of the fact that he can have examined only a few of the countless thousands of guns that exist. But how does [an eyewitness to the crime] that identified the prisoner know that some other person may not look exactly like him? He will say that he has seen enough people in his life to know that the rule is that all people look different, and that the chances are overwhelmingly against finding two people who look so much alike that they cannot be told apart. In precisely the same way, the firearms expert, who admittedly cannot have actually examined more than the tiniest fraction of a percent of all the guns in the world, can still have had enough experience to be well aware of what differences do occur, and to know that the chance of finding two bullets from different guns that are exactly alike in every detail of their surface markings is infinitesimally small. The same thing applies with equal force to the finger-print method of identification. Subsequently, he posited this idea more succinctly: “It may be quite common for two or more prominent individual marks on bullets from two entirely different guns to match exactly, but the chance that there will be a correspondence of a great many of the individual characteristic marks on two bullets that came from different guns is so remote as to amount to a practical impossibility” (Hatcher, 1935:288). The basic approach to identification in forensic science developed into the concept of class characteristics and individual characteristics. Thornton and Peterson (2002:5–6) define these concepts: Class characteristics are general characteristics that separate a group of objects from a universe of diverse objects. In a comparison process, class characteristics serve the very useful purpose of screening a large number of items by eliminating from consideration those items that do not share the characteristics common to all the members of that group. Class characteristics do not, and cannot, establish uniqueness. Individual characteristics, on the other hand, are those exceptional characteristics that may establish the uniqueness of an object. It should be recognized that an individual characteristic, taken in isolation, might not in itself be unique. The uniqueness of an object may be established by an ensemble of individual characteristics. A scratch on the surface of a bullet, for example, is not a unique event; it is the arrangement of the scratches on the bullet that mark it as unique.
OCR for page 58
Ballistic Imaging In the context of examining bullet and cartridge case evidence, such parameters as caliber and the number of lands and grooves are class characteristics that can be used to screen or filter out possible comparisons that could not possibly have been fired from the same gun. Other class characteristics include firing pin shape (for instance, the distinctive rectangular firing pin impressions left by Glock firearms2) or pronounced “drag marks” that can be caused by firing pins as the cartridge case goes through the ejection process. Individual characteristics—the fine striations on a bullet’s surface or peculiar microscopic textures in the firing pin impression, for instance—allow for the set of possible guns from which an exhibit could have been fired to be narrowed. However, a binary split between class and individual characteristics proved too limited to describe the full range of phenomena observed by experienced examiners. In 1985, the Association of Firearm and Tool Mark Examiners (AFTE) convened a committee to develop a consensus document on a theory of toolmark identification and a range of basic conclusions that could be reached from comparison of toolmark evidence. The result, produced in 1989 and unanimously approved by the organization’s membership, is reproduced in Box 3-1. Importantly, the 1989 standard also added a new term to the firearms identification lexicon: subclass characteristics, defined as follows (Moran, 2002:228): Discernable surface features of an object, which are more restrictive than “class characteristics” in that they are Produced incidental to manufacture Significant in that they relate to a small group source (a subset of the class to which they belong) Can arise from a source which changes over time Caution should be exercised in distinguishing subclass characteristics from individual characteristics. As a middle ground between class and individual characteristics, subclass characteristics covered marks or features that—arising from specific manufacturing techniques, or flaws in said techniques—could induce similar marks on pieces of evidence even though they originated from different 2 Rectangular firing pin marks were a telltale sign of Glock pistols “prior to the introduction of Smith & Wesson’s Sigma Series, Model SW40F, semi-automatic pistol,” the design of which is very similar to Glock standards. Nichols (1995:133, 134), in documenting the similarity between this firearm’s marks and the known Glock characteristics, notes that the two highly similar gun types can be distinguished by ejector and extractor marks, as well as a characteristic “dimple on the [casing] head above the firing pin drag” on Glock rounds.
