An Epistemologist in the Bramble-Bush

At the Supreme Court with Mr. Joiner

Special issue contains articles from expert meeting, 'Evidence: Its Meanings and Uses in Law, Medicine and Health Care.


Here is Justice Blackmun, struggling valiantly if not quite successfully to articulate the mismatch between science and law that lies at the root of the trouble:

[T]here are important differences between the quest for truth in the courtroom and the quest for truth in the laboratory. Scientific conclusions are subject to perpetual revision. Law, on the other hand, must resolve disputes finally and quickly. The scientific project is advanced by broad and wide-ranging consideration of a multitude of hypotheses, for those that are incorrect will eventually be shown to be so, and that in itself is an advance. Conjectures that are probably wrong are of little use, however, in the project of reaching a quick, final and binding legal judgment—often of great consequence—about a particular set of events in the past.52

Yes, we want the law to settle disputes in a timely manner, while scientific inquiry takes—well, it takes the time it takes. Of course, we want cases settled not just promptly but rightly: Mr. Frye to be acquitted if and only if he didn't do it, Mr. Coppolino to be convicted if and only if he did do it, Mr. Joiner to be compensated if and only if his cancer was promoted by his exposure to PCBs, and so on. When scientific evidence is pertinent, we want scientific evidence which is probably right.53 As Justice Breyer reminds us, one of the goals that the Federal Rules of Evidence set themselves is "that the truth be ascertained."

I don't mean to suggest that juries can never (perhaps with the help of a cross-examining attorney) spot inconsistencies in scientific testimony, realize that a scientist's credentials are dubious, notice that the studies relied on were not controlled, or form a reasonable suspicion that a scientific witness is stretching the facts for the sake of a large fee, or, etc.;54 nor, of course, that mistakes are only made where scientific witnesses are involved. But as I have been maintaining all along, scientific evidence is "more so"—complex, esoteric, often expressed in an unfamiliar and deeply theoretical vocabulary, and hence unusually difficult for a jury or a judge adequately to assess. (On average, that is; nothing I have said implies that it is more difficult for a judge or jury adequately to assess relatively simple scientific evidence than, say, extremely complicated evidence about accounting procedures.)

No legal form of words can come close to ensuring that only the probable-enough is admitted. Of course we want relevant and reliable scientific evidence; but that form of words doesn't tell a judge anything about what, specifically, to exclude and what to admit (as Peirce might have put it, it reaches only the second grade of clarity, not the third, pragmatic or operational grade). Of course, also, scientists in the relevant field are nearly always better judges of the quality of scientific work than the rest of us; but finding a good way to delegate some of the responsibility isn't trivial, and nothing can ensure that even the most competent and honest scientists will always agree about what is probably right, or that they won't sometimes agree that, at the moment, they just don't know.

No wonder scientific evidence provides so many opportunities for opportunism! Often, we are trying to arrive at justice on the basis of imperfect and imperfectly understood information; and not so rarely, we are trying to create justice out of ignorance.

I'm afraid I have been something of an epistemological wet blanket—so much so that by now you may think me an incurable pessimist. So I had better remind you of that nice old Leibnizian joke: "What's the difference between an optimist and a pessimist? They both think this is the best of all possible worlds"—and assure you that in my opinion this is quite far from the best of all possible worlds.

There are no easy answers; but there are, certainly, better questions and worse. Rather than worrying fruitlessly about the problem of demarcation or the distinction of methodology versus conclusions and all that, we would do better to turn our attention to questions of other kinds—and to keep firmly in mind that, though perfection is impossible, better is better than worse; that the cumulative effect of small improvements can be quite large; and that it is inadvisable to restrict our attention too exclusively to issues and strategies internal to the legal system.

Some of the fruitful-looking questions are practical in orientation: What could be done to help jurors deal better with scientific evidence: e.g., consistent with filtering out legally unacceptable questions, to allow them to ask for clarification when they can't follow an expert witness? What could scientists' professional associations do to help serious scientific witnesses communicate better with judges and juries, or to discourage those who abuse their expertise? Could the legal profession and legal educators do more to discourage unscrupulous witness-shopping and related abuses? What could we learn from the experience with Judge Pointer's panel about bridging some of the gaps between the folkways of science and of the legal system? What advice might best be given to court-appointed scientists about what connections should be disclosed, or what kinds of record-keeping will be expected of them? (Should we consider asking court-appointed scientists to provide details of the qualifications and affiliations of any assistants on whom they relied; of which studies they decided to look at in detail, and why; of which studies seemed most strongly to indicate the contrary conclusion to theirs, and why, in their opinion, those studies were flawed?)

