It was with a perverse sense of familiarity that i read in the daily papers in January that catastrophe was once again on the way. “If a huge asteroid crashes into the middle of the Atlantic Ocean,” wrote Pulitzer Prize-winning reporter John Noble Wilford in The New York Times, “say goodbye to Broadway, the beach house on Long Island and just about everything else on the East Coast as far inland as the foothills of the Appalachians.” Washington Post reporter Kathy Sawyer was in a similar funk: “While the economy is booming and there are no major wars,” she wrote, “scientists have come up with something to fill the worry gap: If a space rock three miles in diameter slams into the Atlantic Ocean, it would produce a towering, high-velocity wave that would swamp most of the upper East Coast.”
This “revelation” came from researchers at the Los Alamos National Laboratory, who apparently had too much supercomputer time on their hands-what with the downsizing of the nuclear weapons programs and the disappearance of Star Wars-and so had run simulations on a one-in-ten-million-years asteroid plunking into the Atlantic. The Los Alamos public relations people knew a good story when it fell from the heavens and had thrown a press conference to announce the results.
The ensuing articles were further work in a genre that can be called “Death from Above” stories, after a phrase made famous by the best-selling science writer Timothy Ferris in a New Yorker article of January 1997, that the editors, alas, saw fit to run as a tie-in to an upcoming television disaster movie of the same ilk. The gist of “Death from Above” stories is that asteroids and comets could do to us any day what they apparently did to the dinosaurs. While there are reputable scientists who confess to losing sleep over this, they will admit, if pressed, that in the recorded history of humanity no human being has ever been shown to have been killed by an incoming astronomical object.
“Death from Above” stories make good copy, but they are hardly science. On the contrary, they are pedagogical examples of the relationship between science and the press, which can be described as one of mutually exclusive philosophical challenges. Put simply, the role of the daily press is to report the news, which is by definition what’s new. The job of the newspaper science reporter is to write up the implications of the latest scientific paper-i.e., the one that officially comes out today-although only if the paper is sufficiently at odds with conventional wisdom or has some commentary on the human condition, which means sex, health, sports, aging, or money.
This, however, is fundamentally at odds with the nature of science, which is to establish what constitutes reliable knowledge and does so in fits and starts-false starts, generally, since most of them are either wrong or meaningless. In August 1996, for instance, when NASA scientists announced that they had discovered signs of life in meteorites that had apparently come from Mars (the genre here might be dubbed “We Are Not Alone” stories), a good bookmaker would have put the odds of them being right at 1000-to-1 against, because there were so many ways the researchers could have misinterpreted their data, and so few ways (one) that they could have had what they said they had. Remarkable results, after all, demand remarkable evidence, and the NASA data were decidedly unremarkable. The science reporters covered the story with the complete lack of skepticism it demanded if the goal was to keep it on the front page for a few weeks. Within six months it was obvious even to employees of NASA that the apparent signs of life were most likely artifacts of the experimental technique.
John Ziman, a physicist and philosopher of science, has suggested that the front line of scientific research, where reporters make their livelihood, is simply not the place to find reliable knowledge. He describes it aptly as “the place where controversy, conjecture, contradiction, and confusion are rife.” He quantifies his point by suggesting that the physics in undergraduate textbooks is 90 percent true, while that in the primary research journals is 90 percent false.
I once did my own calculation to this effect: In 1987, back when high-energy physics was still a viable field of research, I wrote an article for Discover magazine in which I tabulated all the discoveries in the field that had made The New York Times over the preceding decade. In nine of 12 cases the researchers had “discovered” something that turned out not to exist. The three discoveries that panned out were predictions of a theory that had been repeatedly validated and was already sufficiently conventional to be called the Standard Model. Not surprisingly, the nine errors were the more interesting claims, which means the better stories, because they were all at odds with the Standard Model or extended it deep into the unknown.