• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Learn the science in your cases

Posted on: Wednesday, 29 October 2003, 06:00 CST

Something has gone wrong with Daubert. To fix it, courts must first acknowledge that science cannot be defined simply.

The "Daubert revolution" has gone astray. Born of an effort to ensure that courts base verdicts and decisions only on genuine and reliable science, it has ossified into a sometimes-nonsensical effort to capture science in a simple definition or checklist of factors.

The Daubert trilogy-Daubert v. Merrell Dow Pharmaceuticals, Inc.,1 General Electric Co. v. Joiner,2 and Kumho Tire Co., Ltd. v. Carmichael3-sent the message that federal district judges should hold scientists to the standards of the day-to-day practice of science.4 But too often, courts apply a judicial version of science that no scientist would understand or recognize.5 Daubert is not wrong, but it has been distorted by lawyers and courts seeking to avoid the difficult task of truly understanding the underlying science in litigation.

The seeds for the current problems with the Daubert trilogy were sown during the oral argument of Daubert itself. When asked what makes something scientific, counsel for the defendant-appellee had no good answer. Counsel for the plaintiff-appellant was equally at a loss when asked what the admissibility test should be beyond the qualifications of the expert.

Since the argument in Daubert, the U.S. Supreme Court has made clear that a trial court should not confine its analysis to qualifications, but too many plaintiff lawyers still refuse to accept this break with the past. The defense bar, however, has not allowed the Court's question about science to go unanswered. Instead, it has convinced some courts that science follows a recipe- like process consisting of a virtually endless series of difficult hurdles quite at odds with the actual practices of scientists.6

Lawyers and courts persist in making errors about science for several reasons. First is the all-too-understandable urge to turn complex new ideas into simple old ideas. Second, and related to this urge, is a desire for simple, bright-line rules. Third is the fixation on precedent drilled into every lawyer's mind from day one of law school. Science, on the other hand, is more concerned with empirically supported rational explanation--which should be the law's focus, too, if courts are truly to follow and apply the trilogy's mandate that science in the courtroom should reflect the actual practice of science.

Of course, some lawyers and some courts have heeded the message of the trilogy, which has led to decisions that seem paradoxically at odds with those by courts that have not.7 One court essentially holds that science requires epidemiology to determine the cause of a disease; another recognizes that science, by its very nature, does not confine itself to what is known or to established methodologies. Most of these seeming paradoxes can be resolved by considering how the lawyers presented the science and whether the court really tried to understand it, or instead applied a legalistic misconception of science.

Why science defies simple definition

One need look no further than the second edition of the Federal Judicial Center's Reference Manual on Scientific Evidence, published in 2000, to find that the so-called scientific method does not exist. One of the biggest problems with the first edition was its failure to include any general discussion of science. This oversight was remedied with a chapter called "How Science Works" by David Goodstein, a physics professor from the California Institute of Technology. He makes crystal clear that efforts to define science as a process for discovery founder on "an important shortcoming. We don't really know what the scientific method is. . . . Everything about the way we do science-the language we use, the instruments we use, the methods we use-depends on clear presuppositions about how the world works."8

In a 1988 article, I wrote that "the misconceptions about the scientific method that permeate legal thinking reflect a once deeply entrenched philosophical tradition known as 'logical positivism.' The philosophers who developed this tradition sought to sustain a vision of science as the source of absolute truth."9

The logical positivists viewed a theory as completely distinct from the empirical observations it explained. Moreover, the reasoning leading to a theory did not count in determining whether the theory was valid. Instead, a theory's validity depended only on whether consequences deduced from it could be observed.

Known as the "hypothetico-deductive method," this version of science "implies that a theory, once verified, becomes an indelible part of our knowledge, though it eventually may become incorporated into a broader, more inclusive theory. At the same time, however, hypotheticodeduction, if strictly applied, invalidates a theory entirely if a single prediction proves false."10 Consider how completely at odds this approach is with one famous scientist's view that figuring out how the natural world works is an ad hoc process akin to a "constant dialogue between imagination and experiment."11 Thus, science is not about an inflexible process; rather, it is about evolving ideas that change and adapt as research progresses. Lawyers need to learn the language of the "constant dialogue" if they hope to explain how science evolves.