Could we make the legal system more responsive when new evidence comes in to the scientific community?55 Could the scientific community be more responsive when legal disputes turn on scientific issues irresoluble by the presently available evidence? Can we think of ways to provide incentives for scientists to study such issues even when they are of much less scientific than practical interest?

Other fruitful-looking questions are more policy-oriented: How significant a gatekeeping role is it appropriate for judges to take? (What exactly do we value about trial by jury, and why?) Given that mistakes are inevitable, should we be more willing to tolerate some kinds than others—not forgetting that scientific evidence plays a role both in civil and in criminal cases, and on both sides?56 Do we think it appropriate for policy considerations about, for example, how to manage the risks inherent in our reliance on synthetic materials, chemicals, drugs, etc., also to determine what evidence is admissible in criminal cases? (What exactly do we value about uniformity in the legal system, and why?) Are the problems of scientific evidence significantly exacerbated by the contingency-fee system? If so, is it worth the price—presumably, more limited access to the legal system for those without large resources—of changing it? What, ideally, would be the role of tort litigation vis à vis other means of ensuring that, when there is a question about the safety of this or that product, it is carefully looked into, and appropriate action taken?57 —a question prompted in part by the singularly unfortunate interaction of the FDA and the tort litigation system in the silicone-implant affair.

And, of course: Are these things done differently elsewhere, specifically in the legal systems of other scientifically and technologically advanced countries? If so, what are the benefits, and what the drawbacks?

But it might be prudent, before I begin to tackle such questions, to take Mr. Lec's very shrewd advice, and Think Before I Think . . .

Dedicated to the memory of Richard A. Hausler

Return to Contents

Appendix: Cases Cited

Commonwealth v. Lykus, 327 N.E. 2d 671 (Mass. 1975)

Coppolino v. State, 223 So. 2d 68 (Fla. Dist. Ct. App. 1968), appeal dismissed, 234 So. 2d 120 (Fla. 1969), cert. denied, 399 U.S. 927 (1970)

Daubert v. Merrell Dow Pharmaceuticals Inc., 509 U.S. 579, 113 S. Ct. 2786 (1993)

Frye v. United States, 293 F. 1013 (D.C. Cir. 1923)

Joiner v. General Electric Co., 864 F. Supp. 1310 (N.D. Ga. 1994), reversed, General Electric Co. v. Joiner, 78 F.3d 524 (11th Cir. 1996), reversed and remanded, 522 U.S. 136, 118 S. Ct. 512 (1997)

Kumho Tire Co., Ltd., v. Carmichael, 526 U.S. 137, 119 S. Ct. 1167 (1999)

People v. Williams, 164 Cal. App. 2d. Supp. 858, 331 P.2d 251 (Cal. App. Dep't Super. Ct. 1958)

Reed v. State, 391 A.2d 364 (Md. 1978)

United States v. Addison, 498 F.2d 741, 744 (D.C. Cir. 1974)

United States v. Starzecpyzel, 880 F. Supp. 1027 (S.D.N.Y. 1995)

Return to Contents


a. This article is adapted from a paper presented at a conference on epistemology and the law of evidence organized by the School of Law and the Department of Philosophy at the University of North Carolina-Chapel Hill, and in the Schools of Law at Boston University, the University of Pennsylvania, the College of William and Mary, the University of Iowa, the University of Virginia, Creighton University, and the University of Maryland. It was also discussed with faculty in the School of Law at Duke University, and was distributed as part of the briefing packet for a workshop on science-based medical evidence organized by the Institute of Medicine, National Academy of Sciences. I would like to thank Paul Gross, Richard Hausler, Robert Heilbroner, Mark Migotti, and Edgardo Rotman for reading this article in draft form and giving me their reactions; Clare Membiela and Janet Reinke of the University of Miami Law Library for their help in locating relevant materials; and the students in my class on Scientific Evidence in Theory and in Court, who taught me a lot.

