A major theme of The Will to Power’s treatment of science is the idea that, despite having largely stripped from our notion of nature that which reflects the contingent perspective of the observer, physics still -- insofar as it rests on sensory evidence -- inevitably reflects that perspective to some extent. Of course, to remedy this, physics tries to frame its description of nature in the abstract language of mathematics. But Nietzsche anticipates Bertrand Russell’s theme (to which I have often called attention) that insofar as physics gives us only knowledge of the physical world’s mathematical structure, it does not give us knowledge of the intrinsic character of that which has that structure, and thus actually tells us relatively little about objective reality.He concludes that one would be
hopelessly naïve in supposing that  and that science gives us much in the way of objective knowledge.This I find puzzling. What does Ed mean by the 'intrinsic character' of, presumably, the physical world, and what does he think possessing knowledge of this would be like? There is a clue in a Nietzsche quote:
The calculability of the world, the expressibility of all events in formulas -- is this really "comprehension"? How much of a piece of music has been understood when that in it which is calculable and can be reduced to formulas has been reckoned up?The performance of a piece of music can be a complex business, as can our response to it. Maybe there's not much more to comprehend in an electron other than its mass, charge, and spin? Nietzsche and Ed seem to think that there is: that the world might be knowable in a way analogous to the way a piece of music or perhaps a person can be known. A rather romantic idea, I think.
Nietzsche also says,
It is an illusion that something is known when we possess a mathematical formula for an event: it is only designated, described; nothing more!Perhaps, in the end, description is all we can achieve, in any domain. I could be content with that.
Mechanistic theory can… only describe processes, not explain them.