• 沒有找到結果。

論漢語中的終結點選擇

N/A
N/A
Protected

Academic year: 2021

Share "論漢語中的終結點選擇"

Copied!
79
0
0

加載中.... (立即查看全文)

全文

(1)國立臺灣師範大學英語學系 碩. 士. 論. 文. Master Thesis Department of English National Taiwan Normal University. 論漢語中的終結點選擇. On Telicity Selection in Mandarin Chinese. 指導教授:吳. 曉. 虹. Advisor: Dr. Hsiao-Hung Wu 研 究 生:張. 家. 豪. 中華民國一 百 零 二 年 六 月 June, 2013.

(2) 摘. 要. 本文主要研究漢語中的終結點是如何決定。林若望(2003)指出漢語中,現在式選擇已終 結事件,而過去式選擇未終結事件。 與林若望(2003)不同的是,我提出過去式的情境裡,終結點是由時態以外的的因素 決定,例如受詞的定指與不定指、主要動詞的自然語意、以及漢語的內部時貌標記。關 於內部時貌,我會引用 Travis (2006)與廖偉聞(2004)在文中另外介紹。 我主要的結論是,時態為過去式時,漢語的終結點是由自然語意和內部時貌標記決 定的。在分析過程中,我會提供實質的語料證據。另外,我會區分語意中實現貌與終結 點的差別。除此之外,我還會提出漢語內部時貌標記與動詞的共現狀況,並且進一步推 斷出漢語內部時貌標記的自然語義 關鍵字:終結點、漢語、內部時貌. 1.

(3) ABSTRACT This thesis mainly focuses on the selection of telicity in Mandarin Chinese. Lin (2003) suggests that covert present tense selects atelic events while covert past tense selects telic events. Partially different from Lin (2003), my primary proposal is that when the tense is past, there are other semantic and syntactic factors that select Chinese telicity, including definiteness of the object, the semantic nature of the main verb, and overt inner aspectual markers. Chinese inner aspectual markers will be introduced in my thesis mainly based on Travis (2006) and Liao (2004). My primary conclusion is that semantic nature and inner aspectual markers are the keys to the selection of Chinese telicity while the tense is past. During the analyses, I will provide substantial empirical evidence for my proposal. Besides, I will clarify the difference between the semantic realization and telicity in Chinese. Moreover, I will discuss the fact that each Chinese inner aspectual marker has clear co-occurrence with certain verb types, and the semantic nature of Chinese inner aspectual marker can be indicated by the rules of each co-occurrence. Key Words: Telicity, Mandarin Chinese, Inner Aspect. 2.

(4) ACKNOWLEDGEMENT This thesis is my first long-term study as a graduate student. With the interest in syntax and semantics, I decided to work on the theory of telicity and also on Chinese inner aspect in my M.A. thesis. Although a lot of problems appeared, I still tried my best and eventually finished the final version. As an M.A. student, of course my study cannot be an influential work for the academic circle of linguistics, but I hope that the Chinese data organized in this thesis will still contribute to later researches on Chinese linguistics. My great thanks are firstly given to my committee members: Prof. Hsiao-Hung Wu, Prof. Miao-Ling Hsieh, and Prof. Chen-Sheng Liu. Working with Prof. Wu, who was a thesis advisor, was a valuable experience and she always encouraged me to express my own ideas toward linguistics; Prof. Hsieh gave me inspirations on the production of the Chinese data; also, Prof. Liu raised many substantial questions on my study so that I realized how to enhance the content. In addition to the excellent professors, many students helped me a lot as well. I want to thank my Chinese informants, who are mostly my classmates in college and graduate school, for their contribution to the analyses of my Chinese data. In the end, I want to thank my mother, who always encouraged me to choose my own way to lead a life. She taught me not to stop pursuing a goal no matter how many frustrations there are. Although I will leave the academic circle after gaining the M.A. degree, this thesis has been a memorable journey of linguistics in my life. 3.

(5) TABLE OF CONTENTS. CHAPTER 1 INTRODUCTION 1. Overview. 7. 2. Linguistic Aspect. 10. 2.1 The Syntactic Semantic Concepts of Aspect. 10. 2.2 On Inner Aspect. 12. 3. Chinese Verbs and Telic Events. 16. 4. The Classification of Chinese Verbs. 19. 5. Definitions of Semantic Features. 23. 6. The Semantic Nature of Verbs. 26. CHAPTER 2 ISSUES ON CHINESE TELICITY 1. Overview. 28. 2. Semantic Realization vs. Telicity. 28. 3. Issues on Chinese Verbs and Telicity. 35. 4. Summary. 41 4.

(6) CHAPTER 3 EMPIRICAL ANALYSES 1. Overview. 42. 2. The Selection of Telicity in Chinese. 42. 3. The Effect of Objects. 53. 4. On Chinese Achievement Verbs. 56. 5. The Co-occurrence of Chinese Inner Aspectual Markers and Verb Types. 58. 6. Summary. 69. CHAPTER 4 CONCLUSION 1. Overview. 71. 2. A Theory of Chinese Telicity. 71. 3. Linguistic Behavior of Chinese Inner Aspect. 73. 4. The Issue on Chinese Tense. 75. REFERENCES. 76. 5.

(7) TABLE 1 Verb Types and Feature Specification. 19. FIGURE 1 The Selection of Chinese Telicity. 72. 6.

(8) On Telicity Selection in Mandarin Chinese. CHAPTER ONE INTRODUCTION. 1. Overview. This thesis mainly focuses on the selection of telicity ([±telic]) in Mandarin Chinese 1. Lin (2003) suggests that covert past tense ([+past]) selects [+telic] events, as illustrated in (1).. (1) Ta he. da-po. yi-ge. hua-ping. (Lin 2003). hit-broken. one-CL flower-bottle. “He broke one vase.”. The tense of (1) is covert since there is no overt tense markers, such as temporal adverbs zuotian ‘yesterday’ or jintain zaoshang ‘this morning’, and the semantic tense of (1) can only be [+past]. On the other hand, if (1) is uttered in real life, the event of (1), breaking the vase, is definitely finished because the action dapo ‘break’ must be achieved in an instance, and therefore the event of (1) is telic.. 1. Mandarin is a Chinese dialect which originates mainly from Beijing, and Mandarin has been spoken as the. official language in China since 1728. I will only conduct a study of Mandarin without any discussion on other Chinese dialects in this thesis.. 7.

(9) On Telicity Selection in Mandarin Chinese Also, Lin suggests that covert present tense ([+present]) selects [-telic] events, as illustrated in (2).. (2) Wo xiangxin ni I. believe. [+present][-telic] (Lin 2003). you. “I believe you.”. In Chinese, the tense of (2) can only be [+present] since the state verb xiangxin ‘believe’ needs overt tense markers, such as a temporal adverb yiqian ‘in the past’, if it denotes a [+past] event, to receive [+past]; also, due to the fact that a state verb is imperfective, the event of (2) is [-telic]. Lin (2003) organizes the selection of [±telic] as shown in (3a-b).. (3) Lin’s Analysis in His 2003 Paper. (Lin 2003). a. Covert present tense must select imperfective AspP as its complement. b. Covert past tense must select perfective AspP as its complement.. Partially different from Lin (2003), in which he suggests that [+present] selects [-telic] while [+past] selects [+telic], my primary proposal is that when the tense is [+past], there are 8.

(10) On Telicity Selection in Mandarin Chinese other semantic and syntactic factors that select Chinese [±telic], including definiteness / semantic restriction of the object, the semantic nature of the main verb, and overt inner aspectual markers (InAsp). Besides, Lin (2003) proposes that Chinese semantic tense is combined with aspect since there is a strong relation between semantic tense and aspect, as exemplified in (1-3). Also, Lin (2005, 2010) raises the possibility that tense-like components are combined with aspect. Based on this idea, he suggests that the tense phrase (TP) is unnecessary in the syntax of Chinese due to the consideration that the T head is not needed to contain the tense feature since the aspect head already bears semantic tense. If Lin is correct, there is a possibility that TP does not exist in Chinese according to the Economy Condition (Chomsky 1995), which indicates that an unnecessary syntactic element should not exist in theory. In order to state Economy Condition clearly, I adopt Radford’s (2004) definition as in (4).. (4) Economy:. (Radford 2004). Economy considerations require that (all other things being equal) syntactic representations should contain as few constituents and syntactic derivations involve as few grammatical operations as possible.. As indicated in (4), a syntactic representation should have as few constituents as 9.

(11) On Telicity Selection in Mandarin Chinese possible. Consequently, if Chinese TP is unnecessary, as suggested by Lin (2003), TP should be eliminated from the syntax of Chinese. This conclusion opposes some proposals in which TP is needed in Chinese (Lin 2003, 2007, Sybesma 2007). I will have further discussion on the issue of whether or not T can completely be indicated by aspect in Chinese in the end of this thesis. In the rest of this chapter, I will discuss inner aspect (InAsp) in Section 2, as well as Chinese verbs and [+telic] events in Section 3; classification of Chinese verbs will be introduced in Section 4, and finally the definitions of semantic features proposed in this study will be shown in Section 5. The organization of the chapters in this thesis also includes a deeper discussion of previous issues on Chinese telicity in Chapter 2, and my main analyses in Chapter 3, in which Chinese empirical evidence on the selection of [±telic] will be provided to support my proposal; in addition, there will be deeper investigation of the effect of objects and verbs on [±telic]. In the end, I will conclude this thesis in Chapter 4.. 2. Linguistic Aspect 2.1 The Syntactic and Semantic Concepts of Aspect In syntax, aspect is considered a head which projects the aspect phrase (AspP) and as well bears verb-related semantic features, which primarily includes the perfect, the progressive, 10.

