>>60
Because logic and the way proofs work.
P1 A
P2 A -> B
This allows me to infere:
c B

Yes P1 is a presumption, maybe, empirical false. But if P1 would be the cause, the inference itself is valid.