Expert Article Library

Out of the fire and into the Fryeing pan of back to the future

by Ned Miltenberg

By Ned Miltenberg

Nearly a decade has passed since the U.S. Supreme Court issued its ruling in Daubert vs. Merrell Dow Pharmaceuticals, Inc. (1) and it is time to evaluate where we’ve been, where we are, and, most important, where we can go regarding the admissibility of expert opinion testimony.

This article suggests that plaintiffs should sidestep the ever-growing dangers posed by Daubert (and its companion cases, General Electric Co. v. Joiner(2) and Kumho Tire Co. v. Carmichael (3)) by taking a page from the successful strategy developed by ATLA’s Legal Affairs Department for challenging tort “reform” statutes: Plaintiffs should consider filing suit in state courts, particularly in jurisdictions that follow the traditional “general acceptance” test established by Frye v. United States. (4)

Daubert was intended to “liberal [ize]” federal evidence practice and abolish what was then regarded as Frye’s hidebound insistence that expert opinion must represent consensus views (5).  In hindsight, however, Daubert has not met expectations.  Indeed, as many have feared, the Daubert-Joiner-Kumho trilogy has bad a devastating effect on civil plaintiffs, at least in federal courts and in those states that have fully adopted Daubert.

By contrast, Frye may now offer more hope for plaintiffs, particularly in states that restrict the “general acceptance” test to new scientific techniques and seemingly definitive devices like polygraphs—sometimes referred to as “black boxes”—that may overawe the jury and determine the outcome of the case.  These states do not apply to Frye test to expert opinions about scientific causation, let alone to expertise that does not rise to the level of hard science.  This recommendation is bases on an analysis of the following historical trends:

  • Before Daubert, federal courts rarely scrutinized the scientific validity of expert opinion testimony in any kind of case and were particularly reluctant to do so in civil cases although experts have testified on various matters since at least 1620( 6), their opinions were not subject to special scrutiny until Frye was decided in 1923. Even so, for the first 50 years after Frye, the case was cited in only 96 federal and state cases, or fewer than 2 cases per year (7). Significantly, until 1984, Frye was never used to exclude an expert in a civil case. (8)
  • i>Frye began to be used in general much more often in the 1970s.  In the 1980s, defendants began using it heavily in toxic tort cases in particular.  Federal courts published more than 50 opinions on experts in the six years that preceded Daubert, decisions that sparked the widespread fear that defendants would use Frye to strangle many toxic tort and products liability cases in their beds. (9)

  • Although it once was hoped that Daubert would reduce the frequency and severity of judicial scrutiny of expert opinions, in reality it had the opposite effect, “trigger[ing] a deluge” of motions to exclude expert testimony, “especially [motions] in…civil cases(10).” In the six years after Daubert, federal courts published 1,065 opinions on expert admissibility on these motions, 871 of which involve civil cases, or 36 times the number of rulings in civil cases in the previous six-year period. (11)

  • Viewed together, the Daubert-Joiner-Kumho trilogy stands for the following propositions: Federal trial judges must act as vigilant “gatekeepers” (12) regarding the admissibility of expert testimony (Daubert). They must do so regarding expert testimony of all types, both scientific and nonscientific (Kumho). They must ensure, at minimum, that testifying experts honor the same theories, employ the same methodologies, uses the same tools, and follow the same standards as they do in the course of their ordinary, nonlitigation work (Kumho). And trial court decisions to exclude expert witness testimony will be reversed only for abuse of discretion (Joiner).

  • Federal judges are not only hearing many more motions to exclude experts, but they are becoming ever more willing to grant them. (13)

  • The great majority of Daubert motions are filed by civil defendants, and they have been winning nearly 70 percent of the time. (14)

Prospects for the future are even more forbidding, particularly in federal courts.  There are at least six reasons why.

First, dozens of corporate-funded tort “reform” groups, industry and insurance trade associations, right-wing think tanks, and corporate foundations (led by Peter Huber and the Manhattan Institute (15)) have continued their decade-long attack on plaintiffs’ “junk science.”  Indeed, these groups are redoubling their efforts at “judicial [re-]education” on this issue through week-long educational retreats for federal judges at luxurious resorts. (16)

The influence of these groups and the effectiveness of their efforts at corporate brainwashing cannot be doubted.  Tort  “reform” propaganda in general, and judicial education campaigns in particular, are among the principal reasons why plaintiffs’ fortunes declined in the 1980s and 1990s. (17)

Second, groups like the American Tort Reform Association and the Defense Research Institute are urging defense lawyers to file Daubert motions whenever possible.  These groups advise that the motions provide a relatively low-cost way to exhaust the finances and patience of all but the most well-funded and persistent plaintiff lawyers.