OCR for page 59
Ballistic Imaging BOX 3-1 Association of Firearm and Tool Mark Examiners (AFTE) Theory of Identification and Range of Conclusions Theory of Identification as It Relates to Toolmarks The theory of identification as it pertains to the comparison of toolmarks enables opinions of common origin to be made when the unique surface contours of two toolmarks are in “sufficient agreement.” This “sufficient agreement” is related to the significant duplication of random toolmarks as evidenced by the correspondence of a pattern or combination of patterns of surface contours. Significance is determined by the comparative examination of two or more sets of surface contour patterns comprised of individual peaks, ridges, and furrows. Specifically, the relative height or depth, width, curvature, and spatial relationship of the individual peaks, ridges, and furrows within one set of surface contours are defined as compared to the corresponding features in the second set of surface contours. Agreement is significant when it exceeds the best agreement demonstrated between toolmarks known to have been produced by different tools and is consistent with the agreement demonstrated by toolmarks known to have been produced by the same tool. The statement that “sufficient agreement” exists between two toolmarks means that the agreement is of a quantity and quality that the likelihood another tool could have made the mark is so remote as to be considered a practical impossibility. Currently, the interpretation of individualization/identification is subjective in nature, founded on scientific principles and based on the examiner’s training and experience. Range of Conclusions Possible When Comparing Toolmarks The examiner is encouraged to report the objective observations that support the findings of toolmark examinations. The examiner should be conservative when reporting the significance of these observations. The following represents a spectrum of statements: IDENTIFICATION: Agreement of a combination of individual characteristics and all discernible class characteristics where the extent of agreement exceeds that which can occur in the comparison of toolmarks made by different tools and is consistent with the agreement demonstrated by toolmarks known to have been produced by the same tool. INCONCLUSIVE: Some agreement of individual characteristics and all discernible class characteristics, but insufficient for an identification.
OCR for page 60
Ballistic Imaging Agreement of all discernible class characteristics without agreement or disagreement of individual characteristics due to an absence, insufficiency, or lack of reproducibility. Agreement of all discernible class characteristics and disagreement of individual characteristics, but insufficient for an elimination. ELIMINATION: Significant disagreement of discernible class characteristics and/or individual characteristics. UNSUITABLE: Unsuitable for comparison. SOURCES: Excerpted from AFTE Criteria for Identification Committee (1992) and Grzybowski et al. (2003). sources. Two examples of carryover of subclass characteristics are described in Box 3-2. The AFTE theory of identification is rooted in the recognition that “the interpretation of individualization/identification is subjective in nature.” However, it melds that recognition with more objective, quasi-quantitative benchmarks—“sufficient agreement,” “significance,” “likelihood … so remote,” and agreement in both “quantity and quality”—but no specific empirical definition is given for these terms. Importantly, the AFTE theory does set up two cognitive thresholds that must be crossed in order to arrive at the conclusion that two exhibits share the same source, to the exclusion of all others. In describing the comparison of fingerprints, Stoney (1991; quoted in Moran, 2002:233) wrote of the basic cognitive process: When more and more corresponding features are found between two patterns, scientists and lay person alike become subjectively certain that the patterns could not possibly be duplicated by chance. What has happened here is somewhat analogous to a leap of faith.3 It is a jump, an “extrapolation,” based on the observation of highly variable traits among a few characteristics, and then considering the case of many characteristics. Duplication is inconceivable to the rational mind and we conclude that there is absolute identity. 3 The “leap of faith” involved in extrapolating to all possible sources in the world was echoed, in specific reference to firearms identification by Rowe (1991). Understandably, some firearms examiners have objected to some overtones of the “leap of faith” phrasing; for instance, Miller and McLean (1998:17–18) “respond [to the phrasing] by stating that the published data which does exist, certainly presents enough statistical data to indicate more substance to the identification theory than a ‘leap of faith.’’’
OCR for page 61
Ballistic Imaging BOX 3-2 Examples of Subclass Carryover Lomoro (1974:18) described test firings from F.I.E. Titanic revolvers, in a particular serial number range (but later shown to include some of a different make from the others). These firings produced seemingly similar striation patterns on fired bullets that could be naively matched to each other even though they had been fired from different weapons. “This pronounced stria [pattern] was only present on the lands of the bullets and very little if any matching stria was found on the grooves of the bullets. Examination of the bore revealed that either a worn or a very poor rifling tool was used during manufacturing,” causing the similarity be imparted within a batch of separate barrels. Matty and Johnson (1984) observed that the particular tooling used on some Smith & Wesson firing pins results in a pattern of concentric rings that is repeatable in consecutively tooled pins. This pattern appears in the firing pin impression, meaning that examiners need to downweight the ring patterns and focus on other individual features to make correct matches on the firing pin impression. He continues that this “leap” occurs—in fingerprinting, as in other branches of forensic science, “without any statistical foundation.” In a sense, the AFTE theory of identification requires two extrapolations: first, that marks are sufficiently consistent with true matches (produced by the same tool) to have come from the same source, and, second, that the quality and similarity of corresponding features exceed the best known apparent agreement between true nonmatches (produced by different tools). It follows that both of these extrapolations—but particularly the latter—require considerable experience in comparing exhibits and training in recognizing significant features.4 3–B.2 Historical Evolution Some of the major advances in the field of traditional firearms identification are described in Box 3-3. Calvin H. Goddard, a founder of the Bureau of Forensic Ballistics in the New York City Police Department, is typically credited as the “father” of forensic ballistics in the United States; 4 Biasotti and Murdock (2002:219–220) advocated the addition of the known nonmatch criterion, beginning with their conclusion in 1984 that “existing research was insufficient to validate the quantitative objective criteria necessary to conclude that a working surface is unique.” The known nonmatch criterion “added a quantitative dimension” to the mix.
