ABSTRACT
This chapter argues that explanations in archaeology and in our daily encounters with reality take the form of inference to the best explanation, a kind of inductive reasoning. As widespread as reasoning by inference to the best explanation may be, there are significant differences in how this process of explaining observations is pursued in and outside systematic investigations of human behavior. Unlike worldviews, theories require an explicit statement of variables and their relations along with repeated testing of inferences to the best explanation. Social scientists face much greater problems defining their variables and testing causal relations among them than do practitioners of the physical sciences. Archaeological theories, like all conceptual schemes in the social sciences, are difficult to test in ways that yield results accepted by the majority of researchers in the field. Learning guided by theory involves paying close attention to the operation of the relatively few factors and relations to which that theory draws attention.
