Slightly Against Scout Mindset
In important matters, be a scout, if you're just shooting the shit at a party, be a soldier
1 There are things more important than being right
“Brilliance in a scientist does not consist in being right more often but in being wrong about more interesting topics.”
― Kent Beck
I remember in middle school, back in the days when I’d go and tell every person under the sun about my political views, I had one interesting conversation. I was chatting with a fellow middle-schooler, and, after being presented with my utterly overwhelming case for all my political views, they said roughly the following: “sure, that all sounds plausible, but like, we’re 13-year-old children, and so we shouldn’t place too much confidence in our judgments.”
I remember being pretty irritated by that answer. It had the infuriating property of being technically true—we were 13, and likely to be wrong. And yet…
The problem with this answer isn’t that it’s wrong; it’s that it’s boring. If every time you have a discussion, you just appeal to higher-order evidence about why you shouldn’t trust the judgments of that conversation, you’re not wrong. But you are boring.
And it’s not just that it’s boring—it’s actively unhelpful. I’d imagine that middle schoolers who spent their time thinking through these issues are more likely to form true beliefs than those who don’t. Thinking through complex issues in interesting ways isn’t just fun—it’s useful. I know about lots of random things merely because I was obsessed with politics as a middle-schooler—and this is true even though I was wrong.
David Friedman, as well as many others, has discussed the issues with Democracy. Voters have no incentive to be informed—informed voters have a chance of influencing the election roughly equal to their odds of being married to Richard Pryor at some point. Thus, just as people don’t spent that much time thinking about their wedding arrangements with Richard Pryor, neither do they spend much time trying to be right about politics.
If you’re wrong about politics, you bear no personal cost. Thus, voters have an incentive to signal with their vote rather than to vote for the right person. So being right about politics isn’t very important—and yet it’s almost surely more important than being right about most things. If a philosopher is wrong about the trolley problem or whether the correspondence theory of truth is correct, it doesn’t matter at all.
2 The scout mindset
“Learning to spot your own biases, in the moment, is no easy feat.”
― Julia Galef, The Scout Mindset: The Perils of Defensive Thinking and How to Be Right More Often
Julia Galef, in her book The Scout Mindset, talks about two mindsets: one of them is the scout mindset, the other is the soldier mindset.
The soldier mindset involves treating ideas the way a soldier treats their side of the war. When fighting, your goal is not to discover truth, not to discover things as they really are. Instead, the goal is to crush your enemies in whatever ways are possible. A soldier is not interested in having accurate beliefs, they’re interested in winning the war, damnit.
Contrast that with a scout. A scout wants their side to win the war, but the way they do that is by gaining accurate information. A scout wants to see the world as it actually is, to know where their enemy is, rather than merely crush their enemy. Thus, a scout’s aim is to gain information, a soldier’s is to win the war. Scott Alexander summarizes the distinction in the following way.
A Soldier’s goal is to win the argument, much as real soldiers want to win the war. If you’re an American soldier fighting the Taliban, you want to consider questions like “What’s the most effective way to take that mountain pass?” or “How can I shoot them before they shoot me?”, but definitely not “What are the strongest arguments for defecting and joining the Taliban?” Likewise, someone with Soldier Mindset considers questions like “What’s the most rhetorically effective way to prove this point?” or “How can I embarrass my opponents?”, but not “Am I sure I’m on the right side?” or “How do we work together to converge on truth?”
Scout Mindset is the opposite. Even though a Scout is also at war, they want to figure out what’s true. Although it would be convenient for them if the enemy was weak, if the enemy is in fact strong, they want to figure that out so they can report back to their side’s general. They can go on an expedition with the fervent hope that the enemy turns out to be weak, but their responsibility is still to tell the truth as they understand it.
Galef argues that we should be scouttier in our thinking. All too often, we see ideas as soldiers. We want to prove our side rather than be on the correct side. For example, Galef notes
When law students prepare to argue for either the plaintiff or defendant in a moot court, they come to believe that their side of the case is both morally and legally in the right—even when the sides were randomly assigned.
