Michael Rescorla

When P(E) > 0, conditional probabilities P(H|E)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(H|E)$$\end{document} are given by the ratio formula. An agent engages in ratio conditionalization when she updates her credences using conditional probabilities dictated by the ratio formula. Ratio conditionalization cannot eradicate certainties, including certainties gained through prior exercises of ratio conditionalization. An agent who updates her credences only through ratio conditionalization risks permanent certainty in propositions against which she has overwhelming evidence. To avoid this undesirable consequence, I argue that we should supplement ratio conditionalization with Kolmogorov conditionalization, a strategy for updating credences based on propositions E such that P(E) = 0. Kolmogorov conditionalization can eradicate certainties, including certainties gained through prior exercises of conditionalization. Adducing general theorems and detailed examples, I show that Kolmogorov conditionalization helps us model epistemic defeat across a wide range of circumstances.

A striking thesis lies at the core of Davidson's philosophy: when we attribute intentional content to another creature's mental states and speech acts, we must treat the creature as largely conforming to our own rational norms. I will discuss how this thesis informs Davidson's treatment of rationality and intentionality. After reviewing some historical background, I present basic aspects of Davidson's position. I then examine various worries about the position. I conclude by highlighting some key Davidsonian insights into rationality.

In this dissertation, I explore the relation between language and thought. I criticize theories that explain the representational dimension of language in terms of the representational dimension of thought. I also investigate the prospects for theories that treat neither thought nor language as explanatorily prior to the other. ;The first chapter, "The Diversity of Representational Content," examines one of the most important arguments for treating thought as explanatorily prior to language. Many creatures seemingly lack language while manifesting serious cognitive activity. These examples of non-linguistic cognition apparently favor treating thought as explanatorily prior to language. In response to this argument, I urge that non-linguistic creatures display proto-thought rather than genuine thought. Genuine thought involves propositional content, while proto-thought involves different types of representational content. Once we distinguish proto-thought from genuine thought, we can maintain that genuine thought intimately involves language even though proto-thought does not. ;In the next two chapters, "The Norms of Reasoned Discourse" and "Why Should I Speak the Truth?", I develop a theory of assertion. I argue that assertion falls under irreducibly linguistic norms: the norms of reasoned discourse. When speakers violate these norms, they cease to reason correctly with one another. ;In the final chapter, "The Code Conception of Language," I deploy my theory of assertion against the conception of language propounded by Paul Grice and David Lewis. Grice and Lewis hold that sentences inherit content from the thoughts they communicate. For instance, a simplified form of the view holds that the sentence "Frank is hungry" inherits content from the belief that Frank is hungry because it serves to transmit that belief. This approach, which currently enjoys widespread popularity, treats thought as explanatorily prior to language. I argue that the approach ignores the norms of reasoned discourse and therefore enshrines a false theory of assertion