中文主題串英譯之研究 - 政大學術集成

全文

(1) . 國立政治大學語言學研究所碩士論文 National Chengchi University Graduate Institute of Linguistics Master Thesis. 指導教授：尤雪瑛博士. 政治大. Advisor: Dr. Hsueh-ying Yu. 立. ‧. ‧ 國. 學中文主題串英譯之研究. sit. y. Nat. io. n. al. er. A Study on the English Translation of Chinese Topic Chains. Ch. engchi. i n U. 研究生：康恆銘撰 Student: Heng-ming Kang. 中華民國一零四年一月 January, 2015. v.

(2) . A Study on the English Translation of Chinese Topic Chains. By. 立. 政治大 Heng-ming Kang. ‧. ‧ 國. 學 er. io. sit. y. Nat. A Thesis Submitted to the. n. aGraduate iv l C Institute of Linguistics n h eFulfillment in Partial the n g c h iof U Requirements for the Degree of Master of Arts. January 2015.

(3) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(4) . 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. Copyright © 2015 Heng-ming Kang All Rights Reserved. iv. i n U. v.

(5) To my mother 獻給我的母親. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. v . i n U. v.

(6) . 謝詞. 辛苦了這麼久，總算把這本論文生出來了。能夠有今天的成就，要感謝的人很多。首先，要感謝我的語言學啟蒙老師，李櫻老師。相當年在師大求學時，還對自己的英文文法觀念相當自負。一直到大三修習李櫻老師的「英語功能語法」後，對英語語法，甚至是對「語言」的理解，整個改觀。李櫻老師帶領我們分析坊間參考書的文法，鼓勵我們思考文法背後的道理。感謝李櫻老師讓我看見自身對英語了解的不足，將我對語法的理解從句子帶到篇章層次，讓我對功能語言學產生興趣，因而進入政大語言所求知。接著要感謝我的指導教授尤雪瑛老師。進入政大之前，其實我早就打定主意要跟著尤老師做研究，因為尤老師是台灣少數研究篇章語法的學者。進入研究所後，我一路跟著尤老師上課，到最後做論文研究。尤老師除了強調理論外，更重視應用，讓我將學到的知識化為可利用的工具。從一開始思考論文題目，到討論研究方法，感謝尤老師總是能夠一針見血，突破我的盲點，給予我不同的思考面向。除了研究之外，尤老師對寫作的高標準也讓我重新省視自己的寫作風格，並且更加精進。另外，感謝我的論文口試委員：李櫻老師及賴惠玲老師。兩位老師願意抽空閱讀我的論文，並給予我寶貴的建議，真是我的一大榮幸！在政大語言所的求學過程中，相當豐富精彩。感謝所上每位老師帶我認識語言的各個面向：蕭宇超老師、何萬順老師、黃瓊之老師、徐嘉慧老師、張郇慧老師、萬依萍老師、莫建清老師。另外感謝所上惠玲助教總是在我慌張無措時給我適時的幫助。感謝一起在季陶樓奮鬥過的研究所同學，每天五點下課後可以一起去吃晚餐繼續討論語言學，真是人生難得的經驗。感謝客家庄的詩敏和秋杏學姐，在我感到困惑時總能給我一些建議。另外要感謝在我寫論文時一路陪伴我的 K。寫論文時壓力很大，你總是能夠體諒，給我許多鼓勵。有時候怠惰下來，你也會不斷鞭策我。儘管學科不同，你也很樂意幫我整理語料和檔案，非常感謝你的付出。最後，感謝我的父親康振義與母親李滿菊。感謝你們的養育與照顧，雖然你們對我所學只有一點模糊的概念，但仍然默默的支持。未能讓母親來得及看見我拿到畢業證書，是我一生的遺憾。這本論文完成，希望能慰藉母親在天之靈。. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. vi. i n U. v.

(7) 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：中文主題串英譯之研究指導教授：尤雪瑛博士研究生：康恆銘論文提要內容：（共一冊，23,943 字，分五章）. 政治大. 中文可說是個篇章導向的語言(discourse-oriented language)，以篇章為其基本. 立. 單位，數個中文子句不藉連接詞即可連結成主題串(topic chain)。當如此龐大的篇. ‧ 國. 學. 章單位翻譯成主語顯著的英文時，譯者會面臨的困難是，如何將主題串切割成數. ‧. 個英文句子。然而，此議題卻鮮少前人研究。因此，本論文試圖探討中文主題串. y. Nat. al. er. io. sit. 英譯時的翻譯策略，並著重於斷句的影響因素。分析的語料來自漢英對照版的台. n. 灣光華雜誌以及翻譯教科書。為了瞭解翻譯策略，本研究分析中文的語意分段標. Ch. engchi. i n U. v. 記與資訊順序，並與譯文比較。分析結果顯示譯者在翻譯時會採用三種策略：反映段落標記(Reflecting the Markers)、建立語句關係(Establishing Textuality)、重整資訊(Rearranging Information)。第一個策略是將中文段落標記作為英譯斷句的依據。段落標記包括三類：主題的形式、連接詞、標點符號。第二個策略，建立語句關係，表示英譯斷句依據中文主題串各句子之間的篇章關係。第三個策略，重整資訊，表示透過增加、刪除、調換順序等方式調整原文的資訊。關鍵詞：翻譯、中譯英、翻譯策略、篇章分析、主題串、篇章標記、篇章關係. vii.

(8) . Abstract Chinese is considered a discourse-oriented language. The basic unit of the Chinese language is discourse-based. Several Chinese clauses can be linked together without any connectives to form a topic chain. When such a large discourse is translated into English, a subject-prominent language, translators may have difficulty deciding how to segment a Chinese topic chain into English sentences. However, little research has been done on this topic. The present study aims to explore translation. 政治大. strategies used in translating Chinese topic chains into English. In particular, the. 立. demarcation mechanism will be the focus. Chinese-to-English translation data from. ‧ 國. 學. Taiwan Panorama, a Chinese-English bilingual magazine, and from translation. ‧. textbooks are collected for analysis. The demarcation markers and information flow in. Nat. sit. y. Chinese are analyzed and compared to understand how they are treated in the English. n. al. er. io. translation. Three strategies have been found: Reflecting the Markers, Establishing. Ch. i n U. v. Textuality, and Rearranging Information. Reflecting the Markers is to reflect the. engchi. Chinese boundary markers as English demarcation points. Boundary markers contain nominal references of topic, connectives, and punctuation marks. Establishing Textuality is to organize the Chinese topic chain based on the internal textual relationships. Rearranging Information is to add, delete, or reorder the information.. Keywords: translation, C-E translation, translation strategy, discourse analysis, topic chain, discourse marker, textual relationship. viii.

(9) Table of Contents Acknowledge……………………………….……….………………………………..vi Chinese Abstract…………………………….……………………………………...vii English Abstract……………………………..……………………………………..viii Table of Contents……………………………...………………………….………….ix List of Tables……………………………………...………………………………….xi Chapter One Introduction........................................................................................... 1 1.1 Background and Motivation .......................................................................... 1 1.2 Purpose of the Study...................................................................................... 4 1.3 Organization .................................................................................................. 5 Chapter Two Literature Review ................................................................................. 6 2.1 Subject and Topic .......................................................................................... 6 2.2 Topic Chain ................................................................................................. 12 2.3 Translation Studies ...................................................................................... 20 2.3.1 Translation strategies ................................................................... 21 2.3.2 Contrastive analysis ..................................................................... 23 2.3.3 Translation studies of Chinese topic and topic chain ................... 25 2.4 Discourse Segmentation .............................................................................. 28 2.4.1 Discourse markers ........................................................................ 29 2.4.2 Referent forms ............................................................................. 32 2.4.3 Chinese punctuation system ......................................................... 33 2.5 Summary ..................................................................................................... 35. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. i n U. v. Chapter Three Methodology ..................................................................................... 36 3.1 Data Collection ............................................................................................ 36 3.2 Definition of Topic Chain ............................................................................ 38 3.3 Preliminary Analysis ................................................................................... 39 3.4 Analysis Procedure ...................................................................................... 42. Ch. engchi. Chapter Four Results and Discussion ...................................................................... 50 4.1 General Description of the Data .................................................................. 50 4.1.1 Types of topic chains ................................................................... 50 4.1.2 Numbers of units in topic chains ................................................. 58 4.2 Translation Strategies of Topic Chains ........................................................ 60 4.2.1 Overall frequencies of the three strategies ................................... 61 4.2.2 The first strategy: Reflecting the Markers ................................... 64 4.2.3 The second strategy: Establishing Textuality............................... 76 4.2.4 The third strategy: Rearranging Information ............................... 84 4.3 Summary ..................................................................................................... 93 ix.