(12) On Telicity Selection in Mandarin Chinese and the perfective (Comrie 1976, Liao 2004, Travis 2006). Moreover, semantic aspect is not only indicated by the syntactic aspect head, but also in some cases by verbs which contain the semantic nature of aspect. Examples in English are shown in (5a-c) and those in Chinese are shown in (6a-c).. (5) a. Samantha has learned five languages in Asia. (perfect) b. Larry is running after the bus. (progressive) c. Paula ate a hamburger in the morning. (perfective). (6) a. Wo I. kan-le. na-ge. jiemu. watch-ASP. that-CL. (TV) show. (perfect). “I watched that TV show.” b. Wo I. zhengzai zhu. fan. PROG. rice. cook. (progressive). “I am cooking (a meal).” c. Wo I. xie-wan. gongke. le. write-InAsp. homework. ASP. “I finished writing the homework.”. 11. (perfective).

(13) On Telicity Selection in Mandarin Chinese In (5a), the perfect aspect is indicated by the auxiliary has, which is a common indicator of the perfect aspect in English; on the other hand, running in (5b) shows the progressive aspect via the suffix -ing; (5c) is considered with the perfective aspect since the English verb ate itself contains the semantic nature which indicates that the event of eating something in the sentence is fully accomplished. Also, Chinese has ways to express aspect (Liu 1988, Lin 2003, Liao 2004, Lin 2005). In (6a), the aspectual marker le expresses the meaning of realization of an action and therefore indicates the perfect aspect; the progressive aspect in (6b) is indicated by the aspectual marker zhengzai; the event of (6c) is strictly perfective since there is an InAsp wan. As can be observed, aspect can be indicated by different ways in different languages. An event in a sentence gains its aspect through auxiliaries, suffixes or aspectual marker, and the semantic nature of a verb. My thesis primarily focuses on lexical aspect rather than grammatical aspect; also, in this study, I will mainly focus on the perfective aspect in order to conduct my investigation of a theory of telicity. The perfective aspect, if indicated by the syntactic structure, is contained in InAsp, which will be further discussed in the following section.. 2.2 On Inner Aspect The study of InAsp has been conducted in many languages. Travis (2006) proposes the 12.

(14) On Telicity Selection in Mandarin Chinese syntactic position of InAsp, which is inside the VP-shell (Larson 1988), as illustrated in (7).. (7) VP V’ V. InAspP InAsp’ InAsp. VP. Travis (2006) initially discusses the syntactic position of aspect. Firstly, based on Chomsky (1965), aspect is under AUX in a phrase structural tree; therefore, in a tree with TP, aspect should be under T. Moreover, Travis (2006) concludes that there should be a VP at [Comp, AspP] since aspect is a syntactic head determining a VP and aspect has scope over V. Therefore, if there is an InAspP, it should be at the potential position between the two VPs of a VP shell, as in (7). Besides, in the study of Chinese, Liao (ms.) proposes that there are three layers of aspects in the syntactic structure, which he calls Three-layerd Aspectual Structure (TAS), as shown in (8).. 13.

(15) On Telicity Selection in Mandarin Chinese (8) higher-AspP …… VP …… middle-AspP …… VP InAsp. In (8), Liao’s middle-AspP is in the identical position of InAspP by Travis (2006). Moreove, Liao further discusses the semantic nature in his Chinese middle-AspP, which bears the perfective aspect and imperfective aspect. In my talk, the perfective aspect leads to [+telic] events while imperfective aspect leads to events of underspecified telicity ([0-telic]), as exemplified in (9a-b).. 14.

(16) On Telicity Selection in Mandarin Chinese (9) a. Wo I. chi-le. yi-tiao. yu,. [+telic]. eat-ASP one-CL fish. “I ate one fish,” (*Keshi but. hai-mei chi-wan) yet-not. eat-InAsp. “but I did not finish eating it.” b. Wo I. du-le. yi-ben. shu,. [0-telic]. read-ASP one-CL book. “I read one book,” (Keshi but. hai-mei du-wan) yet-not. read-InAsp. “but I did not finish reading it.”. The middle-Asp of (9a) bears the perfective aspect so the event is [+telic], and therefore (9a) cannot have the supplementary information keshi hai-mei chi-wan ‘but I did not finish eating it’. In contrast, the middle-Asp of (9b) bears the imperfective aspect so the event is [0-telic], and therefore, (9b) can have the supplementary information keshi hai-mei du-wan ‘but I did not finish reading it’. In my paper, InAspP by Travis and middle-AspP by Liao are considered the same 15.

(17) On Telicity Selection in Mandarin Chinese syntactic projection which bears the same sematic nature and is called InAspP.. 3. Chinese Verbs and Telic Events Comrie (1976) defines [±telic] based on whether there is a set ending point. As a result, English verbs such as sing is [-telic] while sing five songs is [+telic] because there is a set ending point at which the five songs are completely sung. In my talk, I define [+telic] events as the events that must have been completed in reality rather than that have set ending points. In Comrie’s theory, read War and Peace is [+telic] since there is a natural ending point with the fact that a book is going to be finished by certain readers anyway; however, I suppose that the [±telic] of read War and Peace is basically underspecified ([0-telic]) due to the fact that the natural ending point of read War and Peace is a set termination instead of an actually achieved termination in reality, as in Mary read War and Peace, in which Mary did not absolutely finish reading the whole book if there is no supplementary information in the context. The reason why I insist on a stricter definition of [+telic] compared with Comrie’s is based on the fact that there is a natural ending point for every event. For instance, in Mary swims, the swimmer is somehow going to stop swimming since a normal human being is unlikely to swim forever; for another real-life example, in Peter lives in Taipei, the action of living in Taipei will somehow naturally end since Peter, a natural human being, will definitely 16.

(18) On Telicity Selection in Mandarin Chinese die someday and no longer live in reality. In still another example, the star shines, a star is definitely going to stop shining someday; this is due to the fact that a star will explode as a supernova and finally becomes a neutron star 2 when its chemical reaction of nuclear fusion, which fuels the star to shine, stops while there is no sufficient atoms available for the chemical reaction 3. As can be observed, even a star in the sky will naturally stop shining. Therefore, realistically, every event that happens in the real world should have a natural ending point. With the consideration that there should be a natural ending point for every event in the universe, all events can be analyzed as [+telic] with Comrie’s definition. Consequently, I use the strict definition of [+telic], which requires an event to be completely accomplished in reality to be viewed as a [+telic] event. Let us take English for example. Some verbs, such as build and write, obligatorily select [+telic] events, as shown in (10a-b).. (10) a. I built a house, (*but I have not finished building it.) b. I wrote a letter, (*but I have not finished writing it.). 2. A neutron star is a high-density ball composed almost by neutrons. Since there is no nuclear reaction in a. neutron star, it cannot shine. 3. If the reader is interested in how a star dies and stops shining, an introductory article in Scientific American by. Hillebrandt, Janka and Müller (2006) is an excellent reference.. 17.

(19) On Telicity Selection in Mandarin Chinese. The main verbs in (10a-b) select [+telic] events since the events entailed by build and write must be accomplished. On the other hand, in the cases of Chinese [+telic] events, Lin (2004) discusses Chinese monosyllabic verbs that do not obligatorily select [+telic] events, such as gai ‘build’, and xie ‘write’, as illustrated in (11a-b).. (11) (Lin 2004) a. Na-dong that-CL. fangzi,. ta. yijing. gai-le. san. nian le,. house,. he. already. build-ASP. three. year ASP. dao. xianzai. hai-mei gai-hao. to. now. yet-not. build-InAsp. “That house has been under construction for three years, but it is still not fully built.” b. Na-feng that-CL. xin,. wo xie-le. san. tian, cai xie-wan. letter,. I. three. day, just write-InAsp. write-ASP. “I spent three days finishing writing that letter.”. As can be observed, gai ‘build’, and xie ‘write’ in (11a-b) do not obligatorily select [+telic] events since verbs which can co-occur with ‘for a period of time’ do not entail 18.

(20) On Telicity Selection in Mandarin Chinese complete events, as suggested by Vendler (1967); however, as can be indicated in (11a-b), gai-hao ‘build-InAsp’ and xie-wan ‘write-InAsp’ both indicate that the events are completed. In my analyses in Section 2, I will also argue that InAsp is a strong syntactic factor that selects [+telic].. 4. The Classification of Chinese Verbs Vendler (1967) classifies English verbs as in Table 1, adopted from the version in Rothstein (2004).. Table 1 Verb Types and Feature Specification. (Rothstein 2004). [±stage]. [±telic]. States. [-stage]. [-telic]. Activities. [+stage]. [-telic]. Achievements. [-stage]. [+telic]. Accomplishments. [+stage]. [+telic]. In English, examples of Table 1 are illustrated in (12a-d).. 19.

