A corpus-assisted critical discourse analysis of hate speech in German and Polish social media posts

Abstract

The present paper investigates linguistic data that include hate speech motivated by the alleged or real ethnic or national identity of its addressee in German and Polish Internet-based communication. Focusing on hate speech on Facebook, this paper contributes to the studies of hate speech (e.g. Meibauer, 2013; Meibauer, 2021) in online communication (e.g. Baider & Constantinou, 2020; He et al., 2021; Founta & Specia, 2021) and, in a more general sense, to the vast area of research on the use of aggressive language. Based on a representative corpus, the predominant categories of user-generated hateful statements are teased out relative to their form and stance by means of critical discourse analysis and corpus linguistics tools. In the analytical part, we discuss the discursive practices used in the posts, together with the meanings they communicate, followed by a comparative analysis of their frequencies in comparable corpora. Our findings confirm that hate speech is linguistically conditioned by its socio-cultural context. For instance, our contrastive analysis of German and Polish online data indicates that the two nationalities use different discursive practices to express their aversion to the Other, and that Polish comments are more overtly insulting than German comments.