OCR for page 62
Ballistic Imaging BOX 3-3 Highlights in the History of Traditional Firearms Identification 1900: The Buffalo Medical Journal published “The Missile and the Weapon” by Albert L. Hall, commenting on his observations that firearms of different mark impart different striated marks and impressions on fired bullets. 1907: Cartridge case evidence was examined as part of the investigation of the 1906 Brownsville, Texas, riots in which black soldiers allegedly fired upon a white crowd in retaliation against racial slurs. Thirty-nine cartridges were recovered; all but six were matched to four rifles, based on examination of enlarged photographs of firing pin impressions. [Examiners concluded that no identifications could be made on the basis of recovered bullet evidence.] 1915: Charlie Stielow, an illiterate farmer in New York state, was convicted and sentenced to death for killing his employer and the employer’s housekeeper, the latter of whom was found dead on the doorstep of Stielow’s nearby home. Stielow’s conviction was due in part to testimony by an examiner who claimed to find nine abnormal scratches on the recovered bullets that corresponded to particular defects in the alleged murder weapon. However, a more careful subsequent analysis of the evidence by Charles Waite concluded that Stielow’s revolver could not have been the murder weapon; Stielow was pardoned and released from prison. 1925: Philip Gravelle was credited with the development of the comparison microscope for side-by-side comparison of ballistics evidence in Calvin Goddard’s article “Forensic Ballistics” in Army Ordinance. Goddard used the technique in the article, and the comparison microscope technique grew in popularity after the Saturday Evening Post’s two-part “Finger-printing Bullets” article in 1926 profiled New York City’s Bureau of Forensic Ballistics, which Goddard, Gravelle, and Waite helped found. 1929: On February 14, six members of George “Bugs” Moran’s North Side gang (and one acquaintance) were brutally killed in a Chicago warehouse by six men impersonating police officers, believed to be operatives of rival Al Capone’s South Side gang, in what quickly became known as the “St. Valentine’s Day Massacre.” As part of the investigation, Goddard prepared a detailed examination of recovered bullet and cartridge case evidence, connecting them to two Thompson submachine guns and a 12-gauge shotgun. Goddard—considered the father of modern firearms examination—subsequently left New York City to establish the Scientific Crime Detection Laboratory of Chicago, affiliated with Northwestern University. 1934–1935: First editions of two seminal texts on firearms examination were published: Gerald Burrard’s The Identification of Firearms and Forensic Ballistics and Julian Hatcher’s Textbook of Firearms Investigation, Identification and Evidence. 1958: John Davis’ An Introduction to Tool Marks, Firearms and the Striagraph argued for the use of the striagraph, an early tool for direct measurement of the surface contours of ballistics evidence. Though the striagraph never advanced beyond the research stage, it was a precursor to the use of imaging and profilometry techniques for firearms identification.
