APMason comments on What are your benchmarks of rationality? - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (14)
Well, after the list given in the OP (which, while they are in fact necessary conditions for rationality, seem to me to not even constitute a "lowest standard"; they're the surface-level attributes that are adopted almost automatically upon entering into the sceptic community) I tend to use their reaction when I say "everyone should be immortal". Strangely enough it does seem like you need an abnormal clarity of thought to reliably come to the right conclusion about death.
Beware inferential distance. "Everyone should be immortal" includes a lot of unstated assumptions that the person you say it to may not be aware of. They could easily think you mean "Everyone should basically be as they are now, except live forever", which would mean either malthusian misery or draconian restrictions on reproduction. Unless you have already discussed tranhumanism with them, this is a terrible benchmark.
That's not the way I usually phrase it - I don't know how that would fit into a conversation anyway. I was just summarising the subject matter. Sorry for the confusion.
"Everyone should be immortal" is a claim about values, not facts. There's no such thing as "the right conclusion about death".
I see this general idea espoused by rationalists rather often. But despite my months on here, I have yet to change my mind into agreement on this.
Aumann's Agreement Theorem leaves us with three options:
There are a great many people who don't meet the complete standard - in fact, the great majority of people don't; and it seems worthwhile to be able to differentiate between a reasonably rationalist deist and a Californian cloud cookoo-lander.
Of course, any way to differentiate amongst people who do meet the 'lowest standard' is valuable, as well.
I believe APMason's point is that your benchmarks are testing for anti-non-mainstreamism