||Koo, Jong-Nam. (2015). On the the morphology, syntax and semantics amu. The Linguistic Association of Korea Journal 23(3), 85-107. This paper has two goals. The first one is to explain why 'amu'(아무) cannot be realized in usage form in itself, but should they have the form 'amudo'(아무도), 'amuna'(아무나), and also to show that 'amureohada'(아무러하다) can be realized in the same way as 'amu' based on the semantic feature of 'amu'. And the second one is to elucidate the morphological structures, lexicalization, and the grammatical properties(including negative polarity) of 'amu'-based words by way of the semantic features of 'amu'. 'Amu' cannot be referential unlike other indeterminate pronouns because it has semantic feature [-fixed]. In middle Korean it didn't have the feature, but the referentiality has changed in contemporary Korean. Because of that it needs particle -'do(도)', or '-na(나)' to be realized in usage forms. 'Amureohada' also needs the special connective endings '-geodeun'(-거든), '-deunji'(-든지), semantically corresponding to '-do', '-na', to be realized. I discussed that all the forms and semantic features of 'amu'-based words can be explained by the contemporary semantic feature of 'amu' except for the fossilized pronouns.