Angier, Natalie. "Court Ruling on Scientific Evidence: Adjust Burden," New York Times 1 July
1993, p. A12.
In making its first ruling even on the proper use of scientific evidence in the courtroom, the
United States Supreme Court has assured that a great many judges will be frantically playing
catch-up on the basics of scientific method.
Yet lawyers and scientists on all sides of the contentious debate over how much so-called junk
science is findings it way to juries said yesterday that Monday's decision, Daubert v. Marrell
Dow Pharmaceuticals, was a thoughtful, fair and rational response to a messy and complex issue.
In the case involving a morning-sickness drug blamed for birth defects, the High Court ruled that
Federal judges must insure than scientific evidence and testimony admitted in trials "is not only
relevant, but reliable." The decision could have a vast and immediate effect on cases that rely in
part on disputed scientific or technical matters, from the use of DNA fingerprinting and
computerized ballistic analysis as evidence in criminal cases to putative links between illness and
chemical emissions from industrial plants in civil cases.
"This ruling could affect almost every piece of litigation in the Federal courts from this point
forward," said Paul Rothstein, professor of law at Georgetown University in Washington, who
wrote a brief filed on behalf of the pharmaceutical company. "Today's world is so complicated
that experts are used only almost every issue that comes to trial."
The Court rejected the most rigid arguments from the defendant that scientific testimony could
be dismissed from a case just because it had not been previously published in scientific journals
or was not widely accepted by mainstream scientists. But the Justices insisted that this somewhat
liberal allowance did not mean that the courts should serve as circuses for any crackpot scientific
theory a self-described expert witness might care to offer.
The Court said that Federal judges must assume strong responsibility for the quality of the
scientific evidence presented in their courts, and that they must look at the data under question to
make sure scientific conventions were followed in generating the results. The Justices suggested
that Federal judges must consider the basics of scientific method - for example, whether the
theory presented is subject to reproducible experimentation without being falsified, a hallmark of
Lawyers and scientists agreed that the ruling could swiftly improve the caliber of science in the
courtroom, cutting down on the number of cases that rely on dubious scientific ground but that
lawyers might have tried to ram through to litigation because the had an expert with a university
degree willing to testify for them.
The ruling will also mean that a case cannot be dismissed simply because one side's expert
witness is a scientific maverick whose name does not often appear in the prestigious pages
journals like the New England Journal of Medicine.
"I can't think of how the court could have come up with a better opinion," said Steven G.
Gallagher, senior staff associate at the Carnegie Commission on Science, Technology and
Government, a research branch of the Federal judiciary in Washington. "The interaction between
science and low is extremely difficulty, and the Justices understood exactly what the problems
The Carnegie Commission has spent the last three years studying the nexus of science and the
courtroom, and hopes to have a preliminary guide available by the end of the summer to help
judges discriminate between good science and bad.
The Justices also insisted that courts should not base decisions about the admissibility of
testimony on agreement with scientific conclusion, but rather on whether the methods used to
reach the conclusion were sound.
"This means you don't determine merit by asking for a show of hands," said Edward
Imwinkelried, a professor of law at the University at California at Davis who was a consultant to
the plaintiffs in the Daubert case. "Judges and attorneys will now have to roll up their sleeves and
dig into such questions as : Has the theory been tested? What is the size of the data base? What
error rate was discovered in the experiments? Those are the right questions to ask."
Though the ruling applies only to Federal counts, state courts are expected to use the decision as
"In a sense, the real losers in the case are trial judges," said Bert Black, a lawyer at Weinberg and
Green in Baltimore who helped write a brief on behalf of the defendants. "It's going to make a
lot more work for judges to apply it intelligently."
Dr. F. Sherwood Rowland, president of the American Association for the Advancement of
Science, said, "The ideal solution would be to have judges who specialize in scientific matters,
but I don't see that happening for some time."
In the issue before the court, two families in California had brought a lawsuit against Merrell
Dow, saying sons' limb deformities were caused by Bendectin, an anti-nausea drug the mothers
had taken while pregnant. To support the case, lawyers for the family had called upon scientists
who took a novel statistical approach to show a possible link between Bendectin and birth
A lower court, the United States Court of Appeals for the Ninth Circuit, in San Francisco, had
dismissed the case, saying many epidemiological studies had failed to demonstrate any risk of
birth defects from Bendectin. The appeals court added that the unorthodox statistical approach
used by the plaintiff was not legitimate because it lacked "general acceptance" among scientists
and had never been published in a scientific journal.
The Supreme Court dismissed that argument and sent the case back to the lower courts, which
means that in the short term the plaintiffs are the victors. But whether the Bendectin lawsuit will
go anywhere once judges begin to scrutinize the methods behind the novel science in the suit
remains to be seen.