Corresponding lexical domains: A new resource for onomasiological bilingual dictionaries

The purpose of the current study is to develop a model for analysing relations between lexical items across languages in terms of lexical domains. The model combines a corpus-informed distributional approach with the language-in-use theory of meaning to identify sets of semantically similar linguistic items across languages. It also applies the principle of differentiation to establish differences between individual items. The model overcomes the limitations of previous methods which rely heavily on introspection, and focus on single words. It will be proposed that the results of the study can be used as resources in the compilation of a new type of onomasiological bilingual dictionary. Such a dictionary would provide users with direct access to multi-word units across languages, and help them distinguish and choose between available options.