漢語動賓複合詞處理歷程研究

全文

(1)國立臺灣師範大學英語學系碩士. 論. 文. Master’s Thesis Department of English National Taiwan Normal University. 漢語動賓複合詞處理歷程研究. Verb-object Compound Processing in Mandarin. 指導教授：詹曉蕙博士 Advisor: Dr. Shiao-Hui Chan 研究生：許雅雯 Student: Ya-Wen Hsu. 中華民國一百零四年七月 July 2015.

(2) 摘要動賓複合詞的語法屬性，因其可離可合的特殊表現，一直以來都是語言學家討論的議題。許多學者認為動賓複合詞無論在黏著或是離析的情況下，其語法屬性都還是一個詞彙(Chao, 1968; Li & Thompson, 1983; Yi, 2007; Wang, 2009)；也有學者提出只要動賓複合詞有可以被插入語分離的情形，該詞便應視為短語而非詞彙(Lu, 1979; Paul, 1988; Sybesma, 1999)；另有一派學者認為，動賓複合詞在其黏著型態時應被視為詞彙，但在離析型態時則應視其為短語 (Liu, 1967; Zhang, 2010)。本文旨在以行為實驗來探討具有不同語意透明度的中文動賓結構，在黏著句式和離析句式的情況下，其語法屬性和處理機制是否會因為語意透明度的差異而有所不同。本實驗假設，動賓結構在黏著句式的處理應與離析時不同，而語意透明度在處理的方式上應也扮演了重要的角色。實驗的結果符合假設：動賓結構在黏著和離析的句式裡，處理方式確實有差異。在黏著的句式裡，含有語意透明度低（亦即傳統上的動賓複合詞）或者透明度中等的動賓結構之句子，其處理速度都較語意透明度高（亦即傳統上的動賓詞組）的組別來得快；但是在離析的句式裡，三種不同語意透明度的動賓結構則無顯著差異。亦即，當動賓結構被分開了，受試者將會受其插入詞影響，即便是語意黏著性高的動賓複合詞，也會被當成是短語而非單一詞彙處理。簡而言之，本實驗結果顯示動賓複合詞合則為詞彙，離則為詞組，而雙重路徑模型對此型態較能提供合理的解釋。. 關鍵詞：動賓複合詞，離合詞，語意透明度，複合詞處理機制，雙重路徑模型 i.

(3) Abstract The current study examined the controversial issue of Verb-Object Compounds (VOCs). Some scholars in previous literatures contended that a VOC should be treated as a lexical unit no matter it is in a separated form or unseparated one (Chao, 1968; Li & Thompson, 1983; Yi, 2007; Wang, 2009). In contrast, the other group of scholars contended that as long as a VOC can be separated by interposing element, it should be considered a phrase instead of a lexical unit (Lu, 1979; Paul, 1988; Sybesma, 1999). Combining the two views above, a group of researchers think that VOCs are lexical units when they are in unseparated form but are phrases when they are in separated form (Liu, 1967; Zhang, 2010). By adopting a grammatical judgment task, how VOCs with various semantic transparency are processed in unseparated and separated forms in a sentence was examined in this study. It was hypothesized that VO sequences with lower semantic transparency (i.e. traditionally defined as VOC compounds) might be processed differently when they are in an unseparated and in a separated form. Also, when these VO sequences are in an unseparated form, they might be processed similarly to VO sequences with higher semantic transparency (i.e. traditionally defined as a Verb-Object phrase (VOP)). The results revealed that VO sequences with lower semantic transparency were indeed processed faster than those with higher semantic transparency in the unseparated condition, but all the VO sequences were equally taxing in the separated condition, suggesting that VOCs were processed as a lexical ii.

(4) unit in the unseparated condition, while they were processed like phrases in the separated condition. The theoretical claim that VOCs are compound words in an unseparated form and are phrases in a separated form (Liu, 1967; Zhang, 2010) is thus a more plausible explanation. From a psycholinguistic point of view, the duel-route model is a more likely explanation for VOC representations in the mental lexicon.. Key words：Verb-Object compound, separable words, semantic transparency, compound processing, dual-route model. iii.

(5) 謝辭終於，拿到碩士這個學歷。這個學歷對我來說，不僅僅只是一張文憑，它更像是一段漫長旅行後得來不易的紀念品，提醒著我這一路走來，身邊的人事物如何成就現在的我。用著我最熟悉的語言，我想向一路陪伴我的美好人們致上我深深的謝意。我很幸運的在神經語言學這門課和一位很神的教授相遇了。詹曉蕙教授，我論文的指導教授。一位對學術，對生活，對生命都很有熱情的一位教育家。她不只給了我學術上的支持，她對事物抱持的那種正向積極的態度，更帶給了我前進的勇氣和能量。從教授身上，我學到的何止是專業知識而已！真的非常感謝曉蕙教授一直以來的幫忙和支持！同時，我的論文口試委員，李佳霖教授、李佳穎教授、李甄儀教授，同樣也不只是給了我學術上的知識。這三位老師做研究的嚴謹態度，和幫助學生的熱誠，讓我獲益良多。能有機會和上述這四位教授學習真的十分幸運！在寫論文的過程當中，實驗室裡的歡樂氣氛讓我論文之路走來始終覺得心情愉快。我很幸運地在實驗室認識了一群溫暖又可愛的朋友。首先，我想先謝謝已經完成論文的「前輩」Gracie、Matt、Helen、Ken 的幫忙！感謝 Gracie 總是溫柔地回答我的問題，不厭其煩。感謝 Helen 總是細心地提醒我細節，聽我沙盤推演口考。感謝 Matt 下班後還特地跑來學校幫我解決程式的問題。感謝 Ken 儘管人在荷蘭，還是隨時隨地無國界的解答我的疑惑。實驗室的其他成員也在各個層面幫忙且陪伴我走過寫論文這整個過程。感謝 Jeff 將吉他帶入我的生命中，每當壓力大時彈彈吉他總是能瞬間紓壓。感謝 Elvis 每次都跟我鬥嘴，跟你講話我都覺得精神百倍。感謝 Terry 七早八早還義不容辭地跑來幫我做實驗。感謝 Ronald 在我跑實驗跑得昏天暗地的時候溫暖地給了我一碗豆花還載我回家。感謝 Lin 總是很貼心的陪在實驗室幫忙，在實驗後跟你一起吃的飯總是特別好吃。感謝 Lilian 總是能聽我分享心事，討論實驗，聽妳講娃娃的故事真的很有趣。感謝 Lucy 陪我聊了一大堆有的沒的，能跟妳一起走路回家真的很幸福。感謝 Kevin 總是讓我覺得在研究室不孤單，肚子餓也不孤單。感謝 Julia 總是跟我一起瘋一起笑，跟妳一起寫論文愉快的不可思議。 iv.

(6) 除了實驗室的夥伴，我也很感謝我的同學 Becky、Brian 和 Edea。在我的研究所生活裡，如果沒有你們一定失色不少。一起去玩過的地方，吃過的東西，講過的笑話我都不會忘記，謝謝有你們陪我。還有每一位因為篇幅而沒辦法一一打出名字的好朋友，很謝謝有你們陪在身邊，你們豐富了我的人生，給了我溫暖。如果沒有你們，這一路我一定會走得更辛苦，所有的一切因為有你們變的美好。我也想謝謝曾先生、慕涵助教、淑慧助教、豪谷助教、姿儀助教和卉喬助教。謝謝你們一路上的幫忙。每次去系辦看到你們都很開心，也因為有你們，我的研究所生活才能那麼充實又順利，真的謝謝你們！每一位在師大教導過我的老師，謝謝你們。因為有你們的啟發，我的思想才能更成熟更深刻。從你們身上學到的，我會努力的實踐在我的生命中。時時刻刻提醒自己，要努力站在社會最需要我的地方。最後，我想把最深的感謝，獻給我的家人。謝謝我的妹妹，雖然妳每次都不太情願的樣子，但是需要妳幫忙的時候，很謝謝妳還是願意幫我。謝謝我的大弟，雖然你嘴巴蠻壞的，但都會默默的幫我一把，我其實都記在心上。謝謝我的小弟，在我身邊親眼看著我從無到有，陪著我寫完這本論文。真的很謝謝你在艱難的時刻，陪在我身邊。謝謝我家的狗Ｍulu，在我焦慮的時候都會乖乖地給我抱。只要摸摸妳，就能一下子冷靜不少。謝謝我的爸爸媽媽，一直以來都讓我沒有後顧之憂地去追逐我的理想。謝謝你們的支持、包容和陪伴。謝謝你們，讓我有一個很棒的家，不管面對怎樣的困難，我知道我永遠都能從這個家裡得到力量，繼續走下去。謝謝我自己，堅持到最後一刻。最後的最後，感謝神，一切歸給祢。. v.

