卡群布農語的結構重整與詞組結構

Restructuring and Phrase Structure in Takibakha Bunun

施朝凱(Chao-Kai Shih)

指導教授 ： 張永利

國立清華大學/人文社會學院/語言學研究所/博士(2017年)

摘要

本論文以微言主義(Chomsky 2001、2003)為理論框架探討布農語的結構重整、施用結構和謂語居首等三個句法現象。本文所調查的方言以卡群布農語為主，兼及與郡群布農語的比較。
本文的研究發現如下。首先，布農語的結構重整動詞包括副動詞、動貌動詞以及部分的主語控制動詞(含能願情態動詞與情緒動詞)，但不涵蓋跨語言常見的移動動詞和難易動詞。其二，帶主事者語態的結構重整動詞和帶非主事者語態的結構重整動詞所接的動詞補語在句法—構詞上有著顯著差異。其三，結構重整動詞與其動詞補語成分不可接鄰，後者必須位於句末。
本文進而論證，主事語態形式的結構重整動詞所接之動詞補語實為一完整、具有外元之語態詞組，而非主事語態的結構重整動詞所接之動詞補語實為不具外元之輕動詞詞組。該輕動詞不具外元，因而無法定值其內元的格位屬性，從而使得該內元必須進而跟主句之時制中心語進行句法呼應，使自身的格位獲得認可。因此，結構重整現象只發生於帶非主事語態結構重整動詞的句子。
本文提出，將輕動詞詞組進一步分層可以解釋上述在不同語態的結構重整動詞的動詞補語所觀察到的構詞—句法上的不對稱現象。詳言之，下層以輕動詞為中心語，而上層的語態中心語引介主事者，兩者分別認可不同的施用詞組。本文還提出，布農語的動詞居首語序是動詞中心語移位至補語層所造成的。
本論文對語言學理論的研究貢獻如下。一，本研究支持並深化前人所提出輕動詞詞組的分層結構。二，本文從新的面向證明動詞層本身無法認可施用詞組，因而無Pylkkänen’s (2008)所提出的低施用詞組。三，本研究發現，中心語移位比詞組移位更適合解釋卡群布農語及西部郡群布農語的動詞居首語序。最後是關於南島語語態的本質，本文提出新證據反對Rackowski (2002)以及Rackowski & Richards (2005)所提出的格位呼應分析。

關鍵字

結構重整 ； 施用結構 ； 動詞居首

並列摘要

This thesis investigates three different aspects of syntax, namely restructuring, applicativization and verb-initiality, under the Minimalist framework (Chomsky 2001, 2003), with special reference to Bunun, an Autronesian language spoken in Taiwan.
It is found that the class of the restructuring predicates in Takibakha includes adverbial heads, aspectual verbs, part of the “subject-control” lexical verbs, including desirative and emotion verbs, but does not contain motion verbs, easy-predicates. Moreover, a morphosyntactic asymmetry on the verbal complements is observed between AV-marked and NAV-marked restructuring predicates. What is more, the restructuring complements are never fronted with the restructuring predicates and thus appear clause-finally.
It is argued that the verbal complement in AV contexts is a full-fledged vP whereas the verbal complement in NAV contexts is a defective agent-less vP. This defective vP cannot value Case, which is why the lowest DO must enter into Agree relation with the upper T. Hence, restructuring actually occurs in NAV contexts only.
It is proposed that the observed asymmetric applicative morphology and complementation can be captured by a more fine-grained vP, in terms of Split vP analysis, where the expletive applicative, spelled out as is-, is licensed by lower v while the other applicatives as well as the agent are licensed by Voice. It is also proposed that the verb-initial word order of this language is attributed to V-to-C (Fin0) movement.
This study provides some clues to the debating linguistic issues. First, it provides supporting evidence for a split vP structure. Second, it provides new evidence against Pylkkänen’s (2008) proposed “low” applicative. Third, V raising analysis is favored over VP raising analysis in the case of Takibakha. Lastly, new evidence is offered contra Case Agreement approach proposed by Rackowski (2002) and Rackowski & Richards (2005).

並列關鍵字

restructuring ； applicativization ； verb-initiality

參考文獻

Manqoqo, Atul. 2011. A preliminary study on noun formation in Takibakha Bunun.

Li, Paul Jen-kuei. 1988. A comparative study of Bunun dialects. Bulletin of the

Chang, Henry Y. to appear. The AV-only Restriction and Locality in Formosan

Mercado, Raphael. 2003. Transparency is not less structure: A look at Tagalog

Wei, Ting-Chi. 2009. Wh-interrogatives in Amis. Language and Linguistics 10.2: 315-374.