(10) . Chapter Five Conclusion ........................................................................................... 95 5.1 Summary of the Present Study .................................................................... 95 5.2 Pedagogical Implications ............................................................................ 96 5.3 Limitations and Suggestions for Future Research ....................................... 99 References ................................................................................................................. 100 Appendix I ................................................................................................................ 107. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. x. i n U. v.

(11) List of Tables Table 1. Strategies Discussed in Translation Textbooks ...................................... 21. Table 2. Numbers and Percentages of Each Topic Chain Type ........................... 56. Table 3. Numbers of Units in the Data ................................................................. 58. Table 4. Overall Frequencies of Translation Strategies in Topic Chains ............. 63. Table 5. Frequencies of Markers at the Beginning of a Topic Chain................... 68. Table 6. Frequencies of Markers of in the Middle of Topic Chains .................... 68. Table 7. Reflected and Non-Reflected Full NPs and Pronouns in the Middle of. 政治大 Reflected and Non-Reflected 立 Connectives in the Middle of Topic Chains ..... 72. Topic Chains........................................................................................... 71. Distribution of Reflected and Non-Reflected Punctuation Marks ......... 73. 學. Table 9. ‧ 國. Table 8. Table 10 Types of Topic Chains Showing Use of Establishing Textuality .......... 78. ‧. Table 11 Numbers of Each Textual Relationship Found in Establishing Textuality ...... 83. n. al. er. io. sit. y. Nat. Table 12 Numbers of Topic Chains Showing Use of Rearranging Information ... 92. Ch. engchi. xi. i n U. v.

(12) 1. Chapter One Introduction 1.1 Background and Motivation With the development of technology, international communication has become easier than ever. The needs for translation and interpretation have been increasing, either for the purpose of business, travel, politics, or academy.. 政治大. Translation studies is a new academic discipline, formally established after the. 立. Third International Congress of Applied Linguistics in Copenhagen in 1972 (Liao,. ‧ 國. 學. 2011), but it has received great attention since then. In Taiwan, since 1988, eight. ‧. graduate institutes of translation and interpretation studies, two undergraduate. y. Nat. al. er. io. sit. departments, and much more non-degree programs have been established. Translation. n. as a profession is also emphasized in the field of teaching English as a second/foreign. Ch. engchi. i n U. v. language. Some scholars include translation into the basic skills for English and propose five skills—listening, speaking, reading, writing and translation (Naimushin, 2002; Newmark, 1991). The largest-scale English proficiency test in Taiwan, the General English Proficiency Test, even includes Chinese-to-English passage translation into the writing test at Intermediate and at High-Intermediate level. Despite the emphasis received in translation education and test, the studies on teaching translation and translation theory are still insufficient (Liao, 2011). Most .

(13) 2 . translation teachers could not provide a specific guideline for students to follow. When it comes to the standard of good translation, most translation teachers would probably still refer to the three-word maxim: xin (信) ‘faithfulness,’ da (達) ‘expressiveness,’ and ya (雅) ‘elegance.’ The three words, first proposed by Yan Fu in his preface of the translation of Thomas Henry Huxley’s Evolution and Ethics in 1898, are still regarded as golden rules today. According to Shen’s (2000) interpretation, xin. 政治大. (信) means the translation is faithful to the source text, not twisting the original. 立. meaning. Da (達) indicates that the translation is fluent and coherent. Ya (雅) requires. ‧ 國. 學. the language in translation to be artistic or literary. However, the three words are still. ‧. too abstract for students to follow. More specific guidelines are needed for students to. Nat. io. sit. y. follow during the process of translation. The lack of specific teaching guidelines is. al. er. probably due to the lack of research focusing on translation strategies. Most master. n. v i n C hof translation and interpretation and doctoral theses from institutes so far have focused engchi U on aspects of translation criticism, influence of culture in translation works, translation techniques (e.g. note-taking skills, making use of translation tools), or the role of the translator in translation. Their focuses are more of the translator’s internal reflection. Very few of them draw linguistic findings to study the translation strategies of a certain linguistic structure, such as cohesive devices or discourse structure. In fact, linguistic findings of contrastive rhetoric can provide some insight into. .

(14) 3. translation studies. Linguists have demonstrated that Chinese and English are quite different languages in terms of syntactic and discourse structure. Language typologists Li & Thompson (1976) categorize English as a subject-prominent language while Chinese as a topic-prominent language. Subject is a syntactic notion while topic is a discourse notion (Li & Thompson, 1976; Tsao 1979, 1990; Chu, 1998). A topic in Chinese can link several clauses1 together, with or without other. 政治大. conjunctions, to form a “topic chain,” which is a structure that does not exist in. 立. English. An example of Chinese topic chain is given in (1): 句子 i. 真. 難，. Zhe-ge Yingwen juzii zhen nan this-CLASS English sentence really difficult (b) 我不懂 0i，. sit er. io. al. y. Nat. wo bu dong 0i I not understand (c) 他也不懂 0i。. ‧. ‧ 國. 英文. 學. (1) (a) 這個. n. ta ye bu dong 0i he also not understand. ‘This English sentence, (it) is really difficult. I don’t understand (it). He doesn’t understand (it), either.’ (Tsao, 1979, p. 222). Ch. engchi. i n U. v. In this example, the topic is zhe-ge Yingwen juzi (這個英文句子) ‘this-CLASS English sentence.’ The three clauses (a), (b), and (c) are all about this topic. The topic is established in (a), and it is continued in a zero form in (b) and (c). Conjunctions are                                                         1. In several studies, the term sentence and clause have been used interchangeably. Li & Thompson (1976) and Tsao (1979) do not distinguish between the two terms. Chu (1998) proposes another term link to describe the clause/sentence in a topic chain but the three terms are still used interchangeably. He also proposes another term SENTENCE (all capital) to describe a Chinese sentence ending with a full stop, like a topic chain. Li (2005) adopts the term clause to describe a structure which is part of a larger sentence. In the present study, the term clause is adopted for a predicate structure and the term sentence is used for a string of words ending with a full stop.. .

(15) 4 . not required between the three clauses. Such a syntactic/discoursal difference between Chinese and English poses a challenge to translators. Single-clause Chinese sentences are likely to be directly translated into an English sentence. Two-clause Chinese sentences may be translated into two English clauses linked with a conjunction. However, if the translation unit is a Chinese topic chain with more than two clauses, it is more difficult to translate.. 政治大. When translating such a large unit, a translator would not always use conjunctions to. 立. connect all the clauses together because it is awkward to use more than one. ‧ 國. 學. conjunction continuously to link neighboring English clauses. It is more likely for a. ‧. translator to segment a topic chain into several chunks, reorganize the structure, and. Nat. n. al. er. io. sit. y. then translate the ideas into English.. 1.2 Purpose of the Study. Ch. engchi. i n U. v. The purpose of this study is to explore the translation strategies that English translators would adopt when translating Chinese topic chains, a discourse unit with several clauses linked together, into English, a subject-prominent language. The present study will focus on how a translator segments the Chinese discourse into several English sentences. After reviewing the previous literature on the syntactic and discoursal differences between Chinese and English, we would like to examine the. .

