I agree. Philosophy gives you a level of abstract reasoning of the form: "if we agree (with Kant) that we should only take those actions which could be universal law, does it follow that the death penalty is morally justifiable?" There is some degree of reasoning from premises here, but all of the objects you deal with are things that you come into with a bunch of intuition that you never really leave behind.
On the other hand, something like:
> Given a one-dimensional invariant subspace, prove that any nonzero
vector in that space is an eigenvector and all such eigenvectors have the same eigenvalue.
really forces you to grapple with an entirely different level of abstraction
