...
You called for more nerdiness?
If we're talking about the anthropomorphizing function still, why don't we think about it as a closure operation on a partially ordered set, where the ordering is ranked by humanness?
Verifying axioms...
1. f(f(wolf)) = f(wolf) => wolfman == f(wolfman) makes sense to me, because wolfman is already anthropomorphized.
2. wolf < f(wolf) = wolfman. Check, a wolf is less human than a wolfman, as it should be tautologically for any of the members of our set, given that we're talking about an anthropomorphizing function.
3. animal1 < animal2 => f(animal1) < f(animal2), in general, because I can't imagine the anthropomorphized version of more-human thing being less human than that of the "lower beasts."
Good. So we have a closure operation on partially ordered sets. Now I'm wondering if it gives us a topology... well, that's the case if f(empty set) = empty set, OR if the closure preserves the supremum.
...
Can anyone think of a way to anthropomorphize nothingness? I'm having a pretty hard time here.
Hm, I end up wondering if it's metrizable, or if it's anything better than Kolmogorov. We might be able to do calculus on this set, that would be cool...
Oh, and by the way... I'm of the opinion that you can't get more human than human.
Partially because it makes the math work better, and partially because humanness is defined by whatever human is.