| Description |
The goal of this dissertation is to offer an empirically informed evaluation of testimony as a source of knowledge. Epistemologists have assumed that testimony is a generally reliable source of true beliefs because human cognitive faculties would have evolved to be reliable at getting the truth. However, complementary evidence from the signaling theory and social psychology literature shows that testimony is a practical tool with a variety of nonepistemic functions, including forming and maintaining social relationships, coordinating group behavior, and prescribing conduct. Since the value of using testimony is very often independent of its accuracy, humans have evolved to expend as little resources on checking for accuracy as is necessary to satisfy their other needs. In other words, "truth is expendable" to humans trying to get along well in the world and with each other. This implies that testimony is a far less epistemically reliable source of information than philosophers have assumed, and although it is very often prudent to simply believe what people say, it is not epistemically rational to do so. At the end, I offer preliminary empirically informed prescriptions for judging the reliability of testimony. |
