This dissertation addresses the following questions: How should epistemologists conceptualize testimony? What do people use testimony to do? And why does ‘what people do’ with testimony matter epistemically? In response to these questions I both define and characterize testimony. While doing so I argue for the following answers, given here very briefly: What do people do when they testify? They tell each other things and avow that those things are true, offering their statements to others as reasons to believe. More (...) importantly, they interact with each other in order to negotiate about significance. Why do these activities matter epistemically? Because by engaging in them people generate understanding, as well as knowledge, that no one involved may have had prior to negotiation. Not only that, but they generate collective hermeneutic resources—conceptual tools with which to interpret and understand. In so doing, they not only learn, they create significance and (re)construct social worlds, living together as epistemic, moral, and political agents. So, how should epistemologists conceptualize testimony? They should treat it as a particular speech act that most often occurs as part of a testimonial exchange—an interactive, interpretive, dialogical activity that people use in order to negotiate about significance and generate understanding. This characterization of testimony is an important contribution because it: reveals some of the distinctively social aspects of testimonial knowledge and understanding; suggests better answers to epistemic questions about testimony than those based on typical characterizations in the literature; leads discussion on the topic in a number of new directions; and it lays a foundation for an ethics of testimony that cannot be separated from epistemic concerns. Neither can those concerns be separated from social, moral, and political considerations. The position thus pushes epistemologists to investigate the intertwining of epistemic, moral, and political agency. (shrink)