Adequacy of Machine Translations of Biblical Phraseological Units in English Political Discourse

The article discusses the adequacy of Russian translations of biblical phrases used in English political discourse. As part of the study, a deep post-translation structural-semantic analysis of machine translations of contexts containing biblical idioms in political discourse was carried out; the causes of translation inaccuracies were identified; recommendations on the use of machine translation systems by translators were given. The Bible is considered here as a hypertext of world literature, which has become the source of biblical phrases in English and Russian. The authors analysed the use of biblical idioms in political discourse; discussed and developed a classification of techniques for translating biblical phraseological units; outlined the main areas of practical application of machine translation and identified related difficulties and development prospects. A total of 178 cases of use of biblical phrases in political discourse were selected, which were then translated using a machine translation system followed by post-translation analysis. Both cases of adequate translation and mistakes were subject to a detailed review and interpretation. In the latter case, assumptions were made about the causes of inaccuracies and advice was given to translators on how to work with automatic translation systems. Overall, the number of adequate machine translations of Biblical idioms in political discourse was more than three times higher than the number of inadequate ones. Thus, it can be concluded that machine translation systems are relatively efficient when it comes to phraseological units; however, full automation of the translator’s work is out of the question. In order to produce a high-quality translation, a professional translator needs to be involved, who will edit the machine translations of idioms while giving preference to phraseological equivalents and analogues as well as taking into account the context and cultural differences. The most accurate results are achieved by combining machine processing and expert linguistic analysis.