Inferring the frequency and magnitude of past earthquakes from the stratigraphy in exposures of normal-faulted sediments is difficult because colluvial lithofacies assemblages adjacent to faults are complex. Similarities in facies assemblages adjacent to young fault scarps in arid to semiarid areas, such as the Basin and Range province, allow lithofacies to be grouped into two genetic architectural elements: debris and wash elements. Upper and lower facies associations can commonly be recognized within each element. A lithofacies code scheme, similar to those used in the analysis of fluvial and glacial lithofacies sequences, provides a concise way of illustrating lithofacies relations in fault exposures. The source lithology of colluvial lithofacies is shown in the code, and soil-horizon symbols can be included. The architecture of lithofacies assemblages near fault scarps in semiarid areas is explained by a model of colluvial sedimentation in response to a single surface faulting event. Analysis of lithofacies assemblages exposed in three trenches across normal faults in the eastern Basin and Range shows how the model can be used to interpret fault histories. Similar facies analysis methods may be useful in interpreting colluvial sequences formed by non-tectonic processes.