Stay Close (to Morpheme), otherwise I’ll Miss you Yusuke Yoda Kinki University/Kobe City University of Foreign Studies 1. Introduction Determination of an exponent for each head is contextualized by the local relation between X and Y. This local relation is calculated with cyclic head, or phase under the current Minimalist tenet. For instance, a suppletive form, which is phonologically unrelated realization of the single lexical item, such as “good-bett(-er)” or “go-wen(-t)” in English, is widely observed within the local domain. Then, what is the local domain for the selection of exponents? The answer for this question under the current theory should be a PHASE. Suppose the phase determines the locality for the lexical selection, how is it like? In this paper, I will provide a piece of empirical evidence for the existence of domain extension, arguing SUPPRITIVE form of Japanese negative sentences. Moreover, I also provide evidence that Japanese negative morpheme, which has been regarded as an exponent of Neg 0 , is in fact, the an exponent of categorial determining head, a 0 . Then I conclude that only the PHASE EXTENSION is the way to account for the SUPPRITION at v 0 and provide a crucial case for this extension is prohibited. The organization of this paper is following. In the section 2, I will provide a background of locality within the morphological domain. Then, in the section 3, I will provide descriptive generalization on Japanese negative and decline the existing analysis that Japanese negative morpheme is generated on Neg 0 and provide an alternative analysis that negative morpheme is, in fact, an exponent of a categorial determiner and its is not a polarity decider. Then in the section 4, I will compare two Japanese negative morpheme “na(-i)” and “zu” and provide further evidence that “na(-i)” is a regular adjective. Following, the descriptive generalization illustrated in section 5, I will propose the phase based analysis in the section 6. After that, in section 7, I will provide consequence of the current analysis. Finally in the section 8, I will provide conclusion.