(16) 5. authentic linguistic data and investigate the translation strategies applied to Chinese topic chain translation. It is expected that the findings of this study will benefit researchers, translation teachers and translation learners as well.. 1.3 Organization The study comprises five chapters. The background and the purpose of the study. 政治大. have been introduced in this chapter. This shall pave the way for the forthcoming. 立. Chapter Two: Literature Review. In Chapter Two, studies on the properties of subject. ‧ 國. 學. and topic will be introduced. Examples of both typical and atypical topic chains will. ‧. be provided. Following that are previous studies on topic chain translation. In addition,. Nat. io. sit. y. studies on discourse markers and referent forms involved in translation will be. al. er. discussed. In Chapter Three, the researcher will define topic chain in the present study,. n. v i n C hand describe the data explain the data collection method, e n g c h i U analysis procedure. Chapter Four will present a general description of the data, followed by discussion and examples of the strategies found in the present study. Finally, the last chapter will. summarize the major findings of the present study. Pedagogical implications will also be provided.. .

(17) 6 . Chapter Two Literature Review In this chapter, previous studies concerning Chinese topic chain and its translation will be presented. Section 2.1 will present the distinctions between subject and topic. Section 2.2 will present definitions of topic chain from previous studies. Examples of typical and atypical topic chains will be given. Section 2.3 will present. 治政 previous studies on translation of Chinese topic-comment 大structure and topic chain. 立 ‧. ‧ 國. 學. Last, in section 2.4, previous literature on discourse segmentation will be discussed.. 2.1 Subject and Topic. sit. y. Nat. io. n. al. er. The terms topic and comment are first proposed by Hockett (1958) as a more. i n U. v. semantic notion than a syntactic one. They are used to characterize a predicative. Ch. engchi. construction in English and European languages: “the speaker announces a topic and then says something about it” (p. 201). At that time, topics were considered subjects, while comments were considered predicates. The terms are adopted by Chao (1968) as semantic terms to describe Chinese subject-predicate structure. He does not treat topic differently from subject. The term topic is not differentiated from subject until the 1970s. A new typology of languages is proposed by Li & Thompson (1976). World languages are classified .

(18) 7. based on the two different grammatical relations: subject-predicate and topic-comment. Seven characteristics of topic and subject are proposed (Li & Thompson, 1976, pp. 461-465): (1) a. A topic must be definite. Proper and generic NPs are also understood as definite. A subject, on the other hand, need not be definite. b. A topic need not have a selectional relation with any verb in a sentence; that is, it need not be an argument of a predicative constituent. On the other hand, a subject always has a selectional relation with some predicate in the sentence2. c. A topic is not determined by the verb but by discourse while a subject can be predicted by the verb. d. The functional role of the topic is constant across sentences whereas the functional role of the subject can be defined only within the confines of a sentence. e. A topic does not have obligatory agreement to the predicate; however, a subject has obligatory agreement to the verb. f. A topic is always confined to the sentence-initial position but a subject is not. g. While a subject plays a prominent role in some grammatical processes, such as reflexivization, passivization, imperativization and verb serialization, a topic is not involved in such processes.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. From the characteristics proposed by Li & Thompson (1976) in (1), we can see that topic is recognized as a discourse notion while subject as a syntactic notion. A topic is not determined by the verb and can function across sentences whereas a subject bears selectional relation to the verb and its function is confined in a sentence. Based on the characteristics in (1), world languages are typologized into four: Subject-Prominent.                                                         2. The surface subject of some sentences may not be selectionally related to the main surface verb. For example, John in John is easy to please is analyzed selectionally unrelated to the main predicate by some transformational linguists, according to Li & Thompson (1976).. .

(19) 8 . Languages (e.g. English), Topic-Prominent Languages (e.g. Chinese), Subject-Prominent and Topic-Prominent Languages (e.g. Japanese), and Neither Subject-Prominent nor Topic-Prominent Languages (e.g. Tagalog). The four types, according to Li & Thompson, constitute a continuum. Chinese is more topic-prominent than English, but that does not mean subjects cannot be found in Chinese nor does it mean that topics do not exist in such a Subject-Prominent. 政治大. Language as English.. 立. In fact, topic and subject3 are not mutually exclusive in Chinese (Chu, 1998;. ‧ 國. 學. Tsao, 1979; Li & Thompson, 1981). Chinese sentences can have both a topic and a. ‧. subject as in text (2). Chinese sentences can also have a nominal form serving as topic. Nat. io. sit. y. and subject at the same time as in text (3). In addition, Chinese sentences may have. n. al. er. only a topic without a subject as in (4). What is more, both the topic and the subject. Ch. engchi. can be omitted in a Chinese sentence as in (5). (2) 那隻. 狗. 我已經. 看過. i n U. v. 了。. Na-zhi gou wo yijing kan-guo le that-CLASS dog I already see-ASP PAR ‘That dog I have already seen.’ (Li & Thompson, 1981, p. 88) (3) 我喜歡. 吃蘋果。. Wo xihuan chi pinguo I like eat apple ‘I like to eat apples.’ (Li & Thompson, 1981, p. 88)                                                         3. The properties of Chinese subject and topic are still at debate. We do not want to go into the debate about the properties of the two terms. For a more detailed discussion, please see Tsao, F.-F. (1979). A functional study of topic in Chinese: The first step towards discourse analysis.. .

(20) 9. (4) 那本. 書. 出版了。. Na-ben shu chuban-le that-CLASS book publish-ASP ‘That book, (someone) has published it.’ (Li & Thompson, 1981, p. 88) (5) A: 橘子. 壞了. 嗎?. B:壞了。. A: Juzi huai-le ma B: Huai-le. orange spoil-ASP Q spoil-ASP ‘Are the oranges spoiled? ’ ‘(They) are spoiled’ (Li & Thompson, 1981, p. 90) In example (2), na-zhi gou (那隻狗) ‘that dog,’ which appears at the beginning of the. 政治大. sentence, is the topic. The pronoun wo (我) ‘I’ is the subject because it has selectional. 立. relation with the verb kan (看) ‘see.’ This sentence has both the topic and the subject.. ‧ 國. 學. In example (3), the topic is also the subject. Wo (我) ‘I’ in this example is both the. ‧. topic and the subject. In example (4), the topic is na-ben shu (那本書) ‘that-CLASS. Nat. io. sit. y. book,’ and the subject is omitted from the sentence. In example (5), B’s answer does. n. al. er. not contain the topic/subject juzi (橘子) ‘orange’ because it can be understood from the context.. Ch. engchi. i n U. v. Elaborating on Li & Thompson’s (1976) “topic-prominent language,” Tsao (1979) proposes that Chinese is a “discourse-oriented language.” According to Tsao (1979), subject and topic belong to different levels of grammatical organization. He explicitly states that “[t]opic is a discourse notion” and that “it may, and often does, extend its semantic domain to more than one sentence” (Tsao, 1979, p. 88). “[A] topic can be regarded as a topic only at the discourse level; at the sentence level it may be regarded. .

