Abstract
The article focuses on synesthetic word combinations, defined by the author as a phrase, in which both words have a sensory semantic component and refer to different perceptual modalities (for example, “green echo&8j1; refers to vision and hearing). A parallel multilingual corpus has been designed to become the material for the study. The corpus includes the novel “Lolita&8j1; by V. Nabokov in English – the original language of the novel – (1955), and its translations into Russian (1967) and French (1959, 2001). The study aims to classify sensory lexical units, which form synesthetic word combinations, and to quantify the size of each class presented in the corpus. The study reveals that the frequency, with which semantic components refer to sensory modalities, tend to differ. Another conclusion to be drawn is that a stable collocability between semantic components referring to certain sensory modalities can be observed. Thus, the analysis of synesthetic word combinations shows that the following senses appear to be most frequently manifested in sensory lexical units: vision (35%), touch (29%), hearing (22%). The words with the components of taste (8%) and smell (6%) are the least frequent. As for collocability, in the original text the vision-hearing combination is revealed to be most frequent, while in the translated version most sensory collocations tend to include vision-touch components. Collocations whose components refer to vision-olfaction and olfaction-hearing appeared to be least numerous.