121 55 3MB
English Pages 262 [278] Year 2022
Studies in Chinese Language and Discourse
Cantonese GIVE and Double-Object Construction
Andy Chi-on Chin
15 John Benjamins Publishing Company
Cantonese GIVE and Double-Object Construction
Studies in Chinese Language and Discourse (SCLD) issn 1879-5382 The Studies in Chinese Language and Discourse book series publishes works of original research on Chinese from a linguistic, cognitive, socio-cultural, or interactional perspective. We welcome contributions based on systematic documentation of language structure which displays fresh data and analysis from such areas as corpus linguistics, grammaticalization, cognitive linguistics, sociolinguistics, discourse and grammar, conversation analysis, and typological and comparative studies. Both monographs and thematic collections of research papers will be considered. For an overview of all books published in this series, please see benjamins.com/catalog/scld
Executive Editor Hongyin Tao
University of California, Los Angeles
Co-editors K.K. Luke
Nanyang Technological University
Li Wei
UCL Institute of Education
Volume 15 Cantonese GIVE and Double-Object Construction Grammaticalization and word order change by Andy Chi-on Chin
Cantonese GIVE and Double-Object Construction Grammaticalization and word order change
Andy Chi-on Chin The Education University of Hong Kong
John Benjamins Publishing Company Amsterdam / Philadelphia
8
TM
The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.
doi 10.1075/scld.15 Cataloging-in-Publication Data available from Library of Congress: lccn 2022012135 (print) / 2022012136 (e-book) isbn 978 90 272 1104 0 (Hb) isbn 978 90 272 5780 2 (e-book)
© 2022 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Company · https://benjamins.com
Table of contents
Abstract
ix
Glossary and transcriptions
xi
Acknowledgements
xiii
Part I. Grammaticalization of GIVE in Cantonese Chapter 1 Introduction3 1.1 The double-object verb GIVE and its linguistic features 3 1.2 giving in the linguistic sense 5 1.3 Terminology: Double-object, ditransitive, threeplace predicate and dative 7 1.4 Syntactic realization of double-object construction 10 1.4.1 Word order of IO and DO 10 1.4.2 Marked and unmarked double-object constructions 11 1.5 Other syntactic functions performed by GIVE 14 1.6 Organization of this book 16 Chapter 2 Multi-functionality of GIVE in Chinese dialects and neighboring non-Sinitic languages: An areal-typological perspective 17 2.1 Linguistic situation of Southeast Asia 17 2.1.1 Relationship between Chinese and Southeast Asian languages 17 2.1.2 The Southeast Asian linguistic area 20 2.2 A survey of GIVE in Chinese dialects 21 2.2.1 The Yue dialects 21 2.2.2 The Hakka dialects 22 2.2.3 The Min dialects 23 2.2.4 The Gan dialects 25 2.2.5 The Xiang dialects 25 2.2.6 The Wu dialects 26 2.2.7 The Mandarin dialects 27
vi Cantonese give and Double-Object Construction
2.3 Multiple forms of GIVE as a result of language contact 28 2.3.1 Inter-dialectal influence 28 2.3.2 Inter-lingual influence 29 2.4 Multi-functionality of GIVE in Chinese dialects and neighboring non-Sinitic languages 33 2.4.1 Linguistic contact between Northern dialects and Altaic languages 33 2.4.2 Linguistic contact between Southern dialects and Southeast Asian languages 34 2.5 Multi-functions of GIVE in world’s languages 36 2.5.1 As an IO marker or a beneficiary marker 36 2.5.2 As a causative verb 36 2.5.3 As a passive marker (and a causative marker) 37 2.6 Summary 37 Chapter 3 Grammaticalization of GIVE in Cantonese39 3.1 What is grammaticalization? 39 3.2 Grammaticalization of GIVE 41 3.2.1 As an indirect object marker 41 3.2.2 As a beneficiary marker 41 3.2.3 As a causative verb 43 3.2.4 As a passive marker 45 3.2.5 As a verb introducing instruments 49 3.3 Summary 51 Chapter 4 Diachronic development of GIVE and its functions in Cantonese 4.1 Studying the language of the past with authentic textual materials 55 4.2 Pre-modern Cantonese dialect materials 58 4.3 Functions of GIVE in pre-modern Cantonese 60 4.3.1 As a double-object verb 61 4.3.2 As a causative verb 73 4.3.3 As an IO marker 74 4.3.4 As a passive marker 90 4.3.5 As a verb introducing instruments 95 4.4 Summary 99
55
Table of contents vii
Part II. Word order change in Cantonese double-object construction Chapter 5 Word order typology of the double-object construction in Chinese dialects 105 5.1 Typological features in Northern and Southern Chinese grammar 105 5.2 Typology of double-object construction in Chinese dialects 107 5.2.1 The IO DO pattern 107 5.2.2 The DO IO pattern 108 5.3 Syntactic stratification in double-object construction 109 5.4 Double-object construction in Beijing Mandarin 110 5.4.1 The V DO 給 IO pattern in Beijing Mandarin 112 5.4.2 The 給 IO V DO pattern: A beneficiary or a double-object construction? 114 5.4.3 Summary 118 5.5 Double-object construction in the Yue dialects 119 5.6 Summary 122 Chapter 6 Relationship between IO DO and DO IO patterns 125 6.1 Transformational approach 126 6.1.1 Does Cantonese have dative shift? 126 6.2 Discourse approach 129 6.2.1 Discourse approach on Chinese double-object construction 130 6.2.2 Discourse approach on Cantonese double-object construction 131 6.3 Cognitive linguistics approach 134 6.4 An alternative explanation: A loan feature resulting from language contact 135 Chapter 7 On-going word order change in Cantonese double-object construction 7.1 Sociolinguistic situation of Hong Kong 137 7.1.1 Status of Putonghua and Modern Standard Chinese in Hong Kong 140 7.1.2 What does “Chinese” mean in Hong Kong? 141 7.1.3 Contact with mainland China after the 1970s 141 7.1.4 Language attitude toward Putonghua 143 7.2 Fieldwork study on Cantonese double-object construction 144 7.3 Fieldwork data on Cantonese double-object construction 147 7.3.1 Background information of the forty informants 147 7.3.2 The production task 149 7.3.3 The perception task 160 7.4 Development of Cantonese double-object construction 172
137
viii Cantonese give and Double-Object Construction
Chapter 8 Concluding remarks and future work
173
References177 Appendices201 Appendix 1 Survey of the syntactic functions of GIVE in Chinese dialects and neighboring non-Sinitic languages 201 Appendix 2 List of pre-modern Cantonese dialect materials 231 Appendix 3 Frequency distribution of double-object patterns in Wang Shuo’s corpus 233 Appendix 4 Double-object sentences with the IO DO pattern found in pre-modern Cantonese dialect materials 235 Appendix 5 Sentences used in the production task 243 Appendix 6 Sentences used in the perception task 251 Appendix 7 Information sheet for the fieldwork of the double-object construction in Hong Kong Cantonese 253 Appendix 8 Non-native double-object sentence patterns used by the informants in the production task 257 Index259
Abstract
GIVE is a versatile morpheme in many languages. Syntactically, it is a verb that subcategorizes two objects, and the three-argument structure gives rise to a number of constructions with inter-related meanings as a result of grammaticalization and semantic extension of GIVE. The constructions that have been discussed the most include the double-object construction, the causative construction, and the passive construction. While there have been extensive studies on the interplay between the syntax and semantics of GIVE in many languages (see, for example, Newman, 1996, 1998), not much has been done in a similar manner on Cantonese, a member of the Yue dialect group of the Chinese language family. This monograph reports on the study of the morpheme GIVE and its associated functions and syntactic constructions in Cantonese from diachronic, synchronic, and typological perspectives. Drawing on cross-linguistic data, and 19th century dialect materials, this study first traces the chronological development of the various functions played by GIVE in Cantonese: The indirect object marker, the passive marker, and the instrument marker which was relatively under-explored in previous studies. The study then examines the double-object construction in terms of its typological contrast between the Northern and Southern dialects of Chinese with fieldwork data and corpus data. The typological contrast provides the basis for discussing the use of the northern pattern in Cantonese as a result of the increasing influence of Putonghua and Modern Standard Chinese. The extent and the rate of the influence were studied by means of a sociolinguistic survey with 40 native speakers of Cantonese.
Glossary and transcriptions
Abbreviation
Meaning
3sg acc agent art asp causative cl dat de DO IO jiang nom passive past pl prf sfp subj suf VP Vst □
third person singular accussative case agent marker article aspect causative marker classifier dative marker attributive particle de 的 direct object indirect object disposal marker jiang 將 nominative case passive marker past tense plural marker prefix sentence final particle subject marker suffix verb phrase stative verb Morphemes with no Chinese character equivalents
Notes on transcriptions Chinese words in the text are transcribed with Hanyu pinyin 漢語拼音 (in italics). Words that are pertinent to the discussion of a particular dialect will be transcribed with IPA (placed in square brackets and without tonal designation). When discussing the Yue dialects in general, sentences are transcribed with IPA (based on Zee (2001) with tonal designation) on the basis of the pronunciations in Hong Kong Cantonese. Therefore, the same word may be trascribed differently in the text depending on the context of the discussion.
Acknowledgements
The study presented in this book is largely based on my doctoral dissertation (Chin, 2009) completed at the Department of Asian Languages and Literature, University of Washington (UW). The idea of working on the double-object construction can be dated back to my second year of study at UW when I took a graduate seminar class on Chinese dialectal grammar with my mentor, Professor Anne Yue. In this class, I was introduced the topics on linguistic typology, stratification of grammar, language change, language contact, etc. I thus chose to work on the double-object construction in Hong Kong Cantonese by looking at the alternation of the two typological patterns found among young speakers of Cantonese. When I returned to Hong Kong after the qualifying examination, I continued working on the double-object construction as my dissertation topic. I also came across the research on the grammaticalization of the verb give in other languages. I then started to examine the case in Cantonese with some pre-modern dialect materials available to me. The study presented in the dissertation and in this book thus cannot be made possible without the support, guidance and advice from my teachers, colleagues and friends. First and foremost, I owe my deepest gratitude to my supervisor, Professor Anne Yue. Professor Yue has been my best navigator and a good role model for being a serious and rigorous scholar. Without her guidance and patience, I would have long lost track in the gigantic academic jungle and could not complete this work. I would also like to thank the late Professor Jerry Norman and Professor Chris Hamm for their advice and constructive comments on my research as well as studies on Chinese historical phonology, and Chinese vernacular fictions. Of almost equal importance is Sir Edward Youde Memorial Fund Council, which generously provided me with a three-year fellowship to pursue my studies at UW. Lady Pamela Youde was so kind to send me her greetings at festive occassions. Her personal touch has always been heart-warming. I was very fortunate to benefit from many gifted and experienced scholars in UW. These teachers widely broadened my scope of knowledge in different fields of study and I would like to thank them for their mentoring: Professor Alicia Beckford, the late Professor Judith Boltz, Professor William Boltz, Professor Chris Hamm, Professor Zev Handel, Professor Sharon Hargus, Professor David Knechtges,
xiv Cantonese give and Double-Object Construction
Professor Frederick Newmeyer, the late Professor Jerry Norman, Professor Michael Shapiro and Professor Richard Wright. I would also like to thank Professor Samuel Cheung, Professor Anne Yue, Professor Benjamin Tsou, Professor Bit-chee Kwok, Dr Shin Kataoka and Dr Carine Yiu for sharing with me their collections of pre-modern Cantonese dialect materials for studying the linguistic featuers of give. Besides early textual materials on Cantonese, my study of give also examines the oracle-bone inscriptsions (OBI), the earliest record of the Chinese language. I had the fortune to study OBI and bronze inscriptions with Professor Ken-ichi Takashima from Univesity of British Columbia. I would like to thank him for providing me with constructive comments on that part of the work. The research work of my dissertation was conducted while I was working as a research assistant in my alma mater, City University of Hong Kong. I must thank my supervisor, Professor Benjamin Tsou, for giving me all kinds of flexibility while working in his research centre. I also want to express my gratitude to Professor Cheung Hintat at The Education University of Hong Kong, who has given me a lot of encouragement and support for developing Cantonese studies from the perspective of Digital Humanities. My heartfelt thanks also go to my friends and colleagues in Seattle and Hong Kong for their support and assistance in my study and the preparation of this book manuscript: Dominic Lau, Dora Leung, Angela Leung, Steve Lee, Alex Chan, Eric Siu, Deng Lin, Chang Jung-im, Kenny Mok, Tsoi Wingfu, Cherry Yeung. I would like to thank Ms. Ineke Elskamp and her team at John Benjamins for their professional editorial assistance. Last but not least, I have to express my sincere thanks to the editors of the book series Studies in Chinese Language and Discourse: Professor Tao Hongyin, Professor Kang-Kwong Luke and Professor Li Wei, for their interest in publishing my work.
Part I
Grammaticalization of give in Cantonese
Chapter 1
Introduction
give is a versatile morpheme in many languages. Syntactically, it is a verb that subcategorizes two objects, and the three-argument structure gives rise to a number of constructions with inter-related meanings as a result of grammaticalization and semantic extension of give. While there have been extensive studies on the interplay between the syntax and semantics of give in many languages (see, for example, Newman, 1996, 1998, Margetts and Austin, 2007, Malchukov, Haspelmath, and Comrie, 2010), not much has been done in a similar manner on Cantonese, which is usually referred to as the standard variety of the Yue dialect group of the Chinese language family (see, for example, Yue-Hashimoto, 1972, 1991b, J. Yuan, 2001, Zhan, 2004). Drawing on cross-linguistic data, pre-modern dialect materials, and fieldwork data, this study analyzes the chronological development of the various functions played by give in Cantonese, and the increasing use of the non-native pattern of the double-object construction by younger speakers of Cantonese. 1.1
The double-object verb GIVE and its linguistic features
Newman, in his series of studies on the linguistics of giving, argued that the double-object verb GIVE is “experientially basic and semantically complex” (Newman, 1998, p. vii) [italics original]. GIVE is basic because it constitutes one of the core words in many languages (Newman, 1996). In the language of the Dyirbal tribe of Australian aborigines, Dixon (1973) found that some syntactic properties can only be taken up by the nuclear verb of giving, i.e., the verb GIVE, but not other verbs of giving. According to Dixon, speakers of the Dyirbal tribe know two languages: Dyalŋuy, which is used “in the presence of a taboo relative” (p. 206), such as the mother-in-law of a male speaker, and the other is the everyday language, which is used in all other circumstances. The major difference between these two languages is that the former has a restricted set of vocabulary. It only has the core lexical item among the synonyms that can be found in the everyday language. For example, the everyday language has four verbs related to GIVE: bilan ‘to take over to, to send’, gibin ‘to provide food for relatives’, gulŋgan ‘to breastfeed’ and wugan ‘to give’ (Dixon, 1973). Among these four verbs,
4 Cantonese give and Double-Object Construction
wugan is the core item because it can “have a generic meaning, implying any type of giving” (p. 207). It is also the only verb of giving found in Dylaŋuy. In the British National Corpus which consists of 100 million words of contemporary written and spoken English texts (Leech et al., 2001), the verb GIVE is the 72nd high frequency word out of around 6,000 words with frequency greater than 10 (per million words). As a verb, it is the 17th verb out of 1,100 verbs with frequency greater than 10 (per million words). A similar phenomenon is also observed in the Chinese language. According to the data from the synchronous corpus livac,1 the verb GIVE (gei 給) is the first double-object verb in the corpus. In Hong Kong Cantonese, the GIVE verb 畀 [pei] is the 74th lexical item in the master word list, and it is the 18th verb in the corpus with a frequency of 1,518.2 GIVE is also a core linguistic item in child language. According to Benedict’s study, ‘give me’ is one of the first few syntactic phrases acquired by children (H. Benedict, 1979). Fletcher et al. (2000) found that the double-object verb [pei] 畀 ‘to give’ is one of the high frequency verbs in Hong Kong Cantonese child language. According to the two corpora of Cantonese children speech (2–3 years old and 2.5–5 years old) developed by Fletcher et al., [pei] 畀 is the 9th and 15th verb respectively and it is also the verb of giving used the most frequently. In addition, Ho’s (2004) experiment showed that Cantonese children as young as 3 years old are already able to comprehend double-object constructions with overt indirect objects and direct objects.
1. livac stands for Linguistic Variation in Chinese Speech Communities [http://www.livac.org]. It is a synchronous Chinese corpus with language data collected from eight major Chinese speech communities: Hong Kong, Taipei, Beijing, Shanghai, Macau, Singapore, Guangzhou and Shenzhen. Up to 2020, livac has processed over 2.7 billion Chinese characters. For details about livac, see Tsou et al. (1997, 2011) and Tsou and Kwong (2015). 2. The data on Cantonese is based on the second phase of The Corpus of Mid-20th Century Hong Kong Cantonese (https://hkcc.eduhk.hk) developed at The Education University of Hong Kong. The corpus (two phases) has about 1,000,000 character tokens. For more details about the corpus, see Chin (2013, 2019, 2021).
1.2
Chapter 1. Introduction
giving in the linguistic sense
The syntactic construction in which the verb give frequently occurs is usually called the double-object construction. Semantically, a double-object construction denotes an act of giving involving three participants, as Newman (1998) claimed. Zhu (1979, p. 82) gave the following semantic profile for the action of giving: [my translation] a. there exist a giver and a recipient; b. an entity is transferred between the giver and the recipient; c. there is a transfer of the entity in (b) from the giver to the recipient. Semantically the three participants are labelled as giver, recipient and thing. Giving involves a change of possession of the thing from the giver to the recipient. In a deeper sense, the giver causes the recipient to possess the thing.3 The concept of causativity has further led to the claim that give can be associated with the causative construction.4 Since the semantic profile of the action of giving requires three participants, a double-object construction becomes ungrammatical if it only has one object (unless in certain contexts in which one of the objects is already understood or shared by the interlocutors). Sentences (1) and (2), which only have the direct object and indirect object respectively, are ungrammatical. (1) *我 送 禮物 wo song liwu I give gift ‘*I gave a gift’ (2) *我 送 他 wo song ta I give he ‘*I give him’
Verbs that have this type of semantic profile possess the [+GIVE] feature and this type of verbs is known as give-type double-object verbs. Typical examples in English include offer, lend, send, bestow, etc. Examples in Chinese include the following (based on Zhu (1979), p. 82): 3. This is what Newman (1998) meant by “the giving is done intentionally” (p. ix). 4. More will be discussed in Chapter 3. Previous works on the correlation between give and causativity can be found in Oehrle (1976), Larson (1988), Pinker (1989), Harley (2002), Pineda and Mateu (2020), among others. For Chinese, see R. Cheng (1974), Yue-Hashimoto (1993a), L. Cheng et al. (1999), S. Tang (2003).
5
6 Cantonese give and Double-Object Construction
Table 1. Verbs with [+GIVE] semantic feature in Chinese
送 song
賣 mai 還 huan 遞 di 付 fu 賞 shang 獎 jiang 嫁 jia 交 jiao 讓 rang 教 jiao 分 fen 賠 pei 退 tui 輸 shu 補 bu 贈 zeng 賜 ci 傳 chuan
To give as a gift To sell To return To pass To pay To award To award To marry off one’s daughter To handover To give / yield To teach To share To compensate To return To lose To make up for To give as a gift To bestow To pass
獻 xian
找(錢) zhao (qian) 塞 sai 許(女兒) xu (nüer) 借 jie 租 zu 分配 fenpei 遺傳 yichuan 傳染 chuanran 過繼 guoji 轉交 zhuanjiao 移交 yijiao 交還 jiaohuan 歸還 guihuan 退還 tuihuan 贈送 zengsong 轉送 zhuangsong 轉賣 zhuanmai 告送 gaosong
To present To give change To tuck to marry off daughter To lend To rent out To allocate To pass on to offspring To infect To let someone adopt his/her child To pass on To handover To return To return To return To give To pass on To resell To tell
In addition, speech verbs such as jiang 講 ‘to tell’, gaosu 告訴 ‘to tell’ are also members of the give-type double-object verb family. Besides the give-type double-object verbs, there is another set of verbs which can be expressed with the double-object construction. Examples include tou 偷 ‘to steal’, qiang 搶 ‘to snatch’, etc. These verbs also sub-categorize two objects but the semantic roles of the two objects are source and thing instead of recipient and thing respectively. For example, in the sentence wo tou le ta yi ge qianbao 我 偷了他一個錢包 ‘I stole a wallet from him’, the direction of the action goes from the indirect object (i.e., ta 他) to the subject while in the sentence wo gei le ta yi ge qianbao 我給了他一個錢包 ‘I gave him a wallet’, the direct object (i.e., qianbao 錢包 ‘the wallet’) goes from the subject to the indirect object ta 他. The resultant state of the entities denoted by the indirect objects between this set of verbs and the give-type verbs is opposite. For the former, the entity represented by the indirect object does not gain but loses the thing while for the give-type verbs, the entity gains or receives the thing. Therefore, verbs like tou 偷 ‘to steal’ and qiang 搶 ‘to snatch’ carry the semantic feature of [+deprive].5 In the present study, focus will be placed on the give-type double-object verbs (i.e., verbs with the [+GIVE] feature).
5. We use [+deprive] instead of [-GIVE] because [-GIVE] also includes monotransitive or intransitive verbs that are not expressed with a double-object construction.
1.3
Chapter 1. Introduction
Terminology: Double-object, ditransitive, three-place predicate and dative
In the above discussion, we used the terms double-object construction and double-object verb. It is noted that in many studies, the term ditransitive construction is also used and the verb GIVE is labeled a ditransitive verb (see, for example, Haspelmath, 2005, 2006, 2013). To some linguists, the term ditransitive construction not only refers to the syntactic construction in which the verb GIVE occurs, but also other constructions which have two arguments in addition to the subject. For example, Michaelis and Haspelmath (2003) made the following comment when they used the term ditransitive construction to refer to the GIVE-construction: By ditransitive constructions we mean constructions with verbs of transfer like ‘give’, ‘send’, ‘show’ which require two objects, a Recipient (or receiver) and a Theme (or patient), i.e., the entity that is transferred. Thus our class of ditransitive verbs does not comprise verbs like ‘put’, ‘fill’ or ‘load’, which in addition to a Theme role also have a Location role. (p. 1)
The above statement implies that there are linguists including verbs like put, fill and load as ditransitive verbs. Crystal (2003), in A Dictionary of Linguistics and Phonetics, defined ditransitive as “A term used by some linguists to refer to a verb which can take two objects, e.g. give (I gave him a book)” [italics is mine] (p. 147). The term ditransitive (either modifying a verb or a construction) therefore covers not only the GIVE-construction, although some linguists such as Haspelmath used this term with a smaller coverage (i.e., referring to the GIVE-construction exclusively). It is also noted that the term ditransitive verb can be taken as an equivalent to three-place predicate, which is defined by Newman (2005, p. 149) with the following properties:6 a. 3 arguments tend to be present in the overt clause structure and tightly integrated into clause structure; b. 3 arguments seem to be semantically salient; c. the semantics are usually associated with caused motion[s] (this could be [an] “abstract motion”) to a person, thing, or location. According to Newman (2005), “give, show, tell, put and their counterparts in other languages” are “core three-place predicates” [italics original] (p. 150). To 6. A similar categorization of three-place predicates in Chinese can be found in F. Xu (2004) and C. Chen (2002), in which verbs of giving, depriving, saying, putting are treated as three-place predicates. Y. Yuan (1998) included only verbs of giving and depriving as three-place predicates.
7
8 Cantonese give and Double-Object Construction
Haspelmath (2005), ditransitive and three-place predicate are two different concepts. Specifically, he argued that “‘three-place predicate’ is not the same as ‘ditransitive predicate’, because placement verbs like put (‘A puts B in C’) are also three-place [predicates], like give (‘A gives B to C’), but they are not ditransitive” (p. 1 & 2). From the above descriptions, we can see that the terms three-place predicate and ditransitive cover constructions which require two arguments in addition to the subject. Besides GIVE, there are other verbs such as put and load which fulfill this condition. Trask (1993) defined a double-object construction as “a construction involving a ditransitive verb like give or tell, such as Lisa gave me a kiss” [italics original] (p. 86).7 As the term suggests, there are two objects in this type of construction, and they are usually referred to as direct object (DO) and indirect object (IO). The latter is defined by Trask as “the grammatical relation expressing the entity which is the recipient or beneficiary of the action of the verb in sentences in which this entity is clearly distinct from a direct object” [italics is mine] (p. 140). A direct object is defined as “the second obligatory argument of a transitive verb, most typically expressing a patient which undergoes the action of the verb” (p. 82). From the above discussions, we can see that both DO and IO in a give-type double-object construction are defined as the arguments with the semantic roles of thing and recipient respectively.8 These semantic roles are assigned to the arguments sub-categorized by the verb GIVE and other related verbs, such as offer, send, etc. Thus, the sentence I put the books on the table has a three-place predicate put and is a ditransitive sentence but not a double-object sentence because the semantic roles played by the arguments are agent, thing and location and there is no recipient role. In other words, a double-object construction is a sub-type of the ditransitive construction in our analysis.9 Besides ditransitive, the term dative is sometimes used to refer to the constructions involving give-type verbs (see, for example, Peyraube 1981, 1988). It is also used by some linguists to distinguish between the sentences I gave you a book and I gave a book to you. The latter is classified as a dative construction because of the preposition to preceding the IO, which is similar to a dative marker.10 7. Note that Trask (1993) did not have an entry of ditransitive construction. 8. Haspelmath in his studies of double-object constructions used the label theme instead of thing. 9. Haspelmath (2005, 2006) used the term double-object construction for those constructions in which both the IO and the DO are syntactically equal (either marked or unmarked) to the DO of monotransitive verbs. In some studies, double-object construction is used to refer to the construction in which the two objects are not marked. 10. The relationship between these two types of sentences is known as dative-alternation. See, for example, Levin (1993).
Chapter 1. Introduction 9
In this study, we do not use the term dative because dative refers to a type of syntactic cases usually assigned to the recipient in a double-object construction. For example, in Turkish, the dative case a is assigned to the IO (Blake, 1994, p. 1). (3) Mehmet adam-a elma-lar-i ver-di Mehmet-nom man-dat apple-pl-acc give-past.3sg ‘Mehmet gave the apples to the man’
In French and Portuguese, the dative case, which is assigned to the IO, is expressed when the IO is a pronoun. (4) Les parents lui ont donne une education soignee the parents him has give a education good ‘The parents gave him a good education’ (5) Joao lhe deu o livro John-subj him-dat gave the book-acc ‘John gave him the book’
(Melis, 1996, p. 40)
(Berlinck, 1996, p. 120)
Modern Chinese, including its dialects, is an isolating language. There is no syntactic case assigned to nouns and pronouns performing different syntactic functions such as subjects and different types of objects (direct, indirect or oblique). We thus do not adopt the term dative in this study. Some Chinese linguists defined a double-object construction as a construction in which the main verb is followed by two noun phrases. For example, sentences (6) and (7) are classified as double-object sentences by Q. Ma (1992).11
(6) 我 踢 他 一 腳 wo ti ta yi jiao I kick him one foot ‘I gave him a kick’
(7) 我 叫 他 哥哥 wo jiao ta gege I call he brother ‘I called him brother’
Based on our previous discussion, these two sentences are not double-object sentences because none of the arguments in these two sentences carries the semantic role of recipient (see, for example, Gu (1999)).12 For (6), the verb ti 踢 ‘to kick’ is a monotransitive verb which subcategorizes one object only. Yi jiao 一腳 ‘one foot’ is best interpreted as a complement (補語) rather than an argument (T. Tang, 1981). 11. Ma listed altogether 14 sentence types which he classified as the double-object construction. 12. Gu also brought up the notion of semantic roles for defining a double-object sentence.
10 Cantonese give and Double-Object Construction
In other words, this sentence is neither a double-object nor a ditransitive sentence. Jiao 叫 ‘to call’ in sentence (7) is a ditransitive verb because it requires two arguments in addition to the subject. However, it is not a double-object construction because the two objects do not have the semantic roles of thing and recipient. 1.4
Syntactic realization of double-object construction
At the syntactic level, there can be more than one realization for the double-object construction. The variation among these patterns usually lies in the word order and the marking of the two objects. 1.4.1 Word order of IO and DO The English sentences I gave you a book and I gave a book to you are double-object sentences with different word order patterns in terms of the two objects. The former has the IO preceding the DO (i.e., the IO DO pattern) while the latter has the reverse (i.e., the DO IO pattern). However, these two types of word order are not necessarily mutually exclusive within the same language. For example, both word order patterns coexist in English. The distribution of these two word order patterns for the give-type double-object construction is asymmetrical across Chinese dialects: The Northern dialects13 have the IO DO pattern such as wo sung ta yi ben shu 我送他 一本書 (I gave him a book)14 while the Southern dialects mainly have the DO IO pattern. For example, the sentence ‘I gave him a book’ in Hong Kong Cantonese (a member of the Yue dialect group) is rendered as [ŋͻ13 pei35 tsͻ35 pun35 sy55 k’ɵy13] 我畀咗本書佢 ‘I gave him/her a book’, which has the DO IO pattern. More will be discussed in Chapter 5 on the word order typology across Chinese dialects. It is observed that the IO DO word order has started to diffuse into some Southern dialects, especially those which have frequent contact with Putonghua, the official language or the lingua franca in Mainland China, which is based on the Beijing dialect, a member of the Northern dialect group. It is hypothesized that language contact is the key factor contributing to this phenomenon. Hong Kong,
13. Norman (1969) claimed that “several of the isoglosses separating North and South linguistically fall roughly along the Yangtze River” (p. 7). The Northern group is known as the Mandarin dialect while the Southern group comprises the dialect groups of Yue, Min, Hakka, Xiang, Wu and Gan.
14. In Chapter 5, we will show that the DO IO pattern, such as wo song yi ben shu gei ta 我送一 本書給他 ‘I gave a book to him’ is not a native pattern in the Northern dialects.
Chapter 1. Introduction 11
as located in the Southern coastal region of China, shows no exception toward this linguistic change due to its sociolinguistic situation of biliteracy and trilingualism.15 This non-native IO DO pattern is now gradually used by young and educated speakers. It is thus essential and meaningful for us to look into the extent of this on-going syntactic change. 1.4.2 Marked and unmarked double-object constructions Besides word order, another feature pertinent to the double-object construction is whether the objects are marked by prepositions, case markers or postpositions. In the sentence I gave you a book, neither objects are marked by any syntactic markers. We categorize it as an unmarked double-object construction. The IO you in the sentence I gave a book to you is marked by the preposition to and this sentence is a marked double-object construction. In standard Mandarin, the IO can be either marked or unmarked. For example, in the sentences wo gei ta song yi ben shu 我給他送一本書 ‘I give him a book’, and wo song gei ta yi ben shu 我送給他一本書 ‘I give him a book’, the IO ta 他 ‘he’ is preceded by gei 給, which is the same as the double-object verb GIVE in standard Mandarin. There is another double-object sentence wo song ta yi ben shu 我送他 一本書 ‘I give him a book’, in which the IO is not preceded by gei 給.16 Unlike English, in which the morpheme before the IO is unambiguously a preposition, there have been debates over the grammatical category of gei 給 before the IO. Yue-Hashimoto (1971), Peyraube (1988) and S. Tang (1998) among others treated gei 給 as a preposition while C. Li and Thompson (1981) argued that the morpheme is a syntactic unit resulting from grammaticalization and they called it a co-verb. C. Huang and Ahrens (1999), on the contrary, argued that the morpheme in question is still a verb and treated the V DO 給 IO pattern as a serial verb construction.17
15. Some discussions on the linguistic situation and language policies in Hong Kong especially after its handover to China in 1997 can be found in Tsou (1997), Bolton and Bacon-Shone (2008), Bacon-Shone et al. (2015), D. Li and Leung (2020).
16. Zhu (1979) claimed that wo song ta yi bun shu 我送他一本書 is a reduced form of wo song gei ta yi ben shu 我送給他一本書.
17. J. Ting and Chang (2004) argued that the gei-phrase in the double-object construction cannot be modified by adverbs while the second verb phrase in a serial verb construction can be modified by adverbs. In their analysis, the gei-phrase is not a genuine verb phrase.
12 Cantonese give and Double-Object Construction
The analysis of the V DO 給 IO pattern as a serial verb construction is not convincing because of the following:
a. The V DO 給 IO sentence, when analyzed as a serial verb construction, has the syntactic structure V1 NP1 V2 NP2 with gei 給 as V2. In a serial verb construction, there is no sub-categorization relationship, neither syntactic nor semantic, between the two verb phrases. In other words, it is not ungrammatical for a sentence to have only V1 NP1 but without V2 NP2 or vice versa. For example, wo mai youpiao ji xin 我買郵票寄信 ‘I buy a stamp and then send a letter’ is a serial verb construction with two verb phrases mai youpiao 買郵票 ‘buy a stamp’ and ji xin 寄信 ‘send a letter’. Leaving out one of these two verb phrases does not make the sentence ungrammatical. On the contrary, wo song yi ben shu 我送一本書 ‘I gave a book’ is ungrammatical without the gei-phrase containing the recipient. In other words, the gei-phrase is an obligatory constituent in a double-object construction. b. The morpheme preceding the recipient is always the morpheme cognate with the double-object verb GIVE, such as gei 給 in standard Mandarin.18 We do not have cases in which the recipient is preceded by other verbs of giving such as song 送 ‘to give’, di 遞 ‘to pass’, jiao 交 ‘to handover’, etc. This is also the reason why some linguists argued that gei 給 is not a verb but a preposition-like element. In addition, it is also noted that gei 給 always precedes the IO (either pre-verbally or post-verbally) and its function is to introduce the IO in the double-object construction. In this regard, we label gei 給 in the gei-phrase an indirect object marker (IO marker in short). c. It is also noted that non-double-object verbs (i.e., those without the semantic feature of [+GIVE]) can also be expressed with the V1 NP1 給 NP2 structure, as in the sentence wo da le yi jian maoyi gei ta 我打了一件毛衣給他 ‘I knitted a sweater for him’. However, these verbs do not obligatorily subcategorize two objects and NP2 after gei 給 is thus not an argument of the main verb. This type of sentence, though also bearing a gei-phrase, is known as a beneficiary construction in standard Mandarin.19 The crucial distinction between the double-object and the beneficiary construction is the semantic feature of the main verb. The former has the [+GIVE] feature while the latter does not. d. Chinese dialects do not use any marker before the DO (i.e., the so-called “DO marker”) in the double-object construction. Haspelmath (2013) found that 18. The same can be found in other Chinese dialects and this will be discussed in Chapter 2.
19. Another difference between the two types of sentences is that gei 給 can be substituted by other verbs in a beneficiary construction such as wo da le yi jian maoyi song gei ta 我打了一件 毛衣送給他 ‘I knitted a sweater for him’.
Chapter 1. Introduction 13
half of the 378 languages he surveyed have the IO marked differently from the DO of both the double-object and the monotransitive constructions.20 Only 66 languages have the DO marked differently from the IO of the double-object construction and the DO of the monotransitive construction. The ratio of IO marking: DO marking is roughly 3:1.21 In other words, there are more languages having the IO marked differently than similarly to the DO of the double-object and the monotransitive constructions. Chinese, in terms of object marking for the double-object construction, conforms to the typologically dominant pattern with the IO instead of the DO being marked. The gei-phrase before the IO is thus better treated as a grammatical marker instead of a full verb. The above four points illustrate that the double-object construction is a special type of construction which should not be treated simply as a serial verb construction. Figure 1 summarizes the relationship of the concepts on the double-object construction adopted in this study. Ditransitive construction Double-object construction VERB = [+GIVE] VERB = [+DEPRIVE] 送 ‘give’, 給 ‘give’… 偷 ‘steal’, 搶 ‘snatch’ Unmarked IO V IO DO V DO IO
Others VERB = [+LOCATIVE] 放 ‘to put’
Others 叫 ‘to call’
Marked IO IO-Marker IO V DO V IO-Marker IO DO V DO IO-Marker IO
Figure 1. Relationship of concepts on double-object construction adopted in this study
20. The label for this pattern given by Haspelmath (2005, 2013) and Malchukov et al. (2010) is ‘indirect-object construction’, which refers to the double-object construction in which the thing (DO) is coded like the patient of a monotransitive sentence, while the recipient (IO) is coded differently (e.g., marked by case markers, adpositions or prepositions). Such an alignment of the two arguments in relation to the argument type in the monotransitive and ergative constructions was first brought up by Dryer (1986). 21. 84 languages have the DO and the IO marked identically with the DO in the monotransitive construction. 39 languages, including English, have more than one type of marking.
14 Cantonese give and Double-Object Construction
In sum, the present study focuses on the double-object construction with the main verb carrying the [+GIVE] semantic feature which sub-categorizes two objects bearing the semantic roles of recipient (IO) and thing (DO). These verbs are known as give-type double-object verbs. The two objects can have different types of syntactic realization in the surface structure in terms of object marking and word order. The construction can be expressed by means of the IO DO or DO IO patterns and by means of an IO marker, such as the V DO IO-Marker IO, the IO-Marker IO V DO as well as the V IO-Marker IO DO patterns. 1.5
Other syntactic functions performed by GIVE
It has been reported that in many Chinese dialects, the double-object verb GIVE, besides being a double-object verb, can also perform a number of grammatical functions. Below are some examples from standard Mandarin and Hong Kong Cantonese. A. Standard Mandarin
(8) 我 給 他 一 本 書 wo gei ta yi ben shu I give he a cl book ‘I gave him a book’
(9) 我 給 他 送 了 一 本 書 wo gei ta song le yi ben shu I give he give asp one cl book ‘I gave him a book’ (10) 我 給 老師 罵 了 wo gei laoshi ma le I give teacher scold asp ‘I was scolded by the teacher’
In the above sentences, gei 給 is used as a double-object verb, an IO marker and a passive marker respectively. B. Hong Kong Cantonese
(11) 我 畀 咗 本 書 佢22 13 35 35 35 ŋͻ pei tsͻ pun sy55 k’ɵy13 I give asp cl book he/she ‘I gave him/her the book’22 22. Unless stated otherwise, examples of Hong Kong Cantonese are based on my own speech and are transcribed into IPA following Zee (2001).
Chapter 1. Introduction 15
(12) 我 送 咗 一千 文 畀 阿 媽 ŋͻ13 sʊŋ33 tsͻ35 jɐt5-ts’in55 mɐn11–55 pei35 a33-ma55 I give asp 1000 dollar give mother ‘I gave 1000 dollars to my mother’
(13) 我 畀 佢 睇 戲 ŋͻ13 pei35 k’ɵy13 t’ɐi35 hei33 I give s/he watch movie ‘I let him/her watch movies’
(14) 佢 畀 人 偷 咗 銀包 k’ɵy13 pei35 jɐn11 t’ɐu55 tsͻ35 ŋɐn11-pau55 S/he give person steal asp wallet ‘Her/his wallet was stolen by someone’
(15) 佢 畀 把 刀 切 開 啲 蘋果 k’ɵy13 pei35 pa35 tou55 ts’it3 hͻi55 ti55 p’ιŋ11-kwͻ35 S/he give cl knife cut open cl apple ‘S/he divided the apples with a knife’
In the above five sentences on Hong Kong Cantonese, [pei] 畀 is used as a main verb, an IO marker, a verb expressing permissiveness, a passive marker and a verb introducing an instrument respectively.23 Similar usages of GIVE can also be found in other Chinese dialects and neighboring non-Sinitic languages. In Chapter 2, we will present a survey on the multiple functions of GIVE in these linguistic varieties. Noting the multi-functionality of the verb give, it is then relevant and important to understand why and how the verb give can take up these additional syntactic functions. In the literature, it is claimed that the verb in question underwent grammaticalization through which the verb developed from a content word into a function word (Heine & Kuteva, 2002; Newman, 1996, 1999). However, most of these studies rely on the data drawn from contemporary dialects while some only focus on one or some of these grammatical functions. Little has been worked on the grammaticalization process in an integrated manner to trace its temporal development and examine how these various markers are correlated with the double-object verb give (see, for example, Chin & Tsou, 2005; Yiu, 2010). We 23. [Pei] 畀 in these syntactic constructions have different grammatical status. In the sentence expressing the permissive meaning and introducing instruments, [pei] 畀 is a verb because it can take aspect markers, such as [ŋͻ13 pei35 tsͻ35 k’ɵy13 hɵy33 t’ɐi35 hei33] 我畀咗佢去睇戲 ‘I let him/her watch movies’. In the passive sentence, [ŋͻ13 pei35 a33-ma55 nau22] 我畀阿媽鬧 ‘I was scolded by my mother’, [pei] 畀 cannot take aspect markers, such as [tsͻ] 咗, and is thus treated as a grammatical marker rather than a full verb.
16 Cantonese give and Double-Object Construction
will first discuss the grammaticalization process of the double-object verb give in Hong Kong Cantonese from a synchronic perspective. As claimed by A. Yue (1995, 2004), one of the urgent tasks in Chinese dialectal study is the examination of its historical development. After examining the synchronic data, we will examine the diachronic development of these syntactic functions of [pei] 畀 in Cantonese with the aid of pre-modern Cantonese dialect materials published between the 19th and mid-20th centuries. The diachronic and synchronic studies of the syntactic constructions related to the double-object verb give in the Chinese language can help us to work out the mechanisms and chronology of the linguistic change. 1.6
Organization of this book
This book is divided into two parts. Part I consists of Chapters 2, 3 and 4, which deal with the syntactic functions performed by [pei] 畀. Chapter 2 provides a survey on the multi-functionality of GIVE in Chinese dialects and in the non-Sinitic languages spoken in China as well as in the Southeast Asian linguistic area. Notions pertinent to language contact and areal linguistics will be brought up and discussed with reference to the multi-functionality of GIVE in these linguistic varieties. Chapter 3 discusses the grammaticalization processes of GIVE with data drawn from contemporary Hong Kong Cantonese and other languages. Chapter 4 examines the syntactic functions of [pei] 畀 in the 19th century Cantonese based on pre-modern dialect materials. Through these materials, we can observe some major changes of the syntactic functions of [pei] 畀 during the 150 years. Part II of this book, made up of Chapters 5, 6 and 7, focuses on the on-going word order change in the double-object construction in Hong Kong Cantonese. Chapter 5 examines the word order typology of the double-object construction in Chinese dialects. Chapter 6 reviews some early proposals accounting for the relationship between the two word order patterns (i.e., the IO DO and the DO IO patterns). The review shows that none of the current linguistic theories can adequately explain the co-existing word order patterns in some Southern dialects, such as Hong Kong Cantonese, because they overlook the formation of linguistic strata in the dialects as a result of linguistic contact. To understand the extent and causes of the word order change in Hong Kong Cantonese, a sociolinguistic study with 40 native Hong Kong Cantonese speakers was conducted. Chapter 7 discusses the rationale and the design of the fieldwork, followed by the results of the fieldwork. Chapter 8 summarizes the major findings of this study, discusses the implications of the current study, and possible research work on GIVE in future.
Chapter 2
Multi-functionality of GIVE in Chinese dialects and neighboring non-Sinitic languages An areal-typological perspective
In Chapter 1, we briefly discussed the various syntactic functions performed by the double-object verb GIVE in standard Mandarin and Hong Kong Cantonese. It is found that the multi-functionality of GIVE is an areal feature of the mainland Southeast Asia (see, for example, Matisoff, 1991). This chapter provides a survey on this phenomenon. When studying the double-object verb GIVE and its multi-functions, we are able to examine the linguistic strata embedded in Chinese dialects due to their contact with the neighboring non-Sinitic languages. 2.1
Linguistic situation of Southeast Asia
2.1.1 Relationship between Chinese and Southeast Asian languages It has been pointed out that GIVE in standard Mandarin and that of Thai share many similar functions, such as dative marking, beneficiary marking and causative marking (Thepkanjana & Uehara, 2008). Besides Thai, GIVE is also found to perform multiple functions in the languages spoken in the Southeast Asian region24 which is “home to more than a thousand languages, belonging to five major language families: Austronesian, Austroasiatic, Tibeto-Burman, Tai-Kadai, and Hmong-Mien” (Matisoff, 1992, p. 44). Geographically speaking, these languages are spoken in the regions of “Vietnam, Cambodia, Laos, Thailand, Peninsular Malaysia, Burma, parts of Northeast India and Bangladesh, and extensive areas of Southern and Southwestern China” (Enfield, 2003, p. 45). Furthermore, these languages are found to share some linguistic features and this consequently leads to the question of whether these languages are genetically related or their similarities are due to
24. See the survey on Southeast Asian languages such as Vietnamese, Thai, Malay, Akan and Khmer by Yap and Iwasaki (2003).
18 Cantonese give and Double-Object Construction
inter-lingual borrowing.25 When comparing some lexical pairs between Tai and Chinese, F. Li (1976) found that the lexical correspondence between these two groups of languages could “offer some material for the consideration of Sino-Tai relationship” (p. 230). This is the reason why F. Li (1973) considered the following languages members of the Sino-Tibetan language family: (a) Chinese; (b) Kam-Tai; (c) Miao-Yao; (d) Tibeto-Burman. Manomaivibool (1976) examined more than 200 lexical correspondences between Chinese (both Old Chinese and Middle Chinese) and standard Thai.26 She argued that the regular correspondence suggests that these words “are cognate words derived from the same stems rather than from loanwords” (p. 26). She further commented that “it seems implausible that Tai had to borrow that many items of such common vocabulary from Chinese” (p. 26).27 However, the view that Chinese has a genetic relationship with the Kam-Tai languages has been challenged by some scholars such as Paul Benedict. In his Sino-Tibetan: A Conspectus, P. Benedict (1972) overtly argued that “Thai is related to Indonesian rather than to Chinese, and that the traditional view of a Chinese-Thai relationship must be abandoned” (p. 2).28 Gong (2006) also refuted the proposal that Miao-Yao languages are members of the Sino-Tibetan language family. He claimed that the similarities are due to lexical borrowing or sheer coincidence. The Chinese language, since the ancient time, had been in close contact with its neighboring non-Sinitic languages. Hashimoto (1976, 1985, 1986) found that the Northern and the Southern dialects of the Chinese language exhibit significant typological differences which he attributed to the influence of neighboring non-Sinitic languages on Chinese. Specifically, the Northern dialects are claimed to have received influences from the Altaic languages while the Southern dialects from the Tibeto-Burman, the Kam-Tai, the Austroasiatic as well as the Miao-Yao languages. Schuessler (2003) argued that as early as in the Old Chinese period, Chinese “had absorbed a significant number of words from Miao-Yao, Kam-Tai and Austroasiatic sources as well as loans from Tibeto-Burman” (p. 1). R. Cheng (1985) noted that some syntactic features in Taiwan Mandarin are not found in Beijing Mandarin and he argued that these features are “characteristics of the VO varieties of Southern China” (p. 376). Taiwan Mandarin thus displays features conforming to the typology of the Southern languages such as Tai. At 25. See, for example, X. Ma’s (2003) comments on the debate of the classification of the Kam-Tai, Miao-Yao and Sinitic languages. 26. See also Manomaivibool (1975), Manomaivibool and Tsou (1998). 27. See also P. Ting (2005). 28. Sun and Jiang (2000) provided an overview of the studies on the interrelationship between the Chinese language and its neighboring non-Sinitic languages as well as their genetic classification.
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 19
the same time, there are some features in Taiwan Mandarin “that are universal to Chinese as a whole” (p. 376). In other words, two types of opposite forces shape the development of Taiwan Mandarin. Besides Hashimoto’s typological studies, there are other studies in the areas of lexicon and phonology with an attempt to sort out the non-Sinitic substrata in the Chinese language. For example, Norman and Mei (1976) discussed some loanwords imported from Austroasiatic languages into Old Chinese. Bauer (1996) compared 31 lexical items in Cantonese and Southeast Asian languages and found that seven of them were originated from Tai while the remaining ones “involve the Min and Kejia dialects as well as other Southeast Asian language groups” (p. 1836).29 Yue-Hashimoto (1976) studied the correlation between the aspiration feature and yang-tone words in some Min and Yue dialects and argued that these two dialect groups “share many features with the neighboring Chuang [i.e., Zhuang] and Tai languages, which differentiate them from northern Chinese” (p. 7). Tsou (1999) compared the entomological items such as spider and flea in some Yue dialects spoken in the Pearl River Delta region and claimed that some of these dialects contain a substratum of Austroasiatic languages. Besides lexical items and phonology, there were also extensive studies on the syntactic features commonly found in the Southeast Asian languages. Using the Lahu language of the Tibeto-Burman group as an example, Matisoff (1991) compared the use of particles, namely noun-particles, verb-particles and unrestricted particles, in Southeast Asian languages and Chinese. He argued that the similar characteristics found among these languages are due to “long periods of intimate contact” which “have largely homogenized all aspects of the phonology, grammar, and semantics of these languages, so that it makes eminent good sense to consider phenomena like grammatization in the widest possible areal context” (p. 386). Other studies on the grammar of the mainland Southeast Asia include Sybesma’s (2008) analysis of post-verbal can in Zhuang, Cantonese, Vietnamese and Lao, Enfield’s (2001) study on the polyfunctionality of the verb acquire and Bisang’s (1996, 1999) examination of the grammaticalization process of nouns and verbs as well as classifier usage in Southeast Asian languages, the word order of the potential complement structure in Southern Chinese dialects (F. Wu, 2005), Cantonese grammar as areal features (Matthews, 2006), the grammaticalization process in isolating languages (Ansaldo et al., 2018), and typological variations across Chinese dialects (Szeto, 2019).
29. There are studies on the non-Sinitic lexical substratum in the Yue dialects (Yue-Hashimoto, 1991b) and the statistical analysis on the lexical correspondence between Chinese and Kam-Tai languages (B. Chen, 1996).
20 Cantonese give and Double-Object Construction
2.1.2 The Southeast Asian linguistic area The sharing of some common linguistic features among the Southeast Asian languages of different language families suggests that the mainland Southeast Asia constitutes a linguistic area which is defined as “a geographical region containing a group of three or more languages that share some structural features as a result of contact rather than as a result of accident or inheritance from a common ancestor” (Thomason, 2000, p. 312).30 There are at least five famous linguistic areas which have been identified and studied extensively: (1) The Balkan Sprachbund in southeastern Europe; (2) The Sepik River Basin in Papua New Guinea; (3) The Pacific Northwest of North America; (4) The Ethiopian highlands in Africa; (5) South Asia (Thomason, 2000). Added to these five linguistic areas is the Southeast Asian region which covers the language families mentioned above.31 Some members of the non-Sinitic language groups of the Southeast Asian linguistic area are found in the Chinese territory:32 a. Tibeto-Burman (藏緬語): It is a branch of the Sino-Tibetan language family. The languages are mainly spoken in provinces of Yunnan, Sichuan and Guizhou. Examples include Kazhuo 卡卓語 and Lahu 拉祜語. b. Kam-Tai / Tai-Kadai (侗台語): This includes (i) the languages of Zhuang 壯 語 and Dai 傣語 spoken in the Guangxi Zhuang Autonomous Region; (ii) the languages of Li 黎語 and Lingao 臨高話 spoken in Hainan province and (iii) the language of Shui 水語 spoken in Guizhou province; c. Hmong Mien / Miao-Yao (苗瑤語): This includes the languages of Miao 苗 語, Yao 瑤語 and She 畬語 spoken in Guangdong, Guangxi, Yunnan, Guizhou, Sichuan, Hubei and Hunan provinces; d. Mon Khmer / Austroasiatic (南亞語): This includes the language of Jing 京語 spoken in Guangxi; e. Austronesian (南島語): This family includes those Formosan languages spoken in Taiwan and the Huihui language (回輝話) spoken in Sanya of Hainan province. When examining the morpheme GIVE and its functions in Chinese dialects and non-Sinitic languages in the Southeast Asian linguistic area, we can observe inter-lingual and inter-dialectal influences. 30. A similar definition was also given by Enfield (2005) for the Southeast Asian linguistic area. 31. Goddard (2005) in Section 2.3 provided a brief account on how speakers of different language families came to live in the same region. 32. See also Chapter 1 in Norman (1988).