OCR for page 63
Ballistic Imaging 1962: J. Howard Matthews, a retired chemistry professor from the University of Wisconsin who became interested in forensic science after being called in to assist on a bombing case, published the two-volume Firearms Identification. 1969: At a 35-member conference hosted by the Crime Laboratory of the Chicago Police Department, firearms identification practitioners form the Association of Firearm and Tool Mark Examiners (AFTE). The new association’s newsletter developed into the AFTE Journal in 1973 and continues as a quarterly, peer-reviewed publication; the international association continues to hold an annual Training Seminar. SOURCES: Information from Hamby and Thorpe (1999), Hatcher (1935), and Goddard (1999). the text by Hatcher (1935) is generally considered the text on which the modern field of firearms and toolmark examination is based (Nichols, 2003). The modern field of firearms identification owes much to the development of the comparison bridge, a system of mirrors or prisms that permits the views from two separate microscopic lenses to appear side by side in the same field of view. The device equipped with this optical bridge, called the comparison microscope, was not designed for analysis of bullet or cartridge case evidence, nor was it first used in such forensic applications (see, e.g., Thornton, 1978; Goddard, 1980). Even early advocates of forensic firearms analysis cautioned against making too much of the technology at hand, noting that the comparison microscope, on its own, has limitations. Foreshadowing some contemporary claims about the capability of ballistic imaging systems, Burrard (1962:131–132) lamented: The most fantastic claims have been put forward for, and the most ridiculous descriptions of, [the comparison microscope] which are enough to suggest that it has magical properties, and that it automatically, and wholly of its own accord, rings a bell or utters some similar warning, when the two cartridge cases under examination exhibit the same [individual markings]. Unhappily there is no foundation for this comforting belief. In the hands of a trained microscopist the comparison microscope can be of great value in determining the identity of fired bullets; but for cartridge cases I have come to the conclusion that a high-class single instrument is preferable…. And even for fired bullets the comparison microscope offers
OCR for page 78
Ballistic Imaging fired at all is likely,” missing possible hits. Corrosion is also problematic because some means of dealing with it—including scrubbing with metal or other brushes—may also damage the weapon for the purposes of generating matchable marks. Corrosion is a concern not only for the internal parts of firearms, but also for recovered ballistics evidence. Though it is certainly ideal to process and collect evidence from crime scenes rapidly, such is not always possible. Bullet and cartridge case evidence may be exposed to the elements for days, weeks, or months until they are found and recovered, and bullet evidence may be irretrievable from shooting victims because rapid extraction could be harmful to their health. See Larrison (2006) for a study of the rate of degradation of bullets and cartridges—monitored every 6 months over 2 years—in four different and demanding environmental conditions: soil, water, open air, and an animal carcass. 3–D.2 Countermeasures Smith’s (1971) listing of “jokers,” which introduced this section, includes some deliberate countermeasures that might be taken by criminals—perhaps after firing a lethal shot—to try to mislead investigators. Aside from Smith’s rough ordering of situations by frequency and the rough impressions formed by examiners over their years of experience, there are no firm and systematic data on the frequency with which criminals deliberately alter firearms parts (or attempt to obliterate serial numbers) in order to avoid detection. Some countermeasures, like crude filing of the barrel, are relatively easy to perform (if not always effective).14 As Smith (1971) articulates in fleshing out his list of problematic situations, some countermeasures are particularly rare because they are inherently more difficult to perform. The barrel of a revolver, for instance, is a more integrated part of the firearm assembly than the more modular barrel of a semiautomatic, so swapping out a revolver barrel requires special tools and considerable knowledge. With expert knowledge, the range of countermeasures that might be taken is immense. For instance, Schecter et al. (1996:97) concluded that it was possible to replace the entire breech block of a SIG Sauer P226 9mm Luger pistol, so that “only the ejector will remain to give possible individual marks on the cartridge case base area from the ‘original’ weapon.” However, the registration of “good ejector and chamber marks on weapons 14 Of course, it is possible for quick measures to be very effective at changing toolmarks. At the committee’s December 2004 meeting, firearms examiner Lucien Haag shared an evocative presentation entitled “‘New’ Bores and Breechfaces in 60 Seconds,” which suggests two quick ways in which a knowledgeable user could rapidly alter the striations left by the barrel or the breech face markings left on cartridge casings.
OCR for page 79
Ballistic Imaging of this quality can be poor to non-existent,” and this technique would not be immediately obvious on inspection of the weapon. But the issue is how often people who use guns in crime would go to such lengths, akin to the proverbial question of why more criminals do not wear gloves while committing crimes (given the ease of leaving fingerprints). Smith (1971:18) argued that replacing pistol barrels “may happen not infrequently;” he personally reported having seen two such cases in criminal cases.15 He, however, “never personally encountered [a case of replacing a firing pin in an automatic pistol], and it is obvious that such replacement of other parts, (extractor and ejector), refiling of breech face exists, still the possibility of the replacement of this part alone must be kept in mind” (Smith, 1971:19). Refiling the breech face is problematic, but “much refiling would leave the breech surface bright & fresh in appearance”; for full concealment, “it would be much simpler to discard the arm entirely and replace it with another, rather than go to the effort” to make it appear that an arm with a refiled breech face had not been altered (Smith, 1971:19). It should be noted that while some countermeasures may successfully alter the “signature” markings left on ballistics evidence, they also serve to create new sets of markings—and, in some instances, prominent signs of the alteration—that can be used to match to evidence from later cases. Konior (2006) describes a case in which three cartridge casings and a pistol were recovered in investigating a homicide. The evidence casings did not appear to match test firings from the firearm; the latter all showed much greater levels of flowback around the firing pin impression. Subsequent analysis determined that an attempt had been made to alter the firearm by drilling the entire length of the pistol’s barrel; however, the person attempting the modification apparently inserted the drill bit into the firing pin aperture, widening the hole but also damaging the tip of the firing pin. The flowback and a new distinctive mark in the firing pin impression could be attributed to this attempted modification. 3–D.3 “Settle-In” Effect From the standpoint of establishing an RBID, a phenomenon of extreme interest is what might be called a “settle-in” or “breaking-in” effect: the notion that it takes several firings for a firearm’s unique, characteristic markings to stabilize. When the settle-in effect has been documented, it 15 In this context, Smith (1971:18) notes that the interchangeability of barrels was made vivid “in the famous Sacco-Vanzetti case.” Specifically, “the barrel of the fatal arm was, following a demonstration before the court by an expert for the defense, found present in one of the two other arms of the same type which had been introduced as exhibits, all three having been disassembled and reassembled in the course of the demonstration.”
