Presentations from the 2nd International Symposium on the Forensic Examination of Questioned Documents (Part 6; Forensic Science Communications, October 1999)
July 2000 - Volume 2 - Number 3
Presentations at the
2nd International Symposium on the Forensic Examination of Questioned Documents
Albany, New York
June 14 – 18, 1999
The following abstracts of the presentations are ordered alphabetically by authors’ last names.
A state agency used a proctored, twenty-question essay test to determine if an individual should receive a license to practice his profession. The test consisted of ten test booklets with two answers per booklet. One test grader noticed that the second booklet from one individual contained two different styles of handprinting, one for each of the two questions. One handprinting style was consistent with that found in the other nine test booklets, and the other was distinctly different. All ten of the booklets were submitted for examination in order to determine if there was a common author. Differences in pictorial appearance, spacing, size, use of upper- and lower-case letters, and connecting strokes were found. A lack of many comparable letters and letter combinations (particularly upper- and lower-case letters) resulted in an opinion that there were indications that the questioned text and the remaining material were not written by the same person.
The subject was asked to complete a handwriting exemplar and to bring in nonrequest examples of his handprinting. At the start of the session, the subject stated, “I know what I did. I wrote the way I used to.” For inarticulated reasons, the subject had changed his style of handprinting and, under the stress of the test, had reverted to his old-style of handprinting. The request exemplars were taken in both the old-style and new-style handprinting and compared to the questioned material and nonrequest exemplars. It was concluded that all of the material had been written by the same person.
Old style and new style handprinting. Baselines are approximately the same length.
When comparable material letters and letter combinations were obtained in the request exemplars, it was found that the old-style and new-style handprinting contained some similarities. These consisted of some letters and letter combinations (particularly the connecting strokes of letters with a cross bar (i.e., “T”, “t”, “F”, and “f”). It was also noted that some letter formations were clearly different. When the nonrequest exemplars of the old-style and new-style handprinting were compared, it was noted that an occasional old-style letter or letter combination was found in the new-style handprinting. No new-style handprinting habits (that were exclusive to the new-style) were found in the old-style nonrequest handprinting. When the requested old-style and new-style handprinting were compared, it was found that an occasional letter or letter combination of one style was found in the other. This appeared to be a function of consciously trying to print first one style and then the other, a form of cross-fertilization.
It became apparent that if an examiner were confronted with questioned material of one style of handprinting and exemplars of another style, both from this individual, that a definitive opinion would not be possible.
The Crime Laboratory Accreditation Program, established by the American Society of Crime Laboratory Directors (ASCLD), is a voluntary program in which any crime laboratory may participate to demonstrate that its management, operations, personnel, procedures, equipment, physical plant, security, and health and safety procedures meet established standards. The program is managed by the American Society of Crime Laboratory Directors/Laboratory Accreditation Board (ASCLD/LAB), which is responsible to the Delegate Assembly composed of the directors of all accredited laboratories. Accreditation is part of a laboratory’s quality-assurance program, which should also include proficiency testing, continuing education, and other programs to help the laboratory give better overall service to the criminal justice system. The process of self-evaluation, which leads to accreditation, is in itself a valuable management tool for the crime laboratory director.
ASCLD/LAB has adopted the following four objectives that define the purposes and nature of the program:
- Improve the quality of laboratory services provided to the criminal justice system.
- Develop and maintain criteria that can be used by a laboratory to assess its level of performance and to strengthen its operation.
- Provide an independent, impartial, and objective system by which laboratories can benefit from a total operational review.
- Offer to the general public and to users of laboratory services a means of identifying those laboratories that have demonstrated that they meet established standards.
The following underlying questions are asked during an inspection to determine the quality of the laboratory:
- What is it you say you are doing?
- Are you doing what you say you are doing?
- Have you documented what you say you are doing?
After accreditation has been granted, the principal means by which ASCLD/LAB monitors compliance are an annual review report filed by the laboratory director and proficiency-testing reports submitted by approved test providers. This presentation is a discussion and review of the accreditation process.
