Annotating Argumentation in the Wild

The aim of this project is to build and analyze an annotated corpus of naturally occurring argumentative discourse in online communities. Drawing on advances in computational sociolinguistics, we explore the affordances of using online communities such as Reddit as a data source for studying argumentation in the wild. Using  argumentative discourse data extracted from Reddit forum discussions, we (a) annotate the data for statement types and argument types; and (b) investigate the extent to which these statement types and argument types correlate with specific linguistic markers in the construction of argumentation. In this way, we aim to shed light on the extent to which argumentation in online settings is similar to or different from arguments in other types of discourse. Finally, we extend the scope of the project towards applying the analysis of argument types on Reddit to the study of language attitudes and ideologies and the values different types of language are associated with.

Activities and publications

Workshop Talk – V. Kostadinova & J.H.M. Wagemans (2022). Statement type annotation. Workshop “Empirical studies of discourse and argumentation in public policy: Multidisciplinary and Multilingual Perspectives”. Università Cattolica del Sacro Cuore, Brescia. September 8, 2022.

Journal Article – Visser, J., Lawrence, J., Reed, C.A., Wagemans, J.H.M., & Walton, D.N. (2021). Annotating argument schemes. Published online May 7, 2020. Argumentation, 35, 101–139.