(7) Table of Contents 摘要.............................................................................................................................................i Abstract ......................................................................................................................................ii 謝辭...........................................................................................................................................iv Table of Contents ......................................................................................................................vi List of Tables ...........................................................................................................................vii List of Figures ........................................................................................................................ viii Chapter 1 Introduction ............................................................................................................... 1 1.1. Motivation ...................................................................................................... 1 1.2. Significance of the Study ............................................................................... 4 1.3 The organization of the current study ............................................................ 5 Chapter 2 Literature review ....................................................................................................... 6 2.1 Traditional definition of compound composition ................................................ 6 2.1.1 An introduction to Chinese compound ..................................................... 7 2.1.2 The characteristics of verb-object compound in Mandarin .................... 10 2.2 Empirical accounts on compound composition ................................................. 13 Chapter 3 Methodology ........................................................................................................... 21 3.1 Participants ......................................................................................................... 21 3.2 Materials ............................................................................................................ 22 3.3 Procedure ........................................................................................................... 30 3.4 Data Analysis ..................................................................................................... 33 Chapter 4 Results ..................................................................................................................... 35 Chapter 5 Discussion ............................................................................................................... 40 Chapter 6 Conclusion ............................................................................................................... 44 References ................................................................................................................................ 46 Appendices ............................................................................................................................... 52 A. The rating questionnaire for stimulus’ semantic transparency ......................................... 52 B. The rating questionnaire for stimulus’ familiarity ............................................................ 53 C. The rating questionnaire for stimulus’ concreteness ......................................................... 54 D. The rating questionnaire for stimulus’ naturalness ........................................................... 55 E. A complete list of critical verbs ......................................................................................... 56 G. Experimental Instruction .................................................................................................. 90. vi.

(8) List of Tables Table 1: Example sentences of the current experiment. ........................................................ 22 Table 2. The summary of the statistical results for all the controlled variables. ................... 28 Table 3: The number of the experimental trials in each condition. ....................................... 30 Table 4. Behavioral data of the subjects. ............................................................................... 35 Table 5. Separability effect: T-test results of reaction time in VO unseparated and separated conditions ......................................................................................................... 36 Table 6. Transparency effect: T-test results of reaction times between conditions. .............. 37 Table 7. Transparency effect in separated and unseparated conditions: T-test results of reaction times between conditions ................................................................................... 38 Table 8. Transparency effect in separated and unseparated conditions: T-test results of accuracy rates between conditions ................................................................................... 39. vii.

(9) List of Figures Figure 1: Procedure of stimulus presentation. ....................................................................... 32. viii.

(10) Chapter 1. Introduction. 1.1. Motivation. One of the amazing qualities of a language is that its infiniteness is composed of finiteness. Any user of a language can enrich the system of a language by creating new words. There are three main approaches for word-creating, namely, inflection, derivation, and compounding. While compounding could be considered peripheral in some languages, it is the most significant morphological mechanism in Chinese. Approximately 80% of Chinese words are compound words (Xing, 2006:117). In the corpus of neologisms proposed in The Contemporary Chinese Dictionary (2002), more than 90% of new created words are compounds. Linguistically speaking, a compound can be defined as a lexeme (less precisely, a word) that consists of more than one morpheme. Although the process of compounding is seemingly straightforward, the inner structure of a compound is complex (Chao 1948, 1968; Li and Thompson 1981; Tang 1989; Packard 2000). There are at least five types of compounds in Chinese which can be classified in terms of morphological structures 1.

(11) between the constituents within a compound. These five types are subordinate, coordinative, subject-predicate, verb-object and verb-complement structures (Tang 1989; Yuan and Huang 1998; Liu et al. 2000). Among these structures of compounds, verb-object compounds (henceforth VOCs) have their peculiar usage when functioning as a verb. These VOCs are mostly intransitive, and one of the most fascinating features of them lies in its seperabability , e.g. chi cu le 吃醋了 and chi le cu 吃了醋 “being jealous”. This kind of separable VOC is a rather productive kind of disyllabic verbal construction in Mandarin Chinese. The verb part and the object part of a separable VOC can be used separately by interposing elements in between while the semantic integrity remains (Siewierska et al., 2010). Understandably, the semantic transparency of VOCs is highly opaque, since its semantic integrity remains regardless of what form it is in. In other words, since the meaning of a VOC cannot be derived by simply adding the meanings of its constituents together, semantic transparency thus is a key criterion for distinguishing VOCs from verb-object phrase (VOP for short). Based on VOC’s behavioral pattern, some researchers claim that VOC whose constituents can be separated is actually a phrase (Lu, 1979; Paul, 1988; Sybesma, 1999), while some researchers contend that VOCs should be treated as a lexical unit (Li & Thompson, 1983; Packard, 2000; Chung, 2004; Yi, 2007). Some researchers on the other hand think that the grammatical identity of a VOC is neither an absolute phrase nor an 2.

(12) absolute lexical unit: they considered VOC a phrase while they are separated by interposing element and a lexical unit while there is no interposing element in between (Liu1967; Zhang, 2010). Conceivably, the linguistic identity of a VOC has been a controversial issue, especially when it compares with VOP. The distinction between a VOC and a VOP is rather fuzzy in terms of the structure. Since VOCs can work together as a word-like form and can also act separately as a phrase-like form, this behavioral pattern is basically quite similar to VOPs. Take chi cu 吃醋 “eat vinegar” ‘and chi mian 吃麵“eat noodles” as examples. The former one is a VOC, while the latter one is a VOP. Both of them can be used separately as in 她吃了他的醋 ta chi le ta de cu “she is jealous of him” v.s. 她吃了他的麵 ta chi le ta de cu “she ate his noodles”. It can be seen from above that VOCs have presented a challenge to existing morphological and syntactic theories as to whether they are lexical units or merely just verb-object phrases in Chinese. It seems that in the field of formal linguistics this issue is still feverishly discussed without consensus being reached. This study aims to add to the literature and to examine this issue with an empirical approach.. 3.

(13) 1.2. Significance of the Study. Existing Chinese VOC studies are mainly discussed under the framework of formal linguistics. It is not until recent years that Chinese compound processing draws psycholinguists’ and neurolinguists’ attention. Chinese compounds like coordinate construction (Chung et al. 2010) and modifier-head construction (MacGregor & Shtyrov, 2013) have been explored in recent years, showing that the Chinese compound system is an issue worth discussing and studying. So far there is almost no psychological or neurological study on Chinese VOC issues. Hoping to examine VOCs issue form another perspective, the current study is a behavioral study, examining reaction time differences while subjects processed verb-object (VO) structure. We hope to find out whether VO structures under the manipulation of separability and semantic transparency are processed differently or not; i.e. whether there is any distinction between VOCs and VOPs. It is worth noticing that Chinese compound processing studies in recent years are conducted mostly at the word/phrase level (Myers et al., 2004; Chung et al., 2010; MacGregor & Shtyrov, 2013). Such processing at the sentence level is a rather unexplored area. Moreover, since sentences may be a more natural linguistic context presenting VO separated intransitive verbs; the current study thus aims to investigate VO structured verbs at the sentence level. We hope the current design can answer the research question: Do 4.