(21) On Telicity Selection in Mandarin Chinese (12) a. States Mary believes that the astrology can predict the future. b. Activities The dog runs after cars all the time. c. Achievements Elizabeth died of lymphoma last year. d. Accomplishments Jill built a tree house by herself.. On the other hand, the corresponding Chinese examples, which are translations of (12a-d), are shown in (13a-d).. (13) a. States Mali. xiangxin zhanxing. neng. yuce. weilai. Mary. believe. can. predict. future. astrology. “Mary believe that astrology can predict the future.” 20.

(22) On Telicity Selection in Mandarin Chinese b. Activities Na-zhi. gou. that-CL dog. zong. zhui-zhe. che pao. always. chase-ASP. car run. “That dog always runs after cars.” c. Achievements Yilishabai. qu-nian si. yu. Elizabeth. last-year die of. linba-ai lymph-cancer. “Elizabeth died of lymphoma last year.” d. Accomplishments → Not Found in Chinese Jier ziji. gai-le. yi-jian. shu-wu. Jill herself. build-ASP. one-CL tree-house. “Jill built one tree house by herself.”. The main verb in (13d) gai ‘build’ is actually an activity verb since gai is [+stage][-telic]. Researches on Chinese aspect have suggested that there is no accomplishment verbs in Chinese (Tai 1984, Lin 2004, Liao 2004). The corresponding Chinese verbs of English accomplishment verbs are actually activity verbs, as illustrated in (14a-d).. 21.

(23) On Telicity Selection in Mandarin Chinese (14) a. Wo I. xie-le. yi-feng. xin. write-ASP. one-CL letter. “I wrote one letter.” b. Wo. gai-le. yi-dong fangzi. I. build. one-CL house. “I built one house.” c. Wo I. hua-le. yi-tiao. long. paint-ASP. one-CL dragon. “I painted a dragon.” d. Wo I. du-le. yi-ben. shu. read-ASP one-CL book. “I read one book.”. In (14a-d), the English translations are [+stage][+telic], so write, build, paint, and read are accomplishment verbs; however, all of the Chinese examples in (14a-d) are [+stage][-telic], or the events are [0-telic], which means that xie ‘write’, gai ‘build’, hua ‘paint’, and du ‘read’ are activity verbs. The main classification of Chinese verbs, therefore, only includes states, activities, and achievements. Besides, some other semantic features will also be considered in my analyses 22.

(24) On Telicity Selection in Mandarin Chinese in Chapter 3, as shown in Section 5.. 5. Definitions of Semantic Features I organize the semantic classification in (15-18) for my primary analyses in Chapter 3.. (15) [+telic] An event is [+telic] iff the event must have been finished entirely in reality. Example: Yilishabai. qu-nian si. yu. Elizabeth. last-year die of. linba-ai lymph-cancer. “Elizabeth died of lymphoma last year.” . The action of dying must be achieved already in reality when this utterance occurs.. (16) [+stage] A verb is [+stage] iff its action takes a process before reaching a resultative point 4.. 4. The process of an action exists if and only if the process, no matter how short it is, can be possibly measured. by time. The action of gai fangzi ‘build a house’ takes an obvious process since it usually takes years to finish building a house. In other cases, some actions, such as gua ‘hang’ and tun ‘swallow’, usually only take an extremely short period of time; however, these actions are still [+stage] even though they can usually be finished in milliseconds in daily lives.. 23.

(25) On Telicity Selection in Mandarin Chinese Example: Wo gai-le. yi-dong fangzi. I. one-CL house. build. “I built one house.” . There must be a process before reaching the resultative point of finishing building the house since it is impossible to build a house without any process in reality.. (17) [+consume] 5 A verb is [+consume] iff the action of the verb makes an element x disappear from a given possible world 67. 5. Although the semantic feature [+consume] is not commonly seen in other researches, [+consume] is used in. my study since I find that this feature is essential to the selection of telicity in Mandarin Chinese. Given that there is a possibility that [+consume] can be decomposed into traditional semantic features, the decomposition of [+consume] may be a possible topic for future research. 6. The idea of [+consume] can be illustrated with sets of possible worlds as in (i).. (i) Assumptions: There exists a set Plate including one element Fish, given Plate: {Fish}. There exists a set Stomach including the empty set when no food is being digested, given Stomach: {{}}. The event of Eating Fish is [+consume] if and only if the event makes an element x disappear from a given possible world.. 24.

(26) On Telicity Selection in Mandarin Chinese Example: Wo chi-le I. yi-tiao. yu,. [+telic]. eat-ASP one-CL fish. “I ate one fish.”. (18) [+achieve] A verb is [+achieve] iff the action of the verb leads to a [+telic] event without any semantic stage 8. Example: Wo ying-le. na-chang. bisai. I. that-CL. game. win-ASP. “I won the game.”. Eating Fish Plate: {Fish}; Stomach: {{}}. Plate: {{}}; Stomach: {Fish}. Fish disappears from a given possible world Plate. → Eating Fish is [+consume]. 7. The Chinese verb si ‘die’ and sha ‘kill’ may be suspected as [+consume] since both dying and killing are to. make “life” disappear from the world in which the dead creature used to live. However, dying or killing scientifically does not make “life” disappear since dying means that the heart stopped beating or the brainwave is a flat line. Therefore, si ‘die’ and sha ‘kill’ are considered [-consume]. 8. Chinese verbs such as si ‘die’ and sha ‘kill’ are also [+achieve] verbs since dying and killing take no semantic. stage.. 25.

(27) On Telicity Selection in Mandarin Chinese . Ying ‘win’ must be [+telic] when the sentence is uttered, and the action of ying is done in an instance without any process in reality.. 6. The Semantic Nature of Verbs It should be noticed that the semantic nature of a verb depends on the semantic features fitting in the context rather than on the verb itself. For example, the verb sha ‘kill’ can have different semantic features in different contexts, as shown in (19a-b). (19) a. Wo I. sha-le. yi-ge. ren. kill-ASP one-CL person. “I killed a person.” b. Wo I. zaoshang zai shichang. sha yu. morning in. kill fish. market. “I was handling fish in the market this morning.”. In (19a), the verb sha is [-stage] since dying takes no semantic stage; however, sha in (19b) means to handle a fish by removing its scales and the visceral while the fish is already dead, so sha in (19b) is [+stage] since handling a fish must take a semantic process. In my later analyses in this thesis, some other verbs may also be fitted in different semantic features 26.

(28) On Telicity Selection in Mandarin Chinese in different contexts, and the semantic features, instead of the verb itself, are the actual keys for the selection of telicity. In Chapter 2, I will give a deeper discussion of issues related to Chinese telicity; moreover, in Chapter 3, I will propose the detailed factors which have influence on the selection of [±telic] in Chinese, and also I will further discuss the behavior of Chinese InAsp.. 27.

(29) On Telicity Selection in Mandarin Chinese. CHAPTER TWO ISSUES ON CHINESE TELICITY. 1.. Overview. Chinese [±telic] has been widely studied and in this Chapter I will discuss some issues on the reading of [±telic]. In Section 2, I will demonstrate the difference between the realization of an event and the telicity of an action. Also, a detailed investigation of Chinese verbs and [+telic] events will be conducted in Section 3. Finally, a summary will be in Section 4.. 2.. Semantic Realization vs. Telicity. As discussed in Chapter 1, in my thesis, an event is [+telic] if ands only if it is fully accomplished in reality, different from Comrie’s (1976) idea toward [+telic] events since Comrie’s definition of [+telic] as events with natural ending points may cause the result that all events are [+telic] due to the fact that every event in reality has a natural ending point. On the other hand, the realization of an event means that the event has been at least partially done or experienced rather than that the event is completely accomplished (Liu 1988, Liao 2004). For example, you only need to read one page of a book to make reading a book semantically realized. Liao (2004) provides examples for the demonstration of realized events as in (1a-b), in which for three hours is used as an indicator for realized but [0-telic] events. 28.

(30) On Telicity Selection in Mandarin Chinese. (1) (Liao 2004) a. Zhangsan Zhangsan. du-le. zhe-ben shu. san-ge. xiaoshi. read-ASP. this-CL book. three-CL. hour. [0-telic]. “Zhangsan has been reading this book for three hours.” (Liao 2004) b. Zhansan Zhangsan. chi-le. zhe-ge. hanbao. eat-ASP. this-CL hamburger. san-ge. xiaoshi. [0-telic]. three-CL hour. “Zhangsan has been eating this hamburger for three hours.”. The book in (1a) is not definitely finished, and the hamburger in (1b) is not definitely fully consumed; as a result, (1a-b) are [0-telic]. Also, based on Vendler (1967) 1, (1a-b) cannot be [+telic] since for three hours can only be combined with events without the absolute [+telic] reading. As can be indicated in (1a-b), in Chinese, the realization of an action can be. 1. Vendler (1967) proposes that for a period of time cannot be combined with [+telic], as exemplified in English. in (ia-d).. (i). a. Cindy jogged for an hour this morning. b. *Cindy ate a banana for an hour this morning. [+telic] c. John studied mathematics for three hours yesterday. d. *John finished his math homework for three hours yesterday. [+telic]. 29.

