Stylistic Deviations in Linguistics Introductions: A Move-Step Analysis of Wordiness, Redundancy, and Communicative Impact

Introduction: Stylistic deviations such as wordiness and redundancy undermine clarity and precision in academic writing. Their frequency and communicative impact, however, are likely to vary across disciplinary traditions. Earlier research has examined these phenomena in education-related corpora and revealed patterned distributions of redundancy in justificatory passages. By contrast, the ways in which such deviations manifest in linguistics research articles remain underexplored. Purpose: To investigate the distribution, functional localization, and communicative impact of stylistic deviations in linguistics Introductions, with a particular focus on how wordiness and redundancy interact with rhetorical structure. Method: A corpus of 40 linguistics Introductions (388 sentence-level fragments) was compiled and annotated. Each fragment was coded by rhetorical Move and Step according to the CARS model and further categorized by deviation class (wordiness or redundancy), subcategory, and communicative impact. The taxonomy of deviations previously validated in education corpora was applied in full. Statistical analysis included descriptive profiling of classes and subcategories, chi-square tests of association, and effect size estimation. Results: Wordiness predominated across the corpus, accounting for 70.6% of all deviations, while redundancy accounted for 29.4%. Class balance was stable across Moves and Steps, but severity was functionally localized: high-impact deviations clustered in M3_S3, M3_S1, M2_S2, and M1_S3. Almost all high-impact cases were linked to wordiness, with syntactic complexity alone responsible for 54 of 61 instances. Redundancy, although frequent in structural and lexical repetition, rarely reached high severity. Conclusion: These findings show that in linguistics Introductions in English the primary stylistic risk lies not in repetition but in syntactic overload at rhetorically dense points of the text. The results extend previous applications of the taxonomy by demonstrating a discipline-specific pattern of risk concentration. The study highlights the value of combining rhetorical segmentation with fine-grained stylistic annotation and suggests that pedagogical efforts should focus on reducing syntactic complexity in high-pressure rhetorical contexts. Limitations include the modest corpus size and the absence of cross-disciplinary comparison, which future research should address to refine understanding of stylistic risk across fields. © 2025 National Research University Higher School of Economics (HSE University). All rights reserved.