CRIM. NO. 97-CR-10336-NG


June 11, 1999

This case raises questions concerning the application of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) and Kumho Tire Co. v. Carmichael, - - U.S. - -, 119 S.Ct. 1167 (1999) to technical fields, that are not, strictly speaking, science. Two fields are involved: The first is an "old" field, handwriting analysis, which has been the subject of expert testimony for countless years. The second is a comparatively "new" field, the psychology of eyewitness identification. The government maintained that handwriting analysis is "science," meeting the Daubert and Kumho tests, while the psychology of eyewitness identification is not. Not surprisingly, the defendant insisted that the opposite is true.

Johannes Hines ("Hines") is charged under 18 U.S.C. §2113 for allegedly robbing the Broadway National Bank in Chelsea, Massachusetts on January 27, 1997.1 The government's principal evidence consisted of the eyewitness identification of the teller who was robbed, Ms. Jeanne Dunne, and the handwriting analysis of the robbery note.2 In connection with the latter, the government offered Diana Harrison ("Harrison"), a document examiner with the Federal Bureau of Investigations, to testify as to the authorship of a "stick-up" note found at the scene of the crime.

Hines sought to exclude the handwriting analysis. This testimony, defense claims, notwithstanding its venerable history, does not meet the standards of Daubert and Kumho. In the alternative, if the court permitted the jury to hear the handwriting testimony, Hines sought to have his expert - -Professor Mark Denbeaux ("Denbeaux") - - testify as to the weaknesses of the methodology and the basis of Harrison's conclusions. The government, on the other hand, argued for its handwriting expert under the applicable tests, and rejects Denbeaux.

Hines also offered the testimony of Dr. Saul Kassin ("Kassin"), a psychologist who studies perception and memory, and who has been qualified as an eyewitness identification expert in other cases. The government opposed. Should I allow the testimony of either defense's handwriting analysis critic or eyewitness expert, the government seeks counter experts, Dr. Ebbe B. Ebbesen ("Ebbesen"), on eyewitness identification, and Dr. Moshe Kam ("Kam") on handwriting.

With respect to handwriting analysis, I held a Daubert/Kumho hearing at which Denbeaux testified, and the government cross examined. In addition, the government offered several articles as well as the curriculum vitae of its expert, Kam. Harrison did not testify, as she had previously testified during the first trial, but I reviewed her prior testimony. With respect to eyewitness identification, the matter was briefed and argued. After I indicated my intention to admit the testimony, the government focused on the qualifications of Kassin, and the particular conclusions he would be drawing. At the eleventh hour, the government waived its motion for a hearing on this issue.

For the reasons set forth below, I DENIED Hines'motion to exclude handwriting analysis to the extent that it sought to exclude the handwriting expert's entire testimony. I GRANTED the motion in part: I did not permit the handwriting expert to make any ultimate conclusions on the actual authorship of the questioned writing. As a result of this ruling, counsel for Hines made the strategic decision not to call Denbeaux as an expert to critique handwriting analysis at trial. Thus, the government's motion to exclude Denbeaux's testimony (and offer Kam's) was MOOT.

I also DENIED the government's motion to exclude the testimony of Hines' eyewitness expert. Defense presented the testimony; the government presented its own witness.

This trial ended in a hung jury (as did the first). Since the issues noted in this memorandum will recur in the next trial, I outline my reasoning for the reasoning.


To some, Daubert/Kumho have considerably raised the bar for the admissibility of expert testimony. In some ways that is true; in some ways it is not. It is true that Daubert/Kumho have focused a great deal of attention on the judge's role as a gatekeeper for expert testimony under Fed.R.Evid. 104(a). To the extent that there is more pre-trial review of expert testimony, there are bound to be more exclusions. In other ways, however, the impact of Daubert/Kumho has been the opposite - - opening the door to testimony previously excluded.

Our evidentiary rules put a premium on firsthand observations. Opinion testimony is disfavored except under certain circumstances; hearsay is generally excluded.3 The jury is to draw reasonable inferences from the firsthand data. When an expert witness is called upon to draw those inferences, several concerns are raised. The rules give expert witnesses greater latitude than is afforded other witnesses to testify based on data not otherwise admissible before the jury.4 In addition, a certain patina attaches to an expert's testimony unlike any other witness; this is "science," a professional's judgment, the jury may think, and give more credence to the testimony than it may deserve.