More than 320 historic documents were placed on the market by Lawrence X. (Lex) Cusack III, a former paralegal, and two of his associates, John Reznikoff and Thomas Cloud. The documents supported the dark stories and rumors that had circulated regarding President John F. Kennedy, Marilyn Monroe, Robert Kennedy, Sam Giancana, J. Edgar Hoover, and many others. These documents, supposedly found by Cusack in his late father’s New York law offices, were sold for an amount between $6 million and $7 million to approximately 160 investors. Seymour Hersh wrote a book, and ABC and Lancer Productions were preparing a two-hour special using the documents as a core portion of their work.
Although a number of document dealers viewed and authenticated portions of the documents, Hersh, ABC, and Lancer decided to hire a forensic document examiner. They became suspicious after examining photographs of some of the documents. Arrangements were made to examine a few of the originals, and it was determined that they were not genuine. Although some of the signatures were found to be forgeries and most of the others probably forgeries, the real keys were the typewriting and a printed letterhead. Most of the documents were dated between 1960 and 1963. It was noted by Ed Gray of Lancer that one of the 1961 letterheads had a ZIP code. This was problematic because ZIP codes were not implemented until July 1963. In addition, the forensic document examination determined that the type style, the ability of the type to be spaced at two different widths on the same document, and the typewriter ribbon and correcting method were not introduced to the market until early 1970.
On the basis of this information, the Hersh book was revised, the ABC/Lancer TV production was edited, two TV specials were developed (ABC’s 20/20: The JFK/Monroe Papers and CBS’s 60 Minutes: The JFK Papers), and Lex Cusack was indicted by a New York federal grand jury. On April 30, 1999, Cusack was found guilty on 13 counts of selling forged documents. Cusack was sentenced on July 30, 1999.
The American Board of Forensic Document Examiners (ABFDE) was established in 1977 and was initially funded by a Law Enforcement Administrative Act grant. The Board is a joint project of the American Society of Questioned Document Examiners and the document section of the American Academy of Forensic Sciences. Its mission is to promote the advancement of the field of forensic document examination.
The primary purpose of the ABFDE is to act as a certifying body. As Professor Kam pointed out in his presentation, one of the characteristics of a profession as opposed to an occupation is the existence of a self-regulatory body. The testing conducted by the ABFDE is intended to establish a comprehensive minimal level of proficiency in the field of document examination.
The Board is a voluntary, self-funding body. There are currently 155 active Diplomates of the American Board of Forensic Document Examiners. The Board of Directors is composed of eleven individuals. The composition of the Board reflects the diversity of the field.
Certification is granted for a five-year period. Individuals seeking certification must be of good character; reside in the United States, Canada, or Mexico; and have a bachelor’s degree from a recognized university. They must also have two years of comprehensive training at a recognized private or public laboratory, three references who attest to their training and character, and be actively engaged as a forensic document examiner.
Qualified individuals may apply for certification upon completion of their training program. Individuals are expected to be recertified by demonstrating ongoing professional activity in the field.
The entire process of certification can take up to two years. First, the background and references are verified. The individual then is expected to successfully pass written, practical, and oral examinations.
Certification by the ABFDE has gained recognition as a meaningful credential in federal and state courts, as a requirement for employment in some laboratories, and as a qualification for membership in professional organizations (e.g., the American Society of Questioned Document Examiners).
The Board is composed of an executive committee and eight active subcommittees. The subcommittees are involved in evaluating and testing applicants, preparing tests, publishing monographs and pamphlets of interest to document examiners, and monitoring changes in the field (e.g., the proposed Forensic Accrediting Board). The ABFDE also has a mechanism for handling ethics complaints. This last function is an unpleasant but necessary element of a meaningful certifying body.
Since its creation in 1977, a significant percentage of professionals in the field of forensic document examination have served in various capacities on the Board. The Board strives to be a meaningful and professional representative of the field of forensic document examination.
