NP != P for the same reasons as to why all formal models are wrong. NP = P would imply an a priori existence of a formal system rather than an organic heap defined arbitrarily into existence by context. The process of language acquisition and the non-existence of a "universal" and "natural" human language (viz. the language of feral children) gives the greatest proof of this.
Or via Chesterton:
The rare, strange thing is to hit the mark; the gross, obvious thing is to miss it. We feel it is epical when man with one wild arrow strikes a distant bird. Is it not also epical when man with one wild engine strikes a distant station? Chaos is dull; because in chaos the train might indeed go anywhere, to Baker Street or to Bagdad. But man is a magician, and his whole magic is in this, that he does say Victoria, and lo! it is Victoria."
If we end up "solving" the class of human-comprehensible NP-hard problems, it'll be because of the expansion of computational power to fully envelop the ergodic space of human cognition. Of course, at that point, there will have been created a literal deus ex machina incomprehensible to man, so it'll be a moot point as far as advancing science.