Junking 'junk science.'

The trustworthiness of scientific testimony is a hotly contested issue that will have repercussions in coming congressional debates. The legal landscape in court is confused, and there are repeated calls to enforce restrictive standards. These new attacks on "junk science" are likely to be reflected in the debate about the Product Liability Bill, particularly the Biomaterials Subtitle.

The Supreme Court's decision in Daubert v. Merrell Dow Pharmaceuticals, Inc., ended the restrictive "general acceptance" test that Frye v. United States had established in 1923. Frye required that scientific expert testimony reflect the consensus of scientific opinion--so new knowledge and innovative techniques were excluded as unreliable.

The Court in Daubert held that the question of admitting scientific testimony and evidence must be resolved solely on scientific principles and methodology. The Court said that "cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof, rather than wholesale exclusion under an uncompromising `general acceptance' standard. is the appropriate means by which evidence based on valid principles may be challenged."

This means that, rather than having to stick to a scientific "party line," researchers using new techniques in the vanguard of scientific knowledge could bring their knowledge to court. Long disease latency periods have forced plaintiffs increasingly to depend on emerging advances in scientific knowledge to show that a particular harm causes a particular injury.

The new standards Daubert established seemed to mark a new era in legal acceptance of a broader range of scientific knowledge. But dicta in Daubert raised a new level of judicial scrutiny even more stifling than the Frye standard.

Courts were told that they still had to exercise a "gatekeeping function" over proffered testimony, and some have taken this function to heart. Some have gone so far as to appoint "independent advisers" to review proposed testimony and prejudge its suitability, rather than allowing cross-examination to expose imperfections in evidence clearly based on scientific methods and reasoning.

In a recent ruling in Hall v. Baxter Healthcare Corp., a federal district judge in Oregon excluded testimony linking silicone from breast implants to connective tissue disease. The judge appointed his own independent advisers on science but overruled their advice when they reached a decision contrary to his preformed judgment--and decided that no expert could testify about connective tissue disease because he found the evidence unreliable.

This ruling was acknowledged to go "farther in evaluating and in eliminating plaintiffs' claims than any other opinion in breast implant litigation in this country." It marks a dangerous turn toward judges substituting their judgment for that of the jury.

The Hall ruling will be applauded by some in the science and legal communities. Peter Huber's 1991 book, Galileo's Revenge: Junk Science in the Courtroom, urged continued adherence to the Frye standard. His theory is that if evidence has received the broadest possible acceptance among those in the scientific community, it is reliable.

Rather than trusting juries to reasonably distinguish good science from bad, using the time-tested method of cross-examination and presenting further expert testimony, Huber would have judges exclude testimony that is outside the mainstream--even if the underlying research was impeccable. For Huber, "junk science" is basically the same as "out-of-the-mainstream" science.

In reality, the only junk science is bad science. Science is the application of the scientific method to study a phenomenon. By systematic observation and experimentation, science provides greater understanding of the world. If a scientist designs valid, well-controlled experiments, using recognized methods to test hypotheses, the results of those experiments cannot be attacked as junk science, even if they do not conform to the prevailing wisdom in the field.

Recently, Marcia Angell, executive editor of the New England Journal of Medicine, joined the fray with editorials in the Journal and with her book, Science on Trial: The Clash of Medical Evidence and the Law in the Breast Implant Case, which also calls for reinstating the Frye standard. Angell treats publication in a peer-reviewed journal as an essential test of reliability--5 not surprising for the editor of a prominent medical joumal.

Angell repeats many of the standard myths and caricatures touted by "tort reform" advocates. She claims that a supposed "tort tax" saps economic strength and that predatory lawyers and greedy clients seek jackpot verdicts against saintly, blameless manufacturers.

However, Professor Sheila Jasanoff, chair of Cornell's Department of Science and Technology, has a different view. She expressed concern that judges who embroil themselves too deeply in their gatekeeping role "inescapably give up the role of dispassionate observer to become participants in a particular construction . . . of scientific fact."

In her book Science at the Bar: Law, Science and Technology in America, Jasanoff argues that "scientific knowledge" is far less monolithic than people like Huber or Angell suggest. Rather than the law following scientific discovery, Jasanoff points out that often science follows the law. As injuries and potential liability crop up, scientists scramble to do the work to test causal links.

She criticizes Angell's view of breast implant litigation as "ahistorical" because reliable epidemiologic studies of breast implants had not been performed at the time Dow Corning began settling liability actions. Indeed, Dow Corning's failure to adequately show the safety of these implants was a key reason the U.S. Food and Drug Administration required their withdrawal from the market.

All these issues will be in play in the upcoming debate on the Biomaterials Subtitle of the Product Liability Bill. This provision would eliminate liability for many suppliers of raw materials or subcomponents of medical implant devices. Silicone is probably the most prominent and controversial of these materials, and this subtitle last year would have exempted silicone for breast implants from liability.

The junk science label may be used to persuade lawmakers that even silicone in gel breast implants should be included in this special treatment. Angell's book is being used as a weapon by the "liability limits" lobby to treat all plaintiffs' scientific evidence with suspicion.

Already, the biomaterials provision has been reintroduced as part of S. 5, which is virtually the same as the Product Liability Bill that President Clinton vetoed last year. The discussion over the nature of scientific evidence, and the nature of science itself, will continue long after the fate of pending legislation is resolved.