Agent Based Models and the Science of Unintended Consequences of Social Action

Abstract

The aim of this article is to defend the thesis that agent-based models (ABMs) permit us to successfully address the problem of the micromacro connection in the social scienes. The argument is developed in four stages: First, we argue that the social sciences requires a theory of action constructed over a concrete explanatory mechanism, referred to as a ?principle of action?. Following, we examine how the micro-macro connection has often been theorised in sociology using what can be referred to as an ?interactionist transition model?. Thirdly, we show that ABMs fit the properties of the interactionist transition model, while contributing specific explanatory mechanisms. Lastly, the argument is illustrated through an analyis of the social reproduction of educational inequalities.