Methodology for creating a parallel corpus of Oybek’s novel “Navoiy” and annotating literary devices used to depict human appearance

Abstract

The integration of corpus linguistics and translation studies has expanded the possibilities for the empirical investigation of literary texts. This article describes the principles of constructing a parallel corpus based on Oybek's novel Navoi and its English translation, as well as a multi-layer annotation model for literary devices representing human appearance. The corpus was compiled according to the principles of representativeness, completeness, consistency, and reproducibility. Each portrait unit was treated as an independent observation unit and assigned a unique identification code that records its semantic category, type of stylistic device, linguistic realization, translation equivalent, and translation transformation. Statistical analysis was conducted using the token principle, while the source and target texts were aligned through the sentence alignment method based on semantic equivalence. The proposed methodological model provides a systematic framework for investigating the lexical-semantic and translational characteristics of portrait units in a parallel corpus and establishes a methodological basis for corpus-based and translation-oriented studies of Uzbek literary texts.