霧台魯凱語格位標記有無探討

Abstract

本論文以功能語法之觀點探討霧台魯凱語的格位標記存在之有無。格位標記在以往的南島語言研究中通常區分兩層次探討。一是句法層次，格位標記通常被認為與主詞、受詞等句法角色有關。二是語意層次，格位標記通常也具有某些語意標記功能，常與其後方名詞論元是否定指與迴指等有所區分。本文試圖以言談分析為主要研究方法來確認格位標記是否於霧台魯凱語中具有句法或語意功能。 本論文共分五章，第一章主要介紹霧台魯凱語的使用人口分佈區域及其語言背景和本論文的研究方法。第二章呈現以往學者對於霧台魯凱語格位標記的理解及在敘述性故事中格位標記的分布情形。我們主要發現三種不同的決定詞 (以往俗稱的格位標記) ka/ku/ki在語料中分布情形相當不一致，其語用頻率的差異相當顯著。ka與ku的使用頻率極低，以至於很難擁有顯著的句法及語意標記能力。第三章主要介紹功能語法中格位系統的分類及其背後之主要驅力，並以及物性分析來檢視霧台魯凱語決定詞是否對語句中的及物性具有顯著的指向性；我們發現霧台魯凱語對及物性的高低並不會反映在決定詞的使用上。第四章試圖進一步進入句法層次探討，並闡明格位標記與句法角色通常不是呈現一對一的對應關係。所以霧台魯凱語決定詞與句法角色的不一致並非是特殊的語言現象。再者，為更深入理解於言談中霧台魯凱語使用者如何區分主受詞的概念，本文使用主題性測量的方法確定霧台魯凱語中主事者已被虛化成為主詞，而受事者常扮演非主詞的句法角色。而此現象第五章則統合前三章的結果，並提出本論文的研究成果。此外，本章也試圖與其他南島語研究做比較，並對未來所要作的近一步研究提出一些看法。

This thesis questions the existence of “case markers” in Budai Rukai. Case markers have long been considered as double-tier pre-nominal indicators. Syntactically, case markers represent the grammatical roles of the nominal argument they precede. Semantically, case markers also manifest certain semantic factors of the argument nouns such as definiteness, specificity, and anaphoric relation. This thesis uses the functional approach of grammar to explain the function of three determiners ka, ku, and ki, which are traditionally considered to be case markers in Budai Rukai. We use discourse analysis on Budai Rukai narrations to see the actual behaviors of these determiners. By doing so, we can reassess the claim that ka, ku, and ki are case markers in Budai Rukai. This thesis contains five chapters. Language background is introduced in chapter 1. Chapter 2 gives a summary of two important previous researches about the case marking of Budai Rukai. We also illustrate the inconsistency of the usage frequency of ka, ku, and ki. The ka and ku determiners appear very rarely in our Budai Rukai narration. This tendency makes them very hard to be connected to any syntactic and semantic functions. In chapter 3, we give the definitions and examples of the three major case marking systems and the adaptive force behind them. We also use the scalar Transitivity to examine our data. We find that the three determiners are not sensitive to transitivity. Chapter 4 further solidifies our claim by illustrating that case marking device is always not compatible with grammatical roles. Therefore, we measure the topicality of the agent and patient in Budai Rukai transitive sentence, and find that agents always have higher topicality than patient. This tendency shows that agents are the default subjects in sentences. The final chapter summarizes the preceding three chapters and provides our conclusion. We also compare our findings with other Formosan studies which are done in identical or similar theoretical framework. We, in the end, provide our suggestions for future research.