Friday, 29 August 2014

Politics and science...

"A soldier, having seen traces of a horse in the sand, will immediately pass from the thought of a horse to the thought of a horseman, and from that to the thought of war...But a farmer will pass from the thought of a horse to the thought of a plough." Spinoza

There's a tendency for some educated scientific people to consider that you can reduce the process of political decision-making to an exercise in the scientific method. Some sort of 'test' is conducted and the results mandate the policy decision. Most often the 'test' isn't an actual experiment (these are, as those scientists should know, pretty hard to construct in real human populations) but some sort of meta-analysis of smaller tests and analyses. From this process we arrive at statements like 'the science is settled' and 'nearly all scientists agree' despite the scientific truth being a whole lot more nuanced.

A related approach is to assume that a human problem identified by 'science' must require the intervention of the state for its resolution. At this point I can hear some on the left peeling away from the argument but bear with me because this really isn't about left/right or big/small government but about the role and purpose of two different things - science and politics. It would be a rum do if we used politics to determine the outcome of science. Why then do some people not see an equivalent problem in trying to use science to make political decisions?

Scientific and political questions are framed very differently. This is for good reason. A scientist would not, for example, ask "Should Scotland be an independent country? Yes/No". Yet this is an entirely valid question in a political context that will lead to a clear decision. A scientist might ask 'what would Scottish independence mean for X' and then try to construct some sort of test to answer the question. And the findings from that scientific enquiry might well help people answer the political question. But the science does not provide the answer.

The second point about that independence question is that is it absolute - the 'right' answer is that answer securing the most votes. Fifty Per Cent Plus One is enough. For science that isn't enough. Read any piece of good scientific research and you will see caveats - a section perhaps entitled 'limitations' setting out the constraints of the test being conducted and towards the end of the paper a piece on the need for further research. Good scientific research doesn't answer questions so much as turn one question into a myriad of other questions. Such an approach is fantastic if we want to understand the world but worse than useless if we want to decide whether or not Scotland should be independent.

So when a writer - as happens here - singles out one political ideology as peculiarly anti-science this is the result of either a profound ignorance of politics and the political process or else the presentation of a personal ideology as if it were scientific fact. Indeed, we can observe the same from other political perspectives - the adherence of many on the left to the view that GMOs are bad, nuclear power peculiarly dangerous and pesticides destructive is a fine example of how scientific evidence is routinely ignored.

If we are to 'respond' to climate change this does not mean that there must be more government, more regulation and some sort of crusade to stamp out any capitalism bigger than the corner shop or the local agricultural co-op. Those who choose to say 'I don't believe a word of it' also cannot be dismissed because they might just be right.  The problem is (and here's some science) that our ideology - political bias - means that we do one of two things: either we make choices that fit our existing bias or else we select particular findings while ignoring others again to fit a prior bias.

To return to that Scottish independence question again. Were I a 'Yes' supporter seeking evidence to support my case, I would select those studies, tests and experiments that support my contention that Scotland should be independent. I would also ignore the element of choice involved in political decisions. There are costs and benefits to every decision but political or ideological bias leans us towards emphasising only the costs or only the benefits.

To give another example, the argument that alcohol costs the UK £21bn each year is, say public health people, derived from scientific enquiry. It may well be a true figure. But against that figure we have to set an estimate of the benefits society gains from alcohol not just jobs, businesses and exports but the pleasure we get from a glass of wine or a pint of beer.

In the end we have become rather too dependent on scientific enquiry in answering (rather than framing) political questions. And the risk here is that the scientist, once removed from the constraints of experimental investigation, is likely to be just as ideologically biased as the non-scientist. Indeed, what we get too often is a complete misrepresentation of the ideology with which the scientist is disagreeing. To suggest that somehow 'libertarians' balance "individual rights against the rights of others" is a complete misrepresentation both of libertarianism and of the small-government conservsatism that the author was actually criticising.