(14) native speakers of Mandarin Chinese process VOCs differently from VOPs? We hypothesize that, if people indeed treat VOCs as a lexical unit, the RT of VOCs should be faster than VOPs in both separated and unseparated forms.. 1.3 The organization of the current study The current study aims to explore the processing of verb-object compound (VOC) in Mandarin Chinese. First, the motivation and the overall introduction of current study are stated in Chapter 1. In Chapter 2, literature related to the current study is illustrated. The methodology is demonstrated in Chapter 3, which is followed by Chapter 4, the results of the experiment. The discussion and conclusion are included in Chapters 5 and 6 respectively. Finally, the references and appendices are provided at the end of the current study.. 5.

(15) Chapter 2. Literature review. In Chinese, compounding is no doubt the dominant approach to create various lexical units. In this chapter, Chinese compound system and compounding mechanism are introduced and classified first which is followed by the detailed account on the characteristics of Chinese Verb-Object Compounds (VOC) in section 2.1. Experimental account on explaining compound processing from both psycholinguistic perspective and neurolinguistics perspective are illustrated in section 2.2 and section 2.3 respectively.. 2.1 Traditional definition of compound composition. In this sub section, compound composition is illustrated from the theoretical linguistic point of view. There are two parts in this section. Firstly, the definition and classification of Chinese compounds are introduced. The characteristics of verb-object compound in Mandarin are stated thereafter.. 6.

(16) 2.1.1 An introduction to Chinese compound. Serving as background information for the current study, some basic issues about Chinese compounds, such as what is the general definition of words and compounds? What are the descriptive characteristics of compounds? What could the compounding process be? are introduced in this section. In the accumulated literature, the definition of “word” in Chinese is controversial (cf. Chao 1948, 1968; Kratochvil 1968; Li and Thompson 1981; Tang 1989; Packard 2000). Nevertheless, according to the phonological structure of syllables, Chinese words can be generally divided into two categories: monosyllabic and polysyllabic. Basically, the monosyllabic word consists of one morpheme, such as shui 水 (water) and cu 醋 (vinegar), while the polysyllabic word is composed of two or more morphemes, such as ‘十字路口 shiziluko’ (intersection). There are also a few cases in which polysyllabic words contain only one morpheme, such as ‘蘿蔔 luo bo’ (radish) and ‘馬拉松 malasong’ (marathon race). These one-morpheme polysyllabic words are typically borrowed from other languages. Owing to the fact that monosyllabic structures leads to a large number of monosyllabic homophones in modern Chinese, in order to distinguish those homophones, word formation consequently tends to become more and more polysyllabic and compounding therefore is a good approach to make a lexeme polysyllabic. Nowadays, roughly two thirds of the basic 7.

(17) lexical items of everyday Chinese are polysyllabic words (Lu 1965; Lieber 2009). The formation of a compound has been an intriguing topic which attracts lots of linguists to devote themselves into the studies related to compounds. Compounds are lexical units yet are composed of two or more linguistic parts without grammatical marks expressing its inner structure and showing how the structure is obtained. In other words, the relation between the components in a compound is not shown on the surface, e.g. 槍傷 qiang shang “gun wound”, the wound caused by the gun, 腿傷 tui shang “leg wound”, the wound on the leg, and so forth. Moreover, the meaning of the compounds usually cannot be obtained by decomposing its components, which became another reason why compound formation has been under the spotlight for a long time. Compounding is the dominant approach on creating new lexemes and the classification of compounds varies among different linguists (citations). According to Tang (1989), there are five major categories of compounds in Mandarin Chinese based on different internal constructions of words:. (1). Subject-predicate construction: there is a syntactic subject-predicate relationship between the components of a given compound. For example, the compound ‘tou tong 頭痛’ (headache) consists of the morpheme ‘頭 tou’ (head) as the subject and the morpheme ‘痛 tong’ (ache) as the predicate.. 8.

(18) (2). Modifier-head construction: there exists a syntactic modifier-head relationship between the components of a given compound. For example, the compound ‘dahong大紅’ (crimson) consists of the morpheme ‘大 da’ (big) as the modifier and the morpheme ‘紅 hong’ (red) as the head.. (3). Coordinate construction: The antonymous or the synonymous relationship between the components manifests on this kind of compound. For example, the compound ‘hei bai 黑白’ (black and white) consists of the morpheme ‘hei 黑’ (black) and the morpheme ‘bai 白’ (white), which is in an antonymous relationship. The compound ‘dao lu 道路’(path and road) consists of the morpheme ‘dao 道’ (path) and the morpheme ‘lu 路’ (road), which is in a synonymous relationship.. (4). Verb-complement construction: there exists a syntactic verb-complement relationship between the components of a given compound. For example, the compound ‘shui bao 睡飽’ (to sleep till one is full) consists of the morpheme ‘睡 shui’ (to sleep), which can be considered the verb and the morpheme ‘飽 bao’ (full), which can be considered the complement.. (5). Verb-object construction: there is a syntactic verb-object relationship between 9.

(19) the components of a given compound. For example, the compound ‘guan xin 關心’ (worry) consists of the morpheme ‘關guan’ (close), which can be considered the verb and the morpheme ‘心 xin’ (heart), which can be considered the object.. The scope of this article is mainly focused on verb-object compounds (VOCs) as illustrated in the 5th category above.. 2.1.2 The characteristics of verb-object compound in Mandarin. A central issue on VOCs is whether to classify them as lexical units or as phrases, like verb-object phrase (VOP). A group of researchers (Chao, 1968; Li & Thompson, 1983; Yi, 2007; Wang, 2009) contend that a VOC should be defined as a lexical unit.. Li and. Thompson (1981) summarized three criteria on determining whether a VO structure is a compound or not: (1) One or both of the constituents being bound morpheme, (2) Idiomaticity of the meaning of the entire unit, and (3) Inseparability or limited separability of the constituents. Although Li and Thompson’s criteria to certain degree are still valid, they are unable to fully cover the diversity of VOCs, especially with regard to the following two perspectives. First of all, it is generally agreed that a VOC’s meaning is “opaque”, which cannot be derived by simply combining the meaning of its morphemes (e.g. chi cu 吃 10.

(20) 醋 “*eat vinegar”) while the meaning of a VOP is usually “transparent”; i.e., the meaning can be derived by combining the meaning of its morphemes (e.g. chi mian 吃麵“eat noodles”). However, the “opaqueness” is not a simple dichotic classification (e.g. chi fan 吃飯 got both a transparent meaning “eat rice” and an opaque meaning “having a meal”; kai men 開門 got both a transparent meaning “open the door” and an opaque meaning “start running business”). Furthermore, while it is usually argued that a VOP can be freely separated (e.g. chi le hao jiu dou chi bu wan de mian 吃了好久都吃不完的麵 “has been eating noodles for a long time but still cannot finish it”) and the interposing elements of a VOC are limited (e.g. * chi le hao jiu dou chi bu wan de cu 吃了好久都吃不完的醋“has been being jealous for a long time but still cannot finish it”), it is unclear why fang le xin 放了心“relived” is fine but *guan le xin *關了心 “worried” is not. Focusing on the separability of constituents of a VOC, a group of scholars contends that as long as a lexical unit can be separated or expanded, the lexical unit should be classified as a phrase instead of a word (Lu, 1979; Paul, 1988; Sybesma, 1999). However, though this assumption can explain VOCs’ separability, it failed to account for VOCs’ limited separability compared with the unlimited separability in VOPs. Meanwhile, this assumption fails to account for VOCs’ integral semantic meaning in separated form as well. Since the two claims above cannot fully capture the characteristics of VOCs, a group of researchers thus proposed that VOCs are words when the verbal and 11.

