So all semester I've had to cope with Hill's embrace of Dretske's theory of representation:
X represents property F if and only if X has the function of indicating F (from here)
I gather this is supposed to be an improvement on the purely causal/correlational/informational account (roughly, "R represents F iff R co-varies with F") because it (a) ( Read more... )