(31) On Telicity Selection in Mandarin Chinese expressed by the aspectual marker le in most cases. Le is analyzed as a realization aspect in Liu (1988), and Liao (2004) suggests that le is a realization marker which indicates the initial point instead of the ending point of an action. Besides, for a deeper investigation of whether Vendler’s (1967) mechanism is fully suitable in Chinese, I will discuss some Chinese examples appear to allow for a period of time to be combined with [+telic] events, as shown in (2a-b).. (2) (Liao 2004) a. Zhangsan Zhangsan. chi-wan. yi-ge. hanbao. eat-InAsp. one-CL hamburger. san-ge. xiaoshi. three-CL hour. le. [+telic]. ASP. “Zhangsan finished eating a hamburger for three hours.” (Lin 2004) b. Ta He. si. wu. die five. nian. le. year. ASP. [+telic]. “It has been five years after he died.”. The acceptability of (2a) is due to the fact that san-ge xiaoshi ‘for three hours’ is actually combined with the period of time after the hamburger has been fully consumed rather than the event of eating the hamburger, so (2a) means that it has been three hours after the 30.

(32) On Telicity Selection in Mandarin Chinese hamburger has been fully eaten; on the other hand, (2b) is acceptable since wu nian ‘for five years’ is not combined with the event of dying, so (2b) means the fact that it has been five years after the person died rather than that the person has been dying for five years. As can be seen, for a period of time is definitely not allowed to be combined with [+telic] events in Chinese. Back to the issue on the Chinese realization aspectual marker le (Liu 1988, Liao 2004, Lin 2005), since le can only indicate the realization of an event, if the event is in fact [+telic], supplementary information must be added in order to express the [+telic] reading, as illustrated in my examples in (3a-c) and (4a-c).. (3) a. Wo I. jiancha-le. Xiaoming. de. mao. examine-ASP. Xiaoming. DE cat. [0-telic]. “I examined Xiaoming’s cat.” b. Wo I. chi-le. yi-wan. mian. [0-telic]. eat-ASP one-CL noodles. “I ate one bowl of noodles.” c. Wo I. xie-le. yi-ben. shu. write-ASP. one-CL book. [0-telic]. “I wrote a book.” 31.

(33) On Telicity Selection in Mandarin Chinese. (3a) means that the vet examined the cat belonging to Xiaoming, but the process of the physical examination of the cat is not necessarily fully accomplished; also, (3b) can be uttered as long as part of the noodles have been eaten; similarly, (3c) does not definitely mean that the book is finished. Therefore, the actions of (3a-c) are semantically realized but [0-telic]. If the speaker intends to express that (3a-c) are [+telic] in reality, s/he must provide supplementary information, as exemplified in (4a-c).. (4) a. Wo I. jiancha-le. Xiaoming. de. mao,. examine-ASP. Xiaoming. DE cat. “I examined Xiaoming’s cat,” yijing. jiancha-wan-le. already. examine-InAsp-ASP. [+telic]. “and the examination was finished.” 32.

(34) On Telicity Selection in Mandarin Chinese b. Wo I. chi-le. yi-wan. mian,. eat-ASP one-CL noodles. “I ate one bowl of noodles,” quan. dou. chi-wan-le. [+telic]. all. DOU. eat-InAsp-ASP. “and I finished eating all the noodles.” c. Wo I. xie-le. yi-ben. shu,. write-ASP. one-CL book. “I wrote a book,” zheng-ben dou. xie-hao-le. all-CL. write-InAsp-ASP. DOU. [+telic]. “ and I finished writing the whole book.”. Different from (3a-c), in (4a-c), the supplementary information provides the [+telic] reading. As can be observed in (3a-c) and (4a-c), le cannot be the key factor that result in the [+telic] reading; instead, le only indicates that the action of an event is realized. I will further propose the key factors that select Chinese [+telic] events in Chapter 3. Moreover, some Chinese data with le can be [+telic] without any supplementary information, as shown in (5a-c); however, in these examples, le in fact cannot be the key 33.

(35) On Telicity Selection in Mandarin Chinese factor determining [+telic] since the events are still [+telic] when le is removed, as demonstrated in (6a-c).. (5) a. Wo I. zhe-xingqi. sha-le. wu-ge. ren. [+telic]. this-week. kill-ASP five-CL people. “I killed five people this week.” b. Wo I. qu-nian ying-le. san-chang. bisai. last-year win-ASP. three-CL. game. [+telic]. “I won three games last year.” c. Na-chang That-CL. yiwai. si-le. ba-ge. ren. accident die-ASP eight-CL. [+telic]. people. “Eight people died in that accident.”. (6) a. Wo I. zhe-xingqi. sha wu-ge. ren. this-week. kill five-CL people. [+telic]. “I killed five people this week.” b. Wo. qu-nian ying. san-chang. bisai. I. last-year win. three-CL. game. “I won three games last year.” 34. [+telic].

(36) On Telicity Selection in Mandarin Chinese c. Na-chang That-CL. yiwai. si. ba-ge. ren. accident die eight-CL. [+telic]. people. “Eight people died in that accident.”. The killing action must have been fully accomplished in (5a), the games must have been won in (5b), and the people must have been dead in (5c); therefore, (5a-c) are definitely [+telic]. The point is that (6a-c) are without any le but still [+telic]. Consequently, as can be observed in (5a-c) and (6a-c), le does not select [+telic]; instead, (5a-c) are [+telic] because the actions of the verbs sha ‘kill’, ying ‘win’, and si ‘die’ must have been accomplished when the sentences are uttered due to the semantic nature of [+achieve] of the three verbs. After clarifying the difference between the realization and the [±telic] of events, I will discuss previous issues on Chinese verbs and [+telic] events in the following section.. 3.. Issues on Chinese Verbs and Telicity. Chinese is considered a language without any accomplishment verbs (Tai 1984, Lin 2004, Liao 2004). Tai (1984) proposes that the semantic accomplishment in Chinese can only be indicated by two-syllable verbs with additional semantic information. For example, Chinese verbs gai ‘build’ and xie ‘write’ are activity verbs instead of accomplishment verbs because they are [+stage] but [-telic]; in contrast, gai-wan ‘build-InAsp’ and xie-wan ‘write-InAsp’ 35.

(37) On Telicity Selection in Mandarin Chinese can indicate semantic accomplishment since the two activity verbs are combined with InAsp wan as the additional semantic information for [+telic], as demonstrated in (7a-b).. (7) (Tai 1984) a. Wo I. zuotian. xie-le. yesterday write-ASP. yi-feng xin, one-CL letter. “I wrote a letter yesterday,” keshi mei xie-wan but. not write-InAsp. “but I did not finish it.” b. Wo I. zuotian. xie-wan-le. yi-feng xin. yesterday write-InAsp-ASP one-CL letter. “I wrote a letter yesterday,” *keshi mei xie-wan but not write-InAsp “but I did not finish it.”. In (7a), the event is [0-telic] in the first sentence and the second sentence provides the information that the event is [-telic]; on the contrary, the first sentence in (7b) is [+telic] since 36.

(38) On Telicity Selection in Mandarin Chinese there is InAsp wan as the semantic information of [+telic], so the second sentence which implies [-telic] is unacceptable since it contradicts the accomplishment of the event of the first sentence. Moreover, some Chinese verbs, such as da-po ‘hit-broken’, seem to indicate accomplishment, as shown in (8).. (8) (Lin 2003) Ta. da-po. yi-ge. hua-ping. he. hit-broken. one-CL flower-bottle. “He broke one vase.”. The reason why (8) is [+telic] is not because the main verb da ‘hit’, which is an activity verb, can indicate accomplishment by itself, but because the suffix -po ‘broken’ provides the semantic information that the vase is broken already. Therefore, the [+telic] reading of (8) is caused by the semantic nature of -po rather than by the verb da. Here, the idea that there is no accomplishment verbs in Chinese is still supported. Furthermore, Lin (2004) suspects that the definiteness of objects may have influence on the selection the [+telic] reading. Lin (2004) points out that the [0-telic] examples in Tai (1984), as shown in (9), can be [+telic] in other examples with indefinite objects, as shown in 37.

(39) On Telicity Selection in Mandarin Chinese (10a-b).. (9) (Tai 1984) a. Wo I. zuotian. xie-le. yesterday write-ASP. yi-feng xin 2, one-CL letter. “I wrote a letter yesterday,” keshi mei xie-wan but. not write-InAsp. “but I did not finish it.”. (10) (Soh and Kuo 2001) a. Ta He. chi-le. liang-ge dangao, *keshi. eat-ASP two-CL cake,. but. mei chi-wan not eat-InAsp. “He ate two cakes, *but he did not finish eating them.” b. Ta He. 2. kan-le. liang-ben. read-ASP two-CL. shu, *keshi. mei kan-wan 3. book, but. not read-InAsp. The example in (9a) can be plausibly [+telic] if yi-feng xin ‘one letter’ is replaced with liang-feng xin ‘two. letters’. This phenomenon may be caused by the special effect by one-CL, which can be indefinite but also specific or referential based on contexts. 3. The first sentence in (10b) can have the [0-telic] reading for 7 out of 10 informants in my survey.. 38.

