Solving a reasoning problem not only requires representing a set of premises and deriving conclusions from it but also requires interpreting the premises. This latter point has been rarely taken into account by traditional approaches to human reasoning. This article presents how linguistic pragmatics helps to understand the way premises used typically in reasoning tasks are interpreted. It provides several theoretical frameworks and presents a detailed review of studies revealing the influence of pragmatic factors in reasoning. Consequently, the analysis shows that some erroneous answers observed in many tasks are not due to a deficiency in reasoning but result from a discrepancy between the interpretation of the participant and that of the experimenter.