Central and eastern United States (CEUS) have experienced large intraplate earthquakes. Yet, at present there is no comprehensive model to explain stresses, strain, and seismicity in this intraplate setting. Models to explain the intraplate stresses in CEUS include glacio‐isostatic adjustment, ridge push effects, local stresses along preexisting fracture zones, and large‐scale convection. In this paper, we present a self‐consistent model of the dynamics of CEUS that explains the stress field responsible for these intraplate earthquakes. The earthquakes represent slow, ongoing deformation associated with forces arising from a combination of lithosphere topography and structure, together with the effects of density‐driven mantle flow. Using GPS data, we calculate strain rates that are likely to arise from tectonic effects and conclude that intraplate strain rates associated with tectonic effects are unlikely to exceed 1 × 10−9 year−1. We test several models of lateral viscosity variations by comparing model stress orientation output with earthquake moment tensors,
SHmax directions from stress inversion, and P axes of earthquakes. A model that satisfies stress and earthquake constraints and also strain rate magnitude constraints requires high viscosity (1025 Pa·s) craton and old oceanic lithosphere of the western Atlantic block and weaker (5 × 1024 Pa·s) accreted Appalachian terrane. Other strength contrasts within the lithosphere are likely present. Incorporation of these into future models using constraints from seismology, along with refined geodetic measurements and improved estimates of crust and upper mantle densities, is needed to further refine long‐term dynamic models and better evaluate the hazards associated with this very slow, ongoing permanent deformation within the eastern and central United States.