(21) 10 . as several different things” (Tsao, 1979, p. 92). A topic has various syntactic and semantic relations to the sentences under its domain. Syntactically, a topic can be the subject, the direct object, or the indirect object. (6) (a) 那棵. 樹i. 花j. 小，. Nei-ke shui huaj xiao that-CLASS tree flower small (b) 0i 葉子 k 大， 0i yezik da leaf big (c) 0i 很難看，. 立. 政治大. ‧ 國. 學. 0i hen nankan very ugly (d) 所以 0i 我 l 沒買。. ‧. suoyi 0i wol mei mai so I not buy ‘That tree, its flowers were small. Its leaves were big. It was ugly so I didn’t buy it.’ (Tsao, 1979, pp. 92-93). io. sit. y. Nat. al. er. In (6), the topic neike shu (那棵樹) ‘that tree’ is established in (a). Following the topic,. n. v i n C h In (b), the topicU hua (花) ‘flower’ in (a) is the subject. e n g c h i is in the zero form, followed by the subject yezi (葉子) ‘leaf.’ In (c), the topic and the subject are identical. In (d), the topic is also the direct object of the verb mai (買) ‘buy.’ On the other hand, a topic can take a non-nucleus syntactic position and bear an adverbial relation of either time or place to the verb. In (7), the topic zuotian (昨天) ‘yesterday’ is time while in (8), the topic Beijing-chen li (北京城裡) ‘inside Beijing city’ is a place.. .

(22) 11. (7) 昨天. (啊)，. 張三. 來. 看我。. Zuotian (a), Zhang San lai kan wo Yesterday (PART) Zhang San come see me ‘Yesterday, Zhang San came to see me.’ (Tsao, 1979, p. 104) (8) 北京. 城. 裡. (啊)，有個. 故宮。. Beijing-cheng li (a), you ge Gu-gong Beijing-city inside (PART), exist a Old-Palace ‘Inside Beijing city, (there) exists a palace called Old Palace.’ (Tsao, 1979, p. 104) Semantically, Tsao mentions four relations between the topic and the clause(s) of. 政治大 The possessor and the possessed 立頭腦簡單，四肢這個人(啊)，. its domain:. 發達。. 學. ‧ 國. (9). ‧. Zhe-ge ren (a), tounao jiandan, si-zhi fada. this-CLASS man (PART), mind simple four-limbs well-developed ‘This man, (his) mind is simple; (his) body well-developed.’ (Tsao, 1979, p. 105) (10) Whole and part 三十六計 (啊)，走為上策。. sit. y. Nat. n. al. er. io. Sanshiliu ji (a), zou wei shang ce. thirty-six alternative (PART) running-away is best alternative ‘Among the thirty-six alternatives, running away is the best.’ (Tsao, 1979, p. 105) (11) Class and member 魚 (啊)，鮪魚現在最貴。. Ch. engchi. i n U. v. Yu (a), weiyu xianzai zui gui. fish (PART), tuna now most expensive ‘Fish, tuna is now the most expensive.’ (Tsao, 1979, p. 105) (12) Relevance 這件事 (啊)，我的經驗太多了。 Zhe-jian shi (a), wode jingyan tai duo-le. This-Class matter, my experience too many-Asp ‘With regard to this matter, I have had a great deal of experience.’ (Tsao, 1979, p. 129) The four examples above are known as “double subject” or “double nominative” .

(23) 12 . constructions.4 The former three relations (9), (10), (11) can be paraphrased with a possessive construction. In (9), the topic this man and the following sentences bear a possessor-possessed relationship. In (10), the topic the thirty-six alternatives is whole while the following running is part of the the thirty-six alternatives. In (11), the topic fish is a class and the following tuna is a member of the class. In (12), however, the semantic relation is merely relevance and cannot be paraphrased with a possessive structure.. 立. 政治大. ‧ 國. 學. 2.2 Topic Chain. ‧. The term “topic chain” is introduced to describe a stretch of Chinese discourse.. Nat. io. sit. y. Typical topic chains are defined by Tsao (1979) and later revised by Chu (1998). The. al. er. definitions capture major features of topic chains. However, there are topic chains. n. v i n C h the definition. These with elements not perfectly matching e n g c h i U structures are termed. atypical topic chains in the present study. In this section, we will review the definition of both typical and atypical chains. Tsao defines a topic chain as “a stretch of discourse composed of one, and often more than one, clause, headed by a topic which serves as a common link among all the clauses” (Tsao, 1979, p. vii). A topic is characterized as having a chaining function                                                         4. Following Tsao (1979), we will call it double nominative construction in the present study. For a detailed argumentation, please see Tsao (1979: 132-149). .

(24) 13. and “often extends its semantic domain over several sentences. The sentences under the domain of a topic form a topic chain” (Tsao, 1979, p. 221). Example (6) on page 10 is repeated below as (13) for illustration: (13) (a) 那棵. 樹i 花j. 小，. shui huaj xiao tree flower small. Nei-ke that-CLASS (b) 0i 葉子 k 大， 0i yezik da leaf big. 政治大. 立. (c) 0i 很難看，. ‧ 國. 學. 0i hen nankan very ugly. ‧. (d) 所以 0i 我 l 沒買。. n. al. er. io. sit. y. Nat. suoyi 0i wol mei mai so I not buy ‘That tree, its flowers were small. Its leaves were big. It was ugly so I didn’t buy it.’ (Tsao, 1979, pp. 92-93). i n U. v. In (13), clauses (a)-(d) form a topic chain. According to Li & Thompson (1976) and. Ch. engchi. Tsao (1979), the topic invariably occupies the S-initial position, like neike shu (那棵樹) ‘that tree’ in clause (a). Then, its semantic domain is extended/shared to the following clauses (b) , (c) and (d). The topic does not need to be stated explicitly in those clauses. The clauses are linked by the zero-form topic. However, Chinese topic chains are not always represented by the neat structure. For example, a topic may not always occupy the S-initial position. Below (14) is an example from Tsao (1979) for illustration. .

(25) 14 . (14) (a) 這. 件. 衣裳. 髒了. Zhe-jian yishang this-CLASS clothes (b) 0i 洗了半天. 一塊 i. zang-le yi kuaii dirty-ASP a spot 沒洗掉. 0i xi-le ban-tian mei xi-diao wash-ASP half-day not wash-off ‘This dress had a dirty spot on it. (I) washed it for a long time and didn’t get it off. ’ (Tsao, 1979, p. 210) In example (14), Tsao recognizes clause (b) as a one-clause topic chain. He does not consider (a)-(b) to be a topic chain because yi kuai (一塊) ‘a spot’ does not occupy the. 政治大. S-initial position and is thus not considered a topic. According to Tsao’s (1979). 立. definition, yi kuai (一塊) ‘a spot’ becomes a topic in clause (b) when it occupies the. ‧ 國. 學. S-initial position. Clause (b) forms a one-clause topic chain. However, Chu (1998). ‧. does not think that a single clause forms a topic chain. He treats (a)-(b) in (14) as a. Nat. io. sit. y. topic chain. To explain (14), Chu (1998) proposes another definition of topic chain:. al. er. “A topic chain is a set of clauses linked by a topic in the form of ZA (Zero Anaphora)”. n. v i n C ha topic does not necessarily (p. 338). According to Chu (1998), occupy the S-initial engchi U position. Following his revised definition, clauses (a) and (b) in (14) form a topic chain.. Although Chu’s revision of the definition of topic chain is more flexible and thus accommodates more topic chain structures, there are several topic chains that still do not fit the definition. First, a topic chain with an embedded structure is pointed out by Chu (1998), in which a main topic chain may contain an embedded sub-topic-chain. .

