Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

William_Tanksley comments on No Safe Defense, Not Even Science - Less Wrong

14 Post author: Eliezer_Yudkowsky 18 May 2008 05:19AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (66)

Sort By: Old

You are viewing a single comment's thread.

Comment author: William_Tanksley 18 May 2008 05:13:44PM 0 points [-]

"@Caledonian: If it is an old and trivial insight, why do most scientists and near all non-scientists ignore it?"

They don't. The mismatch between you and them is that they're busy thinking about something else at the moment. I like the rule Turney gave above: "Doubt everything, but one at a time, not all at once." Of course, a single person can't follow that rule completely (there's not enough time in a lifespan to doubt EVERYTHING), and most people pick the wrong things to doubt or are lazy in applying the rule.

Of course, that rule's going to get in the way of reaching truth in some cases (some falsehoods come in self-reinforcing pairs both of which must be doubted in order to falsify either, and some things can't profitably be denied even for the sake of argument), but that's the case with any process, and this is something we've known since Goedel.

This kind of confuses me about this series... If all he was telling us was that Science is a powerful set of rules, and that therefore it can't eliminate all contradictions nor state all facts, I'd simply agree with him. But he seems to be saying that Bayesianism is different from Science, that somehow applying it instead of Science will have better results. It seems to me that both are processes, and both have blind spots.