The Cutting Edge / COMPUTING / TECHNOLOGY / INNOVATION : Duplication of Research Isn’t as Bad as It Sounds
Want to know about the latest scientific findings on such issues as global warming, weather forecasting or gene research?
You can find several federal agencies involved in each of those disciplines, all seemingly searching for the same answers to the same questions. And that, according to the new congressional leadership in Washington, means that money can be saved by eliminating duplication in federal research.
Rep. Robert Walker (R-Pa.), the new chairman of the House committee on science, believes the best way to do that is through the creation of a U.S. Department of Science. The National Aeronautics and Space Administration, the Environmental Protection Agency, the National Science Foundation and a host of other agencies, including the U.S. Geological Survey, would be folded into the new department. The science czar would have jurisdiction over all federally funded research in the physical sciences, excluding defense.
Walker estimates that 5,000 research positions could be eliminated, some laboratories closed, and duplication virtually eliminated.
It sounds like a great idea. Why should different federal agencies carry out similar research? Why not eliminate duplication and save a bundle?
“I think it’s a bad idea,” says Daniel Kevles, professor of humanities at Caltech and one of the nation’s most astute science historians.
Duplication is the heart of good research, Kevles and others argue. No discovery, regardless of the credentials of those who claim to have made it, can be taken seriously until other scientists at other labs carry out the same experiment and reach the same results. Without a robust checks-and-balances system, research can quickly deteriorate from science to propaganda.
Most of us today believe, for example, that cigarette smoking is hazardous to our health. We don’t believe it because a high-powered federal agency told us to. We believe it because literally thousands of researchers with hundreds of organizations have reached the same conclusions.
Cold fusion, announced by two reputable scientists--at least they were then--at a major university a few years ago fell by the wayside because scores of other scientists at other locations were unable to get the same results.
Reducing duplication may seem like sound fiscal policy, but “efficiency is the enemy of good science,” Kevles says.
Indeed, the strength of the U.S. science program lies largely in its diversity. And duplication, which would seem to be at odds with diversity, is a critical part of it.
Some of the most important findings are made by obscure scientists at small labs: Much of the work on superconductivity, which allows the transmission of electrical current with minimal losses, has been carried out on the table tops of smaller labs, for example.
Scientists need the freedom to follow up on that exciting discovery without trespassing on someone else’s territory. And a wise policy would make it convenient for them to do so.
“A major strength of the American scientific system arises from its pluralism,” Kevles said recently as he sipped a cup of espresso at the Caltech student center. “With pluralism, you have a means of encouraging many different flowers to bloom.”
But no one knows in advance which flowers are most likely to bloom.
“You never know which kind of science, which approach, is going to be the best one,” Kevles adds.
It is easy to find examples of apparent duplication in federally funded research programs. Several federal agencies have their own weather departments. The departments of energy, agriculture, commerce, defense, and transportation all have weather research programs. So do NASA and the Central Intelligence Agency.
But the programs are not as similar as they seem on the surface. The Department of Agriculture needs to know about long-range weather patterns because that directly affects the nation’s food production. NASA, which learned through the Challenger disaster that it is unsafe to blast off when it is too cold outside, needs information that could affect its launch schedule--and it has tools nobody else has to carry out certain projects. The CIA, for its part, needs to know about drought and subsequent starvation, which can cause as many governments to topple as armed insurrection can.
“A lot of research is related to the mission of the agency,” Kevles says. “The Environmental Protection Agency supports research in areas of environmental science that are related to its mission of regulating the environment. The Department of Interior supports research that is related to its mission of protecting, managing and enhancing the natural resources of the United States.
“There may be overlap and duplication, but I think that’s healthy.”
Walker and others on the Hill argue that all those agencies don’t need to do their own independent experiments. A comprehensive research program under the Department of Science could provide the data more efficiently and with less duplication, they insist.
The work of the U.S. Geological Survey, for example, would go on, but it would be streamlined under the science czar to reduce duplication.
But one wonders. Will a massive bureaucracy care as much about research on earthquakes as an agency that recognizes seismic safety as its most fundamental reasons for being? I think not.
Duplication could be reduced, of course, and federal agencies in Washington are already working toward that goal. But some duplication is essential to good science. Is that too high a price to pay to get the right answers to difficult questions?
Lee Dye, a former science writer for The Times, has covered a broad range of science subjects for more than two decades. He can be reached via e-mail at 72040.3515@compuserve.com.