Accordingly, the trial court is supposed to review expert testimony carefully. The court is to admit the testimony not only where it is relevant to the issues at bar, the usual standard under Fed.R.Evid. 401, but when certain additional requirements are met under Fed.R.Evid. 702.5

The first requirement has to do with the necessity for the testimony: expert testimony may be admitted where the inferences that are sought to be drawn are inferences that a jury could not draw on its own. The inferences may be the product of specialized information, for example, beyond the ken of the lay jury. Significantly, Daubert also emphasized the fact that expert testimony is admissible where it would "assist" the trier of fact. In the latter case, even if the inferences may be drawn by the lay juror, expert testimony may be admissible as an "aid" in that enterprise. See 509 U.S. at 591-592. For example, the subject looks like one the jury understands from every day life, but in fact, the inferences the jury may draw are erroneous. (As I describe below, eyewitness identification, and testimony about battered women syndrome fits uniquely into this category.)

The second requirement concerns the nature of the inferences to be drawn. In outlining the standards for admissibility, the Daubert Court noted the difference between information gleaned in a scientific setting and information presented in a courtroom.6 In the former, the decision makers are professionals; there is no need to come to a definitive conclusion; the decision making process comports with certain rules established by the professional scientific community. In the courtroom, the decision makers are lay, a jury; there is a need to come to a definitive conclusion; the decision making process has to satisfy norms of due process and fairness. In our tradition, for example, the adversary system and party examination and cross examination are central. The issue then, is not only how objectively reliable the evidence is, but also the legitimacy of the process by which it is generated. It is not just how valid the data is, but how well the jury can understand it after direct and cross examination, and legal instructions.

It was not surprising; then, that the Court found the rules for admissibility of expert testimony, even with respect to scientific fields, were flexible, bearing in mind their unique setting and purposes. See Daubert, 509 U.S. at 592. A trial judge could consider some or all of the following factors: (1) whether the expert's technique can be or has been tested; 2) whether the method has been subjected to peer review and publication; 3) the known or potential rate of error of a technique or theory when applied; and, 4) whether there is "general acceptance" within the relevant scientific community, to the extent that one exists. See id. at 593-95.7

Significantly, the "general acceptance test" of Frye v. United States, 54 App.D.C. 46, 293 F.1013 (1923), although now one of many standards, was still important. Daubert's principal focus was necessarily on new science. Its rules were intended to open the door to the admission of theories challenging existing orthodoxies. The petitioners, children born with serious birth defects, allegedly caused by their mothers' ingestion of Bendectin during pregnancy, sought to introduce scientific studies, performed by well credentialed medical experts, that contradicted the existing "generally accepted" theory that Bendectin was not a teratogen (a substance capable of causing malformations in fetuses). See id. at 582.8 Frye's "general acceptance" test was inadequate to deal with new theories and approaches. Having relinquished the Frye test, the Court was obliged to map other standards.9

But the standards articulated by the majority in Daubert raised serious concerns.10 The Frye test was perfectly suited to the court's competence. The judge did not have to be a scientist. He or she only had, in effect, to count the"scientific" votes; determine what the majority accepted. Daubert/Kumho, however, required a judge, a notable generalist, to second guess a scientist. Thus, by listing the "scientific" standards and encouraging review, the decision pointed in the direction of more rigor. But, by underscoring the uniqueness of the trial setting, the "assist the trier" standard and flexibility, the decision does the opposite.

One way of reconciling the competing vectors is this: Traditional science, generally accepted by judges for decades, may not need the same kind of rigorous analysis as "new" science which lacks the legitimacy of a chorus of sponsors. As to the latter, the Daubert analysis is critical, and the court should be an especially vigilant gatekeeper.

Kumho11 extended Daubert to non-scientific fields. In this - - category, for example, are the fields that are based on observations, not traditional science. See 119 S.Ct. at 1175 (finding that some experts, for example, as cited in Amicus Brief of Solicitor General at 1998 WL 541947, testify regarding "criminal modus operandi, land valuation, drug terms, agricultural practices," by relying upon other factors such as "personal knowledge or experience").

At the same time, the Court plainly recognized the limitations of the Daubert approach.12 It reaffirmed Daubert's holding that the court must be the gatekeeper, and must assure the reliability and relevance of testimony with respect to all expert testimony included in Fed.R.Evid. 702, not just scientific testing. See Kumho, 119 S.Ct. at 1174-76. But the nature of the test, notably, how much the Daubert standards are modified, was within the court's discretion. And if the Daubert standards were flexible when applied to traditional scientific fields, surely they were flexible when applied to nonscientific, technical fields. See id. at 1175-76 (factors in Daubert may or may not be pertinent in assessing reliability); see also Black v. Food Lion, Inc., 171 F.3d 308, 311-12 (5th Cir. 1999) ("[F]irst the [court] should decide whether the factors mentioned in Daubert are appropriate.")

