This paper is a review of the theoretical and observational evidence bearing on the magnitude of the shear stress which acts on the base of a lithospheric plate. Estimates based on the viscosity of the upper mantle do not yield useful limits. Arguments based on the thermal stability of the upper mantle indicate that the basal shear stress is no larger than a few bars. An indirect measurement of the rheology and shear stress can be made by studying the diffusion of stress and displacement following a large decoupling earthquake. When applied to the 1965 Rat Island Earthquake, this method yields a basal shear stress of about 2 bars. These results indicate that for small plates the forces produced by basal shear stress are probably small in comparison with forces acting on plate boundaries. To a first approximation, the smaller plates act as if they were decoupled from the mantle below. These stress estimates lead to a model in which the motion of the smaller lithospheric plates is governed almost entirely by the forces acting on their edges. Forces due to basal shear stress may be comparable to forces acting on the edges of large lithospheric plates. Thus, complete decoupling may not be a good approximation for such plates.