2.2
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 21
A survey of GIVE in Chinese dialects
In this section, we survey the double-object verb GIVE in Chinese dialects based on published dialect materials. 2.2.1 The Yue dialects In the Yue dialects, there are two types of double-object verb GIVE: 畀 and [ɂi]/ [ei]. 2.2.1.1 The morpheme 畀 This form is found in the Yue dialects spoken in the following sub-groups.33 a. The Guangfu sub-group (廣府片)3435
Hong Kong 香港 Guangzhou 廣州 (H. Gao, 1980) Foshan 佛山 (B. Li & He, 1992)35 Zengcheng 增城 (W. T. He, 1993) CIT110
畀 [pei35] 畀 [pεi35] 畀 [pei35] 畀 [pei35]
Xinyi 信宜 (K. Luo, 1987) Gaozhou 高州34 Beihai 北海 (X. Chen & Chen, 2005) Nanning 南寧 (Y. Lin & Qin, 2008) CIT57
CIT221
畀 [pei35] 畀 [pei35] 畀 [pei35] 畀 [pi35]
b. The Northern Delta sub-group (北三角洲片)36
Nanhai 南海36 Shunde 順德
畀 [pɐi54] 畀 [pei54]
Gaoyao 高要 畀 [pei45] Tengxian 藤縣 (Yue-Hashimoto, 1979) 畀 [ɂbi55]
c. The Southern Delta sub-group (南三角洲片)37 Dongguan 東莞 (X. Chen, 1993) 畀 [pɐi35]
Huazhou 化州37 畀 [ɓei45]
d. The Liang Yang sub-group (兩陽片)38 Yangchun 陽春38
畀 [pei21]
33. The sub-grouping of the Yue dialects is based on A. Yue (2006b). 34. The data of Gaozhou are based on A. Yue-Hashimoto’s unpublished field notes.
35. No phonetic transcriptions are given in 佛山方言志 and the pronunciations here are based on the Foshan dialect recorded in Zhan and Cheung (1987), p. 87. 36. The data of Nanhai, Shunde and Gaoyao are based on A. Yue-Hashimoto’s unpublished field notes. 37. The data of Huazhou are based on A. Yue-Hashimoto’s unpublished field notes. 38. The data of Yangchun are based on A. Yue-Hashimoto’s unpublished field notes.
22 Cantonese give and Double-Object Construction
e. Overseas
Kuala Lumpur 吉隆坡 (X. Chen, 2003) 畀 [pei35]
2.2.1.2 The morpheme [ɂi] or [ei] This form is found in the Wuyi sub-group (五邑片). Taishan台山 (Yue-Hashimoto, 2005)
[ɂi55]
Kaiping 開平 (Deng, 2000)
[ei55]
2.2.2 The Hakka dialects Two major types of double-object verbs are found in the Hakka dialects: 分 and 拿.
2.2.2.1 The morpheme 分 This form is found in the dialects of the Yue-Tai sub-group (粵台片), Fujian and Guangxi.39 a. Guangdong
Meixian 梅縣 (R. L. Li & Chang, 1992)
Dapu 大埔 (Chiang, 2003)
分 [pun44] Qingxi 清溪 (R. L. Li & Chang, 1992)
分 [pin33]
分 [pun33] Liannan 連南 (R. L. Li & Chang, 1992)
分 [pͻn44]
Wengyuan 翁源 (R. L. Li & Chang, 1992)
分 [pun22] Huazhou 化州 (R. L. Li, 1999)
分 [bun55]
Dongguan 東莞 (X. Chen, 1993)
分 [pun33] Dianbai電白 (R. L. Li, 1999)
Jiexi 揭西 (R. L. Li & Chang, 1992) 分 [pun53] Lianjiang 廉江 (R. L. Li, 1999)
分 [pun45]
Xinfeng 新豐 (R. Zhou, 1990)
分 [pun44] Gaozhou 高州 (R. L. Li, 1999)
分 [pən44]
分 [pun35] Ninghua 寧化 (R. L. Li & Chang, 1992)
分 [pε̃ i31]
分 [pun44] Rongxian 容縣 (X. Chen, 2004) 分 [pun33] Luchuan 陸川 (X. Chen, 2004) 分 [pun44] Bobai 博白 (X. Chen, 2004)
分 [pun33] 分 [pun44] 分 [pun44]
b. Fujian
Xiuzhuan 秀篆 (R. L. Li & Chang, 1992)
c. Guangxi
Fumian 福綿 (X. Chen, 2004) Xingye 興業 (X. Chen, 2004) Beiliu 北流 (X. Chen, 2004)
39. The sub-grouping of the Hakka dialects is based on Wurm et al. (1988).
分 [pən34]
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 23
d. Taiwan
Taizhong 台中 (Chiang, 2003) Sixian 四縣 (S. Yang, 1992)
分 [pun35] 分 [pun13]
Hailu 海陸 (S. Yang, 1992)
分 [pun53]
e. Overseas
Tu’nai in Malaysia土乃 (X. Chen, 2003) 分 [pun33]
2.2.2.2 The morpheme 拿 This form is found in the Hakka dialects spoken in southern Jiangxi and Fujian provinces. a. Jiangxi
Dayu 大余 (R. L. Li & Chang, 1992) Nankang 南康 (L. Liu, 1999) Anyuan 安遠 (L. Liu, 1999) Yudu 于都 (L. Liu, 1999)
b. Fujian
Liancheng 連城 (Xiang, 1997)
拿 [la5] 拿 [na44] 拿 [na35] 拿 [na35]
Longnan 龍南 (L. Liu, 1999) Quannan 全南 (L. Liu, 1999) Dingnan 定南 (L. Liu, 1999)
拿 [na24] 拿 [na24] 拿 [na35]
拿 [nu33]
2.2.3 The Min dialects There are four types of double-object verbs found in the Min dialects: 乞, 互/與, 欠 and 納.
2.2.3.1 The morpheme 乞 乞 is found in the dialects of the Northeastern sub-group, the Xinghua 興化 sub-group and some dialects of the Southern Min sub-group (Norman, 1991). According to R. L. Li (1996), the etymon of the double-object verb in these dialects is 乞, which can be found in many early textual materials in the Ming dynasty.40
40. See also R. Wu (2015) on the diachronic development of [kʰi3] 乞 in the Min dialects.
24 Cantonese give and Double-Object Construction
a. The Northeastern sub-group
Fuzhou 福州 (H. Lin, 2002) Changle長樂 (H. Lin, 2002) Fuqing 福清 (H. Lin, 2002) Yongtai 永泰 (H. Lin, 2002) Ningde 寧德 (H. Lin, 2002) Gutian古田 (H. Lin, 2002)
乞 [k’øyɂ23] 乞 [k’yk3] 乞 [k’yɂ12] 乞 [k’y21] 乞 [k’ik2] 乞 [k’yk24]
b. The Xinghua sub-group
Putian 莆田 (Z. T. Chen & Li, 1991)
Fuan福安 (H. Lin, 2002) Shouning 壽寧 (H. Lin, 2002) Zhouning 周寧 (H. Lin, 2002) Fuding 福鼎 (H. Lin, 2002) Cangnan蒼南 (Akitani, 2005) Taishun 泰順 (Akitani, 2005)
乞 [k’øk5] 乞 [k’iɂ5] 乞 [k’yk5] 乞 [k’iɂ4] 乞 [k’əɂ5] 乞 [hεɂ5]
乞 [k’œɂ11]
c. The Southern Min sub-group
Datian 大田 (Z. T. Chen & Li, 1991) 乞 [k’e31] Suixi 遂溪 (Yue-Hashimoto, 1985) 乞 [k’i55] Longxi 龍溪 (Z. T. Chen & Li, 1991) 乞 [k’ia24] Leizhou 雷州 (L. Lin, 2006) 乞 [k’i55] 44 Jieyang 揭陽 (Tung, 1959) 乞 [k’eɂ ] CIT60
CIT60
2.2.3.2 The morpheme 互 or 與 This form is found in the dialects of the Southern Min sub-group. According to Mei (2005), the etymon of [hͻ] (sometimes written as 互) in Min is yu 與, which was the verb GIVE in Classical Chinese (Peyraube, 1988). Xiamen 廈門 (C. Zhou & Ouyang, 1998) 互 [hͻ11] Taiwan 台灣 (H. Yang, 1995) 與 [ho22] Zhangzhou 漳州 (Z. T. Chen & Li, 1991) 與 [hͻ22]
2.2.3.3 The morpheme 欠 This form is found in the dialects of the Central Min sub-group. Yongan 永安 (Z. T. Chen & Li, 1991) 欠 [k’εiŋ35] Shaxian 沙縣 (Z. T. Chen & Li, 1991) 欠 [k’iŋ24] CIT60
2.2.3.4 The morpheme 納
Jian’ou 建甌 (Z. T. Chen & Li, 1991) 納 [na24]
CIT60
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 25
2.2.4 The Gan dialects Most Gan dialects have [pa] 把 or [pai] 擺 as the double-object verb.
2.2.4.1 The morpheme 把 or 擺 The double-object verbs [pa] 把 or [pai] 擺 are found in the dialects of the Changqing sub-group (昌清片), the Fuguang sub-group (撫廣片), the Jicha sub-group (吉茶 片), the Yingyi sub-group (鷹弋片) and the Yiliu sub-group (宜瀏片).41 a. The Changqing sub-group (昌清片) Nanchang 南昌 (Wei & Chen, 1998) [paɂ5]
b. The Fuguang sub-group (撫廣片)
Dongxiang 東鄉 (L. Liu, 1999) Linchuan 臨川 (L. Liu, 1999) Nanfeng 南豐 (L. Liu, 1999)
擺 [pai353] 擺 [pai35] 擺 [pai11]
Yihuang 宜黃 (L. Liu, 1999) Lichuan 黎川 (S. Yan, 1993)
擺 [pai232] 把 [pai44]
c. The Jicha sub-group (吉茶片)
Jishui 吉水 (R. L. Li & Chang, 1992) 把 [pa31]
d. The Yingyi sub-group (鷹弋片) Poyang 鄱陽 (L. Liu, 1999)
把 [pͻ42]
e. The Yiliu sub-group (宜瀏片)
Pingxiang 萍鄉 (L. Liu, 1999) 把 [pa35]
Leping 樂平 (L. Liu, 1999) 把 [pa213]
Liuyang 瀏陽 (Xia, 1998)
把 [pa24]
2.2.5 The Xiang dialects The double-object verb in most Xiang dialects is 把 or 得.
2.2.5.1 The morpheme 把 This form is found in the dialects of the Changyi sub-group (長益片), the Loushao sub-group (婁邵片) and the Jipu sub-group (吉浦片).42 41. The sub-grouping of the Hakka dialects is based on Wurm et al. (1988). 42. The sub-grouping of the Xiang dialects is based on Wurm et al. (1988).
26 Cantonese give and Double-Object Construction
a. The Changyi sub-group (長益片) Changsha 長沙 (Y. M. Li, 1991)
把 [pa42]
b. The Loushao sub-group (婁邵片) Shaoyang 邵陽 (Z. Chu, 1998)
把 [pa42]
Yiyang 益陽 (Cui, 1998)
把 [pa41]
Qiyang 祁陽 (W. Q. Li, 1998)
把 [pa53]
c. The Jipu sub-group (吉浦片) Xupu 漵浦 (K. He, 1999)
把 [pɒ23]
2.2.5.2 The morpheme 得 This form is found in the dialects of the Changyi sub-group (長益片). Hengyang 衡陽 (Y. M. Li, 1986) 得 [te2]
Anren 安仁 (M. Chen, 1995) 得 [te313]
2.2.6 The Wu dialects 2.2.6.1 The morpheme 撥, 不 or 八 This form is found in the majority of the Wu dialects of the Taihu sub-group (太湖片) and the Taizhou sub-group (台州片).43 a. The Taihu sub-group (太湖片)
Shanghai 上海 (B. Xu & Tang, 1988) Nanhui Zhoupu南匯周浦 (N. Qian, 1992) Baoshan Luodian 寶山羅店 (N. Qian, 1992) Baoshan Shuangcao Dun 寶山霜 草墩 (N. Qian, 1992) Songjiang 松江 (N. Qian, 1992) Kunshan 昆山 (N. Qian, 1992) Changshou 常熟 (N. Qian, 1992) Suzhou 蘇州 (N. Qian, 1992)
撥 [pəɂ55] Changzhou 常州 (N. Qian, 1992) 撥 [ɂbəɂ5] Yixing 宜興 (N. Qian, 1992) 撥 [pəɂ5]
Liyang 溧陽 (N. Qian, 1992)
撥 [pəɂ5] 撥 [pəɂ5] 撥 [pᴇɂ5] 撥 [pəɂ5]
Jintan 金壇 (N. Qian, 1992) Hangzhou 杭州 (N. Qian, 1992) Jiaxing 嘉興 (N. Qian, 1992) Huzhou Shuanglin 湖州雙林 (N. Qian, 1992)
撥 [pəɂ5]
Jiangyin 江陰 (N. Qian, 1992)
43. The sub-grouping of the Wu dialects is based on Wurm et al. (1988).
撥 [pəɂ5] 撥 [pəɂ5] 撥 [pəɂ5]
八 [pɑɂ45] 叭 [pɑ323] 不 [pəɂ5] 撥 [pəɂ54] 撥 [pəɂ54]
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 27
Wujiang Lili吳江黎里 (N. Qian, 1992) Wujiang Shengze 吳江盛澤 (N. Qian, 1992) Wuxi 無錫 (N. Qian, 1992)
撥 [pəɂ5]
Shaoxing 紹興 (N. Qian, 1992) 不 [pιɂ5]
撥 [pəɂ5]
Zhoushan 舟山 (Fang, 1993)
撥 [pəɂ5]
b. The Taizhou sub-group (台州片) Tiantai 天台 (Z. Dai, 2003)
撥 [pøɂ5]
Ningbo 寧波 (N. Qian, 1992)
撥 [pɐɂ5]
撥 [pɐɂ55–33]
Huangyan 黃岩 (N. Qian, 1992) 不 [pɐɂ55]
2.2.7 The Mandarin dialects There are two forms: [kei] 給 and [pa] 把.
2.2.7.1 The morpheme 給 The double-object verb [kei] 給 is found in the Mandarin dialects of the Northern sub-group (北方官話片) and the Southwestern sub-group (西南官話片).44 a. The Northern sub-group (北方官話片)
Beijing 北京 (Hou, 1998) Changli 昌黎 (Hebeisheng, 1984) Zhengzhou 鄭州 (Lu et al., 1998) Luoyang洛陽 (W. He, 1993)
給 [kei214] 給 [kei213] 給 [kei24] 給 [kuə33]
Jinxiang 金鄉 (F. Ma, 2000) Laizhou萊州 (Z. Qian, 2005) Wenshui文水 (Hu, 1984) Lingchuan陵川 (Jin, 1983)
給 [kei312] 給 [kei55] 給 [kue423] 給 [kɤɂ32]
b. The Southwestern sub-group (西南官話片)45
Zhongxiang 鍾祥 (Chao, 1939) Chengdu成都 (Y. Zhang et al., 2001)
給 [kə53] Changde 常德 (Q. Zheng, 1999) 給 [ke21] 給45
2.2.7.2 The morpheme 把 This form is found in the dialects of the Jianghuai sub-group (江淮官話片).46 Hangzhou 杭州 (Simmons, 1992)
把 [pᴀ52]
Huanggang 黃崗 (H. Wang, 2004)
44. The sub-grouping of the Mandarin dialects is based on J. Yuan (2001). 45. No transcription is given in this work. 46. No transcription is given in this work.
把46
28 Cantonese give and Double-Object Construction
2.3
Multiple forms of GIVE as a result of language contact
The above survey gives an overview of the major double-object verb give in Chinese dialects. In addition, we find that some forms which are dominant in one dialect group also appear in other dialect groups as well as non-Sinitic languages. This phenomenon is especially common in areas where different dialects and languages are spoken. 2.3.1 Inter-dialectal influence a. In the Yue dialect of Yulin 玉林 spoken in Guangxi, the double-object verb is [fɐn54] 分 (J. Chu, 2007). This is the dominant form of the Hakka dialects, which are also spoken in Yulin.47 In the Yulin dialect, the initial of 分 is the labio-dental voiceless fricative [f] instead of the bilabial voiceless stop [p], which is commonly found in the Hakka dialects (see Section 2.2.2.1). b. In some Hakka dialects, the form of the Yue dialects (i.e., [pi] 畀) is found. These regions include Hong Kong 香港 (R. L. Li & Chang, 1992), Huizhou 惠 州 (R. Liu, 1991), Yangjiang 陽江 and Xinyi 信宜 (R. L. Li, 1999). c. The morpheme [na] 拿 is one of the major forms of the Hakka dialects (see Section 2.2.2.2). The Min dialect of Jian’ou 建甌 (Section 2.2.3.4) also has a similar form [na] (written as 納 in Z. T. Chen and Li (1991)). The form [laɂ] 拿 in the Gan dialect of Xiushui 修水 in Northern Jiangxi is also related to [na] of the Hakka dialects. d. The Mandarin dialects of the Jianghuai sub-group in Hubei province have [pa] 把 as the double-object verb, which is commonly found in the Xiang dialects (see Section 2.2.5.1). e. Some dialects spoken in Northern Guangdong and Southern Hunan (known as tuhua 土話 ‘patois’) have the double-object verbs similar to the ones in the neighboring dialects. For example, the verb GIVE in the patois土話 spoken in Xi’an 西岸 of Lianzhou 連州 (Chang & Wan, 2004) is [tɐi11] 得, which is similar to [te] 得 in the Xiang dialects (see Section 2.2.5.2). In the patois of Huangpu 黃圃 in Lechang 樂昌 (Northern Guangdong) (Chang, 2000) and the patois of Yizhang 宜章 (R. Shen, 1999) in Hunan province, the double-object verb are 拿 [no24] and 拿 [no44] respectively, which are similar to [na] 拿 in the Hakka dialects. f. The influence of neighboring dialects bearing different forms of the double-object verbs can be well illustrated by the jun dialects 軍話. According to X. Q. Qiu (2005), the jun dialects are now spoken in some scattered areas in Guangxi, 47. For the description of the Hakka dialects spoken in Yulin, see X. Chen (2004).
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 29
Guangdong and Hainan provinces, which have been heavily influenced by the local dialects such as Yue, Min and Hakka. Table 2 lists the double-object verbs in the various jun dialects (based on X. Q. Qiu (2005)). Table 2. GIVE in jun dialects GIVE
Guangdong Huizhou 惠州 俾 pi55
Lufeng 陸豐
分pan33
Guangxi Yongan 永安 乞 k’i35
Hainan Basuo 八所
給 kei51
Zhonghe 中和 給 kei51
Yacheng 崖城 給 kei11
The jun dialect in Huizhou is surrounded by the Yue, Min and Hakka dialects and has absorbed many features from these dialects (X. Q. Qiu, 2005). As for the double-object verb, it is the form of the Yue dialects (i.e., [pi] 俾) it has adopted. The jun dialect in Lufeng is surrounded by the Min and Hakka dialects and it is the Hakka form [pan] 分 adopted. The jun dialect of Yongan in Guangxi may be originated from Fujian province and this dialect thus contains features of the Min dialects. The double-object verb for this jun dialect is [k’i] 乞, which is found in Eastern Min and Southern Min dialects (see Section 2.2.3.1). The double-object verbs in the jun dialects of Hainan province are 給, which is the form used in the Mandarin dialects of the Northern sub-group (see Section 2.2.7.1).48 2.3.2 Inter-lingual influence Besides inter-dialectal influence, we also find that the double-object verbs in some non-Sinitic languages are similar to the ones in Chinese dialects. a. Some languages from the Tibeto-Burman group have the double-object verbs similar to [pei] found in most Yue dialects (see Section 2.2.1.1). Examples include the languages of Bisu 畢蘇語 (S. Xu, 2001) and Lahu 拉祜語 (Matisoff, 1991), in which the double-object verbs are [pi31] and [pi53] respectively. The double-object verb [pe24] in Mulao 木佬語 (Bo, 2003) of the Geyang branch of the Kam-Tai family is also similar to [pei]. b. The double-object verb in the Mien language 勉語 of the Miao-Yao family is [pun33] (M. Zhao, 2004). This is not surprising because the Miao-Yao languages are spoken in Guangdong and Jiangxi provinces in which the Hakka dialects are also spoken. 48. X. Q. Qiu (2005) observed that the jun dialects in Hainan province preserve the features of the Northern dialects the most when compared with other varieties of the jun dialect.
30 Cantonese give and Double-Object Construction
c. The double-object verbs in Maonan 毛難語 (Liang, 1980) and Buyang 布央語 (J. F. Li, 2003) languages of the Kam-Tai family spoken in Guangxi are [ɂna:k44] and [na:k11] respectively, which are close to [na] 拿 in the Hakka dialects (see Section 2.2.2.2). d. In the Yue dialects of Lianzhou 廉州and Beihai 北海 (X. Chen, 2004) and the pinghua 平話 dialects spoken in Nanning 南寧, the double-object verb is [hai]. Table 3 lists the double-object verb GIVE in these dialects. It is noted that this form is not found in the Yue dialects spoken in other places (see Section 2.2.1). Table 3. GIVE in the Yue and pinghua dialects spoken in Guangxi495051 Dialects
Lianzhou 廉州49 Beihai Wa dialect北海佤話 Beihai Haibian 北海海邊話 Sitang 四塘 (J. Chu, 2007)50 Shibu 石埠 (J. Chu, 2007) Tianyang 田陽 (J. Chu, 2007) Tingzi 亭子51 Tiandong 田東 Hengxian 橫縣 Rongshui 融水 Xinxu 心墟 Yongning 邕寧 Yizhou 宜州
GIVE hɐi44 hɐi11 hɐi11 hei35 hei35 hei35 həi55 hɐi35 hai53 hɐŋ44 hɐi24 hɐi35 hɐŋ42
Tonal category
yinqu 陰去 shang 上聲 shang 上聲 yinqu 陰去 yinqu 陰去 yinqu 陰去 yinqu 陰去 yinqu 陰去 yinqu 陰去 yangshang 陽上 yinqu 陰去 yinqu 陰去 yinshang 陰上
It is noted that, in addition to Chinese dialects, the Zhuang language is also spoken in Guangxi. Many speakers living in Guangxi are bilingual or even multilingual in Chinese and the Zhuang language. Take Nanning 南寧 as an example. 41% of the speakers (out of 173 trilingual speakers surveyed by H. Chen and Li (2005)) are proficient in Putonghua, Nanning Yue and Zhuang. Therefore, it is likely that these languages mutually influence each other. When examining the double-object verbs in the Zhuang-Dai 壯傣languages of the Kam-Tai family in Guangxi, we find that many of them are similar to the ones found in the Yue and pinghua dialects. Table 4 lists the verb GIVE in these Zhuang-Dai languages spoken in Guangxi, Guizhou and Yunnan. 49. The data of Lianzhou are based on A. Yue-Hashimoto’s unpublished field notes. 50. The following is pinghua dialects. 51. The following pinghua dialects are based on Qin (2005).
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 31
Table 4. GIVE in Zhuang-Dai languages5253 Language/dialect52
GIVE
(a) Northern Zhuang Wuming 武鳴 (F. Li, 1956) Bo’ai 剝隘土語 (F. Li, 1988) Hengxian 橫縣 Tiandong 田東 Douan 都安 Hechi 河池 Liujiang 柳江 Tianlin 田林 Yongbei 邕北 Nandan 南丹 Shanglin 上林 Yishan 宜山 Donglan 東蘭 Pingguo 平果 Huanjiang 環江 Laibin 來賓 Lingle 凌樂 Longsheng 龍勝 Guigang 貴港 Lue (Hudak, 1996)
hɑա55 haï44 hai33 hai55 həա55 ha:ŋ33 haŋ53 hai45 haա44 haա31 haŋ33 haŋ53 hai33 haա55 haա42 haŋ33 hai35 haա42 ha:ŋ33 hաա11
C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C1
(b) Southern Zhuang Longzhou 龍州 (F. Li, 1993) Debao 德保 Qinzhou 欽州 Jingxi 靖西 Yongnan 邕南 Fusui 扶綏 Long’an 隆安 Ningming 寧明 Lungming (Hudak, 1991)
hա24 hͻ:i214 hai55 hͻ:i23 hai53 hաi42 hai55 həա35 hɤպ33
C1 C1 C1 C1 B2 C1 C1 C1 C1
hա13 haա31
C1 C1
(c) The Dai sub-group (傣語) Xidai 西傣 (Yu & Luo, 1980) Dedai 德傣 (Yu & Luo, 1980)
Tonal category53
52. Unless stated otherwise, the data of the Zhuang language are based on Qin (2005). 53. We follow F. Li’s (1977) convention for labeling the tonal category in Zhuang-Tai languages.
32 Cantonese give and Double-Object Construction
The above Zhuang-Dai languages all have the voiceless glottal fricative h- as the initial and most of them have the low vowel [a] as the main vowel. They all have the C1 tone except Yongnan 邕南. The morphemes have similar syllable structures to the Yue and pinghua dialects spoken in Guangxi (see Table 3).54 The similar forms of the double-object verb GIVE in these linguistic varieties (which are not found in other Chinese dialects) deserves our examination of their inter-relationship. To argue that the double-object verbs in the Yue dialects as well as pinghua dialects of Guangxi are related to the Zhuang-Dai languages, we need to find out whether there is any phonological correspondence between these two sets of morphemes. F. Li (1976) noted that there is a tonal correspondence between the loanwords in Chinese and Tai. Generally speaking, a C1 tone in Tai corresponds to the shang tone in Middle Chinese.55 If the double-object verb GIVE in these dialects is a loanword from Tai, we would then expect that the morpheme in question has a shang tone or a yinshang tone if there is a split of upper and lower registers in the dialect. Among the Yue dialects we examined above, only four of them have a shang tone (i.e., Wa of Beihai, Haibian of Beihai, Rongshui and Yizhou). The other nine have a yinqu 陰去 tone. It seems that a regular tonal correspondence cannot be established. A possible explanation for the irregularity is that it was the tonal contour rather than the tonal category that was borrowed as Qin (2005) and Y. Lin (2010) suggested.56 The sound correspondence between the Yue/pinghua dialects and the Zhuang languages suggests a possible link between these groups of languages from the perspective of language contact.57
54. In his description of the pinghua dialects of Sitang 四塘, Shibu 石埠 and Tianyang 田陽, J. Chu (2007) used the character 許 for the double-object verbs of these dialects. However, Qin (2005) argued that there is no cognate relationship between [hai] and 許. The reflex of the ancient rime of 許 in the pinghua dialects of Tingzi, Tiandong, Hengxian, Xinxu and Yongning are [y], [œy], [y], [ɵy] and [əu] respectively but the finals of the double-object verb GIVE in these pinghua dialects are [əi], [ɐi], [ai], [ɐi] and [ɐi] respectively. 55. Following Wulff ’s study on the tonal correspondence between Tai languages and Chinese, F. Li (1976) claimed that the four Tai tones (A, B, C and D) correspond to the tones of Middle Chinese ping, qu, shang and ru respectively.
56. Among those dialects that have the yinqu 陰去 tonal category for the verb GIVE, six of them have a rising contour, two have a level contour, one has a falling contour. Among the thirty-one Zhuang-Dai languages we have examined, sixteen have a level contour, five have a rising contour and eight have a falling contour. The double-object verb GIVE in these two language groups share similar tonal contours. 57. Y. Lin (2010) further argued that [hai] in the Zhuang language was borrowed from the morpheme 𢬿 [hai] in the pinghua dialect.
2.4
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 33
Multi-functionality of GIVE in Chinese dialects and neighboring non-Sinitic languages
As discussed in Chapter 1, the double-object verb GIVE in Cantonese can function as an IO marker, a passive marker and a causative verb. A survey of the four syntactic functions of GIVE in Chinese dialects and the neighboring non-Sinitic languages was conducted and the data are presented in Appendix 1. There are two observations. 2.4.1 Linguistic contact between Northern dialects and Altaic languages In most of the Mandarin dialects in the Northern sub-group, the double-object verb can be used as an IO marker, but not a passive marker or a causative verb.58 The passive marker and the causative verb functions are usually performed by other morphemes such as jiao 教, jiao 叫 or rang 讓.59 Norman (1982) argued that the convergence of the causative verb and the passive marker (i.e., performed by the same morpheme) in the Mandarin dialects “is due to Manchu influence on Chinese” and “the use of identical morphological and syntactic markers for the causative and passive is not so common a feature that we may view the Peking and the northern Chinese developments as cases of spontaneous and independent generation” (p. 245).60 Hashimoto (1988) also claimed that “the development of the jiao (cause/by) / rang (allow/by) type seems to have taken place under the direct contact with the Tungus-Manchu type” (p. 350). For example, the causative and the passive meanings in Manchu are expressed by the same morpheme [bu]. (16) tere cooha niyalma de wa-bu-ha (Norman, 1982, p. 245) he soldier agent kill-passive-past ‘He was killed by the soldier’
58. However, in standard Mandarin, gei 給 can be used as a passive marker, such as wo gei laoshi ma le 我給老師罵了 ‘I was scolded by the teacher’. According to B. Ma (2002), the use of gei 給 as a passive marker in the Mandarin dialects began in the Qing dynasty.
59. J. Ma and Wu (2003) reported that the morpheme jiao 叫 is used as the passive marker in twelve Mandarin dialects of Linxin 臨沂 of Shandong province: Lanshan 蘭山, Hedong 河東, Luozhuang 羅庄, Linshu 臨沭, Ju’nan 莒南, Yi’nan 沂南, Yishui 沂水, Mengyin 蒙陰, Pingyi 平 邑, Feixian 費縣, Cangshan 蒼山 and Tancheng 郯城. 60. We will show in Chapter 3 that the passive usage can be semantically derived from the causative usage regardless the morpheme concerned is a double-object verb or not.
34 Cantonese give and Double-Object Construction
(17) bi terebe gene-bu-he I him go-causative-past ‘I made him go’
In Japanese, a member of the Altaic group to which Manchu belongs, a passive sentence can be interpreted as a causative sentence. (18) John-ga Mary-ni kami-o kir-are-ta (Washio, 1993, p. 49) John-nom Mary-BY hair-acc cut-passive-past ‘Johni had hisi/j hair cut by Mary’
According to Washio (1993), the hair in the sentence can belong to either John or someone else. For the former, the sentence has a passive reading. For the latter, the sentence has the causative meaning: ‘John made Mary cut someone’s hair’.61 2.4.2 Linguistic contact between Southern dialects and Southeast Asian languages With examples from Lahu of the Tibeto-Burman group, as well as Tai, Vietnamese, Chinese and Miao Yao, Matisoff (1991) argued that the use of GIVE as a causative verb and a beneficiary marker is one of the areal linguistic features in the Southeast Asian linguistic area. Notice that Matisoff did not mention the passive marker function of GIVE. For example, the double-object verbs in Thai [hâj] and Vietnamese [cͻ] only function as an IO marker, a causative verb and a beneficiary marker but not a passive marker (Diller, 2001; Thepkanjana & Uehara, 2008). Yap and Iwasaki (2003) surveyed the grammaticalization of GIVE in Southeast Asian languages such as Vietnamese, Thai, Malay, Akan and Khmer. They found that the passive marker function is not performed by GIVE in these Southeast Asian languages. Consider the following sentences from Thai. (19) chûaj rîak thέksîi hâj phǒm (Yates & Tryon, 1970, pp. 442–444) please call taxi for I ‘Please call a taxi for me’ (‘hâj’ as a beneficiary marker) (20) kháw anújâat hâj phǒm rian phaasǎathaj he give I study Thai ‘He allows me to study Thai’ (‘hâj’ expressing the permissive meaning)
The passive marker in Thai is [tաk]. 61. The same can also be found in Korean (Ahn, 1998). There are four allomorphs for the causative and passive suffixes in Korean: i, ki, hi and li. The choice of the suffix is phonologically determined on the basis of the verb stem to which the suffix is attached.
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 35
The same is also found in Vietnamese. The double-object verb [cͻ] can be used as an IO marker, a beneficiary marker, a causative verb, but not a passive marker. (21) ʌ̃ũŋ͡m lam ͻn cͻ toι haι ɗʌ̃ũŋ͡m (L. Thompson, 1965, p. 232) please give me two pastries ‘Please give me two pastries’ (‘cͻ’ as a double-object verb) (22) cιiʌι viet t’ͻ cͻ toι (L. Thompson, 1965, p. 332) she write letter give me ‘She wrote a letter to/for me’ (‘cͻ’ as an IO marker or a beneficiary marker) (23) sιin ʌ̃ũŋ͡m viet t’ͻ cͻ toι ɓiet please write letter let me know ‘Please write a letter to let me know’ (‘cͻ’ as a causative / permissive verb)
The passive marker in Vietnamese is [ɓιi] (meaning ‘to suffer, to undergo’), [zͻ] (meaning ‘to be caused, to be effected by’) or [ɗwŏk] (meaning ‘to receive, to obtain’) but not [cͻ] (L. Thompson, 1965). When examining the multiple functions of GIVE in the neighboring non-Sinitic languages spoken in China, we find that GIVE is used as an IO marker and a causative verb but not many of them use GIVE to perform the passive marker function.62 To explain why some non-Sinitic languages have not yet developed the passive usage of GIVE, Hashimoto (1988) postulated that “the Southern Chinese developments of the passive marker from the verb ‘to give’ could thus be an autonomous development” (p. 350). Although the development is “not necessarily related to the curious connection between the passive markers and the verb ‘to give’ or ‘to give back’ in the southern minority languages of China” (p. 350), there could still be a possibility that these non-Sinitic languages borrowed this mechanism and then developed the passive marker usage. In other words, the passive marker function of GIVE can be taken as a loan feature spreading from the Southern dialects to some adjacent non-Sinitic languages.
62. Out of the 31 non-Sinitic languages we surveyed, only 10 languages (from three families) have the double-object verb GIVE functioning as a passive marker. These include (1) Pulai 仆拉 語 and Kazhuo 卡卓語 from the Tibeto-Burman group; (2) Dai 傣語, Zhuang of Longzhou 龍 州, Mo 莫話, Cun 村語, Mulao 木佬語 and Buyang 布央語 of the Kam-Tai family; (3) She 畬語 and Mien 勉語 of the Miao-Yao family. For a comprehensive study of the passive construction in Sino-Tibetan languages, see J. Li (2008).
36 Cantonese give and Double-Object Construction
2.5
Multi-functionality of GIVE in world’s languages
It has been reported that in languages other than Chinese and Southeast Asian languages, GIVE can also function as an IO marker, a causative verb or a passive marker. However, one point to be kept in mind is that no language (as far as the materials we surveyed concerned) has all the three syntactic functions performed by GIVE. 2.5.1 As an IO marker or a beneficiary marker In Yoruba which is spoken in West Africa (mainly in Nigeria), the double-object verb [fun] (meaning ‘to give’) can function as an IO marker or a beneficiary marker. (24) O ta-a fun mi (Michaelis & Haspelmath, 2003, p. 4) he sell-it give me ‘He sold it to/for me’
In the European-based creole languages of Saramaccan and Fa d’Ambu spoken in Africa, the GIVE verb [da] can also function as an IO marker, as shown in (25) and (26) (Michaelis & Haspelmath, 2003, p. 4). (25) Mi manda biifi da hen I send letter give her ‘I have sent letters to her’ (26) Amu da wan kuzu da bo I give art thing give you ‘I gave something to you’
2.5.2 As a causative verb In Jacaltec of the Mayan language groups spoken in Guatemala, the verb GIVE [a’] or [a’a] can function as a causative verb (Newman, 1996, p. 175). (27) ch-ach w-a’ xewoj asp-you I-give rest ‘I make you rest’ (28) cake x-’a’a-ni-ayoj ixim awal wind asp-give-suf-fall cl cornfield ‘The wind made the corn fall down’
Chapter 2. Multi-functionality of give in Chinese dialects and neighboring languages 37
2.5.3 As a passive marker (and a causative marker) In Malay, an Austronesian language, the verb GIVE can be used to express the causative or the passive meaning (Yap & Iwasaki, 2003, p. 428).63 (29) Ayah bagi orang tipu father give people cheat ‘Father lets people cheat’ / ‘Father was cheated (by someone)’
2.6 Summary In this chapter, we surveyed the double-object verbs in major Chinese dialects and neighboring non-Sinitic languages from the perspective of language contact and areal linguistics. We found that some forms of the double-object verb which are dominant in one dialect group also exist in other dialects or languages. This is usually found in those dialects spoken in the areas where two (or more) dialects or languages are used. As for the syntactic functions of GIVE, the Northern dialects are found to use another morpheme as a causative verb and a passive marker. This may be due to the influence of the Altaic languages which had been in contact with the Chinese language for a long period. Although the passive marker function of GIVE is not an areal feature of the Southeast Asian linguistic area, we have seen that some non-Sinitic languages also have this function performed by GIVE and this is possibly due to the influence of the Southern Chinese dialects.64 With these observations, a pertinent question to raise is why GIVE can perform these functions. Is there any relationship among all these functions or is it simply a coincidence that all these functions are expressed by the same morpheme? It is claimed that the multiple functions of the double-object verb GIVE can be correlated with the basic and core meaning of GIVE. In other words, although these markers are distinct in terms of functions, they should be considered “polysemy rather than homonymy” (Newman, 1993, p. 433). This approach thus argues that meaning extension of GIVE is language independent because the elaboration is related to the semantic profile of the verb concerned. In other words, GIVE undergoes meaning extension. Syntactically, it develops from a content word into a function 63. As noted by Yap and Iwasaki (2003), the use of GIVE to express the passive meaning is only found in some dialects of Malay. 64. For more examples, see Matisoff, 1991; Newman, 1999; Lord et al., 2002; Yap & Iwasaki, 2003; H. Y. Zhang, 2002.
38 Cantonese give and Double-Object Construction
word and this process is known as grammaticalization (Hopper & Traugott, 1997; Heine & Kuteva, 2002). In Chapter 3, we will illustrate and discuss how the theory of grammaticalization can account for the multi-functions of give. We will also show that the passive meaning of give is derived from the causative meaning. Lastly, it is relevant to discuss the relationship between grammaticalization and areal linguistics. On the surface, these two notions are mutually exclusive because grammaticalization, generally speaking, refers to a language internal development while areal linguistics is concerned with language contact which is extralinguistically motivated. As noted by Enfield (2001) in his study of the post-verbal usage of acquire in some Southeast Asian languages, it is the “functional application – not the form – that is shared” [italics original] (p. 287). This type of borrowing is termed as contact-induced grammaticalization by Heine and Kuteva (2003). The use of GIVE as a passive marker in some non-Sinitic languages can serve as an illustrative example. Although Heine and Kuteva (2003) admitted that it is sometimes difficult to determine whether the linguistic developments are “the result of universal principles of grammaticalization” or are “due to the specific factors obtaining in situations of language contact” (p. 561), these two approaches are “in no way mutually exclusive alternatives” (p. 562). Therefore, it is possible to discuss grammaticalization from both areal and typological perspectives.
Chapter 3
Grammaticalization of GIVE in Cantonese
Chapters 3 and 4 focus on the multi-functionality of GIVE in Cantonese. Chapter 3 discusses the relationship of the various syntactic functions performed by GIVE under the notion of grammaticalization. Chapter 4 examines the diachronic development of these functions through early dialectal materials of Cantonese. 3.1
What is grammaticalization?
The concept of grammaticalization was first introduced by Meillet (1958), who defined it as “le passage d’un mot autonome au role d’element grammatical” (p. 131).65 A similar claim can be found in Heine and Kuteva (2002), who defined grammaticalization as “the development from lexical to grammatical forms and from grammatical to even more grammatical forms” (p. 2). According to Hopper and Traugott (1997), grammaticalization refers to “the actual phenomena of language … whereby items become more grammatical through time” (italics is mine) (p. 2). ‘Grammatical’ indicates that the direction of the process usually goes from concrete to abstract. Syntactically, the lexical item in question develops from a content word into a function word. The core lexical meaning of the item is gradually lost and the item then mainly performs grammatical functions in the language. Furthermore, the grammaticalization process is gradual and usually takes a long time to complete. Therefore, it is always the case that some morphemes are in the mid-way of the process and can still function as content words in some contexts while some morphemes might have completed the grammaticalization process and no more core lexical meaning is retained. These two situations can be best exemplified by the morphemes ba 把 and bei 被 in standard Mandarin respectively. Ba 把 in standard Mandarin is mainly used as a disposal marker. At the same time, ba 把 is still a verb in some phrases such as ba guan 把關 ‘to guard a pass’, ba zhe men 把著門 ‘guarding the door’ and ba chi 把持 ‘to control or to dominate’. Bei 被 was
65. My translation: The development of an independent word into a grammatical word.
40 Cantonese give and Double-Object Construction
originally a verb meaning ‘to cover, to suffer’66 and was later grammaticalized into a passive marker. After grammaticalization, the verbal properties of bei 被, such as taking aspect markers le 了, zhe 著 and guo 過, are lost.67 There are some notable features of grammaticalization: a. Hopper and Traugott (1997) observed that it is usually hyponyms or generic terms rather than specific terms that are more prone to be grammaticalized. For example, the future tense marker in English can be rendered by the directional verb go as in the sentence ‘It is going to rain tomorrow’. Go, when compared with other verbs also describing physical movement such as walk, run, jog, simply denotes a physical proceeding without any specific details such as manner or speed. Bybee et al. (1994) thus claimed that “it is lexical items of degree of generality … that enter into grammaticization” (p. 5).68 In Chapter 1, we showed that GIVE is one of the basic or nuclear lexical items in many languages and it is also the core member of the double-object verb family.69 Therefore, this verb is likely to participate in the grammaticalization process. b. Hopper and Traugott (1997) claimed that generalization is one of the means by which the semantic content of a morpheme is weakened and this results in “an increase in the polysemies of a form” (p. 96). Newman (1993) argued that lexical items, after grammaticalization, usually perform more than one grammatical function and appear in more than one type of syntactic construction. These items cannot be treated as homonymies but polysemies which involve meaning extension of the lexical items concerned. c. Grammaticalization involves a series of stages of linguistic development, and it is not impossible for us to observe the various functions of the same source at “different stages along the same grammaticization path” (Bybee et al., 1994, p. 5). If we are provided with sufficient and pertinent diachronic materials, we might be able to trace the chronological development of the various functions performed by the same morpheme. This will be further discussed in Chapter 4.
66. In A Dictionary on Chinese Characters 漢語大字典 (Hanyu dacidian bianji weiyuanhui, 1986), bei 被 has nine senses (pp. 3085 & 3086). Three of them are relevant to the meanings discussed here: fugai 覆蓋 ‘to cover’, shou 受 ‘to receive’, zaoshou 遭受 ‘to suffer’ and jieci 介詞 ‘as a preposition used in a passive construction’. 67. See L. Wang (1996) and H. M. Zhang (2005) for the historical development of bei 被. 68. Notice that some linguists use grammaticization instead of grammaticalization.
69. See Table 1 in Chin (2011) on the list of double-object verbs found in the early textual materials of Cantonese. The double-object verb [pei] 畀 takes up nearly two-thirds of the double-object verbs found in these early materials.
Chapter 3. Grammaticalization of give in Cantonese 41
3.2
Grammaticalization of GIVE
In Chapter 2, we showed that GIVE performs multiple functions in Chinese dialects as well as some non-Sinitic languages in the linguistic area of Southeast Asia. In the following, we will examine the semantic relatedness between the core meaning of GIVE and its extended meanings in modern Cantonese. 3.2.1 As an indirect object marker One of the grammatical functions that GIVE performs is to introduce the indirect object (IO) with the semantic role of recipient in a double-object construction, such as [pei] 畀 in the sentence [ŋͻ13 sʊŋ33 ts’in11–35 pei35 k’ɵy13] 我送錢畀佢 ‘I gave him money’. In Chapter 1, we showed that [pei] 畀 in this double-object sentence is not a main verb but a preposition-like element introducing the IO and it is thus called an IO marker. The association of GIVE with the IO marker function is transparent because the thing is given to the recipient in the action of giving. As Newman (1996) claimed, the use of GIVE to mark the recipient “is not uncommon in serial verb languages [such as Chinese, African and Thai languages], in which verbal morphemes more readily take on some purely relational and prepositional characteristics” (p. 212). However, it is found that prior to the 1940s, the IO marker in Cantonese was performed by the directional verb [kʷͻ] 過.70 In Chapter 4, we will examine the diachronic development of the IO marker in Cantonese and specifically, how the development of [pei] 畀 into an IO marker can be correlated with the beneficiary construction in which the main verb carries the [−GIVE] semantic feature. 3.2.2 As a beneficiary marker In standard Mandarin, gei 給 can mark a beneficiary, as shown in (30).
(30) 我 給 小王 寄 了 一 本 書 wo gei xiao wang ji le yi ben shu I give Wang send asp one cl book Reading A: ‘I sent a book to Xiao Wang’ Reading B: ‘I sent a book (to someone) on behalf of / for Xiao Wang’
70. A preliminary study on the IO marker in pre-modern Cantonese can be found in Takashima and Yue (2000).
42 Cantonese give and Double-Object Construction
Xiao Wang after gei 給 can have the semantic role of either a recipient (Reading A) or a beneficiary (Reading B). As a recipient, Xiao Wang received the book sent by the speaker. As a beneficiary, Xiao Wang is not the recipient of the book. Instead, the speaker sent the book to someone [i.e., other than Xiao Wang] on behalf of Xiao Wang. In this sense, Xiao Wang benefits from the action carried out by the speaker. As noted by Kittlä (2005), the semantic roles of recipient and beneficiary are common so that they “usually benefit from the events they are parts of ” (p. 269).71 These two actions are closely related, and it is thus not unusual for a language to have the same morpheme to mark these two semantic roles. In Hong Kong Cantonese, there are two types of beneficiary constructions. In the first type, the beneficiary is placed before the main verb which is similar to standard Mandarin discussed above. The beneficiary is marked by the morpheme [t’ʊŋ] 同 ‘to be with’ or [pͻŋ] 幫 ‘to help’ instead of [pei] 畀, as shown in (31).72 (31) 我 同/幫 老 王 寄 咗 一 本 書 ŋͻ13 t’ʊŋ11/pͻŋ55 lou13 wͻŋ11–35 kei33 tsͻ35 jɐt5 pun35 sy55 I for/help old Wang send asp one cl book ‘I sent a book (to someone) on behalf of / for Wang’
The second type uses the post-verbal beneficiary marker [pei] 畀 and the main verb is a non-double-object verb. (32) 我 織 咗 件 冷 衫 畀 你 ŋͻ13 tsιk5 tsͻ35 kin22 laŋ55 sam55 pei35 nei13 I knit asp cl sweater give you ‘I knitted a sweater for you’
The surface structure of this type of beneficiary construction is the same as the double-object construction: NP1 + V + NP2 + 畀 + NP3. The crucial difference is the semantic feature of the main verb. For the beneficiary construction, the main verb is a non-double-object verb while in the double-object construction, the main verb has the intrinsic [+GIVE] feature which subcategorizes two objects. In other 71. See also the discussions in Vietnamese and Khmer (Bisang (1992), as cited in Kittlä & Zúñiga (2010), p. 9), Thai (Jenny, 2010) and Tibeto-Burmese languages (LaPolla, 2007, p. 33).
72. See the rendition of the sentence gei dajia banshi 給大家辦事 ‘work for everybody’ in the Yue dialects of the Pearl River Delta region collected in Zhan and Cheung (1988). None of the 31 Yue dialects uses GIVE as the beneficiary marker when the beneficiary is placed pre-verbally. According to X. Huang and He (2018), GIVE used as a beneficiary marker is mainly found in Mandarin dialects (except the Southwestern and Jianghuai Mandarin sub-groups), Hui dialects and some Wu and Min dialects. Yue dialects do not use a pre-verbal beneficiary marker cognate with GIVE.
Chapter 3. Grammaticalization of give in Cantonese 43
words, the 畀-NP3 phrase is not an obligatory constituent in a beneficiary construction. In (32), knitting a sweater does not necessarily require giving it to another person.73 Sometimes, the beneficiary marker can be preceded by a double-object verb, as (33) shows.
(33) 我 買 層 樓 送 畀佢 ŋͻ13 mai13 ts’ɐŋ11 lɐu11–35 sʊŋ33 pei35 k’ɵy13 I buy cl flat give give s/he ‘I buy a flat for him/her’ (lit.: I buy a flat and give it to him/her)
However, adding a double-object verb before the IO marker [pei] 畀 in a double- object construction is not possible, as (34) shows. (34) *我 遞 枝 筆 送 畀佢 ŋͻ13 tɐi22 tsi55 pɐt5 sʊŋ33 pei35 k’ɵy13 I pass cl pen give give s/he ‘Lit.: I passed a pen and gave it to him/her’
Thus, [pei] 畀 in the beneficiary and the double-object construction are two types of markers as formulated in (35) although the two constructions share the same surface syntactic structure. Therefore, beneficiary construction is more like a serial verb construction. (35) Double-object: NP1 V[+GIVE] NP2 畀[IO marker] NP3[recipient] Beneficiary: NP1 V[-GIVE] NP2 畀[beneficiary marker] NP3[beneficiary]
3.2.3 As a causative verb The causative meaning of GIVE can be illustrated by the double-object verb [hͻ] in the Southern Min dialect of Taiwan.
(36) 我 hͻ 伊 得 第 一 名 I give s/he get prf one cl ‘I caused him/her to get the first prize’
(L. Cheng et al., 1999, p. 150)
In Hong Kong Cantonese, [pei] 畀 cannot express the causative meaning. (37) is ungrammatical if [pei] 畀 is interpreted with the causative sense. 73. In this sentence, the beneficiary can also be a recipient of the sweater. X. Huang and He (2018) identified three types of beneficiaries: beneficiary of a service (e.g., gei dajia banshi 給大家辦事), beneficiary as a recipient (e.g., gei haizi da maoyi 給孩子打毛衣), and beneficiary as a result of acting on one’s behalf (e.g., ti ta mai dongxi 替他買東西).
