Irrational rotations motivate measurable sets (Q5957064)

The constant tension between the wish to present mathematics in a prolix way that reflects the process of discovery, and the need to present mathematics in elegant, terse definition-theorem-proof style is explored here in a specific setting. Carathéodory's definition of measurable sets via outer measures is presented as an instance of a definition whose underlying `intuition' or rationale remains mysterious. This much-discussed observation shows the difficulties very clearly: mathematical notions of such lasting impact must sometimes be the product of huge amounts of (now hidden) trial, error, and effort. They may of course also involve some intuition on the part of an exceptionally far-seeing mind. Nonetheless, it is hard to argue with the idea that it is a long path from Carathéodory's definition to finding that the collection of measurable sets give a \(\sigma\)-additive measure on a \(\sigma\)-algebra. A pedagogical consequence is that the notion of ergodicity, which formally requires measure theory to state, can only be presented after much preliminary work. Here an interesting reversal is described. Using the usual outer measure on the circle using sequences of arcs, the outer measure of any set invariant under an irrational circle rotation is found. From this discussion, the notion of measurability emerges quite naturally.