1. Gross 1983: 262.

2. Huber 1991, 1992; Huber and Foster 1997.

3. General Electric Co. v. Joiner, 522 U.S. 136, 148, 118 S. Ct. 512, 520 (1997) (Breyer, J., concurring).

4. For a brief summary of controversies in recent philosophy of science, see Haack 1995 and Haack 1996. McErlean 2000 is a useful anthology.

5. The term, and the idea, come from Bacon 1620.

6. I borrow this happy phrase from Quine 1995: 16.

7. See for example Haack 1990 and Haack 1993. I am also drawing, in this section, on Haack 1995, Haack 1996, and Haack 1999.

8. Einstein 1936, in Einstein 1954: 295; drawn to my attention in 1996 by John Norton.

9. For the relevant history (up to the date of its publication, naturally) see Portugal and Cohen 1977.

10. Readers who have reservations about the concept of truth are referred to Haack 1998b: 7-30 and Haack 1999.

11. Bauer 1993 chapter 3 is good on this.

12.Watson 1968 chapter 26.

13. This puts me in mind of geneticist S. C. Harland's comment on trying to talk about biology with Trofim Lysenko: "it was like discussing the differential calculus with a man who did not know his 12-times table" (Gardner 1952: 147, referring to Huxley 1949).

14. Again, I rely on Watson 1968.

15. Hand 1901: 40-49; the date (1620) is given on p. 45.

16. "The defendant in Frye was subsequently pardoned when someone else confessed to the crime," writes Paul Giannelli 1980: n. 42. Giannelli cites Wicker 1953; Wicker, he says, cites Fourteenth Annual Report of Judicial Council of the State of New York, 265 (1948). But according to the most complete account I have been able to find of the many twists and turns of Mr. Frye's story—Starrs 1982—none of this is true.

17. My source in Starrs 1982: 694; he refers to Transcript on Appeal, File 3968, retired files, National Records Center, Suitland, MD.

18. From Judge Van Ordsel's opinion for the appellate court in Frye. At the time, the D.C. Court offered little in the way of rationale for its ruling. Much later, however, when the influence of Frye was waning, the same court argued that "[T]he requirement of general acceptance in the scientific community assures that those most qualified to assess the validity of a scientific method will have the determinative voice" (United States v. Addison; my source is Giannelli 1980: 1207).

19. See Black, Ayala, and Saffran-Brinks 1994: 735 ff., listing Reed v. State and United States v. Addison, excluding voiceprint evidence under the Frye test; and Commonwealth v. Lykus, admitting voiceprint evidence under the Frye test. There is a useful summary of relevant cases in the Symposium on Science and Rules of Evidence, 99 Federal Rules Decisions 188 (1983).

20. Giannelli comments: "if the 'specialized field' is too narrow . . . the judgment of the scientific community becomes, in reality, the opinion of a few experts" (1980: 1209-1210).

21. I rely on Giannelli 1980: 1229-1230. He mentions Saltzburg and Redden 1977: 426 as holding that the Federal Rules are compatible with the Frye test because they don't mention general acceptance; and Wright and Graham 1978: 92, as holding that the Federal Rules are incompatible with the Frye test because they don't mention general acceptance.

22. Daubert, 509 U.S. at 598, 113 S. Ct. at 2794.

23. Daubert, 509 U.S. at 580, 113 S. Ct. at 2790.

24.Daubert, 509 U.S. at 593, 113 S. Ct. at 2796.

25. Green 1992: 645. A footnote (12) refers to Popper, but I can find no reference to Hempel.

26. Daubert, 509 U.S. at 600-601, 113 S. Ct. at 2800.

27. In ordinary speech, of course, corroborated usually means "confirmed by another witness," but Popper has given the word a quite different, technical meaning. Black, Ayala, and Saffran-Brinks 1994: 750 ff. seem to have confused corroboration, in Popper's sense, with confirmation. Green—who, incidentally, introduces Popper's philosophy of science in Kuhnian terms, as "the existing paradigm under which scientists work"!—acknowledges that Popper holds that "[t]heoretically . . . hypotheses are never affirmatively proved" but continues, "of course, if a hypothesis repeatedly withstands falsification, one may tend to accept it, even if conditionally, as true" (1992: 645-646).