(21) nominal/complement morphemes appear together, and are phrases when used separately (Liu, 1967; Zhang, 2010). Along with this line, another group of researchers considered VOCs are neither absolute word nor absolute phrase as well. They considered VOCs are in a “middle-state” or transitional categories in morpho-syntax. Packard (2000) proposed that the diversity of the VOCs can be explained in terms of the continuum of lexicalization, the process of creating items out of syntactic units (Cabrera et al., 1998). The newly created items usually lose their compositionality and acquire a new idiosyncratic content. The separability of a VOC in this assumption thus depends on its position in the continuum of the lexicalization. If a VOC is highly lexicalized, then it is unlikely to insert anything between V and O (e.g.關心 guan xin “worry”, *關了心 guan le xin “worried”). On the other hand, for the VOC which is still in the transition of lexicalization, the insertion will be allowed ( fang xin 放心“feel relived”, fang le xin 放了心 “felt relived”). In this view, VOCs can be considered a dynamic language phenomenon. In sum, VOC is seemingly better accounted for in terms of the claims of “lexicalization” and “unseparated = words, separated = phrases”. Actually, these two are assumptions which most of the linguists agreed on when explaining VOC patterns (Liu,1967; Li&Tompson, 1989;Smith, 1999; Packard 2000 ; Chung, 2004; Fang, 2008 ; Zhang, 2010). Following these assumptions, when used separately, a VOC is like a VOP. The main purpose of this study is to find out whether such a view has psychological reality. 12.

(22) 2.2 Empirical accounts on compound composition. In the previous section, it seems that most linguists agreed to use lexicalization to explain VOC patterns. However, there is still no consensus on the processing mechanism of VOCs among psycholinguists. In other words, how people process VOC is still obscure. Are we retrieving a unitary word representation out of our lexicon? Or do we comprehend a VOC with a combinatory mechanism as in understanding a phrase? In the field of psycholinguistics, the representation and processing of compound words, and morphologically complex words more generally, remains a controversial issue. Much psycholinguistic research has focused on the question whether morphologically complex words are stored in the mental lexicon in their full form or whether only their morphemes are stored and then combined to process complex word forms. The former idea is so-called full-listing models (Butterworth, 1983; Bybee, 1995) and the latter view is termed full-parsing models (Libben, Derwing, & de Almeida, 1999; Taft, 2004; Taft & Forster, 1976). Alternatively, it is proposed by another group of researchers that both mechanisms may be invoked, which is known as dual-route models (Gunter, & Friederici, 2003; Koester, Gunter, & Wagner, 2004, 2007; Zwitserlood, 1994). In dual-route models, two routes of processing are assumed. A complex word can either be stored completely or be decomposed 13.

(23) into its morphological constituents (Baayen, Dijkstra, & Schreuder, 1997; Gunter, & Friederici, 2003). In order to decide among these models, many studies have been designed to explore semantic decomposition of compounds in the auditory and visual modality (Coolen et al. 1993; Isel et al., 2003; Libben et al., 1999; Libben, 1993) with two variables being manipulated: word frequency and semantic transparency. It is conceivable that the frequency of the word mattered in compound processing since more frequent words are more likely to benefit from readily available whole-form storage, whereas less frequently used compounds might have to be processed through a combinatorial mechanism. Semantic transparency is an influential factor on compound processing as well. Since transparent compounds do not contain idiosyncratic meaning, they do not need distinct lexical representations and consequently their processing mechanism may be very similar with syntactic rules linking words in a sentence (e.g. blue+berry). On the contrary, the meaning of opaque compounds cannot be derived by combining the meaning of its constituents (e.g. straw+berry?) and thus may rely on whole-form lexical storage. However, it should be noted that it is still controversial to contend that opaque compound is accessed solely in a full-listing fashion, as illustrated below. To explore whether the meanings of individual constituents are accessed during compound processing or not, a number of behavioral studies are conducted with semantic 14.

(24) priming paradigms. It is shown that, regardless of the preceding prime being semantically related to the first or second constituent of the target compound, the lexical decision times to two-constituent transparent compound words were speeded up (Sandra, 1990; Zwitserlood, 1994). Based on the result above, it is argued that combinatorial processing is carried out for transparent compounds. Trying to provide more evidence for the study above, a cross-modal priming study is conducted by Zhou et al. (2000). In their study, the result showed that visually presented transparent compound words were primed by the prior auditory presentation of both first and second compound constituents, but the effect was absent for opaque compounds. In line with these findings, another cross-modal semantic priming study showed that the prosodic cue of the initial morpheme of a compound is able to assist the processing system in activating a decompositional route at the offset of the morphemes (Isel et al., 2003). The assistant effect only happened in compound words with a transparent head but not in compound words with an opaque head. However, in a lexical decision task using a repetition priming paradigm (Libben, Gibson, Yoon, & Sandra, 2003), the result showed that the presentation of either the first or second constituent as a lexical prime speeded up lexical decisions for both opaque and transparent compounds. This implies that constituent access is activated for both transparent and opaque compounds. Furthermore, in eye movement studies which directly compared 15.

(25) processing of transparent and opaque compounds (frequencies of constituents and the frequency of the whole-word forms were equal between the two types), no differences were obtained on any eye movement measure for either English (Frisson, Niswander-Klement, & Pollatsek, 2008) or Finnish (Pollatsek & Hyönä, 2005) stimuli. Again, the results suggest that both transparent and opaque compounds adopt similar processing mechanism. More interestingly, in a recent study conducted by Gagne et al. (2009), both transparent and opaque compounds were processed more quickly than monomorphemic words, showing that even the opaque compounds were processed more quickly than monomorphemic words which again, indicated that lexical entries of constituents are accessed in compound processing regardless of semantic transparency. In sum, the behavioral studies under review are mostly conducted in terms of semantic priming paradigm. Although the role of transparency in compound processing is still under debate, the above studies generally report decomposition effects, which are in accordance with full-parsing and dual-route models but not with full-listing models.. Accessing to compound constituents has also been studied neurophysiologically in recent years. Event-related potentials (ERPs), with the high temporal resolution, are a suitable tool to investigate such fast psycholinguistic processes.. Before heading into the experiment review, N400 (Kutas&Hillyard, 1980) will be briefly introduced here. The N400 response is a broad negative deflection of the ERP that 16.

(26) starts 200–300 ms after a word has been presented auditorily or visually and peaks after approximately 400 ms. This negative-going wave is usually largest over central and parietal electrode sites, with slightly larger amplitude over the right hemisphere than over the left hemisphere. The N400 is typically seen in response to violations of semantic expectancies. The N400 is typically elicited in response to meaningful stimuli and thought to reflect access or integration of conceptual information (Kutas & Federmeier, 2011). It is also assumed that the N400 effect reflects the difficulty in integrating the local lexical semantics into the sentence/ discourse representation (Van Berkum et al.,1999; Van Berkum, Brown, Hagoort, & Zwitserlood, 2003) or the difficulty in lexical access (Kutas & Federmeier, 2000).. Studies on compounds are rather few, if studies on derivation and inflection morpheme processing are excluded (e.g., Katz, 1991; Li et al., 1993; Carlisle, 2000; Myers, 2006). However, studies on idioms or English verbal phrase can still provide neurophysiological evidence for morphological decomposition or composition. To begin with, there are two recent studies attempting to measure the combinatorial process itself, using the N400 brain response as an index of lexico-semantic integration of compound constituents (Koester et al., 2007; Zhang et al.2013). Koester et al. (2007) showed that transparent compounds elicited a larger N400 than opaque compounds, suggesting a combinatorial mechanism for transparent compounds (Koester et al., 2007). Another recent 17.

(27) ERP study (Zhang et al.2013) aims to investigate the time course of Chinese idiom comprehension and the effects of compositionality. In the study, Chinese idioms with varying degrees of compositionality and non-idiomatic phrases, primed by their literal interpretations, were visually presented to subjects for performing a semantic judgment task. The results show a graded modulation of the N400 for the Chinese idioms, with stimuli with high compositionality (e.g. ju jing hui shen 聚精會神 “concentrate one's attention and energy on”) eliciting the smallest ERP effects and those with low compositionality (e.g. yao ya qie chi咬牙切齒 “gnash the teeth in anger”) the largest. The result again supported that compositionality may induce larger N400.. To summarize, it has been inferred that the processing of compounds, at least transparent compounds, operates combinatorially. As for the opaque compound, the evidence is still not enough to make the conclusion. When it comes to VOCs in Chinese, which can be separated just like a verbal phrase but at the same time their meanings are opaque, it seems that the situation is more complex. Although there is no present literature to refer to, studies on English verbal phrases might be helpful because English verbal phrases have similar patterns as Chinese VOCs.. Similar to Chinese VOCs’ controversy, there is a considerable linguistic debate on whether verbal phrases (e.g., turn up, break down) are processed as two separate words connected by a syntactic rule or whether they form a single lexical unit. The views differ on 18.

