臺語KA字句之第一語言習得

全文

(1)摘要本研究探討閩南語 ka 字句的第一語言習得。目的是要了解閩南語 ka 字句的四大屬性，ka 名詞組的指涉性、ka 名詞的省略性、ka 動詞的動作性，以及現在進行標誌的使用在學齡前兒童母語習得中所扮演的角色。主要的研究議題包含屬性效應、副屬性效應、題型效應、非典型作答模式，以及年齡效應。本研究共設計兩個測驗：文法判斷測驗與看圖回答測驗。研究參與者為 64 位以閩南語為母語的人士，包含 48 位平均年齡為 4 至 6 歲的兒童及 16 位成人。研究結果顯示，不同的屬性、副屬性、題型及年齡，皆影響閩南語 ka 字句的習得。ka 字句的四大屬性對實驗對象而言難易程度有所不同：ka 動詞的動作性最簡單，ka 名詞組的指涉性次之，現在進行標誌的使用再次之，ka 名詞的省略性難度最高。此次序可用 Pinker (1989)的保守理論來解釋。在副屬性效應方面，本研究結果支持 ka 字句的理論性研究，實驗參與者在一般指涉和定指名詞表現優於非定指名詞、非狀態動詞優於狀態動詞、無目的性動詞優於有目的性動詞。根據這些結果，我們更進一步發展 Pinker (1989)的保守理論，提出在習得區別出可出現和不可出現在某特定句型的元素的過程中，本研究之學齡前兒童似乎歷經三個發展階段：階段 1: 無區別階段 2: 部分區別階段 3: 完全區別。在題型效應方面，基本上受試者在文法判斷測驗表現優於看圖回答測驗，支持理解先於口語能力的看法 (Oviatt 1980, Brown 1987, Gerken and Shady 1996)。最後，觀察各個年齡層的表現，我們發現 5 歲是閩南語 ka 字句的轉變階段，然而直到 6 歲時，ka 字句發展才達到完全成熟。. i.

(2) ABSTRACT The present study aims to explore children’s acquisition of the Taiwanese ka construction.. Four major constraints, i.e., Referentiality, Omission, Dynamicity, and. Progressive, were examined in the present study.. Property effects, sub-property. effects, task effects, age effects and other related patterns were investigated in this pioneer research.. A comprehension task (Grammaticality Judgment task, the GJ task). and a production task (a Picture-cued Production task, the PP task) were employed. Forty-eight preschoolers in Xinying City, Tainan County and sixteen adults were asked to participate in the experiment. The preschoolers were further divided into three age groups, ranging from 4 to 6. The findings suggested that main properties, sub-properties, tasks, and age were all crucial factors in the acquisition of the Taiwanese ka construction.. The four. constraints were found to impose a different degree of difficulty on the subjects: Dynamicity > Referentiality > Progressive > Omission (from easy to difficult).. The. hierarchical sequence supports Pinker’s (1989) Theory of Conservativism. Concerning the sub-property effects, the present results argued for the previous theoretical analyses in that our subjects did prefer Nonstative to Stative, Definite/Generic to Indefinite, and Atelic to Telic.. The above findings showed that. in acquiring the contrast between elements permitted and prohibited in a certain construction, our children seemed to go through three developmental stages: Stage 1: No ContrastStage 2: Partial ContrastStage 3: Full Contrast.. Moreover, it was. found that our children performed better on the GJ task than on the PP task, supporting the view that comprehension is prior to production (Oviatt 1980, Brown 1987, Gerken and Shady 1996).. In addition, it was found that age 5 was a. transitional point for the acquisition of the Taiwanese ka construction and age 6 was the cutting age where our children achieved an adult-like grammar. ii.

(3) ACKNOWLEDGEMENTS This thesis would not have been completed with the guidance and assistance from my teachers, subjects, friends, and family. First, I would like to show my deepest gratitude to my dearest advisor, Prof. Chun-yin Doris for her academic guidance, spiritual support and sincere concern throughout my thesis writing.. Ever since I was in her class for the first time, her. patience, optimism, and the most important one, excellent sense of humor had rooted in my mind and deeply affected me. concern for my daily life.. I am especially grateful to her for her genuine. Whenever I feel frustrated or pessimistic, she is always. ready to cheer me up and always offers me great advice.. Words cannot express my. deepest gratitude to her, but still, I want to say thank you to her! My sincere gratitude also goes to my committee members, Prof. Miao-ling Hsieh and Prof. Ruei-ling Fahn.. They carefully read my thesis and offered me insightful. and valuable comments and suggestions.. Without their help, I would have had a lot. of difficulty in revising this thesis. Very special thanks must go to Prof. Shu-kai Hsieh, who opened the world of Computational Linguistics (CL) to me. and innovative thoughts.. I enjoyed and benefited from his lectures. Especially, I want to thank him for willing to hire me and. tolerate me as his research assistant even though at the very beginning I had no idea about what Computational Linguistics is.. His great patience and support help me to. survive my thesis writing. I also want to thank the enthusiastic participation of those cute children in the Xinjin Elementary School Kindergarten, Xintai Elementary School Kindergarten, and Xintianshi Kindergarten in Xinying City, Tainan County.. Also, I want to thank the. teachers there for their kindness and support through the whole data collection process. Special thanks go to Shu-mei Li, a teacher of Xintai Elementary School Kindergarten, iii.

(4) who not only kindly let me conduct the experiment there, but also help me to arrange the experimental time and place. Without her great help, it is impossible for me to finish the data collect in two weeks! I also owe many thanks to Hui-jun Yuan in the periodical room, Xuan-yun Chen in the English Department office, and Li-shu Ceng in the English Teaching Center for their constant encouragement during my thesis writing and their warm support and help with my work. I would like to extend my appreciation to my friends.. First, thanks go to my old. friends, Ya-qing Chen for tolerating me every weekend, Ju-fen Zhuang for cheering up at every stage of my thesis writing, and Shu-min Chen for providing me a good laugh.. Thanks also go to my dear friends from NTNU for their love, assistance, and. encouragement over the years of graduate study: David Shen, Sandra Huang, Shirley Geng, Ali Ge, Amanda Zhang, Wan-lin Xu, Catherine Hung, Zhao-ying Lu, Julia Wu, Cathy Gao, Terrence Lai, Roy Li, Dennis Xu. Finally, and the most importantly, I also like to show my deepest appreciation to my beloved family members.. Among them, my dearest brother, Shi-jie Zhao,. always shows great concern about me and provides great supports whatever happens. My adored little sister, Ya-ling Zhao, who helped me to draw the cute pictures of the study, always provides me a good laugh whenever I feel frustrated. My parents, Su-ying Chen and Shan-long Zhao, who have no idea about my major, have been supporting me unconditionally throughout every moment in my life. it is impossible for me to walk this far.. Without them,. Especially, I am grateful to my grandfather,. Zhu-mu Zhao, for his patience, support and love to me. first word in my life and led me to explore the world. teacher, and a friend. Without him, without me.. iv. He taught me to speak the. For me, he is a grandfather, a. To you, I dedicate this thesis..

(5) TABLE OF CONTENTS CHINESE ABSTRACT ................................................................................................ i ENGLISH ABSTRACT............................................................................................... ii ACKNOWLEDGEMENTS .......................................................................................iii TABLE OF CONTENTS ............................................................................................ v LIST OF TABLES ....................................................................................................viii LIST OF FIGURES .................................................................................................... ix LIST OF ABBREBIATIONS...................................................................................... x. CHAPTER ONE INTRODUCTION 1.1 Motivation ............................................................................................................ 1 1.2 Theoretical Background ....................................................................................... 3 1.2.1 Theory of Conservativism ............................................................................. 3 1.2.2 Age Effects .................................................................................................... 5 1.2.3 Task Effects ................................................................................................... 6 1.2.4 The Semantic Feature Hypothesis ................................................................. 7 1.2.5 Vendler’s (1967) Classification of Verbs ...................................................... 8 1.3 Research Questions .............................................................................................. 9 1.4 Significance of the Study ................................................................................... 10 1.5 Organization of the Thesis ................................................................................. 11 CHAPTER TWO LITERATURE REVIEW AND LINGUISTIC PROPERTIES OF TAIWANESE KA 2.1 Theoretical Studies of Taiwanese Ka ................................................................. 12 2.1.1 Teng (1982) ................................................................................................. 12 2.1.2 Li (1995) ...................................................................................................... 16 2.1.3 Hung (1995)................................................................................................. 21 2.1.4 Cheng and Tsao (1995) ............................................................................... 25. v.