OCR for page 80
Ballistic Imaging has been almost exclusively in the context of the striations left on bullets and not on markings left on cartridge cases. This—in addition to the time-intensive process of firing into a water tank or other nondestructive trap and retrieving individual bullets—is a major reason that the restriction of RBIDs to cartridge case entries is sensible. Murdock (1981:90) analyzed 10 test-fired bullets from each of three new and consecutively rifled barrels. He found that, “in general, the comparison between the first, second, and third firing from any one barrel failed to result in an identification. Some good agreement was present, however. The third, fourth, and fifth test firings from any one barrel could, however, generally be identified as having been fired through the same barrel.” From this, “it became obvious that each barrel needed to have a few bullets fired through it before it began marking in a reproducible, identifiable manner.” Similar work by Hall (1983:43) using a set of consecutively reamed polygonal barrels noted “rapid change during the first few shots,” so that firings 1 and 2 from the same barrel looked much different compared to each other than did firings 19 and 20. Hall found a “lack of identifications where bullets from any of the first five shots were compared to any others.” Matty (1985:65) noted similar behavior but was not as precise in suggesting the number of firings after which individual characteristics were discernible; instead, he observed that “it was not possible to make a conclusive match between test bullets #1 and #2 [from the same barrel, but] bullets #9 and #10 … were very similar in appearance.” From these and other works, Bonfanti and De Kinder (1999a:5) concluded that “about five to nine shots” are needed “so that [barrels] transfer a signature to a bullet.” They note that the studies of consecutively manufactured barrels that they summarize in their review meet this standard. However, at least one of those studies—work by Lutz (1970:25) using two new and consecutively manufactured .38 Special Smith & Wesson barrels—began with the barrels in a completely unfired state and downplays any settle-in effect. Indeed, Lutz described the completely unfired nature of the barrels as the key advantage of the study, as it “offer[s] the ultimate as far as similarities between two successive barrels. It is certain that the similar markings visible now would become less prominent after being subjected to normal firing, cleaning, and wear.” Even though all the test firings Lutz examined were among the first 12 uses of the barrel, “sufficient matching individual striae were noted on the bullets to enable the examiners to easily identify the barrel of origin for each of the bullets. Similarities in class characteristics were noted; however, microscopic comparison of the bullets revealed that each barrel had caused different markings to such an extent that each land and groove impression on each of the bullets had a great number of individual identifying striae.” Though a prominent settle-in effect for the breech face or firing pin—
OCR for page 81
Ballistic Imaging the prominent markings on cartridge cases—has not been documented, there may be structural or design features that may cause a similar effect. The Thompson (1996) investigation of highly similar marks resulting from different Lorcin L9 9mm pistols (see Section 2–D.1) included some firings from guns right off the manufacturing line, with no previous proof firing. It was noted that the “breech face impression … change considerably shot-to-shot [in early firings] due to the paint wearing/chipping off.” 3–E COMMENTARY As detailed in Chapter 1, it is not the function of this committee to assess the general validity of firearms identification and toolmark examination nor their admissibility in court proceedings. The discussion in this chapter on the nature of toolmark evidence and the context in which it is applied, as well as an overview of existing research among firearms examiners on the uniqueness and reproducibility of toolmarks, is presented to frame the discussion in the rest of this report. For instance, understanding situations that may pose particular challenges for associating two images of evidence requires some knowledge of situations that are generally known to be complex in the field; likewise, recommendations for the setup and maintenance of any ballistic imaging system that would do harm to the maintenance of clear chain of custody—so important in the legal context of toolmark evidence—would be ill-advised. However, as we also note in Chapter 1, we understand that some readers may try to infer a position—a leaning, one way or the other—based on the preceding analysis. Accordingly, we believe it important to make the committee’s finding clear and unambiguous: Finding: The validity of the fundamental assumptions of uniqueness and reproducibility of firearms-related toolmarks has not yet been fully demonstrated. There is one baseline level of credibility, however, that must be demonstrated lest any discussion of ballistic imaging be rendered moot—namely, that there is at least some “signal” that may be detected. In other words, the creation of toolmarks must not be so random and volatile that there is no reason to believe that any similar and matchable marks exist on two exhibits fired from the same gun. The existing research, and the field’s general acceptance in legal proceedings for several decades, is more than adequate testimony to that baseline level. Beyond that level, we neither endorse nor oppose the fundamental assumptions. Our review in this chapter is not—and is not meant to be—a full weighing of evidence for or against
