Will climate change cause conflict? Policymakers seem convinced that, at the very least, climate change will exacerbate groups' tendencies toward violence. Yet, climate change-conflict researchers have failed to uncover consistent linkages between environmental shifts and intrastate contention. This article examines recent research in the field in order to ascertain the reasons for this disconnect. It argues that, although climate change-conflict research has been characterized by significant methodological innovations, theoretical propositions have not kept pace. Rather than developing claims that are specific to climate change-conflict relationships, analysts tend to import arguments from the earlier environmental conflict and civil wars literatures, with little modification. As a result, it is often unclear whether the climate change-conflict relationship is actually being tested. The article offers three recommendations for further specification of theoretical arguments: explicit incorporation of agricultural variables into climate change-conflict models; maintenance of the recent emphasis on conditional effects, with greater attention to local institutions; and a shift in analytic focus from climate change to climate uncertainty.