Moreover, if the Daubert standard takes into account the unique trial setting - - how well the lay trier will understand the testimony after examination and instructions - - Kumho plainly does as well. In fact, cross examination and limiting instructions may be more effective in "technical" fields because they are more accessible to the jury, than fields with the charisma of science.

Again, a mixed message: Apply Daubert to technical fields, even though the scientific method may not really fit, but be flexible. Moreover, in this setting, because few technical fields are as firmly established as traditional scientific ones, the new science/old science comparison is less clear. The court is plainly inviting a reexamination even of "generally accepted" venerable, technical fields.

A. Handwriting

Handwriting analysis is one such field. The Harrison testimony may be divided into two parts: Part 1 is Harrison's testimony with respect to similarities between the known handwriting of Hines, and the robbery note. Part 2 is Harrison's testimony with respect to the author of the note, that the author of the robbery note was indeed Hines. I concluded that Harrison could testify only as to the former.

One qualification: This handwriting challenge was raised at the eleventh hour. The hearing was necessarily constrained by the demands of the imminent trial and the schedules of the experts. The Court was unwilling on this record to throw out decades of "generally accepted" testimony.13 Hines' challenge was far reaching. To be sure, it would not be the first time that a "field" was invented and legitimized only by those within it, that then dissolved when tested by independent standards. Moreover, the compromise solution I accepted, allowing testimony about similarities and dissimilarities, not conclusions, derived largely from the case law that pre-dated Kumbo. Finally, the trial testimony suggested that the line I drew between testimony on similarities and conclusions about authorship, may well be a difficult one to enforce. Hines, for example, objected to the government's closing on the ground that she implied that Harrison was a scientist, that her identification of similarities and dissimilarities between the Hines sample and the robbery note met that level of verification. Given the implications of the challenge, on retrial, I would be willing to review this again on a more full record, with amicus support, if possible.

Hines challenges Harrison's testimony under Daubert/Kumho. If I were to give special emphasis to "general acceptance" or to treat Daubert/Kumho as calling for a rigorous analysis only of new technical fields, not traditional ones, then handwriting analysis would largely pass muster. Handwriting analysis is perhaps the prototype of a technical field regularly admitted into evidence.14 But, if I were to apply the Daubert/Kumho standards rigorously, looking for such things as empirical testing, rate of error, etc., the testimony would have serious problems. See Starzecpyzel, 880 F.Supp. 1027, 1036 (S.D.N.Y. 1995) (finding that if the court had to apply Daubert to the preferred handwriting testimony, it would have to be excluded.)

According to Denbeaux,15 handwriting analysis by experts suffers in two respects. It has never been subject to meaningful reliability or validity testing, comparing the results of the handwriting examiners' conclusions with actual outcomes.16 There is no peer review by a"competitive, unbiased community of practitioners and academics." Starzecpyzel, 880 F.Supp. at 1038. To the extent that it has been"generally accepted," it is not by a"financially disinterested independent community, like an academic community," Id.; only other handwriting analysts have weighed in. It has never been shown to be more reliable than the results obtained by lay people. Some tests have been done, but all lacked a control or comparison group of lay persons. Thus, Denbeaux concludes, there is no need for expert testimony on handwriting analysis. Lay people can do as well. In this regard, I accept Denbeaux's testimony, and the article supporting it.

I do not believe that the government's expert, Kam, and the studies he has cited suggest otherwise).17 While Kam has conducted several interesting and important tests, purporting to validate handwriting analysis, they are not without criticism. They cannot be said to have"established" the validity of the field to any meaningful degree.

There is no question that lay witnesses are permitted to draw inferences of authorship from handwriting. Where the lay witness is familiar with the handwriting, he or she can testify about it. No expert testimony is necessary. But just because lay witnesses can evaluate handwriting in some circumstances, does not necessarily mean that they can do it as well as an expert can in all.

Handwriting analysis typically involves reviewing two samples, a known sample and an unknown one, to determine if they are similar. Both defense and government experts agree that unlike DNA or even fingerprints, one's handwriting is not at all unique in the sense that it remains the same over time, or uniquely separates one individual from another. Everyone's handwriting changes from minute to minute, day to day. At the same time, our handwriting is sufficiently similar to one another so that people can read each other's writing. Given that variability, the "expert" is obliged to make judgments - - these squiggles look more like these, these lines are shaped more like these, etc. And those judgments are, as Harrison conceded, subjective.

When a lay witness, the girlfriend of the defendant for example, says "this is my boyfriend's writing," her conclusion is based on having been exposed to her paramour's handwriting countless times. Without a lay witness with that kind of expertise, the government is obliged to offer the testimony of "experts" who have looked at, and studied handwriting for years. These are, essentially, "observational" experts, taxonomists - - arguably qualified because they have seen so many examples over so long.18 It is not traditional, experimental science, to be sure, but Kumho's gloss on Daubert suggests this is not necessary. I conclude that Harrison can testify to the ways in which she has found Hines' known handwriting similar to or dissimilar from the handwriting of the robbery note; part 1 of her testimony.