In addition to the misconception that science produces absolute truth, the law also sometimes views science as a purely inductive process. That, too, is wrong. Like logical positivists, inductivists were seeking

"a method of achieving objectivity while avoiding subjective preconceptions, and of obtaining empirical rather than abstract or metaphysical knowledge." . . . In its extreme form, the inductive approach holds that scientists should observe any phenomena they encounter in their experience and record their observations without any preconceptions as to what to observe or what the truth about the observed phenomena might be.12

Both logical positivism and inductivism fail to account for the presuppositions that Goodstein tells us are so embedded in science, or for the way in which scientists pursue new leads and then modify and adapt their theoretical explanations.

Consider the following hypothetical example of how the interaction between explanation and observations occurs.

[T]wo observers waiting by a highway at night may have very different theoretical ideas about what they will see. One might expect a UFO, and the other an automobile. When two lights appear in the distance, each observer will see something different. Observations, however, also shape theories. As the UFO/automobile draws closer, the observer who believes in extraterrestrial visitors will have to reshape his or her thinking to accommodate the empirical evidence. At some point, he or she will have to abandon the theory of UFOs, or at least modify it so that it allows for the fact that cars, as well as flying saucers, move along highways. Based on the evidence, a scientist would conclude that the car theory is far better supported and far more valid. Science thus proceeds by weaving observation and theory together, not by viewing them as parallel but separate threads.13

When science is understood as evolution derived from the "constant dialogue" between explanation and observation, it becomes clear that scientific validity can be defined no further than a requirement for an empirically supported, rational explanation.14 In many cases the explanation will have to fall within the bounds of normal scientific discourse, but that is not an absolute requirement.

In the introduction to the Federal Judicial Center's Reference Manual on Scientific Evidence, U.S. Supreme Court Justice Stephen Breyer tells the story of physicist Wolfgang Pauli: "After a colleague asked whether a certain scientific paper was wrong, Pauli replied, 'That paper isn't even good enough to be wrong!' Our objective is to avoid legal decisions that reflect that paper's so- called science."15 In other words, the law should recognize the great range of ideas in science and should exclude expert evidence only if it clearly falls outside what scientists would consider meaningful dialogue.

Methodology v. reasoning

Given the complexity of scientific reasoning, attempts to reduce it to a methodology are futile. However, vague language in Daubert has led some courts to insist that an expert lay out a precise, step- by-step methodology, and that it be subject to peer review.

In fact, the Supreme Court did not refer to just "methodology." Instead, it held that a district judge must conduct "a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and properly can be applied to the facts in issue."16

In some cases, "methods" and "methodology" are the proper focus. For example, if the issue before the court is the validity of handwriting analysis, there is a well-defined methodology, and the question to be considered is reduced to whether or not analysts can really distinguish one person's writing from another's.17

Perhaps the best example of a case in which methodology was properly the focus is Libas, Ltd. v. United States, decided in 1999.18 The issue on appeal was whether fabric imported from India had been made with a power loom or woven by hand. If made with a machine, it was subject to a higher tariff. Indian authorities hadcertified the fabric as hand-woven, but the U.S. Customs Service determined it was power-loomed based on laboratory tests.

The federal government argued that the tests were widely accepted, but it could not cite "'any published methods that specifically address [the] issue' of whether fabric has been power- loomed or hand-loomed."19

The Federal Circuit found room to doubt the federal government because "the reliability of the test had not been established by the obvious and natural method of double-blind testing."20 Citing the appellant's argument with approval, the court went on to hold that

non-double-blind testing is worth little because knowing the "right" result before-hand biases the results. The textile literature, by Customs' own admission, contains no published methodologies for distinguishing hand-loomed from power-loomed fabrics. And the personal experience and judgment of five analysts, one of whom is a weaver, is no basis for a technically supportable conclusion.21

Libas represents a dramatic change in approach, because the court analyzed the evidence just as a scientist would-by demanding empirical support for claims that a well-defined test could in fact distinguish one kind of fabric from another.

