This corpus-based study examines how English as a Lingua Franca (ELF) speakers utilize metadiscourse (MD) features to structure and regulate the flow of academic spoken discourse across various speech events. It adopts Ädel’s (2022) move-based taxonomy as a framework for a genre-comparative analysis of real-time ELF academic speech events from the ELFA corpus, encompassing various types, including doctoral defense discussions, conference presentations, conference discussions, lecture discussions, and seminars. The study is grounded in a reflexive approach, treating metadiscourse as ‘discourse about ongoing discourse’ (Ädel, 2006). It focuses on discourse organizing MD features such as announcing and concluding topics, endophoric marking, enumerating, previewing, reviewing, and contextualizing. A corpus-based qualitative and quantitative analysis has been adopted, utilizing the UAM Corpus Tool for qualitative coding identifying contextualized MD functions and quantitative analysis comparing their normalized frequencies across seven event types. Our findings show substantial variation across speech events, although the strength of this variation differs by function. Lectures and presentation events show relatively high rates of prospective resources, particularly topic announcement and previewing. Previewing is relatively infrequent overall but strongly differentiated across events; Reviewing is more frequent overall but does not meet the Bonferroni-adjusted significance threshold for cross-event variation. Doctoral Defense Discussions display a hybrid profile; despite their interactive format, they have the highest normalized rate of Enumerating and substantial rates of Endophoric Marking and Contextualizing, while topic announcements and closures are comparatively rare. We argue that these patterns suggest that prospective and retrospective discourse-management resources respond differently to the structural and interactional demands of academic speech events. The results may be helpful to those working to improve the discourse management skills of students in English for Academic Purposes (EAP) settings, as well as academics involved in international conferences and research.
This corpus-based study examines how English as a Lingua Franca (ELF) speakers utilize metadiscourse (MD) features to structure and regulate the flow of academic spoken discourse across various speech events. It adopts Ädel’s (2022) move-based taxonomy as a framework for a genre-comparative analysis of real-time ELF academic speech events from the ELFA corpus, encompassing various types, including doctoral defense discussions, conference presentations, conference discussions, lecture discussions, and seminars. The study is grounded in a reflexive approach, treating metadiscourse as ‘discourse about ongoing discourse’ (Ädel, 2006). It focuses on discourse organizing MD features such as announcing and concluding topics, endophoric marking, enumerating, previewing, reviewing, and contextualizing. A corpus-based qualitative and quantitative analysis has been adopted, utilizing the UAM Corpus Tool for qualitative coding identifying contextualized MD functions and quantitative analysis comparing their normalized frequencies across seven event types. Our findings show substantial variation across speech events, although the strength of this variation differs by function. Lectures and presentation events show relatively high rates of prospective resources, particularly topic announcement and previewing. Previewing is relatively infrequent overall but strongly differentiated across events; Reviewing is more frequent overall but does not meet the Bonferroni-adjusted significance threshold for cross-event variation. Doctoral Defense Discussions display a hybrid profile; despite their interactive format, they have the highest normalized rate of Enumerating and substantial rates of Endophoric Marking and Contextualizing, while topic announcements and closures are comparatively rare. We argue that these patterns suggest that prospective and retrospective discourse-management resources respond differently to the structural and interactional demands of academic speech events. The results may be helpful to those working to improve the discourse management skills of students in English for Academic Purposes (EAP) settings, as well as academics involved in international conferences and research.
This article is derived from the first author’s unpublished thesis under the supervision of the second author, submitted in 2025. The content has been revised and extended for the purposes of this article. We would also like to thank the Proofreading & Editing Office of the Dean for Research at Erciyes University for additional copyediting and the proofreading services that assisted with this manuscript.