(26) 15. (or sub-chain). Chu (1998, p. 330) cited an example (15) from Shi (1989, p. 232) to explain: (15) (a) 李四這個. 傢伙 i，. Lisi zhe-ge jiahuoi Lisi this-CLASS dude (b) 我 j 因為救他 i， woj yinwei jiu tai I because save him (c) 0j 受了傷， shang 0j shou-le receive-ASP wound (d) 0i 居然不來看我 j，. 立. 政治大 kan wo. ‧ 國. 去了。. 學. 0i juran bu lai j, even not come see me (e) 0i 跑到紐約度假. ‧. 0i pao-dao Niuyue dujia qu-le. run-to New-York have-vacation go-ASP ‘Lisi that dude, I was wounded for saving him. (He) didn’t even come to see me but went to New York for vacation.’. sit. y. Nat. io. n. al. er. In (15), portion (a) is not a clause but just a referent mentioned for later pick-up. It is. i n U. v. picked up in (b) but not as a topic. It does not become a topic until (d) and continues. Ch. engchi. to (e). The relationship between the main chain and the sub chain can be sketched below in (16). (16) (a) Lisi zhe-ge jiahuoi (b) woj……tai (c) 0j…… (d) 0i……woj (e) 0i…… In (16), it can be clearly seen that (a) and (d)-(e) form a main topic chain under the domain of Lisi zhe-ge jiahuo and that clauses (b)-(c), under the domain of wo, form. .

(27) 16 . another topic chain (called a sub-topic-chain), which is embedded in the main topic chain. Based on this example, Chu makes a point that “not all clauses are equal in their status in relation to the chain” (p. 332). In addition to the embedded sub-chain, another less typical topic chain structure, called the telescopic chain is also observed by Chu (1998). In a telescopic chain, “two topic chains merge into each other at the end of one and the beginning of. 政治大. the other” (Chu, 1998, p. 332). According to Chu, the telescopic chain is more. 立. 學. ‧ 國. common than the embedding within a topic chain. Below is an example that Chu cited from Ho (1993, p. 207): 臥室. 裡面. 呢，. ‧. (17) (a) 0i 到了. n. al. Ch. sit er. io. tai jiu juede shoubule, she at-once feel can’t-bear (c) 0i 馬上就現出了. y. Nat. 0i Dao-le woshi limian ne, arrive-ASP bedroom inside PART (b) 她 i 就覺得受不了，. i n U. 原形 j，. engchi. v. jiu xianchu-le yuan-xingj 0i mashang immediately at-once betray-ASP original-form (d) 0j 是一條巨大的白蛇 k 0j shi yi-tiao judade bai shek be a-CLASS huge white shake (e) 0k 盤在床上。 0k pan zai chuang shang coil at bed on ‘Entering the bedroom, she felt she couldn’t take it anymore. In no time (she) was transformed back into her original form, (which) was a huge white snake, (that) coiled on her bed.’ This example has three short topic chains. The first one consists of (a)-(c) under the. .

(28) 17. domain of ta (她) ‘she;’ the second one consists of (c)-(d) under the domain of yuan xing (原形) ‘original form;’ the last one includes (d)-(e) under the domain of judade bai she (巨大的白蛇) ‘huge white snake.’ Clause (c) as well as clause (d) is termed the pivot, the former linking the first and the second chain while the latter linking the second and the third chain. Chu notes that the telescopic chain, though common, presents a problem to analysts because there is no natural break between clauses. This. 政治大. structure is why Chinese sentences are sometimes regarded as liushuiju (流水句). 立. ‘water-flowing sentence,’ meaning it is like non-stop flowing water.. ‧ 國. 學. The third atypical topic chain is topic chain with zero cataphora observed by Li. ‧. (2005). She observes two atypical topic chain structures when analyzing topic chains. Nat. io. sit. y. for the purpose of teaching Chinese as a foreign language. She adopts Chu’s (1998). al. er. definition that a topic chain consists of at least two clauses which are linked by a. n. v i n coreferential relation between anCovert topic and its zero form. However, against h eNP ng chi U Tsao’s (1979) and Chu’s (1998) definition, she observes that a topic chain does not necessarily require an overt topic in the initial clause of a chain. Zero cataphora can occur in the first clause of a topic chain. Take (18) as an example. What is atypical about this topic chain (a)-(c) is that the clauses are linked by zero cataphora in (a) and (b). The zero forms in (a) and (b) cannot find their antecedent in the previous discourse, so they are interpreted cataphorically. The topic is not overtly mentioned. .

(29) 18 . until (c), the third clause in this topic chain. (18) (a) 0i 扎了. 兩針，. 0i Zha-le liang-zhen, get-ASP two-shot (b) 0i 服了劑藥， ji yao, 0i fu-le take-ASP a (dose of) medicine (c) 他 i 清醒過來。 tai qingxing guolai. he wake up ‘After he had an injection of camphor and a dose of medicine, he opened his eyes again.’ (Li, 2005, p. 57). 政治大. 立. The fourth atypical topic chain, the topic chain without any overt topic, is also. ‧ 國. 學. observed by Li (2005). This type of topic chain can be established without any overt. ‧. topic in the chain. Please see the following example.. y. Nat. (19) (a) 他 i 是如此的年輕，. sit. n. al. er. io. Tai shi rucide nianqing, he be so young, (b) 0i 軀幹挺直， 0i qugan tingzhi, stature straight, (c) 0i 唇角永遠. Ch. engchi. 浮著. i n U. 含情的. v. 微笑。. 0i chun-jiao yongyuan fu-zhe hanqingde weixiao. lip-corner forever wear-ASP loving smile.. (d) 0i 每. 星期日自. 講壇上. 下來，. 0i Mei xingqiri zi jiangtan-shang xialai, every Sunday from platform-on come-down, (e) 0i 一定是挾著聖經， 0i yidingshi jia-zhe Shengjing, must tuck-ASP Bible, .

(30) 19. (f) 0i 站. 在琴旁，. 0i zhan zai qin-pang, stand at piano-side, (g) 0i 等著施女士. 出去。. 0i deng-zhe Shi Nushi chuqu. wait-ASP Mrs. Shi go-out. ‘He was a well-built and handsome young man, and forever wearing the blissful smile of one in love. Every Sunday after his sermon, the pastor would tuck his Bible under his arm and step down to the piano to escort Mrs. Shi out of the church.’ (Li, 2005, p. 73) If the punctuation is neglected, this example contains one topic chain (a)-(g).. 政治大. However, the author marks the boundary between (c) and (d) with a full stop. It is. 立. widely acknowledged that a Chinese full stop marks a major discourse break. In. ‧ 國. 學. addition, it is clear that (a)-(c) and (d)-(g) are not continuous in terms of semantics.. ‧. Clauses (a)-(c) are about the pastor’s appearance while clauses (d)-(g) are about what. Nat. io. sit. y. he does every Sunday after the sermon. Clauses (a)-(c) and (d)-(g) each form a topic. al. er. chain with the full stop as a boundary marker. Topic chain (a)-(c) is linked by ta (他). n. v i n ‘he’ in (a). However, topic chainC(d)-(g) linked by zero anaphora, without h e nis merely gchi U any overt topic. This kind of topic chain does not match Tsao’s (1979) and Chu’s (1998) definition that a topic chain requires an overt topic.. In one of the recent studies on the Chinese topic continuation, a less restrictive definition of topic chain is proposed to include both typical and atypical topic chains. A topic chain in Wang’s (2013) study is defined as: a series of clauses led by the same. .

(31) 20 . topic5 (p. 109). According to Wang’s interpretation, the topic does not have to be overt as long as the clauses are under the domain of the same topic. In addition, a topic chain can have more than one covert topic, either in the form of full NP or pronoun. That is, a topic can appear first in a full NP and show up again in the same chain in the form of either a full NP or a pronoun. So far, in this section, we have reviewed the definition of typical topic chains. 政治大. by Tsao (1979) and Chu (1998). We have also reviewed four examples of atypical. 立. topic chains: topic chain with an embedded structure, telescopic chain, topic chain. ‧ 國. 學. with zero cataphora, and topic chain without any overt topic. In order to incorporate. ‧. both typical and atypical topic chains in our analysis, the present study adopts the. Nat. sit. n. al. er. io 2.3 Translation Studies. y. more flexible definition of topic chain by Wang (2013).. Ch. engchi. i n U. v. This section has three parts. Translation strategies discussed in translation textbooks is presented in 2.3.1. Previous studies on Chinese/English contrastive rhetoric will be briefly reviewed in 2.3.2. Translation studies on the topic-comment structure and on Chinese topic chains will be discussed in 2.3.3..                                                         5. . Wang’s original words are 由同一话题引导的系列语句..

