That’s actually a surprisingly good reason. In real life, the best rationalist you know is probably not a character in a story and feeling a sense of opposing pressure when you disagree with them is probably a pretty good idea.
This should cause you to update down your view of Aumann’s Agreement theorem.
(I am reminded of many professional scientists tricked by charlatans when magicians were not fooled- because the scientists knew where to look for truth, and the magicians knew where to look for lies.)
Could you explain what you mean by this? I’m having trouble parsing “update down your view of”.
Aumann’s Agreement theorem is a neat true result about fictional entities. Its applicability to real entities is subjective, and based on how close you think the real entities are to the fictional entities. Increasing that distance makes AAT less relevant to how you live your life, and increasing that distance is what I mean by “update down your view of.”
My feeling is that those entities are really distant, to the point where AAT should not seriously alter your beliefs. “I trusted X because Y trusted X” is a recipe for disaster if you trust Y because of different domain-specific competence, rather than their deep knowledge of X.
I think the reason I was reluctant to accept that Quirrell is Voldemort is that Harry is a lot smarter than me and he trusted Quirrell.
That’s actually a surprisingly good reason. In real life, the best rationalist you know is probably not a character in a story and feeling a sense of opposing pressure when you disagree with them is probably a pretty good idea.
This should cause you to update down your view of Aumann’s Agreement theorem.
(I am reminded of many professional scientists tricked by charlatans when magicians were not fooled- because the scientists knew where to look for truth, and the magicians knew where to look for lies.)
I have updated by learning of it’s existence.
Could you explain what you mean by this? I’m having trouble parsing “update down your view of”.
Aumann’s Agreement theorem is a neat true result about fictional entities. Its applicability to real entities is subjective, and based on how close you think the real entities are to the fictional entities. Increasing that distance makes AAT less relevant to how you live your life, and increasing that distance is what I mean by “update down your view of.”
My feeling is that those entities are really distant, to the point where AAT should not seriously alter your beliefs. “I trusted X because Y trusted X” is a recipe for disaster if you trust Y because of different domain-specific competence, rather than their deep knowledge of X.
Right, ok. I’d already thought that AAT is essentially irrelevant to actual human behavior, so I was confused what brought it up.
ETA: No idea why you were downvoted so far.
On fictional evidence?
Harry is eleven.
I’m twenty-one, and I’m hell of a lot dumber than him in every aspect—despite having an IQ in the top one percent of humanity (135).
I generally expect that learning who to trust is something that comes from age and experience more than IQ.