Science and consensus
There’s a tendency I’ve seen fairly often, and that features in a recent blog debate I’ve been involved in, to misunderstand how science works. This is a bad thing because it causes people to misuse scientific evidence and, even worse, to base political positions on incorrect or even dishonest grounds. The issue is what science actually is. Science is not, as is often imagined, the discovery of universal Truth. Rather, science is a search for understanding.
As such, it works through a simple but very powerful method – hypothesis, experiment, evidence, conclusion. That is, a scientist comes up with a hypothesis (such as “rocks fall at a constant acceleration when dropped”) and then carries out experiments designed to test that hypothesis. Based on the evidence produced by these experiments, the scientist then makes a conclusion (in this case, whether rocks really do fall at a constant acceleration). So far, so good. The problem is that there are a couple of complications that are not immediately obvious. First, science can only test things if we can ask scientifically meaningful questions – that is, the hypothesis must be testable. If the hypothesis is not testable then, however interesting or important it is, there can be no scientific answer. Scientists may nontheless hold strong opinions about what the correct answer would be! But we should not mistake these opinions for facts – only the raw data are facts, and they always require interpretation to make sense.
Second, and more importantly, the last step in the process (conclusion) demands a subjective evaluation by the scientist – does the evidence support the hypothesis or not? The problem arises because, in the real world, the evidence is rarely completely unambiguous, because it is hard to construct experiments that produce absolutely repeatable results or that bear directly on the hypothesis that you’re testing – in other words, we must usually either extract an answer from messy data (which involves selecting the “best” data and then using statistics to work out what they mean) or try and use indirect evidence (we cannot, for example, actually see whether light is a wave or a particle, we can only conduct experiments that suggest one or the other). All science does is to produce models of reality – simplified, idealised models but hopefully also accurate and powerful models.
This is all true even of the poster-child of scientific objectivity, of “absolute scientific truth”, the Law of Gravity. Sadly, Newton’s Law is wrong – or, more exactly, it is accurate only for small masses that are not moving quickly. Observations that showed Newton to be wrong were part of what drove Einstein do develop his Theory of General Relativity, which provides a more accurate picture of what gravity is actually like. However, even Einstein was wrong. Relativity fails on the very small scale, where quantum theory rules, and possibly on the very large scale as well (look up MOND, superstrings and quantum gravity if you’re interested).
This all means that different scientists can (and do) come to different conclusions based on the same data, because they bring different biases, different attitudes to risk and different levels of stringency to the table. As a result, the scientific community works on the basis of consensus – what do most people believe to be true? This is very powerful because, if enough different people agree on something, we assume that their individual biases are submerged by the divergence of views.
The outcome of all this is that science is essentially opinion. Informed, expert, communal, reliable opinion, but opinion nonetheless. And this causes problems when we need to use science to affect politics, because politics knows all about opinion and how to manipulate it. Science uses diverse opinions to arrive at consensus based strictly on the merits of the evidence, as judged by the majority. Politics, however, uses strong single voices to sway to majority to accept a single opinion regardless of the evidence. And this is not necessarily a bad thing because, outside the laboratory, the evidence is rarely as useful because it is vastly messier. More factors must be brought to bear on the issue in order to make sense of the evidence. Indeed, politics must take more issues into account – cost and feasibility, legal issues and international tensions. These issues must inform the eventual decision, just as the science should.
However, when science and politics must interact (as in the debate on climate change, but also in, for example, the management of fish stocks or the exploitation of oil reserves) this can be a problem. Politicians can always find a divergent opinion within the scientific community and then parade this single opinion as “proving” that there is real debate on an issue and that they can therefore ignore the huge weight of scientific consensus on the other side. And that is how politics works – exploiting difference to get your own opinion into a position of power.
Unfortunately, this means that science (and scientists) are ill-used by politicians to further their own aims, regardless of the majority view. And this might seem strange to those living in democracies. Surely we, of all countries, should accept that the majority view is the one that should be given the most weight. We base our entire theory of government upon this principle. When it comes to crucial issues of food, clean water, health and wellbeing, we should surely try to get our politicians to take account of the scientific consensus, rather than a minority view that happens so suit their prejudices. And when the issues are long-term rather than short-term, they should listen just as closely.
What we cannot do is listen to whatever voices happen to suit us and claim that “science” supports our actions. Either we listen to the consensus or we ignore it – but, if we ignore it, we must not claim to be acting in accordance with the science, either to ourselves or to anyone else. To do so is dishonest and will lead to disaster.
pax et bonum
Follow comments using Co.mments.com
Add to your del.icio.us bookmarks
