What roles should values play in social scientific inquiry given the aims of social science and the nature of its subject matter? That is a big question. This talk will focus on a related but narrower question: is there a role for values in guiding and assessing idealizations in social science and humanities projects? I will argue that the answer is yes.

There has been much recent work in philosophy of science on idealization – the way inaccurate representations can be used to understand a target system. Much of this work has focused on the use of idealization in the context of models used for predicting and explaining the behavior of systems. However, idealization is often employed in the social sciences and humanities for reasons besides facilitating prediction and explanation. I will provide examples of some such cases, which I will call cases of “rational reconstruction.” A rational reconstruction is an idealized representation of the attitudes of some individual or group that aims to clarify those attitudes in order to improve communication, facilitate discussion, and aid understanding the other’s point of view. When idealization is used for this purpose, the goodness of the idealization is partially a function of the coherence and truth of the idealized attitudes. Consequently, when the attitudes being idealized include normative attitudes, part of what makes the idealization good are normative facts.