There’s lots of evidence that we’re naturally soldiers. Kahan conducted a study in which people were given math problems. Being better at math made them more likely to get the right answer. At first. Then, Kahan made the math problems politicized, so that getting the right answer meant betraying people’s political instincts. People who were anti gun control were given math problems, for which the correct solution found that gun control reduced crime. People who were pro gun control were given problems for which the correct answer found it increased crime. After this twist was introduced, being better at math made subjects less likely to get the correct answer. That’s… weird.
Well, people who were better at math were able to use their superior math skills to rationalize away the fact that the math contradicted their political views.
Another study found that Republicans and democrats have wildly inaccurate views of what their opponents believe. Republicans and democrats think that the other party is more than twice as likely to hold views that they consider extreme as the other party actually is to hold those views. Now, maybe you think that because you’re smart and educated you’re immune from this. Well, it turns out that more educated people were less accurate. Additionally, the more news people consumed, the less accurate they became. This provides solid evidence that we generally think like soldiers.
3 A scout or a soldier: which one to be?
“A journalist asked … what he would do if … observations failed to match his theory. [He] famously replied: Then I would feel sorry for the good Lord. The theory is correct.”
Imagine you’re deciding whether to believe your friend. You know your friend is probably wrong—a third party wouldn’t judge them to be correct. Should you trust them—believe that they’re right?
In this case, the answer is “it depends.” It depends primarily on whether being right is more important in the case than being a good friend. If a friend tells you that their ex partner is substantially more terrible than most ex partners, and maybe even a psychopathic monster, generally you should nod along. After all, it would probably bother your friend if you didn’t, and there’s no upside—being right doesn’t matter.
But if the fate of the world depended on a correct judgment about your friend’s ex, you shouldn’t believe them. In this case, being right is more important than being a good friend.
I think this also determines whether you should be a scout or a soldier. If being right is what matters, be a scout. If something else matters, being a soldier is generally better.
I think I’m a pretty impartial judge of most things. I don’t have particularly strong political views, I have probabilistic credences in various political views. Politically, I’m pretty scoutish. I’m also pretty scoutish about lots of philosophical views—I don’t have a super strong view about Millianism vs Fregeianism about proper names, nor about the causal theory of reference, nor about most things.
But I’m not very scoutish about utilitarianism. There’s lots of evidence for this. For one, I feel a rush of joy every time I find a new reason to be a utilitarian. Also, I feel a bit guilty assigning a credence to utilitarianism, the same way I do when deciding which of my friends I like the best.
But the biggest evidence for this comes from my behavior. I’m obsessed with defending utilitarianism. I’ve probably spent hundreds of hours writing in defense of it—I don’t do this about most things. Now sure, I could try very hard to be a scout when it comes to utilitarianism—to seek out evidence fairly on both sides, to form rational beliefs.
But that would make me boring! If I were a scout, I would survey the landscape equally. I wouldn’t spend hours thinking of the best new arguments for utilitarianism. I sure as heck wouldn’t write hundreds of blog articles arguing for utilitarianism.
Now sure, maybe this means by assessment of whether utilitarianism is correct is going to be unreliable. But that’s not what makes my writing worth reading, on the subject of utilitarianism. It’s not worth reading because I’m a wise disinterested impartial observer—it’s worth reading because I write with an obsessive fervor, and come up with lost of new arguments, as well as compiling old arguments. Soldiers are able to design creative mechanisms for their side to win the war.
If I didn’t go into normative ethics with the conviction that utilitarianism were correct, I might have abandoned it when presented the organ harvesting counterexample. Had I done that, I wouldn’t have discovered all the reasons the organ harvesting is not actually a good objection.
If you don’t go in with a conviction that your view is correct, oftentimes you’ll abandon the view in the face of some problem. But often the problem isn’t decisive if you really reflect. Even though I think the combination problem is plausibly lethal for panpsychism, I’m glad there are lots of smart panpsychists working on the problem—because if it’s right, we should know.
