Before the passage of the Federal Rules of Evidence in 1975, the rule of Frye v. United States governed the admissibility of scientific evidence. This rule required that a scientific technique be "generally accepted within the relevant scientific community" before an expert could testify to an opinion based on that technique. Commentators have criticized the Frye rule for excluding reliable evidence from the jury. The Federal Rules of Evidence broadened the standards for admissibility of evidence generally, and Rule 702 commissions judges to admit expert testimony on scientific matters if it "will assist the trier of fact." This rule effectively establishes a relevance standard that leaves questions of scientific validity to the jury.

These liberal standards of admissibility have opened the gates to dubious scientific claims and. allowed paid expert witnesses to give the aura of scientific legitimacy to virtually any claim. In recent years, the pendulum has begun to swing back, and courts have called for restrictions on the admissibility of expert testimony. In a recent toxic tort case, Christophersen v. Allied-Signal Corp. ,the Fifth Circuit has signaled this change by giving renewed vitality to the Frye rule. The irony of Christophersen is that, although the rules of admissibility of expert scientific testimony need to be made more rigorous, toxic tort is one context in which the Frye rule is too strict.

Albeit Christophersen died in 1986 from a rare form of liver cancer. For fourteen years before his death, he worked as a manager in a manufacturing plant, where he was exposed to battery fumes. Christophersen's wife brought a wrongful death suit in federal district court and alleged that exposure to the fumes caused his cancer. At trial, the plaintiff relied on the testimony of a single expert witness, Dr. Lawrence Miller, who claimed that Christophersen's cancer had been caused by prolonged exposure to the battery fumes. Dr. Miller based his opinion on the affidavit of a co-worker who described Christophersen's exposure to fumes but did not know their chemical composition. Questioning both the reliability of the co-worker's affidavit and Dr. Miller's causation theory, the district court excluded Dr. Miller's opinion and granted summary judgment for the defendants.

Sitting en bane, the Fifth Circuit affirmed the district court's ruling. The court, in a per curiam opinion, began by stating that the Federal Rules of Evidence and Frye provided the proper test for determining the admissibility of expert testimony. The testimony of an expert witness should be admitted only if sufficient facts support the expert's opinion, as required by Rule 703, and the expert's method has met general acceptance, as required by Frye. The court found that Dr. Miller's testimony failed both tests.

The court held that Rule 703 excludes an expert's opinion based on underlying facts so unreliable that the opinion would not assist the jury. The court stated that " the inquiry into the--- 'facts and data' underlying an expert's testimony is not limited to the admissibility of that data," but extends to the sufficiency and accuracy of the underlying facts. The co-worker's affidavit on which Dr. Miller relied was so "critically inaccurate or incomplete" that the district court did'not abuse its discretion in excluding his testimony.

The Frye test, the court explained, requires a court to examine the validity of the expert's reasoning by considering its acceptance in the scientific community. Both defense experts agreed that Dr. Miller's theory " ' has no support in medical science and is without foundation. '" The court similarly concluded that Dr. Miller's theory was "a scientific hunch, which as far as the record shows, no one else shares," and as such was properly excluded by the district court.

In his concurrence, Chief Judge Clark criticized both the majority's interpretation of Rule 703 and its application of Frye. He argued that Christophersen implicates Rule 703 only insofar as the expert's opinion derives from "inadmissible facts or data. " If the affidavit was admissible, Rule 703 was inapplicable. Even if the affidavit was inadmissible, the proper inquiry was not whether experts in the field would reasonably rely on the particular facts used by this witness, but whether "similar experts use facts or data of the same kind to form opinions on the subject in issue.” Judge Clark also criticized the Frye rule, but he concluded that it did not survive the adoption of the Federal Rules of Evidence. He concurred with the judgment, however, because the risk of prejudice outweighed the testimony's minimal probative value and warranted exclusion under Rule 403.

In dissent, Judge Reavley criticized the majority for changing the Federal Rules of Evidence "without benefit of [ an ] amendment. " He reaffirmed the traditional proposition that the courts "properly entrust determinations of evidentiary weight and credibility to the jury-even in ' scientific ' cases-because of our faith in the adversarial process. " Judge Reavley maintained that although the coworker's affidavit was "imprecise," Dr. Miller's reliance on imprecise data was a question for the jury, not the district court, to decide. He argued that Rule 703 requires deference to "an expert's reliance on chosen facts and data. "

Judge Reavley also faulted the court for relying on Frye. He noted that, although the Fifth Circuit had not expressly abandoned the Frye test, its use always had been limited to the admissibility of "novel device [ s ] or technique [ s ] " such as lie detector tests, and it had never been applied to the admissibility of expert " reasoning" such as whether exposure to certain chemicals causes cancer. Judge Reavley concluded that excluding "otherwise relevant evidence strictly on the basis of lack of general scientific acceptance. . . would impose a significantly heavier burden of proof and persuasion on the offering party.

Although the court in Christophersen plainly erred in reading Rule 703 to allow judges to inspect the validity of an expert's factual foundations, the real import of the case is that it breathes new life into the Frye rule and counters "the movement away from the general acceptance test." In the years after the passage of the Federal Rules, courts have severely limited and modified Frye, and one circuit has explicitly rejected it. At least in the Fifth Circuit, Christophersen extends the Frye requirement of general acceptance to an expert's theory of causation in toxic tort cases.

On the one hand, the court was correct to recognize the importance of general acceptance in determining the validity of scientific reasoning梐nd that the Federal Rules of Evidence do not preclude such recognition. An appeal to general scientific acceptance is justified because a scientific consensus is a good indicator of the validity of a scientific theory. Courts that admit untested theories solely because an expert is willing to testify about them, without any inquiry into the validity of the reasoning, run the risk of allowing bad science into the courtroom. Such questionable but seemingly authoritative testimony often persuades lay juries to award damages for unfounded liability, especially in toxic tort cases. Moreover, an appeal to the larger scientific community is consistent with a conception of judicial competence to assess scientific claims. The general acceptance standard recognizes that courts, ill-suited for evaluating the validity of scientific theories, should defer to the scientific community's determinations.

On the other hand, the court was wrong to follow Frye to the point of making general acceptance a threshold requirement for the admission of a new scientific theory. Although general acceptance is clearly important, it is simply not true that a scientific theory is never valid until the scientific community endorses it. Few would argue that Galileo's heretical theory that the earth revolves around the sun was not "true" until it was later adopted by a majority of scientists. Toxic tort causation theories such as Dr. Miller's occasionally will be, like Galileo's, ahead of their time.

The importance of not allowing general acceptance to be an absolute requirement is highlighted when, as in toxic tort cases, scientific consensus is especially slow to form. Determining whether certain chemicals have caused specific diseases often takes enormous amounts of time and money, and scientific consensus develops only after considerable research and widespread publicity.

A better approach than the Fifth Circuit's extension of Frye-one that would give general acceptance its proper due-would have been to adopt a strong but rebuttable presumption in favor of excluding theories not shared by a substantial part of the scientific community. Only such a presumption can ensure that junk science stays out of the courtroom while the Galileos of the world, who come to court with a valid albeit novel, theory, are allowed in.