28. By January of 1998, according to Frankel 1998: 3, there had been more than eleven hundred such cases.

29. General Electric Co. v. Joiner, 522 U.S. 136, 136, 118 S. Ct. 512, 514 (1997), citing Joiner v. General Electric Co., 864 F. Supp. 1310, 1326 (N.D. Ga. 1994), which in turn cites Daubert, 509 U.S. 579, 113 S. Ct. 2786 (1993), where the phrase occurs three times: at 597 and 2786; at 590 and 2795; and at 599 and 2800.

30. See General Electric Co. v. Joiner, 522 U.S. at 140, 118 S. Ct. at 516.

31. But the question with regard to furans and dioxins, according to the Supreme Court ruling, remained open.

32. Brief for Petitioners, General Electric Co. v. Joiner, 47.

33. Brief for Petitioners, General Electric Co. v. Joiner, 49, citing Skrabanek and McCormick 1997: 35, quoted in Huber and Foster 1997: 142. I notice that on the same page, Skrabanek and McCormick refer to what they call the "weight of evidence fallacy"; this, they claim, is not scientific because science, according to Popper, focuses on negative evidence (which cannot be outweighed by confirming instances). While I am noting that GE's lawyers cite Peter Huber, I will also note that Kenneth Cheseboro was one of Mr. Joiner's lawyers.

34. Oral Argument of Michael H. Gottesman, General Electric Co. v. Joiner, 43-44. Mr. Gottesman was also one of the attorneys for Mr. Daubert.

35. General Electric Co. v. Joiner, 522 U.S. at 155, 118 S. Ct. at 523 (emphasis mine).

36. General Electric Co. v. Joiner, 522 U.S. at 151, 118 S. Ct. at 521 (Stevens, J., dissenting).

37. And of any reassuring noises about jurors' ability to assess the weight of scientific evidence.

38. The term, and the idea, come from Sellars 1965: 172.

39. 61 Federal Register 19,760-1: 17972 (1996) (my italics).

40. Submission of Rule 706 National Science Panel Report: 2, In re: Silicone Gel Breast Implant Products Liability Litigation (N.D. Ala. 1998) (No. CV 92-P-10000-S), available on-line at fjc/BREIMLIT/SCIENCE/report.htm.

41. "Only" not only because the sum is trivial relative to the compensation awarded in some implant cases, but also because the amount is quite modest relative to the task undertaken.

42. Submission of Rule 706 National Science Panel Report: 8.

43. Pointer Rules Federal Science Panel Report Not Tainted by Payments to Panelist, 7.5 Legal Aspects of Breast Implants: 4 (April 1999). I conjecture that the discrepancy between reports about the sum of money involved—plaintiffs say $750, court says $500—may be a matter of Canadian versus U.S. dollars. Plaintiffs also object that a colleague who had assisted Dr. Tugwell in his work for the panel had received support from a company wholly owned by BMS.

44. It is just this that, as a philosopher, I find most disturbingly unfamiliar when I tackle legal matters. Perhaps that is why, though Karl Llewellyn can write (1930: 141), "To me there is more joy than pain, by a good deal, in the thorns of such a thicket as that through which I have just dragged you," I am starting to feel as if I have been dragged through a hedge backwards!

45. And United States v. Starzecpyzel (880 F. Supp. 1027 [S.D.N.Y. 1995]) raises some interesting epistemological issues about learning and skill in perception, the relation of knowing-that and knowing-how, etc. But I shall have to set these aside.

46. General Electric Co. v. Joiner, 522 U.S. at 148, 118 S. Ct. at 520 (Breyer, J., concurring).

47. Washington Times, 11 December 1998, byline by Eric Peters.

48. Wall Street Journal, 10 December 1998, A22.

49.Jay Reeves, No Implant-Disease Link? ABC News on-line at (visited 8 July 1999).

50. I note that in June 1999, a thirteen-member committee of the Institute of Medicine also concluded that "silicone breast implants do not cause chronic disease, but other complications are of concern." News from the National Academies on-line at news.nsf/isbn/0309065321?opendocume. June 21, 1999 (visited 10 August 1999).

