In this case, however, what about situations where we can get Q ^ ~Q (sorry, unfamiliar with this formatting) without depending on P? For instance, the proof of EFQ:
1 (1) P A 2 (2) ~P A 3 (3) ~Q A 1,2 (4) P ^ ~P 1^2 1,2 (5) ~~Q 3,4 RAA 1,2 (6) Q 5 DN
Why isn't it a problem that line 5 does not depend on line 3, when it is the negation of line 3 that we are proving? If we take the above definition of RAA to be true, don't we need 3 to be a dependency of P ^ ~P?