(6) 2.1.5 Summary...................................................................................................... 27 2.2 A Comparison between Taiwanese Ka and Mandarin Ba.................................. 28 2.3 Properties of Ka in Taiwanese ........................................................................... 31 2.3.1 Referentiality Constraint on the Ka-NP ...................................................... 31 2.3.2 Omission Constraint on the Ka-NP ............................................................. 32 2.3.3 Dynamicity Constraint on the Ka-Verb ....................................................... 33 2.3.4 Progressive Constraint on the Ka-Verb ....................................................... 33 2.3.5 Summary...................................................................................................... 35 2.4 Experimental Studies of the Ba construction ..................................................... 35 2.4.1 Cheung (1992) ............................................................................................. 35 2.4.2 Fahn (1993) ................................................................................................. 40 2.4.3 Summary...................................................................................................... 43 2.5 Summary of Chapter Two .................................................................................. 43 CHAPTER THREE RESEARCH DESIGN 3.1 Subjects .............................................................................................................. 45 3.2 Methods and Materials ....................................................................................... 46 3.2.1 Materials of the Grammaticality Judgment Task ........................................ 48 3.2.2 Materials of the Picture-cued Production Task ......................................... 49 3.3 Procedures .......................................................................................................... 50 3.3.1 Pilot Study ................................................................................................... 50 3.3.2 The Study..................................................................................................... 52 3.3.3 Scoring and Statistical Analysis ................................................................ 56 3.4 Summary of Chapter Three ................................................................................ 56 CHAPTER FOUR RESULTS AND DISCUSSION 4.1 Property Effects .................................................................................................. 57 4.2 Sub-Property Effects .......................................................................................... 65 4.3 Task Effects ........................................................................................................ 76. vi.

(7) 4.4 Other Patterns Elicited ....................................................................................... 83 4.5 Age Effects ......................................................................................................... 89 4.6 Summary of Chapter Four .................................................................................. 91 CHAPTER FIVE CONCLUSION 5.1 Summary of the Major Findings ........................................................................ 92 5.2 Limitations of the Present Study and Suggestions for Future Research ............ 94 REFERENCES ........................................................................................................... 95 Appendix A: Task Design........................................................................................ 105 Appendix B: Scenarios Used in the PP Task ......................................................... 108 Appendix C: Test Items Used in the GJ Task ....................................................... 115 Appendix D: Statistic Data...................................................................................... 117. vii.

(8) LIST OF TABLES Table 1.1. Vendler’s (1967) Verb Classification ..................................................... 9. Table 2.1. Comparisons between Ka and Ba ......................................................... 31. Table 3.1. Demographics of the Subjects .............................................................. 46. Table 3.2. Test Items Examined in the GJ Task .................................................... 49. Table 3.3. Test Items Examined in the PP Task .................................................... 50. Table 3.4. A Sample Used in the PP Task ............................................................. 54. Table 3.5. A Sample Used in the GJ Task ............................................................. 55. Table 4.1. Subjects’ Correct Responses on the Four Constraints of the Taiwanese ka Construction (Average Means) ....................................... 57. Table 4.2. Subjects’ Correct Responses on the Sub-Properties of the ka Construction (Average Means) ............................................................. 66. Table 4.3. A Summary of the Statistic Significant Differences between the Experimental and Control Groups ........................................................ 67. Table 4.4. Subjects’ Correct Responses on the Sub-Properties of the ka Construction (Average Means) ............................................................. 68. Table 4.5. Spontaneous Speech Samples: Error Ratio ........................................... 70. Table 4.6. A Summary of the Statistic Significances for Each Age Group .................................................................................................... 71. Table 4.7. Subjects’ Correct Responses to the Two Tasks (Average Means) ....... 76. Table 4.8. Subjects’ Correct Responses on the Properties of the ka Construction in the Two Tasks (Average Means) .......................................................... 78. Table 4.9. Each Group’s Correct Responses on the Properties of the ka Construction in the Two Tasks (Average Means) ................................. 80. Table 4.10. Other Patterns Elicited in the PP Task .................................................. 84. Table 4.11. Other Patterns of Each Group Elicited in the PP Task ......................... 86. viii.

(9) LIST OF FIGURES Figure 2.1. Locative-Full (LF) ................................................................................. 36. Figure 2.2. Theme-Gone (TG) ................................................................................. 36. Figure 4.1. Each Age Group’s Correct Responses on the Properties of the ka Construction (Average Means) .............................................................. 59. Figure 4.2. Each Group’s Correct Response to the Two Tasks (Average Means) .. 77. ix.

(10) LIST OF ABBREVIATIONS ASP. Aspect marker. Ba. The morpheme of ba. CL. Classifier. DUR. Durative aspect marker. GJ. Grammaticality judgment task. Ka. The morpheme of ka. NP. Noun phrase. PP. Picture-cued production task. PT. Particle. SFH. Semantic Feature Hypothesis. teh. Present progressive marker ‘teh’. VP. Verb phrase. x.

(11) CHAPTER ONE INTRODUCTION 1.1 Motivation Issues concerning nature of the ba construction in Mandarin Chinese have received a wide range of discussion in the field of Chinese linguistics possibly because of its peculiar features and challenges it imposes on different theoretical frameworks, for example, the Theta Criterion (Chomsky, 1981) 1 (Chao 1968, Thompson 1973, Li 1974, Li and Thompson 1981, Ross 1991, Gao 1997, Li 2001). Roughly speaking, the ba construction in Mandarin Chinese is considered as a ‘disposal’ form, which conveys “how a person is handled, manipulated, or dealt with; how something is disposed of; or how an affair is conducted.” (Wang 1957, pp. 160-161, cited from Li 1974, pp. 200-201).. Like the Mandarin ba construction, the. ka construction is viewed as serving a similar function in Taiwanese Southern Min. During the past few decades, little attention has been paid to this structure partially because the ka construction is considered as a counterpart of the Mandarin ba construction and they share many peculiar features.. For example, like the Mandarin. ba construction, an indefinite NP cannot show up in the ka construction and a stative verb is impossible to occur in the ka construction. hit. king. pangking sau. KA that. CL. room. (1) a. Ka. sau. sweep sweep. le.. (definite NP). PT. ‘Clean that room!’. 1. According to the Theta Criterion, each argument NP must bear one and only one theta role. (i) Ta ba pingguo chi le liang ge. he BA apples eat ASP two CL ‘He ate two of the apples.’ In (i), liang ge ‘two’ is supposed to be a constituent of an NP. Also, if it is a constituent of the NP, the NP must have a theta role. In (i), the theta role is assigned by chi ‘eat’ since the NP is the object of the verb. In this way, pingguo ‘apples’ is left without a theta assigner because the verb chi ‘eat’ only has a theta role to assign. Thus, this sentence will be predicted ungrammatical by the Theta Criterion. 1.