OCR for page 82
Ballistic Imaging the assumptions, but it is ample enough to suggest that they are not fully settled, mechanically or empirically. Another point follows directly: Additional general research on the uniqueness and reproducibility of firearms-related toolmarks would have to be done if the basic premises of firearms identification are to be put on a more solid scientific footing. A designed program of experiments covering a full range of sources of variability is important to test the fundamental assumptions, as well as to better document phenomena like “settle-in” effects. In such a program, it could be useful to study the level of agreement of marks generated by the whole system of parts that make up a firearm, rather than treating each mark type in isolation. For example, in a large number of test firings, how comparable is the quality of breech face marking with firing pin impressions, and how do those compare with the clarity of striations etched on bullets? Fully assessing the assumptions underlying firearms identification would require careful attention to statistical experimental design issues, as well as intensive work on the underlying physics, engineering, and metallurgy of firearms, but is essential to the long-term viability of this type of forensic evidence. A third point is important in reading this report—stopping short of commenting on whether firearms toolmark evidence should be admissible: Conclusions drawn in firearms identification should not be made to imply the presence of a firm statistical basis when none has been demonstrated. Specifically, as described in Section 3–B.4, examiners tend to cast their assessments in bold absolutes, commonly asserting that a match can be made “to the exclusion of all other firearms in the world.” Such comments cloak an inherently subjective assessment of a match with an extreme probability statement that has no firm grounding and unrealistically implies an error rate of zero. Thornton and Peterson (2002:24–25) note the basic flaw in this reasoning: Since the basis of all forensic identification is probability theory, examiners can never really assert a conclusion of an “identification to the exclusion of all others in the world,” but at best can only assert a very small (objective or subjective) probability of a coincidental match…. It is ironic that those areas of forensic science that have real underlying data offer more modest statements of individualization, while those limited to subjective or impressionistic data make the strongest statements, sometimes of absolute certainty. As described in Box 3-4, recent court decisions in which admissibility of firearms toolmark evidence was in question have generally accepted that the field has validity but have refused to accept “exclusion of all other firearms” arguments. The committee agrees with the basic point: statements
OCR for page 83
Ballistic Imaging BOX 3-4 Recent Court Decisions: “To the Exclusion of All Other Guns” As part of a larger trial on racketeering, assault, and gun charges, Judge Nancy Gertner of the U.S. District Court for Massachusetts rejected a motion to exclude ballistics evidence but strictly limited the scope of the testimony (United States v. Green, 405 F.Supp. 2d 104; 2005 U.S. Dist. LEXIS 34273). Fourteen .380 caliber shell casings were recovered from two sites in September 2000, six of them at the site of a shooting; about a year later, a loaded .38 caliber pistol was found in the front yard of a residence. As part of the trial, prosecutors wanted to introduce testimony from a Boston Police Department sergeant that all of the casings came from the same gun, namely, the recovered pistol. At a preliminary hearing, the sergeant indicated that he could make this determination “to the exclusion of every other firearm in the world.” Gertner’s opinion noted that the “exclusion of every other firearm in the world” claim was “needless to say, extraordinary, particularly given [the sergeant’s] data and methods”: the examiner “took no notes, recorded no measurements, made no photographs, and drew no diagrams.” After reviewing toolmark issues in some detail, the court concluded: [The examiner in question] is a seasoned observer of firearms and toolmarks; he may be able to identify marks that a lay observer would not. But while I will allow [him] to testify as to his observations, I will not allow him to conclude that the match he found by dint of the specific methodology he used permits “the exclusion of all other guns” as the source of the shell casings. Separate rulings in United States v. Monteiro—also from the U.S. District Court for the District of Massachusetts, and coming before and after the Green ruling—adopted similar stances on testimony “to the exclusion of all other firearms.” Defendants in Monteiro challenged the specific examiner (a Massachusetts State Police sergeant) as not being qualified in firearms identification; challenged firearms identification as unreliable under Daubert; and asserted that—even if firearms identification were reliable—that the examiner did not apply the techniques properly. The defense also challenged the validity of any identification because the examiner replaced—“among other parts”—“the firing pin, recoil spring, barrel, and trigger lever (but, significantly, not the breech face)” of the gun in order to get it back into firing condition. Prior to Green, Judge Patti Saris ruled on a motion to exclude ballistics evidence (2005 U.S. Dist. LEXIS 39062), allowing testimony in part but giving the government “two weeks to ensure that its proffered firearms testimony comported with the established standards in the field.” As in Green, no notes or photographs were made on the identification: Saris ruled that “until the basis for the identification is described in such a way that the procedure performed by [the sergeant] is reproducible and verifiable, it is inadmissible under Rule 702.” Saris further
