An extended analysis is given of the program, originally suggested by Deutsch, of solving the probability problem in the Everett interpretation by means of decision theory. Deutsch's own proof is discussed, and alternatives are presented which are based upon different decision theories and upon Gleason's Theorem. It is argued that decision theory gives Everettians most or all of what they need from `probability'. Some consequences of (Everettian) quantum mechanics for decision theory itself are also discussed.
It is often stated that quantum mechanics only makes statistical predictions and that a quantum state is described by the various probability distributions associated with it. Can we describe a quantum state completely in terms of probabilities and then use it to describe quantum dynamics? What is the origin of the probability distribution for a maximally specified quantum state? Is quantum mechanics `local' or is there an essential nonlocality (nonseparability) inherent in quantum mechanics? These questions are discussed in this paper. The decay of an unstable quantum state and the time dependence of a minimum uncertainty states for future times as well as past times are also discussed.