51. Dr. Diamond, one of the members of Judge Pointer's panel, acknowledging that she knows X, Y, and Z, who turn out to be connected in some way with the defendants, remarks that she feels "extraordinarily naive." Transcript of Rule 706 Panel Hearing: 91, available on-line at February 4, 1999 (visited 9 July 1999). I suspect that the involvement of large industrial concerns in the funding of scientific research is at this point so ubiquitous that it may be quite difficult to find scientists who are both competent to the task required and without such associations as Dr. Diamond acknowledged.

52. Daubert, 509 U.S. at 596-97, 113 S. Ct. at 2798.

53. While according to Popper, remember, no scientific claim is ever probable.

54. The truthfulness of a witness is a matter of (1) whether what he is saying is what he believes to be true and (2) whether what he believes to be true is true. Where scientific testimony is concerned, in general I would think juries are more likely to be able to judge the former than the latter.

55. Michael Graham observes: "Once [the status of this or that scientific evidence] is set in appellate concrete, a long time might be required to change it when scientific skepticism begins to overtake the original scientific optimism about the validity of the principle or procedure." 99 Federal Rules Decisions 188: 222-23 (1983). Or vice-versa.

56. In the same 1983 Symposium on Science and Rules of Evidence (99 Federal Rules Decisions 188: 206), Paul Giannelli writes: "For me Frye functions much like a burden of proof. . . . If [in criminal cases] we are going to make mistakes in assessing the validity of a novel technique, they should be mistakes of excluding reliable evidence rather than mistakes of admitting unreliable evidence." Ironically enough, however, in Frye, where the novel scientific evidence was proffered by the defense, Giannelli's argument would go exactly the other way.

57. In his concurring opinion in General Electric Co. v. Joiner (522 U.S. at 148, 118 S. Ct. at 520), Justice Breyer remarks shrewdly on our ubiquitous dependence on synthetic substances and the importance of ensuring that the "powerful engine" of tort litigation discourages the production only of the harmful stuff (though it spoils the effect somewhat that the case in question concerns PCBs, so dangerous that they have been banned for decades!).

Return to Contents


Bacon, F. 1620. The New Organon.

Bandow, D. 1999. Keeping Junk Science Out of the Courtroom. Wall Street Journal, 26 July, A23.

Bauer, H. 1993. Scientific Literacy and the Myth of Scientific Method. Urbana: University of Illinois Press.

Begley, S., and A. Rogers. 1997. War of the Worlds. Newsweek, 10 February, 56-58.

Black, B., F. J. Ayala, and C. Saffran-Brinks. 1994. Science and the Law in the Wake of Daubert: A New Search for Scientific Knowledge. Texas Law Review 72:715-802.

Bridgman, P. 1955. Reflections of a Physicist. New York: Philosophical Library.

Cheseboro, K. 1993. Galileo's Retort: Peter Huber's Junk Scholarship. American University Law Review 42:1637-1726.

Einstein, A. 1936. Physics and Reality. Journal of the Franklin Institute 221.3; reprinted in Einstein 1954.

Einstein, A. 1954. Ideas and Opinions, translated by Sonja Bargmann. New York: Crown.

Frankel, M. S. 1998. The Role of Science in Making Good Decisions. American Association for the Advancement of Science. Testimony before the House Committee on Science, 10 June.

Gardner, M. 1952. Fads and Fallacies in the Name of Science. New York: New American Library.

Giannelli, P. 1980. The Admissibility of Scientific Evidence: Frye v. United States, a Half-Century Later. Columbia Law Review 80:1197-1250.

Goldberg, C. 1999. Judges' Unanimous Verdict on DNA Lessons: Wow! New York Times, 24 April, A10.

Graham, M. 1987. Federal Rules of Evidence. St. Paul, MN: West.

Green, M. D. 1992. Expert Witnesses and Sufficiency of Evidence in Toxic Substance Litigation: The Legacy of Agent Orange and Bendectin Litigation. Northwestern Law Review 86:643-699.

Gross, J., ed. 1983. The Oxford Book of Aphorisms. Oxford: Oxford University Press.

Haack, S. 1990. Rebuilding the Ship While Sailing on the Water. In Perspectives on Quine, ed. R. Barrett and R. Gibson. Oxford: Blackwell.