(12) b. *Ka. cit. king. pangking sau. CL. room. KA a. sau. le.. (indefinite NP). sweep sweep PT. ‘Clean a room!’ (2) a. Muemue. (Li 1995, p. 13). ka. hue. tantiau.. the little sister KA flowers. (nonstative verb). throw away. ‘The little sister threw away the flowers.’ b. *Muemue the little sister. ka. hue. kahi.. KA flowers. (stative verb). like. ‘The little sister likes the flowers.’ However, though similar, significant differences are observed.. Recently some. studies have shown that the ka construction is an even more complicated structure than the ba construction and the ka construction has some very different syntactic and semantic properties from the ba construction (Teng 1982, Li 1995, Cheng and Tsao 1995, Hung 1995, Tsao 2003, Yang 2006).. For instance, a syntactically absent ka-NP,. and the occurrence of a progressive marker teh is sometimes permitted in the Taiwanese ka construction but never in the Mandarin ba construction (Teng 1982, Li 1995, Cheng and Tsao 1995, Hung 1995, Tsao 2003, Yang 2006). (3) a. Titi. oh, cang. the little brother PT, yesterday. mama. ka __ phah.. mother. KA. (Taiwanese). hit. ‘The little brother, Mother hit him yesterday.’ b. *Didi. o, zuotian mama ba __ da. the little brother PT, yesterday mother. BA hit. le yi dun. (Mandarin) ASP a. CL. ‘The little brother, Mother hit him yesterday.’ (4) a. Mama. teh. mother PROG. ka. i. phah.. (Taiwanese). KA him hit. ‘Mother is hitting him.’ 2.

(13) b. *Mama mother. zai. ba. ta. da.. (Mandarin). PROG BA him hit. ‘Mother is hitting him.’ The distinct properties of the ka construction thus motivate us to see if the same developmental trend found in the Mandarin ba construction can be applied to the Taiwanese ka construction.. 1.2 Theoretical Background In this section issues in language acquisition related to the present study will be discussed such as Theory of Conservativism, age effects and task effects.. In addition,. the present study assumes the Semantic Feature Hypothesis proposed by Clark (1971), and Vendler’s (1967) classification of verb types. The main notions of the Semantic Feature Hypothesis will be summarized briefly and then Vendler’s analysis will be reviewed in the following section, too.. 1.2.1 Theory of Conservativism To argue against Baker’s (1979) view that lexical structures are arbitrarily represented and thus have to be learned piecemeal, Pinker (1989) put forward the Theory of Conservativism.. He claims that lexical structures are not completely. arbitrary, and are learnable by rule. According to Pinker, there are three levels of semantic representations: i) The lexical specification of each individual verb ii) The narrow-range (NR) subclasses, which are the sets of verbs that share particular semantic meanings, and whose argument structures can in fact alternate. 3.

(14) iii) The broad-range (BR) classes, which are larger sets of verbs sharing general semantic meanings that all verbs participating in a particular syntactic alternations have in common; the semantic commonalities are presumably necessary for the syntactic alternations to occur The three levels are exemplified by English dative constructions.. The verbs. bring and pull both show up in the form of NP1 V NP2 to NP3 but only bring but not pull can alternate with the form of NP1 V NP3 NP2: (5) a. John brought the present to Mary. b. John brought Mary the present. (6) a. John pulled the chair to Mary. b.*John pulled Mary the chair. As shown in (5), bring can appear in the prepositional dative form as well as the double object dative form.. However, pull can only show up in the form of the. prepositional dative as in (6).. The three levels of representations for them are as. follows: (7) bring a. The verb level: X acts on Y and then Y goes with X to Z b. The BR level: cause to go c. The NR level: continuous force, no specific manner of motion, that is, one can bring by many different kinds of manner, including pulling (8) pull a. The verb level: X acts on Y and then Y goes to Z alone b. The BR level: cause to go c. The NR level: continuous force, specific manner of motion, that is, one can only pull by the manner of pulling Pinker argues that the alternations are primarily semantically driven. Once children 4.

(15) learn the semantic contrast between bring and pull, they will then apply it to the syntactic alternations straightforwardly.. For instance, children will learn the verb. level representations of bring and pull and other verbs at first, then distinguish the kinds of bring verbs from those of pull verbs by the existence of a specific manner or not and finally apply the contrast to the dative alternations. The implication for the theory is that children will not use the syntactic alternations (the BR level) before they acquire the semantic contrast between different kinds of verbs (the NR level) since the contrast is the element necessary for the syntactic alternations to occur.. In this sense, children are conservative learners.. However, once they master the contrast, they will immediately apply it to the construction.. According to the theory, we predict that children have to learn the. contrast between the element exhibited and that prohibited in the ka construction before they apply the contrast to the ka construction because only when they possess the semantic contrast can they apply the contrast to the ka construction.. 1.2.2 Age Effects It seems to be a natural tendency that age plays a crucial role in children’s language acquisition.. Results of many empirical studies have illustrated that. children’s language comprehension and production improve with age (Lenneberg 1967, Erbaugh 1982, Cheung 1992, Fahn 1993).. Lenneberg (1967) found that. children at the age of one could produce one word which may convey more than one meaning.. At the age of three, children could produce two words with. well-established semantic relations. were comprehensible.. And, most of the four-year-olds’ utterances. In a longitudinal study of children’s acquisition of the. Mandarin ba construction, Erbaugh (1982) found that the ba construction appeared in children’s speech as early as they were 2;3 years old. 5. However, the use of this.

(16) construction was not mastered by children until the age of 3;6.. Cheung (1992). conducted a statistical analysis on the longitudinal data provided in Tse, Tang, Shi, and Li’s study (1991) and indicated that children’s error ratio principally decreased with age. Fahn (1993) examined children’s acquisition of the ba construction and found that age five was a crucial cutoff point for certain constraints while age six was a demarcation point for the other constraints2.. Therefore, in this study age will be. one of the issues under investigation.. 1.2.3 Task Effects The other focus of this study is on task effects.. In order to examine children’s. linguistic competence and performance, both comprehension and production tasks will be designed in the present study.. Comprehension tasks are generally considered. as much easier than production tasks because it is assumed that children have to acquire the necessary linguistic knowledge before they can actually use it (Miller 1963, Oviatt 1980, Brown 1987, Gerken and Shady 1996, Wu and Wen 1997). In an investigation of children’s phonological development, Miller (1963) pointed out that a subject clearly perceived the contrast between English s and th, yet she could not produce the contrast herself.. Brown (1987) also stated that children may be. incapable of producing a sentence with an embedded relative clause in it but they may be able to understand it.. Wu and Wen (1997) found that their subjects’. performances on the grammaticality judgment task was better than on the translation task.. To verify this view, task effects will be discussed in this research.. 2. Fahn (1993) examined five constraints of the ba constraint: (1) The Progressive constraint: a progressive marker is impossible to occur in the ba construction; (2) the Verb Selection constraint: certain types of verbs cannot show up in the ba construction; (3) the Modifier constraint: certain verbs require modifiers to co-occur with the ba construction; (4) the Compound Verb constraint: certain compound verbs cannot show up in the ba construction; (5) the Definiteness constraint: the ba-NP cannot be indefinite. 6.

(17) 1.2.4 The Semantic Feature Hypothesis Based on a componential or feature analysis of words, Clark (1971) develops the theory of the Semantic Feature Hypothesis (SFH).. Two properties of the. componential analysis of words are emphasized in the theory: i) The meanings of words can be represented as a set of components arranged in a hierarchy. ii) Feature representations are given positive and negative values. For example, the temporal prepositions before, after and while can be represented in a feature hierarchy as follows: (9) Feature Representations of when, before and after. The first feature specification that comes into play is [+Time] which includes the words referring to some aspect of time. The next feature, Simultaneous, which is dominated by [+Time], has two values: [+Simultaneous] and [-Simultaneous]. Finally, the feature [+/-Prior] only occurs in words specified as [-Simultaneous]. These features are organized in a hierarchical order, with [+Time] dominating [+/-Simultaneous], and [-Simultaneous] dominating [+/-Prior]. According to the hypothesis, it is predicted that superordinate features in a feature hierarchy would be acquired first since these features are usually common to several different words and thus are general concepts, for instance, [+/-Time] or [+/-Simultaneous].. Also, plus values would be mastered before minus values 7.