(32) 21. 2.3.1 Translation strategies A number of translation textbooks have provided plentiful strategies in dealing with translation from source language (SL) to target language (TL). Based on different goals of the translation, translators can choose from the rigid word-for-word translation to the flexible free translation, which includes approaches of addition, deletion or summarizing6 (Liu, 1993; Zhou, 1996). Adjustment is inevitable in. 政治大. translation. Adjustment strategies commonly discussed in translation textbooks (Liao. 立. et al., 2003; Liu, 1993; Si, 1982; Zhang, 1979; Zhang, Yu, Li & Pong, 1993; Zhou,. ‧ 國. 學. 1996) can be categorized into three levels: semantic level, syntactic level and. ‧. discourse level (Table 1).. n. al. Substitution. Semantic Level. Ch. y. sit er. Description (E.g. SL  TL). io. Strategies. Nat. Table 1 Strategies Discussed in Translation Textbooks. i n U. v. A complicated cultural term such as an idiom or metaphor is replaced by similar ideas in the TL culture.. engchi. E.g. mao ku haozi (貓哭耗子) ‘cat cry-for mouse’  shed crocodile tears. Transliteration. A word of the SL is presented in the sound system in the TL. E.g. ma jiang (麻將) ‘a Chinese game played with tiles’  ma jiang. Syntactic Level Synchronization A TL sentence is translated in the same sequence as that in the SL, sometimes word-by-word. E.g. Zheng-Da zai Taipei. (政大在台北。) ‘NCCU in Taipei’  NCCU is in Taipei..                                                         6. . Summarizing is often applied in newspaper articles or press releases..

(33) 22 . Conversion. The part of speech of a certain word in the SL is converted to another part of speech in the TL. E.g. Zhe jian shi hen chenggong. (這件事很成功。) ‘this CLASS thing very successful’  This is a big success.. Amplification. Words are added in the TL without adding meaning. E.g. Ta dang nian ershi you ba. (他當年二十又八。) ‘He that year twenty and eight’  He was twenty eight years old that year.. Omission. Words are omitted in the TL without omitting meaning. E.g. Touyizao jie-qian, xiayizao jiu tao-fa. (頭一遭借錢，下一遭就討飯。) ‘first borrow-money, next will beg-food’  First you borrow, then you beg.. 學. Negation. 治政大 E.g. Wo xi-le, ye shai-le yifu.(我洗了，也曬了衣服。) ‘I wash-ASP, and dry-ASP立 clothes’  ‘I washed the clothes and dried them.’ Words are repeated in the TL, usually for emphasis.. ‧ 國. Repetition. An affirmative sentence is translated into a negative one, and vice versa.. ‧. E.g. Jixu gongzuo. (繼續工作。) ‘Continue working’  Don’t stop. Nat. sit. Discourse Level. al. er. A sentence in the SL is divided into several sentences in the TL.. io. Cutting. y. working.. n. E.g. Ta saodi tuodi hou cai xiyi. (他掃地拖地後才洗衣。) ‘he sweep mop. Ch. i n U. v. after then laundry’  He swept and mopped first. Then he did the laundry.. Combining. engchi. Several sentences in the SL are combined into one sentence in the TL. E.g. Na tian shi 2014 nian 2 yue 26 ri, wo yongyuan buhui wangji na tian (那天是 2014 年 2 月 26 日。我永遠不會忘記那天。) ‘that day is 2014 year 2 month 26 day. I forever not forget that day’  I will never forget that day, February 26, 2014.. Reordering. The sequence of sentences in the SL is reordered in the TL. E.g. Ta sanyue qu Meiguo。na-shi cai xue Yingyu (他三月去美國。那時才學英語。) ‘He March go USA. That-time start learn English’  He started learning English in March. It was after he arrived in the US.. .

(34) 23. From Table 1, we can see that translators enjoy freedom to a certain degree whether at semantic level, at syntactic level or at discourse level. In addition, most translation strategies discussed in translation textbooks are at syntactic level. Very little is said about discourse level strategies. Moreover, how and when to apply the strategies of cutting, combining and reordering is not discussed in the textbooks.. 2.3.2 Contrastive analysis. 立. 政治大. The contrastive analysis between Chinese and English is the core of translation. ‧ 國. 學. theory (Cheng, 1997). The similarities between languages make translation possible;. ‧. the differences, on the other hand, result in various approaches and techniques to. Nat. io. sit. y. translation. Chinese rhetoric is often regarded as indirect while English rhetoric is. al. er. often viewed as direct. Matalene (1985) analyzes her Chinese students’ English. n. v i n C h between the U writing and reports that “the connections e n g c h i sentences are not explicitly signaled but are almost all left up to the reader” (p.801). This feature is one of the major differences between Chinese and English (Cheng, 1997). Chinese sentences tend to be paratactic; the relation of sentences relies on the internal logic. However, English sentences tend to be hypotactic; the relation of sentences is overtly indicated by grammatical structures. The rhetorical differences are existent in even larger discourse. Five different. .

(35) 24 . culture-specific paragraph development patterns7 are identified by Kaplan (1966). These paragraph development patterns are closely related to culture-specific thought patterns (Kaplan, 1966). Anglo-European expository essays are orderly, clear and proceed in a straight line. Semitic discourse is full of parallel constructions. Oriental essays are indirect and circular. Romance languages show digressions from the central topic. Russian essays are loosely constructed with abrupt changes. These patterns can. 政治大. be illustrated by the following graphs:. Semitic. 立. Oriental. Romance. Russian. 學 ‧. ‧ 國. English. sit. y. Nat. io. n. al. er. (Kaplan, 1966, p. 21). i n U. v. Since the differences between thought patterns are reflected in paragraph. Ch. engchi. structures, difficulties arise in translating one paragraph pattern into another different pattern. As the sketch suggests, Chinese paragraph pattern tends to be circular. The relation between ideas tends to be marked vaguely or indirectly and parallel structures are preferred. On the other hand, English paragraph pattern is more straightforward. The textual relations are marked in a clear way. When translating Chinese paragraphs into English, translators cannot simply translate sentence by sentence accordingly.                                                         7. These patterns are rather sketchy. Discourse patterns in fact are subject to a number of different factors, including genre, formality, theme, purpose, audience, etc (Xu, 2010).. .

(36) 25. Reorganization for appropriate paragraph pattern might be necessary from time to time. This textual difference between Chinese and English thus causes difficulty in translation.. 2.3.3 Translation studies of Chinese topic and topic chain A number of studies have been done on the translation of Chinese topic-comment. 政治大. structure into English (e.g. Jiang, 2012; Wang, 1996; Zhang, 2006). Most of them. 立. focus on the description that Chinese topic is different from English subject and that a. ‧ 國. 學. Chinese topic can/should be translated into different syntactic arguments (e.g. subject,. ‧. object, or indirect object) in English. However, their discussion only focuses on the. Nat. io. sit. y. translation of single-clause sentences instead of topic chains. Since a Chinese topic. n. al. er. chain is widely recognized as a discourse concept, a larger unit of translation must be studied.. Ch. engchi. i n U. v. A few translation textbooks put emphasis on the translation of topic chain structure. They give some general descriptions about the different sentence structures between English and Chinese: English sentences tend to be hierarchically structured while Chinese sentences tend to show parallel structure. The idea is realized in different sets of terms: hypotactic/paratactic (e.g. Cheng, 1997; Huang, 2007), or hierarchical level/mono level (e.g. Yang, 2006). Some scholars (e.g. Cheng, 1997;. .

