Non-ideal Epistemic Contextualism

© 2021 William TuckwellEpistemic contextualists claim that in order for knowledge ascribing sentences, i.e., sentences of the form ‘S knows that p’, to be true S must meet different epistemic standards in different contexts. Some contextualists, those who I’ll label conversational contextualists, claim that speakers can change which standards are contextually operative by making certain conversational moves, e.g., by raising an alternative to the proposition that features in the knowledge ascription to salience. The examples that these theorists work with and the theories that they develop off the back of them tend to build in the assumption that all speakers have equal abilities to pull off standards raising moves. But this is an idealisation that neglects the fact that in our non-ideal world, speakers have unequal abilities to successfully pull off conversational moves, where these unequal abilities vary with unfairly and unjustly distributed power. The project of this thesis is to show that paying greater attention to this fact has three substantial payoffs. Firstly, it allows us to see unrecognised problems with various existing forms of contextualism. In chapter 3 I argue that Keith DeRose’s gap view is vulnerable to an objectionable form of trolling. In chapter 4 I argue that the combination of Michael Blome-Tillmann’s presuppositional epistemic contextualism, unjust power relations, and certain assumptions about the functions of knowledge ascriptions give rise to a host of problematic implications that are grounds for rejecting the view. In chapter 5 I argue that all current versions of conversational contextualism are unable to diagnose an important sub-class of cases of testimonial injustice. Secondly, it helps us to develop an account of an unrecognized form of epistemic injustice that I call irrelevance injustice. Irrelevance injustice occurs either when a speaker raises an alternative that is not taken seriously when it should be, or when a speaker raises an alternative that is taken seriously when it should not be. Thirdly, I argue that the problems that I identify with existing forms of conversational contextualism motivate two new forms of contextualism. Chapter 7 develops a virtue epistemology inspired version of conversational contextualism that excludes the conversational moves of viciously motivated actors from influencing the standards for ‘knows’. Chapter 8 develops a stakes-based view that divorces the determination of the standards for ‘knows’ from conversational dynamics altogether, and instead ties them to what’s at morally at stake in the context of ascription. I stop short of a full endorsement of either of these views, and instead rest content with showing that they provide solutions to the problems I identify for existing forms of contextualism in this thesis whilst retaining some of their central selling points, and that as such both are worthy of further development