Computational Stylistics

Herrmann JB, Jacobs AM, Piper A. Computational Stylistics. In: Kuiken D, Jacobs AM, eds. Handbook of Empirical Literary Studies. De Gruyter; 2021: 451-486.Computational Stylistics (CS) is a field of enquiry that examines the forms, social embedding, and the aesthetic potential of literary texts by means of computational and statistical methods. Operating on larger data sets with more transparent methodologies, CS offers literary studies new scales of obser- vation and new methods of interpretation, to test existing theories and form new ones. As in many data-driven fields, methods range across exploratory, ex- planatory, and predictive modeling, with important debates addressing the af- fordances and limitations of each. From its multiple heritages in authorship at- tribution, stylistics, and natural language processing, CS has evolved to tackle ever more ambitious theoretical questions, including style, genre, and epoch; literary topoi, plot, and character networks; narrative perspective, figure char- acterization, and emotion; gender, race, and social status; canonicity, literari- ness, and textual quality; and cognitive representations of word beauty, meta- phor, and rhyme. Situated within the data sciences, CS comprises distinct knowledge domains, in which the affordances of the digital (method, medium) and the statistical interact with the epistemic to produce new knowledge at the analytic levels of “text,”“context,”“author,”and “reader.