(18) because the plus values are usually simpler or unmarked in a pair of relational terms. With regard to the temporal prepositions above, a distinction between before and after would be acquired later than the general distinction between before/after and when. Moreover, the precedence of positive over negative features expects that when would be mastered prior to before/after, and before prior to after. The psychological validity of SFH has been proved in children’s acquisition of relational terms.. Clark (1971) examined the acquisitional process of before, after. and when and found that when was mastered first, followed by before and finally after. Also, it was shown in Donaldson and Balfour’s (1968), and Donaldson and Wales’s (1970) studies that at the first stage, children seemed to treat more and less as synonyms.. They performed well on questions about more but they consistently. misinterpreted less as if it means more.. Clark claimed that the results were. explicable with SFH if it is supposed that children had mastered the feature [+Amount] which was shared by more and less but had not yet acquired the feature [+/-Polar] for differentiating the pair.. 1.2.5 Vendler’s (1967) Classification of Verbs Vendler (1967) classifies verbs into four types, that is, Stative, Activities, Accomplishments, and Achievements, as stated in Table 1.1:. 8.

(19) Table 1.1 Vendler’s (1967) Verb Classification Type. Example. 1. Statives3. Static, durative, atelic. know, love. 2. Activities. Dynamic, durative, atelic. run, walk. 3. Accomplishments. Dynamic, durative, telic. build a house, walk to school. 4. Achievements. Dynamic, instantaneous. recognize, find. Based on the discussion, States are verbs lacking continuous senses, for instance, know and love belong to this category. they do not possess continuous senses.. They can not have a progressive aspect since Activities such as run and walk are verbs. denoting processes going on in time with no terminal point.. Accomplishments are. verbs denoting processes going on in time with a terminal point, and after achieving the goal, the action can no longer continue. Verbs like build a house and walk to school are members of this type. Achievements are verbs that have a terminal point, but the time for reaching this point is instantaneous, for example, recognize and find are verbs of this type. In the present study it is assumed that the classification of verbs here is also applicable to Taiwanese.. Also, later in Chapter Four, we will develop the feature. specifications of Taiwanese verbs based on Vendler’s (1967) classification of verb types.. 1.3 Research Questions This study addresses the following four questions about the acquisition of the ka construction in Taiwanese:. 3. Vendler (1967) uses the term States for this type of verbs. term Statives. 9. However, in this study, we employ the.

(20) 1) What is the developmental sequence for the acquisition of various properties of the Taiwanese ka-construction? 2) Do children perform similarly on each property’s sub types? 3) Are there any age effects on children’s comprehension and production of the ka construction? 4) Do the children’s comprehension and production pattern alike in the acquisition of the ka construction?. 1.4 Significance of the Study The Mandarin ba construction has been recognized to be a very complicated syntactic structure in the literature. And the ka construction is considered to serve a similar function in Taiwanese.. However, children’s acquisition of the ka. construction in Taiwanese has not been fully investigated in literature. Previous studies of the ka construction usually limit their analyses to the theoretical aspect. The purpose of this study, hence, is to investigate the ka construction in Taiwanese from an acquisitional perspective. The Taiwanese ka construction has at least four functions: patient, source, goal and benefactive (Teng 1982, Cheng and Tsao 1995, Hung 1995, Tsao 2003). However, due to time limitation not all four functions will be investigated in this study.. Only ka as a patient marker will be examined.. The. reason is that ka as a patient marker seems to correspond to that of ba in Mandarin Chinese (Cheng and Tsao 1995, Hung 1995). This study is a pioneer study of the acquisition of the ka construction. The purpose is not to exhaust the subject but to bridge the gap between the theoretical and the experimental analyses in this field. Thus, only ka as a patient marker will be investigated.. 10.

(21) 1.5 Organization of the Thesis This thesis is organized as follows. Chapter Two reviews some previous theoretical and empirical studies of the Taiwanese ka construction, and discusses the properties of the construction. study.. Chapter Three reports the experimental design of this. Chapter Four presents and reports the experimental results.. Finally, Chapter. Five provides concluding statements and suggestions for further research.. 11.

(22) CHAPTER TWO LITERATURE REVIEW AND LINGUISTIC PROPERTIES OF TAIWANESE KA In this chapter previous analyses and linguistic properties of the ka construction in Taiwanese will be discussed. Also, experimental studies of the acquisition of Mandarin ba will be reviewed instead of Taiwanese ka since the two constructions are said to be similar and no empirical studies of the ka construction have been reported in the literature. examined.. In section 2.1 theoretical analyses of the ka sentences will be. Then, in section 2.2 comparisons between Taiwanese ka and Mandarin. ba will be discussed. In section 2.3 properties of the construction examined in the present study will be investigated followed by reviews of two empirical studies of the ba construction in section 2.4.. Finally, section 2.5 summarizes the main points of. this chapter.. 2.1 Theoretical Studies of Taiwanese Ka As mentioned in Chapter One, few studies have been done on the Taiwanese ka construction.. To our knowledge, Teng 1982 is probably the earliest one, followed by. Tsao and Lu 1990, Li 1995, Hung 1995, Cheng and Tsao 1995 and then Tsao 2003. Since both Tsao and Lu’s and Cheng and Tsao’s analyses mainly focus on the historical origin of ka, and Tsao’s study is very similar to Hung’s, in this section we will review Teng 1982, Li 1995, Hung 1995, and Cheng and Tsao 1995.. 2.1.1 Teng (1982) As pointed out by Teng, unlike the Mandarin ba construction, which carries a disposal meaning, the Taiwanese ka has various functions such as a goal, source, patient or benefactive marker.. Given the various functions of ka, Teng states that 12 . .

(23) “ka is a disposal marker only when its object is at the same time the true object of the main verb in the sentence.. A true object is defined in terms of transitivity (p. 334)”. as in (1): (1) I. ka. he. gua me.. KA I. blame. ‘He blamed me.’ In Teng’s study, the disposal ka is treated as a preposition.. Teng further points out. that it is possible to have a disposal ka sentence even when its non-ka counterpart is ungrammatical, as can be seen in (2): (2) a. Li you. ka. chiu giah-khi-lai.. KA hand lift-up-come. ‘Raise your arms!’ b. *Li you. giah chiu khi-lai. lift hand up-come. ‘Raise your arms!’. (pp. 334-335). (2)a is a grammatical disposal ka sentence since chiu ‘hand’ is the object of the transitive verb giah ‘raise.’ However, its non-ka counterpart in (2)b is illicit. Teng’s definition of the disposal ka construction implies that the construction exhibits the following characteristics. Like the ba-NP, the ka-NP can be preposed to the sentence-initial position. (3) a. Hit-e. lang,. li. beh. that CL person you want. ka. tua-lai. KA him bring-come. ‘Are you bringing that man?’. 13 . i. o? PT.

(24) b. Hit-pun. cheh,. li. ka1 be-khi. a. that CL. book. you. KA sell-go. PT PT. o?. ‘You sold that book?’. (pp. 336-337). In (3)a hit-e lang ‘that person’ is moved from the post-ka position, leaving a co-referential pronoun i ‘him’ in situ.. Furthermore, the pronominal copy in situ can. be deleted2 when it is a third person singular as in (3)b, resulting in a syntactically absent ka-NP which is not possible in the ba construction.. The missing ka-NP is. considered as a result of deletion, and then the morpheme ka undergoes a phonetic change, ending with a glottal stop. Also, unlike the ba-verb which requires an aspect marker as its minimal complement, the ka-verb can occur without one, as in (4): (4) a. Mama mother. ka. i. phah.. (Taiwanese). KA him hit. ‘Mother hit him.’ b. *Mama mother. ba. ta. da.. (Mandarin). BA him hit. ‘Mother hit him.’ In (4)a, the ka-verb phah ‘hit’ is a monosyllabic verb, and the sentence is grammatical.. However, in (4)b the ba-verb da ‘hit’ is also monosyllabic, but the. sentence is illicit. In addition, the ka construction is semantically neutral.. Thus, many different.                                                         1. Teng (1982) and Li (1995) claim that when the ka-NP is syntactically absent, the ka will undergo a phonetic change, ending with a glottal stop, that is, ka becomes ka?. However, Cheng and Tsao (1995) and Hung (1995) propose that ka? is further reduced to ka under fast speech. The present study is not going to tackle this problem and will simply adopt the view that ka? is further reduced to ka because we also cannot perceive the existence of a glottal stop. 2 Some studies consider the syntactically absent ka-NP as a result of contraction instead of omission (Li 1995, Hung 1995, Tsao 1995). The validity of the two camps can not be verified in this study. Hence, in this study the statement of omission will be adopted for simplification. 14 .