(40) On Telicity Selection in Mandarin Chinese “He read two books, *but he did not finish reading them.”. The first sentence in (9) is [0-telic] and the second sentence makes (9) [-telic]. Based on (9), xie ‘write’ is an activity verb instead of an accomplishment verb since xie cannot select the [+telic] reading by itself; on the contrary, Soh and Kuo’s (2001) examples in (10a-b) with chi ‘eat’ and kan ‘read’ are [+telic]. Moreover, Lin (2004) doubts the acceptability of (9) 4, and he argues that in Chinese an event should be [+telic] when the object is indefinite according to (10a-b). However, Soh and Kuo have examples in their paper later in 2005 that provide agreement with Tai’s (1984) example in (9), as shown in (11a-b) 5.. 4. I conducted a survey on Tai’s (1984) example shown in (9) in order to examine the acceptability for Chinese. native speakers. The result was that 10 out of 15 informants accepted (9). Besides, Soh and Kuo (2005) agree with the acceptability of (9). Therefore, I will regard (9) as acceptable in this paper. On the other hand, the unacceptability raised by Lin (2004) may be caused by dialectal differences and it is common to see that there are different readings in Chinese examples in studies on Chinese linguistics. 5. Soh and Kuo (2005) mainly focus on the linguistic phenomena rather than the explanations so they do not. further explain the reason why examples shown in (11a-b) are acceptable while (10a-b) in Soh and Kuo (2001) are not.. 39.

(41) On Telicity Selection in Mandarin Chinese (11) a. Ta he. hua-le. yi-fu. hua. draw-ASP. one-CL picture. “He drew a picture/a circle,” (keshi mei hua-wan) (but. not paint-InAsp). “(but he didn’t finish it.)” b. Ta he. xie-le. yi-feng. xin. write-ASP. one-CL letter. “He wrote a letter/a character,” (keshi mei xie-wan) (but. not write-InAsp). “(but he didn’t finish it.)”. The first sentences in (11a-b) are [0-telic] since both the picture and the letter are not absolutely finished. Soh and Kuo’s (2005) examples show that hua ‘draw’ and xie ‘write’ cannot select the [+telic] reading although the objects are indefinite. Consequently, there is space for reconsideration for Lin’s (2004) suggestion that Chinese verbs can select the [+telic] reading with indefinite objects. Also, Lin (2004) does not further discuss the semantic difference between xie ‘write’ in Tai (1984) and chi ‘eat’ and kan ‘read’ in Soh and Kuo 40.

(42) On Telicity Selection in Mandarin Chinese (2001). With the consideration that different verbs may result in different readings, I will lead a further investigation into whether the semantic nature of a verb together with the definiteness of objects have effect on the selection of [±telic] in Chapter 3.. 4.. Summary. In this chapter, I clarify the difference between the semantic realization and [±telic] in Chinese. The main function of the Chinese aspectual marker le is to bear the realization of an event rather than to indicate the [+telic] reading. Moreover, I also discuss whether there are accomplishment verbs in Chinese and my conclusion is negative, as suggested by Tai (1984). Therefore, for Chinese speakers, in order to express the [+telic] reading, especially when the event is [+stage], certain linguistic forces other than verbs must play crucial roles in the selection of the [+telic] reading. As a result, I will propose a detailed framework for the selection of [+telic] in Chapter 3.. 41.

(43) On Telicity Selection in Mandarin Chinese. CHAPTER THREE EMPIRICAL ANALYSES. 1.. Overview. According to Lin (2003), [+present] selects [-telic] and [+past] selects [+telic]. Partly agreeing with Lin’s proposal, in this thesis, I will give empirical evidence that supports my proposal: Although [+present] selects [-telic], [+past] does not necessarily contribute to [+telic]. In particular, in Chinese, if the tense is [+past], certain semantic and syntactic factors other than tense select [+telic], such as the definiteness / semantic restriction ([±def]) of the object, the semantic nature of the main verb ([±consume]), and overt Chinese inner aspectual markers (InAsp) such as hao, wan, diao, and guang. In Section 2, I will show my preliminary observation of the factors that select Chinese [±telic], and the effect of objects will be reviewed in Section 3. In Section 4, [+achieve] will be attributed to be still another factor that selects Chinese [±telic]. Besides, the behavior of Chinese InAsp will be discussed in Section 5. Finally, a summary and a brief conclusion will be provided in Section 6.. 2.. The Selection of Telicity in Chinese. First of all, as illustrated in (1a-d), [+consume] and [-def] together selects [+telic]. When 42.

(44) On Telicity Selection in Mandarin Chinese the main verb is [+consume] and the object is [+def], [±telic] of the event is underspecified ([0-telic]).. (1) a. Wo I. chi-le. yi-pan. cai (*keshi. eat-LE. one-CL dish (*but. hai-mei chi-wan) yet-not. “I ate one dish, (*but I did not finish eating it).” b. Wo. chi-le. na-pan. cai (keshi. I. eat-LE. that-CL dish (but. I. he-le. yi-bei. yet-not. jiu. drink-ASP. one-CL wine. [+consume] [-def] [+telic]. hai-mei chi-wan) 1. “I ate that dish, (but I did not finish eating it.)” c. Wo. eat-InAsp). eat-InAsp) [+consume] [+def] [0-telic]. (*keshi. hai-mei he-wan). (*but. yet-not. drink-InAsp). “I drank one cup of wine, (*but I did not finish drinking it.)[+consume] [-def] [+telic] 1. In Chinese, na-pan ‘that-CL’ can also be said as na-yi-pan ‘that-one-CL’. Na-yi-pan ‘that-one-CL’ is [+def]. and has the same effect of making [±telic] underspecified with a [+consume] verb. As shown in (i).. (i). Wo. chi-le. na-yi-pan cai. I. eat-LE. that-CL. (keshi. dish (but. “I ate that dish, (but I did not finish eating it.)”. hai-mei. chi-wan). yet-not. eat-InAsp) [+consume] [+def] [0-telic]. Due to this fact, na-yi-CL will not be discussed respectively in the following data.. 43.

(45) On Telicity Selection in Mandarin Chinese d. Wo I. he-le. na-bei. jiu. drink-ASP. that-CL wine. (keshi. hai-mei he-wan). (but. yet-not. drink-InAsp). “I drank that cup of wine, (but I did not finish drinking it.) [+consume] [+def] [0-telic]. As can be observed, (1a-d) are [+past], but (1a, c) have to be [+telic] while (1b, d) do not. Since the only semantic difference between (1a, c) and (1b, d) is [±def], [±def] should be the key factor of the selection of [±telic] in (1a-d). As a result, (1a-d) evidence that [+consume] [-def] together selects [+telic] 2. 2. Based on my observation, the semantic emphasis of the object also blocks the selection of [±telic] although. the main verb is [+consume] and the object is [-def], as demonstrated in (iia-b).. (ii). a. Wo I. chi-le yi-kuai. te-da-hao. niupai. eat-ASP one-CL. super-big-size. steak. “I ate one super-large steak,” (chi-le. san. xiaoshi. (eat-ASP three hour. hai. chi-bu-wan). still eat-not-InAsp). “and I could not finish eating it even three hours later.” b. Wo I. [+consume] [+def] [0-telic]. he-le. yi-bei chaoji. nong. de. jiu. drink-ASP. one-CL super. concentrated. DE. wine. “I drank one cup of wine with super-high alcohol concentration.” (he-le. yi. xiaoshi. hai-mei. he-wan). (drink-ASP. one. hour. still-not. drink-InAsp). “(and I could not finish drinking it even after an hour.)”. [+consume] [+def] [0-telic]. Yi-kuai te-da-hao niupai ‘a super-sized steak’ in (iia) is semantically emphasized since the size is. 44.

(46) On Telicity Selection in Mandarin Chinese Besides, the way of examination of [±telic] by Vendler (1967) also provides supporting evidence for the [+telic] readings of (1a-d), as shown in (2a-d). Based on Vendler (1967), the temporal phrase in a period of time, such as in an hour or in three minutes, can only be combined with [+telic] events.. (2) a. Wo I. san. fenzhong. nei. chi-le. yi-pan. cai. three. minute. within. eat-ASP one-CL dish. [+telic]. “I ate one dish in three minutes.” b. Wo I. san. fenzhong. nei. chi-le. na-pan. cai. three. minute. within. eat-ASP that-CL dish. [+telic]. “I ate that dish in three minutes.” c. Wo I. san. fenzhong nei. he-le. yi-bei. three. minute. drink-ASP. one-CL wine. within. jiu. [+telic]. “I drank one cup of wine in three minutes.” emphasized; on the other hand, in (iib), yi-bei chaoji nong de jiu ‘a cup of wine with super-high alcohol concentration’ is as well semantically emphasized since the concentration of the alcohol is emphasized. In both examples, [+consume] and [-def] seem not to select the [+telic] reading as suggested in (iia-d); however, this effect is caused by daily experience in real life rather than linguistic factors. For a normal human being, it is really difficult to eat up a steak in the super-large size, say as large as a basketball. Therefore, I will not consider examples like (iia-b) in my paper.. 45.

(47) On Telicity Selection in Mandarin Chinese d. Wo I. san. fenzhong nei. he-le. na-bei. three. minute. drink-ASP. that-CL wine. within. jiu. [+telic]. “I drank that cup of wine in three minutes.”. As can be indicated, (2a, c) can be perfectly combined with temporal phrases in three minutes; on the other hand, (1b, d) are originally [0-telic], but with in three minutes in (2b, d), only the [+telic] readings emerge. Secondly, [-consume] has an effect on blocking the selection of [±telic], as shown in Soh and Kuo (2005), as demonstrated in (4a-b). As shown in Chapter one, I restate the definition of [+consume] in (3).. (3) [+consume] A verb is [+consume] iff the action of the verb makes an element x disappear from a given possible world.. 46.