Third, plaintiff lawyers have not yet felt the full force of Kumho’s injunction to apply Daubert to all kinds of expert testimony, not just scientific testimony.

Fourth, many federal judges may not yet appreciate the fact that Joiner has given them virtual carte blanche to exclude testimony, as these decisions are reviewed on appeal under the highly deferential “abuse of discretion” standard.  As the judges begin to recognize the breadth of their authority, it is likely that more expert testimony will be excluded.

Fifth, lawyers and the experts they enlist must anticipate having to satisfy not just the five factors that comprised the now-famous “flexible,” “nondefinitive checklist” outlined in Daubert, but an innumerable roster of other factors, too (18).  Many courts have interpreted the Supreme Court’s declaration that a “trial judge must have considerable leeway in deciding…how to go about determining whether particular expert testimony is reliable” as an invitation to devise new tests for expert witnesses (19).  At the last count, at least 50 different hurdles had been erected for experts to surmount. (20)

Sixth and finally, President George W. Bush is sure to appoint more tort “reformers” to the federal bench.

It seems unlikely that the Supreme Court will reverse or even revisit the Daubert-Joiner-Kumho trilogy anytime soon, particularly in light of the newly effective amendments to Federal Rule of Evidence 702, in which Congress codified Daubert.  (Given the prospect that any new justices nominated by Bush will cleave to his preferred Scalia-Thomas model, no news from the Court will truly be good news.)


b>The state solution

Because of these trends, plaintiffs who litigate in federal court face time-consuming and expensive battles against long odds.  Indeed, it is fair to say that Daubert has made trying cases in federal court a riskier and more expensive enterprise than at any time in the last 50 years.

Although there are ways to shorten the odds in federal court—essentially through infinitely more careful, even more time-consuming, and increasingly expensive pretrial screening and preparation—this article will not survey ways plaintiffs can survive in jurisdictions where Daubert holds sway.  Other, more exhaustive essays have already attempted to fill that bill. (21)

Instead, this article proposes a more modest, and hopefully less expensive, solution modeled on the strategy used by ALTA’s Legal Affairs Department in challenging the constitutionality of state tort “reform” statutes.  (This strategy has been successful in striking down statutes in Illinois, Indiana, Ohio, and Oregon, and we expect it to prove equally effective in a pending challenge to Florida’s recently enacted omnibus “reform” legislation.)  The strategy simply involves taking pains to file suit in (and avoid removal from) state court, where the constitutions, rules of evidence, and judicial precedents tend to be much more hospitable.

In a nutshell, because it’s difficult to see light at the end of the Daubert tunnel, plaintiffs must take another tunnel.  In fact, there are 51 other tunnels, 51 other venues where lawsuits can be tried, and 51 other jurisdictions where the odds against plaintiffs’ experts and plaintiffs’ fortunes can hardly be worse than they are in federal court and, as explained below, are often better.

  These options are available because Daubert involved an interpretation of a federal rule—Federal Rule of Evidence 702—not the U.S. Constitution, and thus is not binding on the states.  As a result, 23 states, including many of the most populous in the nation and the District of Columbia have either rejected Daubert or declined to adopt it.  The upshot is that, although Daubert has garnered the lion’s share of attention in the past decade, most of the people—and the overwhelming majority of the cases—in the country are still covered by Frye.

There are two crucial differences between jurisdictions that apply Frye and those (both federal and state) that apply Daubert. The first difference concerns the types of evidence to which Frye and Daubert apply, and the second lies in who ultimately determines admissibility.