(28) whether meaning (transparency vs. opacity) plays a role in determining their syntactically-connected or lexical status. As linguistic arguments could not reach a consensus, Cappelle et al. (2010) adopted megnetoencephalography (MEG) to address the issue. By applying a multi-feature Mismatch Negativity (MMN) design with subjects instructed to ignore speech stimuli, Cappelle et al. recorded magnetic brain responses to particles (up, down) auditorily presented as infrequent “deviant” stimuli in the context of frequently occurring verb “standard” stimuli. Already at latencies below 200 ms, magnetic brain responses were larger to particles appearing in existing phrasal verbs (e.g. rise up) than to particles appearing in non-existing combinations (e.g. *fall up), regardless of whether particles carried a literal or metaphorical sense (e.g. rise up, heat up). Previous research found that MMN is relatively enhanced if speech is linked to a single word, but relatively reduced in the case of a syntactic and semantic match between two words linked by phrase-structure rules (Pulvermüller & Shtyrov, 2003; Pulvermüller et al., 2008). The increased brain activation to particles in real phrasal verbs reported in Cappelle et al.’s study thus provided neurophysiological support that a congruent verb–particle sequence is not in syntactic relationship but more like a lexical unit.. In short, according to the literatures in both psycholinguistics and neurolinguistics, transparent compounds (i.e. similar to VOPs in Chinese) are processed with a decomposition/integration mechanism. The larger N400 effect thus can be considered as the 19.

(29) cost of integration. On the other hand, the processing mechanism of opaque compounds (i.e. similar to VOCs in Chinese) is still obscure.. 20.

(30) Chapter 3. Methodology. In this chapter, the current experiment is illustrated. First of all, the participants of the experiment are described in section 3.1. Materials are introduced in section 3.2. Settings and procedure of the experiment are reported in section 3.3. Finally, the process of data analysis is illustrated in section 3.4. 3.1 Participants. Thirty-nine Chinese native speakers (20 to 35 years old, mean age = 23, 25 females) were recruited for the experiment. All participants were right-handed according to a simplified version of the Edinburg handedness inventory (Oldfield, 1971). They all had normal or corrected-to normal vision. None of the subjects had neurological/psychiatric disorders. Written informed consent was obtained from all participants before the experiment started. They were paid for their participation after they completed the task of experiment.. 21.

(31) 3.2 Materials The materials were sentences embedded with VO-structured verbs with two factors being manipulated: transparency (Transparent, In between and Opaque) and sentence pattern (Separated, Unseparated). They could be divided into six conditions: OS (opaque, separated), OU (opaque, unseparated), IS (in between, separated), IU (in between, unseparated), TS (transparent, separated), TU (transparent, unseparated). The example sentences are provided in Table 1.. Table 1: Example sentences of the current experiment (TS, transparent, separated; TU, transparent, unseparated; IS, in between, separated; IU, in between, unseparated; OS, opaque, separated; OU, opaque, unseparated). Conditions. Transparency. Sentence structure. Example sentence 司. 機. 熬. 了. 夜. OS. Opaque. Separated. si ji ao le ye ‘The driver stayed up late’ 司機熬夜了. OU. Opaque. Unseparated. si ji ao ye le ‘The driver stayed up late’ 胖子跑了步. IS. In between. Separated. IU. In between. Unseparated. pan zi pao le bu ‘The fat guy ran’ 胖子跑步了 pan zi. pao bu le. ‘The fat guy ran’ 助理犯了. 錯. TS. Transparent. Separated. zhu li fan le cu ‘The assistant made a mistake’ 助理犯錯了. TU. Transparent. Unseparated. zhu li fan cu le ‘The assistant made a mistake’. 22.

(32) For the separated conditions (OS, IS and TS), the VO sequence was interposed with a fixed morpheme le 了 for the following two reasons: (1) it was reported that in the case of inserting frequently used aspectual morpheme like –le 了, -zhe 著, and –guo 過, the VO sequence should be viewed as words (Siewierska et al., 2010); (2) –le 了 was reported to be the most frequently used interposing element for the VO sequence (Smith, 1999; Wang, 2009). As results, all the experimental stimuli had the following sentence structures: {[Subject]N.+[V le O] Com./ Phr.}Sen. or {[Subject] N.+[VO le] Com./ Phr.}Sen. Note: “N” means Noun, “Com.” means Compound, “Phr.” means Phrase and “Sen.” means Sentence.. First, Verb-Object structured verbs were selected for the experiment. It should be noted beforehand that in order to observe lexicalization of the VO sequence, the final division of semantic transparency into three conditions (Opaque/In between/Transparent) was made by the result of a pilot test done by native speakers of Chinese Mandarin. Therefore, the classification of three groups in terms of semantic transparency before the pilot test was only temporary. The materials in the Opaque group and the In between group were mainly from the following references: Smith’s dissertation (1999) , Wang’s (2009) and Zhang’s (2013) 23.

(33) master theses. The materials in Smith’s and Zhang’s were from the corpus of Academia Sinica, with the tags related to the VO sequence, [spv.] and [spo.]1. As for the materials in the Transparent group, PRACTICAL AUDIO-VISUAL CHINESE (新版視聽華語) and the Sinica Corpus were the main resources. A total of 623 critical verbs were selected. They were further eliminated with the following steps. The first step was to control the word frequency, which was done by looking up the log frequency with the Chinese Word Sketch Engine (http://wordsketch.ling.sinica.edu.tw/). There were at least a billion Chinese lexical items in the corpus of gigaword2all, which was constructed by the Institute of Linguistics, Academia Sinica. Verbs with no frequency recorded were filtered out. The rest of the materials (577 verbs in total) were then classified roughly by its grammatical tag in the corpus. In the Opaque group (226 verbs) and the In between group (152 verbs), the verbs were mainly from the literatures mentioned above. The grammatical tags for these two groups of verbs were all intransitive verbs, usually tagged as VA, VB and VH2. As for the Transparent group (199 verbs), since this group was composed of transparent VO sequence, most of the verbs in this group were tagged as VC+Na, with VC referring to a transitive verb and Na referring. 1. 2. [spv] and [spo] are two features designed by the Acdemia Sinica Corpus. [spv] means the Verb while [spo] means the Noun of a separable V N compound e.g. 吃 Vc[+spv]了他的虧 Na[+spo]. [VA] Intransitive Action Verb 動作不及物動詞. [VB] Intransitive-like Action Verb 動作類及物動詞. [VH] Intransitive Stative Verb 狀態不及物動詞. 24.

(34) to a noun. (e.g. 寫[VC]詩[Na] “write a poem”). To finalize the classification of the selected verbs into the Opaque, In between and Transparent conditions, a pilot test on semantic transparency was conducted. The verbs from the temporary classified three groups were equally distributed into four questionnaires for subjects to rate. Three questionnaires contained 144 verbs and one contained 145 verbs, and each of them was rated by 30 Chinese-speaking subjects ranging from 20 to 40 years old (See Appendix A for the details of the semantic transparency questionnaire). The subjects were asked to judge how different a verb’s meaning in use is from the meaning combination of its constituents on a 5-point scale (1: least different, 5: very different). Taking chi cu 吃醋“being jealous ” as an example, since chi means “eat” and cu means “vinegar”, the combination of chi cu is “eating vinegar”, which is very different from its meaning in use “being jealous”. Subject may consequently rate this verb 5 (very different). Based on the subjects’ ratings, the mean scores for the verbs in these three groups were: Opaque group =3.65 (SD=.45); In between group =2.40 (SD=.25); Transparent group =1.53 (SD=.14). Statistical analysis on the results showed that there was a significant difference among these three groups, F (2, 58) = 297.02, p<.001. After these three groups (Opaque/In between/Transparent) were formed, other variables that might confound the experiment were further controlled. First of all, word frequency (obtained from the corpus of gigaword2all in Chinese Word Sketch Engine 25.

