The fifteen years following World War II saw the development of a wide range of theories of decision making and resource allocation. Notable fields arising from slim foundations included: sequential analysis, linear programming, dynamic programming, game theory, and the rehabilitation of Bayesian statistics. Operations research, which began as a highly empirical activity during World War II, swiftly reoriented itself around the development and application of these sorts of tools. At the same time, these models were embraced by scholars in the social sciences as a powerful means of generating and organizing descriptions of human behavior. This talk contends much of this work was united and motivated by a new attention to problems of "meta-calculation," that is, the formalized consideration of the conditions under which a calculation could be considered valid, or in what ways an incomplete or suboptimal expression would have to be amended to produce a more complete or more nearly optimal expression.