粤語的句末助詞相當豐富，能表達時間、焦點、語氣等多種意義，在過去曾引起廣泛的討論。在日常話語中，助詞往往可以多個連用，它們在句法上處於不同的層次，連用時，是一層一層的疊加。另一點值得注意的是，除了單音節助詞之外，粤語也有一些雙音節的助詞，如「罷啦」（baa2laa1）、「係啦」（hai2laa1），它們整體和前面的句子結合，學者一般認為它們跟助詞連用的情況不同。本文對粤語複合助詞進行了深入的討論，先考察它們的語義功能和語法特點，再探討它們的句法結構，並主張複合助詞是由句法產生的。 / 除了典型的複合助詞「罷啦」和「係啦」之外，本文也把「定啦」（ding2laa1）和「得㗎」（dak1gaa2）分析為複合助詞，四者都由一個謂詞性成分和一個功能性成分組成。在語義上，它們的意義跟其組成成分都有密切的關係，是兩者語義的相加。本文探討了它們的各種語義和用法，並指出它們的一詞多義現象可以透過「行」、「知」、「言」三域的跨域投射理論去解釋。 / 在語法特點方面，複合助詞與前面小句的配搭存在較多的限制；相反，單音節語氣助詞與前面小句的配搭則非常自由，這反映出複合助詞與前面的小句之間存在著一種緊密的關係，可以理解為一種選擇的關係。 / 因此，我們主張複合助詞在句法的深層結構可以分解為〔詞根（謂詞性成分）+詞綴（功能性成分）〕，謂詞性詞根是VP的中心語，選擇小句作為其補足語，通過謂詞性詞根由V到T再到C的移動，與處於T和C的後綴結合成複合助詞，另外，通過小句的移位，推導出表面的語序。 / 透過以上分析，本文對所謂「複合」作出了重新思考，指出「複合」和「連用」的本質其實是一致的。在深層句法結構中，各個成分處於不同的句法層次，一層一層的疊加，分別只在於助詞可以獨用，但複合助詞的謂詞性成分不能脫離後綴獨立使用。此外，本文的討論亦揭示了粤語句末成分由實變虛的可能性。 / Cantonese is renowned for having a rich inventory of sentence final particles which can express a large variety of meanings, such as time, focus and mood, etc. They have attracted a lot of interest in the literature. In daily conversations, particles can be used individually or successively. When they co-occur, they are located at different levels in the syntactic hierarchy. Even more intriguing is that there are some sentence final particles consisting of two or more syllables, which are known as "compound particles", such as baa2laa1 and hai2laa1. Previous studies suggest that they are different from the successive use of particles, because the two or more syllables in a compound particle attach to the preceding clause as a whole. The thesis investigates the semantic meanings and grammatical properties of compound particles. It also explores the syntactic structure of compound particles and argues that they are derived in syntax. / In addition to baa2laa1 and hai2laa1, the thesis considers also ding2laa1 and dak1gaa2 as compound particles. All of them consist of a predicative element and a functional element. Semantically, they are closely related to their components. The thesis demonstrates that different semantic meanings conveyed by a single compound particle are derived by metaphorical projections among various linguistic domains, such as the content, epistemic and speech act domains. / In terms of grammatical properties, the thesis has found that compound particles impose more constraints on the various constituents in the clauses that they are attached to, whereas the combinations of monosyllabic sentence final particles and the preceding clauses are less restricted or almost free. This phenomenon reveals the close relationship between compound particles and their preceding clauses, which can be interpreted as a kind of selectional relation. / For this reason, the thesis proposes that compound particles are decomposed into〔root（predicative element）+ suffix (functional element)〕in the underlying syntactic structure. The predictive element is the head of VP which selects a clause as its complement. The predicative element moves from V to T to C and forms a compound with the suffixes that are located at T and C. The surface word order is derived by further movement of the clause. / From the above analysis, the thesis re-examines the nature of "compound particles" and "successive use of particles". It suggests that they are consistent in their underlying structures with each of the elements occupies the head position of different phrases. The major difference between them is that every particle in a sequence can be used independently, whereas the predicative element in a compound particle cannot be separated from the suffix. Moreover, the thesis reveals the possible grammaticalization of the sentence final elements. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / 黃卓琳. / Thesis (Ph.D.) Chinese University of Hong Kong, 2014. / Includes bibliographical references (leaves 178-183). / Abstracts also in English. / Huang Zhuolin.

published_or_final_version / abstract / Linguistics / Doctoral / Doctor of Philosophy

This research targets the gender differences in the use of utterance-final particles in Cantonese with respect to the act of sajiao (撒嬌) under the speech act framework proposed by J. L. Austin. Sajiao is defined as the adorable petulance of a spoiled child or young woman who seeks material or immaterial benefits from an unwilling listener. It is hypothesized in this study that the use of utterance-final particles is not generally gender-linked. It is further suggested that some of the utterance-final particles are loaded with gender features which could feminize the utterances, thus performing the act of sajiao. This research has identified five utterance-final particles, including a newly emerged particle that has never been examined in the previous literature, which are believed to be gender-linked, namely 添 “tim1”, LU “lu3”, 噃 “bo3”, 喇喎 “laa3wo3” and 咖呵 “gaa3ho2”, and it has also analyzed meanings that are denoted by these five particles and the two genders’ usage of them. The investigator proposed that the first three particles could help soften the tone of an utterance, while at the same time, they add cuteness, so that the speakers present a more amiable impression of themselves, and thus could help them please or flatter the opposite sex. The two utterance-final particle clusters 喇喎 “laa3wo3” and 咖呵 “gaa3ho2”, on the other hand, are related to implicit demands. By skillfully employing these seemingly softer particles, female speakers could make the men who are appeared to be stronger to yield and satisfy their demands without creating any hard feelings. / published_or_final_version / Linguistics / Master / Master of Arts

published_or_final_version / abstract / toc / Linguistics / Master / Master of Philosophy

published_or_final_version / abstract / toc / Linguistics / Master / Master of Arts