The Judicial Gatekeeping Project
Home | About the Gatekeeping Project | Judicial Gatekeeping Book | Educational Resources |
| Message Board | News & Developments | Subscribe to the Judicial Gatekeeping Newsletter |
| Site Map | Search | Contact Us | Credits | Back

& National Conferences

New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Rhode Island
South Carolina
South Dakota
West Virginia


Scope of Judicial Gatekeeping:
Does Daubert Apply To Areas Other Than New Science?

by Chris Kelly and Derek Squire - Harvard Law School '99

[Please note: the Supreme Court's decision in March 1999 in Kumho provides a definitive answer to this question, which is, in short, "yes."  Click here for our analysis of Kumho.  In light of Kumho, this paper is now useful for examining the differing approaches states may be considering on the breadth of Daubert's applicability.]
     The Supreme Court's decision in Daubert v. Merrell Dow Pharmaceuticals is often thought of as two separate holdings.(1) In the first, unanimous holding, the Supreme Court asserts that the long-standing Frye rule for the admissibility of expert testimony has been replaced by the more recently enacted Federal Rules of Evidence.(2) This holding has been generally well received and universally applied by the federal district courts. The second holding, however, has been a source of great confusion and controversy. In the second part of the Daubert decision, over the dissent of Justices Rehnquist and Stevens, Justice Blackmun writes that federal judges have a duty to ensure that "an expert's testimony rests on a reliable foundation and is relevant to the task at hand."(3) In addition, Justice Blackmun lays out a number of criteria by which the reliability of an expert's testimony may be judged.(4) While it is clear that the majority of the Court envisions a more active gatekeeping role for federal judges, it is not clear exactly what the limits and scope of that role should be. Perhaps most significantly, it is uncertain whether this more active judicial role applies only to determining the admissibility of "hard" scientific evidence, or whether the Daubert decision is to be used to evaluate the admissibility of non-scientific and quasi-scientific expert testimony as well.(5)

     In his dissent to Part II of the majority decision, Justice Rehnquist writes,

Questions arise simply from reading this part of the Court's opinion, and countless more questions will surely arise when hundreds of district judges try to apply its teaching to particular offers of expert testimony. Does all of this dicta apply to an expert seeking to testify on the basis of 'technical or other specialized knowledge'--the other types of knowledge to which Rule 702 applies--or are the "general observations" limited only to "scientific knowledge"?(6)
Justice Rehnquist's confusion seems well grounded. The Frye test for admissibility applied only to "new scientific evidence."(7) The Federal Rules of Evidence, however, apply to "scientific, technical, or other specialized knowledge," while the Blackmun opinion only discusses criteria for judging the admissibility of scientific evidence.(8) Thus, it is not surprising that the several district courts have applied the Daubert decision in differing ways.

The Different Approaches
     The district courts have generally followed three different formulas in applying Daubert:

(1) some have interpreted the Daubert decision narrowly and applied it only to scientific evidence;

(2) others have applied the decision broadly and used the Daubert criteria for judging the admissibility of all expert testimony;

(3) still others have taken a middle ground, assuming a Daubert-like duty to judge the admissibility of all expert testimony but using different criteria depending on the type of expert.

A. Argument that Daubert Should Be Restricted to Scientific Expert Testimony
     There are strong arguments that Daubert should properly be applied only to the testimony of scientific experts and not non-scientific or quasi-scientific experts. First of all, lay jurors might "give more weight to testimony purportedly based on a scientific theory rather than an opinion or theory based in another discipline."(9) Thus, the usual skepticism of jurors along with the rigors of cross-examination should be enough to ensure that a non-scientific expert's testimony is reliable. More importantly, Justice Blackmun's criteria for assessing the reliability of testimony may have little relevance to non-scientific testimony. He suggested that the reliability of scientific testimony be judged using four criteria: 1) whether or not it could be tested and "falsified"; 2) whether it had been subject to peer review and publication; 3) its known or potential rate of error; and 4) whether or not it was generally accepted within the scientific community.(10) Edward Imwinkelried has argued that these factors cannot be applied to fields of expertise which are not based on the scientific method, the process by which scientific hypotheses are tested and re-tested using duplicable experiments:
In a sense, Daubert was an easy case for the Court; the nature of the scientific process made it a simple matter for the Court to identify external validation standards. The possibility of re-testing, by duplicating the original experimental conditions, provides an external check...In contrast, nonscientific disciplines do not lend themselves to external re-testing.(11)
The Second and Ninth Circuits have followed this reasoning, applying Daubert only to scientific testimony. For all other types of expert testimony, these courts rely more closely on the language of the Federal Rules of Evidence. The judge will focus mainly on Rule 702 which allows testimony by a "qualified expert", so long as the proffered "knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue."(12)

