Abstract

The problem of shared background, i.e., the question of what assumptions, knowledge, attitudes, etc. participants in interaction have to share in order to mutually understand each other, has been the focus of discussion in philosophy and cognitive psychology for several decades. ;This study approaches this topic from the point of view of ethnomethodological conversation analysis and on the basis of German conversational data. First, shared background is defined as a holistic network of implicit and explicit assumptions that are a necessary prerequisite for interaction. It will be argued that the only way to prove reliably the relevance of a particular assumption as an item of the shared background is by means of negative empirical evidence in cases where interactional problems arise and are treated by the interactants. ;From a linguistic point of view, conversational repair has been described as a way for interactants to deal with interactional trouble. It is suggested in this study that the problem of shared background can be approached in an empirical manner on the basis of analyzing natural conversation with a focus on repair sequences. In particular, a typology of conversational problems that participants in conversation treat by repair will be suggested and it will be shown what interactive and, especially, what linguistic means interactants employ in dealing with conversational trouble and how the relevance of particular items of the shared background thus become reflected by the interactants' activities