臺灣客語強化程度副詞之分析：以口語語料庫為本 - 政大學術集成

全文

(1)國立政治大學語言學研究所學位論文 National Chengchi University Graduate Institute of Linguistics Master Thesis. 臺灣客語強化程度副詞之分析：以口語語料庫為本. 治政 Degree Intensifiers in Taiwan 大Hakka: 立. ‧. ‧ 國. 學. A Study Based on a Spoken Corpus. er. io. sit. y. Nat 指導教授：賴惠玲博士. n. a. v. i l C Professor Huei-lingnLai Advisor:. hengchi U. 研究生：胡雪瀅撰 Student: Hsueh-ying Hu 中華民國一百零六年六月 June, 2017.

(2) 臺灣客語強化程度副詞之分析：以口語語料庫為本 Degree Intensifiers in Taiwan Hakka: A Study Based on a Spoken Corpus. BY. 立. 政治大 Hsueh-ying Hu. er. io. sit. y. ‧. ‧ 國. 學. Nat. A Thesis Submitted to the. Graduate Institute of Linguistics. n. al. ni C h Fulfillment ofUthe In Partial engchi. v. Requirements for the Degree of Master of Arts. June 2017.

(3) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. Copyright © 2017 Alyssa Hsueh-ying Hu All rights reserved. v.

(4) Table of contents List of figures ................................................................................................................ v List of tables ................................................................................................................ vi Acknowledgement ......................................................................................................... i Chinese abstract ..........................................................................................................iii English abstract........................................................................................................... iv Chapter Ⅰ Introduction............................................................................................ 1 1.1 Motivation and purpose ................................................................................... 1 1.2 Conventions of the data ................................................................................... 7 1.3 Organization of the thesis ................................................................................ 8 Chapter Ⅱ Literature review ................................................................................... 9 2.1 Defining degree modifiers and degree intensifiers .......................................... 9 2.2 The degree intensifiers, gradability and boundedness ................................... 11 2.3 The degree intensifier and subjectivity .......................................................... 16 2.4 Cross-linguistic studies of degree intensifiers ............................................... 20 2.4.1 Studies of degree intensifiers in English and other languages ............ 20 2.4.2 Studies of degree intensifiers in Mandarin Chinese and Taiwan Hakka ...................................................................................................................... 23 2.5 Remarks ......................................................................................................... 25 Chapter Ⅲ Data and methods ................................................................................ 27 3.1 Data collection ............................................................................................... 27 3.1.1 Materials ............................................................................................. 27 3.1.2 Procedures ........................................................................................... 27 3.2 Coding scheme ............................................................................................... 31 3.2.1 The syntactic structure of X ................................................................ 31 3.2.2 Syntactic subcategories of X ............................................................... 32 3.2.3 Grammatical functions of the [DI. + X] construction ......................... 42 Chapter Ⅳ Analysis and discussion ....................................................................... 45 4.1 Data distribution............................................................................................. 45 4.1.1 Corpus distribution of the six degree intensifiers across dialects ....... 45 4.1.2 Distribution of the syntactic categories of X among the six degree intensifiers .................................................................................................... 47 4.1.3 Distribution of the grammatical functions of the [DI. + X] construction .................................................................................................. 53 4.2. Analysis of modified elements of the six degree intensifiers ....................... 58 4.2.1 Canonical modified elements with the six degree intensifiers............ 58 4.2.2 Non-canonical modified elements with the six degree intensifiers .... 87 4.3 Grammaticalization and lexicalization in [DI. + X] construction.................. 92 Chapter V Conclusion ............................................................................................. 104 5.1 Summary of the thesis .................................................................................. 104 5.2 Directions for future studies......................................................................... 106 References ................................................................................................................. 108 Appendix ................................................................................................................... 113. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(5) List of figures FIGURE 1 The non-degree/ degree dichotomy and the interaction between scale and boundedness ......................................................................... 14 FIGURE 2 The bidirectionality of semantic pressure between the degree modifier and the adjective .................................................................... 15 FIGURE 3 Data collection ........................................................................... 28 FIGURE 4 Extraction of the concordance lines .......................................... 29 FIGURE 5 Encoding structure of syntactic categories and subcategories .. 33 FIGURE 6 Scale-based interpretation on the relative adjectives................. 70 FIGURE 7 Scale-based interpretation of absolute adjectives ...................... 80 FIGURE 8 Host-class expansion of the [DI. + X] construction .................. 93 FIGURE 9 Synchronic clines of lexicality and grammaticality of the [DI. + X] construction..................................................................................... 97. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(6) List of tables TABLE 1 Illustration of target usages of six degree intensifiers ................... 3 TABLE 2 Tone diacritics of four sub-dialects ............................................... 7 TABLE 3 Various categorizations on the degree modifiers ........................... 9 TABLE 4 Synchronic analysis of the degree intensifier .............................. 16 TABLE 5 Different aspects of the intensifiers ............................................. 19 TABLE 6 Details of thirty-two recordings of daily conversations .............. 27 TABLE 7 The diagnostics of scale types ..................................................... 39 TABLE 8 The four sub-dialects and the six degree intensifiers .................. 46 TABLE 9 Distribution of syntactic categories in X slot among six degree intensifiers ............................................................................................ 47 TABLE 10 Distribution of relative adjectives and absolute adjectives in APs .............................................................................................................. 48 TABLE 11 Distribution of grammatical functions of [DI. + X] construction among the six degree intensifiers and the syntactic categories............ 52 TABLE 12 The frequencies of 多 do1 ‘many/much,’ 大 tai3 ‘big’ and 好 ho2 ‘good’ ............................................................................................ 60 TABLE 13 The distribution of [DI. + 多 do1] construction and grammatical functions among the six degree intensifiers ......................................... 61 TABLE 14 List of instances modified by [DI. + 多 do1] construction ...... 61 TABLE 15 The distribution of [DI. + 好 ho2] construction and grammatical functions among the six degree intensifiers ......................................... 64 TABLE 16 The distribution of the [DI. + 大 tai3] construction and grammatical functions among the six degree intensifiers .................... 67 TABLE 17 The numbers of collocation of 大 tai3 ‘big’ and NUM-CL morphemes ........................................................................................... 69 TABLE 18 The distribution of three types of closed scale .......................... 74 TABLE 19 The distribution of types of stative verbs .................................. 82 TABLE 20 The comparison of each degree construction by frequency ...... 94 TABLE 21 The top 10 collocability of the [DI. + X] construction (X= relative adjectives) ............................................................................. 113 TABLE 22 The top 10 collocability of the [DI. + X] construction (X= absolute adjectives) ............................................................................ 114. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(7) 謝辭 Acknowledgement “Si no escalas la Montaña jamás podrás disfrutar el paisaje. Y aunque parezca interminable, nunca mires abajo.”—Pablo Neruda 智利詩人聶魯達曾寫下：「若未曾親自攀至頂峰，怎看得見山頭美景？雖然仍有一座又一座的山等著我們征服，但你會看得更高更遠。」短短一年撰寫論文的日子，常常讓我有正在攀登一座比一座更難的高山之感。但最後的喜悅之情難以表述，從最初決定要攀怎樣的山、看怎樣的景色都不假手他人，即使迷路跌倒受苦受傷，這些旅程中所有的磨難、恐懼與不屈服的淚水與汗水，都將雕琢自己，成為更好及更善良的人。走完這段旅程得感謝路上相遇的引路人，他們有的是祝福（blessings）、有的則是課題（lessons），不論哪一種，我都心懷感激，是他們砥礪了現在挺直背. 政治大. 脊站得平穩的自己。感謝我的指導教授賴惠玲老師，她帶領我進入「語言學」的世界，自大學部的必修《語言學概論》及選修《語意學導論》始，便讓我對這塊領域有了熱忱，而後指導我撰寫大專生研究計畫，開啟我的研究之路。老師對學術的嚴謹與認真，讓我在撰寫論文的過程中，不敢有一絲懈怠；同時，老師也讓我在 Hakka Lab 學習很多客語知識。沒有老師嚴格的把關，這本論文無法完成。亦謝謝兩位清大口委：葉瑞娟老師與黃漢君老師對這本論文的肯定，以及提出許多讓這本論文更臻完美的建議。其次，謝謝語言所的老師：何萬順老師的幽默風趣把抽象的句法學講得生動有趣，上課絕無冷場，帶領我們如何運用邏輯思考在句法領域有力地辯證；徐嘉慧老師在言談分析及語意領域的豐富學識及對學術的嚴格要求，讓我在這兩個領域有了紮實的訓練；黃瓊之老師的第二語言習得，讓我對領域有更廣的認識；張郇慧老師永遠是語言所最溫暖的大家長，口試前幾天以陽光的笑容勉勵我，讓焦慮不安的我心安不少；鍾曉芳老師讓我對語料庫有深入的研究，也讓我在處理資料庫的議題更得心應手；曹逢甫老師學識淵博，深入淺出地將語用學講得平易近人。老師們像引路人給我這趟旅途將會用到的知識與工具，之後放手讓我走自己的路。此外，亦感謝助教惠鈴學姐在行政上的協助與時常關心我們的課業、生活。. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 除了引路人外，也一一感激旅途上出現的每個珍貴的旅伴── Gracias a mi amiga Alba Ho, 何欣宜。Ella siempre escucha mis quejas con paciencia, al encontrar los altos y bajos en la vida cotidiana, me animó a seguir avanzando durante los días más difíciles, nos confortamos y nos enterramos con la aromaterapia. Gracias a Catalina Méndez, mi maestra de español que me enseña todo sobre este hermoso idioma.. i.