     In Thomas v. Newton International Enterprises, the Ninth Circuit Court of Appeals found that the district court had abused its discretion by excluding the testimony of a longshoreman who offered to testify to the hazards of an unguarded safety hatch on a tanker.(13) The Court wrote that "Daubert was clearly confined to the evaluation of scientific expert testimony. . . Special concerns arise when evaluating the proffer of scientific testimony that do not arise when evaluating the type of expert testimony offered here."(14) The Court went on to evaluate and accept the longshoreman's testimony based on Federal Rule of Evidence 702 noting that Rule 702 "is broadly phrased and intended to embrace more than a narrow definition of qualified expert."(15)

     The Second Circuit took a similar approach in United States v. Locascio.(16) In this case, the district court allowed the testimony of a police officer regarding the behavior of organized crime families without explicitly evaluating the testimony based on the Daubert criteria. The court upheld the lower court ruling and found that the Daubert criteria should be applied to scientific testimony only and should not restrict the judge's freedom to allow all other expert testimony admissible under the Federal Rules of Evidence:

We decline, however, to shackle the district court with a mandatory and explicit trustworthiness analysis. The district judge, who has the ideal vantage point to evaluate an expert's testimony during trial, already has the authority under Fed. R. Evid. 403 to conduct an explicit trustworthiness analysis should she deem one necessary...We will not, however, circumscribe this discretion by burdening the court with the necessity of making an explicit determination for all expert testimony.(17)
These decisions demonstrate the direction taken by some courts in restricting Daubert to apply only to the consideration of "hard" scientific evidence.

B. Argument that Daubert Should Be Applied to All Expert Testimony
     Conversely, several commentators have taken the opposite view and have argued that the Daubert decision and Justice Blackmun's criteria for judging the admissibility of evidence should be applied to all expert testimony. The arguments for such an approach are twofold. First, FRE Rule 702, which the Daubert decision interpreted, includes all types of expert testimony, not just purely scientific evidence. Arguably, then, the Court intended its decision to apply to all types of testimony as well. Second, it is desirable to have a single set of criteria to use for evaluating all expert testimony. Attorney Merlyn Clark argues, "There is value in having standards that help to produce reliable expert opinion evidence and an argument can be made that a single standard is highly desirable."(18) Such a simplified approach would make the judicial gatekeeping role easier since it would not require developing and implementing separate criteria for evaluating different types of experts. Additionally, advocates would be in a better position to counsel clients due to their increased ability to assess which experts would be admitted prior to trial. This reasoning has been followed by the Seventh and Eighth Circuits.

     In Gier v. Educational Service Unit No. 16, an action was brought against a school on behalf of mentally retarded students for sexual, physical, and emotional abuse.(19) The plaintiffs intended to rely on evidence presented by three psychiatric experts whose testimony was based on their evaluations of the children. At a Daubert hearing for the case, the court severely limited the scope of the experts' testimony due to its determination that such testimony was unreliable because of problems of "testability" and high error rate.(20) In its conclusion, the Gier court cited a Louisiana case, State v. Foret, which held that such psychological syndrome theories are essentially unable to be tested. Generally, this has not been accepted within any scientific community, and possesses a high potential rate of error: "this type of evidence is of highly questionable scientific validity, and fails to unequivocally pass the Daubert threshold test of scientific reliability."(21)

      In the Seventh Circuit, the U.S. District Court for the Northern District of Illinois attempted to measure the reliability of an engineer's testimony against Justice Blackmun's criteria. In Stanczyk v. Black & Decker, Inc., the plaintiffs produced a mechanical engineer to testify that the guard on a power saw could have been designed to cover more of the blade area.(22) The judge rejected the testimony because the engineer offered no testable design to support his concept, because "[t]here is a high potential rate of error for mechanical concepts," and because the engineer's theory had not been subject to peer review and publication.(23)

C. Argument that Daubert Should Apply to All Experts, But in Modified Form
     There is a third view of the Daubert standard's application to non-scientific expert testimony that recognizes the strengths of each of the preceding arguments and proposes a compromise. This interpretation argues that rather than take an all-or-nothing approach, courts need to adopt a modified version of the Daubert test aimed at customizing the reliability assessment they make to different types of expert testimony.

