Magellan altimetry measurements indicate that the mountain belts and plateau margins of Ishtar Terra have topographic slopes in the range 2–30°. Magellan radar images reveal that numerous sets of narrow, closely spaced troughs and lineations, which we interpret as graben and normal faults, are located in many of these regions of high slope. The orientation of many of these graben sets perpendicular to the downslope direction suggests that they formed as a result of gravitational spreading. In several of the mountain belts, the extensional structures are parallel to the apparent direction of shortening, consistent with the interpretation that gravitational spreading occurred while mountain building was still active. To explore the implications of this spreading for the tectonic evolution of Ishtar Terra, we model the process by means of a finite element algorithm for viscoelastic deformation in a vertical section of the crust at the margin of a plateau or broad mountain range. Flow in the crust is assumed to be governed by a nonlinear, depth-dependent rheology. We predict brittle failure and relaxation of topographic relief as functions of time, for ranges of crustal thickness (10–30 km) and thermal gradient (5–25 K/km) and for observed ranges in topographic elevation (1–6 km) and slope (1–30°). This parameter range predicts a large spread in the time scales for initial failure and significant relaxation of relief. For a crustal thickness greater than 10 km, a thermal gradient of 15 K/km or more, and average values of relief (3 km) and slope (3°), the topography relaxes to 25% of its original height in less than 10 m.y. Although initial failure, in the form of normal faulting on the highlands and margin and occasional shallow thrusting in the lowlands, is predicted to occur much earlier, values of horizontal surface strain that are likely to be observable (∼1%) do not accumulate until significant relaxation of relief begins. Given the predicted rate of topographic relaxation, either the crust of Ishtar Terra must be anomalous in some way (either much stronger than predicted by standard flow laws or very thin or cold) or the topographic relief and high slopes have been actively built and maintained until times significantly younger than the average crater retention age for Venus of 500 Ma. The large apparent depth of compensation, the abundance of volcanism, and the uncertainty in the local age favor the view that mountain building in Ishtar Terra has occurred until times at least as recent as 10 Ma.