(35) mentioned above) of all the VO structured verbs in the three groups were equivalent no matter in VO unseparated form, F(2, 58) = 1.65, p=.200 or in VO separated form F(2, 58) = 2.67, p=.078. Moreover, two other pilot tests were conducted to ensure that the degree of concreteness and familiarity were equal among the three groups of verbs (See Appendices B and C for the details of the familiarity questionnaire and concreteness questionnaire). Another two groups of 30 subjects (20-40 years old, native speakers of Chinese Mandarin) were recruited to judge the concreteness and familiarity of the verbs on a 5-point scale (1: least concrete, 5: very concrete; 1: least familiar, 5: very familiar). These two groups of subjects did not participate in the previous transparency pilot test and the later formal experiment. The results of these two pilot tests showed that there were no significant difference among these three groups of stimuli in terms of concreteness, F(2, 58) = 2.21, p=.118, and familiarity degree, F(2, 58) = 1.08, p=.346. Finally, the neighborhood size of the verbs in these three groups was controlled. The value of neighborhood size was computed automatically by adopting python-based Load and Analysis Chinese Corpus-Natural Language Toolkit (LACC-NLTK) based on Academia Sinica Balanced Corpus with tagged texts (http://tm.itc.ntnu.edu.tw/CNLP/?q=node/6). There was no significant difference among the three groups, F (2, 58) = 1.43, p=.247. As soon as the critical VO constructed verbs were finalized, the construction of sentences stimuli was started. All the sentences, including fillers, were built with five characters. In 26.

(36) those critical sentences, the subject of a sentence was always a two-character animate noun. Since all the sentences should be natural enough for the experiment, the subject of a sentence stimulus was chosen to be correlated to its critical verb, e.g. 胖子+跑步了 “The fat guy + ran.” The correlation between the subjects and the verbs of the three groups of sentence stimuli were checked. The statistic results showed that there was no significant difference among the three groups of sentence stimuli in terms of correlation, F(2, 58) = 1.48, p=.236. Furthermore, the number of the strokes of the verbs (F(2, 58) = .42, p=.657) was statistically equivalent across three groups of sentences. Finally, a final pilot test (See Appendix D for the details of the sentence naturalness questionnaire) was conducted to ask a new group of 30 subjects (20-40 years old, native speakers of Chinese Mandarin) to rate the naturalness of the sentences on a 5-point scale (1: least natural, 5: very natural). Ideally, all the sentence stimuli should be rated equally nature across all the conditions, but VO unseparated sentences got statistically higher score compared with VO separated sentences. However, this was not surprising since previous literature had pointed out that the VO unseparated form is the main usage of VOCs (Yi,2007). Though the statistic result showed that there was a main effect of sentence naturalness between VO separated sentences and VO unseparated sentences (F(1, 29) = 54.75, p<.001), and a significant separablility x transparency interaction, F(2, 58) = 4.58, p=.014, with unseparated condition having higher ratings than separated condition , all the sentences could still be counted as natural since the mean score of all the sentences were 27.

(37) higher than 4.37 out of 5 (TS = 4.53, TU =4.65 , IS =4.48 , IU =4.76 , OS =4.37 , OU =4.75). The results of the pilot tests are summarized in table 2. Table 2. The summary of the statistical results for all the controlled variables. Ratings/values. F Value. p. Semantic transparency. F (2, 58) = 297.02. .000**. Word Frequency-VO unseparated. F (2, 58) = 1.65. 1. F (2, 58) = 2.67. .078. F (2, 58) = 2.21. 1. Familiarity of verbs. F (2, 58) = 1.08. 1. Neighborhood size of verbs. F (2, 58) = 1.43. 1. Stroke number of verbs. F (2, 58) = .42. 1. Separability F (2, 58) = 54.75. .000**. Semantic transparency F (2, 58) = .76. 1. Separability x Semantic transparency F (2, 58) = 4.57. .014*. Word Frequency-VO separated VO verbs. Concreteness of verbs. Sentence naturalness. Factors. Semantic transparency (3 levels). Separability (2 levels), Semantic transparency (3 levels). Sentences. Sentence naturalness in VO unseparated form. Semantic transparency (3 levels). F (2, 58) = 2.12. 1. Sentence naturalness in VO separated form. Semantic transparency (3 levels). F (2, 58) = 2.74. .072. Correlation between subjects and verbs. Semantic transparency (3 levels). F (2, 58) = 1.48. 1. Note. p*=<.05, p**=<.001 28.

(38) There were 30 sentence stimuli for each condition, with a total of 180 critical sentences included in the experiment. Ninety fillers were also added, including23 ungrammatical VO separated sentences (e.g. 粽子違著規 zong zi wei zhe gui ‘The rice dumpling is breaking rules’), 23 ungrammatical VO unseparated sentences (麵粉摸彩著 mian fen mo cai zhe ‘The flour is drawing lots’) and 44 non-VO grammatical sentences (蚊子很討厭 wen zi hen tao yan ‘Mosquitos are hateful’). Since a participant could not read the same VO verb in both separated sentence and unseparated sentence, two lists were set up to avoid stimuli repetition. Each list then contained 90 critical sentences and 90 fillers (180 sentences in total) without any repetition on both the verb part and the noun part of a VO constructed verb. In addition to the verbs, all the subjects of the sentences were not repeated on the lexical level. The exact sentence numbers in each experimental condition are provided in Table 3. (See Appendix E for a complete list of critical verbs and Appendix F for a complete list of sentence stimuli). 29.

(39) Table 3: The number of the experimental trials in each condition.. List Conditions. Trial no.. Example sentence. List Conditions. Trial no.. Example sentence. Trial number per condition. 1. OS. 15. 司機熬了夜. 2. OS. 15. 小孩說了謊. 30. 1. OU. 15. 小孩說謊了. 2. OU. 15. 司機熬夜了. 30. 1. IS. 15. 胖子跑了步. 2. IS. 15. 衛兵敬了禮. 30. 1. IU. 15. 衛兵敬禮了. 2. IU. 15. 胖子跑步了. 30. 1. TS. 15. 助理犯了錯. 2. TS. 15. 小丑摔了跤. 30. 1. TU. 15. 小丑摔跤了. 2. TU. 15. 助理犯錯了. 30. 1. Fillers. 23. 粽子違著規. 2. Fillers. 23. 粽子違著規. 23. 1. Fillers. 23. 麵粉摸彩著. 2. Fillers. 23. 麵粉摸彩著. 23. 1. Fillers. 44. 蚊子很討厭. 2. Fillers. 44. 蚊子很討厭. 44. 180. List 2 Total. List 1 Total. 180. Total 270. 3.3 Procedure The experiment was carried out in a sound attenuated room. Participants were seated in a comfortable chair in front of a computer screen used for presenting stimuli. All the stimuli were presented with the E-Prime 2.0 software (Psychology Software Tools, Inc.). When the experiment started, the participants were asked to press space bar as soon as they saw the 30.

(40) fixation point (a plus sign, "+") at the center of the computer screen. The fixation point would automatically move on to the stimulus if the space bar was not pressed within 5 sec. After the fixation point, sentences were presented segment-by-segment in serial visual presentation mode at the center of the screen. Each sentence was segmented into two frames: subject (noun) and verb (e.g., [胖子 pan zi “The fat guy”]frame1 +[跑了步 pao le bu “ran”] frame2).. All the stimuli were presented in white against black background, with a visual angle. of less than 1o. The presentation duration of frame 1 and frame 2 was self-paced; an upper limit of 5 sec was set so that the experiment could proceed automatically. Subjects were asked to press space bar on seeing frame 1, but were required to press F or J on seeing frame 2. Since the presentation of frame 2 meant that a sentence had finished displaying, subjects then needed to judge whether the sentence was acceptable or not by pressing the J or F button on the keyboard with their index fingers of the right and the left hand and their response time was recorded. The assignment of response buttons was counter-balanced across participants: half of the subjects used their right hand and the other half used their left hand for “Acceptable” sentences. After the response to the sentence judgment task, there was a 1.5-sec blank and then a central fixation point (a plus sign, "+") would appear again, signaling the start of the next trial. See Figure 1 for the experimental procedure of stimuli presentation. (See Appendix G for the experimental instruction). 31.