However, a more complex issue, such as disease causation in toxic tort litigation, would not be amenable to such easy resolution. Instead, a court would have to delve into the expert's reasoning to see whether it made sense. As the Eighth Circuit held in Sorensen v. Shaklee Corp., the "error rate" factor listed in Daubert is inapplicable to the issue of disease causation because "the testimony advanced involved theories and not any particular technique."22

Paradoxes in case law

The boom in toxic tort litigation in the 1970s and 1980s began with horrendously dangerous products like asbestos, the Dalkon Shield, and DES, and with horrendous working conditions and waste- disposal practices. Although the expert battles were hard fought, most scientists did not doubt the causal link between the plaintiffs' injuries and the defendants' conduct or products. With some justification, plaintiffs and their lawyers assumed that if a disease seemed connected to a defendant, discovery would probably prove that the connection was real.

As plaintiff lawyers pushed the limits of causation, however, the defense bar dug in and fought back. As the claims became more speculative, judicial skepticism about experts grew sharper and deeper. There developed a general sense that courts had to take control of experts.

Some judges responded by applying the 1923 Frye "general acceptance" test, which had lain dormant for many years. The Frye approach ended in 1993 when the Supreme Court held in Daubert that Federal Rule of Evidence 702 had displaced it. But Daubert and the subsequent decisions in Joiner and Kumho went well beyond narrow interpretation of evidentiary rules.

The Daubert trilogy wrought radical change by encouraging judges to be gatekeepers and by suggesting how they might ensure that an expert "employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field."23 When judicial skepticism joined this new gatekeeping mandate, it became imperative that plaintiff lawyers make clear to courts that their experts have provided an empirically supported, rational explanation within the bounds of normal scientific discourse. When this is done, the testimony probably will be admitted. When it is not done, the result will probably be exclusion.

Any attempt to understand seeming paradoxes in the case law-in terms other than the quality of the evidence, how it was presented to the court, and the court's willingness to apply the standards of science instead of wooden precedent-is doomed to failure. The focus has to be on the science itself, not abstract, legalistic precedents about science.

Two Fifth Circuit cases-Moore v. Ashland Chemical, Inc. and Pipitone v. Biomatrix, Inc.-make this point abundantly clear. In Moore-a case alleging that exposure to solvents had caused the plaintiff's reactive airway-dysfunction syndrome (RADS)-the en banc court, in an opinion written by Judge W. Eugene Davis, reversed a three-judge panel and affirmed summary judgment for the defendants.24

The plaintiff expert, pulmonary specialist Daniel Jenkins, "examined Moore, performed a series of tests, and reviewed Moore's medical records. He concluded that Moore was suffering from RADS. Based on his examination and tests, Jenkins expressed the opinion that Moore's RADS had been caused by Moore's exposure to [toluene solvent] vapors."25 The material-safety data sheet for Dow Toluene, the solvent at issue, indicated it could be a respiratory irritant, and there was literature supporting the theory that such irritation could cause RADS.

The expert, however, "made no attempt to explain how any of the other chemicals that he believed caused RADS in his earlier patients had properties similar to the Dow Toluene solution."26 Instead, the plaintiff argued in legalistic, precedential terms that Daubert should not apply to this kind of medical testimony because it was not scientific.

