DIGITAL HUMANITIES AND COMPUTATIONAL STYLISTICS IN CONTEMPORARY ENGLISH LITERARY STUDIES
Keywords:
computational stylistics, digital humanities, distant reading, authorship attribution, Burrows' Delta, scalable reading, natural language processing, cognitive stylometry, corpus linguistics, literary analysisAbstract
This study provides a comprehensive examination of digital humanities and computational stylistics within contemporary English literary studies, tracing the evolution of stylistic analysis from classical rhetoric to twenty-first-century computational methodologies. The paper systematically surveys the theoretical foundations, mathematical frameworks, and methodological innovations that define this rapidly expanding interdisciplinary field, with particular emphasis on the dialectical relationship between traditional hermeneutic practices and quantitative text analysis. Through critical analysis of Burrows' Delta metric and its variants including Quadratic Delta and Cosine Delta the review elucidates how mathematical modeling of high-frequency function words enables robust authorship attribution and stylistic distance measurement across large digital corpora. The paper explores the paradigm of scalable reading as a mixed-methods approach that integrates macro-level distant reading with micro-level close textual analysis, demonstrating how computational tools function as heuristic mechanisms for literary discovery rather than replacements for human interpretation. Additionally, the review addresses corpus-driven narratology, structural parsing, and computational character modeling, highlighting both the achievements and persistent challenges of applying natural language processing to literary texts. Critical controversies, particularly Nan Z. Da's computational case against computational literary studies, are examined alongside defenses of exploratory data analysis as a valid critical practice. The review concludes by investigating frontier developments in generative AI and cognitive stylometry, including LLM-induced stylistic normalization and perplexity-based modeling of reader surprise. By synthesizing theoretical debates, methodological innovations, and emerging research directions, this paper demonstrates how computational approaches are reshaping literary scholarship while underscoring the necessity of maintaining critical rigor in quantitative text analysis.