(41) Figure 1: Procedure of stimulus presentation. Each trial began with a central fixation point (+), and the subjects were visually presented with a segmented sentence (frame 1+ frame 2= a complete sentence). On seeing the second frame, the participant needed to perform a self-paced sentence judgment task to judge whether the sentence was making sense or not and his/her RT was recorded. After the response to the sentence judgment task, a 1.5-sec trial interval appeared, followed by a central fixation point indicating the upcoming trial.. 32.

(42) Ten practice trials were provided to familiarize participants with the experimental procedure until reliable performance was demonstrated. For the formal experiment, each participant read 180 sentences in total, with 15 sentences from each of the six conditions (90 in total) and 90 filler sentences. The critical and filler sentences were randomized. The experimental time lasted about 15 minutes.. 3.4 Data Analysis. In the current experiment, the E-Prime 2.0 software from Psychology Software Tools, Inc. was used under the Windows XP system to present the experimental stimuli and to record the subjects' behavioral data (i.e. reaction time (RT) and response accuracy). The software SPSS 18 was used to analyze the data collected from subjects. Only data with accuracy rate higher than 95% were included and only trials with correct responses were analyzed. Also, if the reaction time was slower/faster than mean score plus/minus two SDs, the data were excluded. Behavioral data (RTs and accuracy rate) were analyzed with a two-way repeated analysis of variance (ANOVA) with the factors of separability (separated, unseparated) and semantic transparency (Opaque (similar to VOCs in the traditional term), In between, and Transparent (similar to VOPs in the traditional term)). An alpha value of 0.05 (two-tailed) 33.

(43) was adopted, with the Greenhouse-Geisser correction applied when the Mauchly's test of Sphericity was violated. As for follow-up paired t-tests, a two-tailed alpha level of 0.05 was chosen and Bonferroni corrected.. 34.

(44) Chapter 4. Results. The current study recruited 39 subjects for the experiment. Five subjects’ data were excluded due to the fact that their RTs were slower than the criterion (mean score plus two SDs). There was no data excluded due to RTs which are faster than mean score minus two SDs. The data from the rest of the 34 subjects were further analyzed. First of all, 34 subjects’ mean RT and accuracy rate are illustrated in Table 4. The data revealed that the accuracy rates were high among all the conditions. As for the RTs, no matter in separated condition or unseparated condition, the RTs in the transparent groups were the highest. .. Table 4. Behavioral data of the subjects. (OS, opaque, separated; OU, opaque, unseparated; IS, in between, separated; IU, in between, unseparated; TS, transparent, separated; TU, transparent, unseparated). Note: S.D. is indicated in parenthesis Conditions. OS. OU. IS. IU. TS. TU. 799.10. 796.24. 797.50. 800.45. 830.55. 854.65. (144.18). (109.47). (129.72). (132.73). (134.92). (177.59). Accuracy. 98. 99. 98. 99. 97. 99. rate (%). (.03). (.01). (.03). (.02). (.03). (.02). RTs (ms). For the subjects' RTs, a 2 x 3 repeated ANOVA with the factors of separability 35.

(45) (separated, unseparated) and transparency (opaque, in between, transparent) was conducted. The results showed that there was a main effect of transparency (F(2,66)=7.91, p<.005), while no main effect of separability (F(1,33)=.61, p=.44) and no interaction between the two factors (F(2,66)=.59, p=.555) were found. The results of follow-up paired t-test on comparing the VO separated condition and the VO unseparated condition in each semantic transparency level are summarized in Table 5. The results confirmed that there was no significant RT difference between the two conditions.. Table 5. Separability effect: T-test results of reaction time in VO unseparated and separated conditions (OS, opaque, separated; OU, opaque, unseparated; IS, in between separated; IU, in between, unseparated; TS, transparent, separated; TU, transparent, unseparated). Transparency. Separablilty (S/U). Mean(S.D)ms. t(33). p. Opaque (O). OS vs. OU. -2.86 (89.13). -.188. .852. In between (I). IS vs. IU. 2.94 (84.52). .203. .840. Transparent (T). TS vs. TU. 24.09 (137.25). 1.024. .313. On the other hand, the follow-up t-tests on the transparency effect revealed that there were significant differences between the Opaque and the Transparent groups and also between the Transparent group and the In between groups (Opaque vs. In between: t (67) =-.12, p = 1; In between vs. Transparent: t (67) =-3.05, p < .01, Transparent vs. Opaque: t 36.

(46) (33) =3.30, p < .01) The summary of t-test is provided in Table 6.. Table 6. Transparency effect: T-test results of reaction times between conditions. Comparison. t(67). p. Opaque vs. In between. -.12. 1. In between vs. Transparent. -3.05. .009*. Transparent vs. Opaque. 3.30. .006*. (* p < .01) Although there was no interaction between transparency and separability, planned comparisons were conducted to further examine the effect of transparency in the separated and the unseparated conditions. The results showed that there was no RT difference regarding transparency in the separated conditions (F(2,66)=2.01, p=.142 ;), but there was a significant RT difference in the unseparated conditions (F(2,66)=6.54, p<.005. Follow-up t tests in the unseparated conditions revealed that the transparent condition took longer to process than the other two conditions (TU>OU=IU) (Opaque vs. In between: t (33) =-.31, p = 1; In between vs. Transparent: t (33) = -2.58, p < .05, Transparent vs. Opaque : t (33) =3.08, p < .05). The results are summarized in Table 7.. 37.

(47) Table 7. Transparency effect in separated and unseparated conditions: T-test results of reaction times between conditions (OS, opaque, separated; IS, in between separated; IS, in between separated; TS, transparent, separated; TS, transparent, separated; OS, opaque, separated /OU, opaque, unseparated; IU, in between, unseparated; IU, in between, unseparated ; TU, transparent, unseparated; TU, transparent, unseparated OU, opaque, unseparated). Comparison. t(33). p. Comparison. t(33). p. OS vs. IS. .09. 1. OU vs. IU. -.31. 1. IS vs. TS. -1.69. 1. IU vs. TU. -2.58. .042*. TS vs. OS. 1.61. .34. TU vs. OU. 3.08. .012*. (* p < .05). As to the accuracy rate of subjects' responses, a two-way repeated measures ANOVA with the factors of semantic transparency and separability showed that there was a main effect of separability (F (1, 33)=4.33, p=.045), with the unseparated condition having a higher accuracy rate than the separated one. There was no main effect of transparency (F(2, 66)=.39, p=.594) and the interaction between the two factors did not reach significance, either (F(2,66)=1.20, p=.293). To find out if transparency played a role in processing the VO sequence in the separated and the unseparated conditions, planned comparisons between separated and unseparated conditions at each semantic transparency level were conducted and are summarized in Table 8. As Table 8 shows, no accuracy difference was present. 38.

(48) between separated and unseparated condition in all the semantic transparency conditions. (In the separated condition: Opaque vs. In between: t (33) =-.94, p = 1; In between vs. Transparent: t (33) =1.15, p =.771, Transparent vs. Opaque: t (33) =-1.00, p =.975 ; in the unseparated conditions:. Opaque vs. In between: t (33) =.70, p = 1; In between vs.. Transparent: t (33) =-.32, p =1, Transparent vs. Opaque: t (33) =-.44, p =1). Table 8. Transparency effect in separated and unseparated conditions: T-test results of accuracy rates between conditions: (OS, opaque, separated ; IS, in between separated; IS, in between separated; TS, transparent, separated; TS, transparent, separated; OS, opaque, separated /OU, opaque, unseparated; IU, in between, unseparated; IU, in between, unseparated ; TU, transparent, unseparated; TU, transparent, unseparated OU, opaque, unseparated). Comparison. t(33). p. Comparison. t(33). p. OS vs. IS. -.94. 1. OU vs. IU. .70. 1. IS vs. TS. 1.15. .771. IU vs. TU. -.32. 1. TS vs. OS. -1.00. .975. TU vs. OU. -.44. 1. 39.

