A lexicogrammatical perspective in encoding dictionaries ―with reference to ‘pain’ examples in English and in Japanese

This paper addresses a problem in the current encoding bilingual dictionary models, and argues for the necessity of having a lexicogrammatical perspective in such dictionaries, by discussing a pair of equivalents in English and in Japanese. Because existing encoding dictionaries base their models on decoding dictionaries, they force users to start from a fixed lexical item. This paper demonstrates why such a lexically bound encoding process is not preferable, and argues that a more lexicogrammatical approach is necessary in order to enhance the level of naturalness in target languages. First of all, the paper examines the textual grammar of „pain‟ both in English and in Japanese. Secondly, it analyses the predicate information according to types of adjectives in Japanese with reference to corpus data. Thirdly, it establishes a lexicogrammatical unit larger than a single lexical item. These arguments will finally lead to a conclusion that an alternative model is necessary for encoding bilingual dictionaries.