Rationality Is Counter-Intuitive

Table of Contents

Rationality requires some attention be allocated contrary to what most people find draws their attention. If different creators each independently try to draw attention, that isn't a truth-seeking process. Getting popularity or attention are different things than having correct ideas. For a rational world, some people will have to sometimes pay attention to something that they initially dislike, or which initially fails to interest them, or initially fails to draw their attention, or initially fails to seem promising or valuable to them.

And some people will have to have a positive attitude while paying attention to something they're not yet impressed by. A bad attitude sabotages the attention people give something, so they'll probably conclude it's bad even if it's good. A bad attitude means not giving ideas a fair, unbiased chance.


One of the standard ways of paying attention to something you think may be bad is debate. It's typical to debate with people you think are wrong. It's typical to debate against ideas you don't spend your time studying and trying to learn from. It's typical to debate with ideas you rejected. And it's not rare to debate against ideas you think are really bad, and wouldn't give attention to other than through debate or outrage. (Outrage attention is widespread on social media platforms like Twitter, but problematic, not rational.)

The tradition of debate is one of the reasons people learn about other sides, other schools of thought, at all. If they didn't learn about something before rejecting it, then they'd do poorly in debate about it. You need to know enough about something to point out a flaw in order to be able to debate against it well. This suggests investigating every school of thought at least until you find a flaw – a reason to reject it – rather than just ignoring all but the one that seems most initially appealing to you or has some other incentive favoring it. (You don't have to figure out all the flaws yourself. Reading criticism that someone else thought of and wrote down is more common.)

By contrast to debate, in popularity contests (or contests for attention), people tend to just find stuff they like and never investigate alternatives much. You don't need to be able to point out flaws in alternatives; you don't need to be able to debate; you just need to be able to say "I like this one that I'm engaging with". If it's a relatively intellectual topic, maybe you'd be expected to give some positive reason you like it, which is different than being able to criticize alternatives, and doesn't require familiarity with alternatives.

Some people like debate. Liking debate is a way to have a positive attitude to paying attention to ideas you think are bad. People who like debate are a minority. And then, of the minority of people who like debate, the majority of them like debate in an irrational way: they like beating people. Many debaters use it as a substitute for war that doesn't get their hands dirty or risk their death: they see debate as a way to fight people and win victories. Many people talk about landing "blows" or even "punches" in their debates. One consequence of that mean attitude is they enjoy debate less when they lose, so they sometimes make very biased, unreasonable arguments to deny losing and avoid changing their mind.

A rational person should typically enjoy losing a debate at least as much as winning. When you lose a debate, you have more opportunity to learn. It's the so-called "loser" of a debate who gets informed about some better ideas and gets to change his mind and improve his thinking. Sometimes the "winner" loses some sub-points within the debate, so he gets to learn something too. Rational people seek out debate losses and get bored if their debates are too easy, and too repetitive, so it's hard to learn much. Debate losses, unlike wins, aren't repetitive (unless you lost a debate then kept the same ideas anyway).

Rational debate is about sorting through conflicting ideas to try to learn something about the truth and correct some errors. It should usually be mostly cooperative, and even when people have strong disagreements about long-held beliefs, it should still be more cooperative than adversarial. Explaining what you think to a debate partner, and answering their questions in productive ways that they'll understand, is inherently cooperative. And trying to criticize their errors is better done to help them learn better than to try to beat them. Arguments made in good faith, to try to educate not browbeat, are better arguments.


Another well-known, traditional way of paying attention to ideas you don't think are very good is research. One type of research where this comes up is surveying a field. Sometimes people try to look at all the ideas on a topic and get an understanding of them before deciding which are good or bad and why.

Often, people only survey the popular ideas in a field. I understand the need to save time and limit how many ideas they research. But popularity contests aren't truth-seeking. They should be dissatisfied with that method. More people should be seeking better methods.

Critical Fallibilism proposes a potentially better method: making organized debate trees that focus on decisive refutations of ideas. Ideas with no known refutations shouldn't be skipped. In other words, if no one has written down any refutation of an idea (that you know of), then it shouldn't be ignored. Every idea should get enough attention to find at least one error in it.

A milder version of this idea is to give every idea that much attention (find a decisive refutation) only if you can find at least one living advocate who is willing to respond to criticism and answer questions. If a school of thought has no one who will defend it, clarify issues, or learn from criticism, then it still might be correct, but there are major disadvantages to engaging with it. Those disadvantages provide a reason to spend your energy elsewhere.

Another reason to give a school of thought attention is for completeness in your own opinion. Like if you're looking at the possible positions on a topic, there might be several that you immediately think of yourself, and leaving one of those out would leave an obvious (to you) hole in what positions you've covered. You should survey a field in a way that seems reasonable complete and satisfactory to you (and to your best understanding of a hypothetical well-educated, smart, neutral person).

For example, if you're researching economic systems, you might think there could be high freedom, low freedom, or a medium amount of freedom. If you (hypothetically and unrealistically) only found literature advocating high or medium freedom, that would seem incomplete to you. You'd want to consider low amounts of freedom in some way before reaching a conclusion about economic systems. You would look for some more obscure literature discussing it. If you couldn't find any, or what you found had arguments that seemed unsatisfactory to you, then you could brainstorm some ideas favoring low freedom yourself. You'd want a third alternative to consider besides high or medium freedom, even if it wasn't popular, or else your research would be incomplete. Ends of ranges, like 100% or 0% freedom, also tend to be worth considering. Even if the end of a range isn't actually good, it can still be productive to analyze because it's a purer position that avoids compromises found at 90% or 10%. It's often useful to consider arguments that apply to the ends of ranges, then see how/if they apply to ideas near the ends of the ranges or in the middle.


Debate and research are two of our main tools for engaging with ideas that we don't have a great first impression of. They help make the world more rational. They are alternatives to popularity contests, social status hierarchies, and designing ideas to get attention (like clickbait articles).