(49) Chapter 5. Discussion. The purpose of the current study was to empirically examine whether native speakers of Mandarin Chinese process an opaque VO sequence (similar to VOCs in the traditional term) differently from a transparent VO sequence (similar to VOPs in the traditional term). By manipulating semantic transparency as well as the separability of the VO sequence, we hoped to capture the process of lexicalization in the subjects’ mind. To begin with, the results of reaction times showed that the participants reacted faster to unseparated VO structures in the Opaque and In between groups than to those in the Transparent group, but did not have such an RT difference among the three conditions in processing separated VO structures. Since unseparated VO structures in the Opaque and In between groups fall into the traditional categorization of VOCs, our data revealed that compared with transparent VO sequence, VO structures in the Opaque and the In between groups might be processed as a lexical unit without decomposition. Once VO structures were separated, the VO structures in the Opaque and In between groups would be processed like phrases, as those VO structures in the Transparent group. Also, our results revealed that the accuracy rate mirrored the result of sentence 40.

(50) naturalness. It was reported in Chapter Three that VO separated sentences were rated less natural than VO unseparated sentences. In line with the result of sentence naturalness, people were more likely to give wrong answers when the stimuli were separated VO structures. It is not surprising since in previous literature (Yi, 2007), it was attested that the unseparated VO form was the default form of VOCs. However, it should be noted that though the accuracy rate was statistically different between separated and unseparated sentences, the accuracy rate was higher than 97% in all the conditions. As mentioned in Chapter 2, there are three main standpoints about the identity of VOCs: (1) VOCs should be defined as a lexical unit (Chao, 1968; Li & Thompson, 1983; Yi, 2007; Wang, 2009), (2) VOCs that could be used with separate constituents should be treated as phrases (Lu, 1979; Paul, 1988; Sybesma, 1999) and (3) VOCs are words when they are in unseparated form, and are phrases when they are in separated form (Liu1967; Zhang, 2010). Some researchers furthered the last point of view by suggesting that VOCs are in a “middle-state”: they are in a continuum of lexicalization, the process of creating items out of syntactic units (Cabrera et al., 1998). The results of the current study support the third standpoint. It can be seen from the RTs that people indeed treated unseparated opaque VO structured verbs ( as in verbs in Opaque and In between group) as a lexical unit because the time they used to process these items was much shorter than that in processing unseparated transparent VO structured verbs (verbs in the Transparent group). Also, 41.

(51) subjects treated separated opaque VO structured verbs like phrases since there was no significant RT difference among separated VO structured verbs in the Opaque, In between and Transparent categories. It should be mentioned that the In between condition behaved more like the Opaque group in the unseparated condition. Future studies are needed to see if semantic transparency is indeed a defining factor for the status of VOCs and to see if there is a way to better quantify it: how “opaque” a VO structure should be to be treated as a lexical item by language users. From a psycholinguistic viewpoint, the processing mechanism of VOCs in people’s mind can be inferred from the current results. As mentioned in Chapter 2, whether morphologically complex words are stored in the mental lexicon in their full form or whether only their morphemes are stored is a controversial issue. Three related models are: full-listing models (Butterworth, 1983; Bybee, 1995), full-parsing models (Libben, Derwing, & de Almeida, 1999; Taft, 2004; Taft & Forster, 1976) and the dual-route models (Gunter, & Friederici, 2003; Koester, Gunter, & Wagner, 2004, 2007; Zwitserlood, 1994). In the dual-route models, a complex word can either be stored completely or be decomposed into its morphological constituents. Based on this type of models, frequently used compounds and opaque compounds are usually stored in their full form since it would be more efficient to retrieve a whole unit than repeatedly (de)compose it. In contrast, the meanings of transparent and less-frequently used compounds are more likely to be derived by 42.

(52) decomposing/combining its constituents (Sandra, 1990; Zwitserlood, 1994). The results of the current study showed that the processing of the opaque VO sequence can be adequately captured in terms of the dual-route models. The RT difference between Opaque and In between groups(similar to VOCs in traditional term) and Transparent group (similar to VOPs in traditional term) showed that the opaque VO sequence might be stored as a whole unit while the transparent VO sequence need to be decomposed. On the other hand, the equal amount of RT between the opaque VO sequence and the transparent VO sequence in the separated forms showed that, under the influence of interposing elements, the opaque VO sequence might be processed through the decomposition / combinatory route, just like how transparent VO sequences were processed. The result thus reflected that dual-route models are a more plausible approach on explaining the opaque VO sequence processing. It should be noticed that within each semantic transparency condition (Opaque, In between and Transparent), there was no RT difference between separated and unseparated forms, suggesting that the processing of lexical units and short phrases, especially when they are of the same length (3 characters long in the current study), is equally automatic. Future studies using electrophysiological techniques are needed to find out whether there are qualitative differences in brain responses in processing separated vs. unseparated VO sequences despite the quantitative similarity in the RTs.. 43.

(53) Chapter 6. Conclusion. To sum up, the current study adds to the literature of theoretical linguistics and psycholinguistics. It confirms previous theoretical linguists’ view (Chao, 1968; Li & Thompson, 1983; Yi, 2007) and shows that semantic transparency is indeed a crucial factor in defining VOCs. It also shows that separability does play a role in defining the status of VOCs: opaque VO sequences are lexical units when unseparated and are phrases when separated. Finally, the study demonstrates that opaque VO sequences may be accessed via dual routes in the mental lexicon: they are retrieved as a whole unit when they are in an unseparated form but are retrieved morpheme by morpheme when they are in a separated form. Finally, the current study has a limitation. Since a behavioral study can only reflect how fast a participant processes the VO sequences, how he/she processes the inner structure within a VO sequence is still unknown. It was mentioned in Chapter Two that a recent study conducted by Cappelle et al. (2010) showed that English phrasal verbs are perceived as a lexical unit and that the relationship between the verb and the particle is not a syntactic one. Future ERP studies on the VO sequence inner relationship may shed some light on this 44.

(54) issue.. 45.

(55) References Baayen, R. H., Dijkstra, T., & Schreuder, R. (1997). Singulars and plurals in Dutch: Evidence for a parallel dual-route model. Journal of Memory and Language, 37, 94–117. Butterworth, B. (1983). Lexical representation. In B. Butterworth (Ed.), Language production (pp. 257–294). San Diego, CA: Academic Press. Bybee J. (1995). Regular morphology and the lexicon. Language and Cognitive Processes. 425–455. Cappelle, Bert, Shtyrov, Y., & Pulvermuller, F. (2010). Heating up or cooling up the brain? MEG evidence that phrasal verbs are lexical units. BRAIN AND LANGUAGE, 115(3), 189–201. Carlisle, J. F. (2000). Awareness of the structure and meaning of morphologically complex words: Impact on reading. Reading and Writing, 12(3), 169-190. Chao, Yuen-ren. (1948). Mandarin Primer. Cambridge, MA: Harvard University Press. Chao, Yuen-ren. (1968). A Grammar of Spoken Chinese. Berkeley and Los Angeles: University of California Press. Chung, K. S. (2004). Mandarin compound verbs. Universiteit Leiden. Chung K. K. H., T. X., Liu P. D., McBride-Chang C., Meng X. (2010). The processing of morphological structure information in Chinese coordinative compounds: an event-related potential study. Brain Res. 1352, 157-166. Christina L. Gagné , T. L. S. (2008). Constituent integration during the processing of compound words: Does it involve the use of relational structures? Journal of Memory and Language, 20-35. Fang, C.方瑾（2008）。論現代漢語詞素、詞、詞組之界定及其教學啟示。Master thesis, National Taiwan Normal University Department of Chinese as a Second Language.. 46.