> The statement "all mathematics is divided into cryptography, hydrodynamics, and celestial mechanics" seems untrue. He's obviously using an over-the-top gener

> The statement "all mathematics is divided into cryptography, hydrodynamics, and celestial mechanics" seems untrue. He's obviously using an over-the-top generalization to be provocative and funny. I guess we are not supposed to understand these words literally. The adscription of topology to mechanics is not entirely casual in his case. He's essentially the father of topological methods in dynamics, and he proved (with Kolmogorov and Moser) the famous "KAM" theorem about the long term stability of the solar system with probability one. Notice that Hardy also uses exaggeration to state some of his finest claims, and he's probably a better writer than Arnold because he manages to do so without the reader noticing.