More courage, less fear

Don't fight "cancel culture" by exaggerating its power

Nobody really knows what anyone means by “cancel culture,” but I think it usually pretty clearly has something to do with a climate of fear.

That’s why I had such a negative reaction to Sarah Owermohle’s April 12 article in Stat News titled “A controversial hire at Arnold Ventures raises questions about the donor’s stance on addiction.” The story is about Jennifer Doleac, a professor at Texas A&M who’s been a podcast guest of mine and who recently announced she’ll be leaving academia after this school year to join Arnold Ventures as a VP for criminal justice policy. The thrust of the piece is that some people don’t agree with one of her empirical papers in which she tried to estimate the impact of naloxone access laws on overdose mortality. Doleac, along with co-author Anita Mukherjee of the University of Wisconsin, found that while having more naloxone available does increase the survivability of any specific overdose episode, the aggregate impact is “more opioid-related emergency room visits and more opioid-related theft, with no net measurable reduction in opioid-related mortality.”

Not everyone likes this conclusion, and not everyone finds the paper convincing. The empirical turn in economics has been a positive development in a lot of ways, but it does often run into the reality that even with the best possible methods for causal inference, it’s hard to reach definitive conclusions. You can look at any of Doleac’s papers — or any comparable paper by anyone in the field — and poke holes in it if it has some normative implication that you’re displeased with.

But the Stat News article wasn’t really about academics quibbling over research methods, nor was it some kind of stunning exposé — it was simply broadcasting that it was problematic of Arnold Ventures to hire someone who had published a paper indicating that one popular harm reduction strategy might be misguided.

And the point of this kind of article is to send a message that this kind of conclusion is not welcome, that if researchers know what’s good for them, they’ll desk drawer that kind of empirical result and focus their energies on other areas. This tactic can be very effective. I know plenty of people who are ensconced in the progressive universe and who have doubts about some aspect of the progressive consensus that they don’t want to voice publicly because they think it will end badly for them. I don’t think those fears are entirely unwarranted and I see why people worry. But as someone who thinks that approach is bad for society’s epistemic institutions, I generally try to send a message that is reassuring and encouraging, that people should not be too afraid of bullies. My tweet about the Stat article was popular and well-received, for example, with hundreds of likes and dozens of RTs, many from journalists and academics across the political spectrum.

But not everyone takes that approach.

Exaggerated criticism can generate exaggerated fears

This is something I’ve been thinking about for almost a year now since I attended the Heterodox Academy conference in Denver last June.

HxA is basically an organization for conservative, libertarian, moderate, or just center-left-but-ornery professors and graduate students, and I think the mission of fighting for ideological diversity and open-mindedness in academic spaces is worthwhile. The conference itself often had a kind of support group vibe, and people who clearly felt isolated in their respective institutions were excited to network with other people. And that’s an important function for professional organizations to fulfill. Non-leftist professors arguably need some safe spaces.

At the same time, a bit ironically for an organization in which Jonathan Haidt plays a major role, this sometimes seemed to me to tip over into a kind of counterproductive pity party, with people telling each other spooky campfire stories about the cancel cult and the DEI gestapo.

There are different ways to frame this. The positive framing, of which Haidt himself is very much emblematic, is something like “look, a lot of people think this way, we can support each other, we’ve got some donors who are interested, and here are some constructive steps we can take.” But the negative framing sometimes, in effect, does the cancellation bullies’ work for them. After all, while it’s true that our epistemic institutions can’t function if everyone is compelled to toe the party line, they also can’t function if people are prohibited from issuing sharp critiques of those who dissent. The question is whether those sharp critiques create a spirit of dialogue or a climate of fear. The answer hinges in part on how the people involved respond, but also on what lessons actors in wider society draw from these episodes.

I found the mentality behind the Stat piece disturbing, but at the end of the day, Doleac is going to be running a criminal justice program for a major philanthropy.