(8) 謝謝一起撐過最難的某堂課的孟潔好朋友，我們大笑、抱怨、認認真真地過好生命的每分鐘。即使前路難行，仍一步步手把手地相互扶持。沒有妳耐心傾聽我所有的難過與難言，這趟旅途我可能走不了這麼遠。謝謝俐蓁，人在異鄉依舊惦念著我，偷偷寄來布達佩斯的黃昏，寫了滿滿的話鼓舞我、支持我。謝謝韶君姐姐常常關心我的心情，在我低潮快撐不下去的時候，給我最溫暖的話語。謝謝認識很久老朋友們：勻芊，義不容辭地當我的編碼員，在我語感死光光的時候，提供很多見解；嘉純，懷念我們相聚的時光，即使彼此在不同領域打拼，也不忘相互鼓勵；劭欣，總是耐心回答我關於華語教學的各種問題，懷念我們仨的蛋糕時光及厭世寶貝；Wayland, thanks for always listening to my thoughts & exploring every aspect of the academia together；也謝謝英文系的師長易道老師、Prof. John Corrigan、Prof. Rey Bianchi 及 Prof. Ruth Martin；謝謝學弟妹們倍華、亦凡、雅庭、倢君、蓓怡、培綸；謝謝英文所的渝庭學姐，常常跟我分享厭世之心；謝謝華文所的佩玲學姐、雅筑大大、阿鐘。謝謝在開始這趟旅程前，已經陪了我九年的好麻吉徐子涵與鄭郁涵，尤其是郁涵，工作繁忙之餘還得聽我訴苦，安慰我眼淚擦乾繼續往前走。最後，感謝我自己的不輕言放棄，即使一路磕磕碰碰，心碎心傷心苦心累，咬著牙也要看見山頂的絕世美景。我始終相信：「Fortuna iuvat.」（幸運眷顧勇敢之人），兩年唸完研究所，期間發表兩三篇論文，我想除了旅途豐碩外，亦終可稱得上是「qualified graduate student」，無愧於心。𠊎𠊎當承蒙吾屋下底肚个阿公、阿爸阿姆、老妹崴崴摎老弟緯哥。佢兜个支持同鼓勵係這本論文完成个動力。寫論文个時節，𠊎𠊎總想等愛仰般正做得將客家話个靚摎特別寫落來。好在自家煞猛研究，總算對麼个都無到一本賁賁个，盡暢、盡歡喜。做得為客家話个保存出一份力，係𠊎𠊎个福氣也係責任。語言消失當遽，希望這本研究使得為想曉得客家話底背个強化詞仰般仔用个人，提供一个盡詳盡个紹介！. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 最尾，這本論文𠊎𠊎乜愛獻分今這下佗佛祖身項修行个阿婆，𠊎𠊎毋摎你忘忒。. 於一百零六年季夏伏月梅雨季裡的政大. ii.

(9) 國. 立. 政. 治. 大. 學. 研. 究. 所. 碩. 士. 論. 文. 提. 要. 研究所別：語言學研究所論文名稱：臺灣客語強化程度副詞之分析：以口語語料庫為本指導教授：賴惠玲. 博士. 研究生：胡雪瀅論文提要內容：（共一冊，貳萬柒仟貳佰參拾陸字，分五章）日常生活中，由於表達的需要，說話者常會使用強化程度副詞，作為增強. 政治大灣客語六個常見強化程度副詞「恁、當、蓋、盡、異、還」視作強化程度副詞立或強調其所欲言的一種手段。本論文旨在藉由觀察口語語料庫中的對話，將臺. 詞群，進一步以[DI. + X] 結構討論以下面向：(一) 探討其後接修飾之成分在句. ‧ 國. 學. 法、語意層面之表現。本文發現其後接成分可分為典型成分類與非典型成分. ‧. 類，以形容詞一類為大宗，其後接成分會受到等級性（gradability）、有界與無. y. Nat. 界性（boundedness）等參數影響語意的解讀。(二) 探討強化程度副詞[DI. + X]. er. io. sit. 結構於句法所扮演的角色。本文發現此結構多半出現在謂語之位置，除了強化語意外，同時也展現說話者之態度或對某主題之評價。(三) 探討強化程度副詞. al. n. v i n 結構於口語使用中的語法化與詞彙化現象，本文認為這些現象由於溝 Ch engchi U. [DI. + X]. 通需要，亦會伴隨提升主觀性（subjectivity），尤其是強化程度副詞「恁」的主觀性為詞群中最高。藉由上述不同面向探討與分析臺灣客語六個強化程度副詞「恁」、「當」、「蓋」、「盡」、「異」、「還」之表現，以期能更系統性解釋強化程度副詞是如何在口語使用中展現與運作。關鍵字：強化程度副詞、等級性、有界性、語法化與詞彙化、主觀性、臺灣客語. iii.

(10) Abstract. The aim of the present study is to examine the [DI. + X] construction manifested by the six degree intensifiers 恁 an2, 當 dong1, 蓋 goi3 , 盡 qin3 ,異 i3 , and 還 han5 in Taiwan Hakka based on their authentic usages in a spoken corpus. Three findings can be reported as follows. First, by exploring the modified elements from the aspect of their structure and meaning, the study classifies the modified elements into canonical and non-canonical usages. X as AP takes up the majority of the. 政治大. occurrences of the syntactic categories. Such manifestation can be accounted for. 立. various linguistic behaviors of gradability and boundedness. Second, the [DI. + X]. ‧ 國. 學. construction usually serves as a predicate, the prominent grammatical function, in a clause. This implies that the [DI. + X] construction is generally adopted by different. ‧. speakers to not only add intensity but also make judgements or comments. Third, the. y. Nat. sit. [DI. + X] construction displays grammaticalization and lexicalization. Due to. n. al. er. io. communication needs, these linguistic phenomena can also motivate the subjectivity.. i n U. v. It is found that the [恁 an2 + X] construction can display the highest subjectivity. Ch. engchi. among the six. To sum up, this study, with its analysis of different syntactic and semantic aspects of the six degree intensifiers in Taiwan Hakka, offers useful insights into explaining the manifestation of degree intensifiers during speech.. Key words: degree intensifiers, gradability, boundedness, subjectivity, grammaticalization and lexicalization, Taiwan Hakka. iv.

(11) Chapter Ⅰ Introduction 1.1Motivation and purpose In daily life, when people plan to be more expressive or emotional in speech, it is common for them to use intensification. Intensification is usually realized through the use of adverbs, the category of language which has long been discussed fervently (a category of language which has been the subject of much discussion amongst linguists) due to its vagueness. In terms of grammatical function, many researchers have reached a consensus that an adverb can be plausibly utilized as degree, scope,. 政治大 positioned usually on the predicate 立 or pre-predicate modifying adjectives, some verbs, modal particle or negation. In terms of syntactic behavior, an adverb is said to be. ‧ 國. 學. or the entire predicate (Zhang, 2000; Liu & Pan & Gu, 1996; Biq & Huang, 2016). Among all the adverbs, it should be noted that the intensification of adjectives relies. ‧. on the use of a degree intensifier, which serves to situate the meaning of the adverb on. sit. y. Nat. a certain scale of gradability such as that of quantity, scope, time, or height. Its. n. al. er. io. modified elements normally adjectives, are usually gradable. A degree intensifier is to. i n U. v. produce the heightening effect on the target entity (Hoye, 1997).. Ch. engchi. In the extant literature, some of the attention paid to degree intensifiers in the extant literature seeks to sort out the roles played by degree intensifiers in the context of the development of language use and to build up inventories. Stoffel (1901) conducted pioneering research to explore the development of degree intensifiers in both middle and late English. In Bolinger’s study (1972), degree intensifiers are labeled as boosters, a quality scales upward with the capacity to scale a quality upward in intensity. They offer a picture of "fevered invention" (Bolinger 1972: 18) which implies that these special words could motivate the renewal of a language constantly. Next, the interest in degree intensifiers shifted to the discussion of 1.