Courtroom Computer Graphics, Animation, and Interactive Multimedia
K. S. Runyon
Minneapolis Police Department
D. W. Davis
Expert Digital Solutions, Incorporated
This presentation demonstrated how the judicious use of computer graphics can aid medical expert testimony and how trial attorneys can enhance their effectiveness by using an interactive multimedia presentation in the closing argument.
- Augment expert testimony with computer-generated/enhanced graphics so that juries can clearly understand the complicated injury patterns in adults and children.
- Become aware of what is generally acceptable and unacceptable to use for demonstrative computer evidence.
- Understand that interactive multimedia has significant potential to aid the trial attorney in presenting a cohesive and compelling closing argument.
Forensic pathologists frequently testify in homicide trials. Some pathologists prepare digital or computer graphics for use in the courtroom to support their testimony.
The term graphics refers to evidence consisting of color autopsy photographs with added highlights (e.g., circles, arrows, boxes, labels, cropping, overlays, insets, and scaled comparisons). Two-dimensional graphics and three-dimensional models can also be used as demonstrative evidence for explaining injury patterns (e.g., child abuse injuries, bruises, and knife and gunshot wounds). Computer-generated 35-mm slides can be projected onto a screen. Attorneys have had little or no trouble admitting computer-generated graphics into court, and in most cases, judges and juries find carefully prepared graphics informative and a tasteful alternative to autopsy photographs only.
Interactive multimedia presentations have been developed from computer-generated graphics by prosecutors to deliver as part of the closing argument in homicide trials, making use of key pieces of evidence presented in the trial. Evidence may consist of crime scene and autopsy photographs, witness accounts, defendant interviews, 911 audios, documents, surveillance videos, or other media. The closing argument presentations may include text, graphics, audio, and video. Some cases require scene animations. Using the evidence, the prosecutor is able to vividly depict the crime scenario and reinforce key elements of the argument to establish the defendant’s guilt. The presentations are given from a portable computer attached to a video/data projector and projected onto a screen in front of the jury. Audio from the computer connects directly into the courtroom’s public address system or speakers. The prosecutor advances through the presentation using an infrared remote control that actuates predetermined effects on the computer. Prosecutors who have used the technology think that the computer presentations give them a distinct advantage in difficult or complex cases.
The purpose of the Scientific Working Group for Questioned Document Examination (SWGDOC) is to assemble forensic document examiners, academians, and legal professionals to perform the following tasks:
- Define the scope and practice areas of the profession;
- Standardize operating procedures, protocols, and terminology;
- Consolidate and enhance the profession of forensic document examination; and
- Promote self-regulation, documentation, training, continuing education, and research in the area of forensic document examination.
In January 1999, the Technical Working Group for Questioned Document Examination (TWGDOC) was renamed the Scientific Working Group for Questioned Document Examination (SWGDOC). The first TWGDOC meeting was in the spring of 1997 in Washington, DC. That meeting consisted of only federal laboratory representatives. Shortly after this meeting, the first subcommittee began standardizing operating procedures and terminology. In December 1997, the TWGDOC expanded to include regional and organizational representatives. The current composition of the SWGDOC is one committee with five subcommittees.
The main committee of the SWGDOC consists of eight representatives from United States federal laboratories, eight national and regional organizational representatives, five at-large representatives, the Coalition of Private Practice Examiners, and two academic representatives.