(48) On Telicity Selection in Mandarin Chinese (4) (Soh and Kuo 2005) a. Ta he. hua-le. yi-fu hua. /. ?yi-ge. quan-quan,. draw-ASP. one-CL picture. /. one-CL circle. “He drew a picture/a circle,” (keshi mei hua-wan) (but. not paint-InAsp). “(but he didn’t finish it.)” b. Ta he. [-consume] [-def] [0-telic]. xie-le. yi-feng. xin /. write-ASP. one-CL letter /. ?yi-ge. zi,. one-CL character. “He wrote a letter/a character,” (keshi mei xie-wan) (but. not write-InAsp). “(but he didn’t finish it.)”. [-consume] [-def] [0-telic]. Yi-ge quan-quan ‘one circle’ in (4a) and yi-ge zi ‘one word’ in (4b) are awkward since nobody normally says, “I drew a circle.” or “I wrote a word” while the circle or the word is only partially completed in daily life. For example, if I write the word calculus as calc, none will say that I wrote “a word”. As the result, the unacceptability of incomplete yi-ge quan-quan ‘one circle’ or yi-ge zi ‘one word’ does not result from linguistic factors but from 47.

(49) On Telicity Selection in Mandarin Chinese the experience in daily life or world knowledge. If non-linguistic factors are removed, (4a-b) demonstrate that [-consume] verbs such as xie ‘write’ and hua ‘draw’ block the selection of [±telic] and overrides the effect of [-def] objects. Besides, I also classify some Chinese verbs, such as xue ‘learn’ and xiazai ‘download’, as [-consume] due to the fact that learning is to copy the knowledge from books or TV programs to one’s brain without resulting in any disappearance of some element in a given possible world; on the other hand, downloading is to copy files from websites without making anything disappear from the Internet. Therefore, xue ‘learn’ and xiazai ‘download’ are categorized as [-consume] based on the definition in (3), and the two verbs also block the selection of [±telic], as shown in (5a-b).. (5) a. Wo I. xue-le. yi-zhong ruanti. learn-ASP. one-CL software. “I learned (how to operate) one software,” (xue-le. liang. (learn-ASP two. nian hai-mei xue-hao) year yet-not. learn-InAsp). “(but I did not successfully learn it after two years.)” 48. [-consume][-def] [0-telic].

(50) On Telicity Selection in Mandarin Chinese b. Wo I. xiazai-le. liang-ge dangan. download-ASP. two-CL file. “I downloaded two files,” (xiazai-le. yi. xiaoshi. (download-ASP one hour. hai-mei xiazai-wan) yet-not. download-InAsp). “(but I did not finish downloading it after an hour.)”. [-consume][-def] [0-telic]. As can be observed, (5a) xue-le yi-zhong ruanti ‘learn one software’ does not indicate the accomplishment of full-acquiring the knowledge and technique of the software; on the contrary, the event can just be an experience of leaning, and therefore the event in (5a) is [0-telic] instead of definitely [+telic]. Similarly, (5b) xiazai-le liang-ge dangan ‘download one file’ does not imply that the file is fully downloaded; in fact, the event in (5b) can just indicate that the file is still undergoing the process of downloading, and therefore (5b) is also [0-telic]. As shown above, xue ‘learn’ and xiazai ‘download’ are [-consume] based on the definition of (3) and both of the verbs have the influence on the blocking of the selection of [±telic]. Still, some Chinese verbs that are [-consume] also show the force to block the selection of [±telic], such as jiancha ‘examine’ and zhunbei ‘prepare’. Based on (3), jiancha ‘examine’ and zhunbei ‘prepare’ are [-consume] since the two 49.

(51) On Telicity Selection in Mandarin Chinese verbs do not entail any action to remove elements being examined or prepared for. These two verbs also block the selection of [±telic], as demonstrated in (6a-b).. (6) a. Wo I. jiancha-le. zaoshang de. mao. examine-ASP morning DE cat. “I examined the cats sent this morning.” b. Wo I. zhunbei-le. mingtian. prepare-ASP tomorrow. de. [-consume] [0-telic] kaoshi. DE test. “I prepared for the test tomorrow.”. [-consume] [0-telic]. In (6a), jiancha-le yi-zhi mao ‘examine the cats’ means that an animal doctor has operated the process of a physical examination of some cats, but the event does not definitely indicate that the process has been fully finished; consequently, the event of (6a) is [0-telic]. Also, zhunbei-le mingtian de kaoshi ‘prepare for tomorrow’s test’ in (6b) implies that the student has studied the subject, such as mathematics, for the test tomorrow, but the event does not absolutely mean that the student has finished reviewing all the sections in the textbook that the test will cover tomorrow; as a result, (6b) is also [0-telic]. As discussed above, jiancha ‘examine’ and zhunbei ‘prepare’ block the selection of [±telic]. So far, my analyses provide the generalizations stated in (7a-c). 50.

(52) On Telicity Selection in Mandarin Chinese (7) a. [+consume] [-def] selects [+telic]. b. [+consume] [+def] blocks the selection of [±telic]. c. [-consume] blocks the selection of [±telic].. Moreover, above all the forces in (7a-c), overt Chinese InAsp, such as hao, wan, diao, and guang, select [+telic] and overrides the effect of (7a-c), as demonstrated in (8a-d).. (8) a. Wo I. xie-hao-le. yi-feng. xin. write-InAsp-ASP. one-CL letter. “I finished writing one letter.” b. Wo I. [-consume] [+telic]. jiancha-wan-le. na-zhi. mao. examine-InAsp-ASP. that-CL cat. “I finished examining that cat.” c. Wo I. [-consume] [+telic]. chi-diao-le. na-kuai dangao. eat-InAsp-ASP. that-CL cake. “I ate that cake.” d. Wo I. hua-guang-le. [+consume] [+def] [+telic] naxie. spend-InAsp-ASP those. qian money. “I spent all of the money.”. [+consume] [+def] [+telic] 51.

(53) On Telicity Selection in Mandarin Chinese. In (8a), the main verb is xie ‘write’ is [-consume], which should block the selection of [±telic]; however, the letter must be completed since there is InAsp hao. In another case, wan in (8b) forces the event to be [+telic] with the main verb being [-consume], which should lead (8b) to be [0-telic]. In (8d), diao indicates the fact that the cake has been fully eaten while the [+consume] verb chi ‘eat’ and the [+def] object na-kuai dangao ‘that cake’ should together block the selection of [±telic]. Similarly, guang in (8d) indicates that all the money must have been spent although the main verb hua ‘spend’ is [+consume] and the object is [+def]. In sum, [+present] select [-telic], as proposed by Lin (2003), but different from Lin (2003), [+past] does not select [+telic]; instead, other semantic and syntactic factors discussed above select [±telic] when the tense is [+past]. My current conclusion is illustrated in (9a-d).. (9) The Generalization of the Selection of Chinese Telicity. (First Version). When the tense is [+past]: a. Overt Chinese InAsp, such as hao, wan, diao, and guang, select [+telic]. When there is no overt InAsp: b. [-consume] blocks the selection of [±telic]. When the main verb is [+consume]: 52.

(54) On Telicity Selection in Mandarin Chinese c. [-def] selects [+telic]. d. [+def] blocks the selection of [±telic].. In the following, I will further investigate the effect of objects and verbs on the selection of Chinese [±telic].. 3.. The Effect of Objects. The direct argument of a predicate, as suggested by Verku yl (1972), has an influence on the selection of telicity. The English examples are shown in (10a-f).. (10) a. Mary ate a banana.. [+telic]. b. Mary ate some bananas.. [+telic]. c. Cindy wrote a letter.. [+telic]. d. Cindy wrote some letters.. [+telic]. e. Mary ate bananas.. [-telic]. f. Cindy wrote letters.. [-telic]. In (10a-d). the events are [+telic] while those in (10e-f) are [-telic]. Based on Verku yl (1972), direct arguments like that in (10a-d) have a certain property, specified quantity of A’, 53.

(55) On Telicity Selection in Mandarin Chinese to license the selection of [+telic]. On the contrary, as proposed by Chierchia (1995), the direct arguments in (10e-f), which are a bare plural, assign a universal interpretation to the predicate; therefore, (10e-f) only have the meaning that the individual had a habit of eating bananas / writing letters in the past or before she died, so the events cannot be [+telic]. Similarly, as demonstrated in Section 2.1, Chinese direct arguments also affect the selection of [±telic], as shown in (11a-c).. (11) a. Wo I. chi-le. yi-kuai. dangao. [+telic]. eat-ASP one-CL cake. “I ate one piece of cake.” b. Wo I. chi-le. yixie. eat-ASP some. dangao. [+telic]. cake. “I ate some cake.” c. Wo I. chi-le. dangao. [0-telic]. eat-ASP cake. “I ate (certain) cake.”. Both direct arguments in (11a-b), yi-kuai dangao ‘one piece of cake’ and yi-xie dangao ‘some cake’, are considered with the property licensing the selection of [+telic] in Verku yl’s 54.

