In the spirit of model revision, researchers continue to refine the notion of a learning progression. Despite many advances in learning progressions research, one key design element has eluded scholarly critique, the upper anchor. Drawing on science education research and studies of science, this essay argues for a shift from the predominant model of the upper anchor as the fixed, "most sophisticated" way of thinking toward a more expansive "upper reach" that acknowledges plurality and context-dependence in ways of knowing. Three possible models for context-dependent upper reaches are offered.