WHEN 'SCIENCE' PLAYS POLITICS
By MICHAEL FUMENTO
NEW YORK Post
Opinion
February 28, 2006
LAST May, a Korean report in Science magazine prompted headlines around the world by declaring it had made tremendous advances in the heretofore disappointing field of embryonic-stem-cell (ES cell) research. News that it was faked has now prompted much soul-searching in media land. "How could we have been fooled?" reporters are asking themselves in print.
Well, wake up guys, because the major science and medical journals have been fooling you for years. And what appeared to be a trickle when I first wrote on it in 1999 has become a torrent.
In fairness, for many submitted papers it's quite difficult for journal editors and assigned peer-reviewers to spot data manipulation. This is especially true for that generated from a single lab. But not so if it's pulled from some public source.
Last September, after Hurricane Katrina, activists in lab coats saw a grand opportunity to tie the exceptionally violent hurricane season to global warming. A study in Science declared, "A large increase was seen in the number and proportion of hurricanes reaching categories 4 and 5."
But the researchers simply cut off their data at 1970, though public statistics go back to 1850. Using the full data set would have reversed the conclusion. Why did the editors and peer-reviewers at both JAMA and Science not insist on use of the full data set? Because slicing off inconvenient data is a time-honored tool of advocacy science.
Editors can even ignore papers in their own publication if it serves their purpose. A report in a recent (Feb. 17) issue of Science uses a computer model to show that glaciers along the coast of Greenland are rapidly melting and leading to rapid sea level rise; the study (naturally) blames global warming. Yet, just three months earlier, Science published a study based on actual data that showed increasing snowfall in Greenland was leading to greater ice accumulations than previously measured, slowing Greenland's contribution to sea-level rise.
When there's a political cause, such oversights come easy.
Yet published studies at least are subject to debunking. Try reading between lines that don't exist because journals refuse to publish them.
Such was the case this month when Science killed a paper at the very last minute by respected British scientist Peter Lawrence. It criticized "the cult of political correctness" that insists men and women are born thinking alike. Editor-in-chief Donald Kennedy explained it didn't "lead to a clear strategy about how to deal with the gender issue" — as if Science hasn't published countless papers on global warming with no strategy on how to deal with it.
In fact, of the hundreds of global-warming articles in Science and its British counterpart, Nature, just try finding one that doesn't take the doomsayer party line.
Some journal editors are completely unabashed about their chicanery. In 2004, The Lancet released ahead of publication and right before the 2004 U.S. presidential election an outrageous report claiming 100,000 Iraqi civilians had been killed since the U.S. invasion. Yet other calculations showed a range of 15,000 to 24,000 — and even Osama bin Laden claimed just "over 15,000."
No matter, the Lancet's editor took the opportunity to blast "democratic imperialism" and said "the evidence we publish today must change heads as well as pierce hearts."
Even Science's awful stem-cell embarrassment wasn't purely a matter of fraud. I have written repeatedly on how both Science and Nature have turned themselves into cheerleaders for any supposed advance in ES cell science, while opening their pages to laughable attacks on what many see as both medically and ethically superior — namely adult stem cells.
Perhaps the best explanation for why the Korean paper slipped by is that the editors so desperately wanted to believe it was real that they missed all the warning signs of fraud.
If you think I'm cherry-picking, consider I've only cited papers from three of the world's top four medical and science journals. Yet the problems with all these papers were readily apparent before publication and all reports were on politically-charged issues.
Bottom line: First, there needs to be an outside body of peer-reviewers not picked by the journals themselves. Second, the media need to stop treating medical and science journals as somehow sacrosanct. Like seemingly everything in today's world, they've gone political.
Michael Fumento (fumento@pobox.com) is a Hudson Institute senior fellow and author of "Science Under Siege: Balancing Technology and the Environment."
nypost.com |
