This essay makes two principal points about the role of preferences in explaining international politics. First, for most analytical purposes, preferences must be kept separate from other things–most important, from characteristics of the strategic setting. Otherwise, we are unable to distinguish between the causal role of actors' interests and that of their environment. Second, scholars need to be explicit about how they determine the preferences of relevant social actors. Whether preferences are variables of interest or control variables, it is essential that they be derived clearly and unambiguously.