Modelling the Semantic Field of Ma'naviyat in a Specialised Parallel Corpus: Theoretical and Methodological Foundations of Cross-Genre Discourse Analysis

This article develops a theoretical and methodological model for identifying, grouping, and comparing items within the semantic field of ma'naviyat across genres in a specialised parallel corpus. The semantic field of ma'naviyat is treated here as a network of concepts relating to spiritual and moral life, education, and social values. Rather than automatically classifying every related word as a term, the study aims to distinguish its conceptual content, terminological status, and discursive function. Using the Uzbek–English language pair as a working model, it clarifies the distinctions between parallel and comparable corpora, genre and register, and lexical item and concept. Core, adjacent, and contextual layers of the field are proposed, together with a multilevel annotation system. Criteria for corpus construction address source reliability, translation direction, genre balance, segment alignment, and the recording of uncertainty. The outcome is not a set of statistical findings calculated from an existing corpus, but an operational model that enables subsequent empirical analysis to be replicated. The model allows concepts associated with ma'naviyat to be examined beyond lexical correspondence, taking account of their genre-specific communicative functions.