Part 2 of the Harrison testimony is, however, problematic. There is no data that suggests that handwriting analysts can say, like DNA experts, that this person is "the" author of the document. There are no meaningful, and accepted validity studies in the field.19 No one has shown me Harrison's error rate, the times she has been right, and the times she has been wrong. There is no academic field known as handwriting analysis. This is a "field" that has little efficacy outside of a courtroom. There are no peer reviews of it. Nor can one compare the opinion reached by an examiner with a standard protocol subject to validity testing, since there are no recognized standards. There is no agreement as to how many similarities it takes to declare a match, or how many differences it takes to rule it out.

Indeed, Denbeaux draws an interesting analogy to eyewitness identification. Courts have concluded, as a matter of law, that one-on-one show-ups are unduly suggestive. Likewise, Denbeaux suggests, are one-on-one handwriting comparisons. The outcome of this analysis, for example, may be different if Harrison were given a "lineup" of similar handwriting exemplars to review, and asked to determine which of this group is most similar to the robbery note author.20 Whether one adopts that specific approach or not, one thing is clear: when Harrison says, "I conclude that Hines wrote the robbery note," she may well be going beyond her expertise.

Moreover, the issue here is not only the validity and reliability of the expert testimony, but its validity and reliability in the context of this- -lay proceeding. Harrison's account of what is similar or not similar in the handwriting of Hines and the robber can be understood and evaluated by the jury. The witness can be cross examined, as she was, about why this difference was not considered consequential, while this difference was, and the jury can draw their own conclusions. This is not rocket science, or higher math.21

Her conclusion of authorship, however, has a difference resonance: "Out of all of my experience, and training, I am saying that he is the one, the very author." That leap may not at all be justified by the underlying data; and in the context of this case, is extraordinarily prejudicial.

The Court faced a similar issue in United States v. McVeigh, 1997 WL 47724 *3-4 (D. Colo. Trans.) (citing to Starzecpyzel, supra):

[T]here is a great difference between a witness who has the requisite training and skill saying, 'Look, I've compared this handwriting on this exhibit with this exemplar and I've used the techniques of microscoping (sic) and, you know, all of those things that are often involved in that kind of comparison, and these are the things I find,' and 'I see these similarities and these dissimilarities and so forth' but does not go on to reach any sort of ultimate conclusion that this was written by the same person or expresses some probability or degree of confidence.

The problem with . . . handwriting is that there is no testing of the - - no verification-type testing of these opinion results; and in addition, there has never been within the discipline of people who practice this skill - - there has never been any agreement on how to express the results. There is no standardized nomenclature, you know. Therefore, it seems to me that we should draw the distinction between somebody getting on the stand and saying 'Yeah, written by the same person,' or 'no, not written by the same person,' vs. 'these are the similarities or these are the dissimilarities'; and the jury can decide.

I find Harrison's testimony meets Fed.R.Evid. 702's requirements to the extent that she restricts her testimony to similarities or dissimilarities between the known exemplars and the robbery note. However, she may not render an ultimate conclusion on who penned the unknown writing. See Paul, 1999 WL 300884 at *4 (letting writing expert identify points of comparison that he recognized between the two writings and allowing jury to conduct its own comparison and reach its own conclusion regarding authorship of the two writings); see also McVeigh, 1999 WL 47724 at *20 (permitting handwriting expert to testify that two writings were consistent with each other but not permitting an ultimate conclusion as to authorship).

B. Eyewitness Identification

The government offered the testimony of Jeanne Dunne ("Dunne"), the teller. Dunne, a white woman, gave the the following identification moments after the robbery occurred: She identified the man as black with dark skin, a wide nose, and a medium build.22 Her description was as close to a generic identification of an African American man as one can imagine. Dunne was unable to identify Hines from a book of photographs of African American men shortly after the robbery.23 She picked out a few photographs, but none of them were as "dark black" as the robber. Working with a police artist, she helped construct a sketch of the robber. Immediately following, she was shown eight photographs, including one of Hines, and indicated that the Hines photograph "resembled" the robber, that it "looked like him," but she was still not sure. (Since the robber was wearing a hat, she tried to envision the man in the photograph with a baseball cap.) Months later Dunne picked Hines out of a lineup.

Hines offered the testimony of Kassin, a psychologist studying human perception at Williams College with substantial credentials, and trial experience. The government offered a similarly credentialed expert, Ebbesen. I allowed the testimony of both.

