Why is the Galactic disk so cool?

The bulk of old stars in the Galactic disk have migrated radially by up to several kpc in their lifetimes, yet the disk has remained relatively cool, i.e., the ratio of radial heating to migration has been small. Here, we demonstrate that this small ratio places very strong constraints on which mechanisms could have been responsible for orbital transport in our Galaxy. For instance, Sellwood & Binney's mechanism of nonlinear horseshoe transport by spirals tends to produce too high a ratio of heating to migration, unless the spirals' amplitudes are heavily suppressed away from their corotation resonances, or their pitch angles are significantly larger than is observed. This problem is only made worse if one includes the effect of the Galactic bar, diffusion due to disk or halo substructure, etc. Resonant (but non-horseshoe) scattering by spirals can drive transport consistent with the data, but even this requires some fine-tuning. In short, reproducing both the observed radial migration and the small ratio of heating to migration is a highly nontrivial requirement, and poses a significant challenge to models of the Milky Way's dynamical history, theories of spiral structure, and the identification of 'Milky Way analogues' in cosmological simulations.