Russian Loanwords in Standard Chinese: Problems of Phonetic and Morpological Adaptation, and Automatic Recognition in Chinese Texts

This work is aimed at providing an integral analysis of the Russian phonetic loans in Mandarin. The present work consists of four parts. The first one is a corpus analysis of the transformation of segmental and super-segmental elements borrowed from Russian into Mandarin; it is conducted within the Optimality Theory paradigm. The second part is a psycholinguistic experiment aimed at evaluating the role of the graphic factor in phonetic transformations of the Russian loanwords. The third part addresses the changes that the Russian morphemes undergo in the process of borrowing. The fourth part involves comparison of state-of-the-art Chinese word segmentation algorithms for the task of phonetic loanword detection The study is conducted on the wide range of materials. We used the Chinese loanword dictionary, experimental data, reference guides on the transliteration of the Russian words, subset of the Chinese Wikipedia articles, the Big Russian-Chinese dictionary (BKRS), subset of the Russian-Chinese parallel corpus of Ruscorpora. As a result of our work, we have got the consistent description of the phonetic transformations, that is comparable to studies on contact phonology of other language pairs. Although the majority of generalizations that are proposed to the English loanword adaptation is relevant for the Russian borrowings, we discovered some peculiarities and proved their regularity in our experiment. The study of the morphemic adaptation has identified a number of the trends that Russian morphemes and allomorphs undergo during their adaptation. The results of this descriptive analysis can be used for the more general models of the loanword adaptation processes, that will include the morphemic transformations. During the comparative analysis of the CWS algorithms, we found the most efficient segmentator for the task of loanword detection. It will be integrated into the document processing pipeline of the Russian-Chinese parallel corpus of Ruscorpora.