It was recalled that, when reasoning in solid-state physics had been purely phenomenological, it had been usual to distinguish only so-called viscous or solid frictions, in which the resistance to motion had been assumed to be proportional to the velocity, or independent of the velocity, respectively. It was stressed here, on the basis of specific examples, that such descriptions were insufficient when applied to the motion of lattice defects at the atomic scale.

J.Friedel: Czechoslovak Journal of Physics, 1995, 45[11], 907-12