Scientists, it’s said, behave more like lawyers than philosophers. They do not so much test their theories as prosecute their cases, seeking supportive evidence and ignoring data that do not fit—a failing known as confirmation bias. They then accuse their opponents of doing the same thing. This is what makes debates over nature and nurture, dietary fat and climate change so polarized.
But just because the prosecutor is biased in favor of his case does not mean the defendant is innocent. Sometimes biased advocates are right. An example of this phenomenon is now being played out in geology over the controversial idea that a meteorite or comet hit the earth 12,900 years ago and cooled the climate.
That the climate suddenly cooled then, plunging the Northern Hemisphere back into an ice age for 1,300 years, is not in doubt. The episode is known as the Younger Dryas, because in Scandinavia abundant pollen from a tundra flower called the mountain avens, Dryas octopetala, reappears in soil from this date, indicating that the forest had once more given way to tundra. With the sudden arrival of cooler, drier and less predictable seasons, early human attempts at agriculture in the Near East ceased, and people returned to nomadic hunting and gathering.
The cause of this cold lurch was seemingly settled some time ago when Wallace Broecker, a Columbia University geochemist, suggested that a North American ice sheet collapsed, flooding the Atlantic with fresh water, which interrupted the normal circulation of the Gulf Stream. Then a marine geologist, James Kennett of the University of California, Santa Barbara, said he had found evidence of the impact of a large object from space 12,900 years ago, in the form of carbon spherules in silt.
Dr. Kennett’s argument is that a swarm of meteorites punched through the atmosphere and caused a vast conflagration, filling the air with dust and soot. This shut out the sun, causing decades of continuous winter —sufficient to trigger an advance of ice sheets that, even when the dust cleared, kept the climate cool for more than a thousand years, at least in the Northern Hemisphere.
Dr. Kennett prosecuted his case with gusto, also suggesting that the impact had extinguished North American mammoths, just as an earlier impact had finished off the dinosaurs (a theory hard to reconcile with the survival of mammoths for thousands of years longer on islands off Siberia and Alaska, where hunters could not reach them). He suffered a key setback in recent years when several groups failed to find the right kinds of spherules or otherwise duplicate the results of his team’s work—and, worse, when a spherule sample from Younger Dryas rocks proved to be only 135 years old.
But spherules, dated to the right period, now have apparently shown up. Dr. Kennett and colleagues have published evidence in the prestigious Proceedings of the National Academy of Sciences that a “black mat” from the sediment of a Mexican lake dates to 12,900 years ago and shows a sudden peak of magnetic and carbon spherules, “nanodiamonds” of a kind known as lonsdaleite, and charcoal: all of it evidence of extreme heat.
Last year Michael Higgins of the University of Quebec published details of an underwater crater in the Gulf of St. Lawrence, arguing that it may also date from as little as 12,900 years ago. The crater, three miles across, has the characteristic central mound of a fresh meteorite impact. Its [impactor] was probably too small to shift the climate, but perhaps it was part of a swarm.
After the previous debacles, the jury will take much convincing that the new results can be replicated. But the burden of proof has shifted a little in Dr. Kennett’s favor. After all, Dr. Broecker and his followers, too, may be emotionally invested in his ice-sheet theory: Confirmation bias can affect us all.