44 Cantonese give and Double-Object Construction
(37) *我 畀 佢 攞 第 一 名 ŋͻ13 pei35 k’ɵy13 lͻ35 tɐi22 jɐt5 mιŋ11 I give s/he get prf one cl ‘I caused him/her to get the first prize’
Instead of the causative meaning, (37) expresses the meaning of permissiveness: ‘I let him/her get the first prize’. Although the causative meaning is not obtained for [pei] 畀, it is still possible to consider that the meaning of causative subsumes the meaning of permissiveness because permitting someone to do something can be understood as causing someone to do something. The link between the causative meaning involving GIVE was discussed in Kemmer and Verhagen (1994) who claimed that “causatives of transitive predicates … are seen as modeled on simple three-participant clauses” (p. 115). The use of GIVE as a causative verb shows that “the participants of the causative structure are recurrently seen as analogous to the participants of a ditransitive clause [i.e., double object construction]” because double object verbs also “involve the idea of an agent causing an entity to come into possession of something” (pp. 128–129). Newman (2005) claimed that “‘causative’ is also common as a grammaticalized extension of ‘give’” (p. 158). The giver, the recipient and the thing in the action of giving correspond to the causer, the causee and the caused event respectively. H. Lai (2001) argued that the meaning of the transfer “can be easily extended from a noun phrase to a verb phrase” (p. 146) in the action of giving (see also Yue-Hashimoto (1993a, p. 131)). What H. Lai (2001) suggested is that “the meaning of giving someone something is transferred into giving someone the permission to do something” (p. 146). In some studies on the causative function of the double-object verb GIVE, the sense of ‘permissiveness’ is also included.74 In this regard, [pei] 畀 in Hong Kong Cantonese can also express the causative meaning. (38) 我 畀 你 出 去 玩 ŋͻ13 pei35 nei13 ts’ɵt5 hɵy33 wan35 I give you go out play ‘I let you play outside’
Although the causative meaning can be derived from the meaning of GIVE, the syntactic pattern of the causative construction is different from the double-object construction. (39) and (40) give the structural formulae of the two constructions with the semantic feature of the three noun phrases (see also Chin, 2011, p. 539). (39) Causative:
NP1[causer] 畀 NP2[causee] [V NP3[affectee]][caused event]
(40) Double-object: NP1[giver] 畀 NP2[thing] NP3[recipient]
74. Matisoff (1991) labelled this function of GIVE as a ‘permissive-causative’ function.
Chapter 3. Grammaticalization of give in Cantonese 45
3.2.4 As a passive marker Hong Kong Cantonese uses [pei] 畀 to express the passive meaning. Matthews and Yip (1994) claimed that “the usual spoken form béi [with a mid-rising tone [pei35]] originates as beih [i.e., 被 with a mid-low level tone [pei22]] with a changed tone” (p. 150).75 In other words, they argued that the passive marker in colloquial Cantonese is [pei] 被 instead of [pei] 畀. However, we do not support this claim.76 Syntactically, passive sentences using the marker bei 被 can leave out the agent, such as Zhang San bei da le 張三被打了 ‘Zhang San was beaten’. On the contrary, the agent cannot be elided after the Cantonese passive marker [pei] 畀.77 If the Cantonese passive marker is [pei] 被, then one has to explain why there is a difference in the grammaticality of the omission of the agent. As the Chinese dialect examples in Chapter 2 show, the passive marker in many dialects is identical to the double-object verb GIVE in the respective dialects. It is then more accurate to claim that the colloquial form of the passive marker in Hong Kong Cantonese is associated with the double-object verb [pei] 畀 instead of [pei] 被. 3.2.4.1 Relationship between causative and passive constructions In the following, we discuss how [pei] 畀 can function as a passive marker. Specifically, we argue that the passive marker function of [pei] 畀 is derived from the causative function of [pei] 畀. Compare the following pair of sentences in standard Mandarin.78 (41) 我 讓 他 打開 了 箱子 wo rang ta dakai le xiangzi I let he open asp box ‘I let him open the box’
(42) 我 讓 他 打傷 了 頭 wo rang ta dashang le tou I let he beat-injure asp head ‘My head was beaten (and then injured) by him’ or ‘I let him beat my head’
75. This claim was not found in the second edition of the same book (Matthews & Yip, 2011). 76. Cheung (1996) also had a similar comment on this claim. 77. This obligatory presence of the agent after the passive marker is also commonly found in other Southern dialects (Zhan, 1981). 78. The same also applies to jiao 叫/教.
46 Cantonese give and Double-Object Construction
It is noted that both (41) and (42) have the same surface structure and rang 讓 expresses the meanings of causative in both sentences. In addition, (42) also expresses the meaning of passive. Then how does the passive reading of (42) come about? Newman (1993) claimed that it is not easy to figure out how the meaning of GIVE (as well as causative) derives the meaning of passive.79 According to L. Jiang (2000, pp. 231–232), the passive meaning is obtained when a. the verb phrase after rang 讓 is transitive and carries the meaning of passive (shoudong 受動); b. the noun phrase before rang 讓 carries the semantic role of a patient.80
A similar claim is also found in S. Jiang (2003), who argued that the passive meaning develops from the causative meaning. S. Jiang (2003, p. 215) listed three conditions under which a causative construction is interpreted as a passive construction:81 a. There is no morphological distinction between the active and passive constructions in Chinese; b. The verb of the causative construction is transitive; c. The subject of the causative verb is not an agent but a patient.
The common feature shared by these two proposals is that there is a switch of the semantic role of the noun phrase preceding rang 讓 (i.e., the subject of the sentece) from an agent to a patient. However, the two proposals still cannot adequately explain why there is a switch of the semantic role for the subject and also why and how the verb is interpreted with the passive (shoudong 受動) meaning. In (41) and (42), both verbs dakai 打開 ‘to open’ and dashang 打傷 ‘to injure by beating’ are transitive. Both sentences have a causative meaning but the passive meaning is only obtained in (42) but not in (41). One of the crucial features shared by both the causative and passive constructions is that their surface structures are identical, as summarized in (45) on the basis of sentences (43) and (44) from Hong Kong Cantonese. 79. D. Xu (2005) compared the directionality of the actions denoted by the markers gei 給, bei 被 and ba 把. Among these three markers, D. Xu (1994) found that gei 給 is neutral because the directions of the action can be either toward the agent or the patient. In particular, she quoted the use of gei 給 in the dialect of Hankou 漢口 in Hubei (a Southwestern Mandarin dialect), in which the sentence lang gei yang chi le 狼給羊吃了 can either mean ‘the wolf was eaten by a sheep’ (a passive reading) or ‘the wolf ate the sheep’ (a disposal construction which is similar to lang ba yang chi le 狼把羊吃了). 80. The original text reads as: 當“與+VP”或“與+N+VP”結構中的VP是及物性的、並且表示 受動時,就轉為被動結構;當這兩式前面有名詞、並且表示受事時,就轉為被動句。
81. The original text reads as: 漢語動詞主動和表被動在形式上沒有區別;“教”字句的謂語動 詞是及物的;“教”字前面的不是施事主語,而是受事主語。
Chapter 3. Grammaticalization of give in Cantonese 47
(43) 我 畀 佢 出去 ŋͻ13 pei35 k’ɵy13 ts’ɵt5hɵy33 I give s/he go out ‘I let him/her go out’
(44) 我 畀 阿爸 打 ŋͻ13 pei35 a33pa11 ta35 I give father beat ‘I was beaten by my father’ or ‘I let my father beat me’ (45) NP1 + 畀 + NP2 + VP
Given the same syntactic patterns, it is therefore sometimes difficult to distinguish between the two meanings, as shown in sentence (46). (46) 我 畀 隻 狗 食 晒 啲 飯 ŋͻ13 pei35 tsεk3 kɐu35 sιk1 sai33 ti55 fan22 I give cl dog eat suf cl rice Causative: I let the dog eat up all the rice. Passive: All my rice was eaten up by the dog.
The interpretation of the causative and the passive readings depends on whether the second action (i.e., [sιk1 sai33 ti55 fan22] 食晒啲飯, ‘eat up the rice’) is a desirable action. If this action is not favored by the agent, the passive reading is thus preferred. The subtle difference between the two readings lies in semantics instead of syntax.82 According to Yap and Iwasaki (2003), the causative or permissive sentence obtains the passive reading when the permissive meaning becomes an unwilling permission where the subject involuntarily controls the action carried out by the object.83 This gives rise to the adversative and unfavorable meanings for the passive construction, which are one of the major characteristics of the Chinese passive construction.84 L. Zhang (2006) held a similar view and claimed that the causative 82. See also Fagerli (2001) on the discussion of the possible malfective meaning of GIVE. Matthews et al. (2005, p. 274) adopted the notion “bridging context” to account for the multiple readings of the same syntactic structure. Furthermore, Kittlä and Zúñiga (2010) found that there are languages in which malfective and beneficiary are introduced by the same marker. 83. Syntactically, Yap and Chen (2018) claimed that the morpheme GIVE undergoes valence reduction. In the double-object and permissive constructions, GIVE sub-categorizes two objects (i.e., a three-place predicate), but in the passive construction, it is demoted to a 2-place predicate. 84. It has been claimed that the passive construction in Chinese is used to express unfavorable or undesirable events. According to L. Wang (1996), under the influence of western languages, this situation had changed and the passive construction can now refer to desirable or favorable situations such as ta bei ren zan le 他被人讚了 ‘he was praised by someone’. The passive sentence [k’ɵy13 pei35 lou13 pan35 tsan33] 佢畀老闆讚 ‘S/he was praised by the boss’ is also acceptable in Hong Kong Cantonese.
48 Cantonese give and Double-Object Construction
construction first derived the meaning of involuntary permission ‘fei ziyuan yunrang 非自願允讓’ and later developed into the passive construction. The route of the development is shown as follows (L. Zhang, 2006, p. 141): Causative/Permissive > involuntary permissive > passive Such a semantic difference between the causative and passive constructions is also found in German. A German sentence containing the causative verb lassen ‘let, make’ with an inanimate subject can have a passive reading. Compare (47) and (48), which are taken from Haspelmath (1990, pp. 46–47). (47) Nesrin lasst sich fotografieren ‘Nesrin has herself photographed’ (48) Das Fahrrad lasst sich reparieren ‘The bike can be repaired / is repairable’
In some languages such as Indonesian, the causative and the passive constructions share the same suffix -kan (Sneddon, 1996, pp. 76 & 252). (49) Siti sudah memkesihkan kamar ini clean-causative the room Siti has ‘Siti has caused the room to be clean’ (meaning: Siti has cleaned the room) (50) Uang itu sudah diberikan kepada adiknya Money that has given to brother ‘The money has been given to his brother’
Thus, the passive construction has a close relationship with the causative construction on the semantic level. Since [pei] 畀in Hong Kong Cantonese can express the causative/permissive meaning, it is possible and natural for it to develop into a passive marker. 3.2.4.2 Passive morphology The association of GIVE with the passive construction can also be accounted for by the typology of the passive morphology in world’s languages. In many languages, the passive construction can be rendered by means of morphological devices, such as by attaching an affix to the verb stem which then converts the verb from active to passive (see, for example, Keenan (1985) and Haspelmath (1990)). Some languages such as English use auxiliary verbs together with the main verb to form the passive structure. According to Keenan’s (1985) cross-linguistic survey on passive morphology, these auxiliary verbs usually are ‘verbs of being or becoming’, ‘verbs of reception’ or ‘verbs of motion’ (pp. 257–261). The passive construction made up of bei 被 in standard Mandarin belongs to the category of ‘verbs of reception’ since
Chapter 3. Grammaticalization of give in Cantonese 49
bei 被 was originally a verb meaning ‘to receive’. On this basis, we can extrapolate the concept of reception to include giving since both reception and giving are complementary to each other. One can receive something only when s/he is given something.85 The four syntactic functions of GIVE discussed above are commonly found in modern Chinese dialects as well as other languages (see the survey in Chapter 2). We now examine the fifth function of [pei] 畀 which was seldom reported before. 3.2.5 As a verb introducing instruments In Hong Kong Cantonese, [pei] 畀 can be used to introduce a noun phrase with the semantic role of instrument. This function is not common or productive in modern speech but can still be found among some older speakers’ speech. The following sentences were collected from a 65-year-old Hong Kong Cantonese speaker when he was asked to demonstrate how to prepare a dish. (51) 畀 個 袋 包 住 隻 雞, 跟住 擠 入 雪櫃 pei35 kͻ33 tͻi22–35 pau55 tsy22 tsεk3 kɐi55 kɐn55tsy22 tsɐi55 jɐp1 syt3-kʷɐi22 give cl bag wrap suf cl chicken then put into fridge ‘Use a bag to wrap the chicken and then put it into the fridge’
(52) 首先 畀 水 洗 乾淨 隻 雞, 跟住 畀 sɐu35sin55 pei35 sɵy35 sɐi35 kͻn55-tsεŋ22 tsεk3 kɐi55 kɐn55tsy22 pei35 first give water wash clean cl chicken then give 豉油 腌 佢 兩 個 鐘頭 si11-jɐu11 jip3 k’ɵy13 lœŋ13 kͻ33 tsʊŋ55-tɐu11 soy-sauce marinate it two cl hour ‘First, clean the chicken with water. Then, marinate it with soy sauce for two hours’
Younger speakers of Cantonese tend to use [jʊŋ] 用 instead of [pei] 畀 to introduce the instruments (i.e., the bag, water and soy sauce). The function of introducing an instrument by [pei] 畀 can still be found in the crystallized expression [pei35 sɐm55 kei55] 畀心機‘to use one’s heart’. The standard Mandarin equivalents are also expressed with the verb meaning ‘to use’, such as hua xinji 花心機, yong xin 用心.86 85. According to the typological study of the syntactic constructions in her five linguistic areas in Mainland China, Chappell (2015) classified the passive markers in the Northern linguistic area as the causative type while those in the dialects in the Southeastern and Far Southern linguistic areas as the give type. 86. Chin and Tsou (2005) and Yiu (2010) found this usage in 19th century Cantonese. This will be discussed further in Chapter 4.
50 Cantonese give and Double-Object Construction
This function of GIVE is still common in the Taishan 台山 dialect of the Siyi area, as shown in the following examples.87
(53) □ 心機 ɂei55 ɬim33-kei33 give heart ‘Study hard’
讀書 uk32-ʃi33 study
(54) □ 條 繩 綁 個 狗 55 11 11 55 ɂei hiu ʃaŋ pͻŋ kui33 kau55 give cl string tie cl dog ‘Tie up the dog with a string’ (55) □ 筷箸 夾 餸 55 33 55 ɂei fai -tu kap3 ɬuŋ33–21 give chopsticks clip dish ‘Clip the dish with chopsticks’
(56) 件 衫 □ 凍 水 洗 ken32 ʃam33–21 ɂei55 uŋ33 ʃui55 ɬai55 cl shirt give cold water wash ‘Wash the shirt with cold water’
The use of [pei] 畀 to introduce an instrument is seldom reported in other Chinese dialects88 and is not cross-linguistically attested (Heine & Kuteva, 2003). It is relevant for us to explore how the core meaning of [pei] 畀 is associated with this function. A double-object construction is used to express a three-participant event. When viewing an action involving an instrument, we also obtain a three-participant event. According to Stolz (2001), “[p]rototypical situations involving instrumentals require three participants, viz. an agent – the one who deploys the instrument –, the instrument itself, and a patient” (p. 171). Syntactically, the two constructions are realized as follows.
87. This is based on an informal interview in 2008 with an 88-year-old female speaker living in Seattle. According to the informant’s daughter-in-law who also speaks the same dialect, [ʃͻi] 使 (meaning ‘to use’) can also be used in addition to [ɂei] to introduce the instrument. 88. In Zhan and Cheung’s (1988) survey of 31 Yue dialects in the Pearl River Delta region, the verb introducing an instrument is either yong 用 or shi 使. In First Book of Reading 啟蒙淺學 written by the Basel Missionary Society in the Hakka dialect in 1879, the double-object verb [pin] (written with the character 俾) could introduce an instrument, as in the sentence 在山裡鋤倒 個不過係石頭, 人愛俾火燒過佢正成石灰 (Zhuang & Huang, 2014, p. 289) ‘The things that people dug up in the mountain were only stones. People used fire to burn them to make lime’ [my translation].
Chapter 3. Grammaticalization of give in Cantonese 51
a. GIVING: NP1 + 畀 + NP2 + NP3 b. INSTRUMENTAL: NP1 + 畀 + NP2 + VP Notice that the surface structure of the instrumental usage is identical to the ones of permissive and passive. This will be further discussed in Chapter 4. In many non-European languages, there is a syncretism of the markers for the instrument and the one for the agent (see also Nilsen, 1973). Stolz (2001) argued that the syncretism is not a random process and there must be “some reason for the empirical fact that certain categories combine more easily in syncretism than others do” (p. 170). The common feature shared by both the agent and the instrument again is causativity. The agent deliberately causes and instigates an action to take place while the instrument is an assistant or a catalyst to cause the action to take place. 3.3
Summary
In this chapter, we discussed, with the notions of grammaticalization, how and why the double-object verb [pei] 畀 develops into an IO marker, a beneficiary marker, a verb introducing an instrument, a causative verb which further develops into a passive marker.89 We have also shown that some of the subtle differences among these five functions are closely related to semantics than syntax as a result of reanalysis which is a common mechanism in language change. These various functions of [pei] 畀 can be parameterized by the semantic attributes causative and receive, as claimed by A. Yue (2003): “‘to give’ is ‘to cause to receive’ (causative) while ‘to receive’ (passive) is ‘to cause to give’ (causative)” (p. 110). These four constructions can be parameterized by the semantic features of [±causative] and [±receive] with the following formulae: A. B. C. D.
double-object90: [+causative, +receive, NP2(DO), NP3(IO)] permissive: [+causative, +receive, NP2(causee), VP] passive: [− causative, +receive, NP2(causer), VP] instrument: [+causative, −receive, NP2(instrument), VP]
89. A more comprehensive cross-linguistic investigation of the grammatical functions developed from GIVE, including purpose, beneficiary, dative, causative and concern, can be found in Heine and Kuteva (2002). 90. This also includes the beneficiary marker and the IO marker.
52 Cantonese give and Double-Object Construction
In a double-object construction, what the agent (NP1) does is to cause the recipient (NP2) to receive the thing (NP3). In a permissive construction, the agent (NP1) causes someone (by giving a permission) to carry out an action (VP). In this sense, the recipient (NP2) receives something. As our previous discussion showed, in a passive construction, NP1 (i.e., the subject) semantically speaking is an unwilling causee to receive the action performed by NP2 who is a causer instigating the action denoted by VP. In this sense, NP1 is affected (i.e., receives) by the action. In the instrumental construction, the agent uses the instrument to perform another action. The agent thus causes an action to take place. The inter-relationship of GIVE in these syntactic constructions fits into the semantic map (Figure 2) proposed by M. Zhang (2015, p. 47) for the inter-relationship of the grammatical relations in Chinese. Semantic map is “a method for visually representing cross-linguistic regularity or universality in semantic structure” and for displaying “the interrelationships between meanings and functions across languages” and “to describe patterns of polysemy” (Georgakopoulos, 2019).91 The various functions taken up by GIVE in Hong Kong Cantonese can be covered by a continuous area in the semantic map.92 Furthermore, M. Zhang (2003, as cited in Yiu (2010, p. 168)) discussed the inter-relationship of the various functions of GIVE in the Idealized Cognitive Model (理想化認知模式) [my translation]: The action of giving refers to a situation where A gives a thing to B by hand, and B receives the thing with his hands. This action has the following properties: a. b. c. d. e.
Giving is an action with intention; In the action of giving, the giver (i.e., A) has the original possession of the thing; Having the possession of the thing means having control of the thing; Giving refers to the transfer of the possession from A to B; With (c) and (d), giving refers to the transfer of the control of the thing from A to B; f. The control is not limited to an entity, but also an event; g. After giving, the giver loses something while the recipient gains something.
91. For an overview of studying Chinese with the semantic map approach, see M. Zhang (2015). For the application of the semantic map approach on the double-object constructions, see Phua (2009, 2013), Ding and Zhang (2015). 92. In the semantic map, the grammatical function of “pretransitive” is included but this has not been discussed in this chapter. According to Kataoka (2007), there is a morpheme 𢬿 [kaai] in Cantonese which has the basic meaning of ‘to hold’. In 19th century Cantonese, 𢬿 [kaai] was found to have the functions of (a) a disposal marker (similar to ba 把 in Modern Standard Chinese), (b) a verb introducing instrument and (c) an IO marker.
Chapter 3. Grammaticalization of give in Cantonese 53
受益者 Beneficiary
原因 cause
方向/接受者 Direction / recipient 使役 Causative
並列連接 conjunctive
(共同施事者) (co-agent)
處置 pretransitive
伴隨 comitative
工具 Instrumental
被動 Passive
來源 source
方式 manner
Figure 2. Semantic map of the double-object construction in Cantonese (based on M. Zhang, 2015, p. 50)
This chapter provides a synchronic account of the various functions performed by give in Cantonese. These functions may not evolve from the source at the same time. It will be meaningful if we can find out the chronology of the development of these functions of give to understand better the mechanism and conditions of the change. Chapter 4 examines the diachronic development of these grammatical functions of [pei] 畀 in Cantonese with the aid of early Cantonese dialectal materials.
Chapter 4
Diachronic development of GIVE and its functions in Cantonese
In this chapter, we focus on the diachronic development of the various functions performed by [pei] 畀 in Cantonese by examining some dialectal materials published in the 19th and early 20th centuries. The diachronic analysis offers us a better understanding on the grammaticalization process of [pei] 畀, including its sequence, time frame and conditions. The various functions we examined for GIVE in Chapters 2 and 3 are based on contemporary languages or dialects in which all the functions of GIVE already co-exist. According to the grammaticalization theory, when a morpheme develops into grammatical markers performing various syntactic functions, these functions usually do not emerge at the same time. This suggests that some of the markers emerge first in the language, followed by others. Chappell (2001) suggested that “clues to the pathways of grammaticalization and semantic change can only be clearly delineated with reference to precise analyses of earlier stages of the Chinese language” (p. 3). Therefore, if we are provided with sufficient and pertinent diachronic language materials, we would be able to work out the relative chronology of the development of these grammatical markers as well as the conditioning factors triggering the change. In addition, we may also discover some functions that had become obsolete in the contemporary language and the possible reasons for their disappearance. 4.1
Studying the language of the past with authentic textual materials
There are at least two approaches with which one can study the early grammar of a language (see Chapter 3 in Labov, 1999). The first one is the apparent time approach. With this approach, the data is collected from speakers of different ages which represent the language of the periods in which the speakers were born or raised. For example, the language of an 80-year-old speaker in year 2020 is assumed to reflect the language of the 1940s. However, this method becomes useless if we want to study the language of 150 years ago because finding a speaker of this age is impossible. Another important factor we need to take into consideration is that linguistic change may have taken place on the speaker since s/he was born. In other words, we cannot always safely assume that the language of an 80-year-old speaker
56 Cantonese give and Double-Object Construction
in year 2020 is identical to the language of the 1940s. This approach sometimes may draw misleading conclusions if such factors are not taken into serious consideration and well controlled. In their studies of neutral questions,93 Yue-Hashimoto (1991a, 1993b, 1994) and Cheung (2001) found that, on the basis of documents of early Cantonese and Min dialects, the pattern of the neutral question was VP-Neg94 such as [nei13 jɐu13 ts’in11–35 mou13] 你有錢冇 (you-have-money-no, ‘do you have money?’). According to Cheung, this pattern was dominant in Cantonese before the 1940s. After the 1940s, the V(P)-Neg-VP pattern such as [nei13 jɐu13 mou13 ts’in11–35] 你 有冇錢 (you-have-no-money, ‘do you have money?’) became dominant. However, in his fieldwork investigation on the Cantonese dialect, Tsou (1981) found that the V-Neg-VP pattern was also used substantially by speakers over 60 years old in the 1970s.95 Notice that these 60-year-old speakers were born in the early 20th century in which VP-Neg was still the dominant pattern in the dialect. This means that linguistic change had taken place among the speakers. Furthermore, it is also noted that the speed of change varies from speaker to speaker. Thus the apparent time approach may not elicit the most appropriate responses and may not accurately reflect the actual scenario of the language in the early stage. The alternate approach one can take is the real time approach. As Labov (1999) suggested, for the real time approach, “comparative data are drawn from earlier studies with entirely different aims, methods, and theoretical perspectives” (p. 29). This approach is more reliable than the apparent time approach because it is the data of the respective period being investigated. Regarding the syntactic pattern of neutral questions in the Yue dialects, A. Yue (2004) remarked that “[i]f it had not been due to earlier written records, one would not even be aware that the latter form [i.e., VP-Neg] ever occurred in the said dialect” (p. 246). To take the real time approach, one has to refer to the records of the language of the period one intends to study. The use of authentic and real data is thus important and necessary in the diachronic studies of a language. As for the Cantonese dialect, we are fortunate that a considerable amount of textual materials compiled since the 19th century are available to us. These early dialectal materials have different aims and nature: Translation of the Bible, descriptions on the dialect, and documents about Chinese culture, customs, and things.96 93. Neutral questions are also known as “yes-no question” or “polar questions”. 94. ‘Neg’ stands for the negator such as [m̩] 唔, [mou] 冇 in Cantonese.
95. According to Tsou, around 30% of the neutral questions produced by the older speakers were in the form of V(P)-Neg-VP. 96. For more details on the nature of these early Cantonese materials, see A. Yue (2004).
Chapter 4. Diachronic development of give and its functions in Cantonese 57
In spite of the availability of these extant materials, “lexical and syntactic research using such materials did not surface until the present decade [i.e., the 2000s]”, as commented by A. Yue (2004, p. 248). This is partially due to the accessibility of these materials.97 Furthermore, one may raise doubts about the accuracy on the description of the dialects in these texts because many of these texts were compiled by non-Chinese people such as western missionaries whose native language was neither Chinese nor its dialects. Since the 16th century, many western missionaries came to China mainly for religious purposes such as preaching. In order to help their preaching and attract more followers, these missionaries studied the Chinese language and the local vernaculars. They also compiled many dictionaries as well as grammar notes for other missionaries to study. Some of these works became important references on studying the Chinese language and its dialects of earlier periods. The quality of the missionaries’ works on Chinese language and dialects have been highly appreciated and valued by many linguists and Sinologists.98 These missionaries thus left us with valuable materials for historical linguistics research. Among these missionaries, Robert Morrison (1782–1834) was one of the key contributors to the study of Cantonese. He came to China in 1807 and stayed in Guangzhou and Macau for 25 years. During his stay, he compiled many dictionaries and translated the Bible into Chinese. He also compiled language materials on the Yue dialects. One of his famous works is A Vocabulary of the Canton Dialect (廣東省土話字彙).99 Bolton (2001), in the introductory chapter of the reprint of this work, commented that it seems clear that Morrison’s achievement in his Vocabulary [i.e., A Vocabulary of the Canton Dialect] was not simply to provide a practical dictionary and phrase book for one particular dialect of Chinese. His wider achievement was rather to establish a precedent for missionary studies of Chinese dialects that would prepare the way for dialect dictionaries and dialect studies, which would reshape Chinese language study and eventually influence the baihua [白話] movement and the later reforms that paved the way for the creation of the putonghua national language. (p. xxxii) 97. Some of these materials were not found in China or in the areas where the Cantonese dialect is spoken. Quite a few of them were found overseas such as Japan and Europe. See also You (2002) for the list of dialect materials produced by western missionaries. 98. See, for example, C. Luo’s (1930) lengthy study of the phonological systems described by early Jesuits; Standaert (2001) had a chapter on missionaries’ contributions to linguistics. For other related studies, see G. Xu (2000), Ye (2004) and Cheung (2017). 99. Details on Morrison’s knowledge and contributions to the Chinese language as well as Cantonese can be found in C. Lee and Kataoka (2006).
58 Cantonese give and Double-Object Construction
A. Yue (2001) compared the materials on the Chao’an (潮安) dialect of Southern Min recorded by the late Professor Y. R. Chao against the one compiled by the foreign missionaries in the 19th and early 20th centuries. She found that Chao’s data “not only set their own value [in the studies of Chinese dialectology] but often affirm the value of earlier materials” (p. 198). Furthermore, A. Yue (2001) added that “a comparison of these ‘less professional’ materials with the late Professor Chao’s materials often provided confidence of different degrees in the former: the latter serving as some kind of check of possible change from the former, or affirming the possibility of the existence of certain categories in the former” (p. 198). In the above quote, “less professional” materials refer to the ones “most frequently compiled by missionaries” (p. 197). This demonstrates that the works compiled by the missionaries are reliable for dialectal study. Furthermore, we find that the linguistic features pertinent to Cantonese, such as sentence final particles and tone change, were well described and explored in these materials compiled by missionaries.100 Therefore, one cannot dismiss the value of these pre-modern dialect materials. Another important advantage of these early dialectal materials is that they sometimes can allow cross-dialectal comparison, especially in those texts which involve the translation of the Bible because the contents of these materials are the same, as claimed by You (2002). Our pre-modern dialect materials also include three translations of the Gospel of Luke published in different periods.101 4.2
Pre-modern Cantonese dialect materials
A corpus was constructed to analyze the [pei] 畀-related sentences collected from 44 pre-modern Cantonese dialect materials. These 44 materials have different natures and were compiled by different kinds of writers. Sentences related to [pei] 畀, be they double-object, passive or causative, were collected, categorized and then analyzed. In addition to [pei] 畀, we also include sentences with the same constructions but expressed by different markers. For example, sentences with the passive marker [pei] 被, double-object sentences with [kʷͻ] 過 as the IO marker are also the target of the current study. The earliest and the latest materials collected in the corpus were compiled in 1828 by Morrison and the 1990s respectively, covering a time span of about 170 years.102 Among the 44 works, 16 were compiled in the 19th century and the other 100. Relevant studies on these significant linguistic features in both early and modern Yue dialects can be found in Cheung (2000, 2009), C. Leung (2005), and Tsou (1994a). 101. These three editions were published in 1870, 1931 and 1997.
102. Yiu (2010) also examined the various functions of [pei] 畀 in early Cantonese dialect with 16 textual materials.
Chapter 4. Diachronic development of give and its functions in Cantonese 59
28 in the 20th century. The bibliographical details of these 44 works are given in Appendix 2. Table 5 lists the nature of these works and their places of publication. Table 5. Publication details of pre-modern Cantonese dialect materials103 Author
Year
Nature
Place of publication
Morrison Bridgman Devans Hobson Bonney Bunyan Gospel of Luke Dennys Harper Anonymous Stedman & Lee Fulton Ball Mai Mai Ball Wisner103 LeBlanc Ball Cowles Ball Caysac Wisner Gospel of Luke Wells Brouner & Fung Hoh & Belt O’Meila Wells Chao Chiang Chan Chiang Oakley Bruce
1828 1841 1847 1850 1853 1871 1873 1874 1874 1877 1888 1888 1888 1893a 1893b 1894 1906a, 1906b 1910 1912 1920 1924 1926 1927 1931 1931 1935 1936 1941 1941 1947 1949 1951 1951 1953 1954
vocabulary & phrase chrestomathy textbook dialogue vocabulary religious text Bible grammar religious text textbook textbook Sentence grammar annotation annotation reader textbook grammar grammar textbook textbook textbook textbook Bible textbook textbook textbook textbook dialogue primer textbook textbook textbook textbook textbook
Macau Macau Hong Kong Guangzhou Guangzhou Guangzhou Shanghai London Guangzhou Hong Kong New York Shanghai Hong Kong Guangzhou Guangzhou Hong Kong Guangzhou Hanoi Hong Kong Hong Kong Hong Kong Hong Kong Guangzhou Hong Kong Hong Kong New York Guangzhou Hong Kong Hong Kong New York Singapore Hong Kong Singapore Kuala Lumpur Kuala Lumpur
(continued) 103. Two volumes under the title Cantonese Romanized were combined together.
60 Cantonese give and Double-Object Construction
Table 5. (continued) Author
Year
Nature
Place of publication
Whitaker Whitaker Wu Yuan Lau104 Chapman Gospel of Luke
1954 1959 1960 1960 1972 1973 1997
sentence drill textbook textbook textbook conversation Bible
London London Hong Kong Hong Kong Hong Kong Hong Kong Hong Kong
These materials have different genres: Language instructions, narrative texts and translation of foreign works such as the Bible. These materials with diversified genres provide a balanced coverage of contents for studying the history of Cantonese. In terms of place of publication, eighteen were published in Hong Kong, nine in Guangzhou, two in Macau and thirteen outside of China such as London, Kuala Lumpur, Hanoi, Singapore and New York. 104 4.3
Functions of GIVE in pre-modern Cantonese
Given the importance and usefulness of the extant pre-modern dialect materials, it is hoped that we can, with concrete textual evidence, trace the sequence of the grammaticalization process (see Chapter 3). More important, we also need to examine the mechanism and conditions of the syntactic development. The preliminary study on the grammaticalization process of [pei] 畀 carried out by Chin and Tsou (2005) was based on 12 pre-modern Cantonese materials compiled between 1850 and 1976. The basic finding in their study is that the functions of [pei] 畀 in pre-modern Cantonese were not totally identical to what we find in modern Cantonese. For example, the IO marker was not [pei] 畀 but [kʷͻ] 過 before 1940105 while the passive marker was either [pei] 畀 or [pei] 被, with the latter dominant prior to the first quarter of the 20th century. [Pei] 畀 was also used to introduce instruments, as the following sentence shows.
104. Two series, Elementary and Intermediate, each with two volumes, were combined together.
105. A study on the development of the IO marker from [kʷͻ] 過 to [pei] 畀 in Cantonese can also be found in Takashima and Yue (2000).
Chapter 4. Diachronic development of give and its functions in Cantonese 61
(57) 我 見 佢 畀 棍 打一 箇細 呅 仔106 ⊆ngo kin⊃ ⊆k’ü ⊂pi kwan⊃ ⊂ta yat ko⊃ sai⊃ man ⊂chai ⊆ ⊃ I see s/he give stick beat one cl child 106107 ‘I saw him beating a boy with a stick, your Honor’
Chin and Tsou (2005) however did not provide any explanations for the relationship between this function of [pei] 畀 and the double-object verb [pei] 畀 (cf. Yiu (2010)). Moreover, the amount of early dialectal materials examined was not large enough. In view of these, we re-investigated the various functions of [pei] 畀 with a larger amount of language data comprising different types of materials such as translated texts, grammar books, annotated texts. 4.3.1 As a double-object verb As a basic and core verb in the family of double-object verbs, [pei] 畀 was used frequently as a main verb in the double-object constructions in the corpus. Table 6 lists the frequency counts of the double-object verbs found in each text.108 Items preceded by an asterisk are the double-object verbs with the highest frequency in the text.
106. This example is taken from Stedman and Lee (1888, p. 137). All the sentences taken from the early materials are transcribed according to the ones recorded in the work. For those works that were written in Chinese characters exclusively, no phonetic transcriptions will be provided. 107. Unless stated otherwise, both phonetic transcriptions and translations of examples follow the ones used in the pre-modern materials. 108. We include all instances in which a double-object verb was used as a main verb. Thus relative clauses, sentences with elided DOs and IOs are also included. For example, in Dennys’ (1874), [pei] 畀 occurs once in a relative clause: [ngo pi ni ko pun shü hai pin-ch’ü] 我俾你嗰本書喺邊 處 ‘Where is the book I gave you?’ (p. 68). There are also some sentences without an IO. For example, in Bruce’s (1954), [nei bpei gkei-dtoh ne] 你俾幾多呢? ‘How much do you offer?’ (p. 76).
62 Cantonese give and Double-Object Construction
Table 6. Double-object verbs in pre-modern Cantonese dialect materials109 Author 1. Morrison (1828)
2. Bridgman (1841)
3. Devan (1847) 4. Hobson (1850)
5. Bonney (1853)
6. Bunyan (1871)
Frequency of Double-object verbs
*畀: 7 Others:109 *[kau55] 交 (7) ‘to handover’ [kou33 sou33] 告訴 (1) ‘to tell’ [wa22] 話 (1) ‘to tell’ [tsε33] 借 (1) ‘to lend’ [mai22] 賣 (1) ‘to sell’ [ji11] 移 (1) ‘to transfer’ *畀: 29 Others: [tɐi22] 遞 (10) ‘to pass’ [kau55] 交 (3) ‘to handover’ [mai22] 賣 (2) ‘to sell’ [fʊk5] 覆 (1) ‘to reply’ [jœŋ22] 讓 (1) ‘to give’
*畀: 2 Others: Nil
*畀: 5 Others: [nap1] 納 (1) ‘to pay (tax)’ [sʊŋ33] 送 (1) ‘to give’
*畀: 31 Others: [sʊŋ33] 送 (3) ‘to give’ [kau55] 交 (2) ‘to handover’ [wan11] 還 (1) ‘to return’ [mai22] 賣 (1) ‘to sell’ *畀: 41 Others: [kͻŋ35] 講 (30) ‘to tell’ [ts’i33] 賜 (10) ‘to bestow’ [wa22] 話 (6) ‘to tell’ [pou33] 報 (2) ‘to report’ [tɐi22] 遞 (2) ‘to pass’ [kau55] 交 (2) ‘to handover’ [pɐn35] 稟 (1) ‘to report’ [ts’ιŋ11] 呈 (1) ‘to submit’
109. The verbs are listed in descending order of frequency.
Chapter 4. Diachronic development of give and its functions in Cantonese 63
Table 6. (continued) Author
7. Gospel of Luke (1873)
8. Dennys (1874)
Frequency of Double-object verbs [fat3] 發 (1) ‘to send out’ [kou33] 告 (1) ‘to report’ [kai35] 解 (1) ‘to explain’ [mai22] 賣 (1) ‘to sell’ [p’ui11] 賠 (1) ‘to compensate’ [jœŋ22] 讓 (1) ‘to give’ [sœŋ35 ts’i33] 賞賜 (1) ‘to bestow’ [si55] 施 (1) ‘to give in charity’ [sʊŋ33] 送 (1) ‘to give’
畀: 39 Others: *[wa22] 話 (55) ‘to tell’ [kau55] 交 (9) ‘to handover’ [ts’i33] 賜 (7) ‘to bestow’ [pou33] 報 (4) ‘to report’ [fun55] 分 (4) ‘to share’ [nap1] 納 (4) ‘to pay (tax)’ [kau55 t’ͻk3] 交托 (3) ‘to entrust’ [tsε33] 借 (3) ‘to lend’ [tɐi22] 遞 (2) ‘to pass’ [kͻŋ35] 講 (2) ‘to tell’ [ts’yn11] 傳 (1) ‘to pass’ [ts’yn11 sɐu22] 傳授 (1) ‘to teach’ [fat3] 發 (1) ‘to send out’ [kou33] 告 (1) ‘to report’ [kͻŋ33 lͻk1] 降落 (1) ‘to send down’ [kai35] 解 (1) ‘to explain’ [mai22] 賣 (1) ‘to sell’ [jœŋ22] 讓 (1) ‘to give’ [si55 sε35] 施捨 (1) ‘to give in charity’ [hin33] 獻 (1) ‘to present’ *畀: 45 Others: [mai22] 賣 (3) ‘to sell’ [tsε33] 借 (2) ‘to lend’ [sʊŋ33] 送 (2) ‘to give’ [tsau35] 找 (2) ‘to give change’ [wan11] 還 (1) ‘to return’ [kei33] 寄 (1) ‘to send’
(continued)
64 Cantonese give and Double-Object Construction
Table 6. (continued) Author 9. Harper (1874)
10. Anonymous (1877)
11. Stedman & Lee (1888)
12. Fulton (1888)
13. Ball (1888)
14. Mai (1893a)
Frequency of Double-object verbs
*畀: 2 Others: [kͻŋ35] 講 (1) ‘to tell’
*畀: 4 Others: [tsε33] 借 (1) ‘to lend’
*畀: 33 Others: [mai22] 賣 (4) ‘to sell’ [fɐn55] 分 (1) ‘to share’ [kau55] 交 (1) ‘to handover’ [tsyn35] 轉 (1) ‘to transfer’ [tsou55] 租 (1) ‘to rent out’
*畀: 12 Others: [kau55] 交 (6) ‘to handover’ [tsε33] 借 (3) ‘to lend’ [kei33] 寄 (2) ‘to send’ [tsau35] 找 (2) ‘to give change’ [kau55 t’ͻk3] 交托 (1) ‘to entrust’ [tsi55] 支 (1) ‘to give (salary)’
*畀: 16 Others: [tɐi33] 遞 (1) ‘to pass’ [tιŋ35] 頂 (1) ‘to transfer (the ownership)’ [kau55] 交 (1) ‘to handover’ [tsε33] 借 (1) ‘to lend’
畀: 28 Others: *[ts’i33] 賜 (72) ‘to bestow’ [fʊŋ55 ts’i33] 封賜 (13) ‘to enfeoff and bestow’ [sʊŋ33] 送 (12) ‘to give’ [tsɐŋ22 sʊŋ33] 贈送 (12) ‘to give’ [kͻŋ33 ts’i33] 降賜 (11) ‘to bestow’ [kʷɐi55 ts’i33] 賜歸 (7) ‘to bestow’ [tsɐŋ22 ts’i33] 贈賜 (6) ‘to bestow’ [kιŋ33 tsɐu35] 敬(酒) (5) ‘to toast’ [tsɐŋ22 ts’i33] 增賜 (5) ‘to bestow’ [kʷɐi55] 歸 (4) ‘to return’
Chapter 4. Diachronic development of give and its functions in Cantonese 65
Table 6. (continued) Author
15. Mai (1893b)
Frequency of Double-object verbs
[kau55] 交 (3) ‘to pass’ [kιŋ33 hin33] 敬獻 (3) ‘to present respectfully’ [hin33] 獻 (3) ‘to present’ [ts’i33 fʊk5] 賜福 (2) ‘to bless’ [ts’i33 sɐu22] 賜授 (2) ‘to bestow’ [ka55 ts’i33] 加賜 (2) ‘to bestow’ [kͻŋ33] 降 (2) ‘to send down’ [sœŋ35 ts’i33] 賞賜 (2) ‘to bestow’ [wɐi11 ts’yn11] 遺傳 (2) ‘to pass down’ [jy22 ts’i33] 御賜 (2) ‘to bestow’ [ts’yn11 lͻk1] 傳落 (1) ‘to pass down’ [ts’i33 fʊŋ55] 賜封 (1) ‘to bestow and enfeoff ’ [ts’i33 lap1] 賜立 (1) ‘to bestow’ [ts’i33 sɐu22] 賜受 (1) ‘to bestow’ [ts’i33 tsɐŋ22] 賜贈 (1) ‘to bestow’ [tap3 tsε22] 答謝 (1) ‘to thank’ [fɐn55] 分 (1) ‘to share’ [fɐn55 ts’i33] 分賜 (1) ‘to divide and bestow’ [fɐn55 p’ai33] 分派 (1) ‘to distribute’ [fʊŋ55] 封 (1) ‘to enfeoff ’ [ka33 p’ui33] 嫁配 (1) ‘to marry off one’s daughter’ [kͻŋ33 ha22] 降下 (1) ‘to pass down’ [kau33] 教 (1) ‘to teach’ [pan55 ts’i33] 頒賜 (1) ‘to bestow’ [jœŋ22] 讓 (1) ‘to give’ [hœy35 p’ui33] 許配 (1) ‘to marry off one’s daughter’
畀: 1 Others: *[ts’i33] 賜 (89) ‘to bestow’ [fʊŋ55 ts’i33] 封賜 (18) ‘to enfeoff and bestow’ [jœŋ22] 讓 (8) ‘to give’ [sœŋ35 ts’i33] 賞賜 (5) ‘to bestow’ [ts’yn11] 傳 (4) ‘to pass’ [ts’i33 kʷɐi55] 賜歸 (4) ‘to bestow’ [kͻŋ33 ts’i33] 降賜 (4) ‘to send down and bestow’ [kau55] 交 (4) ‘to handover’ [hin33 sœŋ13] 獻上 (3) ‘to present’ [ts’yn11 kau55] 傳交 (2) ‘to pass’
(continued)
66 Cantonese give and Double-Object Construction
Table 6. (continued) Author
16. Ball (1894)
Frequency of Double-object verbs
[ts’yn11 lͻk1] 傳落 (2) ‘to pass down’ [ts’i33 sɐu22] 賜授 (2) ‘to bestow’ [fɐn55 ts’i33] 分賜 (2) ‘to divide and bestow’ [pan55 kou33] 頒誥 (1) ‘to promulgate’ [p’in33 kou33] 遍告 (1) ‘to spread’ [ts’yn11 tou33] 傳到 (1) ‘to pass’ [ts’yn11 kʷɐi55] 傳歸 (1) ‘to pass’ [ts’i33 ts’ɵt5] 賜出 (1) ‘to bestow’ [ts’i33 kau55] 賜交 (1) ‘to bestow’ [ts’i33 fu22 kʷɐi55] 賜附歸 (1) ‘to bestow’ [fʊŋ22] 奉 (1) ‘to give with respect’ [fʊŋ55 ts’i33 fɐn55] 封賜分 (1) ‘to bestow’ [fʊŋ55 ts’i33 kʷɐi55] 封賜歸 (1) ‘to bestow’ [kʊŋ55] 供 (1) ‘to supply’ [wɐi22 ts’i33] 惠賜 (1) ‘to bestow’ [tsin33] 薦 (1) ‘to recommend’ [kͻŋ33] 降 (1) ‘to send down’ [tsœŋ35 sœŋ35] 獎賞 (1) ‘to award’ [kiu35 ts’i33] 繳賜 (1) ‘to bestow’ [kau55 t’ͻk3] 交託 (1) ‘to entrust’ [lɐu11 ts’yn11] 流傳 (1) ‘to circulate’ [jœŋ22 kʷɐi55] 讓歸 (1) ‘to return’ [jœŋ22 kau55] 讓交 (1) ‘to handover’ [sιŋ55 ts’i33] 陞賜 (1) ‘to bestow’ [sʊŋ33] 送 (1) ‘to give’ [wɐi35] 委 (1) ‘to entrust’ [wɐi35 t’ͻk3] 委託 (1) ‘to entrust’ [fat3] 發 (1) ‘to give (a command)’ [wɐi11 ts’yn11] 遺傳 (1) ‘to pass down’
*畀: 12 Others: [ts’i33] 賜 (3) ‘to bestow’ [fɐn55] 分 (2) ‘to share’ [kau55] 交 (1) ‘to handover’ [si55] 施 (1) ‘to give in charity’ [tsε33] 借 (1) ‘to lend’ [mai22] 賣 (1) ‘to sell’
Chapter 4. Diachronic development of give and its functions in Cantonese 67
Table 6. (continued) Author 17. Wisner (1906a, 1906b)
18. LeBlanc (1910)
19. Ball (1912)
20. Cowles (1920) 21. Ball (1924)
22. Caysac (1926)
Frequency of Double-object verbs
*畀: 26 Others: [kau55] 交 (4) ‘to handover’ [ts’yn11 jim13] 傳染 (1) ‘to infect’ [tɐi22] 遞 (1) ‘to pass’ [wan11] 還 (1) ‘to return’ [kei33] 寄 (1) ‘to send’ [kau55 lͻk1] 交落 (1) ‘to leave to’ [kau55 t’ͻk3] 交托 (1) ‘to entrust’ [kɵy35 tsin33] 舉薦 (1) ‘to recommend’ [mai22] 賣 (1) ‘to sell’ [sœŋ35 ts’i33] 賞賜 (1) ‘to bestow’ *畀: 17 Others: [kau55 tai33] 交帶 (1) ‘to entrust’ [mai22] 賣 (1) ‘to sell’
*畀: 32 Others: [kau55] 交 (5) ‘to handover’ [tsε33] 借 (5) ‘to lend’ [mai22] 賣 (2) ‘to sell’ [pei35 kau55] 俾交 (1) ‘to handover’ [kau55 tai33] 交帶 (1) ‘to entrust’ [jœŋ22] 讓 (1) ‘to give’
*畀: 39 Others: nil
*畀: 20 Others: [tɐi22] 遞 (1) ‘to pass’ [tιŋ35] 頂 (1) ‘to transfer (the ownership)’ [kau55] 交 (1) ‘to handover’ [tsε33] 借 (1) ‘to lend’ *畀: 40 Others: [mai22] 賣 (9) ‘to sell’ [tsε33] 借 (4) ‘to lend’ [kau55] 交 (3) ‘to handover’ [sʊŋ33] 送 (2) ‘to give’ [fɐn55] 分 (1) ‘to share’
(continued)
68 Cantonese give and Double-Object Construction
Table 6. (continued) Author
23. Wisner (1927)
24. Gospel of Luke (1931)
Frequency of Double-object verbs
[ta35 tin22 wa22–35] 打(電話) (1) ‘to give a call’ [kei33] 寄 (1) ‘to send’ [ka33] 嫁 (1) ‘to marry off one’s daughter’ [kau55 t’ͻk3] 交托 (1) ‘to entrust’ [sy55] 輸 (1) ‘to lose’ [hin33] 獻 (1) ‘to present’
*畀: 33 Others: [kau55] 交 (2) ‘to handover’ [kau55 t’ͻk3] 交托 (2) ‘to entrust’ [p’ai33] 派 (2) ‘to distribute’ [ts’yn11 jim13] 傳染 (1) ‘to infect’ [tɐi22] 遞 (1) ‘to pass’ [wan11] 還 (1) ‘to return’ [kei33] 寄 (1) ‘to send’ [kɵy35 tsin33] 舉薦 (1) ‘to recommend’ [mai22] 賣 (1) ‘to sell’ [sœŋ35 ts’i33] 賞賜 (1) ‘to bestow’
畀: 31 Others: *[wa22] 話 (39) ‘to tell’ [kau55] 交 (13) ‘to handover’ [ts’i33] 賜 (10) ‘to bestow’ [fɐn55] 分 (5) ‘to share’ [pou33] 報 (4) ‘to report’ [tsε33] 借 (4) ‘to lend’ [kͻŋ35] 講 (3) ‘to tell’ [ts’yn11] 傳 (2) ‘to pass’ [kʊŋ55] 供 (2) ‘to supply’ [ts’yn11 sɐu22] 傳授 (1) ‘to teach’ [tɐi22] 遞 (1) ‘to pass’ [fʊŋ55] 封 (1) ‘to enfeoff ’ [wan11] 還 (1) ‘to return’ [kau55 t’ͻk3] 交託 (1) ‘to entrust’ [nap1] 納 (1) ‘to pay (tax)’ [jœŋ22] 讓 (1) ‘to give’ [si55 sε35] 施捨 (1) ‘to give in charity’ [wɐi33] 餵 (1) ‘to feed’ [hin33] 獻 (1) ‘to present’ [tsi35 si22] 指示 (1) ‘to give instructions’
Chapter 4. Diachronic development of give and its functions in Cantonese 69
Table 6. (continued) Author 25. Wells (1931)
26. Brouner & Fung (1935)
27. Hoh & Belt (1936)
28. O’Meila (1941)
29. Wells (1941)
Frequency of Double-object verbs
*畀: 25 Others: [jœŋ22] 讓 (6) ‘to give’ [kei33] 寄 (3) ‘to send’ [tιŋ35] 頂 (2) ‘to transfer (the ownership)’ [tsε33] 借 (2) ‘to lend’ [mai22] 賣 (2) ‘to sell’ [sʊŋ33] 送 (2) ‘to give’ [fat3] 發 (1) ‘to send out’ [fat3 k’ɐp5] 發給 (1) ‘to send out’ [kʷͻ33] 過 (1) ‘to transfer’ [kei33 ts’ɵt5] 寄出 (1) ‘to send out’ [kau55] 交 (1) ‘to handover’ [kau55 t’ͻk3] 交托 (1) ‘to entrust’ [tsyn35 mai22] 轉賣 (1) ‘to resell’ *畀: 19 Others: [tsau35] 找 (1) ‘to give change’ *畀: 3 Others: [tɐi22] 遞 (1) ‘to pass’ [wan11] 還 (1) ‘to return’ [kei33] 寄 (1) ‘to send’
*畀: 136 Others: [sʊŋ33] 送 (35) ‘to give’ [kei33] 寄 (6) ‘to send’ [tsε33] 借 (2) ‘to lend’ [mai22] 賣 (2) ‘to sell’ [tsau35] 找 (2) ‘to give change’ [wui22] 匯 (1) ‘to remit’ [kai33 siu22] 介紹 (1) ‘to introduce’ [kau55] 交 (1) ‘to handover’ [sœŋ35 ts’i33] 賞賜 (1) ‘to bestow’
*畀: 43 Others: [sʊŋ33] 送 (5) ‘to give’ [tsε33] 借 (4) ‘to lend’ [mai22] 賣 (3) ‘to sell’ [fɐn55] 分 (1) ‘to share’
(continued)
70 Cantonese give and Double-Object Construction
Table 6. (continued) Author
30. Chao (1947)
31. Chiang (1949)
32. Chan (1955)
33. Chiang (1951)
34. Oakley (1953)
35. Bruce (1954)
36. Whitaker (1954)
Frequency of Double-object verbs
[fɐn55 tsou55] 分租 (1) ‘to sublease’ [sy55] 輸 (1) ‘to lose’ [tsyn35 tsou55] 轉租 (1) ‘to sublease’
*畀: 2 Others: [tsε33] 借 (1) ‘to lend’
*畀: 5 Others: [sʊŋ33] 送 (1) ‘to give’ [tsou55] 租 (1) ‘to rent out’ *畀: 32 Others: [kei33] 寄 (2) ‘to send’ [tsε33] 借 (1) ‘to lend’ [mai22] 賣 (1) ‘to sell’ [sʊŋ33] 送 (1) ‘to give’
*畀: 8 Others: [sʊŋ33] 送 (2) ‘to give’
*畀: 22 Others: [sʊŋ33] 送 (5) ‘to give’ [mai22] 賣 (1) ‘to sell’ [tsou55] 租 (1) ‘to rent out’
*畀: 42 Others: [pou33] 報 (3) ‘to report’ [tsε33] 借 (2) ‘to lend’ [mai22] 賣 (2) ‘to sell’ [ta35 tin22 wa22–35] 打(電話) (1) ‘to give a call’ [wan11] 還 (1) ‘to return’ [kai33 siu22] 介紹 (1) ‘to introduce’ [p’ai33] 派 (1) ‘to distribute’ [sʊŋ33] 送 (1) ‘to give’
畀: 2 Others: *[tɐi22] 遞 (4) ‘to pass’ [kau55] 交 (3) ‘to handover’ [kai33 siu22] 介紹 (1) ‘to introduce’
Chapter 4. Diachronic development of give and its functions in Cantonese 71
Table 6. (continued) Author 37. Whitaker (1959)
38. Wu (1960)
39. Yuan (1960)
40. Lau (1972a, 1972b)
41. Chapman (1973)*畀: 6
Frequency of Double-object verbs
畀: 8 Others: *[ta35 tin22 wa22–35] 打(電話) (10) ‘to give a call’ [tsε33] 借 (2) ‘to lend’ [fɐn55] 分 (1) ‘to share’ [sʊŋ33] 送 (1) ‘to give’ *畀: 4 Others: [tsau35] 找 (3) ‘to give change’ [ta35 tin22 wa22–35] 打(電話) (1) ‘to give a call’ [tsε33] 借 (1) ‘to lend’ [kai33 siu22] 介紹 (1) ‘to introduce’ *畀: 7 Others: [ta35tin22 wa22–35] 打(電話) (1) ‘to give a call’ [kai33 siu22] 介紹 (1) ‘to introduce’
*畀: 68 Others: [ta35 tin22 wa22–35] 打(電話) (34) ‘to give a call’ [sʊŋ33] 送 (34) ‘to give’ [tsε33] 借 (27) ‘to lend’ [mai22] 賣 (10) ‘to sell’ [fʊk1] 覆 (6) ‘to reply’ [tsou55] 租 (5) ‘to rent out’ [kei33] 寄 (3) ‘to send’ [wan11] 還 (1) ‘to return’ [p’ui11] 賠 (1) ‘to compensate’ [p’ui11 sœŋ11] 賠償 (1) ‘to compensate’ *畀: 6 Others: [t’ɵy55 tsin33] 推薦 (2) ‘to recommend’ [ta35 tin22 wa22–35] 打(電話) (1) ‘to give a call’ [wui11 sɵn33] 回(信) (1) ‘to reply (a letter)’ [tsε33] 借 (1) ‘to lend’ [kai33 siu22] 介紹 (1) ‘to introduce’ [mai22] 賣 (1) ‘to sell’ [sʊŋ33] 送 (1) ‘to give’
(continued)
72 Cantonese give and Double-Object Construction
Table 6. (continued) Author 42. Gospel of Luke (1997)
Frequency of Double-object verbs
畀: 28 Others: *[wa22] 話 (50) ‘to tell’ [kau55] 交 (12) ‘to handover’ [ts’i33] 賜 (10) ‘to bestow’ [fɐn55] 分 (6) ‘to share’ [tsε33] 借 (3) ‘to lend’ [ts’yn11] 傳 (2) ‘to pass’ [tɐi22] 遞 (2) ‘to pass’ [fʊŋ22 hin33] 奉獻 (2) ‘to present’ [si55] 施 (2) ‘to give in charity’ [pou33] 報 (1) ‘to report’ [fu22 t’ͻk3] 付託 (1) ‘to entrust’ [wan11] 還 (1) ‘to return’ [kͻŋ35] 講 (1) ‘to tell’ [kͻŋ33] 降 (1) ‘to send down’ [jœŋ22] 讓 (1) ‘to give’ [hin33] 獻 (1) ‘to present’ [tsyn35 tsou55] 轉租 (1) ‘to sublease’ [tsou55] 租 (1) ‘to rent out’
[Pei] 畀 appears in all the 44 works. It is the most frequent double-object verb in 35 works. The exceptions are the three translations of the Gospel of Luke, the two annotations of Chinese classics (Mai, 1893a, 1893b): The Book of Odes (Shijing 詩 經) and the Book of Documents (Shangshu 尚書 or Shujing 書經) and the two texts by Whitaker (1954, 1959). In the Gospel, there are many quotes or sayings of Jesus. For example:
(58) 我 實在 話 你 知110 I indeed tell you know ‘Of a truth I say unto you’111 110111
Since Mai Shizhi’s two works are annotation of Chinese classics, some of the words used by Mai were literary or classical. Besides, the stories in the Book of Documents were related to the emperors of ancient China. There was a unique double-object 110. The English translation of the examples quoted from the Gospel of Luke is based on Carroll and Prickett (1997). 111. See the Gospel of Luke (1873, p. 36). This sentence appears 12 times in this edition. There are 30 sentences which do not have the adverb [sɐt tsͻi] 實在 ‘of a truth’.