(56) On Telicity Selection in Mandarin Chinese (1972) proposal, and (11a-b) are [+telic] as expected; on the other hand, (11c) is [0-telic] because of the effect of [+def]. However, when the main verb is [-consume] in Chinese, the results are different from those in (11a-c), as illustrated in (12a-c).. (12) a. Wo I. xie-le. yi-feng. xin. write-ASP. one-CL letter. “I wrote one letter.” b. Wo I. [0-telic]. xie-le. yixie. xin. write-ASP. some. letters. “I wrote some letters.” c. Wo I. xie-le. xin. write-ASP. letter. [0-telic]. “I wrote (certain) letters.”. [0-telic]. Although (12a-b) should be [+telic] according to Verkuyl’s (1972) proposal, they are [0-telic] instead. As can be observed, Chinese has another factor that overrides the effect of Verku yl’s (1972) property, and I suppose that the factor is the semantic nature of the main verb, as discussed in Section 2.1. Since the main verb xie ‘write’ is [-consume], (12a-c) are 55.

(57) On Telicity Selection in Mandarin Chinese expected to be [0-telic] based on my previous proposal. Briefly speaking, the force of [-consume] is stronger than that of the object in the selection of [±telic] in Chinese. In addition, still another linguistic force, [+achieve], has significant influence on the selection of [±telic], as demonstrated in next section.. 4.. On Chinese Achievement Verbs. Still another issue that I will further investigate is that when the Chinese main verb is a [+achieve] verb 3, such as ying ‘win’ si ‘die’, the event is always [+telic], as shown in (13a-b).. (13) a. Wo I. ying-le. na-chang. bisai. win-ASP. that-CL. game. “I won the game.” 3. [+consume] [+def] [+telic]. In my 2012 conference presentation at Student Workshop on East Asian Linguistics (SWEAL-2012), some. participants raised the idea that gua ‘hang’ and tun ‘swallow are [+achieve] verbs in Chinese. In my response, gua and tun are not real [+achieve] since both of the verbs are [+stage]. In fact, when you hang something, such as a coat, you need to at least spend some milliseconds even if you hang the coat in an instance; likewise, when you swallow something, such as a pill, it also takes some time. In contrast,the typical [+achieve] verbs ying ‘win’ and si ‘die’ do not even take any measurable time before reaching a resultative point. Although not very obvious, both of the actions of gua and tun undergo a process, and therefore they cannot be categorized as [+achieve]. Instead, gua and tun are activity verbs because they are [+stage] [-telic].. 56.

(58) On Telicity Selection in Mandarin Chinese b. Na-zhi that-CL. gou. qu-nian si. le. dog. last-year die ASP. “The dog died last year.”. [+consume] [+def] [+telic]. The action of ying ‘win’ in (13a) and si ‘die’ in (13b) must be achieved when the sentences are uttered since the two actions have to be finished without any stage; therefore, the events in (13a-b) are [+telic]. With the consideration of the effect of [+achieve], I revise my generalization in (9a-e) as in (14a-e). (14) The Generalization of the Selection of Chinese Telicity When the tense is [+past]: a. [+achieve] selects [+telic]. When the main verb is [-achieve]: b. Overt Chinese InAsp determinates [+telic]. When there is no overt InAsp: c. [-consume] blocks the selection of [±telic]. When the main verb is [+consume]: d. [+consume] [-def] selects [+telic]. 57. (Modified Version).

(59) On Telicity Selection in Mandarin Chinese e. [+consume] [+def] blocks the selection of [±telic].. 5.. The Co-occurrences of Chinese Inner Aspectual Markers and Verb Types. There are restrictions on the co-occurrences of wan, hao, diao, guang and verb types. The verb types co-occurring with wan, hao, diao, guang is shown in (15-18), and demonstrated in (19a-d), (20a-d), (21a-d), (22a-d) respectively.. (15) The Verb Types Co-occurring with Wan : A verb can co-occur with wan iff it is [+stage] 4.. 4. In some cases, the verb sha ‘kill’, which is usually classified as [-stage], can co-occur with wan, as in (iii) and. (iv).. (iii). Wo. ba. zhe-xingqi de. ren. dou sha-wan. le. I. BA. this-week DE. people. DOU kill-InAsp ASP. “I killed all the people arranged this week.”. (iv). Wo. ba. yu. dou. I. BA. fish DOU. sha-wan. le. kill-InAsp ASP. “I finished handing the fish.”. (iii) can be uttered by a professional killer after s/he accomplished his or her job of killing a certain number of people for this week. In this special case, the action of sha is actually [+stage] since there is a process. If five people are the targets for this professional killer, s/he needs to kill them one by one before reaching the resultative point of finishing killing all the targets.. 58.

(60) On Telicity Selection in Mandarin Chinese Grammatical Examples: a. chi-wan eat-InAsp. yi-pan. cai. one-CL dish. “finish eating one dish” b. xie-wan. yi-ben. write-InAsp. shu. one-CL book. “finish writing one book” c. kan-wan. yi-ge. watch-InAsp. jiemu. one-CL show. “finish watching one show” d. du-wan. yi-fen. baozhi. read-InAsp one-CL newspaper “finish reading one newspaper”. Ungrammatical Examples:. e. *si-wan. yi-ge. die-InAsp. ren. one-CL person. “A person finishes dying.”. (iv) can be uttered when one finishing handling all the fish. This example surely takes an obvious process because one needs to remove the scales and visceral of the fish before reaching the resultative point of finishing handling the fish. The reasons why sha (iii) and (iv) can co-occur with wan is that sha can be [+stage] in some cases.. 59.

(61) On Telicity Selection in Mandarin Chinese f. *ying-wan yi-chang. bisai. win-InAsp one-CL. game. “finish winning one game”. (16) The Verb Types Co-occurring with Hao: A verb can co-occur with hao iff it is [-consume]. Grammatical Examples: a. xie-hao. yi-feng. write-InAsp. xin. one-CL letter. “finish writing one book” b. hua-hao. yi-fu. draw-InAsp. hua. one-CL picture. “finish drawing one picture” c. jiancha-hao. yi-ge. bingren. examine-InAsp one-CL patient “finish examining one patient” d. zhunbei-hao. yi-chang kaoshi. prepare-InAsp. one-CL exam. “finish preparing for one exam”. 60.

(62) On Telicity Selection in Mandarin Chinese Ungrammatical Examples:. e. *chi-hao eat-InAsp. yi-ge. dangao. one-CL cake. “finish eating one cake” f. *he-hao. yi-bei. drink-InAsp. jiu. one-CL vapor. “finish drinking one vapor” g. *Tun-hao swallow-InAsp. yi-ke. yao 5. one-CL pill. “finish swallowing one pill”. 5. The example tun-hao yi-ke yao ‘finish swallowing one pill’ may be acceptable if the agent is going to swallow. more than one pill, as shown in (v).. (v). Tun-hao. yi-ke. yao hou mashang. zai. swallow-InAsp. one-CL. pill. then swallow. later immediately. tun. di-er-ke number-second-CL. “Swallow one pill and then swallow the second one immediately.”. This phenomenon is very context-sensitive and only appears in situations like that in (v), and the acceptability of (v) is possibly influenced by Taiwanese Southern Min.. 61.

(63) On Telicity Selection in Mandarin Chinese (17) The Verb Types Co-occurring with Diao: A verb can co-occur with diao iff it is [+consume]. Grammatical Examples: a. chi-diao eat-InAsp. yi-kuai niupai one-CL steak. “finish eating one steak” b. he-diao. yi-ping futejia. drink-InAsp. one-CL vodka. “finish drinking one vodka” c. hua-diao. yi-bi. spend-InAsp. qian. one-CL money. “spend the money”. 62.

(64) On Telicity Selection in Mandarin Chinese Ungrammatical Examples:. d. *xie-diao 6 write-InAsp. yi-fen. xin. one-CL letter. “finish writing one letter” e. *xue-diao learn-InAsp. yi-zhong ruanti one-CL software. “finish learning one software”. (18) The Verb Types Co-occurring with Guang: A verb can co-occur with guang iff it is [+consume].. 6. Xie ‘write’ is usually classified as [-consume] but in some special cases can be [+concume]. As shown below,. xie-diao ‘write-InAsp’ is unacceptable in (v) but acceptable in (vi).. (vi). *Wo I. xie-diao. yi-ben. shu. write-InAsp. one-CL. book. “I wrote one book.” (vii). Wo. xie-diao. yi-zhi. bi. I. write-InAsp. one-CL. pen. “I used up one pen.”. In (vi), xie-diao means using up the ink of a pen through writing, so the action is in fact [+consume] and therefore can co-occur with diao.. 63.

(65) On Telicity Selection in Mandarin Chinese Grammatical Examples: a. chi-guang eat-InAsp. yi-guo. mala-guo. one-CL super.spicy-hotpot. “finish eating one super spicy hotpot” b. he-guang. yi-ping. drink-InAsp. kele. one-CL coke. “finish drinking one coke” c. hua-guang spend-InAsp. zhe- xie. qian. this-some. money. “finish spending this money” d. si-guang 7 die-InAsp “finish dying”. 7. Si ‘die’ is [-consume] since the body does not disappear when a creature dies. However, si-guang is acceptable. because the action is to consume a certain number of people or creatures, as demonstrated in (vii).. (viii). Bìng-ren. dou. si-guang. le. Sick-people. DOU. die-InAsp ASP. “All the patients died.”. (vii) can be uttered if and only if each patient in a certain area, such as a hospital, died. When the patients died one by one, the number of patients was consumed gradually. Consequently, this action is [+consume].. 64.