(25) categories of verbs can show up in the structure. In the study, Teng only categorizes the monosyllabic verbs that occur in the construction into the following nine types: i). Verbs that take patient (e.g. ciah ‘eat’). ii). Outward verbs (e.g. be ‘sell’). iii) Inward verbs (e.g. nia ‘collect’) iv) Perception verbs (e.g. khuann ‘see’) v). Verbs of handling (e.g. giu ‘pull’). vi) Verbs that take locative (e.g. phah ‘hit’) vii) Verbs of utterance (e.g. mng ‘ask’) viii) State verbs (e.g. thiann ‘care’) ix) Verbs that take affected object (e.g. khi ‘build’). (pp. 344-347). To sum up, Teng provides a detailed analysis of the Taiwanese disposal ka construction.. Nevertheless, some inadequacies are found in the study.. First, Teng. does not discuss the referentiality of the ka-NP. Later studies such as Li (1995) or Tsao (2003) point out that an indefinite NP cannot occur in the ka construction. Second, Teng does not provide a generalization for the type of verb that can occur in the ka construction.. Aside from lacking a generalization, the claim that all the nine. types of monosyllabic verbs can show up in the ka construction is also problematic. Considering the following example: (5) I cin suiau lang. ka. he very need people KA. thiann/*ai/*kahi/*sionn. care/love/like/miss. ‘He needs other people to care about/*love/*like/*miss him.’ The verbs thiann ‘care’, ai ‘love’, kahi ‘like’, and sionn ‘miss’ all belong to the type of state verbs and yet except for thiann ‘care’, the others are impossible to occur in the ka construction.. The ungrammaticality clearly shows that state verbs are. generally not compatible with the ka construction. Hence, further study will be 15 .

(26) needed to capture a generalization for the verbs that can show up in the ka construction.. 2.1.2 Li (1995) Li investigates the syntactic structure of ka as a disposal marker (the patient ka) in Taiwanese.. Two theoretical assumptions are presupposed in her study: first, the. Government and Binding theory developed by Chomsky (1981, 1982, 1986a and 1986b) and second, Li’s (1990) and Chen’s (1990) proposals. Inspired by Li and Chen, Li in this study assumes that the sentence structure in Taiwanese is S-O-V underlyingly, that is, Taiwanese is a head-final language.. Also, adopting Chomsky’s. (1986b) analysis, the subject is presupposed to be universally base-generated in [Spec, IP] in Li’s study.. Furthermore, directionality and adjacency requirements of case. assignment in Li’s (1990) study is also assumed, and the assumption that accusative case in Taiwanese is assigned to the right is thus presupposed in the study.. Given the. above assumptions, a Taiwanese sentence like (6) will have a D-structure representation as in (7) and an S-structure representation as in (8): (6) Gua kahi hit I. like that. nia sann. CL dress. ‘I like the dress.’. 16 .

(27) (7) D-structure representation. (8) S-structure representation. As in (7), the object hit nia sann ‘that dress’ is base-generated at the left of the verb kahi ‘like’ and at SS, the object moves to a post-verbal VP-adjoined position to be assigned a case by the verb kahi ‘like’ as shown in (8). Based on the above assumptions, Li proposes that ka is “a unique syntactic category, heading its own projection, selecting a VP complement (p. 35).”. It is also. argued that ka is a case assigner, discharging a [disposal] feature to [Spec, KaP] at LF, 17 .

(28) hence the manipulative interpretation of the ka sentences. Based on Li’s analysis, a ka-NP will move to [Spec, VP] to be case-marked and to [Spec, KaP] at LF to receive a disposal meaning.. Li further claims that though the. verb is a qualified case assigner, in the disposal construction, the NP will not move to the postverbal position because the verb only contains case while ka has two features to discharge, that is, [disposal] feature and case.. Hence, the NP will take ka as a. stronger trigger than the verb and move to the preverbal position instead of the postverbal position.. Thus, a sentence like (9) has a DS representation as in (10), an. SS representation as in (11), and an LF representation as in (12). (9) I he. ka. gua. KA me. phah. hit. ‘He hit me.’ (10) D-structure representation. 18 .

(29) (11) S-structure representation. (12) LF representation. In (10), the verbal object gua ‘me’ is base-generated at the left of the verb phah ‘hit’, and to be assigned an accusative case by ka, it is moved to [Spec, VP] at SS as in (11). Furthermore, to receive a disposal feature discharged by ka, it is then moved to [Spec, KaP] at LF as in (12)3.                                                         3. According to Li’s analysis, the retained object construction in Taiwanese as in (i) will have two SS 19 . .

(30) A syntactic analysis of the ka construction is provided in Li’s study.. However, Li. did not discuss the use of a progressive marker in the ka construction. It is pointed out that unlike the ba construction which is impossible to co-occur with a progressive marker, in the ka construction, a progressive marker is possible to show up (Hung 1995).. Generally speaking, a progressive marker is compatible with the ka. construction if the ka-verb is an atelic verb..                                                                                                                                                             representations as in (ii) and (iii): (i) I ka gina sian chuiphue. He KA child slap cheek ‘He slapped the child on the cheek.’ Li claims that the two objects gina ‘child’ and chuiphue ‘cheek’ are originally base-generated on the left of sian ‘slap.’ Also, since ka and sian ‘slap’ are both case assigners, the two objects can be case-marked by either ka or sian ‘slap,’ yielding two possible SS representations as in (ii) and (iii): (ii) SS Representation (iii) SS Representation. In both cases, the two objects can be case-marked by either the verb or ka. In other words, Li’s analysis will predict that both (ii) and (iii) are possible sentences in Taiwanese. However, as shown in (iv), this prediction is wrong: (iv) *I ka chuiphue sian gina. He KA cheek slap child ‘He slapped the child on the cheek.’ The ungrammaticality of (iv) is attributed to the semantic oddness. Li claims that gina ‘child’ and chuiphue ‘cheek’ stand in a possessive relation. A sentence like (iv) is odd because the listener has the concept of the subpart of the possessor first and then the idea of the possessor. 20 .

(31) (13) a. Koko. teh. ka. titi. the brother PROG KA. phah.. the little brother. hit. ‘The brother is hitting the little brother.’ b. *Koko. teh. ka. titi. phah-si.. the brother PROG KA the little brother hit-die ‘The brother is hitting the little brother to death.’ As shown in (13), phah ‘hit’ is an atelic verb while phah-si ‘hit-die’ is a telic one. The grammaticality of (13)a and the ungrammaticality of (13)b show that a progressive marker is incompatible with a telic ka verb.. 2.1.3 Hung (1995) Hung examines the four functions of the ka construction: a source, a goal, a patient and a benefactive marker.. She further divides ka as a patient marker into two. types according to the (non-)existence of a verbal complement: ka with a transitive verb but without any verbal complement and ka with a transitive verb and a verbal complement. Ka without a verbal complement has the following sentence pattern: NP1 ka NP2 V, as in (14). (14) Guan my. kiann. ka. gua. cim.. son. KA. I. kiss. ‘My son kissed me.’. (pp. 70-71). The features of this type of ka are discussed in the study. First, the ka-NP is usually a definite NP as in (15)a or a specific NP as in (15)b.. However, Hung further claims. that a non-specific NP is sometimes possible in this construction as in (15)c4.                                                         4. Hung uses example (15)c to show that a ka-NP can sometimes be non-specific. section we will show that her analysis may be inappropriate. 21 . Latter in this.

