Thread by @juliagalef

For a long time I thought ppl were saying dumb things about rationality, until I realized they were using a different definition of rationality from me. I think their definition is something like "If social consensus is that an idea is unpromising, then that idea is irrational"

https://twitter.com/paulg/status/1090608908534800384

That definition of rationality leads to people saying things like... "Steve Jobs had the irrational belief that people could finish a project really fast if they tried. They didn't believe him, but he kept turning out to be right. That proves that irrationality can be beneficial"

The definition I'd been using is the decision science one -- forming beliefs about the world as accurately as you can, given your priors & the information available to you (epistemic rationality), and making decisions that maximize your expected utility (instrumental rationality)

Admittedly, under that definition, it's tricky to know for sure if someone's beliefs or decisions are rational. It's more of a normative ideal than an easily applicable criterion you can use to judge people's actual beliefs/decisions.

It's tricky because, e.g., you don't know what someone's utility function is. They might be more risk-seeking than you, or derive more utility from working on a weird/cool/unique project than you do. A decision that would be irrational for you might be rational for them.

Oh, another definition ppl use is "A decision is rational iff you can explain why it's good in a way that will make sense to other people."
Problem is, our beliefs about the world are based on intuitions developed over years of experience, which we can't easily articulate

Anyway, all of this is part of why I've been increasingly avoiding the word "rational". Too many hard-to-resolve misunderstandings.

To paraphrase Bertrand Russell, "Everything is ambiguous to a degree you do not realize till you have tried to get to the bottom of a disagreement"