Taking the impact of the Fukushima incident on the global discourse about nuclear energy as a case study, the present paper shows how to integrate computational linguistic methods into corpus-based discourse analysis (CDA). After an extensive literature review with regards to the related hermeneutic work, we present the corpus linguistic methods and point out methodological extensions. These extensions include visualization techniques that might help hermeneutic researchers explore large corpora, and second-order collocates, which help triangulate the semantics of lexical items. In our case study, we firstly give an in-depth analysis of the discourses that have formed around salient lexical items, in particular nuclear phase-out and energy transition, in the German Frankfurter Allgemeine Zeitung (FAZ) and the Japanese Yomiuri (both are conservative newspapers of the respective countries). We then provide preliminary results for the impact that the discourse had on German Twitter data. Last but not least, we show what effect the discourse has had on second-order collocates of the lexical item Germany in the FAZ corpus.
@InProceedings{HEINRICH18.3, author = {Philipp Heinrich ,Christoph Adrian ,Olena Kalashnikova ,Fabian Schäfer and Stefan Evert}, title = {A Transnational Analysis of News and Tweets about Nuclear Phase-Out in the Aftermath of the Fukushima Incident}, booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {may}, date = {7-12}, location = {Miyazaki, Japan}, editor = {Jana Diesner and Georg Rehm and Andreas Witt}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, isbn = {979-10-95546-05-4}, language = {english} }
Summary of the paper