On the status of discourse markers

Research on discourse markers has run into fundamental difficulties recently. First, that group of linguistic items appears to withstand approaches trying to account for them in terms of formal properties, even though the set of discourse markers includes a subset, that of discourse particles, that is somewhat easier to characterise in formal terms. Secondly, a function-based description of discourse markers has its own problems, too: it is not easy to find the common denominator of all the diverse roles that the relevant literature attributes to them. A substantial part of the latter difficulty has its roots in the fact that, to date, no adequate and full-fledged classification has been proposed with respect to linguistic units having a pragmatic role, including discourse markers.No increase in the amount of empirical research does, in itself, present a way out of that situation: successful continuation of such research and general theoretical advancement both require that certain fundamental issues be clarified first. The purpose of this paper, accordingly, is to reveal certain difficulties in referring to and characterizing discourse markers (via a review of the relevant literature), claiming that some of their allegedly canonical properties are oversimplified or downright wrong. Also, an effort is made to highlight crucial functional aspects of the delimitation of the group of discourse markers and to discuss the status of multi-word discourse markers, too