The five subcommittees are
- Standard Operating Procedures and Terminology,
- Proficiency Testing and Quality Assurance,
- Research and Validation Studies, and
- Legal Issues
In scientific inquiry it becomes a matter of duty to expose a supposed law to every possible kind of verification and to take care, moreover, that this is done intentionally, and not left to a mere accident.CHuxley 1974
The best analogy for thinking about handwriting “experts” may be the practitioner of folk medicine. Like folk medicine, handwriting identification may sometimes be efficacious; but no verification yet exists of when, if ever, it is and when it is not.CRisinger et al. 1989
The law has been moving toward a standard under which asserted handwriting identification expertise, with its existing foundations, may cease to be admissible as evidence. Under the Federal Rules, the standard for admission of scientific expert testimony is that the court must be rationally convinced of its validity (Daubert v. Merrell Dow Pharmaceuticals, Inc.). “Were the court to apply Daubert to . . . forensic document examinations . . . it would have to be excluded” (United States v. Starzecpyzel). Some courts concluded that because forensic document examination was not a science and did not employ scientific methods, it was not subject to Daubert. Recently, however, the Supreme Court rejected that distinction, requiring that all expert evidence, science and nonscience alike, satisfy basic validity requirements (Kumho Tire Co. v. Carmichael). This apparently would have become the standard even if the Supreme Court had not so held, because that is what is required under the revision to Federal Rule of Evidence 702. Moreover, because each district judge has the duty to screen expert evidence in every appropriate case, without deference to precedent (General Electric Co. v. Joiner), forensic document examinations will be scrutinized repeatedly. In at least one case, the opinions of forensic document examiners already have been excluded (United States v. McVeigh).
The only way to know whether an empirical claim is valid or not is to subject it to careful empirical testing. The essential feature of those enterprises regarded as sciences is that they engage in extensive, systematic empirical testing of their ideas. Scientists recognize that a finding of fact is only as good as the methods used to find it. The benefit of empirical testing is that it has a powerful capacity to separate true propositions from false ones and thereby to allow us to discard incorrect beliefs and ineffective practices and inventions and to continually replace them with more correct and more effective ones. When fields undertake empirical testing they typically discover that they had been subscribing to many beliefs that in the light of data need to be modified or discarded. Consequently, fields which conscientiously use the results of such testing inevitably stand on firmer ground than those which do not.
The State of Forensic Document Examination Validation
Forensic document examination has no tradition of subjecting its hypotheses and claimed skills to systematic empirical tests. Although virtually any field that claims to be a science, or to be working in a scientific manner, can produce hundreds if not thousands of systematic empirical studies bearing on their field’s subject matter, forensic document examiners can produce almost none (Kam et al. 1994; Galbraith et al. 1995; Moenssens et al. 1998; Risinger et al. 1989). The limited research that does exist raises at least as many concerns as it dispels. The Forensic Sciences Foundation (FSF) studies found considerable differences of opinion and numerous erroneous conclusions made by forensic document examiners. Performance varied from quite good to quite poor depending on the task. The Galbraiths’ (1995) further analyses found that on one-third of the FSF tests, document examiners were no more accurate than would be expected by random guessing and that nonexperts were equally good at identifying true positives, though not as good at avoiding false positives. Kam (1994) found that, overall, experts outperformed nonexperts, but whether this difference was because of differences in skill or in motivation awaits further research. Even so, the distribution of nonexperts’ performance showed extreme bimodality such that the best nonexperts did about as well as the experts. Kam (1997) attempted with questionable success to equalize motivation between experts and nonexperts. Still, the study found nonexperts to be as good as experts at identifying true positives but not as good as experts at avoiding false positives. Kam (1998) made another attempt to test the impact of different reward schemes for nonexperts. On the same task as before, under a variety of reward conditions, this study found an overall 41 percent improvement in the ability of nonexperts to avoid false-positive errors. By any measure, the research has only begun to evaluate the skills claimed by forensic document examiners, whereas suggesting that many of the abilities usually claimed are not supportable.
Agendas for Research
A program of serious, systematic empirical research needs to be launched and sustained, subjecting everything that is important and testable to every possible kind of verification. At least three different and mutually compatible research agendas are possible.
The fundamental propositions on which the possibility of accurate handwriting identification depends need to be tested. Forensic document examiners accept as true a large number of assumptions about writing and its identification (see the list compiled by Risinger in Faigman et al. 1997). These are hypotheses that need to be tested so that the valid ones can be retained and the invalid ones replaced.
Objective Probability Basis
A second agenda would follow the model of DNA typing. Gather adequate writing samples from the population at large, measure important dimensions of variation, determine the relative frequency of attributes, and use that database in casework to calculate the actual probability of coincidental (false-positive) matches. Stop intuiting. Start measuring, computing, and reporting.