Whereas, Kumho extended Daubert’s reach to all types of expert testimony, many Frye jurisdictions limit “validity/reliability” challenges to the 10 percent of experts who offer purely scientific testimony.  These courts do not allow these challenges against the 40 percent who furnish medical expertise, let alone t the remaining 50 percent, who supply all other kinds of expert testimony. (22)

Moreover, many of the state courts that apply Frye typically allow the “general acceptance” test to be used solely to challenge the admissibility of scientific testimony based on new, supposedly perfect (and perfectly definitive) mechanical “black boxes”—that is, machines, techniques, and devices like the ostensibly infallible “lie detector” that was scrutinized in Frye—that automatically spit out authoritative and often outcome-determinative “truths.” Courts fear that these machines (and machine-based testimony) have the aura of mathematical certainty or infallibility and thus the potential to overawe jurors. (23)

California courts are the acknowledged leaders in narrowly restricting the use of Frye to evaluate and exclude expert testimony.  Indeed, there are so reluctant to use their version of the Frye test, formulated in 1979 in People v. Kelly (24), that “there are no reported California cases applying the…Frye analysis to cancer causation and the like.” (25)

The reason for this reluctance was explained by the California Supreme Court in 1989.  According to the court, Frye is not only restricted “to that limited class of expert testimony which is based, in whole or part, on a technique, process, or theory which is new to science and, ever more so, the law (26),” but is also restricted to an

unproven technique or procedure [that] appears in both name and description to provide some definitive truth which the expert need only accurately recognize and relay to the jury.  The most obvious examples are machines or procedures which analyze physical data.  Lay minds might easily, but erroneously, assume that such procedures are objective and infallible. (27)

Thus, “absent some special feature which effectively blindsides the jury, expert opinion testimony is not subject to…Frye.” (28)

The importance of the distinction between expert testimony that is based on “some special feature which effectively blindsides the jury” (and is therefore subject to Frye) and testimony that is not (and is therefore not subject to Frye) was explained in People v. McDonald (29).  In that case, California’s high court unanimously ruled that it was prejudicial error to exclude psychological testimony regarding the accuracy of eyewitness identification.  As the court said in a later holding,

In dispensing with any need for a Kelly/Frye showing in [McDonald], [we] noted that “when a witness gives his personal opinion on the stand—even if he qualifies as an expert—the jurors may temper their acceptance of his testimony with a healthy skepticism born of their knowledge that all human beings are fallible….We have never applied the Kelly/Frye rule to expert medical testimony….” (30)

Significantly, the more certain and exact the scientist is in his or her views and the more “foolproof” and absolutely true the scientist claims the technique or theory to be, the more the Kelly/Frye rule is needed and the more likely its invocation will be successful.  Conversely, the less categorical the expert and the more he or she relies on a variety of techniques and “normative and subjective” thought processes and judgments (for example, “patient interviews, case history, and past experience in forming…[an] educated but debatable opinion”), the less reason exists for the Kelly/Frye doctrine to be applied. (31)

In light of these multiple hurdles, it is hardly surprising that California appellate courts have never sanctioned the use of the Kelly/Frye test in a toxic tort, products liability, or medical negligence case.  In cases where experts do not rely on a single supposedly definitive and infallible truth-telling machine or device, but instead use their scientific judgment to evaluate a variety of “normative and subjective” tools, methodologies, techniques, theories, data, and experience to reach conclusions, Kelly/Frye simply does not apply.  And California courts are not the only ones that restrict Frye to a narrow band of cases. (32)

The advantages of this approach are obvious.  Plaintiffs enjoy a greater chance of having a jury hear their experts testify, and they are spared the considerable expense of conducting endless admissibility hearings.

The second crucial difference between Daubert and Frye lies in who ultimately makes the call on admissibility.  Daubert entrusts federal judges with responsibility for “gatekeeping.”  That term may have been intended to describe a relatively benign presence—as noted above, Daubert was ostensibly designed to liberal [ize]” admissibility standards—and it conjured up images of friendly functionaries checking tickets at a county fair.  But the reality is that the gatekeepers tend to view themselves as the Spartans at Thermopylae or, to use a modern analogy, as Cold War GIs at Checkpoint Charlie. (33)

  Frye certainly is not a panacea, and it may be inherently, unavoidable amorphous—no theory can be admitted into evidence unless it has been approved by some vaguely defined consensus among a vaguely defined “relevant community” of scientists.  But at least Frye courts tend not to play amateur scientist and tend not to devise ever more numerous hurdles to be placed in front of expert witnesses.  It may be difficult to win the “general acceptance” of the scientific community, as Frye requires, but it is even more difficult to win general acceptance (which is still a Daubert factor) and meet the dozens of other tests thrown up by inventive and hostile judges. (34)

Thus, under Frye, trial judges do not evaluate the reliability of all preferred testimony.  They merely assess whether a restricted class of testimony is based on generally accepted principles and is therefore admissible.

