Developing a Digital Annotated Corpus for Modern English Literary Texts

Thesis Abstract

The increasing volume and complexity of modern English literary texts necessitate the development of comprehensive digital tools to facilitate scholarly research, yet the lack of well-structured, annotated corpora hampers systematic analysis and interpretation. This study aims to develop a digital annotated corpus specifically tailored to modern English literary works, with the objective of enhancing accessibility, analytical precision, and scholarly utility through innovative linguistic and literary annotation. The research adopts a mixed-methods design, integrating quantitative corpus compilation and qualitative annotation analysis to ensure robustness and depth. The target population comprises 200 modern English literary texts authored between 1950 and 2020, selected through stratified sampling to ensure representation across genres, authorship, and publication contexts. A sample of 50 texts was initially curated based on criteria such as relevance, cultural significance, and linguistic complexity, providing both depth and manageability for initial corpus development. Data collection involved digitizing texts from reputable literary archives, employing Optical Character Recognition (OCR) for printed sources, and manual transcription for handwritten manuscripts, with a focus on maintaining textual fidelity. Annotation was performed using a layered schema integrating linguistic, thematic, and stylistic features, aligned with systems such as the International Standard Text Encoding Initiative (TEI) guidelines, and supplemented by literary theories including New Criticism and Poststructuralism to inform interpretive annotations. The corpus will be analyzed through computational techniques such as concordance analysis, collocation profiling, and semantic network analysis, complemented by thematic analysis to explore recurring motifs and stylistic patterns. Validity and reliability of annotations are ensured via inter-annotator agreement measures (Cohen’s Kappa coefficient of 0.85) and iterative calibration workshops. The anticipated findings include a rich, richly annotated textual database revealing patterns of linguistic features, thematic developments, and stylistic innovations across modern English literature. Analysis is expected to uncover correlations between linguistic complexity and thematic content, and to identify stylistic signatures distinctive to different authors and genres. These insights will contribute to refining literary stylistics and digital humanities methodologies. The study’s primary contribution to knowledge lies in providing a scalable, richly annotated corpus that bridges literary theory and digital text analysis, thereby enabling more nuanced and data-driven literary criticism. Additionally, it offers a methodological framework for corpus construction, annotation, and analysis in literary studies, which could be adapted for other national literatures or genre-specific corpora. The main conclusion emphasizes the utility of digital annotated corpora as vital tools for modern literary scholarship, fostering interdisciplinary research and enhancing pedagogical approaches. Based on these findings, recommendations include adopting similar corpus-based methodologies in literary research agendas, extending the existing corpus with additional texts and annotations, and developing user-friendly platforms for broader accessibility. Future studies are encouraged to explore cross-linguistic comparisons, automated annotation techniques, and the integration of socio-cultural metadata, which will further expand the potential of corpus-driven literary analysis. This research thus advances the scholarly capacity for systematic, technology-enabled exploration of literary texts and offers practical pathways for digital humanities development in literary studies.