Who’s in, who’s out? Predicting the Inclusiveness or Exclusiveness of Personal Pronouns in Parliamentary Debates

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 5849–5858 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 5849 Who’s in, who’s out? Predicting the Inclusiveness or Exclusiveness of Personal Pronouns in Parliamentary Debates Ines Rehbein, Josef Ruppenhofer Mannheim University, Leibniz-Institut der Deutschen Sprache, Mannheim rehbein@uni-mannheim.de, ruppenhofer@ids-mannheim.de Abstract This paper presents a compositional annotation scheme to capture the clusivity properties of personal pronouns in context, that is their ability to construct and manage in-groups and out-groups by including/excluding the audience and/or non-speech act participants in reference to groups that also include the speaker. We apply and test our schema on pronoun instances in speeches taken from the German parliament. The speeches cover a time period from 2017-2021 and comprise manual annotations for 3,126 sentences. We achieve high inter-annotator agreement for our new schema, with a Cohen’s κ in the range of 89.7-93.2 and a percentage agreement of > 96%. Our exploratory analysis of in/exclusive pronoun use in the parliamentary setting provides some face validity for our new schema. Finally, we present baseline experiments for automatically predicting clusivity in political debates, with promising results for many referential constellations, yielding an overall 84.9% micro F1 for all pronouns. Keywords: clusivity, pronouns, political text analysis 1. Introduction This paper investigates the use of personal pronouns in political discourse. Personal pronouns are a central rhetorical device for building group identities and for othering third parties (Dam, 2015). Tyrkk¨o (2016) con- siders personal pronouns as “one of the primary linguis- tic features used by political speakers to manage their au- diences’ perceptions of in-groups and out-groups”. This makes them an important device for populist rhetoric where the speaker evokes a dichotomous view of soci- ety, us-versus-them (see, e.g., Mudde (2004; Mudde and Kaltwasser (2017)). Studies on the interface between corpus linguistics, prag- matics, discourse studies and political science have pre- sented empirical investigations of the use of personal pronouns in political discourse. One example is Tyrkk¨o (2016) who presents a diachronic investigation of the use of pronouns in political speeches and notes a shift in the early 20th century where many politicians changed their self-centric style and started using more instances of the inclusive plural forms. One of the drawbacks of previous work, however, is that the presented empirical findings are usually based on word frequencies (Vukovi´c, 2012; Tyrkk¨o, 2016; Alavidze, 2017), and only few studies have tried to systematically investigate this topic in more detail, i.e., by trying to measure the agreement between human annotators for disambiguating the referents of personal pronouns in political speeches, or by presenting large- scale studies of the use of personal pronouns beyond word frequencies. More linguistically oriented studies, on the other hand, stress the capacity of 1st person personal pronouns (1PPL) to distinguish between inclusive and exclu- sive uses, depending on whether an addressee (or ad- dressees) are included in or excluded from the set of referents which also contains the speaker (Filimonova, 2005). Cysouw (2002), for example, distinguishes …

Content truncated to 500 words