Corpus-Based Analysis of Russian Translations of Animal Farm by George Orwell

The phenomenon of multiple translations of same classical works has been discussed extensively since 18th century. The dominant approach however is to study retranslating as a cultural, not a linguistic phenomenon. For unknown reasons, little use has been made of corpus data in research on this topic, although corpora of retranslated texts would seem to be a natural source of empirical data. Studying multiple translations with the help of corpus-based methods makes it possible to obtain the general picture of the data and to find its critical points. The quantitative data can be used for developing criteria for evaluation of the texts.
In this paper, six Russian translations of George Orwell's Animal Farm are studied. The translations are compared against an unedited machine translation. A multidimensional scaling of the frequency-list-based distance matrix was performed. The analysis demonstrated that the most frequently republished translations are the most distant from the MT. The keyword analysis of the translations confirmed the findings of the MDS analysis and gave concrete clues on the lexical items typical for certain translations.