Chapter 4. Diachronic development of give and its functions in Cantonese 73
verb [ts’i] 賜 ‘to bestow’ describing specifically the action of giving by the emperor to the vassals or the people in ancient China. Therefore, [ts’i] 賜 ‘to bestow’ was used more frequently than [pei] 畀 ‘to give’ in this text. As for Whitaker’s works, [pei] 畀 ‘to give’ did not really occur infrequently when compared with the highest frequency double-object verbs. In Whitaker (1954), [pei] 畀 ‘to give’ occurred twice while the highest frequency double-object verb [tɐi] 遞 ‘to pass’ occurred only four times. In Whitaker (1959), [pei] 畀 ‘to give’ occurred eight times while [ta tin wa] 打電話 ‘to give a call’ occurred ten times. 4.3.2 As a causative verb [Pei] 畀 was frequently used to express the permissive meaning in the pre-modern period.
(59) 我 俾 你 睇, 等 你 抄 花色 去 呵 ⊆ngó ⊂pí ⊆ní ⊂t’ai ⊂tang ⊆ní , ch’áu fá shik h’ü⊃ ó ⊆ ⊂ ⊂ ⊃ I give you look wait you copy style go sfp ‘I’ll let you see them, so that you may write down the different colors’ (Bridgman, 1841, p. 239) (60) 老爺 唔 俾 人 讀 個 的 書 master not give people read that cl book ‘The officers will not let the people read those books’ (Bonney, 1853, p. 88) (61) 佢 唔 肯 俾 佢 過 去 k’ü m hăng pi k’ü kwo hü s/he not willing give s/he go through ‘He would not let him pass’ (62) 俾 我 睇 你 嗰 部 書 péi ngŏh t’ái nĕi kóh pô shue give I read you that cl book ‘Let me look at your book’
(63) 但 係 父 親 唔 俾 我 去 tâan hâi fôo ts’an m̄ péi ngŏh hùi but father not give I go ‘But my father will not permit me to go’
(Dennys, 1874, p. 67)
(Wisner, 1906a, p. 4)
(O’ Melia, 1941, p. 150)
(64) 佢 買 唨 燕梳 先, 我 至 俾 佢 揸 車 kui5 maai5 joh2 yin3soh1° sin1, ngoh5 ji3 bei2 kui5 ja1 che1° then give s/he drive car s/he buy asp insurance first, I My translation: ‘I will let him/her drive only when s/he had bought the insurance’ (Lau, 1972, p. 669)
In the following, we will examine the other three functions of [pei] 畀 which had significant differences from modern Cantonese.
74 Cantonese give and Double-Object Construction
4.3.3 As an IO marker Three IO markers are found in the corpus: [pei] 畀 ‘to give’, [kʷͻ] 過 ‘to cross’ and null marker. Table 7 gives their frequency distribution.112 Table 7. Frequency distribution of the three IO markers in pre-modern Cantonese materials
113
Work 1. Morrison (1828) 2. Bridgman (1841) 3. Devans (1847) 4. Hobson (1850) 5. Bonney (1853) 6. Bunyan (1871) 7. Gospel of Luke (1873) 8. Dennys (1874) 9. Harper (1874) 10. Anonymous (1877) 11. Ball (1888) 12. Fulton (1888) 13. Stedman & Lee (1888) 14. Mai (1893a)113 15. Mai (1893b) 16. Ball (1894) 17. Wisner (1906a, 1906b) 18. LeBlanc (1910) 19. Ball (1912) 20. Cowles (1920) 21. Ball (1924) 22. Caysac (1926) 23. Wisner (1927) 24. Gospel of Luke (1931)
過
8 23 0 2 9 83 71 21 3 3 8 18 11 26 2 14 21 7 13 6 12 0 23 91
Frequency
畀
2 2 0 1 0 0 5 0 0 0 0 1 2 54 52 1 3 1 9 0 0 35 3 0
null 2 7 1 2 5 17 38 19 0 1 11 3 6 4 10 3 8 1 15 13 7 13 9 11
112. We do not only limit our analysis to main clauses, but also include relative clauses or sentences with elided DOs. Some examples include [kui5 je3 bei2 ngoh5 goh2 bo6 siu2 suet3 sap6 fan1 ho2 tai2] 佢借俾我嗰部小說十分好睇 ‘the novel he lent to me is extremely good’ and 送過我 睇 ‘presented to me to look at’. These two examples are taken from Lau (1972, p. 264) and Bonney (1853, p. 61) respectively. The former is a relative clause while the latter is a double-object sentence with the DO elided.
113. Besides [pei] 畀 and [kʷͻ] 過, the two texts by Mai (1893a, 1893b) also have [pei kʷͻ] 畀過 as the IO marker. This will be discussed later.
Chapter 4. Diachronic development of give and its functions in Cantonese 75
Table 7. (continued) Work 25. Wells (1931) 26. Brouner & Fung (1935) 27. Hoh & Belt (1936) 28. O’Meila (1941) 29. Wells (1941) 30. Chao (1947) 31. Chiang (1949) 32. Chan (1951) 33. Chiang (1951) 34. Oakley (1953) 35. Bruce (1954) 36. Whitaker (1954) 37. Whitaker (1959) 38. Wu (1960) 39. Yuan (1960) 40. Lau (1972) 41. Chapman (1973) 42. Gospel of Luke (1997) Total
過
24 5 0 13 16 0 0 0 0 1 7 1 2 0 0 23 0 9 576
Frequency
畀
8 0 2 40 7 1 2 5 2 8 15 7 13 5 1 138 9 45 479
null 9 8 2 67 13 2 4 18 4 7 21 1 7 4 6 16 3 36 434
A number of phenomena related to the IO marker are observed: a. From the above frequency distribution, it is noted that in the 19th century (items 1 to 16), it was [kʷͻ] 過 that mainly functioned as the IO marker. The two exceptions are the two texts by Mai (items 14 and 15). The use of the IO marker [kʷͻ] 過 continued until the 1940s (except Caysac (item 22)).114 After the 1940s, [pei] 畀 replaced [kʷͻ] 過 to become the IO marker except in Wells (item 29).115 Materials published after the 1940s scarcely used [kʷͻ] 過 as the IO marker which can be considered a residual usage.
114. J. Yang (2006), on the basis of Cantonese Made Easy (1888), 粵語全書 (Handbook on Cantonese) (1905) and 粵音指南 (Instructions on Cantonese Pronunciations) published between 1882 and 1910, also observed the use of [kʷͻ] 過 as the IO marker in early Yue dialects.
115. One may argue that the work by Wells (1941) basically followed his earlier work (Wells (1931)) where [kʷͻ] 過 was dominant. However, it is observed that the percentage of [kʷͻ] 過 dropped from 75% to 69.6% while that of the IO marker [pei] 畀 increased from 33.3% to 43.8%.
76 Cantonese give and Double-Object Construction
b. Among these 44 dialectal materials, three of them are translations of the Gospel of Luke 路加福音 published in different periods: 1873, 1931 and 1997. It is thus meaningful for us to compare these three editions since they contain identical contents. By comparing those sentences originated from the same source in the Gospel, we can clearly observe the development of the IO marker during the 120 years. Table 8 lists six examples.116 Table 8. IO markers in the three editions of the Gospel of Luke Source
1873
1931
Chapter 10, verse 35: (he) took out two pence, and gave them to the host
擰出二錢嚟, 交 過店主
Chapter 11, verse 11: will he (for a fish) give him a serpent?
倒轉畀蠍過佢呢
反畀蠍過佢
就畀蠍佢呢
你去話過個隻狐 狸知
你哋去話過個隻 狐狸知
你哋去話畀嗰隻 狐狸知
亞爸, 𢬿我所應 得嘅家業分過我
請你𢬿我應得嘅家 父親,請你將我應 業分過我喇 得嘅產業分畀我
Chapter 11, verse 5: Friend, lend 請你借三個餅 me three loaves 過我
Chapter 13, verse 32: Go ye, and tell that fox Chapter 14, verse 9: Give this man place Chapter 15, verse 12: Father, give me the portion of goods that falleth to me
讓位過呢個人喇
擰兩個銀錢出來, 交過店主話
1997
第二日拎兩個銀 幣, 交畀客棧嘅 主人
請你畀三個餅過我 請畀三個餅我
讓位過呢個人喇
請讓座畀呢位啦
The above data show that the replacement of the IO marker [kʷͻ] 過 by [pei] 畀 took place after the 1930s. It is also noted that in the 1997 edition, there is no IO marker when the main verb is [pei] 畀 and this null IO marker usage will be discussed later. c. In Mai Shizhi’s two texts (1893a, 1893b), the two IO markers [pei] 畀 and [kʷͻ] 過 co-occurred (i.e., [pei kʷͻ] 畀過) within the same sentence, as shown in (65) and (66). 116. The use of [kʷͻ] 過 as an IO marker can be found in some old Cantonese movies produced in Hong Kong in mid-20th century. Two examples can be found in the movies 英雄難過美人關 (The Hero Becomes a Prisoner of Love, 1950) and 標準丈夫 (Standard Husband, 1965) respectively: 你至少都畀個證據過我呀! (My translation: You have to give the evidence to me!) and 噉我至多畀晒啲旅行證件過你 (My translation: Then I give all my travel documents to you). These examples were taken from the second phase of The Corpus of Mid-20th Century Hong Kong Cantonese [二十世紀中期香港粵語語料庫 https://hkcc.eduhk.hk/] (Accessed on 1 August 2021). For the details of the corpus, see Chin (2013, 2019, 2021).
Chapter 4. Diachronic development of give and its functions in Cantonese 77
(65) 佢 贈送 呢 一 個 木瓜 俾 過 我 s/he give this one cl papaya give pass I My translation: ‘S/he gave me this papaya’
(66) 就 賜 個 一 條 大 天命 來 俾 過 我地 文 王 then bestow that one cl great mandate come give pass our Wen King My translation: ‘Then (the Heaven) gave a great mandate to our King Wen’
Takashima and Yue (2000) claimed that this hybrid form containing both the old and the new IO markers illustrates the transitional change of the IO marker.117 Such a claim can be supported by our quantitative data on the distribution of the two IO markers. In Table 7, some texts prior to 1893 also used [pei] 畀 as the IO marker though the dominant one was still [kʷͻ] 過. We can claim that by the late 19th century, [pei] 畀 and [kʷͻ] 過 already co-existed in the dialect and [kʷͻ] 過 was giving way to [pei] 畀. It is thus not impossible for us to record such a hybrid form [pei kʷͻ] 畀過 in the dialects, which is “something typical of transitional periods in linguistic change” (Takashima and Yue, 2000, p. 41). d. Table 7 shows that besides [pei] 畀 and [kʷͻ] 過, the IO marker can be absent. In contemporary Hong Kong Cantonese, there are double-object sentences in which the IO marker is not present. For example:
(67) 佢 畀 三 本 書 我 k’ɵy13 pei35 sam55 pun35 sy55 ŋͻ13 s/he give three cl book I ‘S/he gave me three books’
With this kind of sentence, one may argue that, in addition to the V DO 畀 IO pattern, Cantonese has one more double-object pattern with the structure V DO IO (see, for example, S. Tang (1998), D. Liu (2001), M. Zhang (2010), Yeh (2020), and H. Wu (2021)118). There are proposals claiming that this pattern is derived from the V IO DO pattern via transformation through which the DO is fronted. L. Xu and Peyraube (1997), on the other hand, claimed that the V DO IO pattern is derived 117. This hybrid form is mainly found in Mai’s two texts. Other materials only had occasional use of [pei kʷͻ] 畀過. For example, in Wells (1931, p. 44), [ngoh2 tei3 chiu3 ka1ng2 so3 muk4 lai1 kai3, ch’ut4 ti1 pei2 kwoh3 ku3 tung1] 我哋照加五數目嚟計,出的俾過股東, ‘we might reckon a 50% advance, and let shareholders have some’. In some cases, the main verbs are not double-object verbs. For example, in Dennys (1841, p. 64), [ning ti kom to t’ong pi kwo k’ü] 擰啲咁多湯俾過 佢 ‘take a little of that soup and give it to her’.
118. These scholars call the V DO IO sentence as inverted or reverse double-object construction 倒置雙賓句 as they use the term double-object construction to refer to the V IO DO pattern.
78 Cantonese give and Double-Object Construction
from the V DO 畀 IO pattern via the deletion of the IO marker. S. Tang (1998) also argued for the same by means of phonological account. He suggested that a pause can be inserted between the DO and the IO in the V DO IO pattern, which in turn demonstrates that a constituent (i.e., the IO marker) is elided.119 The argument for the omission of the IO marker can be supported by the pre-modern dialect materials. Table 7 shows that double-object sentences with null IO markers are found in all the works except Harper (1874). We repeat the same set of data in Table 9 but this time, we also include the frequency (in parentheses) of those sentences that have [pei] 畀 as the main verb. For example, in Morrison (1828) there are two cases of null IO marker but the verbs are not [pei] 畀 since the frequency in the parentheses is zero. In Bunyan (1871), there are 17 cases of null IO marker and 8 of them occur with the verb [pei] 畀. Table 9. Double-object sentences with null IO markers Author
Year
Frequency of null IO marker
Morrison Bridgman Devans Hobson Bonney Bunyan Gospel of Luke Dennys Harper Anonymous Ball Fulton Stedman & Lee Mai Mai Ball Wisner LeBlanc Ball Cowles
1828 1841 1847 1850 1853 1871 1873 1874 1874 1877 1888 1888 1888 1893a 1893b 1894 1906a, 1906b 1910 1912 1920
2(0) 7(7) 1(1) 2(2) 5(5) 17(8) 38(17) 19(18) 0 1(1) 11(10) 3(3) 6(6) 4(0) 10(1) 3(3) 8(8) 1(1) 15(15) 13(13)
119. However, S. Tang (1998) did not have any phonetic evidence to support his claim. According to my native judgment, the IO marker becomes null only when the main verb is [pei] 畀. Therefore, [ŋͻ13 tɐi22 pun35 sy55 k’ɵy13] 我遞本書佢 ‘I passed a book to him/her’ sounds unnatural or even ungrammatical to me.
Chapter 4. Diachronic development of give and its functions in Cantonese 79
Table 9. (continued) Author
Year
Frequency of null IO marker
Ball Caysac Wisner Gospel of Luke Wells Brouner & Fung Hoh & Belt O’Meila Wells Chao Chiang Chan Chiang Oakley Bruce Whitaker Whitaker Wu Yuan Lau Chapman Gospel of Luke Total
1924 1926 1927 1931 1931 1935 1936 1941 1941 1941 1949 1951 1951 1953 1954 1954 1959 1960 1960 1972 1973 1997
7(6) 13(13) 9(9) 11(5) 9(7) 8(8) 2(2) 67(65) 13(13) 2(2) 4(4) 18(18) 4(4) 7(6) 21(21) 1(1) 7(6) 4(4) 6(6) 16(16) 3(3) 36(22) 414 (360)
Among these 43 works,120 27 have null IO markers when the sentences have [pei] 畀 as the main verb (i.e., both numbers are the same). In another set of 7 works, the majority of the null IO marker were used when the main verb is [pei] 畀. These include Ball (1888, 1924), Dennys (1874), O’Meila (1941), Oakley (1953), Whitaker (1959). As for the remaining 7 works, the null IO markers were used with double-object verbs other than [pei] 畀.121 Altogether, around 87% (i.e., 360 out of 414) of the double-object sentences with null IO markers have [pei] 畀 as the main verb. To further examine the null IO marker usage in the double-object construction, we now focus on the frequency distribution of the three IO markers ([kʷͻ] 過, [pei] 畀 and null) in those sentences with [pei] 畀 as the main verb. 120. Harper (1874) does not have double-object sentences with null IO markers. 121. For example, the three editions of the Gospel of Luke used many null IO markers but only around half of them were used with the verb [pei] 畀.
80 Cantonese give and Double-Object Construction
Table 10. IO markers in double-object sentences with the main verb GIVE Author
Year
Morrison Bridgman Devans Hobson Bonney Bunyan Gospel of Luke Dennys Harper Anonymous Ball Fulton Stedman & Lee Mai Mai Ball Wisner LeBlanc Ball Cowles Ball Caysac Wisner Gospel of Luke Wells Brouner & Fung Hoh & Belt O’Meila Wells Chao Chiang Chan Chiang Oakley Bruce Whitaker Whitaker Wu Yuan Lau Chapman Gospel of Luke Total
1828 1841 1847 1850 1853 1871 1873 1874 1874 1877 1888 1888 1888 1893a 1893b 1894 1906a, 1906b 1910 1912 1920 1924 1926 1927 1931 1931 1935 1936 1941 1941 1947 1949 1951 1951 1953 1954 1954 1959 1960 1960 1972 1973 1997
Null 0 7 1 2 5 8 17 18 0 1 10 3 6 0 1 3 8 1 15 13 6 13 9 5 7 8 2 65 13 2 4 18 4 6 21 1 6 4 6 16 3 22 360
過
3 10 0 1 2 29 21 8 2 1 3 2 4 11 0 2 9 4 8 2 4 0 9 27 9 2 0 2 5 0 0 0 0 0 7 0 1 0 0 20 0 3 211
畀
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 10
Chapter 4. Diachronic development of give and its functions in Cantonese 81
When we examine those works published before the 1940s, we find that the IO marker [kʷͻ] 過 was often present even when the main verb of the double-object sentence was [pei] 畀. Although null IO markers are found in some of the works (except Harper (1874)), the frequencies of having an overt IO marker (i.e., [kʷͻ] 過) are generally higher. After the 1940s, the IO marker shifted from [kʷͻ] 過 to [pei] 畀122 but we find that no materials use [pei] 畀 as the IO marker (except Oakley (1953)) when the main verb is [pei] 畀. This is significantly different from the period in which [kʷͻ] 過 was still the IO marker. Furthermore, we also note that when [kʷͻ] 過 was still the main IO marker, it was used in those sentences with short DOs. (68) 佢 畀 信 過 你 ⊆k’u ⊂pi sun⊃ kwo⊃ ⊆ni s/he give letter pass you ‘He gives a letter to you’
(Fulton, 1888, p. 95)
(69) ... 所以 唔 肯 畀 錢 過 佢 ... so not willing give money pass s/he My translation: ‘… so I was not willing to give money to him/her’ (Anonymous, 1877, p. 95)
(70) 我 畀 個 隻 馬 過 佢 嘅 兄 弟123 ⊆ngo ⊂pi ko⊃ tchek ⊆ma kouo⊃ ⊆k’u ke⊃ ⊂hing tai⊇ I give that cl horse pass s/he de brothers ‘I give that horse to his/her brothers’ 123(LeBlanc, 1910, p. 168)
(71) 我 畀 個 部 書 過 你 個 兄弟124 ⊆ngo ⊂pi ko⊃ po⊇ chu kouo⊃ ⊆ni ko⊃ ⊂hing tai⊇ ⊂ I give that cl book pass you cl brother ‘I give that book to your brother’ 124(LeBlanc, 1910, p. 175)
In modern Hong Kong Cantonese in which [pei] 畀 is the IO marker, these four sentences can be expressed without the IO marker. Based on the above discussion on the distribution of the three types of IO markers (null, [kʷͻ] 過 and [pei] 畀), we claim that the V DO IO pattern in modern Cantonese is a reduced form of the V DO 畀 IO pattern as a result of haplology.125 122. Only Bruce (1954), Whitaker (1959), Lau (1972) and the Gospel of Luke (1997) still used [kʷͻ] 過 as the IO marker. 123. The French translation given by the author is: ‘je donne ce cheval à son frère’. 124. 7The French translation is: ‘je remets ce livre à ton frère’. 125. Matthews (2006) argued that V DO IO is the basic structure of Cantonese. He attributed this to an areal feature spreading from the Thai language. However, the IO marker [hay] in Thai can be used when the main verb is also [hay].
82 Cantonese give and Double-Object Construction
When the main verb is [pei] 畀, the IO marker is elided when the DO is short because the main verb and the IO marker are identical and close to each other.126 We can postulate that in the underlying structure of the double-object construction of Cantonese, there is an IO marker and its omission is correlated with the nature of the main verb as well as the length of the DO: V DO IO-Marker IO → V DO IO /V=IO-Marker and DO is short
4.3.3.1 From [kʷͻ] 過 to [pei] 畀 In this section, we examine the conditioning factors for the development of the IO marker from [kʷͻ] 過 to [pei] 畀. We argue that the syntactic change involved both syntactic and semantic factors.127 The surface structure of the double-object construction is identical to the one of the beneficiary construction: V1 NP1 V2 NP2. In modern Cantonese, V2 in both constructions are [pei] 畀. The crucial difference between the two constructions is that V1 in the double-object construction has the intrinsic semantic feature of [+GIVE]. A. Yue (2004) found that in some pre-modern materials, [pei] 畀 was mainly used as the beneficiary marker.128 In other words, V1 that co-occurred with [pei] 畀 as V2 did not have the [+GIVE] feature (i.e., a non-double-object verb). Yue’s conclusion is based on Ball (1894). Prior to 1894, there were other materials in which both [kʷͻ] 過 and [pei] 畀 were used. The distribution of [kʷͻ] 過 and [pei] 畀 as the IO marker and the beneficiary marker is listed in Table 11. Table 11. Frequency distribution of 過 and 畀 as the IO marker and the beneficiary marker129
Author
Year
Morrison
1828
Bridgman Devans
IO marker
過
畀
1841
23
2
1847
0
0
8
4
Beneficiary marker 過129
畀
嚟5
3
0
2
2
0
126. Although in Table 9, there are cases in which the main verb was not [pei] 畀 but no IO marker was used or vice versa, we can still see that the two patterns (i.e., V DO IO and V DO 畀 IO) have a derivational relationship instead of two independent patterns. More will be discussed in Sections 4.3.3.1 and 4.3.3.2. 127. Some of the discussions can also be found in Chin (2010a, 2010b, 2010c).
128. The example she quoted is [ning pooee shooee pay ngoh] 擰杯水畀我 ‘give me a glass of water’.
129. We also include other semantically related morphemes such as [lɐi] 來/嚟 ‘to come’ because they are also directional verbs.
Chapter 4. Diachronic development of give and its functions in Cantonese 83
Table 11. (continued) Author
Year
Hobson
1850
Bonney
IO marker
過
畀
1853
9
0
Bunyan
1871
86
1
Gospel of Luke
1873
76
9
Dennys
1874
21
0
Harper
1874
3
0
Anonymous
1877
3
0
Ball
1888
8
0
Fulton
1888
18
1
Stedman & Lee
1888
11
2
Mai
1893a
26
54
Mai
1893b
2
52
Ball
1894
14
1
Wisner
1906a, 1906b
21
3
LeBlanc
1910
7
1
Ball
1912
13
9
Cowles
1920
6
0
Ball
1924
12
0
Caysac
1926
0
35
Wisner
1927
23
3
Gospel of Luke
1931
98
0
2
1
Beneficiary marker 過129
畀
0
5
過2
過3
俾過 1 過5
過 20 來 14 嚟2 來過 1 俾過 1
俾過 1 歸1 嚟1 0
過1 嚟1 出嚟 1 來1 去5 過2
1 1
3 2
3 2 5 0
1
0
0
0
0
過2
過5 嚟2 俾過 1
1 11
0
2
0
7
俾過 1
8
0
5
0
29
過4 嚟2
過7 俾過 3
9 0
(continued)
84 Cantonese give and Double-Object Construction
Table 11. (continued) Author
Year
Wells
1931
Brouner & Fung
IO marker
Beneficiary marker
畀
1935
5
0
Hoh & Belt
1936
0
2
O’Meila
1941
13
40
Wells
1941
16
7
Chao
1947
0
1
Chiang
1949
0
2
Chan
1951
0
5
Chiang
1951
0
2
0
3
Oakley
1953
1
8
0
3
Bruce
1954
7
15
0
8
Whitaker
1954
1
7
0
2
Whitaker
1959
2
13
0
7
Wu
1960
0
5
0
0
Yuan
1960
0
1
24
8
Lau
1972
23
138
Chapman
1973
0
9
Gospel of Luke
1997
10
60
過129
畀
過
俾過 1 過3 過7 來7 嚟3 0
過1 嚟1
4 0 1 0
3
0
2
0
1
嚟1
嚟2
8
1
0
79
0
4
過1
16
In some materials published before the mid-1930s, the IO marker and the beneficiary marker were rendered by different morphemes. For example, in Bonney (1853), the IO marker was [kʷͻ] 過 (9 times) while [pei] 畀 (5 times) was used as the beneficiary marker. Such a distinction of semantic roles between [kʷͻ] 過 and [pei] 畀 can also be found in Anonymous (1877), Ball (1888), Wisner (1906a, 1906b), LeBlanc (1910), Ball (1912), Cowles (1920), Ball (1924), Wisner (1927).130 After 1940, both functions were taken up by [pei] 畀 almost exclusively, except Oakley (1953), Bruce (1954), Whitaker (1954, 1959), Lau (1972) and the Gospel of Luke (1997). Among the 44 works, Ball’s works deserve our attention the most. The data of Ball’s several works are summarized in Table 12. 130. Although not all these works show an absolute division of functions between these two morphemes, the general trend can still be observed.
Chapter 4. Diachronic development of give and its functions in Cantonese 85
Table 12. Distribution of 畀 and 過 as the beneficiary and IO markers in Dyer Ball’s works
Work
Year
Cantonese Made Easy Readings in Cantonese Colloquial How to Speak Cantonese Cantonese Made Easy
1888 1894 1912 1924
IO marker
過
8 14 13 12
畀
0 1 9 0
Beneficiary marker
過
0 2 0 0
畀
5 1 7 5
In Ball (1888), [kʷͻ] 過 and [pei] 畀 performed different roles: [kʷͻ] 過 as the IO marker while [pei] 畀 as the beneficiary marker. Six years later, [pei] 畀 began to function as an IO marker although the frequency was low with only one occurrence. Ball (1912) used more [pei] 畀 as the IO marker while its role as a beneficiary marker was still prevalent. Apparently, Ball (1924) seemed to contradict this trend. However, Ball (1924) was a revised edition of Ball (1888). Thus, it was normal for these two works to have a similar distribution. The above examination supports Yue’s claim that in the 19th century, [pei] 畀 was used as a beneficiary marker while double-object sentences used [kʷͻ] 過 as the IO marker. Later, the double-object construction shifted to using [pei] 畀 as the IO marker. The development of the IO markers is summarized in Figure 3 (see also Chin, 2010b, p. 12). Beneficiary construction
Double-object construction
NP1 + V[GIVE] + NP2 + 畀 + NP3
NP1 + V[+GIVE] + NP2 + 過 + NP3
NP1 + V [+GIVE / GIVE] + NP2 + 畀 + NP3
Figure 3. Development of IO marker from 過 to 畀 in Cantonese
Another reason for the proposed IO marker development can be attributed to the semantic nature of the markers concerned. The marker [kʷͻ] 過 originates from a verb meaning “to pass” in Cantonese. Schematically, it refers to an action that the agent goes from one place to another place. This schema well describes the action of giving in which the thing passes from the giver to the recipient. In our corpus, there are also instances in which the IO is introduced by other directional verbs such as [lɐi] 嚟 ‘to come’ or [hɵy] 去 ‘to go’ whose schematic description is similar to [kʷͻ] 過 ‘to pass’. For example, in Bridgman’s (1841) work, there is a sentence [⊂ts’ing ⊆ní tai⊇ chik⊃ chí⊃ sai⊃ ké⊃ ⊆lai ⊆ngó] 請你遞隻至細嚟我 ‘I will thank you
86 Cantonese give and Double-Object Construction
to send me one of the smallest’ (p. 178). We find that [kʷͻ] 過 was used more frequently than [lɐi] 嚟/來 ‘to come’ or [hɵy] 去 ‘to go’.131 The reason is related to the directionality of these three verbs. For [kʷͻ] 過, it is neutral in terms of the direction in the sense that the action can be directed to or away from the speaker. On the contrary, [lɐi] 嚟/來 ‘to come’ and [hɵy] 去 ‘to go’ are deictic verbs referring to the movements toward and away from the speaker respectively. This can be confirmed with the frequency counts of the noun phrases following these three types of IO markers, as summarized in Table 13. Table 13. Directions of the actions denoted by IO markers 嚟/來/去過132
IO marker
嚟/來 ‘to come’ 去 ‘to go’ 過 ‘to pass’
Toward the speaker 50132 0 123
Away from the speaker 2 5 244
The semantic extension that the two types of verbs (i.e., the double-object verb and the directional verb) underwent is different when grammaticalizing into an IO marker. The directional verb [kʷͻ] 過 mainly encodes a spatial transfer of the object from the giver to the recipient. The double-object verb [pei] 畀 encodes a transfer of the possession of the object from the giver to the recipient. Since the main verb in a beneficiary construction lacks the [+GIVE] feature, it needs an overt marker (i.e., [pei] 畀) to mark the beneficiary. Double-object verbs which bear the [+GIVE] feature already encode a transfer of possession and their recipients were not necessarily introduced by markers with the intrinsic [+GIVE] feature (i.e., [pei] 畀). The marker developed from the directional verb [kʷͻ] 過 thus suffices. 4.3.3.2 Typological significance of go-type and give-type IO markers The switchover of the IO marker from [kʷͻ] 過 to [pei] 畀 is not only found in Cantonese, but also in other Chinese dialects. Chin (2010b) proposed two types of IO markers: give-type and go-type in the Chinese language. The former is commonly found in modern Chinese dialects as our survey in Chapter 2 shows. The latter is common in the earlier stage of the Chinese language including its dialects. In other words, the go-type IO marker predates the give-type IO marker. The earliest
131. The frequencies of [lɐi] 嚟, 來, [hɵy] 去 and [kʷͻ] 過 are 28, 24, 5 and 648 respectively.
132. Examples of IOs that can illustrate an action directing toward the speaker include [ŋͻ] 我 ‘I’ or [ŋͻ tei] 我哋 ‘we’.
Chapter 4. Diachronic development of give and its functions in Cantonese 87
record of the go-type IO marker can be found in the oracle-bone inscriptions in which yu 于133 played the IO marker function.134 (72) 于父乙㣇㞢匄 (BB 51o = Heji 272o)135 ‘To Fu Yi (we should direct our) request [求] for what (we) lack’ (p. 178)136
(73) 貞于河報 (BB 117o [3] = Heji 672o)137 ‘Tested: (We) should direct the you-cutting sacrifice to (the spirit of) the River He of bao-captives’ (p. 283) (74) 勿五十牛于河 (BB 117o [6]) ‘As for fifty cattle (we) should not direct the you-cutting sacrifice of them to (the spirit of) the River He’ (p. 284)
Pulleyblank (1986), in his discussion of the locative particle usage of yu 于, claimed that yu 于 underwent semantic extension to obtain additional meanings or functions other than locative. Specifically, he mentioned that yu 于 can “indicate the recipient of an action” (p. 4). The example he used comes from the Book of Odes (Pulleyblank, 1986, p. 4).
(75) 獻豜于公 (Book of Odes, Ode 154 七月) ‘We present the older boars to the prince’
The meaning of yu 于 as a recipient marker can thus be associated with the basic meaning ‘to go’ in the following way: the ritual/sacrificial performers present some sacrificial victims which go to the spirits/ancestors/deities.138 133. Mei (2004), quoting Gong Hwang-cherng’s data on Proto-Sino-Tibetan reconstruction, showed that yu 于 is etymologically related to wang 往. See also the discussions in Guo (1997) and X. G. Qiu (2010) on the meaning of yu 于 and its development into other functions.
134. For a more detailed discussion of the IO marker in oracle-bone inscriptions and its development in the later stage of the Chinese language, see Chapter 5 of Chin (2009). 135. BB (Bing Bian) and Heji stand respectively for the Fascicle Three of Inscriptions from the Yin Ruins 殷墟文字丙編 compiled by Chang Bingchuan 張秉權 and the Collections of the OBI 甲骨文合集 edited by Guo Moruo 郭沫若. The suffixes “o” and “r” after the inscription numbers stand for “obverse” and “reverse” respectively. Reference to the oracle-bone inscriptions in these two works and the translation of BB texts is based on Takashima and Serruys (2010). For those texts that have two translations (one by Takashima and one by Serruys), the one done by Takashima is adopted. 136. The page numbers of the translation are based on Takashima and Serruys (2010). 137. Numbers in the square brackets represent the sentence numbers on the plastron.
138. The IO marker yu 于 was then replaced by yu 於 and later yu 與 (which is a double-object verb) starting from the Eastern Han dynasty (Peyraube, 1986).
88 Cantonese give and Double-Object Construction
There are other dialects that also use the go-type IO markers, such as du 度 in the Min dialects of Quanzhou (Lien, 2002), Jinjiang (Tung, 1959) and Shaowu (Kwok, 2008), ti 至 in the Min dialect of Hainan (Lee, 2011),139 the Wu dialect in the Ming and Qing dynasties (R. Shi, 2006a, 2006b, 2006c). R. Shi (2006a, p. 232) argued that the IO markers la 拉 and le 勒 found in Feng Menglong’s collection of Mountain Song 山歌 (compiled in 1612) and The White Sewn Fur Coat 綴白 裘 originated from the directional verb 來 as a result of phonological change. In the Shanghai dialect of the late 19th century, la 拉 was still used as the IO marker as reported by N. Qian (1997, 2003), and it was later replaced by bo 撥 after the 1930s. The same is also observed in the languages of the Southeast Asian linguistic area such as Lahu (Matisoff, 1973, 1991, 2003), and Thai (Kessakul & Ohori, 1998). An on-going switchover from the go-type to the give-type IO markers is observed in the Conghua dialect (a sub-dialect of Yue).140 My fieldwork with four Conghua informants of different ages reported in Chin (2010a, 2010b, 2010c) showed that younger speakers do not use the go-type IO marker [kʷͻ] 過. They only use the give-type marker. Older speakers use both markers but when the double-object verb does not specify a genuine transfer of possession, such as [tse] 借 (to borrow), the go-type IO marker is preferred. The fieldwork data of the four informants is summarized in Table 14 (modified from Chin (2010b), p. 14).141 Table 14. The go-type and give-type IO markers in Conghua dialect %
Informant 1
Informant 2
Informant 3
Informant 4
Marker
K 25
K 47
K 23
K 0
K/P P 26 49
K/P P 3 50
K/P P 16 61
K/P P 10 90
In Heine and Kuteva’s (2002, p. 330) crosslinguistic study of lexicon of grammaticalization, the dative case has three major sources: allative, beneficiary and GIVE. Allative expresses “the meaning of motion ‘to’ or ‘towards’ a place” (Crystal, 2003, p. 19). In their survey of the grammaticalization patterns of the allative case, Rice and Kabata (2007) found that 34% of the 44 “genetically and areally diverse 139. Lee (2011) followed Chin’s (2010b) typological classification of IO markers in Chinese dialects and argued that the directional verb zhi 至 is the “optimal candidate for the etymon” (p. 520) of the IO marker ti in the Hainan dialect. 140. Conghua is about 56km to the northeast of Guangzhou. 141. The four informants were 70, 56, 59 and 28 years old at the time of the fieldwork investigation in 2007. 117 sentences were used in the fieldwork. K and P stand for the go-type and the give-type IO markers respectively. K/P refers to the situation where both IO markers were used by the informant.
Chapter 4. Diachronic development of give and its functions in Cantonese 89
languages” use the allative case to mark the recipient (p. 451 & 479). Typical examples include English and Japanese. The former uses the preposition to while the latter uses the postposition に (ni) to mark locations and recipients. From our survey, the go-type IO marker seems not as productive as the give-type in modern Chinese dialects. What is even more significant is that the go-type predates the give-type as far as the language data from oracle-bone inscriptions, pre-modern Wu and Cantonese dialects are concerned. It is argued that the semantic extension that the two types of verbs (i.e., the double-object verb and the directional verb) undergo is different when they are grammaticalized into IO markers. The most salient semantic attribute associated with the go-type IO marker is that it encodes a spatial transfer of the object from the giver to the recipient.142 The give-type IO marker encodes a transfer of the possession of the object from the giver to the recipient while the spatial transfer is not necessarily implied.143 These two semantic attributes associated with the two types of IO markers are formulated as follows: a. GO-type IO marker: [+spatial, –possessive] b. GIVE-type IO marker: [±spatial, +possessive] According to a native speaker of Thai,144 the IO marker [thա̆ŋ] (with the basic meaning of ‘to arrive’) is used with verbs like send, which principally encodes spatial transfer of an object to a goal. The same is also found in the Conghua dialect. Some speakers of the Conghua dialect use the IO marker [kʷͻ] 過 with verbs such as [tsε] 借 ‘to lend’ and [tsou] 租 ‘to rent out’, which do not involve a transfer of possession. This dichotomy is also observed in English. According to Hovav and Levin (2008), the so-called dative construction (as in I sent a book to him) expresses “caused motion” meaning that “an agent causes a theme to move along a path to a goal” while the double-object construction (as in I sent him a book) expresses “caused possession – causing a recipient to possess an entity” (p. 130).
142. Chor (2018) claimed that the use of [kʷͻ] 過 as an IO marker is a metaphorical extension of a spatial verb (meaning ‘to across’) to a “transfer particle” (p. 118).
143. For example, in the sentence wo song ni yi dong fangzi 我送你一棟房子 ‘I gave you a house’, there is no spatial transfer of the house. Only the possession of the house is transferred. 144. This native female Thai speaker was around 30 years old. As for the difference between [thա̆ŋ] and [hay], she stated that [thա̆ŋ] expresses a goal-recipient while [hay] expresses a recipient-beneficiary.
90 Cantonese give and Double-Object Construction
4.3.3.3 Null IO marker in pre-modern Cantonese We have discussed that the elision of the IO marker is a result of haplology. However, Phua (2007, p. 222) found that when [kʷͻ] 過 was still the dominant IO marker in Cantonese, elision had already taken place (see Table 9 and Table 10). For example, in Bridgman (1841), there are double-object sentences with no IO marker. (76) 畀 棉 布 汗 衫 我 ⊂pí ⊃ hón⊇ shám ngó mín pò ⊆ ⊂ ⊆ shirt I give cotton ‘Give me a cotton shirt’
(77) 畀 個 價 錢 我 ⊂pi kó⊃ ká⊃ ts’ín ⊆ ⊆ngó give cl offer I ‘Make me an offer’
It seems that haplology cannot explain the absence of the IO marker in the 19th century in which [kʷͻ] 過 was the dominant IO marker. When closely examining the null IO marker in the pre-modern Cantonese dialect materials, we obtain two important observations: a. Before the 1940s, nearly 61% (i.e., 360 out of 581) of the double-object sentences with [pei] 畀 as the main verb have null IO markers; b. These 360 sentences account for about 87% of the double-object sentences with null IO markers in the dialect materials. In other words, the null IO marker is mostly found in those sentences with [pei] 畀 as the main verb. This phenomenon can be explained by the fact that [pei] 畀 is the core member of the double-object verb family and sentences made up of this verb have a strong sense of giving which can optionally take the recipient marker. 4.3.4 As a passive marker Chin and Tsou (2005) observed that [pei] 畀 did not function as a passive marker until the turn of the 20th century. Prior to the 20th century, the passive marker was predominantly performed by [pei] 被.
(78) 自己 唔 顧 體面 攞 人 嫌意, 被 人 怒罵 one-self not care face take people advantage PASS people anger-scold My translation: ‘One who did not care about his/her own face and took other people’s advantage would be scolded by other people’ (Anonymous, 1877, p. 31)
Chapter 4. Diachronic development of give and its functions in Cantonese 91
(79) 就 竟然 被 佢 呢 一 個 人 來 盡 行 then actually PASS s/he this one cl person come completely execute (Mai, 1893a) 克服 統管 overcome administer My translation: ‘Then [someone] was completely overcome and controlled by such a person’
(80) 就 連 隻 大 野牛 都 被 射 死 (Mai, 1893a) then even cl big bull also PASS shoot die My translation: ‘Even the big bull was also shot dead’
(81) 賊 有 咁 多, 咁 多 被 兵 打 死?145 ⊇ ⊆ ⊃ ⊃ ⊇ ⊂ta ⊆sz ts’ak yao kóm ⊂to kom ⊂to pi ⊂ping thief has that many that many PASS soldiers beat die? ‘There were so many thieves, and were there many beaten to death by soldiers?’ 145(LeBlanc, 1910, p. 35)
(82) 仲 有 啲 多 飯 都 被 佢 食 埋 chûng yău ti toh fâan to pêi k’ŭi shîk māai still have cl much rice also PASS s/he eat suf ‘Still had a little (cooked) rice, also by him eaten up’
咯 lòk sfp (Wisner, 1906a, p. 31)
Below are two passive sentences with [pei] 畀 as the passive marker taken from Bruce (1954, p. 109) and Chapman (1973, p. 95) respectively. (83) 而 家 畀 個 山 窿 遮 住 晒 yi gka bpêi* gkòh saan lung tse tsûe saài now give cl cave block asp suf Bruce’s translation: Now we’re in the tunnel My translation: ‘Now we are completely blocked by the cave’
(84) 畀 蛇 或者 鱷魚 咬 一 啖 你 就 會 死 bei seh wak-jeh ngok-ue ngau yat damm nei jau wooi say crocodile bite one mouthful you then will die give snake or My translation: ‘A bite by a snake or crocodile can kill you’
It should be noted that [pei] 被 and [pei] 畀 have close pronunciations in Cantonese. The former has a yangqu 陽去 tone while the latter a yinshang 陰上 tone. Sometimes, the author rendered the passive marker with [pei] 被 but transcribed it as [pei35] (i.e., with the yinshang 陰上 tone). For example, Lau (1972) wrote the passive marker with the character [pei] 被 but the pronunciation he gave 145. Due to the special tone marks used in this work which cannot be properly displayed here, I thus show the tones with the traditional 4-corner system. Notice that [sei] 死 ‘to die’ is marked by the author with a yang-shang tone instead of a yin-shang tone. The French translation given by the author is: ‘autant de pirates il y avait, autant furent tues par les soldats’.
92 Cantonese give and Double-Object Construction
is [bei6*].146 This shows that one cannot rely on the written form of the passive marker. Instead, one has to check the transcriptions in order to determine if the passive marker is [pei] 被 or [pei] 畀. Besides the tone, the passive construction made up of [pei] 畀 and [pei] 被 has one syntactic difference: The agent after [pei] 畀 cannot be left out while the one after [pei] 被 can be optional. If the agent after the passive marker is not present, we can then safely conclude that the marker in question is [pei] 被 rather than [pei] 畀. The distribution of [pei] 畀 and [pei] 被 as the passive marker in the pre-modern dialect materials is summarized in Table 15 and Table 16.147 Table 15. Distribution of the passive markers 畀 and 被 (with phonetic transcriptions)
fn148148 fn149149 fn150150 fn15151 fn152152 fn153153 fn154
Author
畀
Frequency Morrison (1828) Dennys (1874) Ball (1888) Fulton (1888) Stedman & Lee (1888) Ball (1894) Wisner (1906a, 1906b) LeBlanc (1910) Ball (1912) Caysac (1926)
(pe)148
0 2 (pí)149 0 (⊂pei)150 2 (⊂pi) 0 (‘pi)151 2 (⊂pei) 4 (péi) 1 (⊂pi)152 3 (⊂pei) 0 (pi2)153
Frequency 4 (pe) 8 (pì) 4 (pei⊇) 1 (pi⊇) 5 (pi•) 14 (pei⊇) 6 (pêi) 6 (pi⊇) 0 3 (pi3)
被
No Agent 1 6 1 0 0 4 4 0 0 0
146. 6 refers to the yangqu 陽去 tone and the asterisk represents tone change and the resultant tone is yinshang 陰上. 147. Yiu (2010) also observed the use of [pei] 被 as the passive marker in pre-modern Cantonese dialect materials, but her analysis mainly focused on the passive marker [pei] 畀. 148. There is no marking for the tones in this work.
149. The two diacritic marks represent the shang tone and the qu tone respectively. The distinction of the yin and yang registers is made by the font style: The italicized font represents the yin register.
150. Although [pei] 畀 was not used as a passive marker in this text, we also include the transcription of [pei] 畀 as the double-object verb for comparison. The same applies to other texts below. 151. The yin register is denoted by an inverted comma while the yang register by a dot.
152. Since the special diacritic marks used in this work cannot be displayed easily, I convert them into the traditional four-corner system. 153. The yin and yang registers are distinguished by the upper and lower scripts of the tones respectively.
Chapter 4. Diachronic development of give and its functions in Cantonese 93
Table 15. (continued) Author
畀
Frequency Wells (1931) O’Meila (1941) Wells (1941) Chao (1947) Oakley (1953) Whitaker (1954) Whitaker (1954) Bruce (1954) Lau (1972) Chapman (1973) Total
(pei2)
1 21 (péi)154 3 (pei2) 12 (pee) 3 (pei) 1 (bee) 14 (bee) 7 (bpêi*)155 138 (pei6*)156 8 (bei) 222
Frequency
被
No Agent
4 (pei3) 18 (pêi) 0 1 (bey) 0 0 0 3 (bpêi) 0 0 80
1 0 0 1 0 0 0 3 0 0 21
154155156
Table 16. Distribution of the passive marker 畀 and 被 (without phonetic transcriptions)
Author
Hobson (1850) Bunyan (1871) Gospel of Luke (1873) Harper (1874) Anonymous (1877) Mai (1893a) Mai (1893b) Wisner (1927) Gospel of Luke (1931) Gospel of Luke (1997) Total
畀
Frequency
Frequency
0 0 0 0 0 1 0 4 0 0 5
1 91 44 1 3 30 29 7 52 59 317
被
No Agent 0 4 4 0 1 5 2 0 19 38 73
The pre-modern Cantonese materials can be divided into two major groups with respect to the representation of the passive marker: The first group (Table 15) has the passive marker transcribed in the texts which are given in the parentheses. The other group only has the markers written with Chinese characters (Table 16). The general trend is that before the 1930s, the majority used [pei] 被 as the passive 154. The acute and the circumflex symbols mark the yin-shang and yang-qu tones respectively. 155. The asterisk indicates tone change and the resultant tone is a mid-rising one. 156. The asterisk represents tone change and the resultant tone is a yinshang tone with the tonal value 35.
94 Cantonese give and Double-Object Construction
marker. After the 1930s, [pei] 畀 became the passive marker (except the Gospel of Luke published in 1997).157 Besides the phonological distinction between the two markers, we also examined the expression of the agent after the passive marker [pei] 被. According to our previous discussion, the agent can be left out after [pei] 被 but not after [pei] 畀. Our dialect materials show that all the passive sentences made up of [pei] 畀 have agents. On the other hand, some of the agents marked by [pei] 被 are elided and the frequencies are indicated in the last column of Table 15 and Table 16. The omission of agents suggests that the morpheme in these sentences is [pei] 被 instead of [pei] 畀. One may argue that even if we can prove that the morpheme [pei] 被 was used as the passive marker, there is still a possibility that this usage was due to the influence of the written language. In the Gospel of Luke (1997), there were sentences containing a mixture of both colloquial and written languages. For example, in the passive sentence 甚至當佢被拘禁 ‘even when s/he was detained’, the pronoun [k’ɵy] 佢 ‘s/he’ was used but there is no agent after [pei] 被. This passive usage is not native in colloquial Cantonese and the use of [pei] 被 is clearly borrowed from the written language.158 Notice that this edition of the Bible translation is the only work that used [pei] 被 as the passive marker in the contemporary period. Prior to the 1930s, [pei] 被 was used as the dominant passive marker in 10 out of 11 works. It is difficult to conceive that all the works prior to the 1930s were influenced by the written language. Even if such an influence did exist, there remains the question why those works compiled after the 1930s did not receive the same influence. Therefore, it is more reasonable to argue that the Cantonese dialect before the early 20th century used [pei] 被 as the passive marker, which was replaced by [pei] 畀 around the 1930s.159 The frequency distribution between [pei] 畀 and [pei] 被 before the 1930s and after the 1930s (based on those works with transcriptions) is summarized in Table 17. 157. In her comparison of two Cantonese translation of《官話指南》(The Guide to Kuan Hua) published in 1895 and 1930, M. L. Zhang (2018, 2020) found that the translated text published in 1930 used bei 被 as the passive marker while the one in 1895 mainly used bei 畀.
158. In her study of two 16th-century Min texts: Li Jing Ji 荔鏡記 and Doctrina Christina, Chappell (2000) also noticed that both colloquial and literal passive markers (乞 and 被 respectively) were used in the same text. The latter is mainly found in the sections that were written with a more classical language style. Modern Cantonese sometimes exhibits a mixture of both colloquial and literary layers. Matthews and Yip (2001) attributed the use of [pei22] 被 in [keoi5 zung1 jyu1 bei6 sik1 fong3] 佢終於被釋放 ‘S/he was released eventually’ in Cantonese (p. 269, with Matthews & Yip’s transcription) an illustration of high register Cantonese which is commonly found in news broadcast or formal speech.
159. J. Yang (2006) argued that the change of the passive marker from [pei] 被 to [pei] 畀 was due to their close pronunciations.
Chapter 4. Diachronic development of give and its functions in Cantonese 95
Table 17. Comparison of the passive markers 畀 and 被 before and after 1930
Passive marker 畀 被
Before 1930
After 1930
15 (21.4%) 55 (78.6%)
207 (90.4%) 22 (9.6%)
4.3.5 As a verb introducing instruments We now turn to the function of introducing instruments by [pei] 畀.160 Consider the following sentences.
(85) 你 都 唔 捭 耳 聽 我 講 ⊆ní ⊂pí ⊆í ⊆ngo ⊂kong to m t’ing ⊂ ⊆ ⊂ you also not give ear listen I speak ‘You do not at all give ear to what I say’
(Bridgman, 1841, p. 58)
(86) 屍 移正, …, 俾 白 布被 蓋 身 上, 又 俾 corpse move straight, …, give white cloth cover body top also give 面巾 蓋 住 頭面 face towel cover suf head face ‘The corpse is set properly …the body is covered by white cloth, the head by a face kerchief ’ (Hobson, 1850, p. 15.6A)
(87) 俾 水 洗 米 give water wash rice ‘Take water and wash the rice’
(Bonney, 1853, p. 54)
(88) 我 見 佢 畀 棍 打 一 箇 細 呅 仔 ngo kin’ k’ü ‘pi kwan’ ‘ta yat, ko’ sai’ man ‘chai I see s/he give stick beat one cl child ‘I saw him beating a boy with a stick, your Honor’ (Stedman & Lee, 1888, p. 145) (89) 呢 的 藥 材 俾 水 嚟 開 ⊇- ts’oi ⊂pi ⊂shui lai ni tik jeuk ⊂ ⊃ ⊆ ⊆ ⊂hoi this cl medicine give water come dilute ‘Dilute this medicine with water’
(Fulton, 1888, p. 21)
(90) 又 得 隻 雀 俾 佢 對 翼 的 毛 來 遮 護 住 佢 also get cl bird give it pair wing cl feather come shelter protect suf it My translation: ‘Also, the bird used the feathers of its wings to protect itself ’ (Mai, 1893a, p. 177) 160. J. Yang (2006) also observed the use of [pei] 畀 to introduce instruments but she did not give any account of its development and its relationship with the double-object verb [pei] 畀.
96 Cantonese give and Double-Object Construction
(91) 呢 陣 時 啲 大火 船 係 俾 鐵 做 嘅 ni chân shī ti tâai fóh shūen hâi péi t’ìt tsô kè this-time cl big fire ship be give iron make sfp ‘This period of time the big steamers are made by iron’ (Wisner, 1906a, p. 32)
(92) 唐人 著 嗰 啲 鞋 俾 草 整 嘅, 你 叫 ⊃ ⊂pi ⊂ts’o ⊂ching ke⊃ ⊆ni kiou⊃ t’ongjan tcheuk ko ti hai ⊂ ⊆ ⊆ ⊆ ⊃ Chinese-people wear those cl shoe give grass make prt you call 做 乜 名?161 ⊇ tso mat⊃ ⊆ming? what name ‘How do you call those shoes worn by Chinese people that are made of grass?’ 161 (LeBlanc, 1910, p. 175)
(93) 先先 要 俾 石 共 坭, 整 一 條路 ⊃⊂pei shek⊇ kung⊇ ⊂ching yat ⊇ sin sin jiu nai ⊆ ⊂ ⊂ ⊃ ⊆t’iu lo first need give stone and mud make one cl road ‘First of all a road has to be made with stones and earth’ (Ball, 1912, p. 148)
(94) 俾 的 梳打 嚟 煠 的 枱布 喇 laī shaāp ºti *t’oī pò la peí ºti ºshoh tá give cl soda come boil cl table cloth sfp ‘Put in some soda when you boil the table cloth’ (Hoh & Belt, 1936, p. 69)
Table 18 gives the frequency distribution of this usage of [pei] 畀 in our pre-modern Cantonese materials. It can be seen that the use of [pei] 畀 to introduce instruments was no more found in the materials published after 1941 except the Gospel of Luke (1997). This decreasing trend can be observed from the three editions of the Gospel of Luke in which the frequencies of [pei] 畀 as an instrument marker are 9, 4 and 2 respectively.162 Why did this function of [pei] 畀 disappear in the dialect?
161. The tone marks used in this work are converted to the traditional four-corner system for easy display. The French translation is: ‘comment appelez vous ces chaussures de paille que portent les Chinois?’.
162. The use of [pei] 畀 to introduce instruments can be found in some old Cantonese movies produced in Hong Kong in mid-20th century. Two examples are taken for illustration from the movies 人面桃花相映紅 (Peach-Blossom Face, 1956) and 龍鳳合歡花 (The Joyful Matrimony, 1960) respectively: 頭先邊個畀呢嚿木頭嚟扑我嘅?(My translation: Who just beat my head with the wood?) and 嗰啲茶好似畀海水煲噉 (My translation: the tea seems to be made with sea water). These examples were taken from the second phase of The Corpus of Mid-20th Century Hong Kong Cantonese [二十世紀中期香港粵語語料庫 https://hkcc.eduhk.hk/] (Accessed on 25 March 2021).
Chapter 4. Diachronic development of give and its functions in Cantonese 97
Table 18. Frequency distribution of 畀 with the function of introducing instruments
Author
Year
Morrison Bridgman Devan Hobson Bonney Bunyan Gospel of Luke Dennys Fulton Stedman & Lee Mai Ball Wisner LeBlanc Ball Wisner Gospel of Luke Wells Hoh & Belt O’Meila Wells Gospel of Luke Total
1828 1841 1847 1850 1853 1871 1873 1874 1888 1888 1893a 1894 1906a, 1906b 1910 1912 1927 1931 1931 1936 1941 1941 1997
Frequency 1 6 2 22 5 7 9 7 4 1 26 7 13 3 12 13 4 2 1 9 3 2 159
4.3.5.1 Disappearance of the instrument marker function The three functions of [pei] 畀, namely causative (or permissive), introducing instruments and passive, share the same surface structure: NP1 + 畀 + NP2 + VP. In Chapter 3, we showed that the passive marker function of GIVE is derived from the causative function and the interpretation of [pei] 畀 between these two functions sometimes depends on the context. (95) is one example.
(95) 落低 個 度 簾 咪 俾 熱頭 晒 lok⊇-⊂tai ko⊃ to⊇ ⊆lim ⊆mai ⊂pi it⊇⊆t’au shaai⊃ lower that cl curtain not give sun shine Original translation: ‘put down that curtain; don’t let sun shine in’ (Fulton, 1888, p. 14) Possible translation: ‘Lower that curtain so that the room will not be shone by the sun’
98 Cantonese give and Double-Object Construction
When taking the instrument introduction function of [pei] 畀 into consideration, we find that these three functions are not always clearly distinguished. Consider the following examples.
(96) 莫 俾 鉸翦 來 剪去 僥 個 的 枝葉 (Mai, 1893a, p. 7) not give scissor come cut away asp that cl leave Possible translation: ‘Don’t cut away those branches and leaves with the scissors’ Possible translation: ‘Don’t let the branches and leaves be cut by the scissors’
(97) 佢哋 … 就 俾 倃 大石 塞 住 個 竉 口 ⊆k’öü-tei⊇ … tsau⊇ ⊂pei kau⊇ tai⊇ shek⊇ sak⊃ chu⊇ ko⊃ ⊂lung-⊂hau they … then give cl big rock block suf cl hole-mouth Original translation: ‘A stone was brought, and laid upon the top of the den’ (Ball, 1894, p. 42) Possible translation: ‘They were blocked by the big rock at the den’
(98) 就 算 係 俾 刀 嚟 拮 佢 個 肚, 我 估 佢 都 tsâu sun hâi péi to lāi kat k’ŭi kòh t’ŏ ngŏh kóo k’ŭi to guess he also even is give knife come prick he cl tummy I 事必 唔 肯 放手 shî pìt m̄ háng fòng-sháu must not willing let-go Original translation: ‘then although [someone] used a knife to prick his stomach, I think he also certainly not willing to let go’ (Wisner, 1906b, p. 180) Possible translation: ‘Even if his tummy is pricked by a knife, I guess he may not be willing to let go’
Although the morpheme [pei] 畀 in sentences (97) and (98) were translated by the author with the function of introducing an instrument, it is not impossible to interpret these sentences with a passive meaning. The subject in the sentences for both functions are animate who can use an instrument to carry out an action, or suffers from an action. As discussed in Chapter 3, the difference between these two readings and the one of the causative construction is whether the subject is willing to experience the action denoted by the VP in the second part of the sentence. The three constructions share the same identical surface structure: NP1 畀 NP2 VP where the interpretation of [pei] 畀 sometimes is ambiguous. This then creates difficulties to the addressees to choose the appropriate reading for [pei] 畀. It is thus likely for one of these three structures be eliminated in the language system in order to save the extra effort required to resolve the ambiguity. As shown by the pre-modern dialectal materials, the function of introducing instruments became obsolete when the passive marker function was gaining its ascendancy around the 1940s. The mechanism of the syntactic change is summarized as follows.
Chapter 4. Diachronic development of give and its functions in Cantonese 99
First, in pre-modern Cantonese, [pei] 畀 in the NP1 畀 NP2 VP pattern was used as a causative verb or a verb introducing instruments. When [pei] 畀 was further grammaticalized into a passive marker, the causative construction involving [pei] 畀 was reanalyzed as a passive construction when the action is involuntary from the perspective of the agent. As a result, the surface syntactic pattern NP1 畀 NP2 VP acquired three meanings. This surface pattern caused ambiguity which was minimized by eliminating one of the competing forms. Based on our dialectal data, it was the function of introducing instruments opted out. 4.4 Summary The chronological development of the various grammatical functions of [pei] 畀 is summarized in Table 19.163 Table 19. Temporal development of the grammaticalization process of 畀 in pre-modern Cantonese Function
Before 1940
After 1940
+ + +/− +/− +
+ + + + −
1. A double-object verb 2. A causative verb 3. A passive marker 4. An IO marker 5. A verb introducing instruments
Chappell and Peyraube (2006, p. 982), based on diachronic materials on the Southern Min dialects, proposed the following two grammaticalization paths for the double-object verb GIVE. a. GIVE > dative marker164 b. GIVE > causative > passive marker Our study on pre-modern Cantonese generally agrees with Chappell and Peyraube’s proposal. The passive marker function of [pei] 畀 developed from the causative function of [pei] 畀. The major difference between Cantonese and the Min dialects is the function of instrument marking which, to my best knowledge, has not been discussed in the Min dialects. 163. The “+/-” sign indicates that the usage was found but was not yet dominant in the period. 164. Chappell and Peyraube (2006) used the notion dative marker instead of IO marker.
100 Cantonese give and Double-Object Construction
This chronological development also fits in the three semantic-pragmatic tendencies put forth by Traugott and König (1991) to account for grammaticalization. Among the three tendencies, the first and the third ones are relevant to our current study. Tendency I: Meanings based on the external described situation > meanings based on the internal (evaluative/perceptual/cognitive) situation (p. 208); Tendency III: Meanings tend to become increasingly situated in the speaker’s subjective belief-state/attitude toward the situation (p. 209). Tendency I is related to the “shift from a concrete, physical situation to reference to a cognitive, perceptual situation” (p. 208). This can account for the use of GIVE as a causative verb because what is given to the recipient is not a concrete object but an abstract one, such as a permission or a right. Tendency III illustrates the grammaticalization of [pei] 畀 into a passive marker. This tendency is concerned with the speakers’ perception toward the situation s/he describes. We have shown that the passive meaning is associated with unfavorable situations which are subject to the perception of the agent. Another important finding in this diachronic study is the rise and fall of the various grammatical functions of [pei] 畀. In particular, we focused on the emergence of the passive marker function and the disappearance of the function of introducing instruments. This change illustrates how reanalysis might give rise to ambiguity, which cost listeners or addressees extra efforts in sentence processing during communication. In fact, [pei] 畀 is not the only example motivating this type of syntactic change in Cantonese. In his study of the Cantonese distal demonstrative [kͻ] 嗰 ‘that’, Cheung (2006) observed that from the early 19th century materials, the distal demonstrative was pronounced as [kͻ33] (with the yinqu 陰去 tone) and it was homophonous with the general classifier [kͻ] 個. The combination of the distal demonstrative and the classifier [kͻ] 個 has the same pronunciation as the reduplication of the classifier [kͻ] 個. The former has a definite reading, meaning ‘that 個 (as a classifier)’ while the latter a distributive reading, meaning ‘every 個 (as a classifier). With the grammar notes from various pre-modern Cantonese materials, Cheung argued that the ambiguity arising from this surface form was then resolved by changing the tone of the distal demonstrative from yinqu to yinshang: [kͻ33]>[kͻ35].165 165. Orthographically, the classifier [kͻ] 個 is also distinguished from the distal demonstrative. According to Cheung, possible written forms (though there is no standardization) for the latter include 嗰, 箇, 果.
Chapter 4. Diachronic development of give and its functions in Cantonese 101
Finally, it is also noted that the chronological development of the grammaticalization of [pei] 畀 coincides with that of child acquisition of the various functions of [pei] 畀 in Hong Kong Cantonese. According to Wong (2004), “the sequence of acquisition is found to be transfer > permissive > dative > passive” (p. 326). Wong (2004) found that the passive construction is first used when a child is about 2.5 years old but the main verb usage and the causative function are acquired as early as 1 year old and 11 months old respectively. The earliest IO marker usage (dative in Wong’s terminology) is found when the child is around 2 years old. Wong argued that the causative function and the IO marker function of [pei] 畀 share the semantic content of transfer which can be easily acquired by children as an extension of the basic meaning of [pei] 畀. On the other hand, the passive marker function of [pei] 畀 is not connected to the semantics of transfer directly. In spite of its identical surface structure with the causative construction, this semantically peripheral construction is not acquired by children until they have mastered those core transfer-related constructions effectively.
Part II
Word order change in Cantonese double-object construction
Chapter 5
Word order typology of the double-object construction in Chinese dialects
In Part I, we examined the various syntactic functions performed by GIVE as a result of grammaticalization and their diachronic development in Cantonese. In the following chapters of Part II, focus will be placed on the typology of the double-object construction in Chinese dialects, and the on-going word order change in Hong Kong Cantonese. Although it has been claimed that the Northern and the Southern dialects exhibit opposite word order for the DO and IO, there are dialects in which the two patterns coexist. This underlines the importance in recognizing syntactic stratification. 5.1
Typological features in Northern and Southern Chinese grammar
Traditional Chinese dialectology has been focusing on the linguistic variations in the phonological aspect across Chinese dialects while the syntactic aspect does not draw much attention or the discrepancies among dialects in syntax are considered negligible. Chao (1968) once commented that “it is in matters of grammar that the greatest degree of uniformity is found among all the dialects of the Chinese language” (p. 13). In fact, studies of dialectal grammar have not received much attention until recent decades.166 Norman (1988), based on ten linguistic features, classified Chinese dialects into three groups: Northern, Central and Southern. The Northern group refers to the Mandarin dialects while the Southern one encompasses the dialect groups of Yue, Min, Hakka and Wu. The Central group is made up of the Xiang and Gan dialect groups, which bear the linguistic features of both the Northern group and the Southern group. In the following discussion, we will subsume the Central group under the Southern group.167 166. For an overview on the studies of Chinese dialectal grammar, see A. Yue (1988, 2003). There were two major publications in the 1990s on the studies of Chinese dialectal grammar, see YueHashimoto (1993a) and B. Huang (1996). 167. Ramsey (1987) classified Chinese dialects into two major groups: Northern and Southern. His Southern group includes the Central group defined by Norman.
106 Cantonese give and Double-Object Construction
There are significant typological features in grammar between the Northern and the Southern dialects. In the past decades, the typology of the neutral question and the comparative construction has been studied extensively.168 For example, according to Yue-Hashimoto (1993a, p. 158), (99) and (100) are the native patterns of the comparative construction in the Northern and Southern dialects respectively. The major difference between these two patterns is the position of NP2 (the object or the standard) with respect to the stative verb (Vst). (99) and (100) have the stative verb appearing after and before NP2 respectively. (99) NP1 + 比 + NP2 + Vst (100) NP1 + Vst + 過 + NP2
It is noted that these two native patterns are not always neatly distributed across the dialects. In other words, the pattern typical to the Northern dialects is found in some Southern dialects or vice versa. For example, in the Hakka dialect of Rucheng 汝城 of Hunan province, the two patterns are found, as shown in the following four sentences taken from B. Huang (1996, pp. 681–682).169
(101) 我 比 你 嗊 壯 I compare you a bit fat ‘I am fatter than you’
(102) 今晡日 比 昨晡日 嗊 冷 yesterday compare yesterday a bit cold ‘Today is colder than yesterday’
(103) 筍 子 高 過 竹 bamboo shoot tall pass bamboo ‘Bamboo shoots are taller than bamboos’ (104) 病人 強 過 郎中 patient strong pass doctor ‘Patients are stronger than doctors’
The co-occurrence of two typologically distinct patterns within a single dialect is known as stratification where “the inherited ‘core’ [stratum] is discernible underneath the subsequent ‘layers’ of innovative influence from outside” (Aikhenvald, 2006, p. 5). Yue-Hashimoto (1993b) emphasized that one should “view the complexities in the interweaving of subsystems as merger from different linguistic strata 168. For the neutral question, see Zhu (1990), A. Yue (1992, 2006a), Yue-Hashimoto (1991a, 1994, 1999), P. Wang and Lien (2001). For the comparative construction, see A. Yue (1997). 169. A similar situation is also found in Hong Kong Cantonese. For details, see A. Yue (1997), Chappell and Peyraube (2015).
Chapter 5. Word order typology of the double-object construction in Chinese dialects 107
formed along a temporal and spatial scale … one has to assume language contact as the cause and linguistic stratification as the result”. (p. 216) In other words, one of the two strata represents the native stratum while the other a loan form imported from other dialects or languages due to language contact. 5.2
Typology of double-object construction in Chinese dialects
Besides the neutral question and the comparative construction, the double-object construction is another example showing typological differences between the Northern and the Southern dialects. The Northern dialects have the IO preceding the DO (i.e., the IO DO word order) while the Southern dialects have the reverse (i.e., the DO IO word order) for the give-type double-object construction (see D. Liu, 2001; Yue-Hashimoto, 1993a and A. Yue, 2003). 5.2.1 The IO DO pattern The IO DO pattern is found in the Zhongyuan Mandarin sub-group (中原官話) and some Min dialects of the Eastern sub-group.170 A. Zhongyuan Mandarin dialects171 Beijing (北京) (Hou, 1998, p. 40) Luoyang (洛陽) (W. He, 1993, p. 110) Zhengzhou (鄭州) (Lu et al., 1998, p. 44) Lingchuan (陵川) (Jin, 1983, p. 23) Wenshui (文水) (Hu, 1984, p. 65) Taiyuan (太原) (Hou & Wen, 1993, p. 300) Linfen (臨汾) (Hou & Wen, 1993, p. 300) Changzhi (長治) (Hou & Wen, 1993, p. 300) Wuhui (烏回) (L. Zhou, 2002, p. 17) Zhongning (中寜) (B. Huang, 1996, p. 675) Tongxin (同心) (A. Zhang, 2000, pp. 271–272)171
給我本兒書 ‘Give me a book’ 給我本兒書 ‘Give me a book’ 給我一本書 ‘Give me a book’ 給我一本書 ‘Give me a book’ 給我一本書 ‘Give me a book’ 給我一本書 ‘Give me a book’ 予我一本兒書 ‘Give me a book’ 給我一本兒書 ‘Give me a book’ 他給我給咧一本子書 ‘He gave me a book’ 給我給一碗水 ‘Give me a bowl of water’ 送給他一雙鞋 ‘Give him a pair of shoes’ 給他送給一雙鞋 ‘Give him a pair of shoes’
170. The sub-grouping of Mandarin and Min dialects follows those in Language Atlas of China 中國語言地圖集 (Wurm et al., 1988) and Norman (1991) respectively.
171. The second example is different from the case in Beijing Mandarin in which the verb song 送 ‘to give’ is also followed by gei 給. According to Zhang, this gei 給 is not an IO marker since the IO marker already appears in the gei ta 給他 phrase at the beginning of the sentence. Therefore, this gei 給 functions as a complement to the main verb and is obligatory even if the main verb carries the [+GIVE] feature.
108 Cantonese give and Double-Object Construction
B. Min dialects (a) The Southern Min sub-group Xiamen (廈門) (C. Zhou & Ouyang, 1998, pp. 390–391) Suixi (遂溪) (Yue-Hashimoto, 1985, p. 358) Taiwan (台灣) (H. Yang, 1995, p. 536) (b) The Northeastern sub-group
周的送林的一雙鞋 ‘Zhou gave Lin a pair of shoes’ 我先給汝三個銀 ‘I gave you three dollars first’ 伊送阿英蜀本冊 ‘S/he gave Ying a book’
Fuzhou (福州) (Z. P. Chen, 1998, pp. 203–204) 依爸乞我蜀把鋼筆 ‘My father gave me a pen’
5.2.2 The DO IO pattern The following lists the dialects in which the double-object construction has the DO IO pattern. These dialects are found in many Southern dialects as well as the Mandarin dialects of the Jianghuai and Southwestern sub-groups.172 The case of the Yue dialects will be discussed in Section 5.5. A. Jianghuai Mandarin dialects
Nanjing (南京) (Fei & Sun, 1993, p. 239) 把一本書我 ‘Give me a book’ Hangzhou (杭州) (Simmons, 1992, p. 159) 我送一本書得他 ‘I give him a book’ Huanggang (黃岡) (H. Wang, 2004, p. 214) 你送五十斤米、二十斤肉(得)二哥 ‘Give 50 catties of rice and 20 catties of meat to brother’
B. Southwestern Mandarin dialects Changde (常德) (Q. Zheng, 1999, p. 320)
給本書我 ‘Give me the book’
C. Wu dialects Shanghai (上海) (Y. Yan, 1994, pp. 245–246) Shaoxing (紹興) (Y. Yan, 1994, pp. 245–246) Wenzhou (溫州) (Y. Yan, 1994, pp. 245–246) Jiangshan (江山) (Akitani, 2001, p. 139) Guangfeng (廣豐) (Akitani, 2001, p. 139) Tiantai (天台) (Z. Dai, 2003, pp. 154–156)
撥張紙頭(撥)我 ‘Give me a piece of paper’ 撥張紙頭我 ‘Give me a piece of paper’ 哈張紙我 ‘Give me a piece of paper’ 渠杯個lᴇ44桃得我 ‘He gave me a peach’ 渠約m0個lɐi445桃我 ‘He gave me a peach’ 我撥本書撥爾 ‘I gave you a book’
172. In their studies of four textbooks 《人中畫》 ( 、《官話問答便語》 、 《白姓官話》and《學官 話》) of the 18th century Mandarin dialect compiled in Ryukyus (琉球), W. Li et al. (2015, p. 51) found that none of these four texts has the IO DO pattern. Instead, the DO IO pattern was used exclusively.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 109
D. Xiang dialects Changsha (長沙) (Bao et al., 1999, p. 335) Hengyang (衡陽) (Y. M. Li, 1986, p. 425) Lianyuan (漣源) (Y. Wu, 2005, pp. 334–335) Xinhua (新化) (Y. Wu, 2005, pp. 334–335)
我把噠一本書他 ‘I gave him a book’ 他送噠一瓶酒我 ‘He gave me a bottle of wine’ 你明日買來,我就得錢你 ‘If you buy it tomorrow, I will give you money then’ 送雙鞋子賜佢 ‘Give him a pair of shoes’ 把本書(把)我 ‘Give me a book’
E. Gan dialects Jishui (吉水) (R. L. Li & Chang, 1992, p. 451) Xinyu (新余) (R. L. Li & Chang, 1992, p. 451) Yifeng (宜豐) (R. L. Li & Chang, 1992, p. 451) Xiushui (修水) (R. L. Li & Chang, 1992, p. 451) Anyi (安義) (R. L. Li & Chang, 1992, p. 451) Yangxin (陽新) (R. L. Li & Chang, 1992, p. 451) Susong (宿松) (X. Huang, 2014, p. 199)
F. Hakka dialects Meixian (梅縣) (R. L. Li & Chang, 1992, p. 451) Liannan (連南) (R. L. Li & Chang, 1992, p. 451) Heyuan (河源) (R. L. Li & Chang, 1992, p. 451) Jiexi (揭西) (R. L. Li & Chang, 1992, p. 451) Ganxian (贛縣) (R. L. Li & Chang, 1992, p. 451) Liancheng (連城) (Xiang, 1997, pp. 360–363)
G. Min dialects of the Western sub-group Jian’ou (建甌) (Z. T. Chen & Li, 1991, p. 95) Jianyang (建陽) (Z. T. Chen & Li, 1991, p. 95) Songxi (松溪) (Z. T. Chen & Li, 1991, p. 95)
5.3
把本書到我 ‘Give me a book’ 把本書我 ‘Give me a book to’ 拿本書到阿 ‘Give me a book’ 拿本書(到)我 ‘Give me a book’ 拿本書(到)我 ‘Give me a book’ 給本書我 ‘Give me a book’ 我把書在佢 ‘I give him a book’ 分(一)本書分亻厓 ‘Give me a book’ 分一本書亻厓 ‘Give me a book’ 把本書亻厓 ‘Give me a book’ 分一本書亻厓 ‘Give me a book’ 拿本書亻厓 ‘Give me a book’ 分一惜韮菜(分/拿)我 ‘Give me a chive’ 拿蜀本書納我 ‘Give me a book’ 拿蜀本書納我 ‘Give me a book’ 拿蜀本書納我 ‘Give me a book’
Syntactic stratification in double-object construction
The two typological patterns, however, are not always neatly distributed in the Chinese dialects. In some Southern dialects listed above, the IO DO pattern is also found. a. In the Wu dialect of Jiangshan 江山 in Zhejiang province, in addition to 渠杯 個lᴇ44桃得我 ‘He gave a peach to me’, Atkitani (2001, p. 139) also listed 渠杯得 我個lᴇ44桃 as another pattern. However, Akitani did not give any information about the differences between these two patterns in the dialect.
110 Cantonese give and Double-Object Construction
b. In the Wu dialect of Shanghai, Y. Yan (1994, pp. 245–246) claimed that younger speakers also use the IO DO pattern such as 撥我一張紙 ‘give me a piece of paper’ and he attributed this phenomenon to the influence of Putonghua. The same observation is also found in B. Xu and Tang (1988, pp. 479–480). c. The Jianghuai Mandarin dialect of Huanggang 黃岡 spoken in Hubei province also has the IO DO pattern such as 你送二哥五十斤米、二十斤肉 (H. Wang, 2004, p. 214). Wang described that this pattern is identical to the one in Putonghua, but no further detail was given. d. Y. Wu (2005) reported that the V IO DO and the IO-Marker IO V DO patterns were found in 30 and 4 Xiang dialects spoken in Hunan province respectively. She further commented that “the word order [of the double-object construction] in the Xiang dialects is now changing under the influence of Mandarin. The Mandarin pattern [i.e., IO DO] has come into the Xiang dialects and coexists with the Xiang word order in many localities” (p. 338). e. The influence of Putonghua is also found in the Wu dialect of Tiantai in which the IO DO is used in addition to the DO IO pattern, as claimed by Z. Dai (2003). It is clear that syntactic stratification in some Southern dialects exists in the double-object construction. In addition to the native DO IO pattern, the non-native IO DO pattern is also found, which can be attributed to the influence of Putonghua or standard Mandarin. At the same time, the Northern dialects are also found to receive the influence of the Southern dialects. In the next section, we will examine the case of Beijing Mandarin by discussing Zhu Dexi’s influential study on the syntactic constructions involving gei 給. 5.4
Double-object construction in Beijing Mandarin
On the basis of the Mandarin dialects presented in Section 5.2.1, we can generalize three patterns for the double-object construction in the Northern dialects. (105) V IO DO (給我一本書, Taiyuan dialect)
(106) 給 IO V DO (給他送給一雙鞋, Tongxin dialect) (107) V 給 IO DO (送給他一雙鞋, Tongxin dialect)
In his analysis, Zhu (1979) also included the V DO 給 IO pattern such as ta song yi ben shu gei wo 他送一本書給我 ‘s/he gave me a book’.173 Zhu divided the verbs into four categories, among which his Va group are double-object verbs. Zhu further 173. It should be noted that not all the sentences in Zhu’s discussion are double-object in nature. Our discussion will only focus on double-object sentences.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 111
divided this set of verbs into two sub-groups: Va and Va/c.174 The members of these two sub-groups are listed in Table 20. Table 20. Examples of Va and Va/c verbs in Zhu (1979)175176 Verb type 175
Va
Va/c
Example
song 送 ‘to give as a gift’, mai賣 ‘to sell’, huan 還 ‘to return’, di 遞 ‘to pass’, fu 付 ‘to pay’, shang 賞 ‘to award’, jiang 獎 ‘to award’, jia 嫁 ‘to marry off one’s daughter’, jiao 交 ‘to handover’, rang 讓 ‘to yield’, jiao 教 ‘to teach’, fen 分 ‘to share’, pei 賠 ‘to compensate’, tui 退 ‘to return’, shu 輸 ‘to lose’, bu 補 ‘to make up for’, zeng 贈 ‘to give as a gift’, ci 賜 ‘to bestow’, chuan 傳 ‘to pass’, xian 獻 ‘to present’, zhao(qian) 找(錢) ‘to give change’, sai 塞 ‘to tuck’, xu(nüer) 許(女兒) ‘to betroth one’s own daughter to’, jie 借 ‘to lend’, zu 租 ‘to rent out’, fenpei 分配 ‘to allocate’, yichuan 遺 傳 ‘to pass on to offspring’, chuanran 傳染 ‘to infect’, guoji 過繼 ‘to let someone adopt his/her own child’, zhuanjiao 轉交 ‘to pass on’, yijiao 移交 ‘to handover’, jiaohuan 交還 ‘to return’, guihuan 歸還 ‘to return’, tuihuan 退還 ‘to return’, zengsong 贈送 ‘to give’, zhuansong 轉送 ‘to pass on’, zhuanmai 轉賣 ‘to resell’, gaosong 告送 ‘to tell’ ji 寄 ‘to send’, hui 匯 ‘to remit’, da(dianhua) 打(電話) ‘to give a call’, fa 發 ‘to send out’, jieshao 介紹 ‘to introduce’, tuijian 推薦 ‘to recommend’, xie(xin) 寫(信) ‘to write a letter’,176 shao 捎 ‘to take something to someone’
The compatibility of these two groups of verbs with Zhu’s four patterns is summarized in Table 21. Table 21. Compatibility of Va and Va/c verbs with Zhu’s four sentence patterns Verb type
IO DO
V給 IO DO Va Va/c
+ +
給 IO V DO +
V IO DO + -
DO IO
V DO 給 IO + +
174. The major difference between these two sub-groups is that the former can be expressed with the V IO DO pattern while the latter cannot. 175. See Zhu (1979), pp. 81, 82 and 85. We follow Yue-Hashimoto’s scheme (personal communication) and discard the following verbs, which are not considered double-object verbs as the recipient is not an obligatory argument for these verbs: ti 踢 ‘to kick’, reng 扔 ‘to throw’, jian 搛 ‘to pick with chopsticks’, yao 舀 ‘to ladle out’, liu 留 ‘to leave’, dai 帶 ‘to bring’.
176. The verb xie 寫 ‘to write’ is not always a double-object verb, according to Zhu (1983). It has the [+GIVE] semantic feature only when the object is xin 信 ‘a letter’ because writing a letter presupposes a recipient. Other objects such as xiaoshuo 小說 ‘a novel’ do not necessarily specify a recipient.
112 Cantonese give and Double-Object Construction
There are two observations from Zhu’s analysis.
a. Both Va and Va/c can occur with the V DO 給 IO pattern, which is the native pattern for the give-type double-object construction in the Southern dialects. In other words, in Zhu’s scheme, Beijing Mandarin possesses both word order types: DO IO and IO DO. b. The Va verbs cannot be expressed with the 給 IO V DO pattern. To be precise, what Zhu claimed is that the 給 IO V DO pattern with Va verbs is not a double-object construction but a beneficiary construction. In the following, we will examine the above two phenomena with corpus data and fieldwork data. 5.4.1 The V DO 給 IO pattern in Beijing Mandarin On the basis of some corpus data of Beijing Mandarin,177 J. Yan (2005) identified eight patterns for the double-object construction, which can be generalized into two patterns bearing typological significance: Pattern 1 has the IO appearing before the DO while Pattern 2 has the reverse. Their frequency distribution is summarized in Table 22.178 Table 22. Frequency distribution of IO DO and DO IO patterns in Yan’s corpus of Beijing Mandarin Pattern Pattern 1: V + 給 + IO + DO Pattern 2: V + DO + 給 + IO
Frequency 245 88
The ratio between the IO DO pattern and the DO IO pattern is around 3:1. This shows that the IO DO pattern is used much more frequently than the DO IO pattern. However, Yan’s study did not provide the frequency distribution of other patterns such as 給 IO V DO and V IO DO. To further investigate this issue, we built a corpus of the writing of Wang Shuo 王朔, whose mother dialect is Beijing 177. The corpus was based on 1,645 texts maintained at the Institute of Linguistics, the Peking University.
178. J. Yan (2005) distinguished V DO 給 IO and V DO 給 IO VP as two patterns, but we combine them as one type as our focus is on the relative order of the two objects. The same also applies for the patterns of V 給 IO DO VP and V 給 IO DO. Furthermore, we excluded the constructions involving ba 把 and preposed DOs such as zhe ben shu wo song gei ta 這本書我 送給他 ‘I gave him/her this book’ because there is no alternate pattern to illustrate the relative word order of the two objects.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 113
Mandarin.179 The frequency distribution of the various double-object patterns is listed in Table 23.180 Table 23. Frequency distribution of the four double-object patterns in Wang Shuo’s corpus Typological pattern IO DO
DO IO Total
Sentence pattern
Frequency
V 給 IO DO
28
V IO DO
305
給 IO V DO V DO 給 IO
233
566
8
574
Out of the 574 double-object sentences found in the corpus, only 8 are expressed with the DO IO pattern. The ratio between the IO DO and the DO IO patterns is around 70:1. The eight sentences with the DO IO pattern are listed in Table 24. Table 24. The V DO 給 IO pattern in Wang Shuo’s corpus
Verb
打電話 ‘to give a call’
Sentence
他很熟悉老師們打電話給他時使用的口氣和措辭 ‘He was so familiar with the tones and wordings when the teachers gave him a call’ 打電話 ‘to give a call’ 我打電話給老紀 ‘I gave Ji a call’ 打電話 ‘to give a call’ 就打電話給小魯 ‘Then, [someone] gave Lu a call’ 打電話 ‘to give a call’ 就打電話給許立宇的車隊 ‘Then, [someone] gave a call to the motorcade of Xu Liyu’ 借 ‘to lend’ 你當這是中國借錢給越南打美國佬 ‘You take this as if China lends money to Vietnam for fighting with America’ 透風 ‘to divulge a secret’ 你不該透風給他們 ‘You should not divulge the secret to them’ 傳授 ‘to teach’ 她也一定傳授了一些基礎知識給方槍槍 ‘She must have passed on some basic knowledge to Fang Qiangqiang’ 讓 ‘to give/yield’ 就這樣讓位給文化精華論 ‘[Someone] then yielded the seat to the essence of culture’
179. Wang Shuo is one of the famous contemporary writers in China. He was born in Liaoning in 1958 and was brought up in Beijing. His writing is based on the Beijing dialect. The writings of Wang Shuo are downloaded from: http://www.shuku.net/novels/wangshuo/wangshuo.html. The corpus has about 1.8 million characters. 180. We excluded those cases in which the double-object construction is used in a relative clause because they cannot reveal the word order between the DO and the IO. In other words, sentences like wo song gei ni de na ben shu hen hao kan 我送給你的那本書很好看 ‘the book I gave you is very good’ were not analyzed.
114 Cantonese give and Double-Object Construction
Among these five verbs, only two (chuanshou 傳授 ‘to pass on’ and toufeng 透風 ‘to divulge the secret’) are used exclusively with the DO IO pattern. The other three (jie 借 ‘to lend’, da dianhua 打電話 ‘to give a call’ and rang 讓 ‘to give/yield’) are also used with the other three patterns.181 From Table 23, we observe that the V IO DO pattern is used the most when compared with the other two IO DO patterns. However, it should be noted that V IO DO is the only pattern with which the main verb gei 給 is expressed in Beijing Mandarin.182 After excluding those sentences with gei 給 as the main verb, the V IO DO pattern only occurs 44 times.183 The ratio between the IO DO and the DO IO pattern becomes 38:1 and the contrast still remains significant. This corpus study thus provides us with additional evidence to argue that the DO IO pattern should not be regarded as a native double-object pattern in the Northern dialects.184 It is possibly a pattern borrowed from the Southern dialects due to inter-dialectal contact. 5.4.2 The 給 IO V DO pattern: A beneficiary or a double-object construction?
One of the controversies arising from the 給 IO V DO pattern is the interpretation of the semantic role of the noun phrase after gei 給. In his analysis, Zhu claimed that the post-gei noun phrases in those sentences with the V[+GIVE] type verbs only have
181. Jie 借 occurs the most with the V IO DO pattern (with a total frequency of 7). The 給 IO V DO pattern is used the most (75 times) with da dianhua 打電話. Rang 讓 is used twice with the V IO DO pattern and five times with the 給 IO V DO pattern. The frequency distribution of the double-object verbs found in Wang Shuo’s corpus with respect to the four patterns is listed in Appendix 3. 182. In some Mandarin dialects such as those in the Northwestern sub-group, the IO marker gei 給 and the main verb gei 給 can co-occur.
183. The following verbs are also used more frequently with the V IO DO pattern: song 送 ‘to give’, jie 借 ‘to lend’, huan 還 ‘to return’, pei 賠 ‘to compensate’, jiyu 給予 ‘to give’, jiao 交 ‘to handover’ and jiang 獎 ‘to award’.
184. According to M. Shen (2002), in the dialect of Taiyuan (太原), a dialect of the Zhongyuan Mandarin sub-group, spoken in Shanxi province, the double-object sentences can be expressed with either the V IO DO pattern or the V 給 IO DO pattern but not the V DO 給 IO pattern. In other words, the Taiyuan dialect only has song ta shu 送他書 ‘give him a book’, song gei ta shu 送 給他書 but not song shu gei ta 送書給他.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 115
the semantic role of beneficiary.185 In other words, in the sentence wo gei Zhang San song yi ben shu 我給張三送一本書, the recipient of the book is someone other than Zhang San (i.e., ‘I gave a book to someone on behalf of/for Zhang San’).186 As Wang Shuo’s corpus data show, the 給 IO V DO pattern is used much more than the other two IO DO patterns.187 In this regard, we conducted a fieldwork survey with six informants of different Mandarin dialects. The native places of the six Mandarin informants are Liaoning, Beijing, Shandong, Yunnan, Taiwan and Hubei.188 The results of the fieldwork survey showed that the scenario is far more complex. In some cases, NP1 in the 給 NP1 V NP2 pattern is interpreted as a beneficiary while in other cases, it can be either a recipient or a beneficiary or even both. All the sentences used in the fieldwork survey were expressed with the 給 NP1 V NP2 pattern and the informants were asked to give their interpretation of the semantic role of NP1, whether it is a beneficiary (B), a recipient (R) or both (B+R).189 The result is summarized in Table 25.
185. Zhu (1979) distinguished two types of gei 給 before the IO. The one marking a beneficiary is a preposition while the one marking a recipient is a verb. He argued that the sentence gei wo song ben shu 給我送本書 is ungrammatical if gei 給 is interpreted as a verb (i.e., wo 我 as a recipient).
186. C. Li and Thompson (1981), however, argued that the preverbal gei-phrase is only indicating that “Mandarin has been moving toward a preferred word order in which noun phrases occur before rather than after the verb” (p. 386). Therefore, in their view, the gei-phrase is only fronted due to word order change in Chinese and Zhang San in this sentence is still a recipient.
187. After excluding those V IO DO sentences with the verb gei 給, the frequencies of the patterns of V 給 IO DO, V IO DO and 給 IO V DO are 28, 44 and 233 respectively.
188. The fieldwork survey was conducted in early 2008. The Beijing informant was born in Shanghai but raised in Beijing. She could hardly speak any dialect of Shanghai except a few phrases. The Shandong informant was a 40-year-old Chinese language teacher in the US. She was raised and received education in Beijing. The Liaoning informant studied in Beijing for some years, and she then did her PhD in the US. The Yunnan informant was 32 years old and was a post-doctoral researcher [in the engineering field] in Hong Kong. The Taiwan informant did not know Southern Min and spoke only Taiwan Mandarin. The Hubei informant had stayed in Wuhan for 15 years. 189. Sometimes, it is not easy to distinguish the semantic roles of beneficiary and recipient, especially when the dialects introduce the two roles with the same morpheme. According to the survey of X. Huang and He (2018), Mandarin dialects (except those of the Southwestern and the Jianghuai sub-groups) use gei 給 as the markers of beneficiary and recipient.
116 Cantonese give and Double-Object Construction
Table 25. Semantic role of NP1 in the 給 NP1 V NP2 pattern190191
Informants from the North LN190
(a) Va type verb B+R191 小張給小王借了三十萬 R 小張給小王遞了一枝筆 R 小張給小王送了一份禮物 B/R 小張給小王獎了二十塊錢 B/R 小張給小王分工資 R 小張給小王補了這個月的工資 B/R 小張給小王傳了一個資訊 R 小張給小王找了三塊錢 R 小張給小王塞了一顆糖 R 小張給小王傳真了一份文件 R 小張給小王分配了一個工作間 B 小張給小王賣了一部車子 B/R 小張給小王賠了六十塊 B 小張給小王租了一個房子 B 小張給小王退了支票 B 小張給小王還了三本書 B 小張給小王還了錢 B/R 小張給小王付了五十塊 B 小張給小王嫁女兒 B/R 小張給小王交了申請書 B 小張給小王轉交了一份檔 B 小張給小王轉送了一份禮物 B 小張給小王轉賣了房子 B 小張給小王交還了三本書 B 小張給小王退還了三件衣服 ba 小張給小王遺傳了一個怪病 ba 小張給小王傳染了一個怪病 --小張給小王讓了他的車子 B 小張給小王過繼了他的女兒
Informants from the South
BJ
SD
TW
YN
HB
B+R R R R B/R R B/R R R B/R R B B/R B B B B/R B/R B B B/R B/R B/R B B ba R ba R
B R R
B+R R R R R R B/R R R R R B B B B R B R B B B B B B B
R R R R R R R R R R R B B B B B B/R R B B R B/R B B B ba R R ba
R R R R R R R R R R R B R B B B B B B B B B B B B ba ba ba R
V-給 --R R R R R R B B B B B B/R B ba B ba B B B B bei bei ba R
V-給 V-給 ba B
190. LN – Liaoning; BJ – Beijing; SD – Shandong; TW – Taiwan; YN – Yunnan and HB – Hubei. 191. B/R: NP1 can be interpreted as either a beneficiary or a recipient. B+R: NP1 at the same time functions as a beneficiary and a recipient. Bei: The sentence has a passive meaning and NP1 is a source rather than a recipient or a beneficiary. Ba: The informant preferred the sentence to be expressed with a ba-construction. ‘V-給’: The informant preferred the sentence to be expressed with the V 給 NP1 NP2 pattern. ‘---’: No data was given by the informant.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 117
Table 25. (continued)
(b) Va/c verb 小張給小王寫了一封信 小張給小王寫了一個便條 小張給小王寄了一封信 小張給小王發了這個月的工資 小張給小王匯了兩千塊 小張給小王介紹了一個朋友 小張給小王推薦了一個秘書 小張給小王打電話
Informants from the North LN190
BJ
SD
R R B/R R R R R R
R R B/R R R R R R
R R B/R R B/R R R R
Informants from the South
TW
YN
HB
R R R R R R R R
R R R R R R R R
R R R R R B/R R R
Below are some observations from the fieldwork: a. The responses given by the six informants of different Mandarin dialects are quite consistent. The interpretation of NP1 does not differ significantly between the Mandarin dialects in the North and the South; b. Within the same type of verbs (in particular the Va type), there are different interpretations toward the semantic role of NP1. For example, NP1 of the following verbs was understood as a beneficiary rather than a recipient by the six informants: mai 賣 ‘to sell’, pei 賠 ‘to compensate’, zu 租 ‘to rent out’, tui 退 ‘to return’, huan 還 ‘to return’, fu 付 ‘to pay’, jia 嫁 ‘to marry off one’s daughter’, jiao 交 ‘to handover’, zhuanjiao轉交 ‘to pass on’, zhuansong 轉送 ‘to pass on’, zhuanmai 轉賣 ‘to resell’, jiaohuan 交還 ‘to return’, tuihuan 退還 ‘to return’.192 For the remaining 11 Va verbs such as jie 借 ‘to lend’, di 遞 ‘to pass’, song 送 ‘to give’, jiang 獎 ‘to reward’, fen 分 ‘to share’, bu 補 ‘to make up for’, chuan 傳 ‘to pass’, zhao 找 ‘to give change’, sai 塞 ‘to tuck’, chuanzhen 傳真 ‘to fax’, fenpei 分配 ‘to allocate’, NP1 was interpreted as a recipient. Zhu’s claim that NP1 has the semantic role of a beneficiary is only partially correct according to the data collected from this fieldwork.193 192. According to M. Shen (2002), in the Taiyuan dialect, NP1 of fa 發 ‘to distribute’, mai 賣 ‘to sell’, huan 還 ‘to return’, tui 退 ‘to return’, zhuan 轉 ‘to pass on’, jiao 交 ‘to handover’ is also interpreted as beneficiary only. Shen only listed seven verbs with this interpretation but based on his discussion, there should be more than seven verbs with the same interpretation.
193. In the Taiyuan dialect, the noun phrase after gei 給 for some V[+GIVE] type verbs can also be interpreted as a recipient. These verbs include fen 分 ‘to share’, pei 配 ‘to allocate’, rang 讓 ‘to yield’, pei 賠 ‘to compensate’, sai 塞 ‘to tuck’ (M. Shen, 2002).
118 Cantonese give and Double-Object Construction
c. For the verb jie 借 ‘to lend’, three informants (LN, BJ and TW) interpreted NP1 as ‘B+R’, which means that NP1 can have both the roles of a recipient and a beneficiary. As pointed out by M. Shen (2002), in the Taiyuan dialect, Xiao Zhang gei wo jie le yi ben shu 小張給我借了一本書 has the meaning of ‘Xiao Zhang borrowed a book from someone for me and then lent it to me’. The informants of Beijing, Liaoning and Taiwan share the same interpretation with the Taiyuan dialect.194
What the fieldwork shows is that for the same sentence pattern (i.e., 給 IO V DO) involving the same type of verbs (i.e., with the [+GIVE] semantic feature), the semantic role of NP1 is not always the same. As for the Va verbs, there is a differential degree of the obligatory use of a recipient with these verbs. According to the six informants, NP1 occurring with some Va verbs such as song 送 ‘to give’ and di 遞 ‘to pass’ has to be interpreted as a recipient. This suggests that the 給-NP1 phrase is obligatory for these verbs. In other words, it is ungrammatical to have *wo song le yi ben shu 我送了一本書 ‘*I gave a book’ without the IO. For those verbs such as mai 賣 ‘to sell’, fu 付 ‘to pay’, zu 租 ‘to rent out’, tui 退 ‘to return’, it is grammatical to have sentences without the IOs, such as wo jintian mai le san bu chezi 我今天 賣了三部車子 ‘I sold three cars today’, wo tui le san tiao kuzi 我退了三條褲子 ‘I returned three pairs of trousers’ and wo gangcai fu le wushi kuai 我剛才付了 五十塊 ‘I have just paid 50 dollars’. The IO, when present, can be interpreted as a beneficiary by some informants because it is less obligatory in the double-object construction. 5.4.3 Summary
With the aid of corpus data and fieldwork data, we have shown that the V DO 給 IO pattern is not a native double-object pattern in the Northern dialects. Zhu’s inclusion of this pattern in his analysis was possibly due to the influence of his dialect since he was a native of Suzhou. Furthermore, the 給 IO V DO pattern is used the most frequently and the semantic role of the IO is not necessarily a beneficiary. The actual situation is rather complex as the semantic role of NP1 between a recipient and a beneficiary varies from verb to verb. To summarize, in Chinese dialects, in terms of the relative word order of the two objects, there are two typological patterns for the double-object construction: 194. Other verbs that have this interpretation in the Taiyuan dialect include zu 租 ‘to rent out’ and zhao 找 ‘to give change’. NP1 for zu 租 was interpreted by the three speakers (LN, BJ and TW) as a beneficiary.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 119
DO IO and IO DO.195 The sentence I gave him a book can be rendered with the following three patterns in the Northern dialects ((108) – (110)) while in the Southern dialects such as Cantonese, only (111) is found. (108) 我 送 他 一 本 書 wo song ta yi ben shu I give he one cl book
(109) 我 送 給 他 一 本 書 wo song gei ta yi ben shu I give give he one cl book (110) 我 給 他 送 一 本 書 wo gei ta song yi ben shu I give he give one cl book
(111) 我 送 一 本 書 畀 佢 ŋͻ13 sʊŋ33 jɐt5 pun35 sy55 pei35 k’ɵy13 I give one cl book give s/he
5.5
Double-object construction in the Yue dialects
As a member of the Southern dialects, the Yue dialects also have the DO IO word order as the native pattern of the give-type double-object construction. Thus, the double-object construction is another frequently cited example to illustrate typological differences between standard Mandarin and the Yue dialects.196 Below are some examples from different sub-groups of the Yue dialects. The sub-grouping of the Yue dialects is based on A. Yue (2006b).197
195. In his book-length discussion of the alignment typology of double-object constructions in Chinese dialects, M. Zhang (2010) argued that the contrast of IO DO and DO IO patterns between the Northern and the Southern dialects is related to the nature of the GIVE verbs. Specifically, Zhang claimed that in the Southern dialects, the GIVE verbs were once monotransitive verbs carrying the meaning of hold or take (see also the discussion of the Gan dialects by X. Li (2015)). This nature of the double-object verbs leads to the placement of the recipient (i.e., the IO) toward the end of the sentence. 196. See, for example, B. Huang (1959), Cheung (1972), H. Gao (1980), J. Huang and Zhan (1983), J. Yuan (2001). 197. The data of Shunde, Nanhai, Huazhou and Yangchun are based on the unpublished field notes of Professor Anne Yue-Hashimoto.
120 Cantonese give and Double-Object Construction
(a) The Guangfu sub-group (廣府片) Hong Kong (香港) Guangzhou (廣州) (B. Huang, 1959, p. 276) Foshan (佛山) (B. Li & He, 1992, p. 59) Zengcheng (增城) (W. T. He, 1993, p. 209) Nanning (南寧) (Y. Lin & Qin, 2008, p. 351)
你畀啲錢佢 ‘Give some money to him/her’ 你送枝筆佢 ‘Give a book to him/her’ 畀本書我 ‘Give a book to me’ 畀本書我 ‘Give a book to me’ 今日下午還部車畀陳老師嘅大佬哂 ‘Return the car to the brother of Teacher Chen this afternoon’
(b) The North Delta sub-group (北三角洲片) Nanhai (南海) 我畀住三個銀錢你先 ‘I gave three dollars to you first’ Shunde (順德) 你畀住三文錢我tʃᴀ35‘You gave three dollars to me first’ (c) The South Delta sub-group (南三角洲片) Dongguan (東莞) (X. Chen, 1993, p. 189) 畀本書阿明 ‘Give a book to Ming’ Huazhou (化州) 畀一半你啦 ‘Give a half to you’ Lianzhou (廉州) (X. Chen & Chen, 2005, hɐi44本書渠 ‘Give a book to him/her’ p. 417) (d) The Liang Yang sub-group (兩陽片) Yangjiang (陽江) (Huang, 1966, pp. 217–218) 其畀三部書你 ‘S/he gave you three books’ Yangchun (陽春) 我畀住三文你先 ‘I gave you three dollars first’
(e) The Wu Yi sub-group (五邑片) Taishan (台山) (Yue-Hashimoto, 2005, p. 225) 我畀半你 ‘I gave you a half ’ 佢交轉乃錢畀你 ‘S/he passed the money to you’ Kaiping (開平) (Deng, 2000, p. 91) 借兩本書畀你 ‘Lend you two books’
The above data show that DO IO is the native word order for the double-object construction in the Yue dialects.198 The same typological observation for the double-object pattern is also found in the pre-modern Cantonese materials. The frequency ratio between the patterns of IO DO and DO IO is 1:6.5. The distribution of the two patterns is summarized in Table 26.199
198. Y. Shi (2005) argued that the V DO IO (without the IO marker) pattern is only found when the IO is a pronoun. He then accounted for this word order by means of phonological stress (i.e., how the IO DO pattern is transformed into the DO IO pattern when the IO is a pronoun). However, the Yue dialect examples show that this is not the case. Even if the IO is a noun phrase (as in the example of the Dongguan dialect), the word order is still DO IO. 199. Sentences with the IO DO word order found in the pre-modern Cantonese materials are listed in Appendix 4.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 121
Table 26. Distribution of IO DO and DO IO patterns in pre-modern Cantonese materials
200 fn201201
Author
Year
Morrison Bridgman Devans Hobson Bonney Bunyan Luke Dennys Happer Anonymous Ball Fulton Stedman & Lee Mai201 Mai Ball Wisner LeBlanc Ball Cowles Ball Caysac Wisner Luke Wells Brouner & Fung Hoh & Belt O’Meila Wells Chao Chiang
1828 1841 1847 1850 1853 1871 1873 1874 1874 1877 1888 1888 1888 1893a 1893b 1894 1906a,b 1910 1912 1920 1924 1926 1927 1931 1931 1935 1936 1941 1941 1947 1949
IO DO200 2 3 0 0 2 1 0 0 0 0 0 0 11 77 57 0 0 4 0 0 0 6 0 4 0 0 0 0 0 0 0
DO IO 3 40 3 5 13 22 69 68 2 2 16 13 21 43 23 11 34 8 31 21 18 62 33 15 34 32 7 131 30 2 5
(continued) 200. We focus mainly on the word order rather than the choice of the IO marker. Thus, we subsume the structures of V DO [過/畀/嚟/zero marker] IO under the DO IO pattern. Similarly, patterns with V [畀/過] IO DO are considered members of the IO DO pattern. Furthermore, we excluded those sentences with the preposed DO marked by [tsœŋ55] 將 or [pa35] 把 since they cannot reveal the relative word order of the two objects. 201. Notice that in Takashima and Yue’s study (2000, p. 40), there is no record for the usage of the IO DO pattern for Mai’s annotation of the Book of Odes. In a later study, A. Yue (2000) found seven instances of the IO DO pattern.
122 Cantonese give and Double-Object Construction
Table 26. (continued) Author
Year
Chan Chiang Oakley Whitaker Bruce Whitaker Wu Yuan Lau Chapman Luke Total
1951 1951 1953 1954 1954 1959 1960 1960 1972 1973 1997
IO DO200 0 1 4 0 0 0 0 0 1 0 9 182
DO IO 28 12 18 7 38 26 10 9 186 15 27 1193
The IO DO pattern was frequently used in the two texts by Mai Shizhi. When examining Mai’s works, we can find a lot of examples of literary language,202 probably due to Mai’s proficiency in the classical language. If we re-calculate the distribution of these two word order patterns without Mai’s two texts (1893a, 1893b), the ratio of the IO DO and the DO IO patterns becomes 1:24. 5.6
Summary
In this chapter, we examined the typological contrast between the Northern and the Southern dialects in terms of the word order of the DO and the IO in the give-type double-object construction. Based on our fieldwork data, corpus data of Wang Shuo and other published materials on Chinese dialects, the native patterns for the Northern and the Southern dialects are IO DO and DO IO respectively.203 Due to the increasing contact with the standard language, Putonghua and standard Mandarin, the IO DO pattern is now used by speakers of the Southern 202. For example, in the annotation of the Book of Odes, Mai (1893a) used zai 在 instead of [hɐi35] 喺 as the locative verb or the locative particle: zai shuimian shang 在水面上 ‘on the water surface’ and zai tiao he zhongjian 在條河中間 ‘in the middle of the river’.
203. The dominance of the DO IO pattern for the double-object construction is also found in the neighboring non-Sinitic languages such as Zhuang, Miao, Thai, and Gelao (Matthews, 2006; Y. Xu, 2015). This further raises the question whether the exclusive use of this word order type in these genetically unrelated languages in the Southeast Asian context constitutes another areal feature.
Chapter 5. Word order typology of the double-object construction in Chinese dialects 123
dialects. In his sociolinguistic study of pre-verbal and post-verbal gei-phrases in the double-object construction of Chinese dialects, Peng (2020) observed that “[n]ortherners showed significantly higher preference for the pre-verbal gei-phrase than southerners” (p. 349). At the same time, “[t]he post-verbal gei-phrase becomes the dominant variant in the southernmost dialect areas” (p. 349). At the same time, Peng noted that the usage of the pre-verbal gei-phrase in Hong Kong Cantonese is 30%. How can we account for this 30% non-native usage? There have been various proposals discussing the relationship between the DO IO and the IO DO patterns. In the next chapter, we will evaluate these proposals and argue that they cannot convincingly account for the situation. This means that there are other non-linguistic factors that we need to explore to account for the formation of syntactic strata.
Chapter 6
Relationship between IO DO and DO IO patterns
In Chapter 5, we discussed that the two double-object patterns, IO DO and DO IO, have typological significance in Chinese dialects. In fact, some world languages have both word order patterns. English is one of the examples. There are languages that only have either the DO IO or the IO DO pattern. The former includes most African languages and Southeast Asian languages while the latter is mainly found in the languages of Eurasia and South America.204 In view of the co-existence of the two word order patterns in English, different proposals have been put forth to explain their relationships and to compare their differences. Some linguists adopted these proposals to standard Mandarin (as well as Cantonese, see Section 6.1.1) because they considered that standard Mandarin also possesses both patterns. In this chapter, we will review these proposals under the transformational, discourse and cognitive approaches. It should be stressed that the adoption of any of these proposals has an assumption that the language in question has both word order patterns as the native patterns. However, our data of Wang Shuo’s writings and native Mandarin speakers disprove the fact that the DO IO pattern is a native pattern in standard Mandarin. We suggest that there are extra-linguistic factors such as language contact contributing to the co-existence of typologically distinct patterns in a dialect. For example, the IO DO pattern in Cantonese is a loan form spreading from Putonghua or standard Mandarin. Takashima and Yue (2000) further claimed that this kind of linguistic change is “implemented to various degrees and at different speeds by different speakers, beginning with the young and the educated” (p. 44). This will be further discussed in Section 6.4.
204. For the distribution of these two double-object patterns across world’s languages, see Michaelis and Haspelmath (2003). According to Michaelis and Haspelmath (2003), standard Mandarin is a mixed-type language in which both word order patterns co-exist. Their classification of the double-object construction in standard Mandarin is based on the data in C. Li and Thompson (1981).
126 Cantonese give and Double-Object Construction
6.1
Transformational approach
The transformational approach argues that one of the patterns is the deep structure, which can derive the other by means of transformation (see, for example, Chomsky, 1985; Larson, 1988). Linguists adopting the transformational approach usually claim that the IO DO pattern is derived from the DO IO pattern and this process is known as dative shift or dative alternation,205 which was proposed by Chomsky (1985, pp. 492–493), who argued that in the formation of a wh-question on the IO, the wh-word has to be preceded by the preposition to as in ‘To whom did the teacher give the book?’ instead of ‘*Whom did the teacher give the book?’. On this basis, Chomsky claimed that in the underlying structure, the IO is preceded by the preposition to and the DO IO pattern is thus taken as the deep structure while the IO DO pattern is the surface structure resulted from transformation. 6.1.1 Does Cantonese have dative shift? S. Tang (2003) and Peyraube (1981) analyzed the double-object construction in Hong Kong Cantonese under the transformational approach. However, their analyses are not the same, which reflect the different assumptions they took: Peyrabue considered the IO DO pattern a native pattern in Cantonese while Tang does not. 6.1.1.1 S. Tang (2003): Lack of dative shift in Cantonese If the DO IO pattern is the underlying form for the transformation, then we need to explain why the DO IO pattern in Cantonese cannot derive the IO DO pattern. S. Tang (2003) argued that there exists a Functional Phrase (FP) for the derivation of the IO DO pattern in standard Mandarin, under which there is a semantic feature of possession triggering the transformation.206 Furthermore, the IO DO pattern carries the causative meaning. Therefore, Zhang San song Li Si yi ben shu 張三送 李四一本書 ‘Zhang San gave Li Si a book’ has a deeper meaning of ‘Zhang San caused Li Si to possess a book’. S. Tang (2003) further claimed that Cantonese lacks this semantic feature in the FP and thus cannot derive the IO DO pattern. In other words, both standard Mandarin and Cantonese have the same transformational rule, but the realization of this rule is subject to the presence of the semantic feature of possession under the FP. 205. There are studies claiming the opposite, i.e., the IO DO pattern is the basic structure. See Bryun, et al. (1999), A. Li (1990). However, the majority still holds the view that the DO IO pattern is the underlying form. 206. S. Tang’s (2003) proposal is based on the analysis of Larson (1988).
Chapter 6. Relationship between IO DO and DO IO patterns 127
There are two issues for Tang’s proposal: a. If the IO DO pattern carries the meaning of causing someone to possess something, then lacking such a pattern in a language would consequently imply that the double-object construction in the language concerned does not have such a meaning or the DO IO pattern does not carry the meaning of causation. However, it is clear that the Cantonese sentence [ŋͻ13 sʊŋ33 tsͻ35 pun35 sy55 pei35 lei13 sei33] 我送咗本書畀李四 ‘I gave a book to Li Si’ has the underlying meaning of ‘I did something (i.e., giving a book) which caused someone (i.e., Li Si) to possess something (i.e., the book)’. In other words, the meaning of causative is also present in the DO IO pattern. b. If Cantonese lacks the semantic feature of ‘cause to possess’ to trigger the transformation to derive the IO DO pattern, then how can we explain the situation in which the IO DO pattern is found in contemporary Cantonese, as observed by Peyraube (1981)?207 6.1.1.2 Peyraube (1981): Dative shift in Cantonese In his fieldwork conducted in the 1980s, Peyraube (1981) obtained the following four patterns of the double-object construction from his informants, who were college students at that time. a. V DO IO b. V DO 畀 IO c. V 畀 IO DO d. V IO DO (a) and (b) have the DO IO word order while (c) and (d) have the IO DO word order.208 The semantic constraint of Heavy End Shift209 proposed by Matthews and Yip (1994) in Section 6.2.2 cannot convincingly explain the use of the IO DO pattern in the following three sentences (with Peyraube’s transcription in Yale) because the DOs are not long (Peyraube, 1981, pp. 36–38). 207. Notice that S. Tang (1998) considered 我畀咗佢一枝筆 ‘I gave him/her a pen’ unnatural instead of ungrammatical. 208. Although Peyraube followed the classificatory scheme in Zhu (1979), the four patterns listed here are not totally identical to the ones of Zhu. For example, Zhu (1979) did not have the V DO IO pattern while Peyraube (1981) did not have the 畀 IO V DO pattern corresponding to Zhu’s 給 IO V DO. We will discuss in Section 7.3 why the 畀 IO V DO pattern is not used by Cantonese speakers. 209. Matthews and Yip (1994) suggested that the DO is placed at the end of the sentence if it is long.
128 Cantonese give and Double-Object Construction
(112) 我 送 你 呢 本 書 ngóh sung néih nı̄̀ bún syù I give you this cl book ‘I give you this book’ (113) 我 畀 佢 一 本 書 ngóh béi kéuih yāt bún syù I give s/he one cl book ‘I give him/her a book’
(114) 我 送 畀 你 呢 本 書 ngóh sung béi néih nī bún syù I give give you this cl book ‘I give you this book’
Although Peyraube (1981) admitted that the IO DO pattern “is much less natural than the inverse order [i.e., V DO (畀) IO], and one feels that it has been influenced by Northern dialects” (p. 35), he claimed that the transformational rule can still apply.210 Specifically, he argued that although “it is likely that, during recent decades, Cantonese has been influenced by the ‘official language’ [i.e., Putonghua] which is based on Northern dialects. Nonetheless, the IO + DO order must have already existed in the past as well” (p. 48). In other words, Peyraube considered the IO DO pattern collected in his fieldwork one of the native patterns in Cantonese.211 Peyraube (1981) classified the Cantonese double-object verbs into seven classes according to four criteria, among which three are related to the syntactic behavior of the verbs. The four classificatory criteria for the seven types of verbs include the following syntactic and semantic behaviors: a. syntactic patterns (i.e., the compatibility of the verbs with respect to the above four patterns); b. nature of the two objects (DO and IO): this is related to the length and definiteness of the objects; c. whether the objects (DO & IO) of the verb in question allow topicalization, relativization and clefting; d. the addition of a supplementary verb or a possessive marker such as [t’ɐi35] 睇 ‘to read’ in the sentence 我送本書畀你睇 ‘I gave you a book to read’. 210. Peyraube (1981) adopted the IO Movement proposed in Yue-Hashimoto (1971), which is also based on the transformational grammar framework. 211. Notice that Peyraube’s analysis followed the classificatory scheme outlined by Zhu (1979) except the 畀 IO V DO pattern. Thus, he assumed that Cantonese also shares with standard Mandarin the same set of syntactic patterns for the double-object construction.
Chapter 6. Relationship between IO DO and DO IO patterns 129
It seems difficult to generalize the common semantic properties shared by the verbs in each class. We thus incline to argue that what Peyraube observed in fact reflected on-going linguistic changes that had been taking place among his teenage informants in the 1980s. Different double-object verbs exhibit different behaviors toward the same syntactic structures because the rate of change varies from verb to verb and from speaker to speaker (cf. W. Wang, 1969). 6.2 Discourse approach Despite English has both the IO DO and the DO IO patterns, there are some double-object sentences which cannot undergo dative shift. For example, sentence (115) cannot derive sentence (116) (Erteschik-Shir, 1979, p. 452). (115) John gave it to Mary. (116) *John gave Mary it.
Erteschik-shir (1979) argued that there is no transformational relationship between the two patterns. Whether the IO or the DO comes first in a double-object sentence depends on the discourse concept dominance which is defined by Erteschik-Shir (1979) as follows: “A constituent C of a sentence S is dominant in S if and only if the speaker intends to direct the attention of his hearers to the intention of C by uttering S” (p. 443). This approach suggests that the relative ordering of the DO and the IO is determined by the information value of the objects. Objects with new information or focus tend to be placed toward the end of the sentence while known or old information pre-verbally.212 Under this approach, either the DO or the IO can be placed at the end of the sentence, which entirely depends on the speaker’s intention and the context. Pronouns and noun phrases headed by the definite article the or demonstratives this or that are less dominant because the noun phrases they modify are definite whose references have already been identified in the previous discourse. In other words, they are not the focus or new information and thus lack dominance, which tend not to be placed at the end of the sentence.213 The choice between the IO DO and the DO IO patterns relies on the dominance of the two objects, which is formulated by Erteschik-Shir (1979): “In the structure … V NP1 NP2 … NP1 is non-dominant and NP2 is dominant” (italics original) (p. 449). 212. Major references on this approach include S. Thompson (1995), C. Li and Thompson (1981). 213. There is one situation that Erteschik-shir (1979) did not discuss. In English, when the two objects are realized by pronouns, only the DO IO pattern is allowed. It is grammatical to have “I gave it to him” but not “*I gave him it”. Under the notion of dominance, both objects are equally non-dominant, then what are the factors determining the word order in this case?
130 Cantonese give and Double-Object Construction
6.2.1 Discourse approach on Chinese double-object construction C. Li and Thompson (1981) adopted the discourse approach to account for dative shift in standard Mandarin. They argued that there is a slight functional difference between sentences in which the DO precedes the IO (i.e., the DO IO pattern), and those in which the DO follows the IO (i.e., the IO DO pattern) because they are used in different speech contexts. If the speech context is such that the DO but not the IO is a piece of information that has already been mentioned, then the order of DO IO is used. Otherwise, the order of IO DO is used. C. Li and Thompson (1981, pp. 372–373) used the following two sentences to illustrate their argument.214
(117) 我 扔 了 那 塊 肉 給 我 的 狗, 那 個 蛋糕 給 你 wo reng le na kuai rou gei wo de gou na ge dan gao gei ni give you I throw asp that cl meat give I de dog that cl cake 的 朋友 de pengyou de friend ‘I threw that piece of meat to my dog and that cake to your friend’
(118) 我 扔 了 我 的 狗 那 塊 肉, 你 的 朋友 那 wo reng le wo de gou na kuai rou ni de pengyou na I throw asp I de dog that cl meat you de friend that 個 蛋糕 ge dan gao cl cake ‘I threw to my dog that piece of meat and to your friend that cake’
According to C. Li and Thompson (1981), (117) and (118) are used in different contexts. The former is used when someone wants to know what has happened to the meat and the cake while the latter is used when someone wants to know what has been done (or given) to the dog and the friend. However, according to the six native speakers of various Mandarin dialects (Section 5.4.2), these two sentences do not sound natural. Five of them (except the one from Yunnan) indicated that it is better to express these two sentences with the ba 把 construction: wo ba na kuai rou reng gei wo de gou, ba na ge dangao reng gei wo de pengyou 我把那塊肉扔給 我的狗,把那個蛋糕扔給我的朋友. 214. Notice that in our scheme, reng 扔 is not a double-object verb.
Chapter 6. Relationship between IO DO and DO IO patterns 131
6.2.2 Discourse approach on Cantonese double-object construction Matthews and Yip (1994, p. 137) claimed that the IO comes before the DO when the length of the DO is very long.
(119) 我 畀 [你]IO [幾 千 文 同埋 一 張 機 票]DO ngóh béi léih géi chīn mān tùhng màaih yāt jeūng gēi pìu one cl air ticket I give you several thousand dollar and ‘I’ll give you a few thousand dollars plus an air ticket’
Hawkins (1994) also accounted for the preference of the IO DO order in English in terms of sentence processing. He argued that the distance between the phrases (in terms of number of words) determines the choice of the word order: A long DO in the V DO IO pattern creates a bigger gap between the main verb and the IO. The V IO DO pattern can then produce a shorter distance, which facilitates sentence comprehension. However, from the Wang Shuo’s corpus, there are cases ((120) and (121)) in which the IO is exceptionally long but is not moved to the end of the sentence.
(120) 我 給 [市 里 的 一 個 計程車 站]IO 打 過 電話 wo gei shi li de yi ge jichengche zhan da guo dianhua I give city de one cl taxi stand make asp call ‘I gave a call to the taxi station in the city’
(121) 我 給 [幾 個 同學 和 我 的 一些 朋友]IO 打 了 電話 wo gei ji ge tongxue he wo de yixie pengyou da le dianhua I give several cl student and I de some friend make asp call ‘I have given a call to my several classmates and friends’
Furthermore, it is interesting to compare (119) with Gao’s earlier description of the Yue dialect of Guangzhou, in which the relative order of the DO and IO does not change even though the length of the DO (i.e., 三本參考書同埋兩本雜誌, three reference books and two magazines) is exceptionally long. Sentence (122) is taken from H. Gao (1980, p. 220).
(122) 我 琴日 畀 咗 三 本 參考書 同埋 兩 ŋͻ13 k’ɐm11jɐt2 pεi35 tsͻ35 sa:m53 pun35 ts’a:m53ha:u35sy53 t’uŋ11ma:i11 lœŋ13 I yesterday give asp three cl reference book and two 本 雜誌 過 阿黃 pun35 tsa:p1tsi33 kwͻ33 a33 wͻŋ11–35 cl magazine pass Wang ‘Yesterday I gave three reference books and two magazines to Wang’
132 Cantonese give and Double-Object Construction
Both the DOs in (119) and (122) are long but the latter does not have any word order change between the two objects. Therefore, whether the IO DO pattern adopted in (119) is due to discourse factors deserves further investigation.215 In addition to the length of the DO, the syntactic structure of the DO may also play a role in the choice of the IO DO pattern. According to Cheung (1972, p. 83), (124) which has the IO DO pattern with a bare noun phrase as the DO, is considered ungrammatical while (123), whose DO is made up of a numeral and a classifier, is grammatical.216 (123) 佢 畀 三 本 書 我 kéuih béi saàm bún syù ngóh s/he give three cl book I ‘S/he gives three books to me’
(124) *佢 畀 我 書 kéuih béi ngóh syù book s/he give I S/he gives me a book’
Furthermore, even if the IO is less dominant or less topic-worthy, such as a pronoun, it is still placed at the end of the sentence, as shown in (125).
(125) 我 送 咗 一 本 書 (畀) 佢 13 33 35 5 35 ŋͻ sʊŋ tsͻ jɐt pun sy55 (pei35) k’ɵy13 I give asp one cl book (give) s/he ‘I gave a book to him/her’
In (125), the IO is realized by a third person singular pronoun [k’ɵy13] 佢 ‘s/he’ and the DO is an indefinite noun phrase composed of the numeral [jɐt5] 一 ‘one’ and the classifier [pun35] 本. According to Erteschik-shir (1979), the IO has less focus than the DO and should not be placed at the end of the sentence. However, it is still natural for Cantonese speakers to utter (125) instead of (126).
215. If the use of the IO DO pattern is attributed to the length of the DO, then (122) should also be expressed with the IO DO pattern. 216. However, Cheung did not give any reason for the ungrammaticality of sentence (124). What Cheung commented is that (p. 83): 「佢畀三本書我」 “ 若說成 「佢畀我三本書」 也還可以,但是 「 佢畀我書」就根本不成句子。” By comparing these two sentences, we may argue that the grammaticality is related to the structure of the DO (i.e., bare noun phrases vs. non-bare noun phrases).
Chapter 6. Relationship between IO DO and DO IO patterns 133
(126) 我 送 咗 佢 一 本 書 ŋͻ13 sʊŋ33 tsͻ35 k’ɵy13 jɐt5 pun35 sy55 I give asp s/he one cl book ‘I gave him/her a book’
Law (1996) studied the Hong Kong Cantonese double-object construction with the discourse approach. In her study, Law compared the acceptability of the various patterns of the double-object construction with a questionnaire comprising 1,080 double-object sentences.217 There were only five informants who were in their twenties in the mid-1990s.218 In terms of the syntactic analysis, Law combined both the frameworks of Zhu and Peyraube. She abandoned the seven classes of double-object verbs categorized by Peyraube. Instead, she used the three sets of verbs classified by Zhu (1979). As for the double-object sentence patterns, Law followed Peyraube’s (1981) scheme, which excluded the 畀 IO V DO pattern but included the V DO IO pattern. With the studies of Peyraube and Zhu as background, Law (1996) “attempted to examine the acceptability of deviations from the standard word order by introducing consideration of referentiality … and by considering discursive factors” (p. 81). She argued that the acceptability of the V IO DO pattern or the V 畀 IO DO pattern by her informants was due to semantic and/or discourse factors such as the complexity and the lengths of the DOs. As for the possibility that the V IO DO pattern is the result of language borrowing or language contact, Law (1996) claimed that “whether the use of this construction [V 畀 IO DO] reflects a beginning syntactic change in Hong Kong has not been addressed” (p. 81). In fact, when we re-examine Law’s data, we find that in some cases, the V IO DO pattern as well as the V 畀 IO DO pattern were accepted even when the DOs were not too long. The distribution of the V IO DO pattern, the V 畀 IO DO pattern and the different types of objects is given in Table 27 for those verbs with the intrinsic [+GIVE] feature such as [pei] 畀 ‘to give’, [sʊŋ] 送 ‘to give’ and [tɐi] 遞 ‘to pass’. 217. The four double-object patterns analyzed by Law were identical to the ones in Peyraube (1981). Altogether, 30 verbs (ten GIVE type verbs, ten deprive type verbs and ten make type verbs) were used in Law’s study. Each verb formed 36 different sentence patterns (i.e., four patterns with nine types of double-object verbs). 218. According to Law, the informants “are female, in their twenties, and were born and raised in Hong Kong. Three of them are university graduates, and two of them have received Form Five education and two further years of professional training. … They have all continuously lived in Hong Kong. They have to use written Standard Chinese when writing. When speaking, they only use spoken Cantonese in all situations” (p. 35).
134 Cantonese give and Double-Object Construction
Table 27. Acceptability of V IO DO and the V 畀 IO DO patterns with different types of DO
Type of DO
1. Numeral + Classifier + Noun 2. Determiner + Classifier + Noun 3. Determiner + Numeral + Classifier + Noun 4. Listing of several DOs 5. Long modified DO: (long adjectival phrase + Numeral + Classifier + Noun)
V IO DO 64% 68% 56% 80% 80%
V 畀 IO DO 44% 40% 28% 68% 72%
Some alternate explanations should be offered to explain why a substantial amount of sentences with the ‘Numeral + Classifier + Noun’ structure as the DO are expressed with the V IO DO pattern (64%) as well as the V 畀 IO DO pattern (44%) because noun phrases made up of ‘Numeral + Classifier + Noun’ such as [jɐt fɐn lɐi mɐt] 一份禮物 ‘one-cl-gift’ are the shortest among the five types of DOs in Law’s study. The same is also true for the ‘Determiner + Classifier + Noun’ type. To sum up, the IO DO pattern found in Cantonese is not a native pattern but a marked pattern subject to certain semantic or discourse conditions. 6.3
Cognitive linguistics approach
Cognitive linguistics views language as a tool for communication which encodes our thoughts and expresses the speakers’ intentions to exchange ideas with other interlocutors. The study of the linguistic phenomena is associated with the context in which the language is used. In this respect, sentences conveying the same propositions but with different structures represent different coding of the real world situations for the same event. The choice of different sentence structures depends on the emphasis placed on certain constituents in the sentences and there can be no direct relationship among the syntactic variants. According to Newman (1993), ta song wo qian 他送我錢 ‘He gave me money’ and ta song qian gei wo 他送錢給我 ‘He gave money to me’ “involve the elaboration of the single predicate [song 送], whereas [the latter] is a composite structure made up of the integration of two relational predicates, each one a different version … The result in each case is a clausal structure in which the giver, the thing, and the recipient are all profiled and elaborated, albeit with slightly different landmarking” (Newman, 1993, p. 445). In other words, ta song wo qian 他送我錢 has the recipient as the primary landmark while the thing the secondary landmark.219 Ta song qian gei wo 他送錢給我 has the opposite effect on the two landmarks (i.e., 219. ‘Landmark’ can be taken as a kind of cognitive salience or focus.
Chapter 6. Relationship between IO DO and DO IO patterns 135
qian 錢 and wo 我). The two sentences “reflect two different ways of imaging a scene and involve two different versions of the predicate” (Newman, 1993, p. 453) and the two constructions are “not ‘derived’ in any sense … Rather, the two constructions … are based on two versions of the verbal predicate, differing slightly in their profile characteristics” (p. 453). As for the suffixal gei 給 (i.e., V-給), Newman (1993) claimed that this construction “results in new profiling characteristics of the verb” (p. 458) meaning that V-gei is not simply concatenation of two morphemes. Based on his discussion, Newman also considered the DO IO pattern native in standard Mandarin. If the two patterns have different emphases on the DO and the IO, then why were there relatively few IO DO sentences in Cantonese, as our pre-modern dialect materials show (see Section 5.5)? Do Cantonese speakers always view the thing as a primary landmark over the recipient in the action of giving? 6.4 An alternative explanation: A loan feature resulting from language contact In Peyraube’s fieldwork data collected in the 1980s, the non-native patterns such as V IO DO and V 畀 IO DO were used by his informants. Examples include [ngóh sung néih nī bún syù] 我送你呢本書 ‘I give you this book’ and [ngóh sung béi néi nī bún syù] 我送畀你呢本書 ‘I give you this book’ (Peyrabue, 1981). It should be noted that in previous studies on the IO DO pattern used in the Yue dialects, only the V IO DO pattern was discussed. To my best knowledge, there has been no mention of the V 畀 IO DO pattern used in any Yue dialects. Peyraube’s fieldwork data thus recorded the actual use of the V IO DO pattern and the V 畀 IO DO pattern in Hong Kong Cantonese back to the 1980s.220 It should be noted that Peyraube’s informants were in the age of 20 to 30 when Peyraube conducted his fieldwork in the early 1980s.221 One thus cannot dismiss the possibility that the informants were under the influence of Putonghua or standard Mandarin, in which the IO DO pattern is used. As discussed, Peyraube’s analysis for the IO DO pattern in Cantonese is typologically unattested. We thus need to explain the phenomenon from a different perspective. In Section 5.3, we have shown that in many Southern dialects, both 220. B. Huang (1966) had a similar observation on the Yue dialect of Yangjiang 陽江 and claimed that the IO DO pattern such as 其畀你三部書 ‘he/she gave you three books’ does not sound like the native pattern of the Yue dialects and it may be a non-native sentence pattern. 221. The data of Peyraube was based on “Cantonese spoken in Hong Kong by the younger generation; the informants … were all born between 1950 and 1960” (Peyraube, 1981, p. 30).
136 Cantonese give and Double-Object Construction
DO IO and IO DO patterns co-exist. The latter is borrowed from Putonghua as a result of language contact. Bisang (2004) argued that “language contact also shows how structures unattested in the languages involved emerge from the interaction of substrate transfer and the reduction of markedness” (p. 36). Non-native structures can gain popularity and become core structures when the contact has become long and intensive. The emergence of new sentence patterns as a result of language contact is not impossible in Hong Kong. Although Cantonese is spoken as the mother tongue by the majority of the Hong Kong population, this is only limited to the spoken register. Similar to other Southern dialect groups, in the written register, Modern Standard Chinese, which is based on standard Mandarin, is used. Therefore, the two Chinese linguistic varieties are equally important in Cantonese speakers’ linguistic repertoire. It is then not uncommon for these two linguistic varieties to influence each other from time to time. In the colloquial speech, Cantonese speakers will produce (127) but in writing, only (128) is permissible. (127) 我 畀 本 書 你 13 35 35 ŋͻ pei pun sy55 nei13 I give cl book you ‘I’ll give you a book’
(128) 我 給 你 一 本 書 wo gei ni yi ben shu I give you one cl book ‘I’ll give you a book’
Furthermore, in the past three decades, Putonghua has gained its importance as one of the major linguistic varieties in Hong Kong. More and more people learned Putonghua to enhance their competitiveness in terms of language proficiency. The influence of Putonghua on the local dialect should have become much more vigorous and extensive than 30-some years ago when Peyraube conducted his fieldwork on the double-object construction in Hong Kong.222 It is thus a good opportunity for us to re-examine the double-object construction in the context of language contact. 222. Besides the double-object construction, there are other syntactic structures in Cantonese which have undergone linguistic change under the influence of Putonghua. These include the comparative construction discussed in Section 5.1 and the order of the post-verbal adverb [sin] 先 ‘first’ in [ŋͻ13 tsɐu35 sin55] 我走先 ‘I go first’, which now becomes [ŋͻ13 sin55 tsɐu35] 我先走. See also the study of Matthews and Yip (2001) on the use of [pei22] 被 such as in [keoi5 zung1 jyu1 bei6 sik1 fong3] 佢終於被釋放 ‘S/he was released eventually’ in Cantonese (p. 269, with Matthews & Yip’s transcription), which is commonly found in high register Cantonese such as news broadcast.
Chapter 7
On-going word order change in Cantonese double-object construction
This chapter examines the on-going word order change in the Cantonese doubleobject construction with a sociolinguistic approach. We can investigate the extent of the change in terms of a number of social variables: age and educational background of the speakers, etc. In Section 7.1, we will first outline the language situation of Hong Kong and discuss the role of Putonghua and Modern Standard Chinese play in shaping Hong Kong’s sociolinguistic situation. The rest of the chapter will examine the use of non-native IO DO pattern in Cantonese based on a fieldwork study with 40 native speakers of Cantonese. 7.1
Sociolinguistic situation of Hong Kong
To argue that the double-object construction in Hong Kong Cantonese is undergoing syntactic change due to the influence of Putonghua, we first need to understand the sociolinguistic situation of Hong Kong and to evaluate the possible extent of the influence of Putonghua on Hong Kong Cantonese. In the 19th century, Hong Kong was a small fishing village of China. In 1842, Hong Kong was ceded to the Great Britain by the Qing government under the Treaty of Nanjing as a consequence of the Opium War. Hong Kong then became a British colony and English was the official language. In 1851, the population of Hong Kong was only around 330,000, among which 95% (i.e., around 310,000) were Chinese (Information Services Department, Hong Kong SAR Government, 2005). These Hong Kong residents comprised five major ethnic groups (Mcdonogh & Wong, 2005): (a) Indigenous Cantonese people who were also known as bendi ren 本地人 ‘local people’. (b) Hoklos people 福佬人 who came from Fujian; (c) Tanka people 蛋家人 who were seafaring people living on boats; (d) Hakka people 客家人; (e) Chaozhou people 潮州人 who came from Shantou and Fujian. These residents spoke different dialects and some of them (such as Chaozhou and Cantonese, Chaozhou and Hakka, Cantonese and Hakka) are mutually unintelligible. There was then a pressing need for having a
138 Cantonese give and Double-Object Construction
lingua franca for inter-dialect group communication (Z. Zhang, 2009). Among the five ethnic groups, Z. Zhang (2009) claimed that the indigenous Cantonese group had a higher Ethnolinguistic Vitality index, which made Cantonese the common vernacular language of the community.223 The population of Hong Kong grew steadily until the 1940s. After the establishment of the People’s Republic of China in 1949, there was a large influx of people from Mainland China to Hong Kong. Between 1947 and mid-1950, the population of Hong Kong expanded from 1.8 million to 2.2 million (Information Services Department, Hong Kong SAR Government, 2005).224 The majority of these immigrants came from Guangdong province and Shanghai (Information Services Department, Hong Kong SAR Government, 2005). Another influx of refugees from Mainland China took place in the 1960s, especially during the Cultural Revolution.225 From the above demographic data, we can see that the sociolinguistic situation of Hong Kong was not simple. Although “without doubt Hong Kong is, ethnically and demographically speaking a ‘Cantonese world’” (Guldin, 1997, p. 27), there are still a large number of residents speaking other dialects as their mother tongue.226 Some speakers of other dialect groups underwent sociolinguistic alignment by switching their home language to Cantonese. The census data on usual home language from 1961 to 2001 are summarized in Table 28.227
223. The index of Ethnolinguistic Vitality is calculated on the basis of three factors: (a) status factors, which include the economic and social status of the ethnic group; (b) demographic factors, which include the population size and the birth and the death rates of the ethnic group; (c) the institutional and control factors, which include the degree of supports received by the ethnic group in the community. For more details on Ethnolinguistic Vitality, see Giles et al. (1977). 224. See also Census and Statistics Department, Hong Kong Government (1969), p. 14. 225. According to R. J. Li (1997), there were at least 60,000 refugees entering Hong Kong during April and May of 1961. 226. In view of this linguistic diversity, since 1954, Radio Television of Hong Kong broadcast news reported not only in Cantonese and English, but also Putonghua, Chaozhou and Hakka dialects. This practice lasted until the 1970s (Wai, 2004). 227. This is partially based on the data in Tsou (2002). Data on the year of 2001 are based on Census and Statistics Department, Hong Kong Government (2002).
Chapter 7. On-going word order change in Cantonese double-object construction 139
Table 28. Distribution of Hong Kong population in terms of usual home language228 Language/Dialect (%)
1961
1966
1971
1991
1996
2001
Cantonese Putonghua English Siyi Fujian Hakka Chaozhou
79.0 1.0 1.2 4.4 6.3 4.9 ---
81.4 -- 0.8 3.1 6.5 3.3 ---
88.1 -- 1.0 1.2 -- 2.7 ---
88.7 1.1 2.2 0.4 1.9 1.6 1.4
88.7 1.1 3.1 0.3 1.9 1.2 1.1
89.2 0.9 3.2 ---228 1.7 1.3 1.0
It can be seen that the dominant home language has been Cantonese throughout the four decades.229 The population of this home language increased from 79% in 1961 to almost 90% in 2001.230 The second was English. Putonghua did not rank very high. Only around 1% of the interviewees used it as the home language throughout the four decades. However, it should be noted that the gap between Putonghua and other Chinese dialects such as Siyi, Fujian, Hakka and Chaozhou had become smaller. In the 1960s, about 5% and 6.3% of the population spoke Hakka and Fujian dialects at home respectively. Four decades later, the percentages of these two dialects dropped to less than 2%. Tsou and You (2001) argued that these changes reflected language shift undertaken by non-Cantonese people, who switched their home language from Fujian and Hakka dialects to Cantonese. This type of alignment was due to the fact that in Hong Kong, Cantonese has a higher sociolinguistic status than other Chinese dialects. The three sets of languages (i.e., English, Cantonese and other Chinese dialects) form a triglossic hierarchy,231 as shown in Figure 4 (cf. Tsou, 1997). In Figure 4, English, which is labeled as the Supreme language in the triglossic model, is ranked the highest over Chinese dialects. The Supreme language is usually the official language of society and is used as the legal language and the medium of instruction in the education system. Cantonese is widely used as a daily vernacular 228. The data on the Siyi dialects were not provided in the 2001 census report. 229. Notice that in the census reports, the term ‘Cantonese’ (usually glossed as guangzhou hua 廣州話) refers to ‘Hong Kong Cantonese’ in a narrow sense.
230. The percentages were 90.8, 89.5 and 88.9 in the three population (by-)census data collected in 2006, 2011 and 2016 respectively. These data were based on Table A111 of the 2016 Population By-census website https://www.bycensus2016.gov.hk/en/bc-mt.html (accessed on 7/8/2021). 231. The notion of triglossia is a concept extended from Ferguson’s (1959) famous concept diglossia. For the concept of triglossia and its application on the linguistic situation of the Chinese communities other than Hong Kong, see Tsou and You (2001).
140 Cantonese give and Double-Object Construction
English
Cantonese
Hakka
Siyi
Chaozhou
Fujian
Others
Figure 4. Triglossic model for Hong Kong’s language situation
in the community, and it is also the medium used in formal contexts such as legislative council, the media, and education settings. Other Chinese dialects, such as Hakka, Siyi, Chaozhou and Fujian are referred to as Low languages which are mainly used as home language for intra-group communication. When speakers of these dialects go beyond the domestic context, such as doing shopping, attending schools, they need to switch to the High language, which is Cantonese or even to the Supreme language for effective communication.232 7.1.1 Status of Putonghua and Modern Standard Chinese in Hong Kong Although Cantonese is the lingua franca in Hong Kong, its status only remains at the spoken level despite written Cantonese has been shown to become popular and has developed into a new dialect writing system.233 It is also noted that although Cantonese pop songs have gained popularity in Hong Kong since the 1970s, “the great majority is written in the form of Standard Chinese which is neither colloquial nor markedly Cantonese; these songs are ‘Cantonese’ only in the sense that lyrics are pronounced in Cantonese” (Snow, 2004, pp. 140–141). Written Chinese, which is based on Modern Standard Chinese, was not considered another official language on a par with English until 1974. After the handover in 1997, the Hong Kong Government promoted the Biliteracy and Trilingualism (兩文三語) language policy.234 232. As Guldin (1997) remarked, “speaking Mandarin or chiu jau wa (Chaozhou) or Hakkag (Hakka) will usually not get you very far if you’re speaking to a Cantonese vendor” (p. 27). 233. Detailed studies on written Cantonese can be found in Bauer (1988) and Snow (2004). 234. This post-handover language policy gave rise to a lot of debates and discussions, such as in the domain of education. See, for example, Adamson and Auyeung (1997), Wang and Kirkpatrick (2019), and D. Li and Leung (2020).
Chapter 7. On-going word order change in Cantonese double-object construction 141
7.1.2 What does “Chinese” mean in Hong Kong? It is also noted that in the context of Hong Kong, the term ‘Chinese’ is sometimes confusing and ambiguous. In some cases, it refers to Cantonese, the spoken vernacular while in some cases, it refers to Putonghua (as well as Modern Standard Chinese) (see, for example, Bruche-Schulz, 1997). This ambiguity is particularly prevalent on the issue of medium of instruction in education after 1997 when China resumed sovereignty over Hong Kong. After 1997, mother tongue education (母語教學) was proposed and promoted because it is argued that education conducted in the mother tongue of the students can stimulate students’ interest in learning and help students overcome the communication barriers they might face (see, for example, W. Leung & Li, 2020). Interestingly, there are different interpretations toward the term ‘mother tongue’ in the context of Hong Kong: Is it Cantonese or Putonghua? As can be seen in the above discussion, the mother tongue of the majority of the Hong Kong population is Cantonese, which has not yet developed a formal and standardized writing system. However, adopting Putonghua as the medium of instruction also creates considerable problems to both students and teachers since Putonghua is not the native language of most Hong Kong students and teachers. Students cannot use Putonghua as a daily language until they have learned it comprehensively and master it proficiently. In addition, teachers may not have received sufficient professional training on teaching non-language subjects in Putonghua. 7.1.3 Contact with mainland China after the 1970s The status of Putonghua and Modern Standard Chinese in Hong Kong has been rising since the 1970s when the Chinese government implemented open door policies, which provided more opportunities for contact between Mainland China and the foreign world, including Hong Kong. There was thus a need for Hong Kong people to gain proficiency in Putonghua. At the beginning, one may consider that learning Putonghua is simply for the sake of gaining access to China. After 1997, Hong Kong became the first special administrative region of China and sociolinguistic alignment of its language situation thus began to take place. To meet this new sociolinguistic change, more channels have been opened up for Hong Kong people to use Putonghua: a. Media plays a significant role in promoting Putonghua in Hong Kong. An exclusive Putonghua radio channel operating 24 hours a day has been set up in the Radio Television of Hong Kong, which is administered by the government. In addition, financial news and news reports are also broadcast in Putonghua for
142 Cantonese give and Double-Object Construction
b.
c.
d.
e.
f.
g.
about half an hour a day in some local TV channels. More Putonghua programs from the Chinese Central Television (CCTV) are also broadcast in commercial TV companies.235 Another aspect related to media is the development of the entertainment industry. There had been a significant increase in cultural exchange between Hong Kong and the Mainland since the 1990s. Particularly, more movies, pop songs and TV dramas produced in Mainland China are shown in Hong Kong.236 In the past decades, there has been a gain of popularity of printed books published in Mainland China. These books are printed with simplified characters and their low prices attract many Hong Kong readers. Many local bibliophiles visit large book arcades in the Pearl River Delta region, such as Shenzhen, Zhuhai and Guangzhou. The rapid expansion of the Internet at the turn of the new millennium also sped up the contact between Hong Kong and the Mainland. The Internet allows people from both sides to access resources of each other more easily and efficiently. This facilitates language contact between the two communities. Tourism is another contributing factor that should not be overlooked. After 1997 as well as the economic downturn such as SARS in the beginning of the millennium, residents from Mainland China were allowed to visit Hong Kong on an individual basis in order to speed up economic revival. Up to 2020, residents of 49 cities in Mainland China could apply individual permits to visit Hong Kong.237 Universities in Hong Kong also play a significant role in promoting Putonghua. In addition to offering Putonghua courses, they also recruit exchange students and research staff/students from Mainland China. In addition, there are a lot of collaborations between both communities. Although there is no data on the number of people learning Putonghua in Hong Kong, a survey on the student handbooks in the universities of Hong Kong reveals that passing a language proficiency test in Putonghua is now prescribed as one of the compulsory language requirements for graduation. Furthermore,
235. A similar discussion can be found in Pierson (1998). 236. It is noted that in the mid-20th century, movies were rendered in Putonghua as well as Cantonese while the music industry of Hong Kong was dominated by English and Mandarin songs before the 1970s. Before the 1970s, Cantonese music was mainly used in Cantonese operas. See, for example, W. Zhao (2007) and Y. Chu (2001). 237. This type of individual visit is known as ziyou xing 自由行 (literal meaning: free-travel). When this policy was first implemented in 2003, only residents of 28 cities can apply for this permit. Relevant information can be obtained from the website of the Immigration Department, the Government of the Hong Kong Special Administrative Region (http://www.immd.gov.hk).
Chapter 7. On-going word order change in Cantonese double-object construction 143
the number of candidates sitting for the Chinese Proficiency Test ‘漢語水平考 試, Hanyu shuiping kaoshi’, also known as HSK, has increased substantially over time. The candidate population taking the HSK, in both general and advanced levels from 2000 to 2005, is summarized in Table 29.238 Table 29. Number of candidates taking HSK in Hong Kong
2000
2001
2002
2003
2004
2005
Number of candidates
700
800
1,200
1,400
1,550
1,750
Based on the above observations, we can see that there has been an increase of contact and usage of Putonghua and Modern Standard Chinese in Hong Kong since 1997. As a consequence, the linguistic contact between Hong Kong Cantonese and Putonghua should have become more intensive and frequent. Among the Cantonese speakers in Hong Kong, the younger ones tend to have more chances to use Putonghua than older speakers, partially due to their education training, as well as their intensive exposure to the media and the cyberworld. Thus, we are more likely to observe that younger speakers are more influenced by Putonghua than the seniors. In this case, age can be one of the social variables, with which we can investigate the extent of the Putonghua influence on Cantonese. In addition, the exposure to Putonghua and Modern Standard Chinese also plays a vital role in the spread and diffusion. These factors will be taken into consideration when analyzing the fieldwork data. 7.1.4 Language attitude toward Putonghua The attitude toward Putonghua has also changed dramatically in Hong Kong. According to a number of studies on language attitude of Hong Kong students toward English, Cantonese and Putonghua, there is a positive attitude toward Putonghua because “it can help them communicate with people in Mainland China and Taiwan” (M. Lai, 2001, p. 121), especially after 1997 when China resumed sovereignty over Hong Kong. From Lai’s survey, each of the three linguistic varieties was perceived to have different roles and status by Hong Kong people: English is a major tool for “academic and career development”, while Cantonese is still regarded as the “mother tongue and the language of local identity”. Putonghua is mainly used for “nation-wide communication” and carries “the sense of ‘Chineseness’” (M. Lai, 2001, p. 130). CIT179
CIT179
238. The statistical data are based on the annual reports of the Hong Kong Examinations Authority (renamed as Hong Kong Examinations and Assessment Authority after 2002) from 2000 to 2005. (Hong Kong Examinations Authority, 2000, 2001, 2002, 2003, 2004, 2005)
144 Cantonese give and Double-Object Construction
With the rising status of Putonghua and Modern Standard Chinese in Hong Kong since 1997, we find that the change can be attributed to a number of extra-linguistic factors, such as the economic boom of Mainland China which encourages more and more people to learn Putonghua. In spite of its rising status, Putonghua, as a spoken variety, may not replace the local dialect (i.e., Hong Kong Cantonese) within a short time period but it is not impossible for it to be as important as English and Cantonese in Hong Kong in the future (cf. Tsou, 1994b).
Fieldwork study on Cantonese double-object construction
7.2
In Chapter 5, we discussed the typological differences between standard Mandarin and the Yue dialects in terms of the relative word order of the DO and the IO. The verbs to be focused on in the present study are Zhu’s Va and Va/c type verbs. The compatibility of these two verb types with the four syntactic patterns in Hong Kong Cantonese (HKC) and standard Mandarin (SM) is summarized in Table 30. Table 30. Compatibility of the four double-object patterns with Va and Va/c verbs in standard Mandarin and Hong Kong Cantonese239240 Va (給, 送) Va/c (寄, 匯)
V M IO DO239 HKC
SM
− −
+ +
M IO V DO HKC
− −
SM #240 +
V IO DO HKC
SM
− −
+ −
V DO M IO HKC
SM
+ +
− −
Those cells in grey indicate that there is no corresponding syntactic pattern in Hong Kong Cantonese for the verb groups concerned. In other words, the investigation will pay special attention to these patterns. Since both verb types are expressed with the same pattern in Hong Kong Cantonese (i.e., the DO IO pattern), we simply combine them into one single class of double-object verbs with the semantic feature of [+GIVE] (i.e., V[+GIVE]) in our sociolinguistic study.241 It is hypothesized that proficiency in Putonghua or Modern Standard Chinese constitutes one of the key factors contributing to the syntactic change and this can be correlated with a number of social variables, such as educational background, 239. ‘M’ stands for the IO-marker [pei] 畀 and gei 給 in Hong Kong Cantonese and standard Mandarin respectively. 240. In Section 5.4.2, we showed that sentences with this syntactic pattern can have two interpretations: double-object or benefactive. 241. According to Zhu (1979), this pattern is also used in standard Mandarin but we have shown in Chapter 5 that this is not a native pattern in the Northern dialects.
Chapter 7. On-going word order change in Cantonese double-object construction 145
age and social classes of the informants. Therefore, the fieldwork data will be analyzed with respect to the following social variables: a. Educational background: In Hong Kong, speakers who have received formal education possess knowledge of Modern Standard Chinese. These speakers are likely to be influenced by Modern Standard Chinese. b. Age: The social variable of educational background is correlated with the age of the speakers. In 1978, compulsory education policy was implemented, under which any Hong Kong resident must receive formal and compulsory education between the age of 6 and 15. In this regard, speakers above 50 years old in the beginning of the 21st century may not have received full and extensive education. c. Social class: Occupations requiring higher educational background will have more chances to use other languages, such as Putonghua or English in addition to Cantonese. d. Usage of Modern Standard Chinese / Putonghua: The investigation will also survey the usage of Modern Standard Chinese and Putonghua. This can be reflected by speakers’ daily activities such as newspaper reading, business writing, etc. It is obvious that there is a positive correlation of the usage of these languages with the linguistic change concerned. The fieldwork consists of two tasks: Production and Perception. In the production part, informants were asked to produce double-object sentences with the lexical items presented to them. Words or phrases were expressed in colloquial Hong Kong Cantonese such as [ŋͻ tei] 我哋 ‘we’, [ŋͻ tsi pɐt] 我枝筆 ‘my pen’. The words or phrases represent the subjects, double-object verbs, IOs and DOs. The IO marker [pei] 畀, however, was not presented to the informants. Informants were asked to use these lexical items to make up meaningful sentences based on their own speech. They could supply extra words (such as the IO marker [pei] 畀) to make their sentences more natural. The lexical items of each sentence were presented on a slide as shown in Figure 5.242 我 I
你 you
畀 give
對筷子 a pair of chopsticks
Figure 5. Sentence components used in the production task 242. The English glosses were not included in the slides. They are added here for illustration.
146 Cantonese give and Double-Object Construction
The lexical items on each slide were randomly placed and the location of the same sentence constituents (e.g., the subject, the DO, the IO, the verb) differs from slide to slide. This type of elicitation method was found suitable for the current study. All the informants were literate and could understand all the words and phrases presented to them. We did not ask them to translate sentences from English or Modern Standard Chinese into Hong Kong Cantonese because this may give some influence to the informants in selecting the best sentence pattern. For example, they may simply translate the sentence ‘I give you a book’ into [ŋͻ13 pei35 nei13 jɐt5 pun35 sy55] 我畀你一本書 ‘I gave you a book’ and we cannot decide if it is a pattern the informant uses or it is simply a direct translation of the input sentence. There were altogether 121 sentences in the production task. Another feature that we looked into is the effect of the lengths of the DOs and the IOs. Some sentences have long DOs but short IOs or vice versa. Some have long DOs and long IOs. In addition, we pay special attention to some verbs such as [tsε] 借 ‘to borrow / to lend’ and [tsou] 租 ‘to rent / to rent out’ that can have both [+GIVE] and [+deprive] semantic features. When these two verbs are expressed with the V IO DO pattern in Putonghua, the meaning can be ambiguous. It can either refer to a giving or a deprive action. In Hong Kong Cantonese, the giving meaning is expressed with the V DO 畀 IO pattern while the deprive meaning with the V IO DO pattern. Therefore, it is important to confirm with the informants which meaning they refer to when the IO DO pattern is used with these verbs. In my previous pilot study (Chin, 2001), I noted that patterns not used by the informants in the production task do not mean that the informants did not accept these patterns.243 In other words, there can be a continuum of the acceptability for the various double-object sentence patterns. Therefore, we included a perception task in which sentences with different double-object patterns were read to the informants. The acceptability of these sentences was then recorded and analyzed. Informants were also asked to give comments on these patterns even if they did not use them. Altogether 54 sentences were used in the perception task. It is noted that in many cases, informants accepted more than one sentence pattern and we then had to check the degree of acceptability among these patterns. The sentences used in the production task and the perception task are listed in Appendices 5 and 6 respectively. There were forty informants participating in the fieldwork study. The fieldwork was conducted in 2006 on an individual basis. On average, each investigation lasted for about 1.5 hours. Before the study, each informant was asked a set of questions related to their language background. The information sheet is listed in Appendix 7. 243. The pilot study was conducted with twenty Hong Kong Cantonese informants in Seattle in 2001.
Chapter 7. On-going word order change in Cantonese double-object construction 147
It should be stressed that we do not expect any informant adopts the non-native pattern(s) for all the double-object verbs. Instead, we are likely to obtain a scenario in which the influence of Putonghua or Modern Standard Chinese is diffusing in the sense that the change is more prevalent for certain verbs, but less obvious for others. Furthermore, among the three non-native patterns, which one(s) is/are more acceptable? Do all the informants show the same preference toward a particular non-native pattern? In other words, is it simply arbitrary or is there a trend that one pattern is preferred to other non-native variants? These questions can be answered with the fieldwork data collected from the forty informants. 7.3
Fieldwork data on Cantonese double-object construction
In this section, we analyze the data collected from the fieldwork with forty native Hong Kong Cantonese informants. Before each interview, a set of demographic questions (see Appendix 7) were asked. The information we collected includes informants’ native dialects, occupations and their daily use of Putonghua or Modern Standard Chinese. 7.3.1 Background information of the forty informants Table 31 summarizes some key information of the forty informants in terms of age, usage of Putonghua or Modern Standard Chinese. Table 31. Background information of the forty informants244245 Informant 1 2 3 4 5
Age 18 18 19 19 19
Gender F M F F F
Putonghua / Modern Standard Chinese Usage (hours/day)244
Proficiency245
1.5 0.5 3 2 2
16 18 14 18 16
(continued) 244. The usage (in terms of hours per day) includes the amount of time spending on Chinese newspapers (mainly written in Modern Standard Chinese) and TV programs, the daily usage of Putonghua in job and home environments. 245. ‘Proficiency of Putonghua and Modern Standard Chinese’ is based on the informants’ ability in the four areas of language skills, namely writing, speaking, listening and reading. It is rated by adding up the scores of these four skills (based on question 10 of the Information Sheet). The maximum score for each skill is 5 and the maximum score for the proficiency is thus 20.
148 Cantonese give and Double-Object Construction
Table 31. (continued) Informant 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Age 21 21 22 23 23 24 24 24 24 24 25 25 25 27 28 30 30 30 32 32 33 33 33 34 34 35 39 41 47 48 53 60 61 65 65
Gender M F M M M F M F M M F M M F F M M M M M M F F F M M M M F F M M F F M
Putonghua / Modern Standard Chinese Usage (hours/day)244
Proficiency245
1 0.5 3 3 2.5 2.5 2 2 0.5 0 3 1.5 0.5 0.5 0 4 2.5 2 3 3 2 1 0.5 5 0.5 1.5 0.5 0.5 0.5 2 2 3 1.5 2 1
14 16 16 20 10 16 18 16 13 14 20 18 20 18 16 18 18 20 20 17 20 20 15 20 20 16 16 12 7 9 8 7 6 5 5
Chapter 7. On-going word order change in Cantonese double-object construction 149
Generally speaking, older informants have lower proficiency in Putonghua, but they still read newspapers and magazines written in Modern Standard Chinese. Our hypothesis is that younger informants with more exposure to Putonghua and Modern Standard Chinese are more likely to be influenced by these linguistic varieties. 7.3.2 The production task Table 32 presents some global statistics of the double-object sentence patterns produced by the informants. Table 32. Percentage distribution of the double-object patterns for GIVE verbs Typological pattern DO IO IO DO 將-construction Total
Sentence pattern
V DO 畀 IO V IO DO V 畀 IO DO 畀 IO V DO 將 DO V 畀 IO
V[+GIVE] 91.2 5.4 0.7 0 2.7 100
Besides the double-object patterns, some informants expressed the sentences with the disposal construction involving the preposition [tsœŋ] 將, in which the DO is fronted to the pre-verbal position. For example, (129) is the 將-construction of (130). (129) 我 將 隻 杯 打爛 咗 ŋͻ13 tsœŋ55 tsεk3 pui55 ta35 lan22 tsͻ35 I jiang cl cup break asp ‘I broke the cup’
(130) 我 打 爛 咗隻 杯 ŋͻ13 ta35 lan22 tsͻ35 tsεk3 pui55 I break asp cl cup ‘I broke the cup’
In the fieldwork, 56 sentences were expressed with the 將-construction by at least one informant. Among these 56 sentences, two were expressed with the 將-construction by more than 10 informants. Table 33 lists the objects and the number of informants (larger than 5).246 246. Due to the limited space, in the following discussion of the sentences used in the production and perception tasks, English gloss is only provided for the main verbs.
150 Cantonese give and Double-Object Construction
Table 33. Sentences expressed with the 將 DO v io pattern
Verb
分 ‘to share’
退 ‘to return’ 交 ‘to handover’ 還 ‘to return’
遞 ‘to pass’ 嫁 ‘to marry off ’ 分 ‘to divide’ 退 ‘to return’ 嫁 ‘to marry off ’ 畀 ‘to give’
DO
我辛辛苦苦賺返 來啲錢 收多咗嘅錢 收返來嘅假銀紙 十幾年前同你借落 嘅本錢 啱啱點著嘅蚊香 個女 呢批贜物 呢幾件衫同埋架車 個女 呢本書
IO
佢啲仔女同埋細佬妹
將-construction V DO IO
啲人客 差人 你
睏喺細房個阿婆 債主 幾多個人 人哋 嗰個窮鬼 求先我撞到嗰個朋友
14
24
10 7 6
26 32 22
6 6 6 5 5 5
29 32 24 35 34 34
In Chapter 5, we showed that the native pattern for the give-type double-object construction in Cantonese is V DO 畀 IO. As Table 32 shows, most of the double-object sentences (91%) were expressed with the native pattern for the two types of verbs. The patterns of V IO DO, V 畀 IO DO and 畀 IO V DO are considered non-native patterns. In the following, we examine the extent of the use of these non-native patterns. c5
tab32
7.3.2.1 The V IO DO pattern Among the 24 double-object verbs, 17 were expressed at least once with the V IO DO pattern. The frequencies (in descending order) are summarized in Table 34.247 Table 34. Distribution of the V IO DO pattern for V[+GIVE] verbs V[+GIVE] 1. 畀 ‘to give’ 2. 獎 ‘to award’ 3. 還 ‘to return’ 4. 補 ‘to make up’ 5. 輸 ‘to lose’ 6. 送 ‘to give’ 7. 寄 ‘ to send’ 8. 分 ‘to share/divide’
Frequency In the fieldwork
V IO DO
V DO (畀) IO
10 6 6 6 6 5 8 6
8 6 5 3 3 3 2 2
10 6 6 6 6 5 8 6
247. Appendix 8 lists the sentence numbers that each informant expressed with the non-native patterns.
Chapter 7. On-going word order change in Cantonese double-object construction 151
Table 34. (continued) V[+GIVE] 9. 介紹 ‘to introduce’ 10. 教 ‘to teach’ 11. 找 ‘to give change’ 12. 遞 ‘to pass’ 13. 交 ‘to handover’ 14. 撥 ‘to allocate’ 15. 讓 ‘to give/yield’ 16. 派 ‘to distribute’ 17. 賠 ‘to compensate’ Total
Frequency In the fieldwork
V IO DO
V DO (畀) IO
5 2 2 8 7 5 2 2 1 87
2 2 2 1 1 1 1 1 1 44
5 2 2 8 7 5 2 2 1 87
The above table shows that [pei] 畀 ‘to give’ was the verb expressed with the V IO DO pattern the most. There were altogether 10 sentences with [pei] 畀 as the main verb in the production task and 8 of them were expressed at least once with the V IO DO pattern. In the following, we list the actual number of sentences that were expressed with the V IO DO pattern for these 17 verbs. Table 35. Sentences expressed with the V IO DO pattern V[+GIVE]
1. 教 ‘to teach’ 2. 讓 ‘to give/yield’ 3. 畀 ‘to give’ 4. 獎 ‘to reward’ 5. 補 ‘to make up’ 6. 分 ‘to share’ 7. 找 ‘to give change’ 8. 還 ‘to return’ 9. 輸 ‘to lose’ 10. 賠 ‘to compensate’ 11. 介紹 ‘to introduce’ 12. 遞 ‘to pass’ 13. 撥 ‘to allocate’ 14. 寄 ‘to send’ 15. 派 ‘to distribute’ 16. 交 ‘to handover’ 17. 送 ‘to give’ Total
V IO DO 53 28 53 39 14 12 6 12 8 2 2 1 1 2 1 1 9 244
V DO (畀) IO 25 8 255 195 105 91 73 160 108 33 75 36 38 77 39 39 107 1464
% of V IO DO 67.9 77.8 17.2 16.9 11.8 11.7 7.6 7.0 6.9 5.7 2.6 2.7 2.6 2.5 2.5 2.5 0.8 14.3
152 Cantonese give and Double-Object Construction
More than half of the responses with the verbs 教 and 讓 were expressed with the V IO DO pattern. However, we still cannot claim that the use of the V IO DO pattern is due to the influence of Putonghua simply based on these quantitative data. In the following, we examine these sentences closely, with a special focus on the nature of the DOs and the IOs. Table 36. Use of the V IO DO pattern with GIVE verbs Verb 畀 ‘to give’ 畀 ‘to give’ 畀 ‘to give’
畀 ‘to give’ 畀 ‘to give’ 畀 ‘to give’ 畀 ‘to give’ 獎 ‘to award’ 獎 ‘to award’ 獎 ‘to award’ 獎 ‘to award’ 獎 ‘to award’ 獎 ‘to award’
還 ‘to return’
還 ‘to return’ 還 ‘to return’ 還 ‘to return’ 還 ‘to return’
補 ‘to make up’ 補 ‘to make up’ 補 ‘to make up’ 輸 ‘to lose’ 輸 ‘to lose’ 輸 ‘to lose’ 送 ‘to give’ 送 ‘to give’ 送 ‘to give’
IO 大家 我 對面大廈門 口個阿婆 我 我 我 佢 我 你 我 佢 細佬 今次有份破 案嘅同事 你 你 圖書館 人 我哋
大家 你 我兩公婆 我 人 我細佬 佢 佢 我同我細佬
DO 一個交代 兩斤白菜 一疊用花紙包好 嘅嘢 個機會 對筷子 條蕃薯 頂帽 乜嘢 杯士多啤梨雪糕 乜嘢 啲有意義嘅嘢 兩件恤衫 四十萬現金
十幾年前同你借落 嘅本錢 所有嘢 啲書 筆錢 一個花樽、一個花 盆、兩個槌仔 一堂 兩個月嘅雙糧 兩百文 兩盤棋 幾萬文同埋架車 廿萬 一張枱 本書 兩本書同埋其他嘢
No. of informant V IO DO
V DO IO
35 6 2
7 33 32
3 2 2 1 12 10 7 6 3 1
37 37 36 38 28 30 33 32 36 36
5
22
4 1 1 1
35 33 35 35
5 6 3 4 3 1 4 4 1
35 33 37 33 36 39 36 32 39
Chapter 7. On-going word order change in Cantonese double-object construction 153
Table 36. (continued) Verb
IO
DO
分 ‘to share’ 分 ‘to divide’ 教 ‘to teach’ 教 ‘to teach’ 找 ‘to give change’ 找 ‘to give change’ 介紹 ‘to introduce’ 寄 ‘to send’ 寄 ‘to send’ 遞 ‘to pass’ 交 ‘to handover’ 撥 ‘to allocate’ 讓 ‘to give/yield’ 派 ‘to distribute’ 賠 ‘to compensate’
你 我 我個仔 個仔 我 啲客 我 老竇 個仔 董事長 議員 學校 我 大家 我哋
一半 咁少嘢 兩度散手 好多英文生字 幾個五文銀 三百文 幾個朋友 乜嘢生日禮物 一個包裹 份最新嘅公司報告 立法草案 兩幅地皮 三步棋 啲試卷 一千幾百
No. of informant V IO DO
V DO IO
9 3 28 25 4 2 2 1 1 1 1 1 28 1 2
26 35 11 14 35 38 37 38 39 36 39 38 8 39 33
Both functional and discourse approaches suggest that the V IO DO pattern is preferred when the DO is long or when the DO contains new information (see Section 6.2.2). In those sentences expressed with the V IO DO pattern, only six of them have the DOs much longer than the IOs or the DOs are indefinite. Other sentences expressed with the V IO DO pattern have the DOs and IOs with almost equal lengths. This demonstrates that the discourse factor (i.e., long DOs) is not sufficient enough to account for the use of the V IO DO pattern. Among those sentences expressed with the non-native pattern, the sentence [ŋͻ13 pei35 tai22 ka55 jɐt5 kͻ33 kau55 tͻi22] 我畀大家一個交代 ‘I will give you all an account’ has the highest frequency. 33 out of 40 informants (about 82.5%) expressed this sentence with the V IO DO pattern. This sentence in fact sounds literal, such as the DO [kau tͻi] 交代 ‘an account’, and this may show that those informants have started to transform the written language into their spoken language. It should be noted that in this task, only the lexical items were presented to the informants and they were asked to produce sentences based on their own speech. The use of the V IO DO pattern indicates that these informants found this pattern more acceptable and natural than the native pattern [ŋͻ13 pei35 jɐt5 kͻ33 kau55 tͻi22 (pei35) tai22 ka55] 我畀一個交代(畀)大家. There are other sentences that deserve our attention. For the sentence [ŋͻ13 fɐn55 5 jɐt pun33 pei35 nei13] 我分一半畀你 ‘I share one half with you’), nine informants
154 Cantonese give and Double-Object Construction
expressed it with the V IO DO pattern. It is noted that the V IO DO pattern for this sentence (i.e., [ŋͻ13 fɐn55 nei13jɐt5 pun33] 我分你一半) is ambiguous in Putonghua. It can either mean ‘I share one half WITH you’ or ‘I share one half FROM you’. In Cantonese, these two meanings are distinguished by means of opposite word order: The GIVE meaning is expressed with the V DO 畀 IO pattern while the deprive meaning with the V IO DO pattern. In other words, when an informant expressed this sentence with the V IO DO pattern, we had to check what his/her intended meaning was. Hypothetically, if he/she has been receiving influence from Modern Standard Chinese or Putonghua, the V IO DO pattern can also express the GIVE type meaning. Otherwise, the sentence can only have the meaning of deprivation.248 Our hypothesis suggests that the influence of Modern Standard Chinese or Putonghua is more prominent on those informants who are young because they have more exposure to these linguistic varieties. In the following, we compare the use of the V IO DO pattern against the age groups of the informants. Table 37. Use of the V IO DO pattern in terms of age group Age group 20 or below 21 – 25 26 – 30 31 – 35 36 – 45 46 – 55 56 or above
No. of informant 5 13 5 8 2 3 4
Frequency V IO DO 26 96 32 62 12 12 4
V DO (畀) IO 529 1304 595 726 217 296 478
Ratio of IO DO : DO IO 1 : 20.3 1 : 13.6 1 : 18.6 1 : 11.7 1 : 18.1 1 : 24.7 1 : 120
When we combine the data in Table 37 with those in Table 31, which indicates the extent of the informants’ exposure to Putonghua or Modern Standard Chinese, we can immediately see that informants between 21 and 35 years old used the V IO DO pattern more frequently than the older informants. This trend is expected since younger informants have more chances to use Putonghua or Modern Standard Chinese. Particularly, informant 23 was a Chinese medicine practitioner, who needed to use Putonghua very frequently.249 Informant 24 worked in the IT field 248. In other words, more than nine informants gave the V IO DO word order for this sentence but the intended meaning for some of them was only ‘I share one half from you’. For these cases, such occurrences are not counted for the frequency of the V IO DO pattern. It should be noted that such a checking of meaning was carried out only when the informants used the V IO DO pattern. If the informant used the V DO 畀 IO, no checking was done because there was no ambiguity in the sentence pattern concerned. 249. He produced 10 V IO DO sentences with the [+GIVE] verbs.
Chapter 7. On-going word order change in Cantonese double-object construction 155
and often read IT magazines published in Taiwan, in which the V IO DO pattern was also used. In this production task, he produced 13 sentences with the V IO DO pattern. Informant 12, an undergraduate student in computer science, who was also learning Putonghua, produced 16 sentences with this non-native pattern. Older informants also used the V IO DO pattern but to a lesser extent when compared with the younger informants. Their use of the V IO DO pattern is not surprising at all because almost all of them read newspapers, which are mainly written in Modern Standard Chinese. But other than printed media, these older informants have relatively less exposure to Putonghua or Modern Standard Chinese, such as in the areas of writing, speaking and listening.250 In terms of occupation, there are nine categories, as shown in Table 38. Table 38. Use of the V IO DO pattern in terms of occupation Occupation Programmer Researcher Chinese medicine practitioner Lawyer Student Clerical worker Insurance agent Social worker Labor Housewife/retired
No. of informant 6 2 2 1 16 2 2 1 2 6
Frequency V IO DO VDO (畀) IO 61 14 18 6 110 8 7 4 8 8
559 192 190 94 1730 210 207 95 200 668
Ratio of IO DO : DO IO 1 : 9.2 1 : 13.7 1 : 10.6 1 : 15.7 1 : 15.7 1 : 26.3 1 : 29.6 1 : 23.8 1 : 25 1 : 83.5
We can see that computer programmers used the V IO DO pattern the most and because one of them indicated that he often read IT magazines published in Taiwan. Chinese medicine practitioners also used the V IO DO pattern a lot because they worked with practitioners from Mainland China, who speak Putonghua mostly. Informants who were working as researchers often read academic articles written in Chinese and the influence of Modern Standard Chinese is expected. As for the informant who was a lawyer, his specialization was in Chinese business law and thus he had to use written Chinese very often. Housewives and retired people use the non-native pattern the least because they had fewer opportunities to use written Chinese or speak Putonghua.
250. These older informants indicated that they need to read the subtitles when watching TV programs broadcast in Putonghua.
156 Cantonese give and Double-Object Construction
7.3.2.2 The V 畀 IO DO pattern According to our data (Table 32), the V 畀 IO DO pattern is much less preferred than the V IO DO pattern. Only 0.7% of the sentences collected were expressed with this pattern. Table 39 lists those sentences (ranked by the frequency in descending order) that the informants expressed with the V 畀 IO DO pattern. Table 39. Use of the V 畀 IO DO pattern251
Verb
IO
還 ‘to return’
你
DO
十幾年前同你借落 嘅本錢 輸 ‘to lose’ 我 兩盤棋 賠 ‘to compensate’ 我哋 一千幾百 分 ‘to share’ 我 咁少嘢 還 ‘to return’ 我哋 一個花樽一個花盆 兩個槌仔 遞 ‘to pass’ 老闆 份最新嘅公司報告 分 ‘to share’ 幾多個人 呢批贜物 獎 ‘to award’ 我哋 啲有意義嘅嘢 補 ‘to compensate’ 你 兩個月嘅雙糧 退 ‘to return’ 低收入嘅家庭 六成嘅稅 還 ‘to return’ 你 所有嘢 找 ‘to give change’ 我 幾個五文銀同埋 十文紙 獎 ‘to award’ 今次有份破案嘅同事 四十萬現金 交 ‘to handover’ 我哋個會計部 二千五百文入會費 還 ‘to return’ 你 個電視機 撥 ‘to allocate’ 學校 兩幅地皮 介紹 ‘to introduce’ 你 個女仔 寄 ‘to send’ 老竇 乜嘢生日禮物 輸 ‘to lose’ 老李 盤棋 撥 ‘to allocate’ 中小學 好多資源 輸 ‘to lose’ 邊個 兩盤棋
Frequency
V 畀IO DO V DO 畀 IO 6
23
3 2 2 2
33 33 35 35
2 1 1 1 1 1 1
36 25251 32 33 35 35 35
1 1 1 1 1 1 1 1 1
36 37 37 38 38 38 39 39 39
251. Eight informants produced this sentence with the 將 DO V 畀 IO pattern.
Chapter 7. On-going word order change in Cantonese double-object construction 157
22 verbs were expressed with the V 畀 IO DO pattern at least once and the total number of sentences expressed with this pattern is 34. Recall that 17 double-object verbs were expressed with the V IO DO pattern (Table 35). Among them, twelve were also expressed with the V 畀 IO DO pattern. Their distribution is summarized in Table 40. Table 40. give verbs expressed with the V IO DO and the V 畀 IO DO patterns Verb
還 ‘to return’ 獎 ‘to award’ 輸 ‘to lose’ 補 ‘to make up’ 找 ‘to give change’ 賠 ‘to compensate’ 分 ‘to share’ 遞 ‘to pass’ 撥 ‘to allocate’ 寄 ‘to send’ 介紹 ‘to introduce’ 交 ‘to pass’ Total
V IO DO 12 39 8 14 6 2 12 1 1 2 1 1 99
V 畀 IO DO 10 2 5 1 1 2 4 2 2 1 1 1 29
Total 22 41 13 15 7 4 16 3 3 3 2 3 128
Generally speaking, the V IO DO pattern was used more than the V 畀 IO DO pattern for the same verb. 7.3.2.3 The 畀 IO V DO pattern The third non-native double-object pattern under investigation is the 畀 IO V DO pattern. As shown in Table 32, no informants used this pattern. There are a number of reasons for the null usage: a. In Putonghua or Modern Standard Chinese, as discussed in Section 5.4.2, this pattern is sometimes ambiguous: The IO can be interpreted as a beneficiary or a recipient. b. This pattern differs from the other two non-native patterns when the 畀-IO phrase is placed before the verb while the other two after the verb. Structurally speaking, this pattern is less similar to the other two.252
252. The other two patterns can be claimed to have the same structural pattern in terms of the word order in relation to the verb, i.e., both objects appear after the verb. The only difference is the presence of the IO marker [pei] 畀 after the verb.
158 Cantonese give and Double-Object Construction
c. Another important reason for not adopting the 畀 IO V DO pattern is that this pattern has the same structure as the causative and passive constructions in Cantonese. Both causative and passive constructions have the following surface structure: NP1 + 畀 + NP2 + V (+ NP3), which is identical with the double-object pattern NP1 + 畀 + NP2 + V[+GIVE] + NP3 under discussion. The null usage of this pattern in the production task can be attributed to the avoidance of structural ambiguity. In the perception task, which will be discussed in Section 7.3.3, we observe that informants could not obtain the double-object meaning from those sentences expressed with this pattern. Instead, they interpreted the sentences with the causative or permissive meaning. 7.3.2.4 Summary of the production task The production task was designed to examine the extent of the use of non-native double-object patterns by the forty Hong Kong Cantonese informants. Among the three non-native patterns under investigation, the V IO DO pattern was used the most, followed by the V 畀 IO DO pattern. The 畀 IO V DO was not used at all due to its identical structure with the causative/passive construction. We also examined the use of these non-native patterns with respect to the nature of the verbs, IOs and DOs. We note that the adoption of these patterns cannot simply be explained by means of discourse factors. In many cases, the DOs are not long when compared with the IOs. Furthermore, we observed that certain double-object verbs were used more frequently with the non-native patterns. Among them is the verb [pei] 畀 ‘to give’, which was used the most with the non-native V IO DO pattern. This illustrates that the spread of the non-native pattern usage started with the core member in the double-object verb family. Table 41 summarizes the use of the two non-native patterns collected in the production task. The data shows that a total of 4421 double-object sentences were collected in the production task. Among these sentences, 276 (6.2%) were expressed with the non-native patterns. This low percentage illustrates that the influence of Putonghua Table 41. Distribution of non-native double-object patterns collected in the production task Typological pattern
Pattern
Frequency
DO-IO IO-DO
V DO (畀) IO V IO DO V 畀 IO DO
4145 (93.8%) 244 (5.5%) 32 (0.7%) 4421
Total
Chapter 7. On-going word order change in Cantonese double-object construction 159
or Modern Standard Chinese is not yet significant but their influence on Hong Kong Cantonese might have started. Another characteristic of this on-going syntactic change is that it is the younger informants, especially those between 20 and 30 years old, who took the lead in using the non-native patterns. We have argued that younger informants usually have more exposure to Modern Standard Chinese as well as Putonghua. Thus, there is a correlation between the age of the informants and the usage of the non-native double-object patterns. As noted before, some informants were taking Putonghua courses or were working in an environment where Putonghua was frequently used. These external factors undoubtedly sped up the syntactic change. Table 42 gives the distribution of the native and the non-native patterns in terms of ages of the informants. Table 42. Use of the non-native patterns in terms of age groups253 Age group
No. of informant
IO DO253
DO IO
Ratio of IO DO : DO IO
20 or below 21–25 26–30 31–35 36–45 46–55 56 or above
5 13 6 7 2 3 4
27 114 37 70 12 12 4
529 1304 595 726 217 296 478
1 : 19.6 1 : 11.4 1 : 16.1 1 : 10.4 1 : 18.1 1 : 24.7 1 : 119.5
It is noted that the V 畀 IO DO pattern was mainly used by younger informants, especially those between 21 and 35 years old. The youngest group (i.e., those below 20) did not have the highest number of non-native sentences. This may be due to the fact that some of them were secondary school students and their exposure to Modern Standard Chinese and Putonghua was not as high as those informants in the mid-20s, who have started working in different fields. We have also shown that students did not have the highest frequency of non-native patterns (Table 38). Instead, informants whose jobs requiring substantial use of Putonghua or Modern Standard Chinese produced more double-object sentences with the non-native patterns. Although the overall usage is relatively small (only 6.2% of the sentences were expressed with the non-native patterns), we would expect that in the future, the use of these non-native patterns will continuously increase since the contact with Putonghua and Modern Standard Chinese will become more intense, as discussed in Section 7.1. 253. This includes both the V IO DO and V 畀 IO DO patterns.
160 Cantonese give and Double-Object Construction
7.3.3 The perception task In the production task, the informants only gave one syntactic pattern for each sentence, which represented the most natural pattern in their speech. However, there can be other patterns acceptable to the informants in addition to the one they produced. We thus designed a perception task where informants were given different double-object patterns carrying the same meaning. The informants were asked to indicate their acceptance (i.e., (a) Totally acceptable; (b) Marginally or moderately acceptable; (c) Unacceptable) toward these sentences as well as to rank the degree of acceptability among these syntactic variants. The list of sentences used in the perception task can be found in Appendix 6. 7.3.3.1 The native pattern: V DO (畀) IO In Section 4.3.3, we showed that the V DO 畀 IO can be contracted to V DO IO by dropping the IO marker [pei] 畀.254 This frequently takes place when the main verb is [pei] 畀 ‘to give’ which is homophonous with the IO marker. In other words, sentences with other double-object verbs are preferred to have the overt IO marker [pei] 畀. Therefore, we can divide our sentences into two types: One with the IO marker and the other without. The latter is mainly found in those sentences with the main verb [pei] 畀 ‘to give’. The acceptability of those sentences in which the verbs are not [pei] 畀 is summarized in Table 43. Table 43. Acceptability of the V DO 畀 IO pattern with verbs other than 畀255256
Verb255
3a. 遞 ‘to pass’ 4c. 找 ‘to give change’ 5b. 送 ‘to give’ 7. 分 ‘to divide’ 8. 交 ‘to handover’ 9c. 送 ‘to give’ 10. 還 ‘to return’ 13. 送 ‘to give’ 14a. 寄 ‘to send’ Total
Acceptability
High256
Moderate
Nil
40 38 39 39 40 39 39 40 39 431
0 2 1 1 0 1 1 0 1 9
0 0 0 0 0 0 0 0 0 0
254. Recall that when the IO marker was still [kʷͻ] 過 in 19th century Cantonese, the IO marker was usually present even when the verb was [pei] 畀 and the DO is short. 255. For easy reference, the sentence numbers used in the perception task are also listed. 256. The numbers indicate the number of informants (out of 40) giving this option.
Chapter 7. On-going word order change in Cantonese double-object construction 161
It is clear that the native pattern with the IO marker [pei] 畀 was preferred and was well accepted by the informants. When we examine those sentences with [pei] 畀 as the main verb, the result is different. There are four such sentences, as summarized in Table 44. Table 44. Acceptability of the V DO 畀 IO pattern with the main verb 畀
Verb
1b. 畀 ‘to give’ 2b. 畀 ‘to give’ 6a. 畀 ‘to give’ 6d. 畀 ‘to give’ Total
Acceptability
High
Moderate
Nil
6 10 14 12 42
17 16 20 19 72
17 14 6 8 45
From Table 44, we note that there were 45 instances (out of 159) rated unacceptable. This is due to the repetition of the same morpheme within a short distance. In spite of the application of haplology, there are still differences in terms of the acceptability among these four sentences: Sentences 6a and 6d were more acceptable with the overt IO marker than sentences 1b and 2b. This can be possibly related to the lengths of the DOs. In 6a and 6d, the DOs are long (compared with those in 1b and 2b). Informants used the IO marker even though it is identical with the main verb. These four sentences (1b, 2b, 6a and 6d) are listed as (131)–(134) respectively. (131) 唔該 畀 三 個 郵票 畀 我 m̩11-kͻi55 pei35 sam55 kͻ33 jɐu11-p’iu33 pei35 ŋͻ13 please give three cl stamp give I ‘Could you please give me three stamps?’ (132) 我 諗 如果 你 繼續 畀 拔蘭地 ŋͻ13 nɐm35 jy11-kwͻ35 nei13 kɐi33-tsʊk1 pei35 pɐt1-lan55-tei22–35 I think if you continue give brandy 畀 佢, 佢 會 重 飲 得 落 pei35 k’ɵy13, k’ɵy13 wui13 tsʊŋ22 jɐm35 tɐk5 lͻk1 give he/she, he/she will still drink able down ‘I think if you keep giving him/her brandy, he/she can still drink it’
(133) 我 畀 咗 兩 萬 一 千 五 百 文 13 35 35 13 22 5 55 13 3 ŋͻ pei tsͻ lœŋ man jɐt ts’in ŋ̩ pak mɐn55 I give asp two ten-thousand one thousand five hundred dollar 畀 佢 pei35 k’ɵy13 give he/she ‘I gave him/her 21,500 dollars’
162 Cantonese give and Double-Object Construction
(134) 你 畀 咗 兩 萬 一 千 五 百 文 nei13 pei35 tsͻ35 lœŋ13 man22 jɐt5 ts’in55 ŋ̩13 pak3 mɐn55 you give asp two ten-thousand one thousand five hundred dollar 畀 邊個 pei35 pin55-kͻ33 give who ‘To whom did you give 21,500 dollars?’
Sentences with verbs other than [pei] 畀 ‘to give’ were not preferred to be expressed without the IO marker. Their acceptability is summarized in Table 45. Table 45. Acceptability of the V DO IO pattern Verb 4a. 找 ‘to give change’ 5c. 送 ‘to give’ 9b. 送 ‘to give’ 10c. 還 ‘to return’ 14. 寄 ‘to send’ 14c. 寄 ‘ to send’ 1a. 畀 ‘to give’ 2a. 畀 ‘to give’ 6b. 畀 ‘to give’ 6c. 畀 ‘to give’
Acceptability High
Moderate
Nil
2 0 0 2 0 0 33 30 15 8
8 8 9 6 0 0 5 10 22 25
21 9 12 11 3 4 2 0 3 6
With IO marker 畀 9 23 19 21 37 36 0 0 0 1
It is clear from the data that sentences with main verbs other than [pei] 畀 ‘to give’ were preferred to have the IO marker. Sentences with [pei] 畀 ‘to give’ as the main verb were acceptable even if the IO marker is elided. Sentences 6b and 6c were less acceptable than 1a and 2a because the DOs were longer than those in 1a and 2a. Sentence 7c uses the IO marker [kʷͻ] 過. As discussed in Chapter 4, [kʷͻ] 過 was the IO marker in 19th century Cantonese, which was then replaced by [pei] 畀 gradually starting from the 20th century. We suggest that the use of [kʷͻ] 過 as the IO marker is less acceptable than [pei] 畀. Table 46. Acceptability of sentences with the IO markers 過 and 畀 Verb
7c. 分 ‘to divide’ (with the IO marker 過) 7. 分 ‘to divide’ (with the IO marker 畀)
Acceptability
High
Moderate
Nil
3 39
7 1
30 0
Chapter 7. On-going word order change in Cantonese double-object construction 163
In this section, we show that the native pattern V DO 畀 IO was acceptable to the informants. The alternate pattern with the omission of the IO marker was acceptable mainly when the main verb was also [pei] 畀 especially when the DO was long. We also show that the IO marker 過 was less preferred than 畀. We now discuss the non-native patterns of V IO DO and V 畀 IO DO, followed by 畀 IO V DO. 7.3.3.2 The V IO DO and V 畀 IO DO patterns The acceptability of the V IO DO and the V 畀 IO DO patterns was not as high as the one of the native pattern. Both the V IO DO and V 畀 IO DO patterns have similar degrees of acceptability. Let us first examine those sentences with [pei] 畀 as the main verb. Table 47. Acceptability of the V IO DO pattern with the main verb 畀257 Acceptability257
Verb
1. 畀 ‘to give’ 2. 畀 ‘to give’ 6. 畀 ‘to give’ Total
High
Moderate
Nil
3 3 17 23
21 7 18 46
16 30 5 51
Although the main verb of these three sentences is [pei] 畀, their acceptability is not the same. Sentence 6 received the highest degree of acceptability because the DO is long and informants were inclined to move it toward the end of the sentence as a result of Heavy Noun Phrase Shift (Matthews & Yip, 1994). For sentence 1, more than half of the informants (i.e., 24 out of 40) accepted it as a legitimate sentence even though the DO was not long. This again shows that discourse factor is not the only factor contributing to the use of the V IO DO pattern. Table 48 below provides the acceptability of the V IO DO and V 畀 IO DO patterns (before and after the slashes respectively) with other double-object verbs.
257. We do not have the V 畀 IO DO pattern for the verb 畀 in the task because the main verb 畀 and the IO marker 畀 appear together when expressed with this pattern, i.e., V畀畀IO DO.
164 Cantonese give and Double-Object Construction
Table 48. Acceptability of the V IO DO and V 畀 IO DO patterns with other double-object verbs Verb
Acceptability
5a. 送 ‘to give’ 9. 送 ‘to give’ 4. 找 ‘to give change’ 13. 送 ‘to give’ 7. 分 ‘to divide’ 5d. 送 ‘to give’ 14. 寄 ‘to send’ 10. 還 ‘to return’ 3. 遞 ‘to pass’ 8. 交 ‘to handover’ Total
High
Moderate
Nil
2/1 2/1 2/21 1/0 1/0 0/1 0/3 0/1 0/0 0/0 8/9
24/22 24/17 22/18 7/1 0/3 21/22 11/18 3/16 2/16 0/2 114/135
14/17 14/22 16/20 32/39 39/37 19/17 29/19 37/23 38/24 40/38 278/256
Since the V IO DO pattern is not a native double-object sentence pattern, it is then expected that only a few informants consider it (highly) acceptable. But still, for those sentences involving double-object verbs other than 畀, 122 sentences out of 400258 (i.e., 30.5%) were rated acceptable by the informants. The acceptability of the pattern V 畀 IO DO is similar to the one of the V IO DO pattern. 144 sentences out of 400 (i.e., 36%) were rated acceptable. Among these 10 verbs, some verbs were preferred to be expressed by the V 畀 IO DO pattern. These include [kei] 寄 ‘to send’, [tɐi] 遞 ‘to pass’, [wan] 還 ‘to return’ and [sʊŋ] 送 ‘to give’. More than half of the informants found the V 畀 IO DO pattern acceptable with these verbs. The overall acceptability of this pattern with double-object verbs (including [pei] 畀) is shown in Table 49. Table 49. Summary of the acceptability of the V IO DO and V 畀 IO DO patterns
V IO DO V 畀 IO DO
Acceptability
High
Moderate
Nil
31 (6.0%) 9 (2.3%)
160 (30.8%) 135 (33.8%)
329 (63.3%) 256 (64.0%)
The acceptability of the V IO DO pattern is much greater than the actual use of this pattern observed in the production task (6.2% only). This demonstrates that the 258. There are 10 verbs in the above table and there are 40 informants. Therefore, the total number of responses we obtained was 400.
Chapter 7. On-going word order change in Cantonese double-object construction 165
V IO DO pattern is gaining its status in Hong Kong Cantonese and is regarded as one of the alternate double-object patterns in addition to the native V DO 畀 IO pattern. In addition to acceptability, we also note that some sentences were rejected more than other sentences. For example, almost all informants rejected sentences 7b, 8b and 10b (repeated below as (135)–(137) respectively). It is noted that the syntactic structure of the DOs and the IOs are correlated with the degree of unacceptability. Some informants commented that these sentences expressed with the V IO DO pattern sound incomplete or even ungrammatical. (135) 你 分 啲 小朋友 啲 蛋糕 啦 nei13 fɐn55 ti55 siu35-pɐŋ11-jɐu13 ti55 tan33-kou55 la55 you divide cl children cl cake sfp ‘Please divide the cake among the children’
(136) 你 交 咗 邊個 啲 功課 呀? nei13 kau55 tsͻ35 pin55-kͻ33 ti55 kʊŋ55-fͻ33 a33 you submit asp who cl homework sfp ‘To whom did you pass the homework?’
(137) 我 還 返 佢 上 個 禮拜 整 爛 咗 個 ŋͻ13 wan11 fan55 k’ɵy13 sœŋ22 kͻ33 lɐi13-pai33 tsιŋ35 lan22 tsͻ35 kͻ33 I return suf he/she last cl week damage asp cl blue 藍色 公仔 lam11-sιk5 kʊŋ55-tsɐi35 doll ‘I returned him/her the blue doll that I damaged last week’
Most informants did not accept these three sentences because they interpreted the intended IOs and DOs together as possessive noun phrases. In other words, [ti55 siu35 pɐŋ11 jɐu13 ti55 tan22 kou55] 啲小朋友啲蛋糕‘cl-small-kid-cl-cake’, [pin55 kͻ33 ti55 kʊŋ55 fͻ33] 邊個啲功課 ‘who-cl-assignment’ and [k’ɵy13 sœŋ33 kͻ33 lɐi13 pai33 tsιŋ35 lan22 tsͻ35 kͻ33 lam11 sιk5 kʊŋ55 tsɐi35] 佢上個禮拜整爛咗個藍色公 仔 ‘s/he-last week-damage-ASP-cl-blue-doll’ in (135), (136) and (137) were interpreted as ‘the kids’ cakes’, ‘whose assignment’ and ‘the blue doll that s/he damaged last week’ respectively.259 In (137), the two noun phrases were interpreted as a relative clause. Informants found the above sentences incomplete because the sentences lack the IOs.
259. Classifiers in Cantonese have a number of functions that standard Mandarin does not have. For example, noun phrases with the ‘cl + Noun’ structure can have a specific or definite reading. The ‘Noun/Pronoun + cl + Noun’ phrase can be interpreted as a possessive noun phrase. Thus, [ŋͻ tsεk mau] 我隻貓 ‘I-cl-cat’ and [ma ma t’iu kwɐn] 媽媽條裙 ‘mother-cl-skirt’ mean ‘my cat’ and ‘mother’s skirt’ respectively. For details on the classifier usage in the Yue dialects, see Cheung (1989). fn260
CIT65
166 Cantonese give and Double-Object Construction
Those sentences in which the IOs and DOs do not have the possible interpretation of the possessive construction received a higher degree of acceptability. Consider sentences 4, 5a, 5d, 9 and 14d which are listed as (138)–(142) respectively: (138) 你 找 返 我 兩 文 先 啱 nei13 tsau35 fan55 ŋͻ13 lœŋ13 mɐn55 sin55 ŋam55 you change suf I two dollar first right ‘It is only correct if you give me $2 change’
(139) 我 送 你 一 本 書 ŋͻ13 sʊŋ33 nei13 jɐt5 pun35 sy55 I give you one cl book ‘I gave you a book’
(140) 你 想 送 佢 乜嘢? nei13 sœŋ35 sʊŋ33 k’ɵy13 mɐt5-jε13 you want give s/he what ‘What do you want to give him/her?’
(141) 我 送 你 呢 一 枝 筆 ŋͻ13 sʊŋ33 nei13 nei55 jɐt5 tsi55 pɐt5 I give you this one cl pen ‘I gave you this pen’
(142) 你 寄 佢 一 本 好 貴 嘅 書 nei13 kei33 k’ɵy13 jɐt5 pun35 hou35 kwɐi33 kε33 sy55 you mail s/he one cl very expensive de book ‘You mailed him/her a very expensive book’
In the above five sentences, the IOs and DOs can potentially be possessive noun phrases (except (139) and (142)) but the acceptability as a double-object sentence is still high with the other three sentences. This may be due to the fact that the main verbs (送 and 找) are core double-object verbs in the sense that when two noun phrases are present with these core double-object verbs, there is a tendency to interpret them as two objects rather than a possessive noun phrase. To further investigate the acceptability of the V IO DO pattern, three pairs of sentences, with different word order of the DOs and IOs, were checked with the informants. These three sentences, together with their acceptability, are presented in Table 50.
Chapter 7. On-going word order change in Cantonese double-object construction 167
Table 50. Comparison of the acceptability between the V IO DO and V DO 畀 IO patterns
260
15. 我會畀大家一個交代 ‘I will give you all an explanation’ 16. 你不如畀佢多一次機會 ‘Why don’t you give him another chance?’ 17. 我好多謝公司畀我呢個大獎 ‘I would like to thank the company for giving me this top prize’
Acceptability261 (compared with V DO IO) Higher
Equal
Lower
27
12
1
6
24
10
7
28
5
The majority of the informants found the V IO DO pattern acceptable and again, the sentence [ŋͻ13 wui35 pei35 tai22 ka55 jɐt5 kͻ33 kau55 tͻi22] 我會畀大家一個交 代 ‘I will give you all an explanation’ was the most acceptable among the three sentences. The majority found (16) and (17) as equally good as the ones expressed with the V DO IO pattern. In Modern Standard Chinese or Putonghua, some verbs expressed with the V IO DO pattern can have either the meanings of deprivation or giving. Typical examples include zu 租 ‘to rent / to rent out’, jie 借 ‘to borrow / to lend’. The following two sentences in Modern Standard Chinese are ambiguous. (143) 我 借 了 他 三十塊 wo jie le ta san shi kuai I borrow/lend asp he thirty dollar ‘I borrowed thirty dollars from him’ or ‘I lent him thirty dollars’
(144) 我 租 了 她 兩 個 房子 wo zu le ta liang ge fangzi I rent asp she two cl room ‘I rented two rooms to her’ or ‘I rented two rooms from her’
In Cantonese, the two meanings are expressed with different word order patterns. The V IO DO pattern is used for the meaning of deprivation while the V DO 畀 IO pattern for the meaning of giving. There is thus no structural ambiguity for these verbs when expressed with the V IO DO pattern in Cantonese. However, if the informants were under the influence of Modern Standard Chinese or Putonghua, they may have begun using the V IO DO pattern to express the meaning of giving. In other words, this V IO DO pattern becomes ambiguous between the [+deprive] 260. We do not have the V 畀 IO DO pattern for the verb 畀 in the task because the main verb 畀 and the IO marker 畀 appear together when expressed with this pattern, i.e., V畀畀IO DO.
168 Cantonese give and Double-Object Construction
and [+GIVE] meanings. We have shown that, in the production task, eight informants used the V IO DO pattern to express the meaning of giving for these two verbs. To further investigate this kind of change, we included sentences with these two verbs in the perception task. Sentences 11 and 12 from the perception task are listed below as (145) and (146) respectively: (145) 我 借 咗 佢 三 本 書 ŋͻ13 tsε33 tsͻ35 k’ɵy13 sam55 pun35 sy55 I borrow asp s/he three cl book
(146) 我 租 咗 佢 兩 間 屋 ŋͻ13 tsou55 tsͻ35 k’ɵy13 lœŋ13 kan55 ʊk5 I rent asp s/he two cl house
When checking the meanings our informants obtained from these two sentences, we did not specifically use the notions of deprivation and giving. Instead, we asked them ‘to whom the three books belong’ and ‘to whom the two houses belong’. If the informant has been receiving the influence of Putonghua or Modern Standard Chinese, s/he would obtain the two meanings: the books and the houses can belong to the subject or the IO. The results are summarized in Table 51. Table 51. Interpretation of the meaning for 租 and 借 in the V IO DO pattern
Sentence 11 12
Verb 借 租
deprivation only
deprivation & giving
32 34
8 6
Eight and six informants obtained both the meanings of deprivation and giving for sentences 11 and 12 respectively. This indicates that some informants have started to interpret the sentences with the meaning of giving. Again, such dual meanings all come from the young informants. The age groups of these informants are summarized in Table 52. Table 52. Correlation of age groups with the dual meanings of the V IO DO pattern for 租 and 借
Age group
20 or below 21 – 25 26 – 30 31 – 35
No. of informant
租 1 2 3 2
借 1 2 1 2
Chapter 7. On-going word order change in Cantonese double-object construction 169
7.3.3.3 The 畀 IO V DO pattern In the production task, we noted that the 畀 IO V DO pattern was never used by any informants because this pattern has the same surface structure as the causative construction. This is confirmed by the data collected from the perception task. In this task, there were four sentences with this pattern and their acceptability is summarized in Table 53. Table 53. Acceptability of the 畀 IO V DO pattern
Verb
3c. 遞 ‘to pass’ 8c. 交 ‘to pass’ 9d. 送 ‘to give’
Acceptability
Accept
Reject
0 0 0
33 5 25
Causative meaning 7 35 15
Similar to the results obtained in the production task, no informant accepted this pattern as a legitimate double-object sentence. The general comment received from the informants is that this pattern sounds like a causative construction in which the agent allows someone to carry out a second act. For example, sentence 8c, repeated below as (147), was interpreted by some informants as ‘Whom did you allow to submit the assignment?’ instead of ‘To whom did you submit the assignment?’ (147) 你 畀 邊個 交 咗 啲 功課 呀? nei13 pei35 pin55-kͻ33 kau55 tsͻ35 ti55 kʊŋ55-fͻ33 a33 you give who submit asp cl assignment sfp Intended meaning: ‘To whom did you submit your assignment?’ Meaning obtained by informants: ‘Whom did you allow to submit the assignment?’
The reason of rejecting sentences 3c and 9d can be attributed to the semantics of the IO and the double-object verb concerned.261 Sentences 3c and 9d are repeated below as (148) and (149) respectively: (148) 請 你 畀 老闆 遞 呢 份 嘢 ts’εŋ35 nei13 pei35 lou13-pan35 tɐi22 nei55 fɐn22 jε13 please you give boss pass this cl thing Intended meaning: Please pass this document to the boss Meaning obtained by informants: Please allow boss to pass this document
261. These comments were given by some informants only.
170 Cantonese give and Double-Object Construction
(149) 我 畀 你 送 呢 一 枝 筆 ŋͻ13 pei35 nei13 sʊŋ33 nei13 jɐt5 tsi55 pɐt5 I give you give this one cl pen Intended meaning: I give you this pen Meaning obtained by informants: I allow you to give this pen (to someone)
According to some informants, sentence (148) sounds unnatural because it is inconceivable that someone permits his/her superior to do something. In other words, the sentence cannot even convey the meaning of permission. For sentence (149), the pre-verbal 畀-IO phrase was interpreted with the meaning of ‘to permit someone’ and there is then no IO in the sentence because [sʊŋ] 送 ‘to give’ is an intrinsic [+GIVE] verb which takes an IO obligatorily. In this sense, the sentence sounds incomplete and is thus rated unacceptable by the informants. 7.3.3.4 Summary of the perception task The acceptability of the various non-native patterns (in percentage) in relation to the age groups of the informants is summarized in Table 54. Table 54. Acceptability of the non-native patterns with respect to age groups Age group 20 or below 21–35 26–30 31–35 36–45 46–55 56 or above
V IO DO H
M
R
H
9.8 3.6 8.0 7.3 4.2 5.6 8.3
23.0 37.0 26.7 43.9 25.0 41.7 20.8
67.2 59.4 65.3 48.8 70.8 52.8 70.8
2.9 2.9 4.4 3.4 0 0 0
V 畀 IO DO M
R
H
28.6 36.3 37.8 27.1 22.2 44.4 44.4
68.6 60.8 57.8 69.5 77.8 55.5 55.5
0 0 0 0 0 0 0
畀 IO V DO M
R
0 0 0 0 0 0 0
100 100 100 100 100 100 100
‘H’, ‘M’ and ‘R’ refer to the three types of acceptability respectively: ‘High’, ‘Moderate’ and ‘Reject’. Among the three non-native patterns, the 畀 IO V DO pattern was rejected by all the informants as a double-object sentence pattern.262 Younger informants (those below 36) had a higher degree of acceptability toward the non-native patterns. It is interesting to note that it was not the youngest group (i.e., below 20 years old) that had the highest degree of acceptability. We have explained before that the main reason for using non-native patterns is related to the exposure to 262. As discussed before, some informants interpreted this sentence pattern with the causative meaning.
