Tuesday, 24 April 2012

Spreading science's values

Last week I heard Dick Taverne (Lord Taverne in private life) give the Sense About Science annual lecture. He claimed that science increases democracy, tolerance and compassion - you can hear the whole lecture on the Guardian website here. He also said that scientists' values were irrelevant to the value of their scientific work. I challenged this - here's why.

Firstly some values are built into science. Science requires openness to new ideas, without which it cannot advance. It requires a willingness to listen to ideas from any source, since authority is a poor guide to truth. And it requires respect for reason and evidence, since we are all prone to believing what we'd like to be true. As Huxley put it "The great tragedy of science is the slaying of a beautiful hypothesis by an ugly fact." Science is a social process and it works best in societies that share these values.

Secondly, though scientists, like most of us, work for money, they have reasons for choosing to work at science. Perhaps the commonest reason is the intrinsic satisfaction of learning more about the world; of adding to humanity's store of knowledge. This matters. I see this store of reliable knowledge as humanity's greatest achievement but that's because I value reliable knowledge. That is a value. Many people, in practice, think money, or feelings or the confirmation of their prejudices to be more important.

And many scientists are motivated by humanitarian concerns or by concern for the natural environment. Of course having the right goals doesn't guarantee that you'll reach them or even do good science but it does guide at least some scientists in their choice of topics. It also helps to connect science with the concerns of everyone else - since everyone wants health and most people want an environment that is living not dead.

But here's the problem. Scientists can only do research that someone will fund. In practice that usually means a corporation, a government or a private philanthropist. The funders must value what science can DO, or they would not fund it, but they may not value what it IS. And they often have values that are alien to science. Corporations want to profit by monopolising some piece of knowledge. Governments want support for policies decided by prejudice or public opinion. Neither of these concerns is conducive to good science and some branches suffer severely from their effects.

So the values inherent in science are of limited effect unless they also guide the funding of science. That means holding corporations and governments accountable for the ways in which they fund, direct and publish research. It also means looking at what they fail to fund, misdirect and censor.

Science could be greater than it is if its values were shared by more of the people and institutions that shape our societies.


Jayesh Navin Shah said...

A discussion of values is also what the public demands of scientists - the Public Attitudes to Science 2011 study (which I worked on as a member of the Ipsos MORI research team) found that two-thirds (65%) of UK adults "would like scientists to spend more time than they do discussing the social and ethical implications of their research with the general public".

Yesterday, Taverne said this perhaps wasn't necessary (in response to Sir Roland Jackson) because scientists are already among the most trusted professions, so they don't need to start talking about their values to gain public trust. However, people may say they trust "scientists" as a broad group, but don't always trust the science (e.g. GM, nuclear power etc) or the institutions that scientists work for (which, as Taverne pointed out, overrides the trust in scientists themselves). One way scientists can overcome this, as public dialogues have shown, is to talk about their values, their enthusiasm for their work and the principles underpinning this - so talking more about these values is of utmost importance.

Julian Bennett said...

This is an excellent rebuttal of the idea that science is (or should be) value free.