Why so many “experts” make such disastrous errors

If you want to discover why so many of our experts are getting things so terribly wrong, Giles Fraser’s interview on Unherd’s “Confessions” with the former Bank of England governor, Lord King, is essential listening.

King asks what it means to be rational in an age of deep uncertainty. Many experts, he says, have come to believe that mathematical calculations and computer modelling provide us with an authoritative representation of reality.

This is hopelessly wrong because such calculations can never encompass the uncertainties that help form human behaviour. Yet these experts have come to believe that everything can be thus quantified and explained. So anything that can’t be quantified is therefore useless; and anyone who challenges the quantification model is somehow suspect and is disregarded or even disrespected.

The belief that we can quantify everything and suppress whatever we don’t understand, says King, is part of the belief that humans can control everything. But these experts don’t factor in any understanding of how human beings behave – which may be very different from what such mathematical calculations and computers are telling them.

Human beings, says King, are good at dealing with ambiguity, mysteries and complexity. Behavioural economics, however, and its fashionable offshoot “nudge theory” (which has its own unit in Whitehall!) is based on the certainty that human beings will always maximise and optimise their opportunities – and if they don’t behave in this way, well, that must be explained by “bias”, more than 100 types of which economists now feed into their calculations.

But it’s the economists’ models that are themselves biased – against reality. As King says, they assume individuals operate only as individuals. In fact, people usually take decisions by talking to others. We tend to act collectively.

These experts make reductive assumptions in a world which is much more complicated than can be explained away by this balance-sheet mentality. As an example, King cites the Brexit debate which partisans on both sides reduced to a “bogus” row about numbers.

Remainers claimed to know exactly how much the public would lose by leaving the EU; Brexiteers claimed to know exactly how much would be available to spend on the NHS. In fact, says King, the NHS figure omitted the amount of spending on other projects which would be cut back; and, as many said, Remainers couldn’t possibly know how much worse off people would be. What was missed altogether by this argument over numbers, he says, was that the Brexit issue was actually a debate about values, about sovereignty and how the EU was developing.

King says he’s been on a personal intellectual journey. He was first introduced to computer modelling of the economy when he was an undergraduate at Cambridge. Now he realises you can’t solve a problem by numbers. He came to understand that inflation couldn’t easily be measured. What we think will happen to the economy might not be what actually happens.

Economics, he says, used to be defined by problems; now it’s defined by techniques and mathematical methods, which means every problem has to be fitted into the mathematical model. So if this model can’t explain what’s going on in the world, it’s assumed there must be something wrong with the world.

King is speaking principally about economics. What he says, however, is also applicable to other issues where so many scientific experts have got things so terribly wrong – such as the devastating errors made by some of them over coronavirus, not to mention the global madness over climate change (to which he refers in passing as characterised by “quite a lot of uncertainty”).

Rather than look at what was undeniably true about Covid-19 – that the combination of its exponential rate of infectivity and serious or fatal effects in a minority of cases meant any health care system would quickly become overwhelmed – mindlessly mathematical modellers fed inadequate abstract information into their computers and got disastrously false results.

One of the clearest and most dramatic examples was the study by Professor Tom Pike of Imperial college, London who predicted that at the height of the Covid-19 outbreak Britain would have 260 deaths a day.

Just a few days later, Britain’s daily virus death rate reached 260 and Pike was forced to eat his words and admit he had seriously underestimated the problem.

He had based his false calculation on the assumption that the outbreak in Britain would follow a similar trajectory to what had been seen in Wuhan, China. Among many aspects of reality that this failed to take into consideration was surely that the Chinese statistics are almost certainly a gross under-estimate.

Pike’s critics got it right.

“Alan McNally, from the University of Birmingham, said that it showed the perils of looking at numbers in isolation, without considering what we know of the virus. ‘Modelling and projections are fantastic tools but really all we need to do is look at the hard numbers we are seeing in other countries just a few weeks ahead of us. We properly introduced social distancing last week,’ he said. ‘Therefore fatalities will continue to double every three days for another two weeks at least given the lag time between infection and death.’

“Jonathan Ball, professor of virology at the University of Nottingham, added: ‘Models are based on major assumptions and often these assumptions are wrong. Until we do on-the-ground research — go out into the community and test exactly how many people have been infected with this virus — we will have no idea of the rate of severe disease or what proportion of infected people die.’”

Exactly. And the same point can be made about the apocalyptic predictions of global catastrophe through “climate change”, previously known as anthropogenic global warming (AGW).

Right from the start, the problem has been that these predictions have been based on computer modelling. This is wholly inadequate to the task, since the climate is one of the most complex, non-linear and chaotic systems in existence. No computer model can adequately deal with this. The information fed into these models is therefore by definition inadequate – which means that what comes out the other end is next to worthless.

And that’s before you factor in the biases and outright frauds of some of the AGW ideologues who occupy scientific positions – and all of whom who, just as Lord King says, assume that their computer models (and indeed, their whole framework of assumptions) constitute unchallengeable reality. So anyone who dares point out the factual evidence that contradicts the model is a “denier” who must be suppressed.

Experts are often very quick to claim that any critic who doesn’t belong to their particular discipline can be discounted as ignorant. Evidence, reality and reason, however, defy the attempted constraints of any discipline. The key point, articulated by Lord King about economics but with far wider application, is that both the social and physical sciences are now dominated by people who have redefined evidence, reality and reason into nothing less than their formulaic antithesis.

The evidence is all around us.