So a gnostically rational person with low epistemic rationality cannot figure things out by reasoning, yet experiences being rational nonetheless? Could you say more about what you mean by ‘rational’ here? Is it something like frequently having good judgment?
Mmmm, these aren’t orthogonal dimensions within rationality. We wouldn’t call a person who happened to win all the time because they made the right choices without being able to explain why a rationalist; we’d probably just say they are wise or have good judgement.
By “rational” and “rationality” I want to point at the same thing Eli(ezer) does, which he also called the “winning Way”. It’s something like “the ability to take action that you are happy with” although I’d probably describe it in technical terms as “axiologically aligned intention”.
Rationality is an almost inherently epistemic notion because such alignment requires logical reasoning to judge, and in fact understanding how this works thoroughly rigorously seems to be the core of the AI safety problem. Thus even if someone could be accidentally rational without being a rationalist, this is something that is only interesting to those with sufficient epistemic rationality to assess it, and thus there’s not really a strong sense in which you can have someone with lots of gnosis of rationality who doesn’t also have episteme of it because they wouldn’t know rationality in any sense well enough to have gnosis of it.
So a gnostically rational person with low epistemic rationality cannot figure things out by reasoning, yet experiences being rational nonetheless? Could you say more about what you mean by ‘rational’ here? Is it something like frequently having good judgment?
Mmmm, these aren’t orthogonal dimensions within rationality. We wouldn’t call a person who happened to win all the time because they made the right choices without being able to explain why a rationalist; we’d probably just say they are wise or have good judgement.
By “rational” and “rationality” I want to point at the same thing Eli(ezer) does, which he also called the “winning Way”. It’s something like “the ability to take action that you are happy with” although I’d probably describe it in technical terms as “axiologically aligned intention”.
Rationality is an almost inherently epistemic notion because such alignment requires logical reasoning to judge, and in fact understanding how this works thoroughly rigorously seems to be the core of the AI safety problem. Thus even if someone could be accidentally rational without being a rationalist, this is something that is only interesting to those with sufficient epistemic rationality to assess it, and thus there’s not really a strong sense in which you can have someone with lots of gnosis of rationality who doesn’t also have episteme of it because they wouldn’t know rationality in any sense well enough to have gnosis of it.