For European readers, the fact that the Brexit referendum is finally over will be very welcome. Weeks of highly polarised campaigning have seen the rhetoric become increasingly strident and the personal attacks increasingly unpleasant. That, to the great surprise of many of us, the Leave campaign emerged the winner will not be the end of the matter.
Across the UK political scene, there are personal scores to be settled and, hopefully, bridges rebuilt. The issues raised will not disappear overnight and now it is down to the winning side to show their vision of a bright future is a realistic one. Meanwhile, in Brussels and across the other 27 member states, business as usual will restart for the time being, but this referendum may prove to be a seminal moment in the development of the EU. This could be the beginning of the end for the EU as we know it. Similar concerns to those raised in the UK will find a voice and the EU institutions will need to adapt their vision to what European citizens will actually countenance.
When it came to the crunch, the majority of British citizens will have voted with their hearts rather than their heads. Despite a welter of ‘facts’ from both sides of the argument, many undecided voters will have found the arguments equally convincing or equally unbelievable, as the case may be. This was probably exacerbated in many cases by their seeming certainty. Although many campaigners would admit that there were no certain, hard facts, there was no space for ‘could’ or ‘should’, only ‘can’ and ‘will’.
Politics is nearly all like this. One party offers a detailed, quantified vision of what it would do, another rubbishes it and comes back with an alternative proposition. Both may be reasonable, if they are based on credible assumptions, but both can’t be right (indeed, they may both be wrong). Science, we may think, is different, with researchers unwilling to commit themselves to a conclusion without very robust evidence. Not so. In some fields, scientists have also lost the use of the conditional tense.
I came across a piece on the BBC website this week headlined New crop varieties ‘can’t keep up with global warming’. The reported conclusion from a paper published in Nature Climate Change is that temperatures are rising faster than the crop breeding cycle can develop new varieties adapted to a warmer world. The lead author, Prof Andy Challinor of the University of Leeds, is quoted as saying "We can use the climate models to tell us what the temperatures are going to be."
Actually, you can’t. Models are, by definition, not the real thing. They are set up to look at what might happen given a set of starting conditions, and based on the current understanding of how a particular system works. They are used in widely different spheres of activity, from aeronautics and car design to economics. In the former case, shapes and construction details of aircraft wings are derived from models that predict the optimum design to provide a balance of lift, strength and performance under different conditions. These designs can be made up as prototypes for testing in wind tunnels – a significant step towards real world conditions – and tweaked as necessary.
In the case of economic models, the output can only be tested by implementing policy and is almost always wrong. Any deviation from the assumed parameters can have an impact and, in any case, the models themselves are imperfect. In addition, assessing the overall state of a national economy is fiendishly complex. Complete data collection and analysis takes so long that early stage measurements of quarterly changes are often corrected many months later. The models are the best guide governments have, but there’s also a lot of judgement and luck involved in the final outcome of policy implementation.
That is difficult enough, but climate models try to reproduce a far more complex system than a national or even global economy. What makes this even worse is the fact that real trends can only be discerned over decades and, even then, causal effects are difficult to discern. But the only ‘experiments’ we can do with climate are to attempt to change our own influence on it, in particular the current goal of radically reducing our species’ output of carbon dioxide and methane into the atmosphere.
The array of models used by the IPCC to project future climate change produces a range of outputs, despite the fact that all use the key assumption that the warming effect of increased levels of greenhouse gases is subject to a positive feedback as the air carries more water vapour and warming oceans release further carbon dioxide. The scientific way of communicating the results would be something like “the models show that average temperatures could rise by between x and y degrees by year z”; the prescription for action might be “to keep the likely rise in temperature to no more than x degrees by year z, the models suggest we should reduce emissions of carbon dioxide by a gigatonnes by year z”.
Instead, the model outputs are increasingly seen, or at least presented, as reality. True, we hear that temperatures are ‘expected’ to rise by 2, 3 or 4 degrees, but nearly all the dialogue is predicated on this rise being inevitable. It is then just a short step to such statements as "We can use the climate models to tell us what the temperatures are going to be."
This is akin to the depiction of the Remain campaign focussing on ‘Project Fear’, a picture of a disastrous future for the UK outside the EU. In the event, a majority of voters chose to ignore this message. Climate change campaigners similarly paint pictures of an apocalyptic future without radical action, even though temperature trends this century do not support this view. This debate has become a typical political one rather than being rooted firmly in science. Time to bring back the conditional.