(32) (15) a. Alun ka Alun. in. kiann that.. KA his. son. kick. ‘Alun kicked his son.’ b. Alun. ka. cit. Alun KA one. e lang. me.. CL person scold. ‘Alun scolded a person.’ c. M-tang NEG-may. oopeh. ka. lang. me.. randomly. KA person scold.. ‘Don’t scold others randomly.’. (pp. 72-73). Second, verbs of this construction are always transitive action verbs. Stative verbs are forbidden in this construction as shown in (16). (16) a. *I he. ka cinn. u.. KA money have. ‘He has money.’ b. *Asam. ka. Ang. Asam KA Ang. senn. surname. ‘Asam’s surname is Ang.’. (p. 73). Third, the ka construction of this type can be in progressive forms, as in (17): (17) Asam teh. ka. in. Asam PROG KA his. kiann. me.. son. scold. ‘Asam is scolding his son.’. (p. 75). Finally, Hung claims that the function of this construction is to make the verb become the semantic focus.. There is no obvious semantic difference between this. type of the ka construction and its non-ka counterpart.. However, in this construction,. the action expressed by the verb seems to become a semantic focus. A piece of evidence for this claim is that when the object of the verb is a pronominal NP, the ka 22 .

(33) construction is usually used instead of its non-ka construction. A pronoun usually carries old information which is rarely a semantic focus.. Thus, ka preposes the. object NP to make the verb become the focus. The second type of ka as a patient marker is ka with a verbal complement. has the following sentence pattern: NP1 ka NP2 V complement. sentence of this type.. It. (18) is a ka. The verb consists of kong ‘hit’ and a complement phua. ‘break.’ (18) I. ka. he. puea. KA cup. kong-phua. a.. hit-break. ASP. ‘He broke the cup.’ According to Hung, the ka construction of this type corresponds to the Mandarin ba construction, and it shares many features with the ba construction such as taking a definite or generic NP, taking a transitive action verb and a verbal complement. The function of the disposal ka construction is to make the verb with its complement become the semantic focus, and most ka sentences with a verbal complement do not have non-ka counterparts, as in (19): (19) a. Gua I. ka. sann. KA clothes. se. chingkhi. wash clean. a. PT. ‘I washed the clothes clean.’ b. *Gua se I. wash. chingkhi clean. sann clothes. a. ASP. ‘I washed the clothes clean.’. (p. 94). Sentences in (19) can only exist in ka sentences.. The ungrammaticality makes the. preposing of the object NP become necessary. The object NP sann ‘clothes’ can move to the position before or after the subject. Also, when it is placed after the subject, it can be marked by ka or nothing. The preposing of an object NP marked 23 .

(34) by ka makes the verb with its complement become the semantic focus. According to Hung, ka as a source marker, a goal marker and a patient marker (the ka construction without a complement) are similar in that they all mark an object of an action, hence they can be categorized into an object marker 5 .. Ka as a. benefactive marker always marks a non-argument NP, which is often an entity indirectly affected by the action.. Because other types of ka always mark an. argument NP, ka as a benefactive marker is argued to be a different morpheme.. The. final type of ka is a disposal marker, that is, the patient marker ka occurring in the ka construction with a complement. the disposal ba construction.. This type of ka shares many characteristics with. Also, it has many features which are different from the. other types of ka; hence, it is argued to be a different morpheme.. According to these. arguments, there are actually three types of kas in Taiwanese: ka as an object marker, ka as a disposal marker and ka as a benefactive marker. Though Hung systematically compares the four functions of ka, her constraint on the progressive marker in the ka construction is not precise enough6. Also, her claim that a non-specific NP is sometimes possible in the ka construction seems to be ungrounded.. Example (20), taken from Hung (1995), is used to show that a ka-NP. can sometimes be non-specific NP. (20) M-tang NEG-may. oopeh. ka. lang. me.. randomly. KA person scold.. ‘Don’t scold others randomly.’                                                         5. In light of Jackendoff’s (1987) view that a thematic relationship can be more than one tier, Hung argues that “whether ka is a source marker, a goal marker, or a patient marker in another tier, in action tier, it has an identical function: to mark the object of an action (Hung 1995, p. 160).” 6 Her analysis implies that a progressive marker can only occur in the ka construction with a monosyllabic verb. Nevertheless, a bisyllabic verb like siuli ‘hit’ is also consistent with the progressive marker. (i) Koko teh ka titi siuli. brother PROG KA little brother hit ‘The brother is hitting the little brother.’ 24 .

(35) The NP lang ‘person’ in (20) is considered non-specific in Hung’s study.. However,. it seems to us that a more natural interpretation of the NP lang ‘person’ is to consider it as a generic NP.. Thus, to claim that a non-specific NP is compatible with the ka. construction based on an example like (20) may be inappropriate.. 2.1.4 Cheng and Tsao (1995) In light of Jackendoff’s (1987) view that a thematic relationship can be more than one tier, Cheng and Tsao (1995) reanalyze the four thematic roles of the ka construction: source, goal, patient, and benefective.. They propose that the four kas. are the same morpheme, and claim that it is the context with different di-transitive verbs that changes the meaning of the morpheme. Jackendoff separates the semantic structure into two tiers: thematic tier which deals with the motion or location and action tier which deals with Agent-Patient relations. Hence, Agent and Patient belong to the action tier and Theme, Source, and Goal belong to the thematic tier. In addition, the correspondence of the thematic roles between the two tiers are not constant.. Take sentence (21) as an example.. (21) a. Kingchat ka gua huat cinn. policeman KA me. fine. money. ‘The policeman fined me.’ b. I ka gua lau he KA me. leave. citkua cinn. some. money. ‘He left me some money.’ In (21)a, the ka-NP gua ‘I’ is the object affected by the action huat ‘fine’ and it is also the starting point of the movement of cinn ‘money.’ Hence, it has two thematic roles: source in the thematic tier and patient in the action tier. With regard to the ka-NP gua ‘I’ in (21)b, it is the object affected by the action lau ‘leave’ and the ending point 25 .

(36) of the moving of cinn ‘money.’ Thus, it has two thematic roles, too: goal in the thematic tier and patient in the action tier. Also, Cheng and Tsao propose that because huat ‘fine’ belongs to the type of taking verbs while lau ‘leave’ belongs to the class of giving verbs, though the ka-NP in (21)a and b are both the affected objects, the action of huat ‘fine’ is harmful to gua ‘I’ in (21)a while the action of lau ‘leave’ is beneficial to gua ‘I’ in (21)b.. The role of gua in (21)b, thus, is better analyzed as. benefactive. The example also shows that the meaning shift of the morpheme ka results from the context with different di-transitive verbs. Except for the claim that the four kas are the same morpheme, according to the distribution of negative elements and adverbials, Cheng and Tsao further claim that the four functions of ka be derived from the preposition kap ‘with.’ It has been observed in the study that negative elements and time adverbials have to precede the ka phrase and the kap phrase as shown in (22)-(23). (22) a. I bo he NEG. ka gua me. KA I. scold. ‘He did not scold me.’ b. I he. changam. ka. gua. yesterday. KA I. me. scold. ‘He scolded me yesterday.’ (23) a. I bo. kap gua tautin. (Cheng and Tsao 1995, p. 31) khi Taitiong.. he NEG KAP I together. go. Taitiong. ‘He and I did not go to Taitiong together.’ b. I. changam kap. he yesterday. gua. KAP I. tautin. khi. Taitiong.. together go. Taitiong. ‘He and I went to Taitiong together yesterday.’ Examples like (22) and (23) show that ka and kap have similar syntactic behaviors. 26 .

