earticle

영어

This paper argues, contrary to what is widely observed in the literature, that (i) shika ‘only’ in adjunct position, Indeterminate-mo (dare-mo/nani-mo/dokoni-mo 'anyone/anything/anywhere'), 1-Classifiermo( hitori-mo/hitotsu-mo, etc ‘even a person/ even a thing’)’, kessite ‘never’ in Japanese must be categorized as an Negative Concord Item (NCI) and that (ii) a diagnostic test to distinguish an NCI should be added to 5 diagnostic tests, that is, ‘ability to co-occur with other NCIs in the same negative clause’. In past studies, it has been widely accepted that shika, Indeterminate-mo, 1-Classifier-mo, kessite should be categorized as an NPI based on Klima(1964)’s analysis on English NPI anyone/anything/any-where. Assuming that their analyses were correct, they cannot explain that why dare-mo/nani-mo/dokoni-mo in Japanese syntactically behave quite differently from anyone/anything/ anywhere in English in the following 5 properties : a. Ability to appear in non negative contexts, b. Ability to appear in pre-verbal position, c. Ability to be modified by expressions like almost, d. Ability to be used as an elliptical answer, e. Clause-boundedness. In fact, according to Watanabe(2004), an NCI must have the above 5 properties and this fact shows that Indeterminate-mo, 1-Classifier-mo should be categorized as not an NPI but an NCI. This paper aruges that besides Indeterminate-mo, 1-Classifier-mo, shika in adjunct position and kessite have properties as an NCI. Furthermore, this paper argues that an NCI should co-occur with other NCIs in the same clause.