- The Washington Times - Tuesday, January 25, 2005

EUGENE, Ore. - The cover of “Saving Science: A Critique of Science and its Role in Salmon Recovery” suggests a book about salmon. But readers who dive into the first chapter will find a deeper current of thought from Oregon stream ecologist Charley Dewberry.

Drawing from more than 25 years of research, Mr. Dewberry challenges modern science and its peer-review process, its view of objectivity, and the privileged status it has attained in popular culture.

“We get it from grade school on up,” said Mr. Dewberry, who is academic dean of Oregon’s Gutenberg College. “Science is the only way to know. … You have a hypothesis, you set up controlled conditions … you run the experiment, you get the numbers, you run the statistics, and you publish it.”

Mr. Dewberry says that while the scientific method can crunch numbers and perform a statistical test, placing research on a pedestal ignores other important tools humans use to discover truths about the natural world, such as logic, theory construction, experience and history.

In the end, science — and salmon — suffer.

The Pew Charitable Trust contracted a team of blue-ribbon scientists in 1995 to find the historical populations of coho salmon in western Oregon and Northern California.

After two years, the panel concluded it could not determine if a crisis even existed because salmon estimates in the late 19th and early 20th centuries were lacking. There wasn’t enough “scientific” data.

Historical evidence, however, was abundant. At the beginning of the 20th century, Florence, Ore., had three canneries on the Siuslaw River that sustained an average catch of 69,000 coho salmon per year, Mr. Dewberry said.

Assuming the catch rate was about 40 percent, the run in the Siuslaw was about 175,000 coho a century ago. Today, only 4,000 return to the river, according to the Oregon Department of Fish and Wildlife.

While such evidence clearly shows the coho salmon have declined, Mr. Dewberry said, scientists cast it aside as second-rate or anecdotal because scientific counts of salmon didn’t start until the 1950s.

“That gives you a horrifically skewed view,” Mr. Dewberry said. “It’s very difficult to prove that there’s even a salmon problem.”

The blue-ribbon panelists, he said, were trying to find scientific answers for a historical question. Instead of limiting science to current data, Mr. Dewberry said the researchers should have considered a broad range of information, especially from the past.

“Good history is the foundation of salmon recovery,” he said. “We’ve got to deal with these issues on the context of decades to centuries. We need to know what’s going on at those time scales. That’s a matter for history, not science.”

Mr. Dewberry said part of the problem lies in the peer-review process, which tends to discount anything other than data collected by trained researchers as unscientific — including historical information.

The National Science Foundation disagreed.Penelope Firth, deputy director of environmental biology at NSF, said panelists actually expect references to historical data in proposals, but with one catch. It must be peer-reviewed.

Barry McPherson, who worked with the Oregon Department of Fish and Wildlife for 29 years, said a lot of historical information doesn’t get into peer-reviewed journals because it may rely on people’s memories.

“You’ve always had people who said, ‘Hey, I remember when.’ … Some of them were right, and some of them were totally wrong. They had bad memories or biased their information.”

Mr. McPherson said the difficulty comes in sorting good historical information from flawed memories.

Mr. Dewberry said the deeper problem lies within peer review, which has limited its view of science to “objective” facts. But in practice, that doesn’t work, he said, because all knowledge is personal knowledge. Reviewers constantly rely on their own backgrounds and training when rejecting papers and awarding grants.

Without such personal knowledge, there would be no criteria for choosing anything, he said, and the scientists would end up randomly investigating the most common patterns on the planet.

“Science is much more of an art form,” Mr. Dewberry said. “It takes genius, and it takes skill. It can’t be reduced to this mechanical model. … There’s no such thing as ‘objective’ facts.”

Among those who disagree is Jim Lichatowich, who served on five independent scientific panels and studied salmon for more than 35 years in the Pacific Northwest. He said that while personal knowledge affects scientists’ decisions, objective facts do exist.

“If I say the sun is rising above the horizon, I think that’s a fact,” Mr. Lichatowich said. “But I think Dewberry is right, in that our culture, our experience, our educational background all influence what we see when we look at rivers.”

Mr. Lichatowich said that while bias may occur, the process still works.

“Peer review is not a perfect system,” he said. “But I don’t know what the alternative would be.”

LOAD COMMENTS ()

 

Click to Read More

Click to Hide