One of the curiosities about what we might call the 'scepticism movement' is its ideological attachment to a sort of social democratic human engineering view of politics. I compare this to the commonplace view among the centre-right that everything would be fine if only we put businessmen in charge. For the 'sceptics' the solution appears to lie in sort sort of post-democratic technocracy where the task of politics is to implement the 'findings' of the scientific consensus and that politicians should be slaves to the selected and presented evidence. A side effect of this concept in politics is the modern idea of leadership - politicians' task is to 'lead' the reluctant and recalcitrant populace towards to consensus defined by the technocrats' 'science'.

None of this is to dispute the value of science. Rather I want to step back from the fetishising of scientific enquiry as the only worthy decision-making system. Instead of a "science says yes" process called evidence-based policy making, we need to understand that the purpose of the evidence is to inform our decision-making not to do that decision-making for us. And the point and purpose of democracy - lost in translation all too often - is for the people or the representatives of the people to make choices about things that can't be (or aren't) made in the market.

Like everyone else scientists are prejudiced, have irrational attachments to ideological positions and allow personal likes and dislikes to colour their thinking. I take the view - like many of those libertarians - that most of the time decisions are better taken in the marketplace where mutual benefit and added value determine the outcome. But this is an ideological position - one with a great deal of scientific support as it happens - and others will believe differently. As we've seen with the independence debate, scientific enquiry can only inform the choice that people have before them.


1 comment:

Nigel Sedgwick said...

This discussion on science and politics is interesting. Here are a few additional views.

Engineering is often described as the application of science and technology for the benefit of business and society. This leaves the word science to usefully express the concept of (a purer sort of) investigation for the purpose of 'mere' understanding. Thus engineering builds on foundations of science, but is different from it. There is also the expression that an engineer can make for sixpence what any fool can make for a shilling: thus introducing the importance, in engineering, of cost versus benefit. The main blog posting does not mention engineering. Though government has quite a few chief scientific advisors, it has no chief engineering advisors. Both absences worry me.

In the formulation of scientific method (and also and particularly mathematics), there is the concept of axioms: a very modest number of basic and intrinsic beliefs which are extended by gathering of evidence and logical deduction therefrom. A particularly important axiom for all natural science is that the 'laws of physics' (effectively the laws of the whole of natural science) are invariant across time and place - other things being equal. As is pointed out in the main blog posting, this does not really apply in politics (nor economics, nor any other social science) as there is no practical repeatability of circumstances at different times and in different places. Thus these social sciences are much less certain that the natural sciences. This must be remembered by scientist applying their skills in engineering, political and other societal ways (and, I think, often is). It must also be remembered by politicians, economists and other non-scientists when relying on social-science evidence to support their arguments as to what should be done. This seems to be forgotten more often than not. Sometimes, as exposed by Sokal in 1996, the 'extension' of science beyond its true contribution can be so extreme as to be ridiculous. There was a highly worrying example (though less extreme than Sokal's) of this sort of thing yesterday on the BBC Radio 4 Today Programme. The UN, seemingly, is going to listen to some 'ordinary people' on important aspects of world government, including the alleged Catastrophic Anthropogenic Global Warming (CAGW). As an example, we were given a lady from the Philippines pleading for help on CAGW, to reduce the increasing occurrence of (indisputably) destructive severe tropical storms (hurricanes, named tornadoes in the Pacific) - which she put down to CAGW. The BBC went along with this, totally ignoring the absence of any evidence whatsoever that the occurrence of severe tropical storms has increased with the level of atmospheric CO2. Thus, a supposedly scientific and logical argument skips one absolutely vital and evidentially measurable step of logical argument, but the UN and BBC carry on regardless in promulgating this person's view as sound scientific support for the political action she called for.

One thing the main blog posting does not really emphasise is the effective of the free market, on the eventual correct application of science and on wise political decisions. Having government decide on policy makes it a requirement that one solution (within a country or local government region) must be forced on all, no matter how well it fits the many different individuals, families and businesses. Having the market decide does (usually after a delay) allow everyone to contribute wisely to the 'societal' decision by their own (purchasing) actions. Where it is possible for different people to make different decisions for application to their own lives, this will happen: and so more people will get what they want, rather than some (or many) having a disbenefit forced on them. The current fuss from the EU's position on vacuum cleaner power limits is an example of this.

Best regards