Beyond Irrational Politics

Science will not depoliticize politics—nor should it. Policy is not a substitute for politics

What can a behavioral scientist learn by reading the politics section these days? There’s a lot of disagreement, of course, but that disagreement is taking a fresh twist. It’s not new to have differences about what we want to achieve, and differences about what we think are the best ways to get there. But now we disagree about whether something is even true. We don’t just hold different values or hypotheses—we hold different facts!

Political polarization has intensified to the point that we believe “our” people and disbelieve “others.” Populism has intensified to the point where we believe common sense and disbelieve expertise. Trust plays a critical role here: we give or withhold our trust based on who says something, not on what they say.

All this wreaks havoc on ideals of evidence-based policy-making. The grand vision of the policy sciences is that—collectively, as a society—we negotiate a common goal (usually about improving social welfare) and then we rationally weigh the evidence to figure out the best means to achieve that goal. In practice, any such clean division between goal-setting and fact-finding has always been uneasy, but the recent swells in political disagreement have made things even more difficult.

The response from some corners of the evidence-based policy circle have been to double down on evidence and rational processing: using science as a way to “depoliticize” the political sphere, reining in ideology.

Populism: outrage during a crisis of legitimacy 

These are many of the same people who are flabbergasted that a politician can openly lie in public and yet receive the support of voters. Lying should be an instant disqualifier for the purely rational voter. If a politician’s value is measured by how attractive their goals are and how credible their plans are to get us there, then lying should challenge our ability to even assess what a candidate would do if they were elected. And yet candidates (especially populists) who lie openly have managed to draw huge support.

A recent study by Hahl, Kim and Sivan explores this conundrum in a brilliant way. Their hypothesis is that a populist can actually make themselves more popular by lying—because they’re telling some kind of deeper truth. The experimental design primed respondents to affiliate with one of the two candidates: either the incumbent or the outsider. In cases where the incumbent was corrupt, voters rated the outsider as very authentic even when he made statements contradictory to established and openly agreed-upon facts—when he lied flagrantly. In the absence of the legitimacy crisis, the effect disappeared

Basically, if no one trusts the “insiders” who run the system, then the more you scandalize those insiders by lying openly, the more you show yourself to be one of “the people.” This setup works when there is a rift between the people and the establishment that is supposed to represent them, when there is a crisis of legitimacy. Such a crisis can arise under at least three circumstances.

  1. Incompetence:People feel that the government is working in the best interests of constituents, but they are ineffective at achieving their goals.
  2. Corruption: People feel that the government is working in its own best interests rather than those of constituents.
  3. Favoritism: People feel that the government is unfairly privileging some constituents over others. (E.g., the playing field is being leveled between a historically privileged group and everybody else, a.k.a. discrimination against the established class)

In any of these cases, the result is that a politics of resentment sets in, where “the people” (and who that refers to will be different in different cases) feel that the government is advancing a hidden agenda, which “the people” are powerless to stop in its advance. The vote for a populist candidate is an expression of outrage with the system as a whole, calling out the system as illegitimate.

The populist’s lie reveals a deeper truth

In their work, the authors talk about the populist’s lie containing a “deeper truth.” But how can a lie and a truth coexist in the space of the same sentence? They are awkward bedfellows. Maybe we need to dig a bit deeper into what it means for something to be true. Ernst Cassirer, a philosopher of the early 20th century, spent many years doing just that. His theories about language, myth and science provide valuable guidance through this contemporary conundrum.

Cassirer’s final book (The Myth of the State, a diagnosis of the intellectual roots of Nazism, written in 1945) talks about the Western belief that, since the Enlightenment, humankind has cast off its tendencies towards superstition. Perhaps our prehistoric ancestors believed in spirits and supernatural forces, but modern man—and at that time they apparently only talked about or among men—is a fully rational agent. If there are any residual superstitious beliefs or individuals in our society, they are surely just marginalia, leftovers from our past that will shortly be swept away by the winds of modernity.