Haack, S. 1993. Evidence and Inquiry: Towards Reconstruction in Epistemology. Oxford: Blackwell.

Haack, S. 1995. Puzzling Out Science. Academic Questions spring: 25-31; reprinted in Manifesto of a Passionate Moderate: Unfashionable Essays. Susan Haack. Chicago: University of Chicago Press.

Haack, S. 1996. Science as Social?—Yes and No. In Feminism, Science, and Philosophy of Science, ed. J. Nelson and L. Hankinson Nelson. Dordrecht: Kluwer; reprinted in Haack 1998a, 194-22.

Haack, S. 1998a. Manifesto of a Passionate Moderate: Unfashionable Essays. Chicago: University of Chicago Press.

Haack, S. 1998b. Confessions of an Old-Fashioned Prig. In Manifesto of a Passionate Moderate: Unfashionable Essays. Susan Haack. Chicago: University of Chicago Press.

Haack, S. 1999. Staying for an Answer. Times Literary Supplement, 9 July, 12-14.

Hand, L.1901. Historical and Practical Considerations Regarding Expert Testimony. Harvard Law Review 15:40-58.

Huber, P. 1991. Galileo's Revenge: Junk Science in the Courtroom. New York: Basic Books.

Huber, P. 1992. Junk Science in the Courtroom. Valparaiso University Law Review 26:732-755.

Huber, P., and K. B. Foster, eds. 1997. Judging Science: Scientific Knowledge and the Federal Courts. Cambridge: MIT Press.

Huxley, J. 1949. Heredity, East and West: Lysenko and World Science. New York: H. Schuman. Institute of Medicine. 1999. News from the National Academies. Available on-line at 21 June.

Lec, S. 1962. Unkempt Thoughts. New York: St. Martin's.

Llewellyn, K. 1930. Bramble Bush: Our Law and Its Study. 2d. ed. 1951. New York: Oceana.

McErlean, J., ed. 2000. Philosophies of Science: From Foundations to Contemporary Issues. Belmont, CA: Wadsworth.

Peters, E. 1998. Benchmark Victory for Sound Science. Washington Times, 11 December.

Portugal, F. H., and J. Cohen. 1977. A Century of DNA: A History of the Discovery of the Structure and Function of the Genetic Substance. Cambridge: MIT Press.

Quine, W. V. O. 1995. From Stimulus to Science. Cambridge: Harvard University Press.

Reeves, J. 1999. No Implant-Disease Link? Available on-line at www.abcnews.go. com/sections/living/DailyNews/breastimpants981201.html.

Rogers, A. 1996. Come In, Mars. Newsweek, 20 October, 56-57.

Saltzburg, K., and K. Redden. 1977. Federal Rules of Evidence Manual: A Complete Guide to the Federal Rules of Evidence. Charlottesville, VA: Michie Co.

Schmitt, R. B. 1997. Witness Stand. Wall Street Journal, 17 June, A1 and A8.

Sellars, W. 1965. Scientific Realism or Irenic Instrumentalism? In Boston Studies in the Philosophy of Science, eds. R. Cohen and M. Wartofsky. Dordrecht: Kluwer.

Skrabanek, P., and J. McCormick. 1997. Follies and Fallacies in Medicine. Buffalo, NY: Prometheus.

Starrs, J. E. 1982. A Still-Life Water-Color: Frye v. United States. Journal of Forensic Sciences 27(3):684-694.

Wall Street Journal. 1998. "An Unnatural Disaster" (editorial). 11 December, A22.

Watson, J. D. 1968. The Double Helix: A Personal Account of the Discovery of the Structure of DNA; critical edition, ed. G. Stent, 1980. New York: W. W. Norton.

Wicker, W. 1953. The Polygraphic Truth Test and the Law of Evidence. Tennessee Law Review 22(6):711-742.

Wilson, E. O. 1999. Consilience: The Unity of Knowledge. New York: Alfred Knopf.

Return to Contents

Current as of April 2001
Internet Citation: An Epistemologist in the Bramble-Bush: At the Supreme Court with Mr. Joiner. April 2001. Agency for Healthcare Research and Quality, Rockville, MD.