The Kassin testimony was offered to show, inter alia, the following: the decreased accuracy of cross-racial identification relative to same-race identification, the effect of stress on identification, the effect of time on memory as it relates to identification, the "confidence-accuracy" phenomenon which suggests the absence of any correlation between the amount of confidence expressed by an eyewitness in his or her memory and the accuracy of that witness' identification, the suggestiveness of subtle aspects of the identification process, such as the darkness of a particular photo as compared to others in the array, the fact that the eyewitness knows there is a suspect in the mix, the transference phenomenon by which a witness may believe that a face looks familiar but is unable to say whether her familiarity comes from seeing a previous mug shot, or from the robbery, etc.

On direct examination, Kassin identified those factors in the Dunne identification that were implicated in the studies with which he was familiar, and could undermine accuracy - - the cross racial issues, the differences between the photographs of the other men and Hines' in the photo array,24 the differences between Hines and the other men in the lineup.25 He noted problems with what he called relative, comparative judgments: The witness would like to resolve the case and so compares the photographs of one man to another in the array, rather than attempting to compare the photographs to the man she saw. On cross, the government brought up the factors that enhanced accuracy, the nature of the lighting, the distance from the robber, the instructions that were given to the witness, especially at the lineup. The government questioned Kassin about the instructions given to Dunne at the lineup. Kassin agreed that those instructions were "ideal."

Unlike handwriting analysis, there is no question as to the scientific underpinnings of Kassin's testimony. They are based on experimental psychological studies, testing the acquisition of memory, retention, and retrieval of memory under different conditions. Indeed, the central debate before the jury, eloquently articulated by Ebbesen, the government's expert, is the polar opposite of the debate in the handwriting field - - whether conclusions obtained in an experimental, academic, setting with college students should be applied to real life setting. See United States v. Norwood, 939 F.Supp. 1132, 1136 (D.N.J. 1996). Kassin and others believe that these conclusions are appropriately applied to eyewitness identifications in court. Ebbesen disagreed.

The government claimed that the jury did not need this testimony at all, that it was not necessary under Fed.R.Evid. 702 to assist the trier of fact. I disagree. While jurors may well be confident that they can draw the appropriate inferences about eyewitness identification directly from their life experiences, their confidence may be misplaced, especially where cross-racial identification is concerned. See id. at 1137; United States v. Smith, 736 F.2d, 1103, 1106 (6th Cir.), cert. denied, 469 U.S. 868 (1994) (such testimony would "not only 'surpass' common-sense evaluation, it would question common-sense evaluation."). Indeed, in this respect the rationale for the testimony tracks that for battered women syndrome experts. The jury, for example, may fault the victim for not leaving an abusive spouse, believing that they are fully capable of putting themselves in the shoes of the defendant. In fact, psychological evidence suggests that the "ordinary" response of an "ordinary" woman are not in play in situations of domestic violence where the victim suffers from "battered women syndrome." See e.g., People v. Day, 2 Cal. Rptr.2d 916, 924 (1992). Common sense inferences thus may well be way off the mark.

Nor do I agree that this testimony somehow usurps the function of the jury. The function of the expert here is not to say to the jury - - "you should believe or not believe the eyewitness." (Indeed, it has far fewer pretensions to conclusions than does handwriting analysis, with far more science attached to it.) All that the expert does is provide the jury with more information with which the jury can then make a more informed decision.26 And only the expert can do so. In the absence of an expert, a defense lawyer, for example, may try to argue that cross racial identifications are more problematic than identifications between members of the same race, or that stress may undermine accuracy, but his voice necessarily lacks the authority of the scientific studies Kassin cited.27

Finally, the fact that the expert has not interviewed the particular eyewitness makes it less likely that the jury will merely accept the expert testimony and more likely that the testimony will be appropriately cabined. The witness can only be providing the jury with the tools to analyze the eyewitness; he has no more specific information. The science makes no pretensions that it can predict whether a particular witness is accurate or mistaken.

In my judgment, the accuracy of these proceedings was enormously enhanced by treating the jury to all sides of the eyewitness debate, rather than assuming there was no controversy, that the issue, notwithstanding this literature is clear.28


Accordingly, the testimony of the government's handwriting expert was limited as described above, to identifying the similarities and dissimilarities of the known writing and the unknown. The defense expert on eyewitness identification was permitted along with the testimony of the government witness.


Dated: June 11, 1999



1This is the second trial of Mr. Hines on these charges. The first trial ended in a hung jury.

2In addition, the government points to a deposit slip which was in a discarded waste basket near where the suspect had been standing in the bank. The deposit slip included some of the numbers of Hines' social security number. Also, one of the bank customers linked the suspect to a red car in the parking lot; the evidence, the government suggests, linked Hines to a red car on that day.