OCR for page 84
Ballistic Imaging directed that the identification be subjected to independent review and verification in order to be admissible. Judge Saris’ subsequent ruling on the general challenge to firearms identification (407 F.Supp.2d 351, 2006) concluded that “the underlying scientific principle behind firearm identification—that firearms transfer unique toolmarks to spent cartridge cases—is valid under Daubert. However, the process of deciding that a cartridge case was fired by a particular gun is based primarily on a visual inspection of patterns of toolmarks, and is largely a subjective determination based on experience and expertise.” The court ruled that the government must demonstrate basic standards for the qualification of the examiner. However, like Judge Gertner, Saris precluded—in any event—any testimony of a match to the exclusion of all other guns in the world. Saris noted that “examiners testified to the effect that they could be 100 percent sure of a match,” a statement that could not be sustained. An examiner may “testify to a reasonable degree of ballistic certainty,” but “an expert may not assert any degree of statistical certainty, 100 percent or otherwise, as to a match.” Most recently, in United States v. Diaz (2007 U.S. Dist LEXIS 13152), Judge William Alsup of the U.S. District Court for the Northern District of California accepted that: the theory of firearm identification … is reliable under Daubert. While there is some subjectivity involved, it is the subjective judgment of trained professionals with a keen practiced eye for discerning the extent of matching patterns.… The record, however, does not support the conclusion that identifications can be made to the exclusion of all other firearms in the world. Thus, the examiners who testify in this case may only testify that a match has been made to a ‘reasonable degree of certainty in the ballistics field. In defense of firearms identification, Alsup remarked: It is important to note that—at least according to this record—there has never been a single documented decision in the United States where an incorrect firearms identification was used to convict a defendant. This is not to say that examiners do not make mistakes. The record demonstrates that examiners make mistakes even on proficiency tests. But, in view of the thousands of criminal defendants who have had an incentive to challenge firearms examiners’ conclusions, it is significant that defendants cite no false-positive identification used against a criminal defendant in any American jurisdiction.
OCR for page 85
Ballistic Imaging on toolmark matches (including legal testimony) should be supported by the work that was done in the laboratory, by the notes and documentation made by examiners, and by proficiency testing or established error rates for individual examiners in the field and in that particular laboratory, but should not overreach to make extreme probability statements. 3–F ROLE OF IMAGING IN FIREARMS IDENTIFICATION In the next several chapters, we explore the current state of ballistic imaging technology. As context for this discussion, we note that imaging and photography have a long and somewhat controversial history in traditional firearms identification. Moran (2003) summarizes some of the historical debate over the use of comparison photographs, culling relevant quotations from source materials. Some of the pioneers of the field of firearms identification—Goddard, Burrard, and Hatcher—considered photography to be valuable, if not essential, Burrard going so far as to comment that “any evidence unsupported by photographs cannot be regarded as being anything more than an expression of opinion. Photographs are, accordingly, essential; and such as are deemed necessary must be taken through the microscope” (Moran, 2003:175). However, Hatcher sounded a note of concern: “There is a difference in the ability of the various experts to use the microscope and camera, so that in the hands of a very skilled operator they may show the correspondence or lack of correspondence very clearly, while in the hands of a poor or mediocre operator, they may show the same thing faintly, or may even fail to show them at all” (Moran, 2003:176). However, with the passage of time, the practice of using photographs to document identifications fell out of favor, so much so that a 1957 revision of Hatcher’s 1935 text now stipulated that “photo micrographs are now rarely used,” for a variety of reasons (Moran, 2003:177): Courts tended to accept examiners’ testimony on identifications without the “visual proof,” obviating the need to prepare the photographs. Preparation of the photographs took time, time that laboratories were unwilling to commit due to increased caseloads. Static views of an evidence match were deemed unsatisfactory, relative to the full range of panning and rotation possible during direct manipulation of the evidence. “These pictures were not understood by juries,” and “a good deal of knowledge and experience are necessary to evaluate them.” “Some men after years of working in Firearms Identification refuse to make a positive identification from pictures alone.”
