Monday, April 20, 2009

Doubt

"Have you ever held a position in an argument past the point of comfort....given service to a creed you no longer utterly believed?" So asks John Patrick Shaney in the preface to his 2004 play Doubt. It's a very good play (and equally good new movie of the same name), about nuances and our tendency to fail to acknowledge how little we actually know. Did the priest...or didn't he?

Our society is currently commodified, dumbed-down, and rewards self-assurance, assertion, and extreme advocacy--belief. It's reflected in the cable 'news' shouting contests, and the fast-paced media orientation that pervades many areas of our society. There are penalties for circumspection.

Science is part of society and shares its motifs. Scientists are driven by natural human vanities, of course, but also by the fact that as middle-class workers we need salaries, pension funds, and health care coverage. We are not totally disinterested parties, standing by and abstractly watching knowledge of the world accumulate. Academic work is viewed as job-training for students, and research is largely filtered through the lens of practical 'importance'. Understanding Nature for its own sake is less valued than being able to manipulate Nature for gain. This was less prominent in times past (Darwin's time, for example; though at the height of industrialization and empire, when there was certainly great respect for practical science and engineering, he could afford the luxury of pure science).

It is difficult to have a measured discussion about scientific issues--stem cells are a prime example, but equally difficult are discussions about the nature of genomes and what they do, and the nature of life and evolution. Our societal modus vivendi imposes subtle, unstated pressure to take a stand and build a fortress around it. Speakers get invited to give talks based on what people think they're going to say (often, speaking to the already-converted). The hardest sentence to utter is "I don't know." (Another, "I was wrong," is impossible, so it isn't even on the radar).

A theme of our recent postings has to do with what we actually know in science, and what we don't know. It is always true that scientists firmly cling to their theoretical worldview--their paradigm, to use Thomas Kuhn's word for it. Maybe that's built into the system. In many ways that is quite important because, as we've recently said, it's hard to design experiments and research if you don't have a framework to build it on.

But frameworks can also be cages. Dogma and rigid postures may not be good for an actual understanding of Nature, and are costly in terms of wasted research resources. Clinging to an idea leads to clinging to existing projects (often, trying to make them larger and last longer), even when they can, from a disinterested point of view, be seen to be past their sell-by date. Great practical pressures favor that rather than simply saying enough is enough, it's not going to go very far, so let's start investing in something new that may have, or may lead us to insights that have, better prospects. Let's ask different questions, perhaps.

GM failed to respond to such a situation in their persistence to make gas-guzzling SUVs, and oil companies now try to fight off alternative energy sources. So science is not alone. But it's a concern, nonetheless.

Science would be better with less self-assurance, less reward for promotional skills, if experiments were more designed to learn about Nature than to prove pet ideas or provide findings that can be sold in future grant or patent applications. Most null hypotheses being 'tested' are surrogates for set-up studies where at least some of the answer is basically known; for example, when searching for genes causing some disease, our null hypothesis is that there is no such gene. Yet before the study we almost always know there is at least some evidence (such as resemblance among family members). So our degree of surprise should be much less than is usually stated when we reject the null. The doubt we express is less than fully sincere in that sense, but leads to favorable interpretations of our findings that we use to perpetuate the same kind of thinking we began with.

Real doubt, the challenge of beliefs, has a fundamental place in science. Let's nurture it.

No comments: