A corpus-based approach to a thesaurus of Tatar idioms

The main aim of the project is an ideographic description of modern Tatar idioms, i.e., organization of idioms according to semantic fields which are defined as a set of lexical units connected with each other by some salient semantic features. The totality of semantic fields along with the conceptual links between them construct the thesaurus of a given language, which can be represented in the form of a semantic network. The need for a new type of dictionary is motivated by the fact that, at present, there is no a comprehensive dictionary of Tatar idioms built on a thesaural principle. The idioms are explored along two revealing lines of inquiry. The first line focuses on a semantic description of idioms and their placement into appropriate taxa according to their meanings. The second line illustrates advantages of organizing a dictionary based on an analysis of corpus data. The idioms were collected from monolingual and bilingual Tatar phraseological dictionaries by continuous sampling. The illustrative materials were sampled from the Written Tatar Corpus containing 620 million words; the Tatar National Corpus (260 million words); and corpus query system Sketch Engine, its Tatar Web Corpus (102 million words). This made it possible to find all occurrences of idioms under analysis and obtain statistically representative data. The findings were processed manually. The Thesaurus is based on an inductive method, that is, from idioms to semantic fields and not from an abstract logical outline to idioms. The aim of the first stage was to describe the semantic features of each idiom and assign a descriptor or several descriptors to the idiom under consideration. At the second stage, the idioms were grouped into semantic fields/taxa in accordance with their meanings. The third stage of classification is connected with checking the correctness of the conceptual marking of idioms, which is of primary importance for regrouping the existing idiom stock. At this point, the frequency of each idiom was checked in the corpus. After idioms were classified on the basis of their dictionary definitions, it was necessary to clarify the information about their meaning and usage. For this purpose, idioms were searched in text corpora. In our case, a corpus-illustrated approach is used as a source of illustrative material. Its purpose is the selection of contexts that could be a source of paradigmatic examples. This approach is relevant in lexicographic work. There is no doubt that a dictionary in which material is drawn from the corpus is more objective, than one with illustrative examples based on linguistic intuition. The results of the research have proved that the use of illustrative materials based on corpus data presents new information on the semantic features of idioms, which contributed to the achievement of the aim. The author declares no conflicts of interests.

Keywords

Authors

References