Since the 1930s, basic statistical problems have been formulated not in terms of interpreting data as evidence but in terms of choosing between alternative courses of action. This monograph is about the problem of representing and interpreting data as evidence.[...]