20 January 2011

Tales to astonish

Douglas Green has a simple recipe for scientific success:

I can sum up in two words what it is that is asked of you, and really, everyone who works in science: Astonish us. That’s it.

Something about that advice niggled in the back of my head, bothering me. The more I thought about it, the more it bothered me.

What is astonishing? The surprising, the unexpected, the dramatic. There’s nothing wrong with wanting to do that kind of science. I love me a good breakthrough.

We’re already highly biased towards those kinds of stories. It’s easy to get carried away. For instance, saying that bacteria can live without phosphorus is an astonishing claim.* Saying that vaccines can cause autism is an astonishing claim. Saying that the third Monday of January is the most depressing day of the year is an astonishing claim.

But in science, we also put a pretty high value on claims being true. We have seen repeatedly that astonishing claims often have problems associated with them. They may be based on small sample sizes, or they can’t be replicated, or what have you. But because we often like the narrative the astonishing claim makes, it’s hard to overturn. This contributes to the file drawer problem. People have perfectly good research that never gets published because there are no statistically significant effects, or the findings contradict previous ones, or they’re just not deemed “interesting enough.”

Likewise, focusing on astonishment plays into the belief that science is just for geniuses. Part of astonishment is the feeling that “I never could have thought of that!” I’ve always hated, “What would [insert name of terribly clever person] do?” That never helps, because I am never going to have the same working knowledge and experience as that person.

This craving for astonishment feeds into the mindset that if it isn’t amazing enough to get into Science, Nature, or Cell, it isn’t worth doing. So entire foundational fields, like basic taxonomy, suffer because describing a new species of beetle isn’t astonishing.

It doesn’t surprise me that Green is addressing people who do biomedical research. The entire culture and attitude of biomedical research in the United States is largely driven by the funding practices of the National Institutes of Health. And the NIH funds individual projects in increments of a few years. In other agencies in other places, there’s more of an emphasis on supporting the careers of individual researchers. This used to be true of Canada’s Natural Sciences and Engineering Research Council, though I’m not sure if that is still true. I hope it is.

Focusing on the astonishing undervalues incremental progress. It undervalues hard work. It undervalues the long view.

P.S.—More tales to astonish!

* This case is different from the others I cite in that it is still being worked out, scientifically, and may yet prove to be correct. But it was a story that had “legs” because it was astonishing.


Genomic Repairman said...

This sensationalism of scientific work in top journals is making me become more and more of a believer in journals like PLoS. Is it accurate? Yes. Then put it to press.

NitrogenCyclist said...

great article.

PLoS is good; you may also want to try Frontiers journals ......