By the same token, back in 2021, there was an insane Two Minute Hate directed at Samuel Norris, Matthew Pecenco, and Jeffrey Weaver for publishing an article that claimed to find that children end up better off on average when their parents are sent to prison.¹ Noah Smith wrote a long piece about it at the time, but to make a long story short, the Norris/Pecenco/Weaver critics were much more explicit in saying that one should never publish a pro-carceral empirical finding. But I think the most important point to make about this affair is that despite the attempted cancelation, all three authors continue to have good jobs at quality universities, and they’re grinding away with more prison-related empirical work like this paper about how being incarcerated reduces mortality.

If you publish an interesting, empirically rigorous paper with a pro-incarceration finding, people will absolutely get mad and yell at you. But you should still publish it! If you want to create an environment that supports better discourse, then you need to encourage everyone to be a little bolder and more heterodox at the margin.

And that means you need to try to be reassuring at the margin, not just feed their worst fears.

Marketing vs. problem-solving

That was my reaction when I saw The Free Press celebrating a Joe Rogan clip where he quotes Bari Weiss saying “When you’re not able to say out loud and in public that there are differences between men and women, the world has gone mad. When we’re not allowed to acknowledge that rioting is rioting and it is bad, and that silence is not violence, but violence is violence, the world has gone mad.”

I do get where she’s coming from.

At the same time, she said it on CNN. It was quoted, approvingly, on what I believe is the most popular podcast in the world. Weiss decided she didn’t enjoy being an embattled ideological minority at the New York Times, and she now runs an extremely successful publication. I know what Weiss is talking about because I also had moments when I felt like the world had gone mad and got into hot water at work for saying so. But the result is I now also make way more money than I used to, I have a good side gig as a columnist for a reputable news organization, and contrary to my fears when I first struck out as a Substacker, I still generally have my emails answered by people in politics and government.

When I said this on Twitter, I got a lot of pushback that basically amounted to “you know what she’s talking about.”

And I do! But I’m also paying attention to what she literally said. It would truly be a world gone mad if you were not allowed to acknowledge that rioting is rioting and it is bad. But is that what the world is like? I said that rioting was bad. Joe Biden said that rioting was bad, and he’s not rotting in a gulag, he’s sitting in the Oval Office. Conservatives like to pretend that Biden didn’t say that because, if true, that would be politically advantageous to them. Since a lot of people believe lies they are told by ideologically aligned media, that means lots of people believe this was a dark forbidden truth in progressive circles in 2020, but they’re just being misled. I think it would be accurate to say that a lot of media organizations went to extreme lengths to downplay the extent of looting and arson, but that’s different.

By the same token, you’re definitely allowed to say that there are differences between men and women.

The problem conservatives have is that they don’t think liberals are drawing the correct conclusions from this fact. Which is a fine thing to argue about. But “people disagree about the implications of gender differences for trans rights and for understanding sex-linked disparate outcomes” is a different and much-less-crazy world from one in which “you’re not able to say out loud and in public that there are differences between men and women.”

What you’re seeing is rhetoric for effect.

It’s marketing. It’s better business for me, for The Joe Rogan Experience, and for the Free Press to paint an exaggerated portrait of how persecuted and marginalized we are. Especially because it’s not a state of total non-persecution. The push to shun Joe Rogan was a very real thing that happened — there was a serious effort to pressure Spotify to break up with him or change his show. But it didn’t work, just like Dave Chapelle is still getting big-time gigs. Again, I won’t deny that there are problems in this space, but I want to make the point that the problems are not as bad as some people say. While exaggerated claims can be good marketing, they are counterproductive in part because they make more junior people more scared than they need to be.

Credible institutions are valuable

The flip side of all this is that I would like to urge everyone who wishes academic research were more influential in policy to recognize the importance of credibility. I read a cool paper recently that looked at different kinds of institutions and found that when they are perceived as ideologically biased, they are less trusted.

