The current evaluation literature tends to conceptualize evaluator roles as a single, overarching orientation toward an evaluation, an orientation largely driven by evaluation methods, models, or stakeholder orientations. Roles identified range from a social transformer or a neutral social scientist to that of an educator or even a power merchant. We argue that these single, broadly construed role orientations do not reflect the multiple roles evaluators actually assume as they complete the activities encompassing an external evaluation. In contrast to the current literature, this article suggests that typical evaluation activities create functional demands on evaluators, and that evaluators respond to these demands through a limited number of specified evaluator roles. This depiction of a set of specific multiple evaluator roles, generated in response to particular evaluation activities and their associated demands, has implications regarding how evaluation is conceptualized, practiced, and studied. This article concludes with a discussion of these implications.