Anglicisms in Uzbek Online Media Discourse: A Diachronic Corpus-Based Study of Lexical Borrowing, 2010–2024

Abstract

Contact between English and the languages of Central Asia has intensified sharply since the 2000s, yet quantitative descriptions of its lexical consequences for Uzbek remain scarce. This study documents the growth, distribution, and integration of English lexical borrowings (anglicisms) in Uzbek online news discourse through a purpose-built diachronic corpus of 2.4 million tokens sampled biennially from six major Uzbek-language news portals between 2010 and 2024, stratified across four topical domains (technology, economy, culture, sport). Anglicisms were identified through a semi-automatic pipeline combining seed-list matching, orthographic pattern detection, and full manual validation by two annotators (κ = 0.88), and were analysed for normalised frequency, domain distribution, and formal adaptation strategy. The overall anglicism rate more than trebled across the period, from 18.4 to 56.7 items per 10,000 tokens, with statistically significant increases at every biennial step (log-likelihood, p < 0.001) and the steepest domain growth in technology (31.2 to 84.3 per 10,000). Orthographic assimilation to Uzbek Latin script was the dominant integration strategy (62.4% of tokens), followed by unassimilated transfer (24.1%), hybrid formations and loan translations (8.7%), and morphologically integrated forms carrying Uzbek agglutinative suffixation (4.8%). For eight of the fifteen most frequent anglicisms, an established Uzbek or earlier Russian-mediated equivalent lost corpus share to the English item over the period. The findings position Uzbek media discourse in an active phase of contact-induced lexical change in which English has displaced Russian as the primary donor, and demonstrate that borrowing is proceeding with rapid grammatical integration rather than code-mixing at the margins.