Since Kauzmann's seminal 1959 paper, the hydrophobic interaction has dominated thinking on the forces that control protein folding and stability. Despite its wide importance in chemistry and biology, our understanding of this interaction at the molecular level remains poor, with little experimental evidence to support the idea of water ordering close to a non-polar group that is at the centre of the standard model for the source of the entropic driving force. Developments over recent years in neutron techniques now enable us to see directly how a non-polar group actually affects the molecular structure of the water in its immediate neighbourhood. On the basis of such work on aqueous solutions of small alcohols, the generally accepted standard model is found to be wanting, and alternative sources of the entropic driving force are suggested. Moreover, the fact that we can now follow changes in hydrogen bonding as the alcohol concentration is varied gives us the possibility of explaining the concentration dependence of the enthalpy of mixing. Complementary studies of solute association on the mesoscopic scale show a rich concentration and temperature behaviour, which reflects a complex balance of polar and non-polar interactions. Unravelling the detailed nature of this balance in simple aqueous amphiphiles may lead to a better understanding of the forces that control biomolecular structural stability and interactions.