(37) 26 . Pang, 1997) even provide a metaphor, describing English sentences as grapes with a main stem and branches whereas Chinese sentences as straight bamboos with one section after another. So far, only two studies have been found to discuss translation of topic chain in depth. Lin (2002) studies the preferred structures in Chinese-English translation in her master’s thesis. Four structures are examined: topic chains, the passive constructions,. 政治大. relative clauses and connectives. She has two groups of students do a. 立. Chinese-to-English translation task of four short passages (19 sentences in total). One. ‧ 國. 學. group is composed of 95 eleventh-grade students; the other is composed of 72 college. ‧. freshmen. Their English translation data are examined to see what structures are. Nat. io. sit. y. preferred by Chinese students. Lin’s results show that when students are translating. er. topic chain structures, they tend to omit the subject in the English translation. For. al. n. v i n C h topic chain (20), example, when translating the following e n g c h i U students are often not able to retrieve the deleted topic in (b) and often come up with sentences like “So when young people is finding job, should be careful.” (20) (a) 所以，青少年 i. 在求職. 時，. suoyi, qingshaoniani zai qiu-zhi shi, so, teenagers when seek-job when (b) 0i 應特別小心才是。 0i ying tebie xiaoxin caishi should especially careful best She also finds that, because of topic-deletion in the second clause of a Chinese topic. .

(38) 27. chain, some students have difficulty identifying the real syntactic subject. They tend to provide it as a subject holder, as the example: “So when young people take the job, it should be careful.” According to Lin, students’ English translation is largely influenced by the Chinese topic-comment structure. In a similar vein, Kwan (1997) observes students’ translation assignment and finds that Chinese students’ translation often lacks cohesive devices, which often. 政治大. results in parallel structures instead of hierarchical structures. Such a lack of cohesive. 立. devices is a reflection of Chinese sentence structures. Contrary to translation textbook. ‧ 國. 學. writers who describe Chinese sentences as paratactic, Kwan notices that the Chinese. ‧. sentences are just “seemingly parallel” (p. 87), especially the topic chain structure.. Nat. io. sit. y. The lack of sentence relation markers does not entail the fact that they are coordinate.. al. er. It is emphasized that translators should be careful in interpreting the unmarked (i.e.. n. v i n CZero covert) cohesive device: zero NP. implies not only coordinate h eNP n gin Chinese chi U relation but also subordinate relation between sentences. Appropriate choice of coordination and subordination will make the message clearer.. From previous studies, we have found that studies on topic-comment translation (Jiang, 2012; Wang, 1996; Zhang, 2006) put emphasis on the translation of topic in a single-clause sentence. Little research has been done on Chinese-English translation of a larger unit—a topic chain. Translation textbooks (Cheng, 1997; Huang, 2007;. .

(39) 28 . Pang, 1997; Yang, 2006) only give descriptions about the different sentence structures between Chinese and English. They do not go into the details and provide a specific guideline on translation of topic chain. Although both Lin (2002) and Kwan (1997) extend their scope to study topic chain translation, they are still limited to two-clause topic chains. Lin (2002) only focuses on how the topic/subject in Chinese is translated into English by students. Kwan (1997) only focuses on how students treat the relation. 政治大. between two clauses. However, in addition to one-clause sentences and two-clause. 立. sentences, what is more prevalent in Chinese is topic chains with more than two. ‧ 國. 學. clauses. Few previous studies have discussed such larger translation unit. Therefore,. ‧. an investigation of the translation strategies of Chinese topic chains is called for.. al. er. io. sit. y. Nat 2.4 Discourse Segmentation. n. v i n C hnotion, when analyzing Since topic chain is a discourse e n g c h i U the translation of topic. chain, several elements will be involved. In this section, we will briefly review the three elements related to discourse segmentation: discourse markers, referent forms and punctuation marks.. .

(40) 29. 2.4.1 Discourse markers8 When it comes to identifying the internal structure of a discourse, discourse markers play an important role. Discourse markers come in a variety of forms. Schiffrin (1987) classifies them into four groups: conjunctions, interjections, adverbs and lexical phrases. Fraser (1999) defines them as “a class of lexical expressions drawn primarily from the syntactic classes of conjunctions, adverbs and prepositional. 政治大. phrases…They have core meaning, which is procedural, not conceptual” (p. 931). In. 立. other words, the interpretation of discourse markers is achieved through the context;. ‧ 國. 學. the function of discourse markers is to signal the relation between two segments of. ‧. text.. Nat. io. sit. y. According to Shiffrin (1987), discourse markers can be classified into two types. al. er. based on their scopes: local and global markers. Local discourse markers link two. n. v i n C hdiscourse markers U propositions together, while global e n g c h i serve as boundary markers,. separating one larger unit from another. The boundary between the different units may be indicated by a change of time, place, action, participant, or textual relation. In other words, global markers are a realization of major continuity break. Another classification based on the semantic relations carried by the discourse markers is provided by Halliday & Hasan (1976). In their framework, conjunctions9                                                         8. Discourse marker is also known as discourse particle, pragmatic markers or connective. Halliday & Hasan (1976) use the term conjunction to refer to not only sentence conjunctions but also adverbs and prepositional phrases such as alternatively, instead, then, for instance, and in conclusion.. 9. .

(41) 30 . are classified into four types of relations: additive, adversative, causal and temporal. The additive relation includes addition of a similar or alternative idea, afterthought, and exemplification with markers like and, or, in addition, and similarly. The adversative relation means counter-expectation with markers like though, but, however, instead, in fact, to name just a few. The causal relation focuses on cause, effect or purpose with markers like so, therefore, consequently, and otherwise. The. 政治大. temporal relation expresses time sequence of two events with markers like next,. 立. before, first, second, and to sum up.. ‧ 國. 學. The correspondence between Halliday & Hasan’s conjunctions and larger textual. ‧. patterns is examined by Yu (2007). There are eight textual relationship patterns Yu. Nat. io. sit. y. summarizes from previous research and reading/writing textbooks: 1. General-example(s),. al. er. 2. General-specific(s), 3. Group-member, 4. Problem-solution, 5. Question-answer, 6.. n. v i n Cause-effect, 7. Time sequence,C and h8.eClaim-counterclaim. n g c h i U The patterns of. General-example(s) and General-specific(s) describe texts that begin with a general statement then followed by examples or specific details to explain the general statement. The Group-member pattern starts with a general description of a group followed by detailed description of each member. In the next two patterns, Problem-solution and Question-answer, writers provide a problem/question and then give the answer. These two are frequently found in advertisement. Cause-effect and. .

(42) 31. Time sequence patterns are often found in expository essays and narrations. The Claim-counterclaim pattern appears mostly in argumentation. Halliday and Hasan’s additive conjunctions can be found in General-example(s), General-specific(s), and Group-member patterns. The adversative conjunctions can be found in Claim-counterclaim pattern. The causal additive can be found in Problem-solution, Question-answer, and Cause-effect patterns. Last, the temporal conjunctions can be found in Time-sequence pattern.. 立. 政治大. Connectives in Chinese discourse have also been studied. According to Yu (1990),. ‧ 國. 學. the four types of connectives proposed by Halliday & Hasan (1976)—additives,. ‧. adversatives, causals and temporals—can also be applied to Chinese written. Nat. io. sit. y. narratives. They mark discourse segments of different levels. Additives usually. al. er. connect micro-structures under one topic. Adversatives may mark the shift of one. n. v i n C h can serve as both topic to another. Causals and temporals e n g c h i U global markers and local markers. In addition, Qin (1998) studies the linguistic devices used to structure a spoken discourse. She gathered her data by asking 20 teachers to tell an unforgettable teaching experience in Chinese. By analyzing the story structure based on Labov’s model (1972): abstract, orientation, complicating action, evaluation, resolution, coda, she finds that the marker na ‘that’ could mark a story’s orientation, complicating action and evaluation. Evaluation often begins with the markers like jieguo ‘as a result’. .