Pipitone was different, primarily because the focus was on the science and not on precedent and lists of rigid factors. Thomas Pipitone suffered from severe osteoarthritis in both of his knees, but because of other ailments, he was not a candidate for knee replacement surgery. As an alternative, he opted to have his knees injected with an artificial synovial fluid called Synvisc, manufactured by the defendant, Biomatrix. One knee was injected on the morning of January 25, 2000. That evening Pipitone had severe knee pain and a temperature of 101[degrees]. When he returned to the doctor two days later, tests showed he had a salmonella infection, which is extremely rare in joints.27

Pipitone sued Biomatrix, supporting his claim with expert testimony from two of the doctors who had treated him. One expert, Chad Millet, an orthopedic surgeon, was not certain of his opinion and essentially deferred to the other expert, Jeffrey Coco, who had "great expertise in the area of epidemiology and infectious diseases."28

Here, however, Coco had not conducted an epidemiology study, and his literature review had "yielded no report of any salmonella infection arising from a contaminated injectable knee product of any kind"-not even Cortisone, which has been injected into joints for years.29 Citing these supposed flaws in Coco's testimony and a failure to eliminate "many viable alternative sources" for the infection, the district court excluded him as a witness and granted summary judgment.

The Fifth Circuit, however, looked past the rote application of precedent and considered the reasoning that led Coco to his conclusion. He had "based [his] opinion on the timeliness of the infection (symptoms of which began to appear hours after the Synvisc injection), the source of the Synvisc [chicken parts, a known source of salmonella], the type of organism. . . ., and the elimination of all other likely alternatives."30

The appellate court first noted that salmonella-contaminated Synvisc clearly could cause an infection, and that the real issue was whether the fluid injected into Pipitone was in fact contaminated or if he was infected by another source. Coco's finding that there were no other reports of infection from knee injections did not contradict his opinion, and actually supported it by eliminating the possibility that "unsterile injection technique or some other cause unrelated to Synvisc" had caused the infection.31 The lack of other instances also explained the absence of peer- reviewed literature. No one would expect a publication on a phenomenon that had not occurred before. Coco's conclusion was further supported by the fact that Synvisc "is the only knee- injectable product made from chicken parts."32

Having thus disposed of Daubert's "testing" and "peer review" factors, the court went on to find that the "error rate" factor "is not particularly relevant, where, as here, the expert derives his testimony mainly from firsthand observations and professional experience in translating these observations into medical diagnoses."33 As to the "general acceptance" factor, the court held that "Dr. Coco based his opinion on how Pipitone contracted salmonella in large part on accepted medical knowledge of the ways in which salmonella functions as an organism and how it infects humans."34

The Fifth Circuit held that the record did not support the lower court's finding that "Dr. Coco had identified 'many viable alternative sources' of the salmonella infection in Pipitone's knee."35 Ingestion of salmonella was ruled out because Pipitone did not have any symptoms of gastroenteritis. Nor was it likely he had contracted the disease through an ulcerous wound on his toe. Coco explained that "it is nearly impossible to contract salmonella through even an open traumatic wound, much less the scabbed-over surface of Pipitone's toe."36 Unsterile injection had been ruled out based on the absence of reports in the literature.

Finally, the court rejected Biomatrix's argument that the source of the infection could not have been the Synvisc syringe used to inject Pipitone because testing of other syringes from the same lot showed no signs of contamination. The court accepted Coco's explanation for this seeming anomaly:

[Although] he would have expected other samples. . . in the same. . . lot to be contaminated . . . only a small number of salmonella organisms would be required to infect a joint that was directly exposed to the organism [, and] in his epidemiological experience, a batch that produces a contaminated sample may contain no other contaminated samples.37

The same Judge Davis who wrote the Moore opinion wrote the Pipitone opinion reversing summary judgment for the defendant.

Science trumps precedent

Moore and Pipitone \have been cited as examples of inconsistency in the application of Daubert to expert conclusions based on medical examination and differential diagnosis. But the same judge wrote both decisions, so something else is clearly going on.

The real difference is that in Moore the argument was legalistic, but in Pipitone it was scientific. In Pipitone, the explanation was clear and well supported. That is why the plaintiff was able to overcome the kind of judicial skepticism that led to the ruling in Moore.

The lesson for both plaintiff and defense lawyers is that they must learn the science in their cases to carry the day.38 If neither side masters the scientific issues, a case may still be decided on precedent, but lawyers should not rely on the other side's laziness. Moreover, recent precedent has not been kind to plaintiffs.

