The jury listens intently to the scientist on the witness stand. The judge has already decided not to question his testimony. After all, the scientist is an expert above reproach. Everything he says must be accepted as important evidence, right?

Wrong.

Behind the Ph.D., the lab coat and scientific rhetoric is a human being just as fallible as anyone else. And when you're dealing with a subject as broad as science, there is plenty of opportunity for error.

"Scientific literature is filled with reports that may or may not be true," noted Kenneth Foster, associate professor of bioengineering. "A famous Harvard doctor several years ago reported that drinking coffee was linked with pancreatic cancer. It turned out to be wrong. He recanted some years later. But that kind of evidence could easily be used in court to show that drinking coffee more likely than not causes pancreatic cancer."

In his original report, the doctor claimed that people who drink coffee have a 2.5 greater chance of developing pancreatic cancer than those who don't. That conclusion, in itself, doesn't prove much, according to Foster.

"That's a small number, a small signal, to measure with this kind of science," he said. But it's enough to cause public concern, which, in turn, can lead to lawsuits--even though the evidence is questionable.

"Scientific evidence is not always convincing, yet plays a tremendous role in litigation," Foster said. "There are cases where the plaintiff can't prove that he was injured, the defendant can't prove that he wasn't, and this whole murky mess is handed to a jury."

Foster's goal is to clean up this "murky mess." Through his writing and research, he demonstrates that you don't need to be a scientist to measure the validity of a scientific claim. You just need an open ear and a little common sense.

"My interest is conveying science to laypeople," Foster offered, "and writing about the problem with interpreting scientific evidence in nonscientific settings."

In "Phantom Risk: Scientific Inference and the Law" (MIT Press, 1993), Foster does just that. The book, which Foster co-edited, examines the tremendous legal problems caused by scientific controversy. Lawyers and scientists contributed chapters on subjects such as PCBs, asbestos and Bendectin, a drug used to cure morning sickness.

"Bendectin was taken by over 20 million women, and a fraction of these women had children with birth defects, and they thought that the drug caused it," Foster explained. "This lead to over a thousand lawsuits being filed, even though there was never any evidence that the drugs caused the defects. In fact, the evidence is pretty solidly favorable that the drug didn't cause the defects.

"Most of these cases were associated with a handful of scientists who made interpretations of the toxicology that they felt implicated the drug," he added. "In most cases, what these people had to say was not true, was not warranted by their evidence."

Although the majority of Bendectin cases never made it past summary judgments, Merrell Dow, the manufacturer of Bendectin, only achieved a partial victory. The pharmaceutical company didn't have to pay punitive damages, but it did have to pay huge legal fees.

"At one point, Merrell Dow made an offer of $120 million to settle a few cases, and the plaintiffs' lawyers turned it down, in effect, because they thought they could make more money going into individual trials," Foster recalled.

In addition to co-editing "Phantom Risk," Foster wrote a chapter on electromagnetic fields, his area of expertise. "There's a lot of public concern about electromagnetic fields," he said. "There's a lot of litigation about land-siting issues, a lot of litigation in local zoning boards when cellular-phone companies want to put up space stations or power companies want to put up transmission lines or substations.

"There are a few tort cases working through the system. The ones that have been pleaded have been found in favor of the defendants. Again, there's no real evidence that electromagnetic fields at any reasonable level cause injury. But the science is ambiguous enough that you could perhaps make a case."

"Phantom Risk" received such high praise from scientists and nonscientists alike, Foster decided it deserved a sequel. He and Peter Huber, co-editor of "Phantom Risk," just finished writing "Judging Science." The book's title is self-explanatory.

A Ph.D. in mechanical engineering and a former MIT professor, Huber now practices tort law in Washington, D.C., and writes a column for Forbes magazine. "Peter Huber has the idea, and it's probably correct, that a lot of this kind of litigation can be reduced by listening simply to what scientists have to say--judge the total value of the science, basically," Foster said.

The Supreme Court had a similar idea. In 1993, the justices handed down a landmark ruling that defined the acceptability of expert testimony. Their decision solved some of the problems associated with scientific evidence. It created some problems, as well.

"The Supreme Court justices gave a number of criteria--such as falsifiability, theories, potential error rates, these kinds of things--which they thought were hallmarks of validity," Foster said. "The question is: What do they mean?

"It turns out that all of these concepts are extremely problematic," he continued. "Scientists, themselves, don't know what they mean.... So how do laypeople judge scientific claims, and what criteria should they use?"

"Judging Science" attempts to answer these questions. Foster maintains that judges must examine the science when ruling on the admissibility of scientific testimony. Only then can they make rational decisions, allowing the jury to make a conclusion based upon relevant evidence.

"All of the criteria that the Supreme Court said that judges should use in judging science--validity, falsifiability, et cetera--are commonsense criteria that are similar to those a careful layman would use in judging any claims about the world," Foster explained. "The important point is that judges should examine the claims, not just let anything be presented because it is supposed to be 'scientific.' "

Foster's argument may seem obvious. However, he points out that judges who are intimidated by the "long list of initials after a scientist's name" often permit irrelevant scientific testimony during a trial.

"In this book, 'Judging Science,' we've actually looked at the testimony of some scientists, and it's clearly off the wall," Foster said. "It doesn't make sense from a scientific point of view. You could punch holes in it. It has factual errors. Some witnesses were clearly presenting personal views that weren't supported. It doesn't take a great scientist to see that."

If the average person can recognize the absurdity of unsubstantiated scientific evidence, then how does it manage to get into the courtroom? Public perception. People expect scientists to provide irrefutable answers, and this expectation can cloud judgments. As Foster points out, people are sometimes so concerned with getting a scientific answer, they don't always stop to question its plausibility.

Yet that's exactly what they must do. Due to its inherent uncertainties, science can rarely prove things beyond a shadow of a doubt. People must realize that before they place all of their trust in a single scientific claim. "Science simply cannot provide clear-cut, authoritative answers with the kind of precision that people need," Foster asserted.

This isn't to suggest that scientists have no place on the witness stand. Foster's point is that the public must not blindly accept everything that a scientist states. Instead, people must carefully consider all the facts, then draw their own conclusions. "People have to live with scientific uncertainty," Foster said, "but that doesn't mean that should paralyze decision-making."

Return to Compass Features for April 23, 1996