Richard Foley

Beliefs, Degrees of Belief, and the Lockean Thesis

The two accounts would seem to be close cousins. An account of rational degrees of belief simply takes a more fine-grained approach than does an account of rational beliefs. The latter classifies belief-like attitudes into a threefold scheme of believing, disbelieving, and withholding judgment, whereas the former introduces as many distinctions as needed to talk about the levels of confidence one has in various propositions.

Being Knowingly Incoherent

Suppose that your degree of belief in p is twice as great as your degree of belief in its negation, while your degree of belief in q is the same as its negation. Suppose also that p implies q. Since a proposition cannot be more probable than a proposition it implies, your degrees of belief violate one of the rules of the probability calculus; they are probabilistically incoherent. If you post odds that reflect these incoherent degrees of belief-i.e., if you post 2:1 odds on the truth of p and 1:1 odds on the truth of q-you will be vulnerable to a Dutch book, which is to say that your opponent can make a series of bets against you such that you will suffer a net loss, no matter how the events you are betting on turn out. For instance, your opponent can bet

Skepticism and Rationality

1.50 on the truth of q and

The Theory of Epistemic Rationality

1 on the falsity of p. If q is true, you will lose

Working without a Net: A Study of Egocentric Epistemology

1.50 on q. You still could win the

What's Wrong with Reliabilism?

1 bet on p, but even if you do, you will suffer a net loss of

Davidson's theism?

.50. On the other hand, if q is false, you will win

Evidence and reasons for belief

1.50 from your opponents bet on q. But since p implies q, p is also false. So, you will lose

Conceptual Diversity in Epistemology

2 to your opponent on p. Thus, once again you will suffer a net loss. This result can be generalized. It can be proved that if your degrees of belief violate the probability calculus and you are willing to post odds that reflect your degrees of belief, then a skillful enough opponent can make a Dutch book against you. (Ramsey, 1964; Definetti, 1964) This makes it tempting to infer that it is always irrational to have incoherent degrees of belief. And in fact, many philosophers have inferred that this is always irrational. But it is important to note that there is no proof of this latter claim.1 There is no proof even if we assume that it is always irrational for you to make yourself vulnerable to a Dutch book. For even with this assumption, what follows is that it is irrational to accept bets at odds that reflect your degrees of belief when these degrees are incoherent. What degree of confidence it is rational for you to have in a proposition is one matter, what odds if any it is rational for you post on it is another.2

A Trial Separation between the theory of Knowledge and the theory of Justified Belief

Skeptical hypotheses have been allowed to set the terms of the epistemological debate. They convince no one. Yet they have an enormous influence. It is often influence by provocation. They provoke epistemologists into endorsing metaphysical and linguistic positions that antecedently would have seemed to have had little appeal. Skeptical hypotheses, it is said, cannot even be meaningfully asserted, or if they can, the nature of God or the nature of objects or the nature of thought makes it altogether impossible for them to be true. There are those who refuse to be provoked, but even their epistemologies tend to be dominated by skeptical hypotheses. The hypotheses push them into an overly defensive posture from which it can seem that the test of an epistemology is how well it would fare in a hostile environment. There must be a third way. There must be a way to think about skeptical hypotheses that is neither dismissive nor submissive.