3See Fed.R.Evid. 602 and 701. See also, William Twining, Evidence and Legal Theory, in Legal Theory and Common Law 62, 70 (William Twining ed., 1986).

4See Fed.R.Evid. 703, which provides:

The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to the expert at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data need not be admissible in evidence.

5Pursuant to Fed.R.Evid. 702:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

6See 509 U.S. at 596-97. See also Alexander Morgan Capron, Daubert And The Quest For Value-Free "Scientific Knowledge" in The Courtroom, 30 U.Rich.L.Rev. 85, 85 n.2 (1996) ("Science is essentially descriptive, positive and predictive. The goal is to tell us what is and what will be in the future. Law, on the other hand, is prescriptive and normative. The effort is to tell us what ought to be, to define rules of conduct and responsibility grounded in events of the past." Citing Margaret G. Farrell, Coping With Scientific Evidence: The Use of Special Masters, 43 Emory L.J. 927, 942 (1994)); Margaret G. Farrell, Daubert v. Merrell Dow Pharmaceuticals, Inc.: Epistemology & Legal Process, 15 Cardozo L. Rev. 2183, 2186-87 (1994); Peter H. Schuck, Multiculturalism Redux: Science, Law, & Politics, 11 Yale L. & Policy Review 1 (1993).

7The Court's approach - - the extent to which it adopted the premises of science, notwithstanding its attention to the differences between the laboratory and the courtroom - - has been subject to criticism. Chief Justice Rehnquist, for example, suggested that the majority should have given the lower courts the task, not of determining scientific validity, but of developing, on a case-by-case basis, a jurisprudence of science - - a set of legal principles for using science's truth in the interest of justice. See Daubert, 509 U.S. at 600-01 (Rehnquist, J., concurring/dissenting); see also, Farrell, supra. Others have suggested that the Court had simply taken the definition of science from Karl Popper, a definition that others have criticized as deriving from a culturally defined, time-bound paradigm. See Daubert, 509 U.S. at 593 (quoting Karl Popper, Conjectures and Refutations: The Growth of Scientific Knowledge, 37 (5th ed. 1989)); Capron, supra at 23; Farrell, supra at 2195, 2205; Thomas S. Kuhn, The Structure of Scientific Revolutions 17-20 (2d ed. 1970).

8The experts based their conclusions not on human studies, but on "in vitro" (test tube) and "in vivo" (live) studies, which were not "generally accepted" within the field of epidemiology. See id. at 582.

9The Court clarified that the abandonment of the "general acceptance" test as the sole requirement for scientific testimony would not lead to a "free-for-all" of "irrational pseudoscientific assertions." Id. at 595-96. Rather, the judge remains free to exercise her discretion, and between vigorous cross-examination, contrary evidence and careful instructions on the burden of proof, any faulty testimony will be adequately mitigated. See id.

10The fear that Daubert would place ill-suited responsibilities on lay judges prompted a number of states to retain Frye. "[A] courtroom is not a laboratory, and as such it is not the place to conduct scientific experiments. If the scientific community considers a procedure or process unreliable for its own purposes, then the procedure must be considered less reliable for courtroom use." Flanagan v. State, 625 So.2d 827, 828 (Fla. 1993) (citing Stokes v. State, 548 So.2d 188, 193-94 (Fla. 1989)). Other states have asserted that the admissibility of scientific evidence remains a legal question. "[I]n testing for the admissibility of a particular type of scientific evidence, whatever the scientific 'voting' pattern may be, the courts cannot in any event surrender to scientists the responsibility for determining the reliability of that evidence." Taylor v. State, 889 P.2d 319, 329 (Okla. Crim. App. 1995) (quoting United States v. Williams, 583 F.2d 1194, 1198 (2d Cir. 1978)).

The Ninth Circuit panel handling the remand of Daubert was skeptical. The Court notes that "the judge's task under Frye was relatively simple: to determine whether the method employed by the experts is generally accepted in the scientific community," but the judge's responsibility under Daubert is "a far more complex and daunting task," particularly when courts are faced with "matters at the very cutting edge of scientific research, where fact meets theory and certainty dissolves into probability." Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1315-16 (9th Cir. 1995). "Mindful of our position in the hierarchy of the federal judiciary," the appellate court continued, "we take a deep breath and proceed with this heady task." Id. at 1316.

11Like Daubert, Kumho was about "new" fields. It grappled with how to determine the reliability of expert testimony from an engineer who sought to conclude that a manufacture's defect in a tire caused a blow-out. The concern was with the expert's method of "draw[ing] conclusions on the basis of what seemed small observational differences." Id. at 1178. The testimony, in part, relied on scientific principles, but also on specific, rather novel, theories. This was a "new" methodology, not "generally accepted," and not particularly reliable at that. The Court affirmed the district court's exclusion of this testimony. See id. at 1179.

