My counterargument to Humbali would go like this: “Suppose I tell you I’ve already taken you might be wrong into account. If you ask me to do it again, then you can just do the same thing to my more uncertain estimate—I’d end up in an infinite regress, and the argument would become a statement that no matter how uncertain you are, you should be more uncertain than that. And that is ridiculous. So I’m going to take you might be wrong into account only once. Which I already have. So shut up.”
My counterargument to Humbali would go like this: “Suppose I tell you I’ve already taken
you might be wronginto account. If you ask me to do it again, then you can just do the same thing to my more uncertain estimate—I’d end up in an infinite regress, and the argument would become a statement that no matter how uncertain you are, you should be more uncertain than that. And that is ridiculous. So I’m going to takeyou might be wronginto account only once. Which I already have. So shut up.”