In this paper, I examine some sentence-ending markers (= Xs) accompanied with the morpheme -(u)l- ('will') that is ambiguous between non-epistemic will (will 1 ) and epistemic will (will 2 ): -(u)l-lay, -(u)l-kka, -(u)l-kel, etc. Based on empirical data, I try to show that they have the following lexical properties/informations: (a) the person feature information of the subject they allow (b) the speech act information they trigger, (c) their selectional property in relation to whether they select will 1 or will 2, and (d) their lexical property in relation to whether they trigger the opposite polarity readings, which I call nPOS/nNEG (= nominal positive/negative) readings following Choe (2015). During the discussion, I suggest the following two descriptive generalizations: (I) will 2 , but not will 1 , can come with the tense morpheme -(e)ss-, and (II) Xs that come with will 2 allow any person (1 st , 2 nd , or 3 rd person) subject while Xs that come with will may not (probably for lexical/semantic reasons). I also provide evidence in favor of the following generalizations suggested in Choe (2015): (A) Declarative Xs as well as interrogative Xs can trigger nPOS/nNEG readings, and (B) an X with the feature [Speaker's Denial] triggers nPOS/nNEG readings when it is in the same clause as POS/NEG.
1. 서론
2. “-{ㄹ/을}-”을 수반하는 (복합)종결어미들
3. 요약 및 추가논의
(0)
(0)