This chapter attempts to examine critically the function of theory in historical research and particularly in economic history. In any position other than the extreme of antiquarianism, the tools of economic analysis must have some function in the handling of the historical material that deals with economic processes. These tools—theory, statistical inference, and general laws of nature and causality—have seemed to many historians to be unnecessary or useless, even antithetical to a historical point of view. An economic historian interested in finding explanations for specific events has for his goal the ability, given a set of historical data, to predict the likelihood of the most explicit set of admissible outcomes. The economic historian should not be surprised if the economist shows little interest in the social competence and family relationships of some nineteenth-century merchants— unless it is previously or concomitantly established that social competence and family relationships help to explain the successes or failures of individuals in general.