Penrose, 3.3:
Mathematical ideas develop, and various types of problem seem to arise naturally. Some of these … can lead to an essential extension of the original mathematical concepts in terms of which the original problem had been formulated… accordingly, the development of mathematics may seem to diverge from what it had been set up to achieve, namely simply to reflect physical behavior. Yet, in many instances, this drive for mathematical consistency and elegance takes us to mathematical structures and concepts which turn out to mirror the physical world in a much deeper and more broad-ranging way than those that we started with.
Penrose certainly doesn’t have the deft, cutting prose of Dawkins, and he’s prone to loose use of physics-speak, but his observations are salient. This is a question I have spent a good deal of time pondering: just what parts of modern mathematics are inevitable extensions of arithmetic? Supposing we could run the history of mathematics over again, or that we’re able to see a parallel developed by intelligent life elsewhere. Surely, notations would be different, and in many cases, this would be a good thing. But what of the structure, the foundational concepts, of calculus, functional analysis, abstract algebras?
The obvious natural extension of mathematics, to me, ends at complex analysis. It brings closure to so many basic issues, and an elegance and beauty I am still in awe of. Beyond that? There is a theorem for universal computation… that any system capable of (some minimal set of capabilities) is computationally equivalent. Is there an equivalent for formal systems? Is it the same theorem? It seems like the minimal requirements in this case would be the same as that for Goedel’s theorem…