The participation of former White House chief strategist Steve Bannon at The Economist’s Open Future festival this month caused a great deal of controversy. But the actual interview he gave was fairly predictable. Free trade and immigration have been bad for the United States and Europe, argued Bannon, a view he shares with other right-wing populists. Interviewer Zanny Minton Beddoes, meanwhile, defended the liberal viewpoint which sees free trade and immigration having broadly increased prosperity. “We will have you back in a few years,” Beddoes said to close the interview, “when we will see which of these world-views proved to be the right one.” 

Beddoes’s closing statement might have been casual, but it revealed an interesting assumption: that these sorts of disagreements, between liberal and right-wing populist politics, can be settled by empirical evidence. Once the right-wing populist agenda has been implemented for a few years in countries such as Hungary, Italy, the USA, so this assumption goes, experience will show that following populist policies did not bring about the promised results, leading to disillusionment with populism. This is also the logic behind articles addressed at Trump supporters, highlighting the discrepancies between what he promised to deliver, and what he is in fact delivering. The implicit claim is that voting for politicians and policies is like adopting a theory about how the world works, one that experience can later either confirm or falsify.

This line of thought echoes the way the twentieth-century Austrian philosopher Karl Popper thought science works: Scientists put forward a theory which they then test against experience. If experience contradicts the theory’s predictions, the theory is “falsified” and should promptly be discarded. Popper saw science as the model of critical and rational thinking, always open to being shown that it was wrong, always accountable to empirical evidence. He also saw science as a model for democratic politics. In a democracy, the government should always be open to criticism, and it should of course be accountable to voters, who test the degree to which government policies work or not. If not, they get rid of them in the next election and vote in a new government.

The problem is that science doesn’t actually work that way—and neither do democratic politics.

In 1965, the International Colloquium in the Philosophy of Science in London, featured a debate including Popper and a young American historian and philosopher of science, Thomas Kuhn. Popper was by then 63 years old and an eminent professor of philosophy at the London School of Economics, whereas Kuhn was a 43-year-old academic who had failed to get tenure at Harvard, and whose first book on the philosophy of science had only been published three years prior. Unlikely as it might have seemed then, Kuhn turned out to be the more influential philosopher of the two, with The Structure of Scientific Revolutions selling over one million copies.

Whereas Popper saw scientists as always engaged in attempts to test the truth of the dominant theories of their time, Kuhn understood science as for the most part a conservative affair, with scientists conforming to the scientific status quo of their day. Kuhn argued that most of the time scientists work towards developing the available theories, rather than putting them on trial. More importantly, according to Kuhn, scientists stick to a theory even in light of observations that seem to contradict it. Having faith in the overall validity of the theory, they make allowances to explain away any apparent contradictions, putting them down to external factors that have nothing to do with the theory itself. A famous example of this is the now-discarded vision of an earth-centric solar system. Having accepted Aristotle’s theory that the sun and planets moved around the Earth in perfect circles, Ptolemy was faced with observations that showed the planets were moving differently. Instead of taking that empirical evidence to signify that Aristotle’s cosmology was wrong (which of course it was), Ptolemy postulated the existence of some extra circular motions the planets made, called epicycles, that resulted in his observations being made compatible with Aristotle’s cosmological thesis.            

In other words, scientists act towards scientific theories less like dispassionate referees, out to catch any mismatch between theory and experience, and more like partisan supporters who blame anything but their adopted theory. This resembles the behavior of the electorate, too. Instead of holding governments accountable to the results they promised, research shows that voters tend to exculpate the party they support, continuing to support them even in the face of apparent shortcomings. Given the levels of devotion that populist politicians inspire today, their voters are unlikely to recognize any future failures of their policies as evidence: Even if immigration cuts and tariffs fail to bring back jobs to the U.S. and raise workers’ wages, voters are unlikely to blame Bannon’s world-view. After all, there is already evidence that the policies he supports don’t work.  

But Kuhn did see a means through which true change could come—both in science and politics, whose revolutions Kuhn saw as fundamentally similar. Even though revolutions—profound shifts in thought—are a response to inherent problems, mere arguments and empirical evidence pointing out the problem aren’t enough to start them, Kuhn argued. A compelling alternative theory also needs to be available. Moreover, since two competing theories lack enough common ground to agree on an evaluation of the arguments and evidence, a level of faith in the new theory is needed, as well as persuasion by rhetoric and other non-evidence based means.

Although we are not yet facing a revolution, Kuhn’s remarks have a familiar echo in our current political moment. Despite addressing real grievances, the rising tide of populism seems to be less the result of rigorous argument and evidence, and more the result of effective rhetoric and the cult-like devotion populist politicians inspire in their followers. What is more, we are witnessing the absence of any common standards that opponents and supporters of populist politics can use to engage one another in an exchange of arguments, rather than talk past each other. Bannon’s interview was a case in point.

For those who wish to defeat the agenda of populism in Europe, the U.S. and beyond, Kuhn’s insights suggest a specific approach: The battle has to be won just as much at the level of rhetoric and persuasion as anywhere else. For better or for worse, people are easily persuaded when what you’re offering is a quick solution to a problem they’re facing. According to Kuhn, the Copernican revolution didn’t take place when new observations were made. It was the result of society’s need for a more accurate calendar, and astronomers’ faith that this new Copernican astronomy could quickly provide it. (In fact, it didn’t.) Rhetoric as a means of persuasion is also very effective. It is often looked at with suspicion, as the appeal to brute passion and emotion over reason, as a technique of manipulation. But even good arguments are more powerful when delivered with rhetorical verve, something effective change-makers have always understood. Despite Dr. Martin Luther King’s brilliance as a political thinker, it was the emotional impact of his speeches and his practice of non-violence that made him so influential in the struggle for civil rights.

Populists have been aiming very effectively at this emotional impact. Defenders of the liberal order need to as well. Simply waiting for experience to prove populist politics wrong won’t work—that strategy doesn’t even work in science.