12Following the Daubert decision, courts began to distinguish between a standard to be used for admitting scientific verses non-scientific testimony, although Fed.R.Evid. 702 does not make or require such a distinction. See e.g., United States v. Starzecpyzel, 880 F.Supp. 1027, 1029 (S.D.N.Y. 1995) ("Daubert standards do not apply to such "skilled" witnesses"); see also Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 80 (1st Cir. 1998) ("Daubert furnishes the principle source of guidance" for determining the admissibility of scientific testimony). In fact, as noted by the Eleventh Circuit in United States v. Paul, 1999 WL 300884 (11th Cir. (Ga.)), many circuits were split on whether Daubert should also apply to nonscientific expert testimony. See 1999 WL 300884 at *3. On this question, some held that the application of Daubert was limited to scientific testimony, while others used Daubert's guidance to ensure the reliability of all expert testimony. See id. (comparing McKendall v. Crown Control Corp., 122 F.3d 803 (9th Cir. 1997) (limiting the application of Daubert to the evaluation of scientific testimony); with Watkins v. Telsmith Inc., 121 F.3d 984 (5th Cir. 1997) (holding that Daubert is not limited to scientific experts)).

13To be sure, this concern could be equated with "grandfathering old irrationality." D. Michael Risinger, Mark P. Denbeaux, Michael J. Saks, Exorcism of Ignorance as a Proxy for Rational Knowledge: The Lessons of Handwriting 'Expertise, 137 U.Pa.L.Rev. 731, 771 n.182 (1989).

14According to Denbeaux's article, co-authored with Saks and Risinger, Dean Wigmore bears a large measure of responsibility. See Risinger, Denbeaux, Saks, supra at 764-771. The validity of handwriting analysis has been assumed in Wigmore's treatises, and virtually every standard evidence treatise since that point. See id.

15Denbeaux's testimony was cited in United States v. Velasquez, 64 F.3d 844, 852 (3d Cir. 1995) (concluding that the district court erred by refusing to allow defendant's expert document examiner to testify in response to the government's document examiner); but see Paul, 1999 WL 300884 at *5 (court did not err in excluding Denbeaux's rebuttal testimony).

16As Denbeaux suggests: "The effects of a medical treatment innovation can be compared with the effects of other treatments or no treatment (a "Control"). And the classifications made by biomedical tests, phrenologists, polygraph examiners, and document examiners can be compared with some known criterion. For example, the classification (positive/negative) of a test for a certain disease can be compared with later developments in a group of patients' conditions or pathology reports." Risinger, Denbeaux, Saks, supra at 736.

17See D. Michael Risinger, Mark P. Denbeaux, and Michael J. Saks, Brave New "Post-Daubert World" - -- - A Reply to Professor Moenssens, 29 Seton Hall L. Rev. 405 (1998) (replying to Andre Moenssens, Handwriting Identification Evidence in the Post-Daubert World, 66 UMKC. L.Rev. 251 (1998), discussing and critiquing the Kam studies on which Moenssens and the government relied). See Moshe Kam, Gabriel Fielding, and Robert Conn, Effects of Monetary Incentives on Performance of Nonprofessonals in Document-Examination Proficiency Tests, 43(5) J. Forensic Sci. 1000-1004 (1998); Moshe Kam, Gabriel Fielding, and Robert Conn, Writer Identification by Profession Document Examiners, 42(5) J. Forensic Sci. 778-786 (1997); Moshe Kam, Joseph Wetstein, and Robert Conn, Proficiency of Professional Document Examiners in Writer Identification, 39 J. Forensic Sci. 5-14 (1994).

18See Moenssens, supra.

19Kam, the government's proposed expert agrees that "[s]urprisingly, there are only a few studies that examine the reliability of writer screening by document examiners." Kam, et al., Proficiency of Document Examiners in Writer Identification, 39 J. Forensic Sci. at 5 (citing Risinger, Denbeaux, Sax, supra n. 13). Kam also concluded that "[i]t is very likely that many examiner decisions and associations are difficult to verbalize, and that some verbal explanations are post factum re-creations of the reasoning process." Id. at 12.

20Denbeaux posits that the identification would be open to far less criticism if it were similar to that of photo identification. In other words, using several unidentified writings and then determining if any of the writings were produced by the same individual.

21In United States v. Buck, 1987 WL 19300 (S.D.N.Y. 1987), an opinion predating both Kumho and Daubert, the court observed that "the ability of jurors to perform the crucial visual comparisons relied upon by handwriting experts cuts against the danger of undue prejudice from the mystique attached to 'experts.'" Id. at *3.