(12) theoretical issues such as the interaction between degree intensifiers and modified elements or the processes involved in the grammaticalization process of degree intensifiers. For example, Kennedy & McNally (1999) investigated the internal structure of degree intensifiers and modified adjectives. Gradability is a key concept in explaining the felicity in these two sentences: Gina is very tall and?? Gina is completely tall. Paradis (2001) further indicated another prominent concept boundedness, which is said to highly correlate with gradability. She argues that an adjective may select its degree intensifier and cause the interpretation of the holistic. 政治大 significant studies raise the issue that certain constraints exist when a degree 立. construction to be more restricted or more expanded than the normal usage. These two. intensifier co-occurs with the modified elements.. ‧ 國. 學. Many cross-linguistic studies have delved into the issue of degree intensifiers as. ‧. well. Investigations of degree intensifiers in German are mostly about the syntactic. sit. y. Nat. behavior of degree intensifiers and the interplay with modified adjectives (e.g.,. io. er. Löbner 2012). As for Spanish, many researchers take a sociolinguistic approach to observing the use of degree intensifiers in some specific speech community (e.g.,. al. n. v i n C hCortés-Torres, 2013). Martínez López, 2009; Brown and As for Japanese, the main engchi U. research concerns are how a degree intensifier interacts with syntactic structures and nuance usages in regional dialects (e.g., Tsujimura, 2001; Matsubara, 2008). As for the use of degree intensifiers in Mandarin Chinese, much of the previous literature on degree intensifiers has mainly focused on dealing with the semantic changes or idiosyncrasies of particular individual intensifiers. For example, many studies have investigated the degree intensifier 很 hen3 by way of its syntactic, semantic, and pragmatic aspects (e.g., Chang 2004; Lee, 2011; Chen, 2014). Also, other degree intensifiers such as 非常 fei1chang2, 蠻 man2 , and 太 tai4 have been examined in different genres (e.g., Shang, 2010; Hong & Lim, 2012; Chang, 2013). 2.

(13) Unlike those in Mandarin Chinese, the representation of degree intensifiers in Taiwan Hakka is different. Two studies in Taiwan Hakka have indicated that frequently-used degree intensifiers occur mostly in monosyllabic forms (Qi, 2007; Li, 2008). Six degree intensifiers in Taiwan Hakka—當 dong1, 蓋 goi3, 盡 qin3, 恁 an2, 異 i3, and 還 han5 —are identified in both studies to highlight their functions to boost. the meaning during speech 1. The canonical usages according to the Taiwanese. Hakka Dictionary of Common Words and the Hakka Dictionary of Taiwan are summarized in Table 1.. 政治大 TABLE 1 Illustration of target usages of six degree intensifiers 立 ‧ ‧‧ 恁衰過雞仔又走忒三隻。. an2 coi1go3 gie1e2 iu3 zeu2ted2 sam1 zag2. ‧. ‧ 國. 恁 an2. 學. Canonical adjectival usages. Degree intensifiers. sit. y. Nat. very unlucky hen again walk-TED three-CL. ‘It is really unlucky that the three hens have gone. ’. io. er. ‧ ‧ 這領衫看起來當靚，愛幾多錢正買得著？. al. v i n C hsam1 kon3hi2loi5Udong1 jiang1 lia2liang1 engchi. n. 當 dong1. this-CL clothes look like oi3. gi2do1. very pretty. qien5 zang3 mai1ded4 do3 2. want how much money so. buy-DED-DO. Though two relevant studies mainly discuss usage in Taiwan Hakka with examples from the 四縣 xi3ien3 dialect, the author has consulted dictionaries and searched corpora to find that 當 dong1, 蓋 goi3, 盡 qin3, 恁 an2, 異 i3, and 還 han5 also occur in other sub-dialects. Hence, the oral data presented in the present study will include data from four sub-dialects. Notably, one of the sub-dialects 四縣 xi3ien3 shows regional variants. That is, in the northern area, native speakers tend to adopt 當 dong1 and in the southern area, people tend to adopt goi3. In order to discuss Taiwan Hakka thoroughly, the study regards 蓋 goi3 as another independent degree intensifier in the analysis. 2 The following abbreviations are used for the grammatical functions in this thesis: CL (classifier), COP (copular verb), DM (discourse marker), GEN (genitive marker), NEG (negative morpheme), NOM (nominalizer), Q (question marker), PRF (perfective aspect), and PRT (particle). 3 1.

(14) ‘This clothes looks very pretty. How much should I spend buying it?’ 蓋 goi3. ‧ ‧ 細妹仔蓋鬧，佢係麼儕？. 這. lia2 se3moi3e2 goi3 nau3，gi5 he3 ma2sa5 this. little girl very pretty she COP. Q. ‘This girl is very beautiful. Who is she?’ ‧ ‧ 這兜魚仔肚該條魚仔个味道盡壞. 盡 qin3. lia2deu1 ng5e2du2 ge3 tiau5ng5e2 ge3 mi3to3 qin3 fai3. 政治大食毋得了. these school of fish that-CL fish NOM smell very bad 怕. siid5m5ded4. le1. afraid eat-NEG-DED PRF. ‧. ‘This one of these fish smells very bad. I’m afraid that it. io. al. 新開. 个. n. ge3 gien1 xin1koi1. Ch. ge3. 店頭. ‧ 異 i3. sit. 該間. er. y. can’t be eaten.’. Nat. 異 i3. pa3. 學. ‧ 國. 立. diam3teu5. v ni. e n g c h i U store. that-CL newly-open NOM. ‧ 大。 tai3. very large. ‘That newly-opened store is very large.’ 還 han5. 佢. 生著. ‧ 還. ‧ 媸了。. gi5 sang1do2 han5 ze2 le1 she look-DO very ugly PART ‘She is really ugly.’. Speaking of spoken Taiwan Hakka, one can find two interesting linguistic facts when a speaker attempts to employ a degree intensifier to strengthen his or her 4.

(15) utterances. First, syntactic variations can be found when a degree intensifier of Taiwan Hakka is adopted. As above-mentioned, elements such as verb phrases and adjectival phrases which can be modified by a degree intensifier are fixed and predictable in Taiwan Hakka. However, the scope for the modification of elements appears to be more flexible during speech. Observe the following utterances in example (1), wherein a speaker employs the degree intensifier 當 dong1 to modify a noun 鄉下 hiong1ha3.. (1) [HL70:11] 3 F3：頭擺. 治政‧ ‧‧ 大啊嗬... 斯戴當鄉下立. PART SII live very rural. 學. past. ‧ 國. F3: teu5bai2 a3ho2…sii5 dai3 dong1 hiong1ha3. ‧. ‘F3: When it comes to my childhood, I lived in a very rural area in the past.’. sit. y. Nat. io. er. Second, the linguistic strategies which can be used for intensification cannot be doubly used. Yet, interestingly, in Taiwan Hakka reduplicated forms which already. al. n. v i n contain a certain degree of intensity C hcan co-occur withUthe degree intensifier 恁 an2 engchi as in example (2) below. (2) [HL13:149] ‧‧ F1: (0)係...轉南部喔...也係講去...斯去台北喔...就看著想...唉喲...該仰會恁白 ‧‧ 白仔... F1: he3 zon2 nam5pu3 ho1 ia3he3 gong2hi3 sii5 hi3 toi5bed4 ho1 qiu3 kon3do2 xiong2 ai2io2 ge3 ngiong2voi3 an2 pag5pag5e2. 3. The sub-dialect, the number of conversation and the turn of excerpted utterance in the spoken corpus will be annotated in the form of [D Num: Turn]. The abbreviations for the sub-dialects are NXI= 北四縣 bed4xi3ien3 (north); SXI= 南四縣 nam5xi3ien3 (south); HL=海陸 hoi2liug8; TB=大埔 tai3bu1; NP=饒平 ngiau5pin5. 5.

