霧台魯凱語廣義全稱量詞

Abstract

本論文旨在對霧台魯凱語中的廣意量詞進行構詞、句法和語意分析。我們將依據 Partee et al. (1987)、Bach et al. (1995)、Keenan (2012)在句法和語意上的區分，將這些量詞區分為 D-quantifiers 和 A-quantifiers。前者為帶有定冠詞 (determiners) 概念的 makanaelre / patuelre / parlaelre 『全部』和 sana ka 『'每個』，後者則指的是在霧台魯凱中通過詞綴或副動詞的方式來構成「分配數詞 (distributive numerals)」(Gil 1982; Choe 1987; Zimmermann 2002; Cable 2014, etc.)。 在霧台魯凱的 D-quantification 中，makanaelre / patuelre / parlaelre 這三個全稱量詞在句法和語意上，與英文中的 all 相似，例如他們都可以修飾單數、複數和不可數定詞；語意上則允許句子有分配性和集合性的解讀。至於 sana ka 一詞，在一般的情況下，它的語意為「全稱條件陳述 (universal conditional statement)」，有著『只要是 ~』或『如果 ~』的意思。由於大部分的全稱條件陳述 (universal conditional statement)的邏輯式可以被轉寫為全稱量詞的型態，sana ka 同時也具有英文中 every 的意思，因為作為全稱量詞時，它只允許分配性的解讀。 在霧台魯凱的 A-quantification 中，霧台利用分配數詞作為名詞或動詞謂語的方式來表達英文中 each 的概念，這種量詞會自動排除集合性或累積性 (cumulative) 的解讀。被這些霧台的分配數詞給修飾的句子或片語，會被視為「分配額 (distributive share)」，並蘊含出兩種不同的分配性解讀，一個是分配性橫跨多個物體 (即 Participant-key readings)，另一個則是分配性橫跨多個事件 (即 Event-key readings)。

The current thesis aims to provide a morpho-syntactic and semantic analysis of universal quantifiers in Budai Rukai. We will divide these quantifiers to universal Dqunatifiers and A-quantifiers based on the syntactic and semantic distinction made by (Partee et al. 1987; Bach et al. 1995; Keenan 2012). As for D-quantification, the notion of determiners makanaelre / patuelre / parlaelre ‘all’ and sana ka ‘every’ will be examined, while A- quantification refers to the of each in Budai Rukai through affixes or adverbials to form distributive numerals (Gil 1982; Choe 1987; Zimmermann 2002; Cable 2014, etc.). In D-quantification of Budai Rukai, similar to all in English, the universal quantifiers makanaelre / patuelre / parlaelre syntactically singular / plural / mass DP, and semantically allow either distributive or collective interpretations. As for the word sana ka, in default it behaves a universal conditional statement with the meaning of ‘as long as’ or ‘if’. Since a universal conditional statement can be re-written to the form of a universal quantifier. When acting as a universal quantifier, sana ka behaves like English every, which permits only distributive interpretations. In A-quantification, Budai Rukai utilizes distributive numerals to function as either nominal or verbal predicates to express the notion of each, ruling out both collective and cumulative readings. A Budai Rukai sentence containing a distributive numeral specially marked as distributive share, can describe two distinct kinds of ‘distributive scenarios’: (i) a scenario where a distribution is over some plural entity (i.e. Participant-key readings), and (ii) one where the distribution is over some plural event (i.e. Event-key readings).