| Summary: | Belief is messy; but if we are careful, it can be tamed. This guiding slogan anchors my thesis, which navigates the complexities of belief regarding three central topics: the conditions under which beliefs should be revised, how our inductively justified beliefs are structured, and what our practice of ascribing beliefs reveals about belief itself. In each case, I argue that the reality underlying belief is far messier than usually acknowledged. Despite its messiness, I demonstrate that, so long as we are careful, theorising about belief is not only possible, but fruitful. Chapter 1 argues that if ordinary beliefs about the future are justified despite having an on-zero chance of being false, then belief revision is far weaker and messier than is usually maintained. I outline novel cases concerning sequences of coin flips that display an odd phenomena: while one is initially justified in believing a coin will not produce a series of n consecutive heads or tails, this belief must be abandoned on observing the first flip of the coin, regardless of whether it lands on heads or on tails. No theory of belief revision can yet capture this phenomena. I outline a new theory that can. Chapter2appliesthistheorytoStalnaker’s notorious ”Composers” case. Suppose you believe on independent bases that Verdi is Italian, Bizet is French, and Satie is French. You then learn that in fact all three composers are compatriots. Should you be cautious, and become ambivalent as to whether they are all Italian or all French, or bold, and conclude they are all French? Implausibly, approaches to belief revision so far side exclusively with either cautious or bold, despite the fact that both reactions look permissible. I demonstrate how the theory in Chapter 1 can maintain that both reactions are indeed permissible, whereby which one is to be preferred depends just on one’s epistemic temperament. Chapter 3 concerns beliefs formed on the basis of induction. I argue that the anti-skeptical claim that such beliefs are justified and can constitute knowledge is only tenable if this knowledge is asymmetric in various surprising and messy ways. For instance, while you can know that all emeralds are green if all the emeralds in your own sample are green, you cannot know this of other possible equally-sized samples, even if they samples that will be examined by your future self, or by other agents who are just like you. Chapter 4 considers belief ascriptions and the increasingly popular thesis that belief is ‘weak’— the claim that one can justifiably believe a proposition even if one has low confidence in it. Though proponents of weak belief usually take various felicitous belief ascriptions as supporting of their view, matters get messy on considering ascriptions of beliefs in conditionals. I show that proponents of weak belief either cannot consistently apply their preferred methodology when accommodating beliefs in conditionals, or they must deny that beliefs in conditionals can be used in reasoning.
|
|---|