Truth-Directed Testimonial Discriminations: An Introduction and Application of the Discrimination Condition on Knowledge

Imagine that we are hiking together. Suppose that, while I am interested in birdwatching, I am not very good at it. In particular, I just can’t tell the difference between buzzards and kites. This is unfortunate because, in the area where we are hiking, half of the birds are buzzards, and the other half are kites. Say that a buzzard flies down and lands on the trail in front of us. Let’s add that I am disposed to form a belief about the sort of bird I see whenever I observe one. On this occasion, I happen to form the belief that bird is a buzzard. Do I know that the bird I see is a buzzard? Surely, the answer is no. I don’t know that the bird I see is a buzzard because I can’t discriminate between buzzards and other types of bird. I take it that cases like this tell us something interesting about the nature of knowledge, viz., to get knowledge, we need to satisfy a sort of discrimination condition. I apply this discrimination condition to the testimonial domain.
In this thesis, I argue that, to attain knowledge from the word of others, our testimonial discriminations must be truth-directed. The bulk of the thesis serves as an attempt to explain the truth-directedness of our testimonial discriminations. In doing so, I will examine four explanatory hypotheses generated by major theories in epistemology: i) the non-reductive hypothesis; ii) the reliabilist hypothesis; iii) the assurance hypothesis; iv) the evidential hypothesis. I defend the “externalist evidential hypothesis” that our testimonial discriminations are truth-directed because they demonstrate a sensitivity to evidence.