(16) F1: and return southern PART or talk about SII go-to Taipei PART then see-DO ponder PART there how come very white. F1: Toward the south, or to talk about it, to Taipei (i.e., the north), I was wondering aww why there were so many white (tong blossoms). However, a broad picture of how degree intensifiers work with the post-modified elements to be a construction during speech and how the use of degree intensifiers anchors in a speaker's attitude to display the speaker’s subjectivity toward the hearers remain unexplored. Partington (1993: 178) pinpoints that “[t]he importance of. 政治大 praising, persuading, insulting, and generally influence listener's reception of the 立. intensification in the communicative process is that it is a vehicle for impressing,. message.” What’s more, as Athanasiadou (2007: 561) emphasizes, the selection of. ‧ 國. 學. degree intensifiers in spoken contexts indeed works to “mentally transport the. ‧. listeners to another space, allowing them to observe the characterization of the. sit. y. Nat. entities.” Hence, the aim of this current study is to explore the system of degree. io. er. intensifiers in Taiwan Hakka based on their authentic usages from the NCCU Corpus of Spoken Hakka (32 daily conversation recordings; time duration in total: 10 hours. n. al. 19 minutes).. Ch. engchi. i n U. v. Three related research questions are to be tackled in this study: (1) What is the distribution of the syntactic categories of the modified elements for each degree intensifier in spoken Taiwan Hakka? (2) What syntactic and semantic differences can be further teased out with regard to the modified elements? (3) How does each of the six degree intensifiers differ regarding the degree of grammaticalization, lexicalization and subjectivity?. 6.

(17) 1.2 Conventions of the data The oral data treated in the present thesis is obtained from The NCCU Corpus of Spoken Hakka (http://spokenchinesecorpus.nccu.edu.tw/index.php). Further supplementary data is also taken from the Taiwanese Hakka Dictionary of Common Words (教育部台灣客家語常用詞辭典) (http://hakka.dict.edu.tw), and the Hakka Dictionary of Taiwan (台灣客家話辭典). Since four sub-dialects (四縣 xi3ien3, 海陸 hoi2liug8 , 饒平 ngiau5pin5 , and 大埔 tai3bu1 ) will be examined together in this study, the data presented in this thesis are mainly marked in tone categories (調類).. 政治大手冊) on the website of the Ministry of Education, Taiwan 立. The tone diacritics are sourced from the Hakka Pinyin Manual (客家語拼音方案使用. (http://language.moe.gov.tw/result.aspx?classify_sn=42&subclassify_sn=447) and are. ‧ 國. 學. presented in Table 2.. ‧ y yingqu. yingru. ‘上聲’. ‘陰去’. ‘陰入’. ‘陰平’. n. al. Sub-D. 1. C h2. sit. shansheng. yangpin yangqu yangru. er. yingpin. io. Tones. Nat. TABLE 2 Tone diacritics of four sub-dialects. 4n U engchi 3. iv. ‘陽平’. ‘陽去’. ‘陽入’. 5. 7. 8. 四縣 xi3ien3. 24. 31. 55. 2. 11. -. 5. 海陸 hoi2liug8. 53. 24. 11. 5. 55. 33. 2. 33/35. 31. 53. 21. 113. -. 54. 11. 53. 24. 2. 55. -. 5. 大埔 tai3bu1 饒平 ngiau5pin5. 7.

(18) 1.3 Organization of the thesis The thesis is organized as follows. Following ChapterⅠ, Chapter Ⅱ presents significant concepts and reviews previous studies of degree intensifiers crosslinguistically. Data and methodology are presented in Chapter Ⅲ. Subsequently, Chapter Ⅳ gives an analysis of degree intensifiers including the patterns in the collocations and also the structural variants and semantic differences in the discourse under consideration. Finally, concluding remarks and suggestions for further studies will be proposed in Chapter Ⅴ.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 8. i n U. v.

(19) Chapter Ⅱ Literature review This chapter will first define the major terminology used in this present study and introduce the various categorizations. Next, three significant parameters— gradability, boundedness, and subjectivity —will be explored when we deal with degree intensifiers in context. Last, the relevant extant literature on this issue will also be further presented and examined in order to depict a thorough view of the research on degree intensifiers.. 政治大 Intensification is a common 立 linguistic expression when people intend to be more. 2.1Defining degree modifiers and degree intensifiers. ‧ 國. 學. expressive or emotional during speech. The heightening effects that such linguistic usage produces have aroused much research interest. Yet, the use of various different. ‧. terminologies to describe the related phenomena has made the understanding of the. sit. y. Nat. scope of intensification elusive. In the very beginning, Stoffel (1991) merely adopted. n. al. er. io. two terms to classify degree modifiers. Later researchers have enlarged the. i n U. v. classification to include different degree modifiers. Table 3 displays the thorough the. Ch. engchi. categorizations different researchers have employed in their studies.. TABLE 3 Various categorizations on the degree modifiers Stoffel (1901) Bolinger(1972). Intensives Boosters. Quirk et al.. Downtoners. Compromisers. Diminishers. Amplifiers. (1985). Maximizers. Allerton (1987). Scalar modifiers. Boosters. Minimizers Downtoners. Approximators. Compromisers. Telic. Absolutive. Differential. modifiers. modifiers. modifiers. 9. Diminishers. Minimizers.

(20) Paradis (1997). Reinforcers Maximizers. Attenuators. Boosters. Approximators. Moderators. Diminishers. The significant phase of exploration of intensification can be marked by the use of expression degree words which refer to a scaling of a quality either upward or downward or in-between (Bolinger 1972). Such words usually associate with adjectives and adverbs. In addition, degree words not only operate grammatical processes but also reveal semantic loading. Bolinger’s categorization (1972) reflects the broad scope of intensification including adjectives or even the use of the indefinite. 政治大 categorization (Bolinger 1972: 17): 立. article. The term intensifier as a cover term may refer to the four subtypes in his. ‧ 國. e.g. He is a perfect idiot.. 學. Boosters: upper part of scale, looking up.. She is terribly selfish.. sit. She is fairly happy.. Nat. e.g. He is rather an idiot.. io. al. n. Ch. er. Diminishers: lower part of the scale, looking down. e.g. It was an indifferent success.. They were little disposed to argue.. engchi. Minimizers: lower end of the scale e.g. He is a bit of an idiot.. y. ‧. Compromisers: middle of the scale, often trying to look both ways at once.. i n U. v. I don’t care an iota for that.. As mentioned previously, various labels such as intensifiers, degree intensifier, degree intensification, intensive adverb, amplifier, booster, and maximizer are used to indicate the adverbs which aim to boost or maximize the meaning they modify (Stoffel 1901; Quirk et al. 1985; Ito & Tagliamonte 2003). In addition, it is suggested that the intensifiers are inherent with two characteristics: (i) versatility and color and (ii) the capacity for rapid change and the recycling forms. The former refers to a 10.

(21) constant invention and illocutionary influence, while the latter indicates that an intensifier is inclined to change its meaning and recycle through times. To be more precise and consistent, the present study will adopt the expression degree intensifier, which specifically refers to those functions to intensify, as the key term.. 2.2 The degree intensifiers, gradability and boundedness Generally speaking, a degree intensifier is usually positioned adjacent to adjectives, adverbs, and certain types of verbs or broadly, the predicates modified by. 政治大 explain the characteristics of degree intensifiers (Bolinger, 1972). A more fine-grained 立 antecedent. However, the syntactic behavior of degree intensifiers may not fully. semantic analysis should also include gradability or boundedness. Under most. ‧ 國. 學. circumstances, an adjective which can be modified by a degree intensifier is said to be. ‧. gradable on a scale. For example, the gradable adjective tall denotes the scale of. sit. y. Nat. tallness. The predicate very tall in the sentence Gina is very tall will entail a reading. io. er. that Gina’s height exceeds the standard of tallness. Yet, the notion of gradability brings about the constraint that only certain adjectives can be modified. Hence, the. al. n. v i n sentence Gina is completely tallC would be infelicitous U h e n g c h i because there is no closed. endpoint on the scale of tallness. The degree intensifier completely cannot be used with the adjective tall. Some researchers have proposed the idea of scalar structure or a configurational meaning structure to solve this situation triggered by the interaction between the degree intensifiers and the gradable adjectives (Kennedy & McNally 1999, 2005a; Paradis 1997, 2001, 2008; Toledo and Sassoon, 2011; Burnett, 2012). The subsequent sections will review the two most crucial works. First, extending the observation made by Bolinger (1972), Kennedy & McNally (1999) examine the degree (of) modification in de-verbal gradable adjectives. The general semantic type for any adjective which is gradable can be formulated as <d <e, 11.

