The combination of Discourse Markers in Persian

This paper sets out to investigate the ways in which some of the Discourse Markers (DM s) in Persian are used by looking at a corpus of 475 million words. By adopting Fraser’s notion of DM (2009), it will analyse all possible combinations of thirty DMs categorized into three groups: contrastive, elaborative, and inferential. This categorization will be based on the types of semantic relationship they signal between the propositions of the discourse segments preceding and following them. Their deployment in the attested data demonstrates that the ordering of these DM s is by no means arbitrary. The result of our investigation also reveals that Persian contrastive DM s show a strong tendency to combine with the members of their own category, while elaborative DM s tend to combine with inferential DM s. Inferential DM s, too, have a tendency for intra-category combinations; however, such combinations are much less frequent than those of contrastive DM s. Contrastive DM s have the lowest predisposition for combining with inferential DM s. In short, by exploring the frequency of all sequences of DM s under investigation, a hierarchy of DM combinations in Persian will be proposed, which can be argued to predict certain possible configurations of DM sequences in Persian, and such empirical findings will build up some basis for future typological research as well as for the theorization of DM sequences in general.