Against representative agent methodology

Abstract

Garnett and Boettke both seek to identify the appropriate behavior for the representative scientist. The social structure of science is better represented, however, with a heterogeneous agent model. Social epistemology and epistemological naturalism provide context for the argument against representative agent methodology. Asking whether individual scientists should “commit themselves to an approach and pursue it doggedly” or make “a professional commitment to intellectual tolerance, openness, and broad-mindedness” is like asking whether it is better to be a bouncer or a bookkeeper. The question depends on particulars that vary from person to person. Down with representative agent methodology. Up with diversity.