(22) t>> and for a degree morpheme the formulation can be illustrated as <d <e, t>, <e, t>>. The computational equivalents of adjectives and degree modifiers can be formulated as [[A]]= λdλx. mA (d) (x) (<d, et>), where mA(x) represents the projection of x onto the scale associated with the adjective A (Kennedy and McNally 2005: 367). The concept of gradability is usually realized by a so-called scalar structure. As they indicate, gradability as a fundamental semantic property can produce effects “beyond adjectives to other lexical categories” (Kennedy & McNally 2005a: 348). They further claim that a gradable adjective can project either an open. 政治大 of scale variation can be placed on a scale graded from open scale, to lower-closed立. scale or a closed scale according to the characteristics of the adjective. The structure. scale, to upper-closed-scale, and then to totally-closed-scale adjectives. Whether the. ‧ 國. 學. two ends with or without minimal or maximal value can result in the scale variations. sit. y. Nat. (3). ‧. in an adjective. These types can be visualized as follows.. io. al. n. Open (e.g. tall). er. Typology of Scale Structure. Lower closed (e.g. dirty) Upper closed (e.g. clean). Ch. engchi. i n U. v. Totally closed (e.g. full). An open-scale adjective (e.g., tall) has neither a minimal nor a maximal value. A lower-closed-scale adjective merely has a minimal value. Taking the adjective dirty as an example, an entity being described as dirty cannot be 100% dirty. Similarly, an upper-closed-scale adjective can be contrasted with a lower-closed-scale one. It only has a maximal value. Taking the adjective clean as an example, an entity being described as clean is usually at its maximum. There is no minimal degree of 12.

(23) cleanliness. Next, a scale with both minimal and maximal value is the type of totallyclosed-scale adjective. For instance, the adjective full is usually used to describe an entity as completely full. For there to be any values lower than 100% of fullness is unacceptable. The characteristics that are manifested by these scale types can explain their interaction with certain degree intensifiers well. Another crucial contribution of Kennedy & McNally (2005) lies in their discussion of the interplay of gradability and the standard of comparison in context, which can be specified as the relative/absolute distinction. Since the characteristic of. 政治大 describing an entity as being tall, the height of this entity exceeds the norm for the 立. an open-scale adjective is a lack of endpoints, it is context-dependent. That is, when. group to which this entity belongs. The standard of the comparison thus varies. This. ‧ 國. 學. type of adjective is called a relative adjective. Yet, the standard of comparison for the. ‧. closed-scale adjective which has minimal value or maximal value or both is usually. sit. y. Nat. fixed and does not vary through contexts. This type of adjective is called an absolute. io. er. adjective. Kennedy & McNally’s proposal gives insights into interpreting adjectives and motivates other studies of adjectives such as those of Mandarin Chinese (Lin &. n. al. Peck 2016).. Ch. engchi. i n U. v. Second, granted the fact that gradability plays a crucial role in interpreting the meaning of the degree intensifier, Paradis (1997, 2001, 2008) explores the possibilities of multiple layers when analyzing the grouping of degree intensifiers and adjectives. She pays much attention to the concept of boundedness within a scale. The concept is prominent that refers to “a high-level schematic domain mode” (Paradis 2001: 3) and highly correlates with gradability. Three layers—gradability, oppositeness, and boundedness—are identified and configured into the meaning structures (see Figure 1). Degree modifiers are classified into two types: scalar modifiers (e.g., very, terribly, fairly) and totality modifiers (e.g., completely, 13.

(24) absolutely, almost). The former type is usually unbounded and reflects the range of a scale, while the latter one is bounded and inherent with precise value. As for the adjectives, Paradis (1997) divides them into those of non-degree (this discussion will not go into further detail) and degree structure. Next, under degree structure, gradable adjectives fall under two subsets: complementarity and contrariety. According to the operation of boundedness, these adjectives will correspond to different degree modifiers in most cases. However, as Paradis (1997) claims, contextual modulation might cause the matching between the degree intensifiers and the adjectives to be more flexible.. 立. 政治大. Nondegree. Degree. y. al. │. Degree modifiers. [none]. C hBOUNDED engchi. sit. (NONSCALE). n financial. contrariety (SCALE). er. io Boundedness. ‧. complementarity. Nat. Oppositeness. 學. Gradability. ‧ 國. Configurational meaning structures. iv n UNBOUNDED U. BOUNDED. dead. narrow. excellent. BOUNDED. UNBOUNDED. BOUNDED. completely. very. absolutely. FIGURE 1 The non-degree/ degree dichotomy and the interaction between scale and boundedness (Paradis 2008: 323). 14.

(25) In order to scrutinize the operation of this configuration of meaning structures, three pairs of adjectives, death-alive, terrific-disgusting and narrow-wide are adopted to illustrate. (i) The pair of dead-alive belongs to non-scale adjectives which are denoted by an “either-or” feature; that is, such an adjective is always situated at the endpoint and no other possibility can be yielded. It usually combines with a bound degree intensifier such as completely. (ii) The pair of terrific-disgusting is on a bounded scale which selects the bounded intensifiers. Yet, this pair correlates with the subjective-evaluative usage depending on different speakers. In this situation,. 政治大 functions. (iii) The pair of narrow-wide illustrates scalar opposites in that when 立. unbounded degree intensifiers could also modify bounded adjectives for pragmatic. unbounded intensifiers are added, they “move away from one another in opposite. ‧ 國. 學. directions of the scale” (Paradis 2008: 324). Yet, as mentioned previously, Paradis. ‧. (1997, 2001, 2008) finds that the combination might be asymmetric and could yield a. sit. y. Nat. dual interpretation, e.g., absolutely clear and very clear. Once an adjective selects a. io. er. degree intensifier, which may not seem to correspond with its type, that degree intensifier will pressure it to restrict the interpretation (see Figure 2).. n. al. Ch. engchi. i n U. v. FIGURE 2 The bidirectionality of semantic pressure between the degree modifier and the adjective (from Paradis 1999: 62 Figure 3-6). 15.

(26) 2.3 The degree intensifier and subjectivity As discussed above, some basic understandings such as various categorizations or the interplay with modified adjectives have been explored in terms of degree intensifiers. Another significant aspect, which has aroused interest among researchers, is the use of a degree intensifier during speech. The way in which a degree intensifier is selected by a speaker shows the operation of subjectivity. Subjectivity, by definition, refers to “a pragmatic-semantic process whereby meanings become increasingly. 政治大. based on the speaker’s subjective belief state/attitude toward the proposition” (Traugott 1995:31).. 立. In order to explore the correlation between degree intensifiers and subjectivity,. ‧ 國. 學. several relevant works are reviewed below. Athanasiadou (2007) investigates how a. ‧. degree intensifier displays its involved subjectivity at stages. Four degree intensifiers. sit. y. Nat. completely, perfectly, absolutely and totally are discussed and analyzed based on their. io. er. original adjectival forms. Three aspects of the formation of subjectivity in the cline are illustrated by the degree intensifier completely (see Table 4).. n. al. Ch. i n U. v. TABLE 4 Synchronic analysis of the degree intensifier (Athanasiadou, 2007). e n g Examples chi. Stages. Contents. Property ‘completeness’. Objective and spatial configuration. The complete works of Shakespeare are on sale.. subjective. Quantification. Subjective evaluation. I felt a complete fool.. More subjective. Intensification. Involvement of the speaker and the highlight of degree. He felt completely successful. Their engagement came as a completely surprising event.. Highly subjective. Emphasis. Expression on the truth value as subjuncts. A: And we missed that? B: Completely.. Most subjective. 16. Subjectivity.

(27) Athanasiadou (2007) concludes that the mark of subjectivity which is associated with the use of the degree intensifiers can be explored through not only the diachronic approach but also the synchronic analysis demonstrated in her study. One can also observe the interplay between subjectivity and degree intensifiers in studies of the nuances in new usage. Waksler (2012:18) examines over-the-top (OTT) intensification in the use of degree intensifiers which reflect the opinion of the speaker that the action or state of a person, situation, or entity has surpassed the usual limits of the certain action or state in way that is much beyond the usual standard.. 政治大 Several examples are presented as follows. In terms of the syntactic behaviors, she 立. Four intensifiers super, uber, so and totally are re-examined in such specific register.. found that two of the intensifiers so and totally could appear broadly than they usually. ‧ 國. 學. do when functioning as discourse cues to mark subjectivity. Example (4) shows the. ‧. new noun phrase environments in which they can be found.. sit. y. Nat. (4) a. This place is so San Francisco to me.. io. al. n. that.. er. b. It only took about ten minutes, but she totally could have charged us for. Ch. c. Dude. Shrimp is so not vegetarian.. engchi. i n U. v. d. Sean Connery was totally the best James Bond. In terms of semantic aspect, OTT intensifiers could excel the limits set by gradability and boundedness which were described in §2.2. To be more specific, OTT intensifiers can take [-gradable] entities and transform them into [+gradable] ones as in the following example. (5) And that's when I realize something: this dude is SO not my age. The constituent not my age has not been regarded as a gradable property traditionally. However, when the degree intensifier so is placed before the constituent not my age, the expression is mapped onto a gradable continuum. Thus, by saying so, the speaker 17.