(37) Both the ka phrase and the kap phrase modify the verb phrase, and the scope of the negative elements like bo and time adverbials like changam includes the whole verb phrase or the sentence. Thus, the negative elements and time adverbials have to precede the ka phrase and the kap phrase. Based on the similar syntactic behaviors between ka and kap, Tsao and Cheng propose that the use of ka should be historically derived from kap. Cheng and Tsao’s analysis unifies different functions of ka and helps to capture a generalization of the use of Taiwanese ka.. However, the reason why stative verbs. are impossible to occur in the ka construction has not been given and the fact that a progressive marker is possible to occur in the construction under a proper context has not been discussed.. Hence, further research is necessary.. 2.1.5 Summary In this section, four theoretical studies are discussed in detail. analysis focuses on the properties of the disposal ka sentences.. Teng’s (1982). Both semantic and. syntactic properties are discussed in the study. According to him, Taiwanese ka has several functions and the disposal usage is only one of the functions. Li’s (1995) analysis concentrates on proposing a syntactic analysis of the properties of the disposal ka construction.. She claims that Taiwanese ka is a syntactic head, selecting. a VP complement and assigning case to the position of VP specifier. As for Hung’s (1995) study, the four functions of ka-a source marker, a goal marker, a patient marker and a benefactive marker-are investigated. among different functions of ka.. The focus is on capturing the relation. According to the semantic and syntactic properties. of ka, Hung argues that there are actually three types of kas in Taiwanese: ka as an object marker, ka as a disposal marker and ka as a benefactive marker.. With regard. to Cheng and Tsao (1995), they propose that the four functions of ka are the same 27 .

(38) morpheme and that the use of ka should be derived from kap.. In the following. section, we will discuss similarities and differences between Taiwanese ka and Mandarin ba.. 2.2 A Comparison between Taiwanese Ka and Mandarin Ba Based on the three theoretical analyses in section 2.1, it is noticed that there exist some similarities and differences between Taiwanese ka and Mandarin ba. Taiwanese ka and Mandarin ba share two main properties. First, as exemplified in (24)-(26), the ka or ba NP is generally definite or generic.. An indefinite NP. cannot show up in both constructions. (24) a. Mama Mother. ka. hit. KA that. te. uann kongphua. CL bowl. break. a.. (Taiwanese). PT. ‘Mother broke that bowl.’ b. Mama Mother. ba. na. BA that. ge. wan. CL bowl. dapo. le.. (Mandarin). break PT. ‘Mother broke that bowl.’ (25) a. Mama Mother. tianntiann. ka. uann. often. KA bowl. kongphua.. (Taiwanese). break. ‘Mother often breaks bowls.’ b. Mama Mother. changchang ba often. wan. dapo.. BA bowl. break. (Mandarin). ‘Mother often breaks bowls.’ (26) a. *Mama. ka. cit. Mother KA a. king pangking. sau. chingkhi.. CL room. sweep. clean. ‘Mother swept a room clean.’. 28 . (Taiwanese).

(39) b. *Mama. ba. Mother. yi. jian. fangjian. sao. ganjing.. BA a. CL. room. sweep. clean. (Mandarin). ‘Mother swept a room clean.’ The ka and ba NP in (24) are definite because they are modified by definite markers hit ‘that’ in Taiwanese and na ‘that’ in Mandarin, respectively.. In (25), the ka and ba. NP indicate a class of elements, hence they are generic.. The sentences are all. grammatical.. In (26), cit king pangking ‘a room’ in Taiwanese and yi jian fangjian. ‘a room’ in Mandarin are indefinite and the sentences are ungrammatical. Second, in general, a stative verb cannot appear in the ka or ba construction, as shown in (27): (27) a. *Gua I. ka. li. sionn.. (Taiwanese). KA you miss. ‘I misses you.’ b. *Wo I. ba. ni. BA you. xiang.. (Mandarin). miss. ‘I misses you.’ In (27), sionn ‘miss’ and xiang ‘miss’ are stative verbs in Taiwanese and Mandarin respectively and the two sentences are ungrammatical. Though ka and ba are similar in some ways, significant differences exist. are three main differences between them.. There. First, monosyllabic verbs can occur. without an aspect marker in the disposal ka construction while an aspect marker is always required in the disposal ba construction. (28) a. Gua I. ka. i. phah.. (Taiwanese). KA him hit. ‘I hit him.’. 29 .

(40) b. *Wo I. ba. ta. da.. (Mandarin). BA him hit. ‘I hit him.’. (Li 1995, p. 19). In (28), phah ‘hit ’in Taiwanese and da ‘hit’ in Mandarin are monosyllabic verbs. However, (28)a is grammatical, but (28)b is not. Second, in a proper context, a ka-NP can be omitted while a ba-NP can never be absent: (29) a. Cinliong. ka. ___ ciah.. as much as KA. (Taiwanese). eat. ‘Eat as much as you can.’ b. *Wo I. yijing. ba. already. BA. ___ chi le.. (Mandarin). eat ASP. ‘I have finished my meal.’. (Li 1995, pp. 19-20). Third, a progressive marker can sometimes show up in the ka construction whereas it is impossible in the ba construction. (30) a. Mama. teh. ka. muemue. phah.. Mother PROG KA the little sister. (Taiwanese). hit. ‘Mother is hitting the little sister.’ b. *Mama. zai. ba. meimei. da.. (Mandarin). Mother PROG BA the little sister hit ‘Mother is hitting the little sister.’ After the above discussion, we know that the Taiwanese ka construction is not a mere counterpart of the Mandarin ba construction. similarities and differences between them:. 30 . Table 2.1 briefly summarizes the.

(41) Table 2.1 Comparisons between Ka and Ba Taiwanese ka. Mandarin ba. Occurrence of an indefinite ka or ba NP. ×. ×. Occurrence of a stative ka or ba verb. ×. ×. Occurrence of a progressive marker. √. ×. Zero verbal complement. √. ×. Omission of the ka or ba NP. √. ×. 2.3 Properties of Ka in Taiwanese In this section, important properties of ka as a patient marker related to the present study will be discussed in detail.7. First, in 2.3.1 and 2.3.2, Referentiality and. Omission constraints on the ka NP will be presented, and then in 2.3.3, and 2.3.4 Dynamicity, and Progressive constraints will be discussed.. 2.3.1 Referentiality Constraint on the Ka-NP The NPs following ka are generally definite or generic (Cheng and Tsao 1995, Li 1995, Tsao 2003). (31) a. Ka hit KA that. king. chu. sau-sau. le!. CL. house. sweep-sweep PT. (definite NP). ‘Clean that house!’.                                                         7. Except for the properties examined in section 2.3, there are other properties of the patient ka. For example, chiong is another patient marker in Taiwanese. The chiong phrase and the ka phrase can co-occur in the same sentence (Cheng and Tsao 1995, Li 1995, Hung 1995, Tsao 2003). Also, in the retained object construction, the ka-NP and the retained object have either a whole-part or a possessor-possessee relationship. Furthermore, only the NP that serves as a whole but not part and the NP that serves as a possessor but not possessee can occur as a ka-NP. These two properties are beyond the scope of this study and will not be examined in the present study. 31 .

(42) b. Muemue. tianntiann ka. little sister often. cheh. KA books. oopeh. tan.. randomly. throw. (generic NP). ‘The little sister often puts books in disorder.’ c. *Ka KA. cit king. chu. sau-sau. a. house. sweep-sweep PT. CL. le!. (indefinite NP). ‘Clean a house!’ According to Li and Thompson (1981), a noun phrase with a classifier and a demonstrative is definite because the demonstrative serves to indicate an entity which is known to the speaker and the hearer.. In (31)a, hit king chu ‘that house’ contains a. demonstrative hit ‘that,’ hence it is a definite NP.. Also, according to Li and. Thompson (1981), a noun phrase with no modifier is generic if it denotes a class of entities instead of any specific members in that class. a generic NP.. Thus, cheh ‘books’ in (31)b is. In (31)c, cit king chu ‘ a house’ is an indefinite NP since the speaker. does not refer to a specific room that is known to the speaker and the hearer and cit king chu in Taiwanese cannot refer to a class of houses.. The grammaticality of (31)a. and (31)b and the ungrammaticality of (31)c show that an indefinite NP cannot occur in the ka construction.. 2.3.2 Omission Constraint on the Ka-NP It is often observed that a ka-NP does not have to be syntactically present (Teng 1982, Li 1995, Hung 1995). However, not all ka sentences allow a missing ka-NP. Only when the ka NP is third person singular can the ka NP be syntactically absent. Considering the following sentences: (32) a. Asam oh,. bo. lang. Asam PT NEG person. e. ka __. will KA. ‘As for Asam, nobody will help him.’ 32 . tausannkang. help.

