Brandon Holter

The proper role of values in science has been a hotly debated topic for many decades now. The traditional view of epistemic justification holds that scientific objectivity requires suppression of value judgments; theories are only justified by an impartial analysis of the evidence. Traditionalists worry that values bias scientists, leading them to believe what they would like to be true regardless of the evidence. Proponents of value-laden science argue that the justification of scientific theories cannot be isolated from moral and practical judgments. The arguments they advance purport to show that values are not merely biases, values can play a positive role in justifying beliefs. Underdetermination arguments claim that evidence alone cannot determine theory choice.These arguments do not motivate an epistemic role for values in science, however, because values may influence scientists' theoretical beliefs without justifying those beliefs. As a result, philosophers have advanced a number of new arguments for the epistemic necessity of values judgments. This dissertation examines several of these arguments with an eye toward their deficiencies in order to develop an account of values in scientific reasoning that does not legitimate bias or delusion, as the traditionalists fear. Drawing on both old and new arguments, I argue that values help determine when a theory is sufficiently justified by the evidence. While evidence alone provides justifying reasons for belief, scientists must decide how much evidence is needed to justify scientific claims. Different scientific inquiries require different epistemic standards and, I argue, traditional value-free epistemologies cannot account for this variation. Philosophers cannot explain why scientists in different contexts adopt different epistemic standards without appealing to value judgments. On the view I adopt, the evidence justifies a claim when we have good reasons to believe we can act on that claim; a theory is sufficiently justified in the epistemic sense when it is sufficiently justified for practical purposes. Allowing values to guide these decisions, I argue, does not allow values to displace rigorous attention to the evidence.