(28) thinks that the difference in age is beyond what might be considered acceptable when thinking about the difference in ages between his and the dude’s. A similar situation happens to the notion of boundedness, which refers to whether the endpoint of an entity exists or not. The following example (6) illustrates that when the unbounded adjective funny is modified by a bounded intensifier such as totally, funny, it is placed on a scale/continuum. (6) Everyone there is really funky and totally funny. In addition to the discussion above, OTT intensification can also be used to. 政治大 intensifiers super and uber can expand the intrinsic scale and hold subjectivity on 立 extend pragmatic limits in colloquial discourse. Waksler discovered that the. different positions. Super is usually employed by a speaker to show his or her view. ‧ 國. 學. that certain entity has exceeded expectations in some way. Yet, the use of super in. ‧. example (7) presents a different reading. Here, the pragmatic function excels the. sit. y. Nat. intrinsic scale on carefulness.. io. er. (7) The women who work here are super careful; they measure and check with you on the front end to make sure everything is as you want them.. al. n. v i n C h its pragmatic function Similarly, uber is said to operate in the same way as super. engchi U. The distinction between super and uber is that uber provides additional discourse information about the entity on which the speaker is expressing a view of. disassociation. Example (8) illustrates not only the speaker's perspective on the yogurt bar, which is beyond her expectations of what she considers to be the normal attributes of a yoghurt bar, but also her wish to disassociate herself from being from identified as the kind of person who would normally frequent such a place.. (8) It's still an uber-modern... trendy yogurt bar, so if you don't like that vibe— like me—you won't like the vibe here. 18.

(29) By investigating the OTT intensification, Waksler’s study (2012) thus provides insights into another aspect of the use of degree intensifiers with regard to subjectivity in colloquial discourse. In another crucial study, Beltrama (2015) identifies three sources of scale: lexical, pragmatic precision, and the speaker’s perspective to explore the use of degree intensifiers. First, the lexical scale is directly encoded in a lexical item and brings about the compositional manner. For example, if someone is described as very tall, what it can be inferred is that the person has to exceed a certain height. Second,. 政治大 strictly. The combination of the intensifier and the modified predicates will not be 立 pragmatic precision causes the meaning of the target property to be interpreted. change truth condition. Third, the speaker-oriented scale displays the involvement of. ‧ 國. 學. speaker’s attitude or stance. These sources of scale interact with two important. ‧. parameters of variation in the intensifier: (i) the restriction and (ii) influence on the. sit. y. Nat. meaning of the modified predicate. Furthermore, he argues that there is a tendency. io. er. that these three sources of intensification undergo stage by stage to form subjectivity in the end. The operation of these sources and examples are summarized in Table 5. n. al. below.. Ch. engchi. i n U. v. TABLE 5 Different aspects of the intensifiers (Beltrama 2015: 126-8 Table 5) Intensifier. Targeted scale. Distribution. Semantic impact. Examples. Degree modifiers. Lexical scale. Gradable predicates. Changes truthconditions. Mark is very tall.. Slack regulator. Scale of precision. Expressions amenable to imprecision. Changes extension. The ball is perfectly spherical.. Speakeroriented intensifier. Speaker’s attitude. Any expression. Specifies independent content. Try answering the fucking question.. 19.

(30) 2.4 Cross-linguistic studies of degree intensifiers 2.4.1 Studies of degree intensifiers in English and other languages Early studies on degree intensifiers in English mostly pay attention to the structure of the degree intensifiers and the diachronic language change. Stoffel (1901) investigates the degree intensifiers such as very, quite and so from some literary texts from the Middle Ages. He discovers that the degree intensifiers may have possessed their lexical meanings in the beginning; yet, these meanings weakened over time (e.g., very in 16th century denotes “genuine” and loses this meaning later). Having interests. 政治大 as an aid to observe the collocational behaviors of the degree intensifiers. Similar to 立. in the diachronic approach, Partington (1993) adopts the BROWN and LOB corpora. Stoffel (1901), he proves the shifts in the meaning of degree intensifiers. The. ‧ 國. 學. following set of examples exemplifies how the degree intensifier really has performed. ‧. differently (Partington 1993: 182). Really in (9a) still pertains its lexical meaning ‘in. y. sit. io. (9) a. Really, I could hear her thinking.. n. al. er. add intensity.. Nat. truth’ and could occur in the sentence-initial position. In (9b), really only serves to. i n b. When the horsetail is C really hot, wrap it up… hengchi U. v. In addition, by analyzing the collocational behaviors in the corpora, Partington (1993) is able to find out that the combination of a degree intensifier and its modified adjective may undergo lexicalization and grammaticalization over time. Later, researchers tended to extract authentic usages from either written or spoken data to further examine degree intensifiers. Paradis (1997) investigates twenty-three degree modifiers in the London-Lund Corpus (LLC) and the LOB corpus. The use of degree modifiers occurs more frequently in the spoken corpus (LLC) than in the written one (LOB). Further, the intensifiers occur more than other subtypes in the subset of the degree modifiers in spoken data. Having examined the collocational 20.

(31) behaviors in the corpora, Paradis further shows that the traditional view of gradability might not be able to sufficiently explain the meaning of the degree intensifiers and the post-elements (e.g., the semantic distinction between the sets very nice & *absolutely nice or absolutely impossible & *very impossible). She further argues that multilayers such as boundedness or intonation harmony should also be included when analyzing data. Wittouck (2011) also utilizes corpora (BNC and COCA) to investigate the grammaticalization of three intensifiers very, pretty, and damn. He finds that certain degree intensifiers would constantly co-occur with several adjectives and form a fixed. 政治大 frequently in the COCA than in the BNC and usually collocates with good. In short, 立 expression when comparing the two corpora. For instance, damn occurs more. the corpus-derived approach provides a more thorough view on the view of degree. ‧ 國. 學. intensifiers.. ‧. Some researchers are also interested in the effect of social factors such as age,. sit. y. Nat. gender, different language groups or other registers on the use of degree intensifiers.. io. er. Ito & Tagliamonte (2003) investigates the usages of eleven intensifiers (very, really, so, absolutely, pretty, too, that, right, totally, completely, bloody) in the York English. al. n. v i n Cspeech Corpus, collected from particular across three different U h e ncommunities i h gc. generations. Two major findings were reported. First, in terms of the distributional analysis, the elder group tended to use degree intensifiers during speech more times. The results for the use of the intensifiers very and really were found to be different across the three age groups. The degree intensifiers shift their usages from time to time. Second, based on the fact inferred previously, they claim that such differences correlate with the semantic category of adjectives (dimension, physical property, color, human propensity, age, value, speed, position) due to the width of the collocation. The results show that the intensifier very mostly collocates with the adjectival type of speed while really collocates with the adjectival type of physical 21.

(32) property. Yet, they do not analyze the features of the degree intensifiers used among language groups further. On the other hand, some researchers are interested in how the degree intensifiers are selected. Certain degree intensifiers are preferably adopted for communicative purposes in a particular language community. It is reported that the constant usage of degree intensifiers such as so, totally, or super in teen talks emerges in daily conversation and reflects the change of social meaning (Núñez Pertejo & Palacios Martínez, 2014; Beltrama, 2016; Tagliamonte, 2016). Other researchers using examples from languages such as German, Spanish, and. 政治大 semantic or sociolinguistic approach. For example, Löbner (2012) examines the use 立 Japanese have examined the issue of degree intensifiers mostly by way of a syntax-. of the degree intensifier sehr ‘very’ in German to observe the interplay with the. ‧ 國. 學. adjectives which it modifies. Martínez López (2009) discovers that the degree. ‧. intensifier mazo 'a lot, very good' is frequently adopted among teenagers in Madrid.. sit. y. Nat. Brown and Mayra Cortés-Torres (2013: 18) examine the degree intensifiers bien. io. er. ‘good’ and muy ‘very ’of Puerto Rican Spanish across different age groups and conclude that the use of these two degree intensifiers displays the “multifaceted and. al. n. v i n complex nature of language useC patterns.” In Japanese,UTsujimura (2001) notices that hengchi one degree intensifier totemo ‘very’ interacts with the scalar structure.. 22.

