What I meant is that as early as in grad school you start working entirely on your own and your advisor is just that, advisor. He is not going to prove your theorems for you. As far as I know in other fields it takes a while to reach this level of independence.
