From Linguistic to Discursive Patterns: Introducing Discoursemes as a Basic Unit of Discourse Analysis

Corpus-assisted discourse studies (CADS) combine corpus linguistics and (critical) discourse analysis to explore how language reflects and shapes discourse. CADS researchers derive macro-level social explanations based on micro-level description of texts. Interpretation (the meso level of discourse analysis) is mostly accomplished by deriving discursive patterns from linguistic patterns observed across many texts. CADS blends close reading (examining individual examples in concordances) and distant reading (looking at results of keyword and collocation analyses). While effective, we argue that CADS lacks a true integration of qualitative and quantitative techniques, which results in a unidirectional workflow where qualitative-hermeneutic interpretation is detached from quantitative analysis. To bridge this gap, we propose operationalising the grouping of linguistic surface realisations in terms of discoursemes – building blocks for discourse analysis. Discursive patterns can then be approximated by co-occurrences of discoursemes. We demonstrate the usefulness of our approach by means of a case study analysing the discourse related to refugees in the German federal parliament during two salient moments in Germany’s history. The case study is carried out using a new open-source software toolkit that facilitates the construction of a consistent database of discoursemes and overcomes some of the technical limitations faced by most CADS studies. Corpus-assisted Discourse Studies (CADS); Critical Discourse Analysis (CDA); German parliamentary debates (GermaParl); Corpus Workbench (CWB); discoursemes