(33) 2.4.2 Studies of degree intensifiers in Mandarin Chinese and Taiwan Hakka It is not easy to provide a clear-cut definition of the category of adverbs in Mandarin Chinese 4 because of its lack of morphological markers. Generally speaking, an adverb can modify a verb, an adjective or an adverb on the left position of the predicate in a clause. As for degree intensifiers, the categorization in Mandarin Chinese differs from the one often used in western language groups. Degree adverbs of Mandarin Chinese are commonly categorized into two major types: (i) absolute. 政治大 ji3 , 很 hen3 , 怪 guai4, 特 te4 ) and (ii) relative degree adverbs: denoting degree by 立 degree adverbs: denoting degree generally without a comparative meaning (e.g., 幾. comparing (e.g., 一般 yi4ban1 , 最 zui4 , 更 geng4 ). Many researchers follow. ‧ 國. 學. such categorization to analyze the data (Zhang, 2000). The degree intensifier, which is. sit. y. Nat. adverbs.. ‧. the main focus of the present study, appears to fall under the type of absolute degree. io. er. The studies of intensifiers in Mandarin Chinese focus upon several particular intensifiers. The intensifier 很 hen3 has been frequently investigated in terms of its. al. n. v i n C h and meaning in context active changes in syntactic behaviors as shown by many studies engchi U. (Lee, 2011; Chang, 2004; Chen, 2014). For example, Chang (2004) propose the degreeevaluative construction [Degree ADV + X] to investigate the intensifier 很 hen3. They believe that examining the holistic construction could explicate the representation of degree intensifier. They also adopt the “qualia structure” to interpret the X slot within ‧‧‧ the construction to illustrate its range of modification (e.g. 很陳水扁 hen3 Chen shuibian ‘very Shui-bian Chen like). Other intensifiers such as 非常 fei1chang2, 蠻 man2,. 4. Adverbs in Mandarin Chinese can fall into two major categories: descriptive adverbs and functional adverbs. The former are lexical, open-class while the latter carry grammatical meanings such as degree, scope, time, frequency, logical relations, textual relations and the speaker’s stance (Biq & Huang, 2016). 23.

(34) and 太 tai4 are also the commonly analyzed in Mandarin Chinese (Shang 2010, Hong & Lim 2012, Chang 2013). In Shang’s study (2010), he further points out that the intensifier 太 tai correlates with the speaker’s subjectivity. When the intensifier 太 tai is adopted exceedingly, the meaning of the entire modification can turn a positive meaning into a negative one. Since the use of intensifiers reflects the attitudes held by a speaker in a discourse, researchers are interested in whether the intensifiers are sensitive to genre difference. Hong and Lim (2012) discovered that intensifiers are preferred differently in different genres. For example, the intensifier hen is found to be. 政治大 duo1(me) occur in press editorial / reviews and romance / general fictions separately. 立. used overwhelmingly in the spoken register. Others such as 十分 shi2fen1 and 多麼. As for Taiwan Hakka, the studies delving into degree intensifiers are still at the. ‧ 國. 學. initial stage. Many researchers have utilized the categorization of the degree adverbs. ‧. in Mandarin Chinese as a foundation to discuss the linguistic behaviors in Taiwan. sit. y. Nat. Hakka. Peng (2006) broadly investigates the use of degree adverbs in Taiwan Hakka. io. er. and compared them with those in Mandarin Chinese. She does not further explore the differences in each subtype. Qi’s study (2007) gives a general picture of degree. al. n. v i n C hdegree adverbs including adverbs in Taiwan Hakka. Eighteen intensifiers and engchi U diminishers are under scrutiny in terms of their syntactic behaviors and the. correlations with mental verbs, modals, and modified adjectives. Li (2008) focuses on Taiwan Hakka degree adverbs and provides an explanation on the quality, scope and classification of the Hakka degree adverbs. Yet, an analysis of Taiwan Hakka needs to be done to give a better understanding of the features of this language.. 24.

(35) 2.5 Remarks In this chapter, a number of studies have demonstrated and epitomized many aspects of the degree intensifiers. In retrospect, some researchers are interested in how a degree intensifier develops from the lexical category toward the grammatical one. Both a diachronic approach and also a synchronic approach have been utilized to explore this issue. Several significant notions such as gradability, boundedness, and subjectivity have been introduced and further explained when dealing with degree intensifiers. The interplay between degree intensifiers and their modified elements is also important in. 政治大. the analysis (cf. Kennedy & McNally 1999, Paradis 1997, Athanasiadou 2007, Waksler 2012).. 立. However, although there has been much discussion of degree intensifiers in cross-. ‧ 國. 學. linguistic research, little attention has been paid to gathering together a set of degree. ‧. intensifiers to see how they act during speech. Particularly, in spoken Taiwan Hakka a. sit. y. Nat. set of degree intensifiers is frequently employed by speakers and most of the degree. io. er. intensifiers in spoken discourse are monosyllabic words. The ways by which each degree intensifier combines with its modified elements to give rise to differences in. al. n. v i n C hof attention. Whether structures and meanings is worthy the use of degree intensifiers engchi U. correlates with above-mentioned parameters such as gradability, boundedness, and subjectivity is another crucial issue to explore. For example, Paradis (1997) argues that a bounded degree intensifier can only match bounded adjectives. Yet, in spoken Taiwan Hakka, a seemingly bounded adjective 慶 kiang3 ‘capable’ can be freely modified by ‧‧ the six unbounded degree intensifiers as in “F2:厥爸讀書盡慶” (F2: His father is really ‧‧ good at studying.) and “F1:還該玉蘭姑斯當慶了.” (F1: And Aunt Yulan is really capable in doing things.) In addition, the role that degree intensifiers play in a spoken corpus has not been fully investigated. Authentic usage is not confined to just propositional meaning, and the speaker’s construal of a state of affairs along with the 25.

(36) related intentions and emotions should also be treated as of importance. Though several related studies have taken the concepts of subjectivity and intersubjectivity into consideration, most of the studies have discussed the phenomenon of the degree intensifiers merely through self-generated data rather than authentic spoken data. Hence, in order to provide a thorough picture of the use of degree intensifiers in spoken Taiwan Hakka, this study regards six degree intensifiers as a family and attempts to provide a systematic explanation as to how these degree intensifiers play their roles in daily casual conversations.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 26. i n U. v.

(37) Chapter Ⅲ Data and methods 3.1 Data collection 3.1.1 Materials The data was collected from The NCCU Corpus of Spoken Hakka (國立政治大學客語口語語料庫). The NCCU Corpus of Spoken Hakka is one of the sub-corpora under The NCCU Spoken Corpus, documented since 2007 (Chui & Lai, 2008). This open-access Taiwan Hakka corpus consists of two major genres—narratives and daily conversations, which are aligned with audio and visual supplementary. In the present. 政治大 recordings of daily conversation 立 were extracted before August, 2016. Further. study, only daily casual conversations will be treated and analyzed. Thirty-two. ‧ 國. 學. information of the recordings of the daily conversations is given below. TABLE 6 Details of thirty-two recordings of daily conversations. 10. Released: 8. sit. y. Status of publicity. Unreleased: 2. n. al. er. io 海陸 hoi2liug8. ‧. Nat. 四縣 xi3ien3. Num.. 15. Ch. i n U. v. Released: 5. engchi. Unreleased: 10. 大埔 tai3bu1. 1. Released: 1. 饒平 ngiau5pin5. 6. Unreleased: 6. 3.1.2 Procedures First, conversational data was obtained from The NCCU Corpus of Spoken Hakka (http://spokenchinesecorpus.nccu.edu.tw/index.php) including yet-to-be released data in the back-end system of the corpus. All of the files were encoded into UTF-8 in Notepad ++ in order to sort out the data conveniently as shown in Figure 3. 27.

(38) Meanwhile, these files in UTF-8 format will also be parsed in the developmental halfautomatic segmentation system for Taiwan Hakka (Lai, 2016).. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. FIGURE 3 Data collection. Second, the six degree intensifiers 當 dong1, 蓋 goi3, 盡 qin3, 恁 an2, 異 i3, and 還 han5 in Taiwan Hakka were extracted using the software TaiwanDH (Liu, Jin, Liu, Chiu, & Yu, 2011) as shown in Figure 4. The output of the concordance lines will then present in an Excel file. 28.

(39) 政治大 FIGURE 4 Extraction of the concordance lines 立. ‧ 國. 學. Third, in total, 3755 tokens were identified and remain untreated. The tokens were trimmed out according to the following criteria. First, a delayed utterance with. ‧. repetition was counted as once occurrence. In the following example (10), threw. y. Nat. n. al. er. io. (10) M2: ..該 M2:. ge3. M2: that. 湖鰍仔. Ch. so. i n U. v. e n g c h i ‧‧ 趣味... ‧‧‧ 蓋蓋蓋. fu5qiu1e2 goi3goi3goi3 loach. sit. occurrences of goi3 are counted as one token.. so. so. qi3mi3 interesting. ‘M2:..That loach is so so so interesting.’. Second, identified tokens that are not used as a degree intensifier were excluded. As shown in example (11), 恁樣 an2ngiong5 ‘such, kind of, as’ is lexicalized and does not serve to intensify any following elements; hence, it was deleted from the data. Other similar expressions such as 恁形 an2hin5, 恁呢 an2ne5, 恁樣形 29.