(43) 32 . or houlai ‘after that.’ The coda is usually marked by suoyi ‘so.’ From these studies, we have come to an understanding that connectives play an important role in Chinese discourse segmentation.. 2.4.2 Referent forms It should be noted that there is one type of marker that is not usually viewed as. 政治大. discourse marker: anaphora. Although many grammar books would say that the use of. 立. anaphoric forms is to avoid repetition, anaphors in fact serve the function of marking. ‧ 國. 學. the structure of discourse.. ‧. Givón (1983) proposes a scale in the syntactic coding of topic accessibility.. Nat. io. sit. y. From the most continuous/accessible to the least continuous/accessible, the coding. n. al. er. goes from zero anaphora to referential indefinite NP’s as in (21). (21). Ch. engchi. i n U. v. most continuous/accessible topic zero anaphora unstressed/bound pronouns or grammatical agreement stressed/independent pronouns R-dislocated DEF-NP’s neutral-ordered DEF-NP’s L-dislocated DEF-NP’s Y-moved NP’s (‘contrastive topicalization’) cleft/focus constructions referential indefinite NP’s most discontinuous/inaccessible topic (Givón, 1983, p. 17) This scale reflects the iconicity principle underlying the syntactic coding. The more. .

(44) 33. coding material is used, the less continuous the topic is. However, this scale involves various kinds of coding means including phonological size, word order and morphology. Given that the present study focuses on written material, only morphology is of our concern. Thus, the nominal forms of high accessibility to low accessibility would be briefed: zero, pronoun, and NP. The correspondence between topic accessibility and the three referent forms. 政治大. can also be found in Chinese (Chen, 1986; Chen, 2010; Li, 1985). Along the scale of. 立. continuity, zero is used when the referent occurs in the discourse that is high in. ‧ 國. 學. continuity. A pronoun occurs in the discourse with less continuity. A full NP occurs in. ‧. a discourse marked by major breaks. According to Li (1985), the three referent forms. Nat. io. sit. y. serve as formal demarcations of the three levels of discourse units: the clause, the. al. er. topic chain, and the paragraph. Zero occurs within the topic chain and it is used to. n. v i n C hPronouns occur at U combine clauses into a topic chain. e n g c h i the boundary between topic. chains and they are used to indicate a break in semantic continuity. Full NPs occur at the beginning of a new paragraph to introduce a new referent or to reintroduce an old referent.. 2.4.3 Chinese punctuation system There are 15 punctuation marks listed in Manual of Punctuation Marks. .

(45) 34 . (National Languages Committee, 2008). However, only six of the punctuation marks are related to topic continuity. An illustration of continuity scale based on the function of each punctuation mark is given in (22). (22) most continuous marker listing marker (、) colon (：) comma (，) dash (—) semicolon (；) full stop (。). 政治大 least continuous marker 立. ‧ 國. 學. A listing marker is used to list parallel words or phrases; thus it is the most continuous marker. A colon is used to provide examples. A comma indicates a pause in speech. ‧. between any constituents. It can show up after a noun phrase, a preposition phrase, an. sit. y. Nat. io. n. al. er. adverbial phrase or even a clause. It can mark either a minor continuity break or a. i n U. v. major continuity break between clauses. A dash is used to mark a change in meaning. Ch. engchi. or provide extra explanation. A semicolon is used to separate clauses of parallel structures or to mark adversative textual relation. A full stop is used in the end of a sentence to mark the completeness of meaning. Although Chinese punctuation marks are less rigid compared with English ones, tendencies of consistence can still be found among users. Chu (1992) has done a small survey on the use of punctuation marks. He takes a short passage of 11 clauses and makes all the clauses end with commas, except for the last clause. Then he asks .

(46) 35. subjects to put full stops at the end of clauses based on their instinct. Three clauses receive the most full stops. The three clauses are exactly the same clauses where the author puts his full stops. This suggests that Chinese speakers show consistency in the use of a full stop as a boundary marker. In the present study, full stops are adopted as indicating the boundary of topic chains.. 2.5 Summary. 立. 政治大. In this chapter, the differences between subject and topic have been presented.. ‧ 國. 學. Subject is a syntactic notion while topic is a discourse notion. A topic can link several. ‧. clauses and form a topic chain. Examples of typical and atypical topic chains are. Nat. io. sit. y. presented. In addition to previous research on topic chain, we have also reviewed. al. er. studies on the translation of topic chains, and have pointed out that most of them do. n. v i n C hstructure of topic chain. not go into detail about the internal e n g c h i U In order to analyze topic chain, a discourse structure, we have reviewed discourse markers, referent forms and the Chinese punctuation system, which are related to topic continuity in Chinese discourse. The discussion in this chapter shall pave our way to the next chapter: Methodology.. .

(47) 36 . Chapter Three Methodology Due to limited references regarding the research of translation of topic chains, the present study adopts a data-driven approach. Translation strategies would be identified through classifying, comparing, and analyzing the translation data. A preliminary analysis is conducted on a small number of the collected data to test the. 治政 research framework. The tentative framework is then revised 大 based on the findings 立 ‧ 國. 學. from the preliminary analysis. After that, more data are analyzed for the formal study. Section 3.1 presents descriptions of the data. Following that, Section 3.2 gives. ‧. the definition of topic chain in the present study. Section 3.3 describes the preliminary. sit. y. Nat. io. n. al. er. analysis and the revised framework in response to the problematic cases. Last in section 3.4, the analysis procedure will be presented.. Ch. engchi. i n U. v. 3.1 Data Collection Two types of material sources are selected for the analysis: a bilingual magazine and translation textbooks. The Chinese-English bilingual magazine, Taiwan Panorama10, is selected because of its variety of content and because of its professional writing and translation. First published in 1976, Taiwan Panorama covers                                                         10. . Before January 2006, Taiwan Panorama was named Sinorama. Readers may find this title elsewhere..

(48) 37. a wide range of Taiwan topics from politics, economy, society, culture, to Taiwan-related international events. Articles in Taiwan Panorama are translated by experienced native American translators. With its professional Chinese writing and English translation, it has been used as course materials in many college translation classes and has even been used to build a bilingual parallel corpus under the project CANDLE as a learning tool for intermediate learners of English11.. 政治大. In the present study, an issue of Taiwan Panorama, April 2013, is randomly. 立. chosen. It contains 21 articles, covering a wide range of topics, including school. ‧ 國. 學. lunches, a new design hotel, a thriving comedy club, pet funerals as well as tips to. ‧. losing weight. Ten articles are selected for the present study. To avoid preferred. Nat. io. sit. y. structures or translation styles by a certain translator, it has been made sure that the. n. al. er. articles are translated by diverse translators. These 10 articles are translated by seven. Ch. different translators (Appendix I).. engchi. i n U. v. In addition to Taiwan Panorama, five textbooks are also examined. The textbook material includes both practical translation textbooks and research-based books on the topic of Chinese-English contrastive analysis (Appendix I). The topic chain translation sentences are taken either from exemplification sentences or from                                                         11. CANDLE stands for Corpus and NLP for Digital Learning of English (http://candle.cs.nthu.edu.tw). It is a national e-learning project (2003-2006) lead by Professor Hsien-Chin Liou of National Tsing Hua University. CANDLE provides a variety of natural language processing (NLP) tools, including part of speech tagging, bilingual concordance, WordNet, and collocation translator, to name just a few. For more information about CANDLE and application of CANDLE to teaching, please refer to Chang, Wu, Shei, Chang, & Jian (2004), Chan & Liou (2005), Lin, Chen, Liaw, Liou, & Yeh (2005) and Liou & Chang (2005)..