     The justification for this approach can be found in the language of the Daubert decision itself. First, in his majority opinion, Blackmun acknowledged his understanding that FRE Rule 702 was intended to apply to all experts and only limited the scope of the decision to "the scientific context because that is the nature of the expertise offered here."(24) Blackmun's recognition of the additional areas covered by FRE 702 implies that a similar approach should be taken with regard to those types of experts as well. Further, the method proposed by Blackmun for evaluating the reliability of an expert's testimony focuses on an assessment of the validity of the "reasoning or methodology underlying the testimony."(25) While his criteria are aimed at evaluating the methodology of scientific testimony, those who argue against applying Daubert to other types of evidence insist that these same criteria become unworkable when used in evaluating non-scientific expert testimony. However, this argument does not imply that the overall approach of Daubert--assessing the methodology behind an expert's testimony--is unworkable for "soft science"; it just means that different criteria are necessary.

     While the application of Daubert to non-scientific testimony is problematic, the general belief remains that judges do have some gatekeeping role in deciding the admissibility of all expert testimony. Professor G. Michael Fenner argues that "the same rules of evidence that led the United States Supreme Court to its decision in Daubert--Rules 403, 702, 703, and 706--apply to all expert witnesses."(26) However, results like those reached in cases such as Gier and Foret, in which the strict exercise of the Daubert test caused arguably relevant, reliable, and important psychological testimony to be excluded, have met with criticism.(27) Therefore, some commentators have advocated that the courts adopt a middle ground approach and rework the Daubert standard in order to instill some degree of flexibility into judicial gatekeeping.

     One such method for adapting the Daubert test begins with the separation of these different types of experts into quasi- and non-scientific.(28) First, with respect to experts offering non-scientific opinions, no assessment of methodology is possible, and thus no Daubert evaluation can be made. In these situations, "Rule 702 is the proper place to begin and end... The question should be whether the expert has technical or other knowledge that will assist the trier of fact," without any judicial role in evaluating the expert's underlying methodology.(29) Such an analysis will provide for a liberal admission of non-scientific expert testimony, enabling juries to assess its relevance and credibility.

     Under this approach, quasi-science should be treated differently. Experts testifying in areas like psychology and engineering "frequently fall somewhere between a doctor who bases an opinion regarding causation on published studies and the detective gang expert who develops an opinion from experience and training alone."(30) Thus, while a strict application of the Daubert criteria to such evidence would be difficult, a court can still make a reasonable assessment of the expert's methodology. The factors in making such an analysis would be whether the expert has "(1) technical or specialized expertise that will (2) be helpful to the trier of fact and (3) whether the methodology used is a methodology reasonably relied upon or generally accepted in the relevant community."(31) This approach will allow relevant testimony to reach the jury while still reserving the right for judges to exclude wholly unreliable evidence.

      A different approach for reworking the Daubert standard advocates applying a modified Daubert test to all non-scientific testimony.(32) The theory behind this adaptation is that just as scientific evidence is evaluated by considering the reliability of the experimentation conducted by the expert in question, so also can non-scientific evidence be appraised by looking at the background and relevant experiences of the expert in question. Thus, "the courts can deduce restrictions on the quantity and quality of the experiential bases for non-scientific experts' opinion."(33) That is, by reviewing the number of experiences supporting the opinion coupled with the degree of relevancy of those experiences to the issue at bar, a judge can effectively evaluate the methodology underlying any expert testimony, scientific or otherwise.

     While this approach does advocate a greater role for judges by applying Daubert to all expert testimony, it at the same time recommends a liberal application of the test when evaluating the reliability of non-scientific expert testimony. By relying on the legal practices of cross-examination, impeachment, and the production of adverse experts, the weakness of non-scientific testimony can be demonstrated without an overly restrictive judicial gatekeeping role.(34) The effect of this would be to provide greater ease in admitting non-scientific testimony through the use of less strict criteria. Nevertheless, this approach does have disadvantages. The primary drawback to using separate criteria for evaluating different types of expert testimony is precisely the argument set forth by Merlyn Clark, that the predictability associated with a single standard will be sacrificed. Likewise, the possibility of unreliable, non-scientific evidence reaching the jury and adversely affecting the outcome is heightened by the use of a less rigorous Daubert test.

