Tuesday, 6 February 2007

The difference between science and policy

A dirty little secret of mine - I don't have a TV. Haven't had one since, oh, Twin Peaks started - rushed about a bit in the throes of video-withdrawal trying to find what it was like, and then said "stuff it". Haven't missed it.

What I do do is watch DVDs on the computer. Works for me, even if I am a little behind the current events of "Lost".

So I'm watching the first series of "Numbers", for the first time. Charlie-the-mathematician is pondering an equation that might give a statistical chance of predicting "success" for children based on demographic data. The implication throughout the episode is that if the Big Bad Government gets ahold of this, it will use it to skew funding, away from the probable "losers" and towards the probable "winners". Charlie's friend advises him to drop it on the grounds that it will be a self-fufilling prophecy and is thus not science.

Sorry, no. There's again a confusion between science and policy there.

That the chances of a kid's success might be predicted by demographic data is a scientific question - regardless of what anyone does with it. That some people might be genetically better than others in defined areas is a scientific question regardless of the history of eugenics (and, really, if it's true for every domestic animal we've ever bred...). That women, as a group, might have different mental capabilities or tendencies than men due to genetics rather than socialization is a scientific question. That blacks and whites might be different is a scientific question, or at least it would be if anyone could define "black" and "white" in meaningful genetic terms.

That anthropogenic climate change may be occuring is a scientific question, regardless of what doing anything about it might mean for the economy. That neuro-linguistic programming as a description of a functioning mind might have something to it is a scientific question, regardless of the sordid uses it seems to be associated with.

Science asks meaningful questions of the world around us. These are available to be answered whether we like the implications or not. That you don't like the policy they may or may not support does not invalidate the science - cf creationism's continual confusion between evolution and all the myriad sins Fundamentalists like to rail at.

Making policy requires we look at the real world, that we understand the science. But the science does not determine our choices - we are not limited to the relentless march of logic. John Ralston Saul has laid this out better than I ever could - go read him.

So let's get back to "Numbers" - what happens if this is true? If we can predict which kids are more likely to succeed, the obvious policy implication is to skew resources to them in order to get more "bang for the buck". To seek efficiency.

As Saul points out, efficiency is a secondary virtue. First you have to know what goals you desire from a policy. To blindly promote efficiency because we have cold hard data enabling optimization for efficiency is to implicitly make a statement about the ends of education which it is difficult to support explicitly in a liberal democracy. An explicit debate on the purposes of providing education to children - ALL children - would be needed.

Therein lies the difference between science and policy. Science asks the "what is" questions, but policy must go further and explicitly ask "and how does that fit in"?

Not making this distinction leads to two deformations - a choice of policy just because the science is there - as was considered the case for eugenics - and a denial of science because of the implied policy - global climate change.

The rational thing for the anti-climate change people is to say "Okay, the science appears to support it - so what? Do we want to trash the economy to help alleviate environmental problems?" and move the debate into that area. They won't do that, because they're pretty sure they'll lose in an explicit examination of the issues - it won't trash the economy, and the environmental impacts may be more significant to actual human lives. So they smear the science and throw smoke bombs around.

No comments: