Cantonese is not only a lingua franca in Guangdong province, Hong Kong and Macau but also a dialect that is widely used by overseas Chinese communities around the world. Being the most prestigious Yue dialect, Cantonese is definitely an important dialect that is worth investigating. In the past few decades, over a dozen dictionaries on Cantonese have been compiled. Among them, the Guangzhouhua zhengyin zidian 《廣州話正音字典》and Yueyin zhengdu zihui 《粵音正讀字彙》are both widely used and thus influential standards on Cantonese. Despite their importance, systematic studies on the Cantonese pronunciations given in the two dictionaries are few and far between. The present thesis is the first attempt to study the subject. The present dissertation attempts to analyze the differences between the two dictionaries so that a comprehensive list of their differences can be obtained. The analysis is conducted with reference to the number of pronunciations given to each character; the order of listing the pronunciations; tone change; and the listing of commonly used pronunciations. The present thesis also tries to examine the differences between the two editions of Guangzhouhua zhengyin zidian and among the three editions of Yueyin zhengdu zihui. Possible reasons for such differences would also be investigated. It is not surprising to find that many Chinese characters are given different pronunciations in the two dictionaries as the two dictionaries use two different approaches to determine the correct Cantonese pronunciations. The pronunciations recorded in the Guangzhouhua zhengyin zidian are the Cantonese pronunciations commonly used in Guangzhou, Hong Kong and Macau, whilst those recorded in the Yueyin zhengdu zihui are mainly based on the fanqie（反切） recorded in the Guangyun《廣韻》and the Jiyun《集韻》, which were both published in Song Dynasty. Significant differences regarding the criteria of determining the correct Cantonese pronunciations are found in the 3 editions of the Yueyin zhengdu zihui. It is surprising that a tendency of giving more emphasis to the pronunciations commonly used is found in the 2nd and 3rd editions. / published_or_final_version / Chinese Language and Literature / Master / Master of Arts

published_or_final_version / Chinese Language and Literature / Master / Master of Arts

本论文报告了三个研究，旨在探讨声母是否为广东话口语产出过程中的基本加工单元。研究一通过三个不同的实验范式（即，图片命名内隐准备范式、词汇联想内隐准备范式和词图干扰范式）对比了声母和音节的准备效应。结果显示，广东话的声母准备效应比较微弱，只能通过图片命名内隐范式得到；相比而言，广东话音节的准备效应非常稳定，在三个实验范式中都非常明显。研究二旨在探索声母准备效应的时间进程。行为结果显示，只有当准备间隔等于500毫秒时，才能得到稳定的声母准备效应。脑电结果也与此呼应，同质条件和异质条件的神经电活动的差异出现于400-650毫秒之间。具体表现为：同质条件下的关联负变化（contingent negative variation, CNV）波幅在400-450和600-650毫秒间比异质条件下的CNV波幅更大；在异质条件下，CNV电波在700-750毫秒时就从早成分转向了晚成分；而在同质条件下，CNV早成分到晚成分的转换时间提前到了600-650毫秒。研究三旨在考察音节频率是否能够调节声母准备效应。结果显示，声母准备效应只能在低频音节中得到，而无法在高频音节中得到。综合所有的研究结果，在广东话语音编码过程中，1）声母是可以作为基本加工单元的，但其效应受到更高层因素的调节（如：准备时间和音节频率）；2）音节下加工单元（sub-syllable units）能够直接启动发音运动准备过程，而无需等待音节中的其它加工单元。最后，根据本论文的结果，我们建议能够用一种普适词汇产出的观点来调和关于基本加工单元是否为语言特异性的争议。 / Three studies were conducted to investigate whether the onset can be a basic planning unit in Cantonese spoken word production. In Study 1, the onset- and syllable-preparation effects were tested through three frequently used paradigms: the picture-naming implicit form-preparation (picIP) paradigm, the word-associative naming implicit form-preparation (waIP) paradigm and the picture-word interference (PWI) paradigm. The onset-preparation effect was weak in Cantonese and could be obtained only through the picIP paradigm, while the syllable preparation effects were much robust and could be obtained using all of the paradigms. Study 2 was aimed at exploring the time course of onset preparation. The behavioral data showed that the onset-preparation effect was significant only when the foreperiod was 500 ms. The event-related potential (ERP) data echoed this finding; different neural activities between the homogeneous and heterogeneous conditions were found 400650 ms after the time cue. Study 3 showed that the syllable frequency could moderate the onset-preparation effect, which was significant for low-frequency syllables but not for high-frequency syllables. These results indicate that 1) the onset can be a basic planning unit in Cantonese, but its function and effect size are moderated by some higher-level factors, such as preparation time and syllable frequency, and 2) sub-syllable units could directly initiate articulation without waiting for the remaining parts within a syllable. A language-universal position is suggested to reconcile the debates between the language-specific and language-general hypotheses regarding the basic planning units in spoken word production. / Detailed summary in vernacular field only. / Ning, Ning. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 92-98). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese; appendixes includes Chinese. / English abstract --- p.iii / Chinese abstract --- p.v / Acknowledgements --- p.vi / Table of contents --- p.vii / List of tables --- p.x / List of figures --- p.xi / Chapter Chapter 1. --- Introduction --- p.1 / Chapter 1.1. --- The basic planning units in word production --- p.2 / Chapter 1.1.1. --- The phonological encoding process in word production --- p.3 / Chapter 1.1.2. --- The properties of the basic planning units --- p.5 / Chapter 1.1.3. --- Ways to approach the basic planning units --- p.7 / Chapter 1.2. --- Debates surrounding the basic planning units --- p.11 / Chapter 1.2.1. --- Sub-syllable units, syllable units or both --- p.12 / Chapter 1.2.2. --- Language specific vs. language general --- p.18 / Chapter 1.2.3. --- Instantaneous vs. delayed response --- p.20 / Chapter 1.3. --- Characteristics of Cantonese word production --- p.21 / Chapter 1.3.1. --- Phonological properties of Cantonese --- p.21 / Chapter 1.3.2. --- The basic planning units in Cantonese word production --- p.22 / Chapter Chapter 2. --- Overview of the present studies --- p.25 / Chapter 2.1. --- Motivations --- p.25 / Chapter 2.2. --- The present studies --- p.27 / Chapter Chapter 3. --- Study 1 - Onset effects in Cantonese word production - the sensitivity of different paradigms --- p.30 / Chapter 3.1. --- Aim and predictions --- p.30 / Chapter 3.2. --- Materials --- p.31 / Chapter 3.3. --- Experiment 1 - The picture-naming IP paradigm --- p.33 / Chapter 3.3.1. --- Method --- p.33 / Chapter 3.3.2. --- Results --- p.36 / Chapter 3.4. --- Experiment 2 - The word-associative naming IP paradigm --- p.38 / Chapter 3.4.1. --- Method --- p.38 / Chapter 3.4.2. --- Results --- p.38 / Chapter 3.5. --- Experiment 3 - The PWI paradigm --- p.40 / Chapter 3.5.1. --- Method --- p.40 / Chapter 3.5.2. --- Results --- p.41 / Chapter 3.6. --- Discussion --- p.43 / Chapter 3.6.1. --- The onset as a basic planning unit --- p.43 / Chapter 3.6.2. --- The “rightward incremental“ pattern in unit association --- p.44 / Chapter 3.6.3. --- Independence between the word-length effect and the preparation effect --- p.46 / Chapter Chapter 4. --- Study 2 - The time course of onset preparation --- p.48 / Chapter 4.1. --- Aim and predictions --- p.48 / Chapter 4.2. --- Materials --- p.51 / Chapter 4.3. --- Experiment 4 - Evidence from a behavioral experiment --- p.52 / Chapter 4.3.1. --- Pilot experiment --- p.52 / Chapter 4.3.2. --- Method --- p.53 / Chapter 4.3.3. --- Results --- p.54 / Chapter 4.4. --- Experiment 5 - Evidence from an ERP experiment --- p.55 / Chapter 4.4.1. --- Method --- p.55 / Chapter 4.4.2. --- Data analyses --- p.58 / Chapter 4.4.3. --- Results --- p.59 / Chapter 4.5. --- Discussion --- p.63 / Chapter 4.5.1. --- The relationship between the behavioral and ERP indices --- p.64 / Chapter 4.5.2. --- ERP correlates of the onset-preparation effect --- p.65 / Chapter 4.5.3. --- A technique issue --- p.67 / Chapter Chapter 5. --- Study 3 - The moderation of syllable frequency on the onset-preparation effects --- p.68 / Chapter 5.1. --- Aim and predictions --- p.68 / Chapter 5.2. --- Experiment 6 --- p.69 / Chapter 5.2.1. --- Method --- p.69 / Chapter 5.2.2. --- Results --- p.71 / Chapter 5.3. --- Experiment 7 --- p.72 / Chapter 5.3.1. --- Method --- p.72 / Chapter 5.3.2. --- Results --- p.73 / Chapter 5.4. --- Discussion --- p.74 / Chapter 5.4.1. --- The “early locus“ hypothesis --- p.74 / Chapter 5.4.2. --- The reversed syllable-frequency effect --- p.75 / Chapter 5.4.3. --- Comparisons with other studies --- p.76 / Chapter Chapter 6. --- General discussion --- p.78 / Chapter 6.1. --- What is new in the present study? --- p.79 / Chapter 6.1.1. --- A sensitive paradigm --- p.79 / Chapter 6.1.2. --- A sensitive foreperiod --- p.80 / Chapter 6.1.3. --- Neural correlates of the onset-preparation effect --- p.81 / Chapter 6.1.4. --- Locus of the syllable-frequency effect --- p.82 / Chapter 6.2. --- Contributions to the three debates --- p.83 / Chapter 6.2.1. --- Syllable vs. sub-syllable units --- p.83 / Chapter 6.2.2. --- Language specific vs. language general --- p.85 / Chapter 6.2.3. --- Instantaneous vs. delayed response --- p.86 / Chapter 6.3. --- Limitations and future directions --- p.88 / Chapter 6.3.1. --- Dissociation between seriality and incrementality --- p.88 / Chapter 6.3.2. --- More time points for foreperiod length --- p.89 / Chapter 6.3.3. --- The lack of onset-preparation effects for high-frequency syllables --- p.89 / Chapter 6.3.4. --- Other moderators and cross-language studies --- p.90 / Bibliography --- p.92 / Appendices --- p.99 / Chapter Appendix A. --- Prime and target pairs for Experiment 2. --- p.99 / Chapter Appendix B. --- Methods and detailed results of pilot study for Experiment 4. --- p.100 / Chapter Appendix C. --- Response words for Experiments 6 and 7. --- p.102

Leung, Ming Ming Grace. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 147-150). / Abstracts in English and Chinese; includes Chinese characters in appendix. / ABSTRACT (ENGLISH) --- p.i / ABSTRACT (CHINESE) --- p.ii / ACKNOWLEDGEMENTS --- p.iii / TABLE OF CONTENTS --- p.iv / Chapter CHAPTER 1 --- INTRODUCTION / Chapter 1.1. --- Background to this Study --- p.1 / Chapter 1.2. --- Purpose and Significance of this Study --- p.2 / Chapter 1.3. --- Organization of the Thesis --- p.3 / Chapter CHAPTER 2 --- LITERATURE REVIEW / Chapter 2.1. --- Introduction --- p.5 / Chapter 2.2. --- /n/ and /l/ in English and Cantonese --- p.5 / Chapter 2.3. --- /n/ and /l/ Variation in Cantonese --- p.6 / Chapter 2.4. --- /n/ and /l/ Variation in Hong Kong English --- p.7 / Chapter 2.5. --- Language Transfer --- p.10 / Chapter 2.6. --- English Perception and Production by L2 Speakers --- p.20 / Chapter 2.7. --- Task Type --- p.34 / Chapter 2.8. --- Linguistic Environment --- p.40 / Chapter 2.9. --- Language Proficiency --- p.43 / Chapter 2.10. --- Summary --- p.45 / Chapter CHAPTER 3 --- METHODOLOGY / Chapter 3.1. --- Introduction --- p.48 / Chapter 3.2. --- Research Questions --- p.48 / Chapter 3.3. --- Research Methodology --- p.50 / Chapter 3.3.1. --- Participants --- p.50 / Chapter 3.3.2. --- Materials and Procedure --- p.50 / Chapter 3.3.3. --- Data Recording and Data Transcription --- p.56 / Chapter 3.3.4. --- "Data Coding, Data Scoring and Data Analysis" --- p.56 / Chapter 3.4. --- Summary --- p.57 / Chapter CHAPTER 4 --- FINDINGS / Chapter 4.1. --- Introduction --- p.59 / Chapter 4.2. --- General Results in Descriptive Statistics --- p.60 / Chapter 4.2.1. --- General Finding of English Production and Perception --- p.60 / Chapter 4.2.1.1. --- Production of /n/and /l/ in Different Task Types --- p.63 / Chapter 4.2.1.2. --- Production of /n/ and /l/ in Different Following Linguistic Environments --- p.66 / Chapter 4.2.1.3. --- Production of /n/ and /l/ with the Presence of a Second /n/ --- p.68 / Chapter 4.2.1.4. --- Production of /n/ and /l/ by Three Groups of Participants --- p.69 / Chapter 4.2.2. --- General Finding of Cantonese Production and Perception --- p.72 / Chapter 4.3. --- Results of VARBRUL Statistics --- p.76 / Chapter 4.3.1. --- Result of Accurate Production of /n/ --- p.81 / Chapter 4.3.2. --- Result of /n/ to /l/ Variation --- p.83 / Chapter 4.3.3. --- Result of Accurate Production of /l/ --- p.84 / Chapter 4.3.4. --- Result of /l/ to /n/ Variation --- p.85 / Chapter 4.4. --- Descriptive and VARBRUL Statistical Results regarding Four Research Questions --- p.87 / Chapter 4.4.1. --- Data for Research Question 1: To what extent does /n/ and /l/ variation of Cantonese play a role in /n/ and /l/ variation in the onset positionin English? --- p.88 / Chapter 4.4.2. --- Data for Research Question 2: How significant is English perception for English production? --- p.91 / Chapter 4.4.3. --- "Data for Research Question 3: How do Cantonese speakers of English perform in different task types (word list, passage, and natural speech) and is task type significant for English production?" --- p.94 / Chapter 4.4.4. --- "Data for Research Question 4: Which linguistic and/or non-linguistic factors, such as the linguistic environment, the presence of a second /n/ within the same syllable, and English proficiency/university major, may have an influence on /n/ and /l/ production in English?" --- p.95 / Chapter 4.5. --- Summary --- p.99 / Chapter CHAPTER 5 --- DISCUSSION / Chapter 5.1. --- Introduction --- p.100 / Chapter 5.2. --- Research Question 1: To what extent does /n/ and /l/ variation of Cantonese play a role in /n/ and /l/ variation in the onset position in English? --- p.100 / Chapter 5.2.1. --- Discussion about /n/ and /l/ Variation in Cantonese --- p.101 / Chapter 5.2.2. --- Discussion about /n/ and /l/ Variation in English --- p.109 / Chapter 5.2.3. --- Transfer of /n/ and /l/ Variation in Cantonese to /n/ and /l/ variationin English --- p.113 / Chapter 5.3. --- Research Question 2: How significant is English perception for English production? --- p.117 / Chapter 5.4. --- "Research Question 3: How do Cantonese speakers of English perform in different task types (word list，passage, and natural speech) and is task type significant for English production?" --- p.124 / Chapter 5.5. --- "Research Question 4: Which linguistic and/or non-linguistic factors, such as the linguistic environment, the presence of a second /n/ within the same syllable, and English proficiency/university major, may have an influence on /n/ and /l/ production in English?" --- p.128 / Chapter 5.6. --- Summary --- p.137 / Chapter CHAPTER 6 --- CONCLUSION / Chapter 6.1. --- Introduction --- p.142 / Chapter 6.1.1. --- Pedagogical Implication --- p.142 / Chapter 6.1.2. --- Limitations of the Current Study --- p.144 / Chapter 6.1.3. --- Suggestions for Further Studies --- p.145 / REFERENCES --- p.147 / APPENDIX A / APPENDIX B / APPENDIX C / APPENDIX D

University of Macau / Faculty of Social Sciences and Humanities / Department of Chinese