     Regardless of the actual manner in which the Daubert criteria are reworked, this approach remains largely an academic concept. The majority of courts have taken one of the first two approaches, applying the Daubert criteria to non-scientific expert testimony either strictly or not at all. Few courts have attempted to employ a compromise approach. In Borawick v. Shay, the court addressed the issue of whether or not to allow a plaintiff to testify based on hypnotically-refreshed recollections of sexual abuse.(35) The court stated that "even though Daubert does not provide direct guidance, our decision today is informed by the principles underlying the Supreme Court's holding."(36) It then offered a broad interpretation of the Daubert test: "Rather than using rigid 'safeguards' for determining whether testimony should be admitted, the Court's approach is to permit the trial judge to weigh the various considerations pertinent to the issue in question."(37) By way of this thorough approach, the court ultimately disallowed the testimony due to its assessment of the unreliability of the evidence.

     While the courts have been fairly evenly split on the issue of whether or not to apply Daubert to non-scientific expert testimony, just about all of them have chosen to take an all-or-nothing approach. A careful analysis of the language and purpose of the Daubert decision, coupled with the obvious problems of a strict application, points toward a middle ground into which few courts have entered and none have adequately defined. Rule 702 of the Federal Rules of Evidence certainly applies to all types of expert evidence, and it is not unlikely that Daubert was intended to apply to all experts as well. Still, some manipulation of the criteria set forth in Daubert will be necessary before they can be properly used to evaluate expert testimony that falls beyond the scope of "hard" science. Commentators have offered insight into the manner and extent of such modifications, but it is up to the courts to decide this issue conclusively in the future.


1. 113 S.Ct. 2786 (1993).
2. Frye v. United States, 293 F. 1013 (D.C. Cir. 1923).
3. 113 S.Ct. at 2799.
4. Id. at 2796-2798.
5. For the purposes of this debate, Lisa M. Agrimonti defines the terms as follows. "Nonscientific experts do not use scientific principles or methodology to reach their opinions [such as accountants and lawyers]... quasi-scientific experts are [those] whose opinions are based both on scientific principles and other types of knowledge, including technical knowledge [such as psychologists, engineers, and economists]." See Agrimonti, infra note 9, 156 n.12-13.
6. 113 S.Ct. at 2800.
7. 293 F. at 1013.
8. Fed. R. Evid. 702.
9. Lisa M. Agrimonti. The Limitations of Daubert and its Misapplication to Quasi-Scientific Experts, A Two Year Case Review of Daubert v. Merrell Dow Pharmaceuticals. 35 Washburn L.J. 134, 144 (1995).
10. 113 S.Ct. at 2796-2798.
11. Edward J. Imwinkelried. The Next Step After Daubert: Developing A Similar Epistemological Approach to Ensuring the Reliability of Nonscientific Expert Testimony. 15 Cardozo L. Rev. 2271, 2283-2284 (1994).
12. Fed. R. Evid. 702.
13. 42 F.3d 1266 (9th Cir. 1994).
14. Id. at 1270 n.3.
15. Id. at 1269.
16. 6 F.3d 924 (2nd Cir. 1993).
17. Id. at 938-939.
18. Clark, Merlyn W. The Impact of Daubert on the Admissibility of Expert Opinion. 39 Advocate 10, 13 (1996).
19. 66 F.3d 940 (8th Cir. 1995).
20. Gier v. Educational Service Unit No. 16, 845 F.Supp. 1342, 1351 - 52 (D.Neb. 1994).
21. State v. Foret, 628 So.2d 1116, 1127 (La. 1993).
22. 836 F.Supp. 565 (N.D.Ill. 1993).
23. Id. at 567.
24. 113 S.Ct. at 2795 n.8.
25. Id. at 2796.
26. Fenner, G. Michael. The Daubert Handbook: The Case, Its Essential Dilemma, and Its Progeny. 29 Creighton L. Rev. 939, 968 (1996).
27. See Confronting the New Challenges of Scientific Evidence: III. Judicial Responsibilities in Assessing Expert Testimony. 108 Harv. L. Rev. 1509, 1525-1528 (May 1995) (discussing the difficulties in accurately assessing the reliability of psychological testimony).
28. This proposal is described in detail in Agrimonti, supra note 9.
29. Id. at 147.
30. Id. at 148.
31. Id. at 154.
32. For a detailed description of this proposal, see Imwinkelried, supra note 11.
33. Id. at 2290.
34. Agrimonti, supra note 9, 155.
35. 68 F.3d 597 (2nd Cir. 1995).
36. Id. at 610.
37. Id.

Page Last Modified on April 25, 1999 by Dan Fridman - Copyright 1999
All materials are the property of the Berkman Center for Internet & Society and Harvard Law School. 
Materials may be reproduced, distributed, or quoted as long as appropriate credit and citation is given.