No single statistic has been used more effectively to suggest that our use of fossil fuels is leading to a planetary meltdown than this: Nine of the 10 hottest years in American history have occurred since 1995. Based on such evidence, who could dispute that the Earth is getting hotter and that humanity is to blame?

Turns out the statistic is wrong.

Only three years in the past decade -- 2006, 1999 and 1998 -- were among the top 10. Four others -- including the hottest of all, 1934 -- occurred during the Dust Bowl decade of the 1930s, before the burning of fossil fuels could have started ramping up appreciably heat-trapping greenhouse gases in the atmosphere. The 15 hottest years since 1880 are now spread out over seven decades, suggesting a process of ebbs and flows rather than the steady buildup that has been widely reported.

Here's where it gets interesting. A Canadian computer analyst named Stephen McIntyre discovered the error and informed NASA's Goddard Institute for Space Studies. The institute accepted McIntyre's calculations and changed the numbers on its Web site Aug. 7. It did not, however, issue a news release about them, and the mainstream media have largely ignored this finding. A recent search on the Nexis database found a handful of short news reports, a few editorials and a syndicated column by conservative pundit Cal Thomas.

One could argue that the institute and the mainstream media buried this story because there is less to McIntyre's findings than meets the eye. The new data pertain only to readings in the United States and do not contradict broader findings that as a whole, the past decade has been the warmest on record. (Supporting this observation while underscoring that the science here is not settled, the National Climatic Data Center says five of the hottest years have occurred since 1998.)

But this logic crumbles when you remember the symbolic weight attached to the idea of what The Washington Post, in a front-page article in January, called "a nine-year warming streak 'unprecedented in the historical record.' " That record, along with melting glaciers and extreme weather, has been a factual poster child used to convey the threat of global warming.

McIntyre, who had previously disputed the "hockey stick" graph Al Gore and others used as evidence of a sharp increase in global temperatures over the past century, did not challenge a marginal claim but, from a public relations standpoint, a central one advanced by those worried about the issue. Imagine if the reverse had occurred, if new figures found that not three but nine of the past 10 years had been the hottest on record. Does anyone doubt that this would have been plastered across every front page?

Name-calling is no help

It seems clear that the institute and the mainstream media buried McIntyre's findings because his data suggested an inconvenient truth. Without undermining the case for global warming, the findings complicated it. They forced those who believe global warming is the most urgent problem we face to admit an error.

It is easy, and necessary, to criticize those who have suppressed McIntyre's findings. But if we view the episode in a broader context, we can see that global warming is not just the province of objective science but also the rancorous realm of ideas and politics, dominated by two sides that have rejected nuance in the name of battle.

Those alarmed by global warming had good reason to downplay McIntyre's findings. Rush Limbaugh and other skeptics pounced on the new report as proof that human-made global warming is a hoax. Thomas used the incident to compare "global-warming fundamentalists" to those who believe in "Santa Claus or the tooth fairy."

Next page >