The Daubert trilogy is a decade old, and the legal profession has yet to fully grasp the central message that scientists must apply the same kind of knowledge and expertise in court as they do in the day-to-day practice of science. Cases like Pipitone show that lawyers and courts are finally getting the message. The result ultimately will be better-informed and more just verdicts.

Ideally, an expert will make a differential diagnosis only after examining the patient, taking the person's medical history, and reviewing the results of diagnostic tests.

Bolster your Daubert argument with documents from the ATLA Exchange

The following document collections on the admissibility of scientific evidence after the U.S. Supreme Court's Daubert decision are available from the ATLA Exchange.

For additional information, visit www.exchange.atla.org, or contact the Exchange by phone at (800) 344-3023 or by fax at (202) 337-0977.

Baker Valley Lumber, Inc. v. Ingersoll-Rand Co. The plaintiff's appellate brief in a case holding that a trial court erred when it disqualified a plaintiff's expert based on a test that focused partly on the reliability of the expert's conclusions under Frye v. United States, rather than exclusively on the reliability of the expert's methodology under Daubert. (No.LR3905.)

Bonner v. ISP Technologies, Inc. The plaintiff's petition, trial brief, and memoranda opposing the defendant's motions to exclude testimony; depositions of the plaintiffs experts; and the court's order on Daubert motions in a case in which a jury found for a woman who suffered neurological deficits that were marked by constant trembling after she had been sprayed in the face with a solvent containing gamma-buty-rolactrone. (No. LR3652.)

Compton v. Subaru of America, Inc. The parties' appellate briefs and an amicus brief in a case holding that the Daubert standards on admissibility are inapplicable to testimony based on an expert witness's experience or training. (No. PL644.)

Dodge v. Cotter Corp. The plaintiffs' motion for partial collateral estoppel and opposition to the defendant's motion for summary judgment based on Daubert in a case in which a jury found that a uranium mill mishandled hazardous waste when it allowed the materials to leach into groundwater, contaminating domestic wells. (No. LR3837.)

In re Joint Eastern & Southern District Asbestos Litigation (Mairoana v. United States Mineral Products Co.). The plaintiff's appellate briefs in a case holding that Daubert does not change the standards for judging the sufficiency of evidence. (No. LR2444.)

Joiner v. General Electric Co. The parties' briefs in a case holding that the correctness of a scientific expert's conclusions is a determination that should be left to a jury to decide. (No. LR2770.)

Kittleson v. Sandoz Pharmaceutical Corp. The transcript of proceedings in a case in which the court held that the plaintiffs' expert testimony-which relied on case reports, the temporal relationship between drug and incident, animal studies, and inferences based on similar drugs-is admissible under Daubert, concluding the basic relevance standard is a liberal one. (No. LR3605.)

Thornton v. Caterpillar, Inc. The defendant's trial brief in a case holding that the admissibility standards announced in Daubert do not apply to the testimony of a mechanical engineer or other expert witnesses possessing technical and other specialized knowledge. (No. LR2966.)

The ATLA Exchange offers access to MDEX Online Daubert Tracker, a searchable database with more than 12,000 expert records from evidentiary gate-keeping cases. ATLA members receive a substantial discount.

For more information, visit www.exchange.atla.org.

Science is not about an inflexible process; rather, it is about evolving ideas that change and adapt as research progresses.

Notes

1. 509 U.S. 579 (1993).

2. 522 U.S. 136 (1997).

3. 526 U.S. 137(1999).

4. Id. at 152 ("The objective. . . is to make certain that an expert. . . employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.").

5. Margaret A. Berger, Upsetting the Balance Between Adverse Interests: The Impact of the Supreme Court's Trilogy on Expert Testimony in Toxic Tort Litigation, 64 LAW & CONTEMP. PROBS. 289, 302 (2001) (courts have created "new rules in the name of science that do not exist in the scientific community").

