This paper furthers work investigating the non-standard use of the present perfect (PP) in spoken Australian English (Engel & Ritz 2000; Ritz & Engel forthcoming), and focuses particularly on how discourse features support extensions in the meaning of this tense. As will be shown, analysis of rhetorical relations in discourse reveals that, unlike in other English varieties, expression of temporal progression in Australian English frequently involves PP clauses. More generally, data from this variety show that the PP is completely flexible with respect to temporal order and thus can appear in clauses that are part of a large variety of rhetorical relations. The interest of such a study lies in the fact that, by investigating the contribution made by pragmatic factors to our understanding of temporal relations in discourse, we can gain further understanding of mechanisms that may in the course of time lead to semantic change. Indo-European perfects in the present tense are notoriously instable and tend to evolve into aorists (see eg Bybee et al. 1994). In this respect, the English PP has been somewhat puzzling in that it has retained the functions of a true perfect, unlike its morphosyntactic equivalents in neighbouring languages. This situation, in turn, means that any current change occurring in the usage of the PP can be examined in detail.

* The work presented here was carried out as part of a larger project conducted jointly with Dulcie Engel, funded by the Australian Research Council in 1999 and the University of Western Australia in 2000. I wish to thank Joseph Salmons and an anonymous reviewer for their helpful comments, as well as the participants of ICHL XVII for interesting feedback, in particular Jac Conradie and Bridget Drinka.