Noam Chomsky is probably a pretty motivated reasoner. He doesn’t seem to change his mind much, doesn’t seem open to entertaining other views as plausible. But our public discourse is enriched by someone like Chomsky—someone who dedicates his time to obsessively documenting the horrors of U.S. foreign policy. The soldiers can gather the evidence which eventually persuade the scouts.
I don’t know if David Friedman is that motivated—he seems pretty effective at avoiding motivated reasoning. But Friedman’s devotion to defending libertarianism and anarchy lead to one of the best books defending anarchy.
When you’re a lawyer, you want to be a soldier. You want to be single-mindedly devoted to defending your side. But oftentimes, the most important things people do is in their capacity as a lawyer—generating arguments.
The other thing—arguing is fun. It’s nice to have a pet issue that you argue about at parties. Scouts don’t have strong opinions or beliefs about things—they’re all probabalistic. But it’s nice to be able to answer the question: what are controversial beliefs you have? It’s nice for the answer to involve a particular issue that you’ve obsessively focused on.
I think lots of good has come historically through people with the soldier mindset. People motivated to find dirt on factory farms are a lot of the people who go undercover.
The quote I gave at the beginning wasn’t about some motivated reasoning shmuck—it was about Einstein.
A journalist asked Einstein what he would do if Eddington’s observations failed to match his theory. Einstein famously replied: “Then I would feel sorry for the good Lord. The theory is correct.”
If Einstein had just been a disinterested observer, he wouldn’t have so carefully and persuasively argued for his theories. Modern thinking may owe quite a lot to the fact that Einstein was very motivated to argue for his weird-at-the-time theories. I’d take ten people as motivated as Einstein who argue for their own weird theories, nine of which are false, over ten people who just sort of defer to higher order evidence because it’s probably right.
It’s a good thing that there are people with heterodox theories—people who have beliefs as crazy as general relativity was once considered to be, who try as hard as they can to find evidence for their theories. If no one is motivated to argue to new theories, science becomes locked in stale orthodoxy.
It seems like most EAs and rationalists have pretty scoutish mindsets. And as a result of this, they’re not very interesting to talk politics with. Their views tend to be a vaguely leftwing mishmash, without specific object level takes about anything. Okay, they have the virtue of being accurate in their beliefs, but they’re boring.
As section 1 pointed out, a lot of politics is signaling. You’re unlikely to change the outcomes of elections. But politics is fun. I think that the average person in the heart of Kansas who becomes a Democrat has a worse life. If you’re in a swing state, I think you should try to be a scout on altruistic grounds. But if you’re in California and your vote won’t change the election, just have fun with it!
Capturing Christianity is a YouTube channel that I sometimes watch. Now, the claim is that they are presenting the intellecual side of Christianity. It is very clear that they’re just trying to present a sludge of arguments, enough to allay the fears of Christians. But this isn’t necessarily a bad thing! It’s plausibly good that there’s a person whose goal is to try to compile all the intellectual reasons to be a Christian. Especially if doubts and deconversion are distressing—maybe it’s a good thing for people to remain Christian.
There is, however, one crucial domain in which people should be much scouttier. They should recognize that it’s very possible for other smart, rational people to disagree with them without being hateful bigots. They should tolerate everyone, even the outgroup. I’ve talked about this more here. But as far as dogmatically believing things and trying desperately to argue for them, I think that this is mostly fine on most important issues.
If you’re deciding the fate of the world, be a scout. But if you’re just shooting the shit at a dinner party, trying to impress a hypothetical girl with your encyclopedic knowledge of Gettier cases—the girl is hypothetical because real girls at parties aren’t impressed by my encyclopedic knowledge of gettier cases :(—then it’s fine to be a soldier.
Interesting article, but I'm gonna hyper-fixate on a very specific, totally irrelevant part of it: Fregeanism about proper names is obviously correct.
It seems we were incredibly similar as middle schoolers haha. I’ve changed my views a lot since then (for the worse, of course), so perhaps I should be more doubtful of my views now.