Denbaux suggests that it is precisely because the jury can so easily draw their own conclusions that expert testimony even in this limited area is not at all necessary. I am not willing to come to that conclusion - - on this record at this time. I am persuaded for now that the testimony involves more than just identifying that is similar and what is different in the same way a lay person would. It involves taking the next step - - that this or that similarity matters, that it equals a general pattern. Presumably, the expert is helped in drawing general patterns by the numbers of exemplars she has seen, just like the spouse identifies the husband's handwriting because she has seen it numbers of times.

22She also noted that he had on a blue jacket, and a baseball cap.

23Another witness, a customer at the bank, was unable to effect any identification.

24Kassin testified that Hines' photograph was the darkest of the eight in the array. Moreover, the numbering for his photograph was different from the others. All the others were 6 numbers, beginning with 96, Hines' number was 1234.

25The defense pointed out that in the photograph taken of the lineup, which Dunne had indicated was a fair and accurate representation of it, Hines was the darkest man. Furthermore, Dunne could have inferred from the presence of a defense attorney that one of the men in the lineup was a suspect.

26In this respect, it is analogous to expert testimony on Rape Trauma Syndrome, see Roger B. Handberg, Expert Testimony on Eyewitness Identification: A New Pair of Glasses for the Jury, 32 Am.Crim.L.Rev. 1013, 1017 (1995), or pseudologia fantastica. See United States v. Thomas Shay, 57 F.3d 126, 129-35 (1st Cir. 1994). Pseudologia fantastica is categorized as a factitious disorder in the Diagnostic and Statistical Manual of Mental Disorders and is sometimes referred to as Munchausen's Disease. See id. at 129 n. 1. This disorder is a variant of lying, often characterized as an extreme form of pathological lying. See id. (citing R. Sharrock and M. Cresswell, Pseudologia Fanastica: A Case Study of a Man Charged with Murder, 29 Med.Sci.Law. 323, 323 (1989)). People who suffer from this condition Present falsifications that are "disproportionate to any discernable end." Id. Pseudologues represent fantasies as real occurrences. See id.

27In the first trial, I instructed the jury as follows, derived from United States v. Telefaire, 469 F.2d 552 (D.C. Cir. 1972):

One of the most important issued in this case is the identification of the defendant; what we call eyewitness identification. Now, recognize that the government has the burden of proving identity beyond a reasonable doubt.

You must be satisfied beyond a reasonable doubt of the accuracy of the identification of the defendant. If you are not convinced beyond a reasonable doubt that the person was the person who committed the crime, you must find him not guilty.

Identification testimony is an expression of belief or impression by the witness. Its value depends upon the opportunity the witness had to observe the offender at the time of the offense and then to make a reliable identification later.

In appraising eyewitness testimony, you should consider the following:

Are you convinced that the witness had the capacity and an adequate opportunity to observe the offender?

Now, whether the witness had an adequate opportunity to observe the offender at the time of the offense will be affected by a number of things, such as how long or short a time was available; how close the witness was; how good were the lighting conditions; whether the witness had had occasion to know the person in the past or whether the person was a stranger to the witness; and whether the person was in a moment of stress or clear-headed.

Are you satisfied that the identification made by the witness after the offense was the product of his own recollection or of suggestion at the intervening time?

You may take into account both the strength of the identification and the circumstances under which the identification was made.

If you find the identification of the witness at the later point may have been influenced by the circumstances under which the defendant was presented to the witness, you should scrutinize the identification with great care.

You may also consider the length of time that lapsed between the occurrence of the crime and the next opportunity of the witness to see the defendant as a factor bearing - - one of the main factors - - one of the many factors, rather, that bears on the reliability of the identification.

You may take into account any occasions on which the witness failed to make an identification or made an identification inconsistent with his or her identification at trial.

And finally, you consider the credibility of each identification witness as you do with anyone else. Look at whether they had the capacity to see what they said.

One more thing on identification. In this case, the identifying witness is of a different race than the defendant. In the experience of many, it is more difficult to identify members of a different race than members of one's own. If this is also your own experience, you may consider that in evaluating the witness' testimony. This is one of the elements that you may then consider in evaluating the witness' testimony.

28In United States v. Brien, 59 F.3d 274 (1st Cir. 1995), the First Circuit affirmed a lower court's exclusion of expert testimony on eyewitness identification due to inadequate foundation. See id. at 275-277. The Court did not adopt a blanket rule that this form of testimony must routinely be admitted or excluded, but left it for the trial court to make case by case determinations under Daubert. See id. at 277.