(40) an2ngiong5hin5, and 恁仔 an2e2 were not included in the data. (11) M: 無. 就. 搵. M: mo5. qiu3 vun3. [該. 橘醬. ge3 gid4jiong3. 啊].... [[就. ‧‧ 恁樣. 啊]]. a1. qiu3. an2ngiong5. a1. like that. PART. M: or else then dip in that tangerine jam PART then. ‘M: Or dip (it) in into that orange sauce. We could only do such a thing.’. Third, due to the technical function of the extraction software, all of the. 政治大 Some of the tokens were irrelevant to this present study and were excluded. For 立. concordance lines that contained any of the six degree intensifiers were identified.. 學. a degree intensifier. Thus, it is deleted. (12) F1: ...加還. 愛. ‧. ‧ 國. instance, in the example (12), 便當 pien3dong1 ‘lunchbox’ has nothing to do with. 送. ‧‧ 便當. 該下. sung3. pien3dong1. ge3ha3 Cong5on1 iu3. 人. 送.... mo5. ngin5. sung3. lunchbox then Chang-an again NEG person adeliver v i one will do it for Chang-an.’ l n ‘F1: By the time when we haveCto deliver the lunchbox, no hengchi U. deliver. han5 oi3. F1:. also. y er. need to. n. and. io. F1: ga1. 又. sit. Nat. 無. ...長安. 2289 tokens which were not qualified within the requirements for degree intensifiers were excluded and 1466 tokens remained for further analysis. Finally, the main data were coded according to the different criteria. To guarantee consistency, all of the data were coded by the author and to a second coder, Yun-chien, Wang who is a native Hakka speaker in a first round of coding. Then the two coders exchanged their respective parts for a second round of coding and discussed any inconsistencies.. 30.

(41) 3.2 Coding scheme To analyze the data appropriately, the current study encodes and examines the data in four aspects. For the purpose of exploring the internal structure of [DI. + X], the syntactic structure of X, and the syntactic subcategories of X are coded in Sections 3.2.1, and 3.2.2. Section 3.2.3 will encode five grammatical functions of [DI. + X] construction.. 3.2.1 The syntactic structure of X. 政治大 [DI. + X]. X could be an adjectival phrase (AP), verbal phrase (VP), adverbial phrase 立 The syntactic structure with the constructional fashion will be assigned as. (ADVP) and or a nominal phrase (NP). Note that X does not have to be a fully-. ‧ 國. 學. fledged phrasal structure. Simple forms such as nouns, adjectives, verbs, and adverbs. ‧. are allowed. The variable X will be encoded into a constructional fashion as [DI. +. sit. y. Nat. NP], [DI. + AP], or [DI. + VP]. The representative examples are illustrated below: in. io. F3：. 頭擺. al. n. (13) [DI. + NP]. er. (13), X as NP; in (14), X as AP; in (15), X as VP. [HL70:11]. Ch. 啊嗬... 斯戴. ‧ 當. ‧‧ 鄉下. engchi. i n U. v. F3： teu5bai2 a3ho2…sii5 dai3 dong1 hiong1ha3 past. PART SII live very rural. ‘F3: When it comes to my childhood, I lived in very rural area in the past.’. (14) [DI. + AP]. [HL62: 38] ‧ 當. ‧ 老. F2：. 松樹. F2：. qiung5su3 dong1 lo5 la2. F2：. pine tree. very. 啦.. old PART. 31.

(42) ‘F2:. The pine tree is very old.’. (15) [DI. + VP] F2: .... [HL12: 28]. 幾隻字. 還. 聽得識. 但係. ...無.... ‧ 當. F2:. gi2zag2sii3 han5 tang1ded4siid4 tan3he3 …mo5… dong1. F2:. many-CL word. still. hear-DED-know. but. NEG. very. ‧‧ 了解. 啦. liau2gie2. la2. understand PART. ‘F2: I could recognize some words but I do not quite understand.’. 政治大 After the NP, AP, and VP of X were differentiated, X was further subcategorized. 立. 3.2.2 Syntactic subcategories of X. ‧ 國. 學. Considering what can be modified by a degree intensifier as previous studies have shown, not all of the subcategories were selected for inclusion in the encoding. ‧. process. Figure 5 illustrates the subcategories which will be included in the operating. sit. y. Nat. principle. Each subcategory will be provided with a proper example from the. io. n. al. er. Taiwanese Hakka Dictionary of Common Words.. Ch. engchi. 32. i n U. v.

(43) Encoded syntactic categories in Taiwan Hakka. VPs. APs. NPs. Relative adjectives. Common nouns. 政治大. Locative nouns. Stative verbs. Optative verbs. Absolute adjectives. 立. ‧ 國. 學. FIGURE 5 Encoding structure of syntactic categories and subcategories. ‧. 3.2.2.1 NPs. y. Nat. io. sit. The head in a nominal phrase is the noun. Nouns are lexical items which can be. n. al. er. animated, unanimated, real or imagined. The noun category can be broadly divided. Ch. i n U. v. into two subcategories broadly. Common nouns are used for general sense in the. engchi. present study. They usually refer to concrete entities such as 桂花 gui3fa1 ‘osmanthus’ shown in (16) and abstract entities such as 愛情 oi3qin5 ‘love’ shown in (17). Or they refer to unique entities such as 日本 ngid4bun2 ‘Japan’ and 臺灣 toi5van5 ‘Taiwan’ shown in (18). Or else they can refer to a point of time such as 朝晨頭 zeu1siin5teu5 ‘morning’ shown in (19). Next, locatives usually refer to locations such as 學校 hog8gau2 ‘school’ in (20). Some localizers which indicate the direction and location will merge with locative in this study. For example, 後背 heu3boi3 ‘back’ is used as a locative in example (21).. 33.

(44) (16) 秋天. 係. qiu1tien1 he3 autumn. ‧‧ 桂花. 盛開. gui3fa1. siin3koi1 ge3. 个. COP osmanthus bloom. 時節 sii5jied4. NOM period. ‘Autumn is the time when the osmanthus is in full bloom.’. (17) 感情. 有. gam2qin2. 當多種. ‧‧ 愛情. 除忒. 還有. 政治大 many forms except love kinship 立. 親情. iu1 dong1do1zung2 cu2ted4 oi3qin5 han2iu1 qin1qin2. relationship has. and. 友情 iu1qin2 friendship. cai3. Japan. locate. 个. 北方. toi5van5. ge3. bed4fong1. NOM. northern. y. io. al. n. Taiwan. Ch. sit. ngid4bun. ‧‧ 臺灣. er. ‧ 國. 在. Nat. ‧‧ 日本. ‧. (18). 學. ‘Relationships can have many forms apart from those of love, kinship, and friendship.’. engchi. ‘Japan is located to the north of Taiwan.’. i n U. (19) ‧‧‧ 朝晨頭. 个. zeu1siin5teu5. ge3. morning. NOM. 日頭. 盡. 熝. ngid4teu5 qin3 lug8 sun. very heat. 人 ngin5 people. ‘The heat of the morning sun bothers people very much.’. 34. v.

(45) (20) ‧‧ 學校. 係. 培育. 人才. 个. 所在. hog8gau2. he3. pi5iug4. ngin5coi5. ge3. so2cai3. school. COP cultivate the talented NOM. place. ‘School is the place to cultivate the talented.’. (21) 該扇. 門. ge3san3 mun5 door. ge3 NOM. 有一幅山水畫政治大 heu3boi3 iu1 id4bug4 san1sui2fa3 立 back. has. one-CL. landscape painting. 學. ‧ 國. that-CL. ‧‧ 後背. 个. ‘There is a landscape painting on the back of that door.’. ‧ sit. y. Nat. 3.2.2.2 VPs. io. er. Verbs occupying the head position in a verbal phrase usually function as predicates in a clause. The defining properties of verbs are as below:. n. al. a. Verbs refer to an action. b. Verbs can take objects.. Ch. engchi. i n U. v. c. Verbs are used mainly as a predicate. Some of them can be used as complements. Verb phrases can appear in the subject, object, or modifier position. d. Some type of verbs such as psych verbs or epistemic verbs are gradable and can be modified by degree adverbials. e. Verbs can be suffixed with one of three aspect markers: the perfective phase aspect marker 忒 ted4, 煞 sod8, 了 le1, 著 do2; the imperfective progressive/ durative marker 等 den2, 著 do2; or the experiential marker 過 go3. 35.