geoff pynn

home | classes | research | cv

Assistant Professor
Department of Philosophy
Northern Illinois University
g pynn @ niu . edu

The Bayesian Explanation of Transmission Failure (forthcoming in Synthese)
Transmission failure occurs when P entails Q, but a subject cannot acquire a justified belief that Q by deducing it from P. Paradigm cases of transmission failure involve inferences from mundane beliefs (e.g., that the wall in front of you is red) to the denials of skeptical hypotheses relative to those beliefs (e.g., that the wall in front of you is not white and lit by red lights). According to the Bayesian explanation, transmission failure occurs when (i) the subject's belief that P is based on E, and (ii) Pr(Q|E) < Pr(Q). But there are compelling cases of transmission failure where Pr(Q|E) > Pr(Q). No modifications to the Bayesian explaantion seem capable of accomodating such cases, so the explanation is inadequate. Alternative explanations employing subjunctive conditionals can capture all of the paradigm cases, as well as those missed by the Bayesian.

Defending The Hook Analysis of Indicative Conditionals
Grice and Jackson defended the view that `If A, then C' is true if and only if either A is false or C is true. The hook analysis, as I call it, is widely regarded as a failure. Adam Rieger has recently offered a set of assertability conditions for the indicative that suggest that there is hope for the hook analysis. The conditions can be derived from the hook analysis together with the knowledge rule of assertion; no ``conventional implicatures'' or special pragmatic meanings of the indicative ``if'' are required. Here I show how Rieger's conditions can be used to address some of the thorniest problem cases for the hook analysis.

The Problem of False Denials: Invariantism and Error
Moderate invariantists must explain why people can, with apparent propriety, falsely deny that subjects know things in contexts where they are taking skeptical hypotheses seriously. Moderate invariantists often try to solve this problem by appealing to pragmatics. I argue that the pragmatic approach is doomed; invariantists should instead recognize that speakers in hard contexts are wrong about what subjects know.

A Priori Anti-Skeptical Justification
Many philosophers balk at the idea that I have a priori justification to believe that various radical skeptical possibilities do not obtain. Here I attempt to say why we should embrace this idea.

Don't Be Insensitive!
I argue that you shouldn't make an assertion that you would make even if it were false; i.e., you shouldn't assert insensitively. I defend the rule using cases, arguments against other explanations of conversational data, and by reference to various considerations drawn from the epistemology of testimony and Stalnakerian theories of conversational dynamics. I then show how to use the rule to defend a moderate stable invariantist theory of knowledge.