This is true of both left-wing groups (professors) and right-wing ones (cops), and interestingly they find that the distrust extends even to people who are ideologically aligned with the institution. I was glad to see that because I feel that way myself. I happily voted for Joe Biden, so it doesn’t per se bother me that the overwhelming majority of college professors also voted for Joe Biden. But I would generally feel better about the quality of policy-relevant academic research if I felt confident that every paper had been talked over by a Republican or two over the course of its lifespan. One reason I think the relatively-less-progressive discipline of economics tends to generate more valuable work is precisely that it’s less of a monoculture. Of course, you’d need to ask a psychologist or a sociologist to explain exactly why the dynamic plays out that way. But it’s still true.

Meanwhile, it’s worth revisiting the policy conclusion of Doleac and Mukherjee’s naloxone paper: “we conclude that naloxone has a clear and important role in harm-reduction, yet its ability to combat the opioid epidemic's death toll may be limited without complementary efforts.”

Is that even something anyone disagrees with? Again, the mildness of the offense speaks to the effort at bullying. It’s not like Arnold hired some kind of fanatical anti-naloxone crusader. We’re talking about a person who has published a lot of empirical studies of crime-related topics, and this one happened to have a moderately inconvenient result for one particular school of progressive reformers. Another Doleac paper, this one co-authored with Anna Harvey and Amanda Egan, speaks much more directly to a hotter political topic. It’s titled “Prosecutorial Reform and Local Crime Rates” and asks whether electing a progressive prosecutor as district attorney leads to higher crime.

This is an important question because there clearly are more shootings and murders in most of the United States than there were 10 years ago, and also a wave of progressive prosecutors has won election during this period. The rise in crime has generated some political backlash, and some people have tried to mobilize that backlash against progressive prosecutors. But crime has risen in lots of places, so can we really blame progressive prosecutors? Doleac, Harvey, and Egan “use variation in the timing of when these prosecutors took office, across 35 jurisdictions, to measure the effect of their policies on reported crime rates” and “find no significant effects of these reforms on local crime rates.”

That’s an important result for an important political topic, and while it probably won’t convince any conservatives, it very much might convince the kind of moderate Democrats whose opinions are critical to the electoral viability of progressive prosecutors. And this research is more credible because it’s done by someone who we know publishes her findings regardless of which way they come out. If you did successfully intimidate a larger circle of academic social scientists to either avoid crime-related topics or else avoid publishing any non-leftist results, you would undermine the ability of reformers to marshal credible arguments on their own behalf.

Let’s build a more sane world

All of which is to say that I really do sympathize with the broad point that the anti-cancellers are making.

But while exaggeration is useful for certain polemical purposes, fundraising, subscriptions, marketing, or heck, just having fun blowing off steam, it’s not actually helpful for the concrete purpose of creating a more open environment for scholarship and journalism. Bullies try to create a climate of fear, and telling everyone how bad and scary the bullies are is a way of positioning yourself as opposed to them while in practice reinforcing their power. In my experience, the average Biden-voting journalist has several heterodox views he or she is keeping under his or her hat out of an all-around desire to avoid contentiousness. But my belief is that giving in to that desire is actually a form of career self-sabotage, and the people in this situation would be better off being a bit bolder. And that’s what I tell people, encouraging them to do the work rather than endorsing their worst fears.

So I wish the anti-cancellers would chill out a bit, do a bit more helping and a bit less warning, and also try to be more precise and accurate in the claims that they are making.

I also, frankly, wish they’d open their eyes a little bit to the wider scope of events in the world. We have evidence from the Fox-Dominion lawsuit of producers calling election deniers “cousin-fucking terrorists,” of Tucker Carlson going on private anti-Trump rants, and much more. The fact that Fox, for some mix of business and partisan considerations, regularly airs such distorted coverage is a huge epistemic problem for the United States of America. Tons of right-of-center people watch Fox, and not only is the quality of Fox’s programming very low, but conservative (and more broadly “heterodox”) intellectuals don’t seem to be very interested in this topic, I think in part because they are self-censoring out of fear of being exiled from the club.

Such, unfortunately, is life in a sharply polarized country.

1. Note that this does not mean that the average child would be better off with an incarcerated parent, but rather that the marginal incarcerated parent — someone who was caught committing a felony — is an unusually bad influence on their kids.