Aspirational Mathematics and the Behavioral Sciences: The Columbia Behavioral Models Project, 1952–1957

The Behavioral Models project (1952–1957) at Columbia University had a clear mission: to make mathematics legible to social scientists. The project, funded by the Office of Naval Research (ONR), enlisted mathematicians and philosophers to demonstrate how mathematical models could be applied to the study of human behavior. Measured against this aim, the project was a failure. The flavors of mathematical modeling promoted by the project were rarely adopted by mainstream social scientists, the target audience. If anything, the disciplines embraced (or re-embraced) the correlational tradition of inferential statistics—an approach that the project had defined itself against. The paper argues that the Behavioral Models Project (BMP) is an expression of aspirational mathematics. The postwar behavioral sciences movement imagined itself, in a gauzy, rhetorical sense, in the image of the natural sciences. But its more proximate reference discipline was economics, which was already self-segregating by the early 1950s. Economists’ mix of mathematical rigor and policy influence, on early and notably partial display in these years, made their discipline an object of envy and qualified emulation. The BMP—framed as a maths booster shot for the behavioral sciences—was a high-profile index of that dynamic. Mathematical sophistication, if not necessarily the economists’ variety, was something to cultivate. As the BMP case helps to show, the aspiration went unfulfilled.