現在我們是親戚了：臺灣華語中親屬稱謂泛化現象的認知語言學研究

Abstract

本研究旨在從認知語言學出發，探討臺灣華語的親屬稱謂泛化現象及其表現。漢語親屬系統以其相對複雜性而聞名於世，而漢語親屬系統的親屬稱謂有時可以被用於稱呼沒有親屬關係的人身上，例如稱呼年長的男性非親屬「爺爺」，或沒有親屬關係的女性為「姊姊」，這種現象被稱為「親屬稱謂的泛化」。 漢語親屬稱謂用語依地區不同而有所差異，因此研究中國親屬稱謂泛化的研究並不能完全適用於臺灣，而多數的相關研究多集中在親屬稱謂泛化的方言差異、分類、用法、功能以及形式等，不甚注重認知的研究角度。即使有部分偏向認知面的研究，對於發話者的心理運作層面，也未能多做著墨。因此，本研究期望從認知語言學的角度，闡釋臺灣華語中的親屬稱謂泛化現象，以及其運作表現。換句話說，本論文是以新的觀點探討存在已久的現象。文中我們提出三個問題：首先，是否所有親屬稱謂都能被泛化？如果不是，哪一些稱謂才是可以被泛化的？其次，在被泛化的親屬稱謂中，單音節親屬稱謂跟雙音節親屬稱謂彼此的相似、相異點為何？最後，在母語者的認知中，親屬稱謂的泛化跟泛化稱謂的不同結合形式，是如何運作的？ 為了解答這些問題，我們以「教育部重編國語辭典修訂本」附錄中的親屬稱謂、林美容 (1982) 提到的核心的親屬稱謂詞素、四個臺灣華語母語者熟知的臺灣閩南語親屬稱謂 (阿公、阿媽、阿伯、阿姨，參採「教育部臺灣閩南語常用辭詞典」所收錄內容)，以及這些稱謂的變體與其他較為常見的稱謂用法，做為關鍵字，在「中央研究院漢語平衡語料庫」和「PTT語料庫」搜尋，比對出屬於泛化親屬稱謂的用法。並進一步將這些擷取出的語料，依照形式及語意分為五類：無變化型、程度詞前缀型、前加姓名型、轉喻主題型、隱喻/慣用詞型；每類之下，再依親屬稱謂的音節分為雙音節類和單音節類。 在之後的語料分析中，我們主要運用概念整合理論 (Fauconnier and Turner 2002) 和構式語法 (Goldberg 1995, 2006) 作為理論的基礎，並納入有關語言觀點(Sweetser 2012)、隱喻及轉喻 (Lakoff and Johnson 2003)、禮貌理論 (Brown and Levinson 1987)、不禮貌行為 (Culpeper 2011) 等理論進行討論。透過分析，本研究提出「Head-Generalized kinship term」的構式。 在雙音節親屬稱謂中，長輩的親屬稱謂泛化數量最多，其中又以「阿姨」數量最大。在單音節親屬稱謂中，以平輩的親屬稱謂泛化數量最多：單從稱謂來看「妹」是數量最多的，不過若是把「兄」、「哥」這兩個指涉同一親屬的稱謂泛化數量加總，他們就會是平輩稱謂中，泛化情況最多的。 「阿姨」會成為泛化數量多的稱謂有可能因為她是小孩互動最多的親屬，因此這個稱謂容易被泛化。另外我們觀察到，「兄」跟「哥」其實在泛化表現上互有表現，甚至可以說他們處於「相互競爭」的階段，難以判斷哪一個稱謂在被泛化時更為強勢。 此外，就泛化的親屬稱謂用來稱呼說話者自己的情形，我們觀察到以下趨勢：1. 這種用法可以 (適度) 縮短對話人之間的距離；2. 發話者的意向是決定採用年長或年幼親屬稱謂的關鍵；3. 「兄」很少被用於自稱，可能是因為這個稱謂主要用於文字表達，而在口語上少見；4. 有部分發話者為了更尊敬聽者，會自稱為「魯蛇弟」或「魯蛇妹」。 再者，有一些長期以來已經固著的「Head-Generalized kinship term」構式，如「警察-泛化親屬稱謂」、「護士-泛化親屬稱謂」、「記者-泛化親屬稱謂」、「司機-泛化親屬稱謂」，在語序上後面都是接著雙音節的泛化親屬稱謂；而「Head-Generalized kinship term」構式的主題如果是「學生」、「學」、「師」的話，緊接其後的泛化稱謂，都是核心家庭的親屬稱謂。最後，本研究也藉由觀察語料中的兩個個案，探討「Head-Generalized kinship term」構式從單一指涉到任意指涉的語意變化。 總結來說，親屬稱謂的泛化主要是透過概念整合的機制進行，過程中涉及的關鍵因素涉及觀點的切換、顯影 (或側重)、隱喻、轉喻、禮貌與非禮貌的表達，反諷的態度也會有影響。「Head-Generalized kinship term」的組合扮演構式的角色：泛化的單音節親屬稱謂由於結構上容易組合，能產性高於雙音節親屬稱謂，這點在轉喻主題型的泛化親屬稱謂組合中，特別明顯。此外，只要在合適的情境下，「Head-Generalized kinship term」構式具有高度能產性，新的親屬稱謂泛化用法可以源源不絕。

