> If nobody understands a mathematical proof, does it count?

That depends. If it deals with integers, yes. If it deals with real numbers, no.