LIMITATIONS OF AUTOMATIC TRANSLATION IN RENDERING UZBEK CONGRATULATORY EXPRESSIONS INTO ENGLISH

Abstract

This article examines the limitations of automatic translation in rendering Uzbek congratulatory expressions into English. Although modern machine translation systems can quickly process simple informational sentences, they often face serious difficulties when translating speech acts that contain cultural values, emotional tone, politeness strategies and formulaic blessings. Uzbek congratulatory discourse frequently includes concepts such as rizq, baraka, xonadon, niyat, oq yo‘l, hurmat and farzand kamoli, which cannot always be transferred through direct lexical equivalence. The article analyzes typical machine translation errors, including literal rendering, loss of pragmatic force, inappropriate register, incorrect interpretation of religious and cultural units, and unnatural word order. It also explains the role of human post-editing in improving the naturalness, cultural accuracy and communicative effect of translated congratulatory texts. The study concludes that automatic translation may be useful as an initial draft, but culturally sensitive congratulatory expressions require human interpretation, functional equivalence and pragmatic adaptation.