For these reasons, unless a plaintiff happens to be in a state that follows the Daubert trilogy hook, line, and sinker, or in a state whose application of Frye offers more malevolent treatment of expert witnesses than they might receive in federal court, the plaintiff should file in state court if at all possible, and avoid removal to federal court at all costs. (35)


  1. 509 U.S. 579 (1993).
  2. 522 U.S. 136 (1997).
  3. 526 U.S. 137 (1999).
  4. 293 F. 1013, 1014 (D.C. Cir. 1923). Although many states have adopted their own versions of Federal Rules of Evidence, not all of these states have adopted Daubert, Joiner, and Kumho, or all aspects of these cases.
  5. 509 U.S. 579, 588.
  6. See Kimberly M. Hrabosky, Case Note, Kumho Tire v. Carmichael: Stretching Daubert Beyond Recognition, 8 GEO.MASON L. REV. 203, 203 (1999).
  7. See Michael J. Saks, Merlin and Solomon: Lessons from the Law’s Formative Encounters with Forensic Identification Science, 49 HASTINGS L.J. 1069, 1076 (1998).
  8. Developments in the Law—Confronting the New Challenges of Scientific Evidence, 108 HARV.L.REV. 1509, 1529 n.160 (1995).
  9. D. Michael Risinger, Navigating Expert Reliability: Are Criminal Standards of Certainty Being Left on the Dock? 64 ALB.L.REV.99,101,104 (2000).
  10. Id. At 104.
  11. Id.
  12. Daubert, 509 U.D. 579, 597.
  13. According to the Federal Judicial Center, federal judges were much more likely to “scrutinize testimony” in civil cases after Daubert than ever before, and they were at least 20 percent “more likely” to exlude even experts who have been carefully vetted by plaintiff lawyers and consulting experts.  MOLLY T. JOHNSON ET AL., EXPERT TESTIMONY IN FEDERAL CIVIL TRIALS: A PRELIMINARY ANALYSIS (Fed. Jud. Ctr. 2000).
  14. Risinger, supra note 9, at 108; see also Gregory A. Weimer, Expert Evidence: What You Don’t Know About Daubert Can Hurt You, 24 VT.B.J. & L.DIG. 51, 53 (1998).
  15. See Kenneth J. Chesebro, Galileo’s Retort: Peter Huber’s Junk Scholarship, 42 AM.U.L. REV. 1637,1705-22 (1993).
  16. See ALLIANCE FOR JUSTICE, JUSTICE FOR SALE (1993); George Lardner Jr., Report Links Environmental Rulings, Judges Free Trips, WASH. POST, July 25, 2000, at A21.
  17. Theodore Eisenberg & James A. Henderson Jr., Inside the Quiet Revolution in Products Liability, 39 UCLA L. REV.731, 770 (1992).  For example, plaintiff “success rates in published opinions fell from 56 percent in 1979 to 39 percent in 1989, a drop of 29 percent.  At the federal district court level…plaintiff success rates fell from 41 percent in 1979 to 31 percent in 1989, a drop of 24 percent.” Id. at 741.  Eisenberg and Henderson attribute theses developments to the effect of a blizzard of tort “reform” propaganda directed at judges and jurors. Id. at 778-79, 789-95.
  18. Daubert identified five factors that could be used to evaluate whether particular testimony was “scientifically valid” and therefore reliable as evidence: whether the method or technique (1) was falsifiable or testable, (2) had been peer-reviewed, (3) had a known (or potentially knowable) rate of error, (4) was accompanied by established standards controlling the technique’s operation and accuracy, and (5) was generally accepted. 509 U.S. 579, 592-596.  Ironically, a technique that would have passed muster under Frye as “generally accepted” might fail under Daubert if a court concluded that an entire discipline was somehow lacking in scientific rigor.
  19. Kumho, 526 U.S. 137, 152.
  20. See Ned Miltenberg, Out of the Fryeing Pan and into the Fire (ATLA Annual Convention Paper 2000).
  21. Id.; Gerson Smoger, From Rule 702 to Daubert to Joiner to Kumho Tire:  A Review of the Supreme Court’s Analysis of the Admissibility of Expert Testimony (ATLA Annual Convention Paper 2000); Stuart Ollanik, Expert Testimony; Defeating the Kumho Challenge, TRIAL, Nov. 1999, at 28.
