- 學位論文
論漢語「是」分裂句：雙句分析
On Bare Shi Clefts in Mandarin Chinese: A Bi-clausal Analysis
摘要
本研究旨在透過雙句分析，深入探討漢語光桿「是」分裂句型。早先研究泰半使用單句分析解釋漢語的分裂句現象，但仍舊有部分問題亟待解決。本研究的雙句分析模式，將「是」字焦點詞分析為繫動詞(copular verb)。此動詞「是」為主句動詞，並且選擇一個完整的包孕句(CP clausal complement)作為其補語。句法成分(consitutent) 必須要透過顯型移位，移入包孕句補詞層的指示詞位置(SpecCP)得到焦點，並且在語音部門上與「是」相近。如此一來，對於不同的焦點結構(例：主詞焦點、附屬成分焦點、動詞組焦點及動詞焦點)都能夠提供一致的分析。 而另一方面，對於漢語中受詞焦點不合語法的現象，雙句分析也能提供一套新的解釋。如果潛在句法結構是兩個子句的情況下，在受詞顯形移位的時候，由於主要子句的「是」已經存在於主要動詞的位置，並沒有多餘的句法位置讓包孕句的動詞能移入。又，就算能夠移入，此時出現在主要動詞「是」之前的句法成分並不完整，如此便能解釋受詞焦點不合語法的現象。此外，透過雙句分析，漢語中的受詞仍舊能藉由受詞移位得到焦點。本研究中，受詞前置的現象可以大體分為兩種：高於時態詞組(TP)的受詞前置(object-fronting)，以及低於時態詞組的受詞前置(object-preposing)。雖然原本兩種受詞前置的位置不同，但在雙句分析模式裡，卻都移入包孕句補詞層的指示詞位置(SpecCP)，而得到焦點；這項結果也顯示了與上述其他焦點結構一致性的分析。 其他相關的焦點議題也同時在本研究中提出與討論。近來研究證實了時態詞組到動詞組之間的區域(屈折內域，Low IP Area)與補詞層的結構極為相似。而與Cheung (2013b)的研究不同的是，本論文分別對於補詞層以及屈折內域提出了分裂焦點假說(Split-FocusP Hypothesis)，並且提出了焦點排列的順序應為：「就」字焦點 > 「只」字焦點 >「是」字焦點 > 「連」字焦點。可是，在單句結構的情況下，無論是Cheung或是本研究的說法，都無法解釋某些語料中的焦點排列順序。進一步的探討發現，雙句分析可以提供已知的語言事實更佳的詮釋方式，也進一步證實了雙句分析的優點。
並列摘要
This study intends to unpack the complexities of the bare shi cleft construction in Mandarin Chinese. Different from the traditional mono-clausal structure of Mandarin clefts in the literature, a bi-clausal alternative, on a par with Wang (2007), is proposed in this thesis. In this proposal, the focus particle shi is analyzed as a copula verb, taking an embedded CP as its complement. The pre-shi elements are considered to be base-generated topics or adjuncts in the matrix clause. On the other hand, in an attempt to get focused, the constituents undergo overt movement to the embedded [Spec, CP] to be PF-adjacent to shi; hence the derivation of the clefts. In this way, it is found that the current proposal of the bi-clausal account can better accommodate the variants of focalized constituents (i.e. subject focus, adjunct focus, VP-focus, and V-focus). On this basis, the long-standing problem of the ungrammatical object focus in Mandarin is claimed to be caused by the non-constituency of the pre-shi elements. Even so, this issue can be tackled through the proposed bi-clausal analysis, and therefore the object can get focused. That is, the object-raising of focus is divided into two types: (TP-external) object-fronting and (TP-internal) object-preposing. Under the bi-clausal view, the presumed TP-internal object-preposing in the mono-clausal alternative is now unified as an instance of object-fronting since both involve the movement of the focalized object to the embedded [Spec, CP]. From another perspective, some focus-related issues are also dealt with in the current study. The fine structure of focus projections is mapped out within CP and the low TP area in the mono-clausal structure. As a whole, this finding is called Split-FocusP Hypothesis. In particular, it better accounts for the ordering structure of Mandarin focus particles (i.e. jiu, zhi, shi, lian(…dou)) in the mono-clause. However, with some problematic issues spotted, it is found that the bi-clausal analysis fares better in accommodating the split-up focus projections. Hence, it is hoped that the bi-clausal alternative sheds some light on the notoriously complicated issue of the bare shi clefts in Mandarin.