Trump advocated using hydroxychloroquine; a lot of studies said it was unsafe. Now, because two of the studies that said it was unsafe have been questioned on the grounds of the motives of the people engaged in the study, many people are saying that Trump was right after all, and that shows that people who criticized Trump were wrong. I’ll begin by saying that I think it’s possible that hydroxychloroquine is a good choice in treating COVID—studies are all over the place, and I don’t have the expertise to assess treatment choices. But, even if all the experts and researchers end up deciding that it is a good treatment, Trump was wrong to advocate it. Because he doesn’t have the expertise to make that kind of recommendation.
The episode got polarized and factionalized quickly, and in a way that epitomizes most of what’s wrong with current American public discourse. For supporters of Trump, if it turns out that hydroxychloroquine is a valid treatment, then that would be proof that his critics are factional, and their criticism of him is irrational.
First, notice that that’s projection. That defense of Trump, that reframing of a question about whether it was a responsible thing for a President to say into a referendum on whether supporters or critics of Trump are entirely right—that is factionalism. (I’m not saying that they’re the only ones who factionalize everything; as I said, this situation is typical of American public discourse.) And the argument about whether it’s responsible for a popular President with no scientific or medical expertise to give medical advice is that it’s irrational for him to be making medical recommendations.
Second, if it turns out to have been a good recommendation, that doesn’t confirm that Trump was right to make it because he could have been wrong. Part of what is at stake in this disagreement is about how knowledge works. This is hard to explain, but what I mean is that, for some people, the world is a stable place that can be known, directly, by anyone with good judgment. A person with good judgment can, with no training or expertise, look at any situation and see the truth. That fantasy of judgment that transcends fields is often called “universal genius,” and it’s an important part of the myth of charismatic leadership.
For followers who are in a “charismatic leadership” relationship with Trump, the issue of hydroxychloroquine is a referendum on whether Trump has that “universal genius”—for them, if it turns out that it is a good treatment, then that is proof that is a person with that kind of untrained and yet universally valid insight.
Except it isn’t a referendum on whether he has universal genius. Because if it turns out to be an unwise treatment, his followers won’t abandon that belief and his insight. He’s already been wrong about lots of things, including whether hydroxychloroquine has harmful side effects. One of the more important midjudgments on his part, particularly relevant right now, was his calling for the execution of five innocent men. So, Trump doesn’t have universal genius (no one does), but that isn’t really my point. The important point is that, for people who believe in “universal genius” or information-free insight, it is a belief that can be proven, but not disproven.
It’s irrational.
“Universal genius” is supported, I’m saying, through a form of “survivorship bias”—a cognitive bias in which only the survivors are noticed. It happens in popular advice on success and business a lot. An article might look at “what the richest people know about success” or “how the most successful people manage their time.” Looking at what the successful people—the survivors—have in common doesn’t mean we can infer what caused them to be successful; they might simply have been lucky. Nassim Taleb has a nice analogy for this kind of bias. If a thousand people play Russian roulette, some (very small) number of people will manage to pull the trigger five times without harm. Interviewing those people to see what strategy they shared will not get us good information about how to win at Russian roulette.
If we only look at the time that a person happened to be right, then we can believe that a person has this extraordinary insight. And that’s what happens with arguments about police violence.
I was very puzzled after the Trayvon Martin murder because defenders of the killing said it was a justified shooting, since Trayvon Martin had texts describing himself as “gangsta”—a fact that George Zimmerman didn’t know. I kept trying to point out to people that playing a tough guy in texts does not carry the death penalty, and Zimmerman didn’t know any of that anyway, so Zimmerman’s shooting was racially motivated and reckless at best. He had no evidence for shooting Martin. Eventually, I came to understand that they saw the information about Martin that Zimmerman didn’t have as proof that Martin was a bad person, and they believed that George Zimmerman saw the signs. Zimmerman, they believed, had made an information-free judgment because of a kind of “bad guy” detector—an ability to read the signs of badness.
If pundits and reporters can turn up negative information about the victim—information the shooter didn’t have—then a lot of people will perceive the shooting as oddly retroactively justified. If the victim can be framed as a bad person, then they “deserved” getting shot, a desert that the shooter magically perceived via signs (rather than evidence).
This, like the issue of Trump and hydroxychloroquine, is a belief that can be proven, but not disproven. People who want to justify police violence look for information that the victim was a bad person and therefore deserved it, information that the police officer didn’t have at the time. And information about crimes that don’t have the death penalty attached. The officer, they believe, saw the signs.
The signs aren’t universally valid—that Zimmerman has a more problematic record than Martin isn’t retroactively proof that Zimmerman was the bad guy. The signs only point one way.
There isn’t “universal genius.” There isn’t information-free extraordinary insight. There is, however, confirmation bias.