Assistant Professor of TEFL,
Department of Basic Sciences,
Faculty of Paramedicine,
Tabriz University of Medical Sciences, Tabriz, Iran
As a part of pragmatics of language, Metadiscourse (MD) has been widely recognized as playing a pivotal role in the expression and comprehension of messages in academic writing. In view of its significance in effective communication, there have been a lot of attempts to categorize different MD devices within manageable models alongside some descriptive works to demonstrate the use of MD devices on the basis of these models. However, despite all the centrality accorded to MD markers in academic writing, few ESP practitioners have tried to explore the use of these devices by non-native speakers (NNS) who are known to be producing dry, incoherent and sometimes confusing papers - the very shortcomings which might easily be managed by judicious use of MD elements. With the absence of systematic works on the use of MD markers by NNS, there is a paucity of information on their use of MD devices in their productions. This is what the present paper seeks to uncover. Drawing on the Hyland and Tse’s (2004) “Interactive and Interactional” model of MD, and their description of the use of MD devices in research articles produced by native speakers (NS) in Hyland (1998), it was set out to document the frequency of MD markers in papers produced by NNS of English and contrast it with that observed in Hyland (1998). For this purpose, 20 RAs written by NNS which had been accepted for publication in the Tabriz Journal of Dentistry were selected after extensive stylistic and linguistic editing, and were compared against Hyland (1998) with respect to the frequency of particular MD elements. The results showed huge discrepancy in the use of all MD devices in general, and some in particular. The findings can provide useful insights in materials development for academic writing classes where learners could receive explicit instructions on the use of MD elements which have been found to be used least frequently.