One of the most extraordinary things about the equations that describe planetary motion is that they allow us to predict the positions of the planets in the future. We can forecast solar and lunar eclipses with great accuracy. Furthermore, small deviations from predictions allowed astronomers to guess the existence of the previously unknown planets Uranus and Neptune. This predictive power is very impressive – and therefore very, very beguiling. In a complex and messy world, we like being able to predict things.
As a consequence, as David Orrell describes in his book Apollo’s arrow, Newton’s great achievement has led modern western society and most of its practising scientists to believe that all other natural phenomena also can be described mathematically and, more importantly, accurately predicted.
In his book Orrell focuses on present attempts to predict the behaviour of weather, climate, and the economy. The scientific challenge, following Newton’s lead, is to develop the correct set of equations to describe the system, and determine the solution on a sufficiently powerful computer. Then the future of the economy or the climate can be predicted, just like an eclipse of the sun.
There is one major drawback. Systems like the climate possess ‘emergent’ properties and a degree of complexity that means they simply cannot be reduced to a set of equations. Newton’s genius perhaps lay in the selection of a problem (planetary motion) that was completely solvable. Weather systems, climate systems, and economies, on the other hand, as Orrell demonstrates, are inherently uncomputable.
This does not stop scientists from assembling complex series of equations with thousands of manipulable parameters in their attempts to construct computer models of the global climate. But though we may be led to believe that these climate models are based on physical laws, they’re not. There are no equations that describe the formation and behaviour of clouds. We may also be led to believe that, with sufficient fine tuning of parameters and improved computer power, the maths of these models can be solved. It can’t.
Furthermore, Orrell’s work has shown that tiny changes to initial parameters in complex systems can result in significantly different predictions due to inherent model error. Mathematical climate models are permanently unreliable.
In the climate change debate, this is more than a little awkward from a green point of view, as it seems to put the climate modellers in the dock. Aren’t they the good guys? However, as Orrell reminds us, the climate model is not the climate itself. If the climate sceptics scoff because mathematical climate models are unreliable, that does not disprove the existence of global warming. Mathematical models of disease epidemics are also unreliable, but the epidemics happen all the same.
Nevertheless, Orrell is making an exceedingly important point that those involved in the climate debate must take note of: climate scientists will never be able to accurately predict what will happen with the climate. For that reason, there is nothing to be gained by waiting on the sidelines for the ‘expert’ debate over climate predictions to resolve itself. Because of the errors and inaccuracies in the models, there will always be enough ammunition for the sceptics to be able to exhort the politicians to do nothing about climate change until they get their impossible ‘proof’.
Assuredly, we don’t want to wait around for the incontrovertible proof that will come in with the tide one morning in the not-too-distant future. The observations of climate change and increasing atmospheric CO2 are all the notice we need. So we do not have the time to waste in waiting for a final mathematical ‘proof’ that can never arrive. Let us, therefore, politely ask the scientists to stand aside. It is time for political action – the radical action necessary to minimise the potential disaster of global warming.
David Orrell concludes his book with some broader observations on the philosophy of mechanistic science, and its consequences for our environmental crisis. So I shall finish with these ideas in Orrell’s own words:
We in the industrialized world still tend to see the world in objective terms, as something to be manipulated and controlled, slave to the laws of cause and effect. … The failure of our forecasting models, and the ancient dream to mathematically predict and control the future, grows out of this confusion between objects and living things.
Lack of predictability is a deep property of life. … The balance of positive and negative feedback loops, when combined with the computational irreducibility of life processes, makes the behaviour of complex life forms impossible to accurately model. … Life, it seems, evolves towards rich complex structures, which defy simplistic analysis.
By turning the world into an object that we can control, however, we also deny it life. And by closing off our emotional involvement with nature, we become unable to take the necessary decisions to protect it. We might be willing to buy the environmental argument intellectually … But as any good consumer knows, decisions are driven less by logic than by feeling. There is an intense drama going on … but we are still looking for the light switch.
(Also published in the US as The future of everything: The science of prediction. From wealth and weather to chaos and complexity.)