Between Norm and Narration: A Corpus-Based Comparative Analysis of German and Uzbek Legal Press Language

Abstract

Despite the expanding body of cross-linguistic corpus studies in legal discourse, the language of legal press, a hybrid genre mediating between juridical normativity and journalistic narration, remains comparatively understudied, particularly in language pairs that span typologically distant families. No prior corpus-based study has systematically compared German and Uzbek legal press language, a gap that is notable given their divergent morphological profiles (inflectional vs. agglutinative), distinct legal traditions (Romano-Germanic civil law vs. Uzbek legal reform), and markedly different levels of representation in international corpus linguistic research. The present study addresses this gap through a purpose-built bilingual corpus of 120,000 tokens (60,000 per language), drawn from established German legal press outlets (Legal Tribune Online, beck-aktuell) and Uzbek legal periodicals (Huquq va Hayot, Yurist), covering the period 2019–2023. Using AntConc 4.2.0 and Sketch Engine, the analysis examines lexical density, standardized type-token ratio (STTR), nominalization frequency, mean sentence length, keyword profiles, and hedging devices. The findings reveal that German legal press exhibits significantly higher nominalization density (M = 18.7 per 100 tokens) and longer mean sentence length (M = 24.3 words) than its Uzbek counterpart (M = 11.2 and M = 16.8, respectively), while Uzbek legal press displays greater lexical variation and a higher proportion of epistemic hedges. These patterns are interpreted through register theory (Halliday & Matthiessen, 2004) and genre theory (Swales, 1990; Bhatia, 1993). The study contributes the first quantitative baseline for Uzbek legal press language and offers implications for legal translation pedagogy, terminology standardization, and corpus-assisted language teaching in Central Asian legal contexts.