Subjective Probability Basis
Examiners could continue to serve as black-box, intuitive, and subjective human machines for writer identification if the reliability and validity of their judgments were tested systematically. Such research could determine how well or poorly examiners perform in relation to different types of writing, in different amounts, across the range of writing situations, comparing examiners of different training and backgrounds, and using different examination protocols and procedures. This would permit an empirical mapping of the topography of circumstances of handwriting, handwriting examiners, and handwriting identification practices. Examiners and the courts would then know what can be done well, what not so well, and what not at all.
Barriers to the Successful Pursuit of a Research Agenda
Formidable barriers stand in the way of implementing any research agenda in and on forensic document examination. Few if any forensic document examiners know how to design, conduct, or analyze empirical research. They are not affiliated with researchers who are their cognate partners who could do the research for them and with them (as medical researchers do for clinical physicians). There exists no tradition of continual questioning, doubting, testing, and rethinking of beliefs. There are no sources of funds waiting to support an ongoing research program in forensic document examination. Many of these barriers will be all the more difficult to overcome because of the culture of litigation in which forensic document examiners have always been firmly rooted. In contrast to the instincts bred by the litigation culture, the science culture thrives on the following:
- Relentless pursuit of truth, even if it will refute past beliefs and reveal limitation.
- Routine sharing of ideas, problems, studies, and data.
- Open and public discussion of ideas, studies, and data, whether flattering or embarrassing.
- An eagerness, not merely a willingness, to expose research to thoughtful criticism.
Finally, as professionals, forensic document examiners will have to keep themselves informed about research findings as they emerge, adverting to sound findings even when they contradict long-cherished beliefs, and incorporating them into what most likely will become a continually revised understanding of the field.
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
Faigman, D., Kaye, D. H., Saks, M. J., and Sanders, J. (eds.). Modern Scientific Evidence: The Law and Science of Expert Testimony. West, St. Paul, Minnesota, 1997 and 1999 Supplement.
Galbraith, O., Galbraith, C. S., and Galbraith, N. G. The principle of the “drunkard’s search” as a proxy for scientific analysis: The misuse of handwriting test data in a law review article, International Journal of Forensic Document Examiners (1995) 1:7-17.
General Electric Co. v. Joiner, 522 U.S. 136 (1997).
Huxley, T. H. The method of scientific investigation. In: Science: Method and Meaning. eds. S. Rapport and H. Wright, Washington Square Press, New York, 1964.
Kam, M., Wetstein, J., and Conn, R. Proficiency of professional document examiners in writer identification, Journal of Forensic Sciences (1994) 39:5-14.
Kam, M., Fielding, G., and Conn, R. Writer identification by professional document examiners, Journal of Forensic Sciences (1997) 42:778-786.
Kam, M., Fielding, G., and Conn, R. Effects of monetary incentives on performances of nonprofessionals in document-examination proficiency tests, Journal of Forensic Sciences (1998) 43:1000-1004.
Kumho Tire Co. v. Carmichael, ___U.S.___, 119 S.Ct. 1167 (1999).
Moenssens, A. Handwriting identification evidence in the post-Daubert world, UMKC Law Review (1997) 66:251-343.
Risinger, R. M., Denbeaux, M. P., and Saks, M. J. Exorcism of ignorance as a proxy for rational knowledge: The lessons of handwriting identification expertise, University of Pennsylvania Law Review (1989) 137:731-792.
Risinger, R. M. and Saks, M. J. Science and nonscience in the courts: Daubert meets handwriting identification expertise, Iowa Law Review (1996) 82:21-74.
Risinger, R. M., Denbeaux, M. P., and Saks, M. J. Brave new post-Daubert world: A reply to Professor Moenssens, Seton Hall Law Review (1997) 29:405-490.
United States v. McVeigh, 1997 WL 47724 (D.Colo.Trans.).
United States v. Starzecpyzel, 880 F.Supp. 1027, 1036 (S.D.N.Y.1995).