6. The following litany from the Daubert hearing in a products liability case involving the drug Parlodel illustrates this approach: "The scientific method entails accumulating scientific data to formulate a hypothesis; observing, experimenting with, and testing that hypothesis; reviewing new data to come to a general conclusion about the phenomena being studied; and then repeating the process again and again to get closer to a true relationship." Hearing transcript, vol. 1 at 1998, Rider v. Sandoz Pharms. Corp., No. 1:95-CV-3068-TWT, (Dec. 18, 2000).

7. Compare Siharath v. Sandoz Pharms. Corp., 131 F. Supp. 2d 1347, 1370 (N.D. Ga. 2001) (excluding evidence in Parlodel case and holding that to prevail, a plaintiff would have to provide epidemiology, chemical mechanism, general acceptance, a plausible animal model, and "dozens of well-documented case reports"), aff'd sub nom. Rider v. Sandoz Pharms. Corp., 295 F.3d 1194, 1196 (11th Cir. 2002) (holding that there was no abuse of discretion), with Globetti v. Sandoz Pharms. Corp., 111 Supp. 2d 1174, 1180 (N.D. Ala. 2000) (also a Parlodel case; the court held that when epidemiology is a practical impossibility, science turns to other means).

8. David Goodstein, How Science Works, in REFERENCE MANUAL ON SCIENTIFIC EVIDENCE 67, 70 (2d ed. 2000).

9. Bert Black, A Unified Theory of Scientific Evidence, 56 FORDHAM L. REV. 595, 615 (1988) (citations omitted).

10. Id. at 617-18 (citations omitted).

11. FRANCOIS JACOB, THE STATUE WITHIN: AN AUTOBIOGRAPHY 225 (Franklin Philip trans., 1988). A common misconception, Jacob wrote, is that science advances inductively by "accumulating experimental facts and drawing up a theory from them."

12. Bert Black et al., Science and the Law in the Wake of Daubert: A New Search for Scientific Knowledge, 72 TEX. L. REV. 715, 757 n.273 (1994), quoting THEODOSIUS G. DOBZHANSKY ET AL., EVOLUTION 476 (1977).

13. Black, supra note 9,at 619.

14. At least two courts have recognized this formulation. See Rider, 295 F.3d 1194, 1197; Schafersman v. Agland Coop., 631 N.W.2d 862, 876 (Neb. 2001).

15. Stephen Breyer, Introduction, in REFERENCE MANUAL ON SCIENTIFIC EVIDENCE 1, 4 (2d ed. 2000) (citations omitted).

16. Daubert, 509 U.S. 579, 592-93 (emphasis added).

17. See, e.g., United States v. Hines, 55 F. Supp. 2d 62 (D. Mass. 1999).

18. 193 F.3d 1361 (Fed. Cir. 1999).

19. Id. at 1368.

20. Id.

21. Id. at 1368-69.

22. 31 F.3d 638, 649 (8th Cir. 1994).

23. Kumho, 526 U.S. 137, 152.

24. 151 F.3d 269 (5th Cir. 1998).

25. Id. at 273.

26. Id.

27. 288 F.3d 239 (5th Cir. 2002).

28. Id. at 245.

29. Id.

30. Id. at 248.

31. Id. at 246.

32. Id.

33. Id.

34. Id.

35. Id. at 248.

36. Id.

37. Id. at 249.

38. See also DePacpe v. Gen. Motors Corp., 141 F.3d 715, 720 (7th Cir. 1998), in which the Seventh Circuit chastised GM's lawyers for "their cri de coeur [about the plaintiffs expert. This complaint] is not backed up by references to any body of scientific knowledge. What tests do engineers use to resolve questions of the kind [the expert witness] addressed? What tests should he have performed? What data did he overlook? Counsel apparently want appellate judges to make a priori judgments about how scientific inquiry should be conducted. That way quackery lies."

Bert Black practices law with Diamond McCarthy Taylor Finley Bryant & Lee in Dallas.

Copyright Association of Trial Lawyers of America Sep 2003

More News in this Category