This study focuses on the explanation of how kinship terms of Taiwan Mandarin generalized and their performance. The Chinese kinship system is a well-known relatively complex kinship system in all human communities around the world. Due to the richness of lexicons, kinship terms in Chinese kinship system might be used in a way much broader, such as to call an older non-relative male yéyé (爺爺) ‘grandfather’ or a non-relative female jiějiě (姊姊) ‘elder sister’, this is a phenomenon commonly mentioned as “the generalization of kinship terms” in Chinese-speaking areas. Chinese kinship terms of separate regions are not all the same. Hence, research of the generalization phenomenon in China may not appropriately explain the generalized cases occurred in Taiwan. Moreover, most studies account for the generalization of kinship terms from the view of dialectal variance, categorization, usage, function, and form, instead of a cognitive perspective. Even studies that take a cognitive approach, they fail to give us more information about speakers’ mental process. Thus, this thesis aims to elaborate how kinship terms of Taiwan Mandarin generalized and how they perform through cognitive linguistic approach, in other words, this study wants to see an old phenomenon with new eyes. We propose three questions at here: firstly, can all kinship terms of Taiwan Mandarin be generalized? If not, which terms are candidates for generalization? Secondly, what is the similarity/dissimilarity between generalized disyllabic kinship terms and monosyllabic kinship terms? Thirdly, how the generalization and its combination construed in native speakers’ mind? To answer these questions, we adopt kinship terms as keywords from the Appendix of the Revised Chinese Dictionary published in 2015 (Ministry of Education, MOE), the core morpheme of Chinese kinship terms proposed by Lin (1982), four well-known kinship terms of Taiwan Southern Min (viz., āgōng (阿公) ‘father/mother’s father’, āmà (阿媽) ‘father/mother’s mother’, āpeh (阿伯) ‘father’s elder brother’, and āyí (阿姨) ‘mother’s elder/younger sister’) by native speakers of Taiwan Mandarin (adopted from the Dictionary of Commonly Used Words in Taiwan Southern Min (MOE 2011)), as well as some alternative and additional forms of kinship terms. Next, these keywords are searched in the Academia Sinica Balanced Corpus of Modern Chinese and the PTT Corpus. We extract target data from the two corpora and categorize them into five types by their forms and meanings, these types are ‘bare form’, ‘degree modifier as prefix’, ‘modified by surname or name’, ‘metonymic theme preceded’, and ‘metaphorical/idiomatic usage’. Under each type, we further divide the data into two subtypes: the ‘disyllabic kinship term’ and the ‘monosyllabic kinship term’. Following the data collecting and categorizing procedure, this study analyzes these language materials by mainly adopting Conceptual Blending Theory (Fauconnier and Turner 2002) and Construction Grammar (Goldberg 1995, 2006); however, other theories such as viewpoint (Sweetser 2012), metaphor and metonymy (Lakoff and Johnson 2003), politeness (Brown and Levinson 1987) and impoliteness (Culpeper 2011) are also taken into discussion. Through the analysis, we propose there is “Head-Generalized kinship term” construction. Generally speaking, the amount of generalized kinship terms of older generation is the most between all disyllabic terms, and āyí is the predominant one in them. Between generalized monosyllabic kinship terms, kinship terms of peer generation are dominators, and mèi (妹) is the one that with more combinations; however, if we take all the combinations of xiōng (兄) and gē (哥) into consideration (both of them refer to ‘elder brother’), the sum of them occupies the most prominent part in peer generation. The reason of āyí becomes dominant could be attributed to that it is the one that children interact most and feel closer to, thus this kinship term is very possible to be generalized. In addition, the author observes some usage differences between xiōng and gē and argues that they are competing against each other in the domain of generalized usage, to decide which one is more dominant is a challenging task. Moreover, some tendencies of using generalized kinship terms that refer to oneself are found, including: 1. to shorten the mental distance between interlocutors (but not too close); 2. the speaker’s intention determines an older/younger kinship term; 3. xiōng less appears in verbal communication due to its nature of written form; 4. some speakers address themselves as younger siblings and ‘loser’ at the same time, in order to display more respect for addressees. Furthermore, some early-established “Head-Generalized kinship term” combination such as “police-generalized kinship term”, “nurse-generalized kinship term”, “journalist-generalized kinship term”, and “driver-generalized kinship term” persist their preference of preceding a disyllabic kinship term, and keep attracting adequate candidates to fit in the combination. Some recurred head parts of the “Head-Generalized kinship terms”, such as xué-shēng, xué, and shī can be followed by a group that contains kinship terms of a nuclear family. At last, we explore the meaning change of the “Head-Generalized kinship term” combination from referential uniqueness to referential randomness by observing two distinct cases in the data. To conclude, the generalization of kinship terms is mainly processed through conceptual blending, prominent effective factors are viewpoint shifting, profiling, metaphor, metonymy, politeness and impoliteness expression, and also ironical attitude, they collaborate in native speakers’ concept and produce the output. The combination of “Head-Generalized kinship term” behaves as a construction: monosyllabic kinship terms are more productive than disyllabic terms owing to its easily-compatible structure, and this attribute makes monosyllabic kinship terms generalized in a significant amount, especially in the category of metonymic themes preceded. Also, the “Head-Generalized kinship term” construction is highly productive, as long as a suitable context exists, novel usages of generalized kinship terms will always be allowed to create.