Fourteenth Bonn Applied English Linguistics Conference (BAELc14)

Dr. Isobelle Clare

Title: Co-occurrence is key: Trolls, Trump, Tyranny, and the Taliban

Abstract: Linguistic co-occurrence – patterns of co-occurring linguistic features - have captured my fascination and led me to develop three linguistic methodologies aimed at uncovering such patterns across large language datasets predominantly consisting of short texts, which are notoriously difficult to analyse with traditional statistical techniques, yet common in security and forensic contexts. In this talk, I will cover two of these techniques that I have developed (short text Multi-Dimensional Analysis and Keyword Co-occurrence Analysis) and their application to various corpora, including Twitter trolling, Trump’s tweets, texts from websites promoting pseudoscience and conspiracy theories, and British news reports of Islam and Muslims. I will show the range of phenomena that linguistic co-occurrence can uncover and importantly how they have revealed interactions between different levels of language. 

Prof. Dr. Valentin Werner

Title: All Fun and Games? Corpus Approaches to Pop Cultural Linguistics

Pop Cultural Linguistics (PCL; see Werner 2022; Werner et al. 2025) is a thriving subfield of linguistics concerned with the study of performed language, that is, the (largely) scripted and fictional language used in various artifacts with a commercial, entertainment-related purpose, including songs, movies, TV series, comics, etc.

This talk has several objectives: First, it sketches the emergence of PCL and discusses why performed language for a considerable time has been ignored in linguistics. Second, it highlights some of the general challenges when working with pop cultural material from a corpus-based perspective, for example in terms of data availability, corpus compilation, and data processing. Third, by way of several case studies, it illustrates how corpus linguistics has informed PCL. Specifically, it demonstrates how common techniques, such as keyword analysis, n-gram analysis, or MDA, have been fruitfully employed in the study of performed language. In passing, reference will be made to both publicly available online corpora (such as the TV Corpus and the Movie Corpus; Davies 2021), as well as to self-compiled corpora, which have mainly been explored with the help of standalone tools (e.g. Werner 2021; Karpenko-Seccombe 2023).

Overall, it is argued that corpus linguistics is an indispensable component of the research toolkit in PCL for at least three reasons: (i) it allows both quantitative and qualitative descriptive insights into synchronic and diachronic pop cultural data, (ii) it has been usefully applied in research from several sub-disciplines, including stylistics and sociolinguistics, and (iii) it has a substantial practical dimension, potentially informing language education/applied linguistics.