William Conner

This paper concerns the problem of determining whom one should trust and how much in complex testimonial exchanges featuring conflicting reports, unclear communication, and higher-order evidence bearing on the reliability of a speaker. Drawing on a Bayesian model of source reliability, I argue for testimonial underdetermination: the claim that testimonial exchanges can underdetermine whom one ought to trust and how much. The argument proceeds by showing that assessments of a source’s reliability are made relative to reference classes – sets of testimonial exchanges sharing features relevant for predicting the source’s reliability. In complex cases, an agent’s evidence may fail to determine which among several plausible reference classes is most appropriate, where different reference classes underwrite different degrees of trust. I then contend that testimonial underdetermination supports a form of synchronic intrapersonal permissivism about trust: an agent’s evidence may permit her to adopt any of several incompatible degrees of trust in a speaker. I conclude that when purely epistemic considerations underdetermine trust, practical considerations may guide one’s choice between different epistemically permissible options, putting pressure on the idea that we can have a pure epistemology of trust that applies to real-life, complex cases involving non-ideal agents.

Beliefs based on pernicious ideology are widespread, and they often have harmful consequences. Attempts to solve the problems these beliefs cause could benefit from epistemological work on them, so it is heartening to see more epistemologists turning to study ideological beliefs. In this paper, I discuss one recent approach, radical epistemology, which has two aims: (1) offering structural explanations of epistemic justification and (2) putting these explanations to work in opposing ideology. While I share radical epistemologists’ opposition to pernicious ideology, I argue that their position is untenable because it gives rise to a vicious circularity. Its core commitment maintains that theorists’ choice between competing epistemological theories should be guided by their moral and political commitments. These commitments themselves, however, are susceptible to reasonable disagreement and thus stand in need of defense. To defend them, radical epistemologists must employ the very theories those commitments are supposed to undergird. I call this _the problem of theory choice_ and conclude that radical epistemologists cannot solve it. But we shouldn’t despair: the time-honored tools of _non_-radical epistemology offer all we need to successfully theorize about and combat bad ideology.