OCR for page 86
Ballistic Imaging Photographs depict striations and features that do not match as well as those that do, which could create doubt among jurors. In subsequent decades, the use of photographs as part of documentation remained a matter of personal and agency tastes, with some favoring it and others opposing it: for example, Heard (1997:113) stated that “the use of comparison photomicrographs in a court of law to illustrate stria comparisons should be discouraged” (although he suggested that a video might be more informative). Ultimately, in 2005, the AFTE membership approved a “standardization of comparison documentation” stipulating that “photography is the preferred method of documentation” but noting that other forms of documentation (including “narrative descriptions”) “may serve to satisfy the requirements of this standard.” This standard, revised as of June 13, 2005, was promulgated in a 2006 issue of the AFTE Journal (Vol. 38, No. 1). Prior to the use of automated ballistic imaging, law enforcement agencies such as the Los Angeles Police Department (LAPD) relied on cruder means—and a certain amount of luck—to make connections with evidence in their open case files. Quite literally, making those connections depended on reference to Polaroid photomicrographs, posted on a bulletin board or shared with colleagues, in order to jog memories and generate possibilities. As of 1994, the process used by the LAPD was to post Polaroid images on two walls in “a hallway that we pass through constantly. On one side are the ‘unknowns,’ like cartridge cases collected as evidence at a homicide scene. On the other side are ‘knowns.’ When we recovered a gun from a suspect or a crime scene, we test-fire it in the lab—that makes it a known.… When we get a new known, we compare it to all the unknowns. When we get a new unknown, we compare it to all the knowns and all the unknowns,” and the pictures are added to the hallway for future reference (Maruoka, 1994b:214).16 The next chapter begins to describe efforts to go beyond Polaroids in the hallway: to use computer imaging to make it possible to more readily draw connections with exhibits in other open cases.17 16 The Maruoka article is, in fact, a reprint in the AFTE Journal of a profile written for the Polaroid Corporation’s Instant Evidence newsletter for law enforcement officials, circulated in 1993. As such, it touts the specific Polaroid equipment and film used for the standard imaging. 17 As an example of the way a large-scale search of exhibits is conducted using conventional methods, without imaging assistance, Grooß (1995:29, 30) describes a series of three murders in West Germany in 1984–1985. Cartridge cases from the second and third crimes were easily matched by examiners but the cases from the first crime bore only limited similarity in markings to the others. Still, the headstamp information on the casings from all three crimes, and the bullet evidence, strongly suggested that a member of the police might be the killer. So strong was this suspicion to investigators that all 7,862 Walther P5 duty pistols then in use by the police forces throughout the state of Baden-Württemberg were test fired and the cartridges
OCR for page 87
Ballistic Imaging compared with the crime scene exhibits. “As the comparison work in this specific case [was done] using conventional microscopes,” it “was very complicated and time-consuming”; up to 3 firearms examiners at a time worked continuously on the comparisons. Ultimately, test-fired casings from the 3,704th pistol in sequence matched the casings from the second and third murders; this led investigators to an officer in Stuttgart who was ultimately found dead in southern Italy, having also murdered his wife and sons. Grooß concluded that the experience was strong testament to “the individuality of marks on fired bullets and cartridge cases”: that the German examiners were able to observe “marks … which left no doubt that they were identical to those observed on the evidence ammunition,” against a backdrop of “approximately 4000 pistols of the same manufacturer, same model, approximately the same age and same degree of wear.” Apparently, the comparisons with test-fired exhibits were halted with the positive result on the 3,704th pistol, even though the linkage to the casings from the first murder was unclear. Even when the officer’s pistol was recovered and test fired with a variety of ammunition brands, no casing could be generated that matched the evidence from the first murder. It was originally suspected that the pistol “got a new blue finish” (a refurbishing process that may affect the marks left by the gun) at a time between the first and second murder, which might explain the differences. However, “a more careful examination showed that the pistol had been blued in the period between the second and third murder” (why this did not impede the ability to match the second and third crimes is left unspecified). It was also speculated that the officer might have deliberately planted a different casing at the first crime scene to mislead detectives.
OCR for page 88
Ballistic Imaging This page intentionally left blank.