I've read about a quarter of the sequences, but I'm not sure if this topic has been addressed on LessWrong before. If it has, let me know.

The Upward Scaling Importance of Rationality goes like this:

The more influence your thought process and decisions have, the more important it is that you're rationalist. In the grand scheme of things, it is relatively unimportant that a barback at a restaurant is a rationalist, and I say this having done that. It is extremely important that a leader of a highly influential company, or a president of a university or country is a rationalist. Their decisions affect thousands if not millions of people.

The more influential you are, the more your decisions have potential to screw over other people. Influence doesn't necessarily have to be in a management position: elementary school teachers and police officers are highly influential, even though they aren't in control of an organization. Influence can even be by virtue of the people you reach out to. A famous person with a large fanbase or a parent of a child prodigy, both have the capacity to influence the world with their decisions.

Though arguably, this can  be extended to anyone who votes.

So rationality scales upward: the more influential someone is, the more important is it they're rationalists. Neglecting this can have bad consequences.

New Comment
10 comments, sorted by Click to highlight new comments since:
[-]Shmi90

This is true, but why privilege rationality? Integrity, kindness and other desirable traits scale the same way and are probably just as important.

[-]jamesf100

Because this is a site about rationality.

Kindness will only affect decisions where altruistic behavior wouldn't occur if lacking kindness. Integrity I'm even less sure about. Rationality could affect any decision where bias or fuzzy reasoning is involved, which is almost every decision.

More kindness and less rationality could result in spending billions on curing rare diseases of cute puppies. However, more rationality and kindness, that would work great.

It would be hard to do this more that we already are, since pretty much every disease you can think of has been "cured" in lab mice, and many of them in dogs. I was always wondering why we would spend so much money on treating housemice and dogs, now I know it's just the kind thing to do.

[-][anonymous]40

A response from an egoist perspective:

The more influence your thought process and decisions have, the more important it is that you're rationalist.

A professional and popular mixed martial artist must make many rational decisions, but his profession and popularity are more dependent on his physical strength, speed and fighter's heart. A professional and popular fashion model must make many rational decisions, but her profession and popularity are more dependent on her physical beauty and social charm. Some people are born with the potential for conventional physical strength and beauty and many are not. This is relevant because to say the punch of a fighter or the flashing smile of a model are also rational is to say either (a) some people cannot be rational or (b) everything is equally rational, 'ultimately.'

The more influential you are, the more your decisions have potential to screw over other people.

For a tyrant, or a spy, or a propagandist, or a soldier, screwing over other people is making decisions and influencing people. Rationality is as good for making other people suffer as it is for making other people prosper.

So rationality scales upward: the more influential someone is, the more important is it they're rationalists. Neglecting this can have bad consequences.

The more important to what person it is they're rationalist? To the rational agent, or to other people? It might be entirely rational to screw over other people. Fun, even.

Let's not presuppose that influential companies/people share the same terminal values: the big lesson of FAI is that one could very well be rational without being beneficial to humanity. It's also important to point out that, in the present scheme of our society, a more rational person is less influenced by more influential people/organization. Point in case: a large company like Marlboro could be very well be rational in their strategy of make it so that more and more people start smoking heavily, but a more rational person would search the medical data and avoid the danger, possibly donating to a charity that helps people quit smoking.

I can think of a lot of caveats to this (you don't want someone in power to be half a rationalist, or a strong rationalist whose values conflict with yours), but I think the basic principle you're pointing at makes sense. Something along the lines of: "the more influence a person's decisions have, the more one wants them to have good decision making methods". (Assuming that influence means control over things that you (or the person whose perspective you take) value.)

If we're defining rationality as winning (I get the sense that's still controversial?), then the best decision making methods a human could use would be those of a strong rationalist who shares your utility function.

A couple other things you could derive from this:

  • The more influence you have, the more useful it is for you to be a strong rationalist
  • If you are already a strong rationalist, you should try to increase your influence

I'm not sure if this has been addressed before. I think increasing your influence if you are already a strong rationalist has at least been implied, and 18,000 Hours is a specific case of it. I don't immediately recognize the version that you stated from any LessWrong articles, but Harry used that reasoning in deciding to befriend/corrupt Draco Malfoy in HP:MOR.

Would one apply this by focusing on influential people, or people who appear likely to become influential, when trying to make more rationalists?

The more influence your thought process and decisions have, the more important it is that you're rationalist.

Important, to whom?

If you're so very important, likely you can get away with being very irrational. Rationality is for the peasants. How tiresome it can be, to be bound by cause and effect. How relaxing it would be, to be "important", and to have all that tedious cause and effect business dealt with by underlings and hangers on. To merely have to wish, to have it so.

Do the people generally listened to, whose opinions have influence, seem more or less rational to you?

[-][anonymous]00

I wonder how strong the reverse relationship is: "The more rational someone is, the more likely they are to be influential". The only ways I can think of to supply something valuable enough that you become "influential" without exerting a fair amount of instrumental rationality are in entertainment and some kinds of writing, and even then it still seems like it'd help enough to give you a major advantage over whoever's competing for influence with you. Epistemic rationality is of course near-worthless to most people, and they're often right in that assessment.

[This comment is no longer endorsed by its author]Reply