How Much Dissent is Optimal?

The answer is not zero

The TL;DR for this essay:

• The authority to police disinformation will inevitably be used to crush dissent.

• Crushing dissent is bad.

• Therefore, we should not give anyone the authority to police misinformation.

To dissent is to express a belief or opinion or claim that differs from. . .conventional wisdom; the Party Line; elite opinion; the scientific consensus. In terms of freshman statistics, a dissent is the Alternative Hypothesis, which differs from the Null Hypothesis.¹

The Internet has given dissenters an ability to be heard and to find one another. If you dislike the increased visibility of dissent, then to you it is a bug. To others, it is a feature.²

Elite vs. Elite, Elite vs. The Rest

Let me define “elite” for this purpose as someone with both a college degree and an expected lifetime income in the upper third of the population. Well educated and well off, in other words. If someone does not satisfy one or both criteria, then that individual is in The Rest.

Let me define elite opinion as the opinion that most people in the elite would express out loud, if asked. Some remarks:

• There are dissenters among the elite, and their dissenters among The Rest.

• The Rest do not dissent from elite opinion about everything. The Rest are fine with “2+2=4” or “the earth is round.”

• Sometimes, what people in the elite really believe differs from what I am calling elite opinion, which is what they would say out loud. Timur Kuran calls this “preference falsification,” and it can lead to an “information cascade” as people suddenly become willing to say out loud what they truly believe.

After President Biden’s disastrous debate performance, people who privately had believed for a long time that he was cognitively impaired began to say it out loud. As another example, I suspect that privately many in the elite supported gay marriage before it became safe to say so out loud.

On the topic of race, out-loud elite opinion is that affirmative action by universities generally works well and that measures of average IQ by race are “junk science.” I suspect that private beliefs are rather different. I’ll bet there are many more people who agree with Amy Wax secretly than do so openly.

Who are the dissenters?

On some issues, dissent is concentrated in The Rest. Opposition to immigration is an example. In the UK, Labor won a “landslide” because Nigel Farage created a third party that broke with the conservative elite on immigration. As a thought experiment, imagine if the Republican elite had kept control of the party in 2016 and nominated a pro-immigration candidate, leading Mr. Trump to run on a third party ticket. This would probably have produced a similar “landslide” for Hillary in the Electoral College.

But there are issues where the significant dissenters come from the elite, with The Rest not much involved. Think of various efforts to make college campuses more hospitable to conservatives, such as Heterodox Academy or University of Austin.

Are there issues where elite opinion is actually conservative? I have a hard time coming up with any. Conservatives are in the minority within the elite, and according to my definition above, elite opinion means what most within the elite will say out loud.

There are issues on which radical progressives are dissenters. For example, their views on Israel and Hamas are out of synch with the majority of the elite. But I don’t see the elite as conservative just because the majority have not (yet) adopted the radical progressive viewpoint.

I am not saying that it is impossible for elite opinion to be conservative. I just have difficulty coming up with examples. So I think that dissent mostly comes from The Rest or from the elite who are on the right.

Norms on Campus

College administrators police speech, and I tend to think that they are entitled to do so. But I would distinguish between speech that threatens individuals and speech that merely offends members of particular groups. Giving offense without making a threat should be allowed.

This is not a bright line. There is speech that falls somewhere in the middle, which is not physically threatening but treats a student as unwelcome on account of his or her identity (“No Zionists allowed!”). I would lean toward punishing such speech.

As it stands, colleges are not doing what I would like. College administrators are much tougher than I would be against speech that is offensive to women, minorities, and LGBTQ. Jews often demand similar protection, and I do not believe that they should get it. But Jews are entitled to more protection than they have been receiving from speech that entails threats and intimidation.

Dissent lowers morale

Some people are nostalgic for the Walter Cronkite era.³ He would close his newscasts by saying “And that’s the way it is ___” followed by the day and date. As if there was no reason for no dissent.

But with a little reflection, you realize that the optimal amount of dissent is clearly not zero. The conventional wisdom is often wrong, and when it is wrong it helps to correct it. We need to allow dissent.

I think that one can make a case that too much dissent lowers morale. Tyler Cowen calls this the problem of excess negativity. When there is a lot of general distrust, pretty much everyone’s initiative gets stifled. Regulations get thicker. Veto points get stronger.

Think of an organization. Have you ever been in a unit where the mood is very negative? That quickly becomes self-fulfilling.

When I was hiring people, I came to realize that I needed a way to filter out hiring people who were chronically unhappy, because I saw how demoralizing they were. My tactic during an interview was to ask an applicant what they liked and did not like about their previous employer. If they jumped straight to complaining about their previous employer and never came back to answer the question about what they liked, then I took that as sign of a negative person. Now that I’ve learned the lingo of Big Five personality psychology, I would say that I was trying to filter out applicants with high neuroticism.

But censorship lowers morale

I would not want a chronic complainer in my business unit. But the last thing I want to do is censor complaints that well-meaning employees wish to surface. Censoring complaints lowers morale.

I believe that censorship by the boss comes from the same psychological place as chronic complaining by the employee. It comes from a lack of self-confidence. It reflects an inability to cope with the challenges we face in an imperfect world.

I believe that political censorship also comes from a place of weakness. Effective political leaders or scientists can overcome dissent without resorting to censorship.

Are certain types of dissent out of bounds?

People want to place some forms of dissent out of bounds. Holocaust denial is a common example. People want to say that Holocaust denial and similar beliefs should be put in a special category. Rather than calling these views “dissent,” they should be called “misinformation” or “hate speech,” and be subject to censorship.

The problem with trying to draw a line between legitimate dissent and misinformation is that some authority must draw that line. And the authority is likely to be biased, particularly against dissent that goes against the interests of that authority.

I do not find Holocaust denial threatening. I am confident enough that the evidence for the Holocaust will continue to be convincing.

What about Foreign Interference? What about the algorithms?

Foreign interference with U.S. elections is a bad thing. But I can only oppose foreign information warfare in passive voice. That is, in theory it would be great if it could be stopped. But once you tell me in active voice that such-and-such board is in charge of controlling information warfare, it becomes problematic. I fear that the board will be unable to resist scope creep, and we will end up with a censorship regime that protects (some) elites. I feel safer letting Russia or Iran try to run an influence operation than allowing a board to decide what content I am allowed to see.

Another issue concerns social media algorithms. Do I want to let evil corporations control what people see?

I dislike that the algorithms aim for engagement, and this rewards from fear-mongering and vitriol. As you know, I have a preference for an algorithm that grades on quality of reasoning and the ability to be fair to other points of view. But I would rather let the algorithms evolve in a competitive market than have “experts” decide what the algorithms should do.

So, yes, I do want to let evil corporations control what people see. It beats the alternative.

What about misinformation?

Dan Williams writes,

I think the “problem of misinformation” they identify is much more expansive and complex than they acknowledge. As they frame things, it simply boils down to the problem of human fallibility.

…Discovering the truth about reality is extremely challenging. Moreover, most information sources—whether mainstream media, politicians, commentators, pundits, and so on—are not exclusively or even primarily concerned with imparting accurate beliefs to their audiences. Given this, most communication ends up being misleading in some way.

Anyone can be wrong. The challenge is to arrive at beliefs that are more likely to be right. Trusting experts is a heuristic that is often a good choice. But it does not follow that dissent should be stamped out.

Substacks referenced above:

@

Can experts save the public from error?

1. I also use the statistics metaphor when I discuss educational interventions. The term “Null Hypothesis” plays a different metaphorical role here.

2. James Cham made this point in an email, helping to inspire this post.

3. Brian Leiter, for example.