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

Andrew2 comments on Pascal's Mugging: Tiny Probabilities of Vast Utilities - Less Wrong

39 Post author: Eliezer_Yudkowsky 19 October 2007 11:37PM

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

Comments (335)

Sort By: Old

You are viewing a single comment's thread.

Comment author: Andrew2 20 October 2007 02:26:56AM 0 points [-]

OK, let's try this one more time:

3. Even if you don't accept 1 and 2 above, there's no reason to expect that the person is telling the truth. He might kill the people even if you give him the $5, or conversely he might not kill them even if you don't give him the $5.

To put it another way, conditional on this nonexistent person having these nonexistent powers, why should you be so sure that he's telling the truth? Perhaps you'll only get what you want by _not_ giving him the $5. To put it mathematically, you're computing p*X, where p is the probability and X is the outcome, and you're saying that if X is huge, then just about any nonzero p will make p*X be large. But you're forgetting two things: first, if you have the imagination to imagine X to be super-huge, you should be able to have the imagination to imagine p to be super-small. (I.e., if you can talk about 3^^^^3, you can talk about 1/3^^^^3.) Second, once you allow these hypothetical super-large X's, you have to acknowledge the possibility that you got the sign wrong.