Corpus Linguistics And A Mixed Monolingual–Parallel Methodology For Investigating Uzbek – English Action Verbs Synonyms

Corpus linguistics is a branch of linguistics concerned with compiling naturally occurring language data in electronic form and analyzing it systematically. A corpus is understood as a collection of authentic texts that is sufficiently representative to support reliable generalizations about a linguistic unit or variety. The methodological value of corpora lies in their scale and authenticity, which provide extensive empirical evidence that can validate claims that might otherwise rely on subjective intuition. This empirical orientation is especially important for synonym research because near-synonyms often differ in collocation, phraseological behavior, stylistic distribution, pragmatic constraints, and evaluative meaning, distinctions that are not always visible in dictionary definitions. This paper presents a corpus-based research design for analyzing Uzbek and English action-verb synonym sets through a combined methodology: monolingual corpora are used for intralingual semantic profiling, while a custom parallel corpus is compiled for translation-equivalent analysis. The study operationalizes lexical–semantic distinctions via collocation measures (Mutual Information and t-score), multiword pattern extraction aligned with the idiom principle, semantic prosody analysis using a two-stage procedure, and reliability control through Cohen’s kappa. In addition, validity safeguards based on corpus representativeness, normalization, and dispersion are integrated into the analysis. The translation component follows corpus-based translation studies and descriptive translation studies frameworks to quantify translation equivalents, identify asymmetries, and interpret translation strategy tendencies.

Перевод пока недоступен