(66) On Telicity Selection in Mandarin Chinese Ungrammatical Examples:. *xie-guang yi-feng xin write-InAsp one-CL letter “finish writing one letter” *hua-guang. yi-fu. hua. draw-InAsp. one-CL picture. “finish drawing one picture”. The examples in (15-18) are illustrated in full sentences as followed.. (19) a. Wo I. xie-wan. na-feng xin. le. write-InAsp. that-CL letter. ASP. “I finished writing one letter.” b. Wo I. [+stage]. chi-wan. na-tiao. yu. eat-InAsp. that-CL fish. le ASP. “I ate that fish.”. [+stage]. c. Wo. zhunbei-wan mingtian. de. I. prepare-InAsp tomorrow. DE test. “I am prepared for the test tomorrow.” 65. kaoshi. le ASP [+stage].

(67) On Telicity Selection in Mandarin Chinese d. *Wo I. kan-wan. na-zhi. mao le. see-InAsp. that-CL cat ASP. “I saw that cat.”. (20) a. Wo I. [-stage]. xie-hao. na-feng xin. le. write-InAsp. that-CL letter. ASP. “I finished writing one letter.” b. *Wo I. [-consume]. chi-hao. na-tiao. yu. eat-InAsp. that-CL fish. le ASP. “I ate that fish.” c. Wo I. zhunbei-hao. [+consume] mingtian. prepare-InAsp tomorrow. de. DE test. “I am prepared for the test tomorrow.” d. *Wo I. kaoshi. kan-hao. na-zhi. see-InAsp. that-CL cat ASP. le ASP [-consume]. mao le. “I saw that cat.”. [-consume]. 66.

(68) On Telicity Selection in Mandarin Chinese (21) a. *Wo I. xie-diao. na-feng xin. le. write-InAsp. that-CL letter. ASP. “I finished writing one letter.” b. Wo I. [-consume]. chi-diao. na-tiao. yu. eat-InAsp. that-CL fish. le ASP. “I ate that fish.”. [+consume]. c. *Wo. zhunbei-diao mingtian. de. kaoshi. I. prepare-InAsp tomorrow. DE test. “I am prepared for the test tomorrow.” d. *Wo. kan-diao. na-zhi. I. see-InAsp. that-CL cat ASP. I. ASP [-consume]. mao le. “I saw that cat.”. (22) a. *Wo. le. [-consume]. xie-guang. na-feng xin. le. write-InAsp. that-CL letter. ASP. “I finished writing one letter.”. [-consume]. b. Wo. chi-guang. na-tiao. yu. I. eat-InAsp. that-CL fish. le ASP. “I ate that fish.”. [+consume] 67.

(69) On Telicity Selection in Mandarin Chinese c. *Wo. zhunbei-guang. mingtian. de. I. prepare-InAsp. tomorrow. DE test. “I am prepared for the test tomorrow.” d. *Wo. kan-guang. na-zhi. I. see-InAsp. that-CL cat ASP. kaoshi. le ASP [-consume]. mao le. “I saw that cat.”. [-consume]. In sum, each Chinese InAsp has its own restriction on its co-occurrence with different verb types, and the restrictions stated in (15-18) are actually related to the semantic nature of Chinese InAsp, as shown in (23-26).. (23) A verb can co-occur with wan iff it is [+stage]. Therefore: Wan implies that the action has reached its resultative point through a semantic stage.. (24) A verb can co-occur with hao iff it is [-consume]. Therefore: Hao implies that the action does not make anything disappear from a given possible world. 68.

(70) On Telicity Selection in Mandarin Chinese (25) A verb can co-occur with diao iff it is [+consume]. Therefore: Diao implies that the action of the verb makes something disappear from a given possible world.. (26) A verb can co-occur with guang iff it is [+consume]. Therefore: Guang implies that the action of the verb makes something disappear from a given possible world.. 6.. Summary. In this chapter, I show the evidence which indicate that [+past] does not necessarily select [±telic] in Chinese, different from Lin’s (2003) suggestion; instead, overt Chinese InAsp, such as hao, dao, wan, diao, and guang, have strong effect on the selection of [±telic], and also some semantic features, such as [±achieve], [±def], and [±consume], have influence on Chinese [±telic]. My brief conclusion of the selection of Chinese [±telic] is restated from (14a-e) in (27a-e).. 69.

(71) On Telicity Selection in Mandarin Chinese (27) The Generalization of the Selection of Chinese Telicity. (Final Version). When the tense is [+past]: a. [+achieve] selects [+telic]. When the main verb is [-achieve]: b. Overt Chinese InAsp determinates [+telic]. When there is no overt InAsp: c. [-consume] blocks the selection of [±telic]. When the main verb is [+consume]: d. [+consume] [-def] selects [+telic]. e. [+consume] [+def] blocks the selection of [±telic].. 70.

(72) On Telicity Selection in Mandarin Chinese. CHAPTER FOUR CONCLUSION. 1.. Overview. Since there is no genuine accomplishment verbs in Chinese, as suggested by Tai (1984) and discussed in Chapter 2, and Chinese aspectual marker le cannot select the [+telic] reading because le is a realization aspect (Liu 1988, Liao 2004) rather than a perfective aspect, which can be overtly indicated by InAsp (Travis 2006, Liao ms.). Consequently, there should be some linguistic forces other than verbs in Chinese for the purpose of determining the [+telic] reading. The primary conclusion of this thesis is to propose a theory of [±telic] in Chinese based on empirical [+telic] evidence; besides, a detailed discussion on Chinese InAsp is also included.. 2.. A Theory of Chinese Telicity. Lin (2003) generalizes the selection of [±telic], in which [+present] selects [-telic] while [+past] selects [+telic]. However, in my analyses of Chinese [±telic] in Chapter 3, [+present] naturally selects [-telic] events, as suggested by Lin (2003), but different from Lin (2003), [+past] does not have any effect on the selection of [±telic]; instead, the selection of [±telic] with [+past] depends on semantic features of objects and verbs or on Chinese overt InAsp, as 71.

(73) On Telicity Selection in Mandarin Chinese demonstrated in Chapter 3. In the end of this thesis, I organize my generalization in (27a-e) in Chapter 3 as in Figure 1.. TENSE [+present]. [+past]. [-telic]. The main verb is [+achieve].. YES. NO. [+telic] with overt Chinese InAsp YES. NO. [+telic] The main verb is [-consume]. YES [-consume] blocks the selection of [±telic].. [+consume] [-def] selects [+telic]. [+consume] [+def] blocks the selection of [±telic]. Figure 1 The Selection of Chinese Telicity 72. NO.

(74) On Telicity Selection in Mandarin Chinese. 3.. Linguistic Behavior of Chinese Inner Aspect. Moreover, as discussed in Chapter 3, each Chinese InAsp has clear co-occurrence with certain verb types, and the semantic nature of Chinese InAsp can be indicated by the rules of each co-occurrence. The generalizations of the co-occurrences of Chinese InAsp and verbs are restated as shown in (1-4).. (1) The Verb Types Co-occurring with Wan : A verb can co-occur with wan iff it is [+stage].. (2) The Verb Types Co-occurring with Hao: A verb can co-occur with hao iff it is [-consume].. (3) The Verb Types Co-occurring with Diao: A verb can co-occur with diao iff it is [+consume].. (4) The Verb Types Co-occurring with Guang: A verb can co-occur with guang iff it is [+consume].. 73.

(75) On Telicity Selection in Mandarin Chinese The semantic nature of each Chinese InAsp discussed in my thesis is shown in (5-8).. (5) A verb can co-occur with wan iff it is [+stage]. Therefore: Wan implies that the action of the verb has reached its resultative point through a semantic stage.. (6) A verb can co-occur with hao iff it is [-consume]. Therefore: Hao implies that the action does not make anything disappear from a given possible world.. (7) A verb can co-occur with diao iff it is [+consume]. Therefore: Diao implies that the action of the verb makes something disappear from a given possible world.. 74.

參考文獻

相關文件

printing, engraved roller 刻花輥筒印花 printing, flatbed screen 平板絲網印花 printing, heat transfer 熱轉移印花. printing, ink-jet

Teachers may consider the school’s aims and conditions or even the language environment to select the most appropriate approach according to students’ need and ability; or develop

(1) Western musical terms and names of composers commonly used in the teaching of Music are included in this glossary.. (2) The Western musical terms and names of composers

Through a critical examination of some Chinese Christian intellectuals’ discussion on the indigenization of Christianity in China, this paper attempts to show that Chinese

DVDs, Podcasts, language teaching software, video games, and even foreign- language music and music videos can provide positive and fun associations with the language for

assessment items targeting the following reading foci: specific information, inferencing, main ideas. What syntactic and/or semantic clues would you identify in the text to guide

LTP (I - III) Latent Topic Probability (mean, variance, standard deviation).. O

•It directly models prior semantic knowledge units, which enhances the ability to learn semantic representation?. • ERNIE learns the semantic representation of complete concepts by