(43) b. *Li li. oh, bo PT. lang. NEG person. e. ka __ tausannkang.. will KA. help. ‘As for you, nobody will help you.’ c. *Gua oh, bo I. lang. e. PT NEG person. ka __ tausannkang.. will KA. help. ‘As for me, nobody will help me.’. (Hung 1995, p. 7). In (32), only the ka-NP in a can be missing but never the ka-NPs in b and c.. The. ka-NP in a is third person singular, in b it is second person singular and in c it is first person singular.. The sentences in (32) show that only a third person singular ka-NP. can be syntactically absent.. 2.3.3 Dynamicity Constraint on the Ka-Verb Not every type of verb is compatible with the ka construction.. Generally. speaking, a stative verb cannot show up in the construction. (33) a. Muemue. ka. hue. the little sister KA flowers. tantiau.. (nonstative verb). throw away. ‘The little sister throws away the flowers.’ b. *Muemue the little sister. ka. hue. KA flowers. kahi.. (stative verb). like. ‘The little sister likes the flowers.’. 2.3.4 Progressive Constraint on the Ka-Verb The situation with the Taiwanese ka construction and the present progressive marker teh is rather complicated.. According to Hung (1995), the ka sentences. without a verbal complement can have a progressive form while those with a verbal complement cannot.. She claims that a bare action verb only indicates an action but 33 . .

(44) not the accomplishment of an action whereas a verb with a complement always indicates the accomplishment of an action. Given this, the progressive marker can only occur in ka sentences without a verbal complement but not with it. Besides, for the ka sentences without a complement, Hung focuses only on the monosyllabic verb. Hung’s analysis is right but not precise enough. Bisyllabic verbs like siuli ‘fix’ can co-occur with a progressive marker, too.. Also, since the occurrence of a. progressive marker depends on the telicity of the ka verb, a more precise way to state this constraint is that a progressive marker can only show up in the ka construction with an atelic verb but not in those with a telic verb. (34) a. I. teh. ka. in. kiann me.. PROG. KA. his. son. he. scold. ‘He is scolding his son.’ b. Papa. teh. ka. titi. siuli.. father PROG KA little brother hit ‘Father is hitting the little brother.’ c. *I he. teh. ka. png chia-liau.. PROG. KA rice. eat-finish. ‘He is eating the rice all up.’ In (34)a the verb me ‘scold’ is a bare monosyllabic verb and in (34)b the verb siuli ‘fix’ is bisyllabic.. Both of the verbs do not indicate accomplished actions, that is,. they are atelic verbs, and the sentences are compatible with the progressive marker. However, in (34)c, a verbal complement liau ‘finish’ signifies the accomplishment of the action and makes the verb become a telic verb, which is contradictory to the meaning of the progressive marker, resulting in the ungrammaticality.. 34 .

(45) 2.3.5 Summary In this section, four major constraints on the ka construction are discussed: Referentiality,. Omission,. Dynamicity,. and. Progressive. constraints.. The. Referentiality constraint states that only definite or generic NPs can show up in the construction.. The Omission constraint states that only when the ka-NP is a third. person singular pronoun can it be syntactically absent.. As for the Dynamicity. constraint, a stative verb cannot occur in the ka construction.. Finally, the. Progressive constraint states that a progressive marker can only show up in the ka sentences with an atelic verb.. 2.4 Experimental Studies of the Ba Construction Since there are no experimental studies of children’s acquisition of the ka construction, this section recapitulates two experimental studies on the acquisition of the Mandarin ba construction as ka is said to be very similar to ba. One study emphasizes on the effects of the semantic information on the acquisition of the ba construction (Cheung 1992).. The other study investigates children’s acquisition of. the various constraints on the ba construction (Fahn 1993).. 2.4.1 Cheung (1992) Cheung conducts an empirical study on investigating children’s L1 acquisition of the Mandarin ba construction. The focus is on the effect of semantic information on the syntactic status of ba. Two specific questions are addressed: 1) How do children treat ba during their grammatical development? What is the categorical status of ba in children's grammar? 2) Do children rely on the Object Affectedness Linking Rule in the acquisition of the ba construction?. (p. 76) 35 . .

(46) To answer the questions, Cheung suggests that a complex predicate analysis may be the best to account for the syntactic complexity of the ba construction based on the aspectual property of the construction.. Three experimental tasks were designed to. evaluate the developmental sequence of the ba construction: a picture selection task, a sentence imitation task and a picture-cued production task. Thirty-two children, sixteen of the age five and the other six, and sixteen adults participated in the experiment. sets were created.. In the comprehension task, sixteen pairs of picture. Each set consisted of three panels.. In a pair of picture sets, one. depicted Locative ba sentence and the other Theme ba sentence.. Two sets of. example pictures are shown below (p. 81):. Figure 2.1 Locative-Full (LF). Figure 2.2 Theme-Gone (TG) The first picture set presents a pitcher filled with juice and an empty glass, followed by a second panel showing a person pouring the juice from the pitcher into the glass. In the last panel, the person is gone and the glass is full of juice, though some is left in the pitcher.. The second picture set is similar to the first set except that 36 . .

(47) in the first panel, the pitcher has less juice while in the third panel, the pitcher is empty though the glass is not full.. That is, the two sets depict different entities. which are affected by the action, the glass in the first set and the juice in the second. The picture sets are matched with two potential target sentences: (35) a. Locative Ba Loashi. ba. beizi dao. teacher. BA cup. pour. man le. guozhi.. full ASP. orange-juice. ‘The teacher poured the cup full with orange juice.’ b. Theme Ba Loashi. ba. guozhi. teacher. BA orange-juice. dao dao. le. beizi. li.. pour arrive. ASP. cup. inside. ‘The teacher poured orange juice into the cup.’. (p. 81). The subjects were asked to choose a picture set which can better describe the sentence that the experimenter utters.. If a Locative Ba sentence is presented, the. first picture set is expected to be chosen because the ba-NP underscores the location and the verb-compound dao man ‘pour-full’ indicates the fullness of the glass.. On. the other hand, when a Theme Ba is presented, the second picture set is supposed to be chosen if object affectedness determines the acquisition of the ba construction. As for the sentence production task, the same picture sets used in the picture selection task were employed.. One picture set of each pair was selected.. Altogether, there were sixteen picture sets, eight from the Locative version and eight from the Theme version.. The subjects were asked to describe the picture according. to a prompt question. Two types of prompts were used, one focused on the theme and the other focused on the location.. For example, when a picture set depicts the. action of moving books from the floor to the bookshelves, the two possible prompts are shown below: 37 .

(48) (36) a. Locative prompt Xia little. pengyou ba boy. shujia. zhenyang le?. BA bookshelf. what. ASP. ‘What did the boy do to the bookshelf?’ b. Theme prompt Xia little. pengyou boy. ba. shu. BA book. zhenyang what. le? ASP. ‘What did the boy do to the book?’. (p. 110). For half of the Locative picture sets, Locative prompts were used and for the other half, Theme prompts were used.. The same procedure was performed for the Theme. picture sets. In the imitation task, sixteen sentences were used, half of them were grammatical and half ungrammatical.. Among the ungrammatical sentences, four sentences. belonged to the Theme Ba construction and four the Locative Ba construction. Likewise, eight counterparts of the ungrammatical sentences were created in the grammatical set.. The subjects were required to repeat the grammatical or. ungrammatical sentences which the experimenter said. The results of the picture selection task illustrated that the subjects had a preference for matching the Locative sentence with the Locative picture set. However, the children did not prefer the Theme picture in response to a Theme Ba target sentence, indicating that object affectedness did not determine the acquisition of ba.. The result, thus, did not support the Object Affectedness Linking Rule.. According to Gropen et al. (1991), the rule states that “Link the argument that is specified as “cause to change” in the main event of a verb’s semantic representation to the grammatical object (p.159).”. That is, the affected element, the element which is. caused to change its state by the action, tends to be encoded as the direct object of the 38 .