Of course, the rise of Nazism illustrated the depths of our hubris. Put anyone in a dire enough situation and they will look to anything that provides a ray of hope. When “normal” functioning fractures, myth is only too happy to come spewing forth from that rupture. (The Milgram experiments also demonstrated that obedience to authority will push basically anyone to commit unconscionable acts, putting to bed any lofty, racist notions that perhaps the rise of Nazism was just a German problem.)

Expressing ourselves through myth

But what is myth? It is defined as the inherence of the part in the whole, and vice versa. Let’s digest that. Language is mythic when the word and the object are completely identified and indistinguishable from each other. The magic spell works because the incantation literally is the object it calls forth, containing all of the object’s causal powers within the word.

For example, in ancient Egyptian mythology, Isis tricks Ra into revealing his secret name, and in so doing she gains power over him. Children at summer camp do the same thing, trying to find out the “real names” of counselors that go by “camp names” instead—and lording it over the counselors when they do.

It works the same way with mythic art. The icon contains all the power of what it depicts. For instance, to own an image of The Prophet is to claim to have power over him (which perhaps gives some context as to why people get pretty upset when such pictures, especially unflattering caricatures, are published in newspapers).

Myth is all about expression. Each act and each representation is the full expression of its object, and of the power of that object. Our sign is our identity. By contrast, scientific speech is epistemic. Each utterance represents a fact, a state of affairs out there in the world about which one has knowledge.

Desperate circumstances call for desperate stories 

When Hahl and colleagues talk about the coexistence of the lie and the deeper truth in a single sentence, we can understand this by distinguishing the epistemic lie from the expressive truth. “The state of affairs in the world is not as I claim it to be, but my act—my disdain of the establishment—tells you everything you need to know about who I am: I stand for the people.” The lying populist can emerge only when there is a politics of resentment, when people put the expressive function of their voice above the epistemic function of their voice.

Cassirer highlighted that material conditions must never be allowed to degrade too far, that under such desperate circumstances myth would once again come to dominate the political sphere. What we are learning now with the rise of populism is that such crises can be equally well provoked by a symbolic crisis: a crisis of legitimacy, a sense among the people that the political caste is advancing its own agenda, one that the people feel powerless to stop. (It would also be worth considering what other conditions might trigger a symbolic crisis, beyond corruption or favoritism.)

Reason: only one piece of the puzzle

For evidence-based policy, we must recognize that facts and values cannot be cleanly divorced, as some hope. Science will not depoliticize politics—nor should it. Policy is not a substitute for politics, because avoiding material crises still leaves open the specter of symbolic breakdowns. Perhaps unintuitively, evidence-based policy depends on politics, it does not avoid it. Politics cannot just be shunned, it must be done better if we wish to see progress on the use of evidence.


The AI Governance Challenge

For behavioral science, much of its history has been cast as a reaction to classical economics, and especially as an invalidation of the presumption that humans are rational agents. Behavioral science maps out the conditions under which we are rational and the conditions under which we aren’t. Humans are quasi-rational agents. As long as we continue to frame these “departures” as “failures to be rational,” we tacitly endorse the idea that a perfectly rational agent is what a human ought to be—which means that we haven’t really thrown off the heavy mantle handed down to us from classical economics.

To truly move out from under the rationalist shadow, behavioral research and insights need to start mapping out humanity’s expressive tendencies, in addition to the epistemic tendencies that behavioral research has focused on so far. We need to explore what it is that we are doing when we are not acting rationally. Politics is providing a live demonstration that rationality is only one piece of the puzzle.

This blog post summarizes a paper delivered by the author (in collaboration with Chantale Tippett) at the 2019 International Conference on Public Policy.

The image in this post is of British fascists demonstrating in London in 1937. The desire to belong and the surge of emotions in a crowd can sweep up even the enemies of that ideology. Photograph: Daily Herald Archive/SSPL/Getty.

Read Next

farm technology agriculture

Behavioral Science and the Future of Agriculture

Agtech, or agricultural technology, can help farmers keep up with the growing demand for food—but uptake of these new tools has been slow. Behavioral science can help nudge farmers towards adopting them.


How Dogmatism Leaves Us Less Informed

People with dogmatic worldviews are reluctant to seek out new information to refine their beliefs. A new study suggests this might be rooted in fundamental cognitive processes.