The object of the research is the artificial language "nadsat" in Anthony Burgess' novel "A Clockwork Orange". The subject of the study is the specifics of the transmission of "nadsat" in Russian-language professional and neural machine translation. The author considers artificial language in detail as a functionally loaded system that provides a foreign language effect in a literary text, speech labeling of characters, subcultural isolation and a special degree of semantic opacity. A special attention is paid to the difficulties of translating "nadsat" into Russian, due to the fact that a significant part of its units go back to the Russian language base, as a result of which the effect of strangeness, which is fundamentally important for the author's intention, is lost during direct transmission. The article compares the translations of V. Boshnyak and E. Sinelshchikov, as well as the versions made by the neural network systems GigaChat and ChatGPT (via the BotHub platform). The research methodology includes comparative, contextual and linguistic analysis of fragments of the novel and their translations. The comparison is carried out at the sentence level, taking into account the artistic function, consistency and pragmatics of "nadsat". The novelty of the research lies in the fact that the artificial language "nadsat" is considered not only as an object of literary translation, but also as an indicator of the capabilities and limitations of modern neural systems when working with texts of high stylistic complexity. The main conclusions of the study are the following: professional translators strive to reproduce "nadsat" as an integrated artistically motivated system, but they implement this task using various strategies, from more consistent preservation of the effect of foreign language to partial adaptation of the material to the perception of the Russian–speaking reader. A special contribution of the author to the research of the topic is the comparison of professional and neural translation according to a number of uniform criteria. It has been established that neural network systems in some cases are capable of transmitting individual formal features of an artificial language, however, they either normalize it, eliminating artistic specifics, or demonstrate instability of translation solutions. As a result, it is concluded that at the present stage, neural machine translation does not provide a full reproduction of the functional nature of "nadsat" and can only be considered as an auxiliary tool.
Darya Aleksandrovna Sushkova (Wed,) studied this question.