In good science, error is simply part of the game. No progress is made without it. “Science thrives on errors-cutting them away one by one,” is how Carl Sagan put it. “False conclusions are drawn all the time, but they are drawn tentatively. Hypotheses are framed so they are capable of being disproved. A succession of alternative hypotheses is confronted by experiment and observation. Science gropes and staggers toward improved understanding.” Michael Ghiselin, a biologist and MacArthur Fellow, describes error as part of the overhead of doing research. “The best scientists,” he suggests in his 1989 book Intellectual Compromise, The Bottom Line, “can even be expected to make more mistakes than do the mediocre ones, for the best scientists do the most research. It is they who will work on the most difficult problems, and venture into the areas of greatest risk.”
The challenge for the science reporter is how to deal with the onslaught of fascinating-and quite likely erroneous-results. At times this chronic problem shows up in an acute episode like the infection known as cold fusion. In 1989, during the three months of hysteria surrounding the outbreak of cold fusion, a then-Washington Post science reporter described daily science reporting, especially during such periods of extreme activity, as akin to playing goalie in a hockey match. Pucks come whizzing at you fast and furious, he said, and most you block, but a few get by.
What is the solution? the science reporter can hedge his bets through the liberal use of caveats, but the editorial philosophy of daily newspapers works against caveats. When reporters add them to a story, editors are likely to move them to the end. Once at the end, the caveats can be easily cut when editors find themselves short on space.
Another way around the problem of sorting the seed from the husks is for the reporter simply to throw up his hands and say, “It’s not my job, man.” My favorite recent example of the lack of concern that some reporters attach to publishing bad science is that of the New York Times reporter who allegedly told a government expert on nuclear waste technology that his job as a reporter was not to decide what’s good science and what’s bad, but what’s a good story. (I say allegedly, because the Times reporter refused to speak on the record when asked to confirm or deny the remark.) He then went on to write a front-page Times article on a Los Alamos researcher who had concocted a theory that the proposed nuclear waste dump at Yucca Mountain might someday undergo a nuclear explosion. The buried radioactive waste would simply have to leach from its containers and form itself into a bomb with the help of natural forces. This required, in effect, nearly divine (or perhaps satanic) intervention. The Times reporter, however, did make sure that no pucks would slip into the net from behind by adding the requisite caveats and suggesting that even if the work was simply wrong (which it was) and could be debunked (which it would be), “the existence of so serious a dispute so late in the planning process [for the repository] might cripple the plan or even kill it.” It was the one irrefutable statement in the article.The best way, however, for science reporters to deal with the problem of giving publicity to the erroneous is to rely on experts. As Ghiselin puts it: “In the popular press, we are always reading that most scientists believe’ such and such. Who cares what most scientists believe? We want to know what the best ones believe, especially those in the best position to evaluate the topic at issue.” This last clause is a kicker. Most science reporters have their share of reliable researchers whom they consider experts, but it’s unlikely that any one of these will be an expert in the precise discipline of the latest research. What’s more, the more spectacular the announcement, the more likely that a scientist’s expertise will become problematic. If the discovery is truly revolutionary-which is to say, paradigm-busting-then by definition any scientist on the “wrong” (conventional) side of the paradigm is likely to lack sufficient expertise to understand all the ways the reported work is likely to be wrong.
Consider the cold fusion episode. Within three weeks of the purported discovery of room temperature nuclear fusion by researchers at the University of Utah, the pursuit had devolved into a nuclear version of the emperor’s new clothes. On one side were those scientists who believed Nobel laureate Luis Alvarez’s adage: “Only trust what you can prove.” They pointed out repeatedly that no reliable data existed to support the claim of cold fusion-let alone prove it-and that certain fundamental experimental procedures had been consistently ignored. The press treated these scientists as being firmly entrenched on the wrong side of the “new” paradigm. After all, most of them were nuclear physicists who had spent long years not discovering cold fusion; therefore they must be jealous. The rest of the skeptics were chemists, also tarred by their failure to discover cold fusion. That they did not embrace the new finding could only be because of hopeless self-interest.
Judgments like these render science reporting on most controversial subjects perilously close to anti-intellectualism. Science reporters tend to be fans of science who sincerely want to believe that there was once life on Mars, or that fusion power can be achieved in a glass of water. The experts have been trained to be critical, and they are easily seen as the arrogant eggheads we all disliked in junior high school. Non-experts quickly emerge to fill the vacuum, and they become invaluable resources to the reporter. Not only can you find a huge number of non-experts on any given subject, even a new one, but they are considerably more willing to give a bogus idea the benefit of the doubt, particularly if they stand to get funding to pursue research on the subject should funding agencies decide to go that route.
Although it would help if science reporters and their editors were more skeptical and relied more heavily on real experts, I’m not hopeful that the press/science paradox can be resolved. Indeed, because the press is primarily interested in the unconventional and the spectacular (“Man Bites Dog!”), it will always be easier to get press with bad science than with good. Bad science is inevitably more sensational than good science. Bad science has no boundaries: researchers can be sensationally wrong in an infinite variety of ways, whereas they can be right only in ways that are severely bounded by reality.
This is why even high-end journalism favors bad science: Bad science is the better story. So it is that a Princeton engineer who does ESP research gets five pages in The New York Times Magazine. A pair of Florida researchers who suggest that AIDS can be carried by insects, even though the disease doesn’t fulfill any of the requirements for a vector-borne disease, can get eight pages in The Atlantic Monthly. A theory that electromagnetic fields from power lines can cause cancer, even though the theory defies the known laws of physics and much of what we know about biology, can get 100 pages in The New Yorker. And these are the most literate publications in the country.
But if science writers can’t afford too much skepticism for fear of losing their jobs, readers, at least, can afford to be skeptical-and should be. As for me, I try to get through the morning papers by reminding myself of an old saying about the press. It goes something like this: “Trying to tell what’s going on in the world by reading the daily newspapers is like trying to tell what time it is by looking at the second hand of a clock.”