RATIONALITY VS. IRRATIONALITY: QUANTIFYING LEXICAL ATMOSPHERES IN DETECTIVE AND GOTHIC FICTION

Abstract

The dichotomy between rationality and irrationality is a defining feature of 19th and early 20th-century British genre fiction. While traditional literary criticism has extensively explored the logical frameworks of Detective fiction and the emotive, supernatural terror of Gothic literature through qualitative close reading, the quantitative mechanisms used to construct these distinct atmospheres remain underexplored. This study employs computational text analysis to measure the lexical fingerprints of these two genres. Utilising a corpus of ten seminal works sourced from Project Gutenberg—including texts by Arthur Conan Doyle, Agatha Christie, Mary Shelley, and Bram Stoker—the study applies the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm to isolate genre-specific vocabulary. Addressing the methodological challenge of Named Entity Bias (where proper nouns disproportionately skew TF-IDF scores) and semantic overlap, the text was iteratively filtered using custom stopword lists. The results quantitatively validate traditional literary theories: the lexical atmosphere of Detective fiction is heavily weighted toward procedural, forensic, and rational nomenclature (e.g., strychnine, colonel, car), whereas Gothic fiction is fundamentally constructed through emotive, primal, and psychological lexicon (e.g., anguish, wolves, journal). Ultimately, this paper demonstrates how distant reading techniques can mathematically substantiate the atmospheric paradigms of classic genre fiction.