In error we trust: an apology of abductive inference

The recent surge of pragmatist scholarship in International Relations emphasizes the need for a shift towards a more practical social science. Abductive reasoning is supposed to serve as one of the tools of such reorientation. This article pays closer attention to this form of inference and shows that it represents our deep trust in the role of error within a scientific enquiry. Although abduction stands for a converse error of affirming the consequent, it also enables us to generate new explanations of many complex problems. Instead of attempting to apply abduction to International Relations as a discipline, we simply show that this inferential practice is already being implemented to a large extent thanks to the agent-based modelling that frequently follows pragmatist tenets. In fact this function of suggesting plausible explanations for further examination in the following stages of scientific enquiry may be easily seen as one of the crucial purposes of this kind of modelling. Three examples illustrate this claim.