  22. See Samuel J. McNaughton, What is Good Science? 13 NAT. RESOURCES & ENVT 513, 518 (1999).
  23. See Michael H. Graham, The Expert Witness Predicament, 54 U. MIAMI L. REV. 317,331 (2000).
  24. 549 P.2d 1240, 1244 (Cal.1976).
  25. RAOUL D. KENNEDY & JAMES C. MARTIN, CALIFORNIA EXPERT WITNESS GUIDE §4.15, at 56 (2d ed., Supp. Mar.1999) (emphasis added).  Court first adopted Frye in Kelly, the Kelly/Frye doctrine has been the subject of fewer than a dozen published civil opinions.  And even those scant civil cases in which California appellate courts have applied Kelly/Frye have been quasi-criminal in nature.  See, e.g., Ramona v. Superior Court, 66 Cal. Rptr. 2d 766 (Ct. App. 1997) (childhood sexual abuse); In re Mark C., Rptr. 2d 856 (Ct. App. 1992) (child molestation).
  26. People v. Stoll, 783 P.2d 698,710 (Cal. 1989).
  27. Id.
  28. Id.
  29. 690 P.2d 709, 724-26 (Cal.1994), overruled on other grounds by People v. Mendoza, 4 P.3d 265 (Cal. 2000).
  30. Stroll, 783 P.2d 698, 711 (quoting McDonald, 690 P2d 709, 724).
  31. Id. at 712.
  32. See, e.g., Kuhn v. Sandoz Pharms. Corp., No. 83, 266, 2000 WL 1838325 (Kan. Dec. 15, 2000) (holding that Frye could not be used to screen, let alone reject, the admissibility of testimony by three physicians who used differential diagnosis to conclude that defendant’s drug caused plaintiff’s death); Logerquist v. McVey, 1 P.3d 113 (Ariz. 2000) (Frye inapplicable to evaluate admissibility of expert testimony on repressed memory syndrome); Flanagan v. State, 625 So. 2d 827, 828 (Fla. 1993) (“[P]ure opinion testimony…does not have to meet Frye, because this type of testimony is based on the expert’s personal experience and training”); Brooks v. People, 975 P.2d 1105,1110 (Colo. 1999) (“Generally, the Frye test is applied to novel scientific devices and processes involving the manipulation of physical evidence including lie detectors, experimental systems of blood typing, voiceprints, identification of human bite marks, and microscopic analysis of gunshot residue”).  See also Reese v. Stroh, 874 P.2d 200, 205 (Wash. Ct. App. 1994), aff’d on other grounds, 907 P.2d 282 (Wash. 1995); State v. Anderson, 853 P.2d 135, 137 (N.M. Ct. App. 1993), rev’d on other grounds, 881 P.2d 29 (N.M. 1994); State v. Trager, 974 P.2d 750, 750-52 (Or. Ct. App.), review denied, 994 P.2d 125 (Or. 1999); Torres v. State, 962 P.2d 3, 22 (Okla. Crim. App. 1998), cert. denied, 525 U.S. 1082 (1999).  See generally Edward J. Imwinkelried, The Next Step After Daubert: Developing a Similarly Epistemological Approach to Ensuring the Reliability of Nonscientific Expert Testimony, 15 CARDOZO L. REV. 2271, 2285 (1994); MICHAEL H. GRAHAM, HANDBOOK OF FEDERAL EVIDENCE n.10 (4th ed., Supp. 1999).
  33. One commentator has argued that Daubert makes judges into “executioners” rather than “gatekeepers.”  Larry E. Coben, The Daubert Decision: Gatekeeper or Executioner? TRIAL, Aug. 1996, at 52.  Others have suggested that conservative judges are less concerned with gatekeeping than with “industry ‘safekeeping.’”  Janine M. Kern & Scott R. Swier, Daubert v. Marrell Down Pharmaceuticals, Inc.: “Gatekeeping” or Industry “Safekeeping”? 43 S.D.L.REV. 566 (1998).
  34. If you cannot avoid federal court, you should take pains to try to anticipate—and satisfy—all the myriad tests that might be used to challenge the admissibility of your expert’s testimony.  In addition, you might argue that if your expert’s testimony satisfies what the Daubert Court regards as Frye’s more restrictive “general acceptance” test, that fact alone should guarantee the validity and reliability of the testimony.
  35. The easiest way to avoid removal is to sue at least on in-state defendant ( to defeat complete diversity and thus to deprive a federal court of jurisdiction) and to make sure that you serve that defendant first.