Chapter 7. On-going word order change in Cantonese double-object construction 171
Putonghua or Modern Standard Chinese. Informants between 25 and 36 years old had graduated from the universities and had begun working in society. The nature of their jobs could have increased their contact with Putonghua or Modern Standard Chinese. We have shown the correlation between the use of the non-native patterns and the occupations of the young informants (Table 38). When compared with Chinese medicine practitioners and lawyers specializing in Chinese business law, college students had relatively less exposure to Putonghua or Modern Standard Chinese (although they had Chinese language or Putonghua classes). The perception task helps us better understand how acceptable the non-native double-object sentences are. Our assumption is that not using or producing a pattern does not mean that the pattern was entirely rejected by the informants. The design of the perception task was to compare the relative acceptability among the various non-native patterns. We observe that, among the three patterns, the V IO DO pattern was the most acceptable except when the IO and the DO can be interpreted as a possessive noun phrase, which has potential ambiguity. Most of the sentences expressed with the V 畀 IO DO pattern were moderately accepted by the majority of the informants but this pattern was still less preferred when compared with the V IO DO pattern. Finally, the 畀 IO V DO was rejected by all informants. The major reason was that the surface syntactic structure is identical to the causative construction, which causes ambiguity or ungrammaticality. 7.4
Development of Cantonese double-object construction
The assumption underlying the current fieldwork is that Cantonese has been receiving the influence from Putonghua and Modern Standard Chinese. In Section 7.1, we discussed how these two linguistic varieties are situated in the sociolinguistic profile of Hong Kong. Therefore, the crucial factor contributing to the current syntactic change is language contact. Four decades ago, Peyraube’s fieldwork already provided us with some insights into the use of various non-native double-object patterns in Hong Kong Cantonese. With the current fieldwork with forty native informants of Hong Kong Cantonese, we are able to study closely the extent and the possible mechanism involved in this syntactic change. Among the three non-native double-object patterns, some were adopted by the informants more while some were less (such as the 畀 IO V DO pattern). We also examined the reasons leading to the different degrees of preferences toward the non-native patterns. The production task helps us find out which pattern was used
172 Cantonese give and Double-Object Construction
the most by the informants. The perception task allows us to look into the possible reasons for the (non-)preference of certain patterns. The rate of change is also related to the verb type. For example, in the production task, the verb [pei] 畀 was used the most with the V IO DO pattern. This can be explained by the fact that it is the basic and core double-object verb, as discussed in Chapter 1. Its frequent usage made it more susceptible to change. It is thus not surprising to find that sentences made up of this verb underwent the fastest change. The rate of change was also found to vary from person to person and from verb to verb but the general trend was that the extent of the change was correlated to the degree of the informants’ exposure and contact toward the languages that exert influence. Informants who produced and accepted the non-native double-object patterns tended to use Putonghua or Modern Standard Chinese more frequently. Generally speaking, younger informants with occupations requiring substantial use of Putonghua or Modern Standard Chinese produced and accepted the non-native patterns more. It is expected that the extent of the use of the non-native double-object patterns in Hong Kong Cantonese will continue in the future when the contact between Cantonese and Putonghua as well as Modern Standard Chinese becomes more intensive in Hong Kong.
Chapter 8
Concluding remarks and future work
This study focuses on the double-object verb GIVE and its associated functions and syntactic constructions in the Cantonese dialect. In addition to subcategorizing two syntactic objects, [pei] 畀 in Cantonese has similar behaviors as the double-object verbs in many world’s languages. These include the functions of an IO marker and a causative verb. Although the various syntactic functions of GIVE can be succinctly accounted for by grammaticalization, which is a language-internal development, this grammatical process is not immune to borrowing. Our survey of the multiple functions of GIVE in Chinese dialects and neighboring non-Sinitic languages in Chapter 2 demonstrates that the grammaticalization of GIVE can be considered one of the areal features in the Southeast Asian linguistic area and it serves as a good illustration of the notion contact-induced grammaticalization proposed by Heine and Kuteva (2003). The diachronic analysis of the syntactic functions performed by [pei] 畀 allows us to propose the mechanisms for the development. One major finding is the proposal of two types of IO markers in the Chinese language, the go-type and the give-type. The former is mainly performed by directional verbs while the latter by the double-object verbs. Our examination of oracle-bone inscriptions, pre-modern Cantonese materials, and the Yue dialect of Conghua allows us to postulate the semantic and syntactic factors for the switchover from the go-type to the give-type IO markers, as well as the chronology of the various functions of [pei] 畀. The major syntactic construction in which GIVE involves is the double-object construction. Two typological patterns in terms of the word order of the objects, IO DO and DO IO exist in Chinese dialects. The former is dominant in the Northern dialects while the latter in the Southern and Central dialects. Our fieldwork with native speakers of Hong Kong Cantonese shows that the non-native IO DO word order pattern is gaining its ascendency in Hong Kong Cantonese due to the increasing influence of Putonghua and Modern Standard Chinese. The adoption of the non-native pattern can be seen as a result of bilingualism, which is one of the factors causing linguistic change.264 Speakers who are bilingual 264. For the discussion on bilingualism as one of the causes of linguistic change, see Appel and Muysken (1987), Thomason (2001), Myers-Scotton (2002) and Winford (2003).
174 Cantonese give and Double-Object Construction
in two (or more) linguistic varieties possessing distinct typological features will sometimes unconsciously transfer the features from one variety to another. This is comparable to the case of code-switching in the sense that different linguistic varieties are used at the same time. Although all speakers used Cantonese during the interview, some of them have adopted the linguistic features typical to Putonghua (i.e., the use of the IO DO pattern). We argue that younger speakers are more proficient in both linguistic varieties than their seniors, due to their frequent exposure to Putonghua or Modern Standard Chinese used in the media, cyber world, working environments and schools. The observations from the fieldwork support Thomason’s (2001) claim that language change induced from language contact can be attributed to both social and linguistic predictors. The former includes the intensity of the contact between the two languages involved while the latter subsumes the typological differences between the source language and the recipient language. Although it has been pointed out that in contact linguistics, “the lexicon is most easily and radically affected, followed by the phonology, morphology and finally the syntax” (McMahon, 1996, p. 209), the fieldwork study shows that there were on-going changes for the double-object construction in Cantonese. Lastly, it should be pointed out that there are still some important issues deserving further investigations: a. When discussing the function of introducing instruments performed by the double-object verb [pei] 畀 in Cantonese, we also examine how this function stemmed from the core meaning of GIVE. In spite of the semantic correlation, this function is only found in pre-modern Cantonese and the modern Taishan dialect according to the extant materials we have consulted. Studies on Chinese dialectal grammar do not always include sentences involving the semantic role of instrument as well as its marker. b. In Chapter 4, we discussed the significance of the go-type IO marker from a cross-linguistic perspective and its replacement by the give-type counterpart in the Cantonese dialect. It should be noted that some languages spoken in the Southeast Asian linguistic area, such as Lahu and Thai, also have the go-type IO marker (as well as the beneficiary marker). It is meaningful to investigate whether the go-type IO marker is another areal feature in the Southeast Asian linguistic area but this requires more in-depth studies, particularly on the diachronic aspect of the languages concerned. c. Our study of the IO marker shows that the replacement of the go-type by the give-type took place in the first half of the 20th century. The same was also found for the passive marker (i.e., the replacement of [pei] 被 by [pei] 畀). In addition to these two markers, there were other syntactic constructions such as the neutral question that also underwent structural change around the period
Chapter 8. Concluding remarks and future work 175
(Cheung, 2001). It is thus relevant to ask the question why several syntactic changes took place around the same time. Was it a coincidence or a special time frame that initiated a series of changes in the Cantonese dialect? What was the driving force, language internal or external, leading to these changes? The grammaticalization process of Cantonese give reported in this book was examined with some pre-modern Cantonese textual materials, which faithfully recorded the language spoken or used in the time period concerned. However, we notice some of these materials were not colloquial enough. Some of them were textbooks with only individual sentences without context for understanding the actual usage of the morphemes concerned. In the past two decades, a number of Cantonese corpora with different types of spoken language data have been constructed. These corpus data are more dialogic and interactive which can provide authentic language data, and also the context for studying the use of the language. It is hoped that more research will be undertaken in future with the use of corpus data, especially the spoken ones which faithfully record the actual use of the language.
References
Adamson, Bob, & Auyeung, Winnie (1997). Language and the curriculum in Hong Kong: dilemmas of triglossia. In M. Bray & W. O. Lee (Eds.), Education and political transition: implications of Hong Kong’s change of sovereignty (pp. 87–100). Hong Kong: Comparative Education Research Centre, The University of Hong Kong. https://doi.org/10.1080/03050069728541 Ahn, Sang-Cheol (1998). Phonological aspects in causative suffixation. In R. King (Ed.), Description and explanation in Korean linguistics (pp. 67–84). Ithaca, NY: East Asia Program, Cornell University. Aikhenvald, Alexandra Y (2006). Grammars in contact: A cross-linguistic perspective. In A. Aikhenvald & R. M. W. Dixon (Eds.), Grammars in contact: A cross-linguistic typology (pp. 1–66). Oxford: Oxford University Press. Akitani, Hiroyuki (2001). 吳語江山廣豐方言研究 [A study of the Wu dialects of Jiangshan and Guangfeng]. Matsuyame: Ehime Daigaku Hobun Gakubu Sogo Seisaku Gakka. Akitani, Hiroyuki (2005). 浙南的閩東區方言 [The dialects of Eastern Min in southern Zhejiang]. Taipei: Institute of Linguistics, Academia Sinica. Anonymous. (1877). 散語四十章 [Forty chapters of prose]. Hong Kong: St. Paul’s College. Ansaldo, Umberto, Bisang, Walter, & Szeto, Pui Yiu (2018). Grammaticalization in isolating languages and the notion of complexity. In H. Narrog, & B. Heine (Eds.), Grammaticalization from a typological perspective (pp. 219–234). Oxford: Oxford University Press. https://doi.org/10.1093/oso/9780198795841.003.0011 Appel, Rene, & Muysken, Pieter (1987). Language contact and bilingualism. London: Edward Arnold. Bacon-Shone, John, Bolton, Kingsley, & Luke, Kang Kwong (2015). Language use, proficiency and attitudes in Hong Kong [香港居民語言應用、語言能力及語言態度研究]. Hong Kong: Social Sciences Research Centre, The University of Hong Kong. Ball, Dyer (1888). Cantonese made easy (2nd ed.). Hong Kong: China Mail Office. Ball, Dyer (1894). Readings in Cantonese colloquial. Hong Kong: Kelly & Walsh Limited. Ball, Dyer (1912). How to speak Cantonese (4th ed.). Hong Kong: Kelly & Walsh Limited. Ball, Dyer (1924). Cantonese made easy (4th ed.). Hong Kong: Kelly & Walsh Limited. Bao, Houxing (1998). 東安土話研究 [A study of the patois in Dong’an]. Hunan: Hunan jiaoyu chubanshe. Bao, Houxing, Cui, Zhenhua, Shen, Ruoyun. & Wu, Yunji (1999). 長沙方言研究 [A study of the Changsha dialect]. Changsha: Hunan jiaoyu chubanshe. Bauer, Robert (1988). Written Cantonese of Hong Kong. Cahiers de Linguistique Asie Orientale, 17(2), 245–293. https://doi.org/10.1163/19606028-90000305 Bauer, Robert (1996). Identifying the Tai substratum in Cantonese. In Pan-Asiatic Linguistics: Proceedings of the Fourth International Symposium on Languages and Linguistics, Volume V, (pp. 1806–1844). Bangkok: Mahidol University. Benedict, Helen (1979). Early lexical development: Comprehension and production. Journal of Child Language, 6(2), 183–200. https://doi.org/10.1017/S0305000900002245
178 Cantonese give and Double-Object Construction
Benedict, Paul (1972). Sino-Tibetan: A conspectus. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511753541 Berlinck, Rosane de Andrade (1996). The Portuguese dative. In William van Belle & Willy van Langendonck (Eds.), The dative: Descriptive studies, Volume 1 (pp. 119–154). Amsterdam: John Benjamins. https://doi.org/10.1075/cagral.2.07and Bisang, Walter (1996). Areal typology and grammaticalization: Processes of grammaticalization based on nouns and verbs in South-East Asian languages. Studies in Language, 20(3), 519–597. https://doi.org/10.1075/sl.20.3.03bis Bisang, Walter (1999). Classifiers in East and Southeast Asian languages: Counting and beyond. In J. Gvozdanovic (Ed.), Numeral types and changes worldwide (pp. 113–186). Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110811193.113 Bisang, Walter (2004). Dialectology and typology – an integrative perspective. In B. Kortmann (Ed.) Dialectology meets typology: Dialect grammar from a cross-linguistic perspective (pp. 11–46). Berlin: Mouton de Gruyter. Blake, Barry (1994). Case. Cambridge: Cambridge University Press. Bo, Wenze (2003). 木佬語研究 [A study of the Mulao language]. Beijing: Minzu chubnashe. Bolton, Kingsley (2001). The life and lexicography of Robert Morrison (1782–1834). Introduction to A vocabulary of the Canton dialect (originally printed in 1828). London: Ganesha. Bolton, Kingsley, & Bacon-Shone, John (2008). Bilingualism and multilingualism in the HKSAR: Language surveys and Hong Kong’s changing linguistic profile. In K. Bolton & Y. Han (Eds.), Language and Society in Hong Kong (pp. 25–51). Hong Kong Open University Press. Bonney, Samuel Williams (1853). Phrases in the Canton colloquial dialect. Guangzhou: n.p. Bradley, David (2003). Lisu. In G. Thurgood & R. LaPolla (Eds.), The Sino-Tibetan languages (pp. 222–235). London: Routledge. Bridgman, Elijah Coleman (1841). Chinese chrestomathy in the Canton dialect. Macao: S. Wells Williams. Bruce, R (1954). Cantonese lessons for Malayan students. Kuala Lumpur: Charles & Son Ltd. Bruche-Schulz, Gisela (1997). ‘Fuzzy’ Chinese: The status of Cantonese in Hong Kong. Journal of Pragmatics, 27(3), 295–314. https://doi.org/10.1016/S0378-2166(95)00048-8 Bryun, Adrienne, Muysken, Pieter, & Verrips, Maaike (1999). Double-object constructions in the creole languages: Development and acquisition. In M. DeGraff (Ed.) Language creation and language change: Creolization, diachrony, and development (pp. 239–373). Cambridge, Massachusetts: MIT Press. Bunyan, J (1871). The pilgrim’s progress: From this world to that which is to come 天路歷程 (G. Piercy, Trans.). Guangzhou: South China Religious Tract Society. Bybee, Joan, Perkins, Revere, & Pagliuca, William (1994). The evolution of grammar: Tense, aspect, and modality in the languages of the world. Chicago: University of Chicago Press. Carroll, Robert, & Prickett, Stephen (Eds.). (1997). The Bible: Authorized King James version. Oxford: Oxford University Press. Census and Statistics Department, Hong Kong Government. (1969). Hong Kong Statistics: 1947– 1967. Hong Kong: Census & Statistics Department, Hong Kong. Census and Statistics Department, Hong Kong Government. (2002). Main tables of the 2001 population census. In 2001 population census. Retrieved from https://www.censtatd.gov.hk/ major_projects/2001_population_census/main_tables/index.jsp. Chang, Song-hing (2000). 樂昌土話 [The patois of Lechang]. Xiamen: Xiamen daxue chubanshe. Chang, Song-hing, & Wan, Bo (2004). 連州土話研究 [A study of the patois of Lianzhou]. Xiamen: Xiamen daxue chubanshe.
References 179 Chao, Yuen Ren (1939). 鍾祥方言記 [A study of the Chung-hsiang dialect]. Academia Sinica Monograph, Series A, No. 15. Taipei: Institute of History and Philology, Academia Sinica. Chao, Yuen Ren (1968). A grammar of spoken Chinese. Berkeley: University of California Press. Chapman, Tim (1973). A practical guide to Cantonese conversation. Hong Kong: n.p. Chappell, Hilary (2000). Dialect grammar in two early modern Southern Min texts. Journal of Chinese Linguistics, 28(2), 247–302. Chappell, Hilary (2001). Synchrony and diachrony of Sinitic languages: A brief history of Chinese dialects. In H. Chappell (Ed.), Sinitic grammar: Synchronic and diachronic perspectives (pp. 3–28). Oxford: Oxford University Press. Chappell, Hilary (2015). Linguistic areas in China for differential object marking, passive, and comparative constructions. In H. Chappell (Ed.), Diversity in Sinitic Languages (pp. 13–52). Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198723790.003.0002 Chappell, Hilary, & Peyraube, Alain (2006). The analytic causatives of early modern Southern Min in diachronic perspective. In D. Ho, H. S. Cheung, W. Pan, & F. Wu (Eds.), 山高水長: 丁邦新先生七秩壽慶論文集 [Linguistic studies in Chinese and neighboring languages: Festschrift in honor of Professor Pang-Hsin Ting on his 70th birthday] (pp. 973–1011). Taipei: Institute of Linguistics, Academia Sinica. Chappell, Hilary, & Peyraube, Alain. (2015). The comparative construction in Sinitic languages: Synchronic and diachronic variation. In Hilary M. Chappell (Ed.), Diversity in Sinitic Languages (pp. 134–154). Oxford: Oxford University Press. Chen, Baoya (1996). 語言接觸與語言聯盟:漢越(侗台)語源關係的解釋 [Language contact and language union: An explanation on the genetic relationship of Chinese and Kam-Tai languages]. Beijing: Yuwen chubanshe. Chen, Changlai (2002). 現代漢語動詞的句法語義屬性研究 [A study of the syntactic and semantic properties of verbs in Modern Standard Chinese]. Shanghai: Xuelin chubanshe. Chen, Guoqing (2002). 克木語研究 [A study of the Khmu language]. Beijing: Minzu chubanshe. Chen, Guoqing (2005). 克蔑語研究 [A study of the Kemie language]. Beijing: Minzu chubanshe. Chen, Hailun, & Li, Lianjin (2005). 廣西語言文字使用問題調查與研究 [A study on language use in Guangxi]. Nanning: Guangxi jiaoyu chubanshe. Chen, Manhua (1995). 安仁方言 [The Anren dialect]. Beijing: Beijing yuyan xueyuan chubanshe. Chen, Xiaojin (1993). 東莞方言說略 [A brief description of the Dongguan dialect]. Guangzhou: Guangdong renmin chubanshe. Chen, Xiaojin (2003). 馬來西亞的三個漢語方言 [Three Chinese dialects in Malaysia]. Beijing: Chinese Academy of Social Sciences. Chen, Xiaojin (2004). 廣西玉林市客家方言調查研究 [A study of the Hakka dialects in the Yulin city of Guangxi]. Beijing: Chinese Academy of Social Sciences. Chen, Xiaojin, & Chen, Tao (2005). 廣西北海市粵方言調查研究 [A study of the Yue dialects in the Beihai city of Guangxi]. Beijing: Chinese Academy of Social Sciences. Chen, Yunlong (2006). 舊時正話研究 [A study of Jiushi Zhenghua]. Beijing: Zhongguo shehui kexue chubanshe. Chen, Zeping (1998). 福州方言研究 [A study of the Fuzhou dialect]. Fujian: Fujian renmin chubanshe. Chen, Zhangtai, & Li, Rulong (1991). 閩語研究 [A study of the Min dialects]. Beijing: Yuwen chubanshe. Cheng, Lisa, Huang, C.-T. James, Li, Audrey, & Tang, C-C Jane (1999). Hoo, hoo, hoo: Syntax of the causative, dative and passive constructions in Taiwanese. In P. H. Ting (Ed.), Contemporary studies on the Min dialects, JCL Monograph Series 14 (pp. 146–203). Berkeley: Project on Linguistic Analysis (POLA).
180 Cantonese give and Double-Object Construction
Cheng, Robert (1974). Causative constructions in Taiwanese. Journal of Chinese Linguistics, 2(3), 279–324. Cheng, Robert (1985). A comparison of Taiwanese, Taiwan Mandarin, and Peking Mandarin. Language, 61(2), 352–377. https://doi.org/10.2307/414149 Cheung, Samuel Hung-Nin (1972). 香港粵語語法研究 [Studies of the grammar of Hong Kong Cantonese]. Hong Kong: Chinese University of Hong Kong. Cheung, Samuel Hung-Nin (1989). 粵語量詞用法的研究 [A study of the usage of Cantonese classifiers]. In 中央研究院第二屆國際漢學會議論文集(語言文字組) [Proceedings of the Second International Conference on Sinology: Language and Linguistics] (pp.753–774). Taipei: Institute of History and Philology, Academia Sinica. Cheung, Samuel Hung-Nin (1996). Book review on Cantonese: A Comprehensive Grammar. International Review of Chinese Linguistics, 1(1), 129–133. Cheung, Samuel Hung-Nin (2000). 早期粵語中的變調現象 [Tone change in early Cantonese]. Fangyan 4, 299–312. Cheung, Samuel Hung-Nin (2001). The interrogative construction: (Re)constructing early Cantonese grammar. In H. Chappell (Ed.), Sinitic grammar: Synchronic and diachronic perspectives (pp. 191–231). Oxford: Oxford University Press. Cheung, Samuel Hung-Nin (2006). 早期粵語‘個’的研究 [“Ge” in early Yue]. In D. Ho, H. S. Cheung, W. Pan, & F. Wu (Eds.), 山高水長: 丁邦新先生七秩壽慶論文集 [Linguistic studies in Chinese and neighboring languages: Festschrift in honor of Professor Pang-Hsin Ting on his 70th birthday] (pp. 813–835). Taipei: Institute of Linguistics, Academia Sinica. Cheung, Samuel Hung-Nin (2009). Cantonese Made Easy: 早期粵語中的語氣助詞 [Sentence-final particles in early Cantonese]. Bulletin of Chinese Linguistics, 3(2), 131–170. https://doi.org/10.1163/2405478X-90000057 Cheung, Samuel Hung-Nin (2017). 語法講話:傳教士筆下的舊日粵語風貌 [The early stage of Cantonese described by the missionaries]. In S. H.-N. Cheung (Ed.), 一切從語言開始 [It all begins with language] (pp. 1–48). Hong Kong: The Chinese University of Hong Kong Press. Chiang, Chun-lung (2003). 兩岸大埔客家話研究 [A study of the Dapu Hakka dialects in Mainland China and Taiwan] (Unpublished doctoral dissertation). National Chung Cheng University, Jiayi, Taiwan. Chin, Andy Chi-on (2001, June). The development of the double-object construction in Cantonese. Paper presented at the Tenth International Association of Chinese Linguistics (IACL-10), Irvine, CA. Chin, Andy Chi-on (2009). The verb give and the double-object construction in Cantonese in synchronic, diachronic and typological perspectives [PhD thesis, University of Washington]. ProQuest Dissertations Publishing. Retrieved from https://www.proquest.com/ docview/305015809?accountid=11441 Chin, Andy Chi-on (2010a). The GO-type and the give-type indirect object markers in the Conghua dialect. In W. South Coblin & Anne O. Yue (Eds.), Studies in Honor of Jerry Norman 羅杰瑞先生七秩晉三壽慶論文集, pp. 245–276. Hong Kong: The Ng Tor-tai Chinese Language Research Centre, Institute of Chinese Studies, The Chinese University of Hong Kong. Chin, Andy Chi-on (2010b). Two types of indirect object markers in Chinese: Their typological significance and development. Journal of Chinese Linguistics, 38(1),1–25. Chin, Andy Chi-on (2010c). 粵語間接賓語標記的發展和相關語法現象 [The development of Cantonese indirect object markers and their related syntactic phenomena]. 語言學論叢 [Essays on linguistics], 42, 189–210.
References 181
Chin, Andy Chi-on (2011). Grammaticalization of the Cantonese double object verb [pei35] 畀 in typological and areal perspectives. Language and Linguistics (語言暨語言學), 12(3), 529–563. Chin, Andy Chi-on (2013). 粵語研究新資源: 《香港二十世紀中期粵語語料庫》[New resources for Cantonese language studies: A linguistic corpus of mid-20th century Hong Kong Cantonese]. 中國語文通訊 [Current research in Chinese linguistics], 92(1), 7–16. Chin, Andy Chi-on (2019). Initiatives of digital humanities in Cantonese studies: A corpus of mid-twentieth-century Hong Kong Cantonese. In A. W. B. Tso (Ed.), Digital humanities and new ways of teaching (pp. 71–88). Singapore: Springer. https://doi.org/10.1007/978-981-13-1277-9_5 Chin, Andy Chi-on (2021). 漢語方言語料庫的建構和應用––以《二十世紀中期香港粵語語 料庫》為例 [The construction and application of corpora of Chinese dialects: A case study of The Corpus of Mid-20th Century Hong Kong Cantonese]. 漢語語言學 [Chinese Linguistics], 1, 207–217. Chin, Andy Chi-on, & Tsou, Benjamin (2005). 粵語‘畀’字句語法化演變過程初探 [A preliminary investigation of the grammaticalization process of the give constructions in Cantonese]. In K. P. Tang & C. L. Tong (Eds.) 第九屆國際粵方言研討會論文集 [Proceedings of the ninth international conference on the Yue dialects] (pp. 302–311). Macau: Chinese Language Society of Macau. Chomsky, Noam (1985). The logical structure of linguistic theory. Chicago: University of Chicago Press. Chor, Winnie (2018). Directional particles in Cantonese: Form, function, and grammaticalization. Amsterdam: John Benjamins. https://doi.org/10.1075/scld.9 Chu, Junhai (2007). 桂南平話與白話的介詞研究 [A study of the prepositions in the Pinghua and Baihua of southern Guangxi] (Unpublished master thesis), Guangxi University, Nanning. Chu, Yiuwai (2001). 音樂敢言:香港「中文歌運動」研究 [A study of the “Chinese songs campaign” in Hong Kong]. Hong Kong: Infolink Publisher. Chu, Zexiang (1998). 邵陽方言研究 [A study of the Shaoyang dialect]. Hunan: Hunan jiaoyu chubanshe. Cowles, R. T (1920). Inductive course in Cantonese (2nd ed.). Hong Kong: Kelly & Walsh Limited. Crystal, David (2003). A dictionary of linguistics and phonetics. 5th edition. Malden: Blackwell. Cui, Zhenhua (1998). 益陽方言研究 [A study of the Yiyang dialect]. Hunan: Hunan jiaoyu chubanshe. Dai, Qingxia (2005). 浪速語研究 [A study of the Langsu language]. Beijing: Minzu chubanshe. Dai, Zhaoming (2003). 天台方言初探 [A preliminary study of the Tiantai dialect]. Beijing: Zhongguo shehui kexue chubanshe. Deng, Jun (2000). 開平方言 [The Kaiping dialect]. Changsha: Hunan dianzi yinxiang chubanshe. Dennys, Nicholas Belfield (1874). A handbook of the Canton vernacular Chinese language. London: Trubner & Co. Diller, Anthony (2001). Grammaticalization and Tai syntactic change. In M. R. Kalaya Tingsabadh & A. S. Abramson (Eds.), Essays in Tai linguistics (pp. 139–176). Bangkok: Chulalongkorn University Press. Ding, Jiayong, & Zhang, Min (2015). 從湘方言動詞句式看雙及物結構語義地圖 [Studies of the semantic map of ditransitive construction with verbal sentences in the Xiang dialects]. In X. Li, M. Zhang, & R. Guo (Eds.), 漢語多功能語法形式的語義地圖研究 [Semantic map analysis of Chinese syntactic structures with multiple functions] (pp. 234–255). Beijing: Commercial Press. CIT83
182 Cantonese give and Double-Object Construction
Dixon, Robert Malcolm Ward (1973). The semantics of giving. In M. Gross, M. Halle & M. Schutzenberger (Eds.), The formal analysis of natural languages (pp. 205–223). Hague: Mouton. https://doi.org/10.1515/9783110885248-013 Dryer, Matthew (1986). Primary objects, secondary objects, and antidative. Language, 62(4), 808–845. https://doi.org/10.2307/415173 Enfield, Nick. J (2001). On genetic and areal linguistics in Mainland South-East Asia: Parallel polyfunctionality of ‘acquire’. In A. Y. Aikhenvald & R.M.W. Dixon (Eds.) Areal diffusion and genetic inheritance: Case studies in language change (pp. 255–290). Oxford: Oxford University Press. Enfield, Nick. J (2003). Linguistic epidemiology: Semantics and grammar of language contact in Mainland Southeast Asia. London: Routledge. Enfield, Nick. J (2005). Areal linguistics and mainland Southeast Asia. The Annual Review of Anthropology, 34, 181–206. https://doi.org/10.1146/annurev.anthro.34.081804.120406 Erteschik-Shir, Nomi (1979). Discourse constraints on ditransitive movement. In T. Givon (Ed.), Syntax and semantics: Discourse and syntax. Volume 12 (pp. 441–467). New York: Academic Press. Fagerli, Ole Torfinn (2001). Malefactive by means of give. In H. G. Simonsen & R. T. Endresen (Eds.), A cognitive approach to the verb, morphological and constructional perspectives (pp. 203–222). Berlin: Mouton de Gruyter. Fang, Songxi (1993). 舟山方言研究 [A study of the Zhoushan dialect]. Beijing: Shehui kexue wenxian chubanshe. Fei, Jia, & Sun, Li (1993). 南京方言志 [A description of the Nanjing dialect]. Nanjing: Nanjing chubanshe. Ferguson, Charles (1959). Diglossia. Word, 15, 325–340. https://doi.org/10.1080/00437956.1959.11659702 Fletcher, Paul, Leung, Cheung-shing, Stokes, Stephanie, & Weizman, Zehava (2000). Cantonese preschool language development: A guide. Hong Kong: Department of Speech and Hearing Sciences, The University of Hong Kong. Fu, Changzhong (2005). 黎語坡春話概況 [A description of the Pochun dialect of the Li language]. 民族語文 [Minority Languages of China], 6, 55–71. Fulton, Albert Andrew (1888). Progressive and idiomatic sentences in Cantonese colloquial. Hong Kong: Kelly & Walsh Limited. Gao, Huanian (1980). 廣州方言研究 [A study of the Guangzhou dialect]. Hong Kong: Commercial Press. Gao, Yongqi (2004). 布興語研究 [A study of the Buxing language]. Beijing: Minzu chubanshe. Georgakopoulos, Thanasis (2019). Semantic maps. Oxford bibliographies in linguistics. Oxford: Oxford University Press. https://doi.org/10.1093/obo/9780199772810-0229 Giles, Howard, Bourhis, Richard Y., & Taylor, D. M (1977). Towards a theory of language in ethnic group relations. In H. Giles (Ed.) Language, ethnicity and intergroup relations (pp. 307– 348). London: Academic Press. Goddard, Cliff (2005). The languages of east and southeast Asia. Oxford: Oxford University Press. Gong, Hwang-cherng (2006). 漢語與苗瑤語同源關係的檢討 [An evaluation of the genetic relationship between Chinese and Miao-Yao]. Bulletin of Chinese Linguistics, 1(1), 255–270. Gospel of Luke [路加傳福音書]. (1873). Shanghai: Meihua shuguan. Gospel of Luke. (1931). New and old testaments (in Cantonese) (廣東話)新舊約全書. Hong Kong: Hong Kong Bible Society.
References 183 Gospel of Luke. (1997). The holy bible – New Cantonese bible 聖經 – 新廣東話. Hong Kong: Hong Kong Bible Society. Gu, Yang (1999). 雙賓語結構 [The double-object construction]. In Liejiong Xu (Ed.), 共性與個 性:漢語語言學中的爭議 [Universality and uniqueness: Some issues in Chinese linguistics] (pp. 91–110). Beijing: Beijing yuyan wenhua daxue chubanshe. Guldin, Gregory Eliyu (1997). Hong Kong ethnicity of folk models and change. In Grant Evans & Maria Tam (Eds.), Hong Kong: The anthropology of a Chinese metropolis (pp. 25–50). Surrey: Curzon. Guo, Xiliang (1997). 介詞‘于’的起源和發展 [The origin and the development of the preposition yu]. 中國語文 [Studies of the Chinese language], 2, 131–138. Hanyu dacidian bianji weuiyuanhui. (1986). 漢語大字典 [A dictionary on Chinese characters]. Chengdu: Sichuan cishu chubanshe. Harley, Heidi (2002). Possession and the double object construction. In Pierre Pica (Ed.), Linguistic Variation Yearbook 2002 (pp. 31–70). Amsterdam: John Benjamins. https://doi.org/10.1075/livy.2.04har Harper, L (1874). That Sweet Story of Old [悅耳真言]. Guangzhou: Xiguan tongde dajie fuyintang. Hashimoto, Mantaro (1976). Language diffusion on the Asian continent. Computational Analyses of Asian and African Languages, 3, 49–65. Hashimoto, Mantaro (1985). 語言地理類型學 [A geographical typology of languages). (Zhihong Yu, Trans.). Beijing: Beijing daxue chubanshe. Hashimoto, Mantaro (1986). The altaicization of northern Chinese. In John McCoy & Timothy Light (Eds.), Contributions to Sino-Tibetan studies (pp. 76–100). Leiden: E. J. Brill. Hashimoto, Mantaro (1988). The structure and typology of the Chinese passive construction. In Masayoshi Shibatani (Ed.), Passive and voice (pp. 329–354). Amsterdam/Philadelphia: John Benjamins. https://doi.org/10.1075/tsl.16.11has Haspelmath, Martin (1990). The grammaticization of passive morphology. Studies in Language, 14(1), 25–72. https://doi.org/10.1075/sl.14.1.03has Haspelmath, Martin (2005). Argument marking in ditransitive alignment types. Linguistic Discovery, 3(1), 1–21. https://doi.org/10.1349/PS1.1537-0852.A.280 Haspelmath, Martin. (2006). Ditransitive alignment splits and inverse alignment. Functions of Language, 14(1), 79–102. https://doi.org/10.1075/fol.14.1.06has Haspelmath, Martin. (2013). Ditransitive constructions: The verb ‘give’. In by Matthew Dryer, Martin Haspelmath (Eds.), The world atlas of language structures online. Leipzig: Max Planck Institute for Evolutionary Anthropology. (Available online at http://wals.info/chapter/105, Accessed on 2021-03-13.). Hawkins, John. (1994). A performance theory of order and constituency. Cambridge: Cambridge University Press. He, Jiashan. (1983). 仡佬語簡志 [A description of the Gelao language]. Beijing: Minzu chubanshe. He, Kailin. (1999). 漵浦方言研究 [A study of the Xupu dialect]. Hunan: Hunan jiaoyu chubanshe. He, Wei. (1993). 洛陽方言研究 [A study of the Luoyang dialect]. Beijing: Shehui kexue wenxian chubanshe. He, Weitang. (1993). 增城方言志 [A description of the Zengcheng dialect]. Guangzhou: Guangdong renmin chubanshe.
184 Cantonese give and Double-Object Construction Hebeisheng Changlixian Xianzhi Bianzuan Weiyuanhui. (1984). 昌黎方言志 [A description of the Changli dialect]. Shanghai: Shanghai jiaoyu chubanshe. Heine, Bernd, & Kuteva, Tania. (2002). World lexicon of grammaticalization. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511613463 Heine, Bernd, & Kuteva, Tania. (2003). On contact-induced grammaticalization. Studies in Language, 27(3), 529–572. https://doi.org/10.1075/sl.27.3.04hei Ho, Daphne. (2004). On the acquisition of pei2-dative constructions in Cantonese (Unpublished MA thesis). The Hong Kong Polytechnic University, Hong Kong. Hobson, Benjamin. (1850). Dialogues in the Canton vernacular. Guangzhou: n. p. Hoh, Fuk-Tsz, & Belt, Walter. (1936). The revised and enlarged edition of a pocket guide to Cantonese. Guangzhou: Lingnan daxue. Hong Kong Examinations Authority. (2000). Annual Report. Hong Kong: Hong Kong Examinations Authority. Hong Kong Examinations Authority. (2001). Annual Report. Hong Kong: Hong Kong Examinations Authority. Hong Kong Examinations and Assessment Authority. (2002). Annual Report. Hong Kong: Hong Kong Examinations and Assessment Authority. Hong Kong Examinations and Assessment Authority. (2003). Annual Report. Hong Kong: Hong Kong Examinations and Assessment Authority.. Hong Kong Examinations and Assessment Authority. (2004). Annual Report. Hong Kong: Hong Kong Examinations and Assessment Authority. Hong Kong Examinations and Assessment Authority. (2005). Annual Report. Hong Kong: Hong Kong Examinations and Assessment Authority. Hopper, Paul J., & Traugott, Elizabeth. (1997). Grammaticalization. Cambridge: Cambridge University Press. Hou, Jingyi. (1998). 北京話音檔 [Sound file of the Beijing dialect]. Shanghai: Shanghai jiaoyu chubanshe. Hou, Jingyi, & Wen, Duanzheng. (1993). 山西方言調查研究報告 [A research report on Shanxi dialects]. Taiyuan: Shanxi gaoxiao lianhe chubanshe. Hovav, Malka Rappaport, & Levin, Beth. (2008). The English dative alternation: The case for verb sensitivity. Journal of Linguistics, 44(1), 129–167. https://doi.org/10.1017/S0022226707004975 Hu, Shuangbao. (1984). 文水方言志 [A description of the Wenshui dialect]. Taiyuan: Shanxi sheng shehui kexueyuan yuyan yanjiushi. Huang, Borong. (1959). 廣州話補語賓語的詞序 [The word order of complements and objects in the Guangzhou dialect]. 中國語文 [Studies of the Chinese language], 84, 275–276. Huang, Borong. (1966). 陽江話的幾種句式 [Some sentence patterns in the dialect of Yang jiang].中國語文 [Studies of the Chinese language], 217–218. Huang, Borong. (1996). 漢語方言語法類編 [A collection of Chinese dialectal grammar]. Qing dao: Qingdao chubanshe. Huang, Chu-Ren, & Ahrens, Kathleen. (1999). The function and category of gei in Mandarin ditransitive construction. Journal of Chinese Linguistics, 27(2), 1–26. Huang, Jiajiao, & Zhan, Bohui. (1983). 廣州方言中的特殊語序現象 [Some issues on the word order in the dialect of Guangzhou].語言研究 [Studies in language and linguistics], 2, 121–126. Huang, Xiaoxue. (2014). 宿松方言語法研究 [A study of the grammar of the Susong dialect]. Beijing: Zhongguo shehui kexue chubanshe.
References 185 Huang, Xiaoxue, & He, Xuegui. (2018). 漢語方言與“給”義動詞相關的受益格標記 [Benefactive markers related to verbs meaning “to give” in Chinese dialects]. 語言研究 [Studies in language and linguistics], 38(4), 80–85. Hudak, Thomas John. (1991). William J. Gedney’s the Tai Dialect of Lungming: Glossary, Texts and Translations. Michigan: Michigan Papers on South and Southeast Asia, Center for South and Southeast Asian Studies, University of Michigan. Hudak, Thomas John. (1996). William J. Gedney’s the Lue Language: Glossary, Texts and Translations. Michgan: Michigan Papers on South and Southeast Asia, Center for South and Southeast Asian Studies, University of Michigan. Information Services Department, Hong Kong SAR Government. (2005). Hong Kong 2005. Hong Kong: Information Services Department, Hong Kong SAR Government. Jenny, Mathias. (2010). Benefactive strategies in Thai. In Fernando Zúñiga & Seppo Kittilä (Eds.), Benefactives and malefactives: Typological perspectives and case studies (pp. 377–392). Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.92.16jen Jiang, Di, Ouyang, Jueya, & Tsou, Benjamin. (2007). 海南省三亞市邁話音系 [The phonological system of the Mai dialect spoken in Sanya of Hainan Province]. 方言 [Dialect], 1, 23–34. Jiang, Lansheng. (2000). 漢語使役與被動兼用探源 [An investigation of using the same syntactic structure for the Chinese causative and passive constructions]. In Lansheng Jiang (Ed.), 近代漢語探源 [An investigation of the origin of early modern Chinese] (pp. 221–236). Beijing: Commercial Press. Jiang, Shaoyu. (2003). ‘給’字句, ‘教’字句表被動的來源 [The origin of the passive construction with ‘gei’ and ‘jiao’]. In Fuxiang Wu & Bo Hong (Eds.), 語法化與語法研究(一) [Studies of grammaticalization and syntax (1)] (pp. 202–223). Beijing: Commercial Press. Jin, Mengyin (1983). 陵川方言志 [A description of the Lingchuan dialect]. Taiyuan: Shanxi gaoxiao lianhe chubanshe. Kataoka, Shin (2007). 19世紀的粵語處置句: “𢬿”字句 [The disposal construction of kaai in 19th century Cantonese]. In S. Cheung, S.H. Chang, H. K. Chan (Eds.), 第十屆國際粵方 言研討會論文集 [Proceeding of the 10th international conference on the Yue dialects] (pp. 191–200). Beijing: China Academy Social Sciences Press. Keenan, Edward (1985). Passive in the world’s languages. In Timothy Shopen (Ed.), Language typology and syntactic description: Clause structure (pp. 325–361). Cambridge: Cambridge University Press. Kemmer, Suzanne, & Verhagen, Arie (1994). The grammar of causatives and the conceptual structure of events. Cognitive Linguistics, 5(2), 115–156. https://doi.org/10.1515/cogl.1994.5.2.115 Kessakul, Ruetaivan, & Ohori, Toshio (1998). Grammaticalization of deverbal markers: Toward a cross-linguistic study in the semantic extension of motion verbs. In M. Alves, P. Sidwell & D. Gil (Eds.), SELAS VIII: Papers from the 8th annual meeting of the Southeast Asian linguistics society. (pp. 93–108). Canberra: Pacific Linguistics, Research School of Pacific and Asian Studies, Australian National University. Kittlä, Seppo (2005). Recipient-prominence vs. beneficiary-prominence. Linguistic Typology, 9(2), 269–297. https://doi.org/10.1515/lity.2005.9.2.269 Kittila, Seppo, & Zúñiga, Fernando (2010). Introduction. Benefaction and malefaction from a cross-linguistic perspective. In Fernando Zúñiga & Seppo Kittilä (Eds.), Benefactives and malefactives: Typological perspectives and case studies (pp. 1–28). Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.92.01kit
186 Cantonese give and Double-Object Construction Kwok, Bit-chee (2008). 邵武話動態助詞‘度’的來源 – 兼論邵武話和閩語的關係 [The origin of the verbal particle du 度 in Shaowu – On the relationship between the dialects of Shaowu and Min]. 中國語文 [Studies of the Chinese language], 2, 140–146. Labov, William (1999). Principles of linguistic change: Internal factors. Oxford: Blackwell. Lai, Huei-ling (2001). On Hakka BUN: A case of polygrammaticalization. Language and Linguistics, 2(2), 137–153. Lai, Mee-ling (2001). Hong Kong students’ attitudes towards Cantonese, Putonghua and English after the change of sovereignty. Journal of Multilingual and Multicultural Development, 22(2), 112–133. https://doi.org/10.1080/01434630108666428 LaPolla, Randy (2007). Overview of Sino-Tibetan morphosyntax. In Graham Thurgood & Randy J. LaPolla (Eds.), The Sino-Tibetan languages (pp. 22–42). London: Routledge. Larson, Richard (1988). On the double-object construction. Linguistic Inquiry, 19(3), 335–391. Lau, Sidney (1972) Intermediate Cantonese: Volumes I and II. Hong Kong: Government Training Division. Law, Yin Wah Shirley (1996). The dative construction in Hong Kong. (Unpublished MA thesis). Hong Kong Baptist University, Hong Kong. LeBlanc, Charles (1910). Cours de langue Chinoise parlee dialecte Cantonnais. Honoi-Haiphong: Imprimerie d’ Extreme-Orient. Lee, Cream, & Kataoka, Shin (2006). 馬禮遜對中文的認識 [Robert Morrison’s knowledge on the Chinese language]. 中國語文研究 [Studies in Chinese linguistics], 22(2), 21–36. Lee, Hui-chee (2011). Double object construction in Hainan Min. Language and Linguistics, 12(3), 501–527. Leech, Geoffrey, Rayson, Paul, & Wilson, Andrew (2001). Word frequencies in written and spoken English: Based on the British National Corpus. London: Longman. Leung, Chung-sum (2005). 香港粵語語助詞的研究 [A study of the utterance particles in Cantonese as spoken in Hong Kong]. Hong Kong: Language Information Sciences Research Center, City University of Hong Kong. Leung, Wai-mun, & Li, David Chor-Shing (2020). 兩文三語–香港語文教育政策研究 [Biliteracy and trilingualism: Language education policy research in Hong Kong]. Hong Kong: City University of Hong Kong Press. Levin, Beth (1993). English verb classes and alternations: A preliminary investigation. Chicago: University of Chicago. Li, Audrey Yen-hui (1990). Order and constituency in Mandarin Chinese. Dordrecht: Kluwer Academic. https://doi.org/10.1007/978-94-009-1898-6 Li, Beixiang, & He, Shaomei (1992). 佛山方言志 [A description of the dialect of Foshan]. Foshan: Foshanshi difangzhi bianzuan weiyuanhui bangongshi. Li, Charles, & Thompson, Sandra (1981). Mandarin Chinese: A functional reference grammar. California: University of California Press. Li, David Chor-Shing, & Leung, Wai-mun (2020). 香港“兩文三語”格局: 挑戰與對策建議 [Hong Kong SAR’s “biliteracy and trilingualism” language-in-education policy: Major challenges and proposed coping strategies]. 語言戰略研究 [Chinese journal of language policy and planning], 5(1), 46–58. Li, Fang-kuei (1956). 武鳴土語 [The language of Wuming]. Taipei: Academia Sinica. Li, Fang-kuei (1973). Languages and dialects of China. Journal of Chinese Linguistics, 1(1), 1–13. Li, Fang-kuei (1976). Sino-Tai. In Mantaro Hashimoto (Ed.), Genetic relationship, diffusion and typological similarities of East & Southeast Asian languages (pp. 256–289). Tokyo: Japan Society for the Promotion of Science. CIT191
References 187
Li, Fang-kuei (1977). A handbook of comparative Tai. Honolulu: University Press of Hawaii. Li, Fang-kuei (1988). 剝隘土語 [The language of Bo’ai] Volume 1. Taipei: Academia Sinica. Li, Fang-kuei (1993). 龍州土語 [The language of Longzhou]. Taipei: Academia Sinica. Li, Fang-kuei (2005). 水話研究 [Studies of the Shui language]. In 李方桂全集 [Collection of Essays by Li Fang-kuei], Volume 5. Beijing: Qinghua daxue chubanshe. Li, Jie (2008). 漢藏語系語言被動句研究 [A study of passive sentences in Sino-Tibetan languages]. Beijing: Minzu chubanshe. Li, Jinfang (2003). 巴哈布央語概況 [A description of the Baha Buyang language]. 民族語文 [Minority languages of China], 4, 67–80. Li, Jinfang (2004). 戶語概況 [A description of the Hu language]. 民族語文 [Minority languages of China], 5, 69–81. Li, Rulong (1996). 泉州方言‘給予’義的動詞 [The verb give in the Quanzhou dialect]. In Rulong Li (Ed.) 方言與音韻論集 [A collection of essays on dialects and phonology] (pp. 162–166). Hong Kong: T. T. Ng Chinese Language Research Center, Institute of Chinese Studies, The Chinese University of Hong Kong. Li, Rulong (1999). 粵西客家方言調查報告 [A research report on the Hakka dialects in western Guangdong]. Guangzhou: Ji’nan daxue chubanshe. Li, Rulong, & Chang, Song-Hing (1992). 客贛方言調查報告 [A research report on the Hakka and Gan dialects]. Xiamen: Xiamen daxue chubanshe. Li, Ruojian (1997). 中國大陸遷入香港人口研究 [A study of the Hong Kong population migrated from Mainland China]. 人口與經濟 [Population and economics], 2, 24–29. Li, Wei, Shi, Peixuan, Liu, Yanan, & Huang, Yanxuan (2015). 清代琉球官話課本語法研究 [Studies of the grammar in the Mandarin textbooks compiled in Ryukyus in the Qing dynasty]. Beijing: Peking University Press. Li, Weiqi (1998). 祁陽方言研究 [A study of the Qiyang dialect]. Hunan: Hunan jiaoyu chubanshe. Li, Xuping (2015). Ditransitive constructions in Gan Chinese: A case study of the Yichun dialect. In Brian Nolan, Gudrun Rawoens, Elke Diedrichsen (Eds.), Causation, permission, and transfer: Argument realisation in get, take, put, give and let verbs (pp. 177–193). Amsterdam: John Benjamins. https://doi.org/10.1075/slcs.167.06li Li, Yongming (1986). 衡陽方言 [The Hengyang dialect]. Changsha: Hunan renmin chubanshe. Li, Yongming (1991). 長沙方言 [The Changsha dialect]. Changsha: Hunan chubanshe. Li, Yunbing (2000). 拉基語研究 [A study of the Laji language]. Beijing: Zhongyang minzu daxue chubanshe. Liang, Min (1980). 毛難語簡志 [A description of the Maonan language]. Beijing: Minzu chubanshe. Liang, Min, & Zhang, Junru (1997). 臨高語研究 [A study of the Lingao language]. Shanghai: Shanghai yuandong chubanshe. Lien, Chinfa (2002). Grammatical function words 乞, 度, 共, 甲, 將 and 力 in Li Jing Ji 荔鏡記 and their development in Southern Min. In Dah-an Ho (Ed.), Dialect variations in Chinese (pp. 179–216). Taipei: Institute of Linguistics, Academia Sinica. Lin, Hansheng (2002). 閩東方言詞匯語法研究 [A study of the lexicon and syntax of the eastern Min dialects]. Kunming: Yunnan daxue chubanshe. Lin, Lunlun (2006). 粵西閩語雷州話研究 [A study of the Min dialect of Leizhou in western Guangdong]. Beijing: Zhonghua shuju.
188 Cantonese give and Double-Object Construction Lin, Yi (2010). 漢壯語接觸下的廣西粵方言給與義動詞的變異 [Development of the give verb in the Yue dialects of Guangxi: A contact linguistic approach]. Journal of Chinese Linguistics Monograph Series 24, 217–226. Lin, Yi, & Qin, Fengyu (2008). 廣西南寧白話研究 [A study of the dialect of Nanning in Guangxi]. Guilin: Guangxi Normal University Press. Liu, Danqing (2001). 漢語給予類雙及物結構的類型學考察 [A typological study of the givetype ditransitive constructions in Chinese]. 中國語文 [Studies of the Chinese language], 5, 387–398. Liu, Lunxin (1999). 客贛方言比較研究 [A comparative study of the Hakka and Gan dialects]. Beijing: Chinese Academy of Social Sciences Press. Liu, Ruoyun (1991). 惠州方言志 [A description of the Huizhou dialect]. Guangzhou: Guangdong keji chubanshe. Long, Yaohong (2003). 侗語研究 [A study of the Dong language]. Guiyang: Guizhou minzu chubanshe. Lord, Carol, Yap, Foong-Ha, & Iwasaki, Shiochi (2002). Grammaticalization of ‘give’: African and Asian perspectives. In Ilse Wischer & Gabriele Diewald (Eds.), New reflections on grammaticalizationi (pp. 217–235). Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.49.15lor Lu, Jiawen, Guo, Xiaowu & Hou, Jingyi (1998). 鄭州話音檔 [Sound files of the Zhengzhou dialect]. Shanghai: Shanghai jiaoyu chubanshe. Luo, Changpei (1930). 耶穌會士在音韻學上的貢獻 [The contributions of Jesuits on Chinese phonology]. Bulletin of Institute of History and Philology, 1(3), 267–338. Luo, Kangning (1987). 信宜方言志 [A description of the Xinyi dialect]. Guangzhou: Zhongshan daxue chubanshe. Ma, Beijia (2002). 近代漢語介詞 [Prepositions in pre-modern Chinese]. Beijing: Zhonghua shuju. Ma, Fengru (2000). 金鄉方言志 [A description of the Jinxiang dialect]. Ji’nan: Qilu shushe. Ma, Jing, & Wu, Yonghuan (2003). 臨沂方言志 [A description of the Linyi dialect]. Ji’nan: Qilu shushe. Ma, Qingzhu (1992). 現代漢語雙賓語構造 [The structure of the double-object construction in Modern Standard Chinese]. In Qingzhu Ma (Ed.), 漢語動詞和動詞性結構 [The structure of verbs and verbal constructions in Chinese] (pp. 66–74). Beijing: Beijing yuyan xueyuan chubanshe. Ma, Xueliang (2003). 漢藏語概論 [An introduction to Sino-Tibetan languages]. Beijing: Minzu chubanshe. Mai, Shizhi (1893a). 廣州俗話詩經解義 [An annotation of the Book of Odes in colloquial Guangzhou dialect). (n.p.). Mai, Shizhi (1893b). 廣州俗話書經解義 [An annotation of the Book of Documents in colloquial Guangzhou dialect). (n.p.). Malchukov, Andrej, Haspelmath, Martin, & Comrie, Bernard (2010). Ditransitive constructions: A typological overview. In Andrej Malchukov, Martin Haspelmath, Bernard Comrie (Eds.), Studies in ditransitive construction: A comparative handbook (pp. 1–64). Mouton: de Gruyter. https://doi.org/10.1515/9783110220377.1 Manomaivibool, Prapin (1975). A study of Sino-Thai lexical correspondences. (Unpublished doctoral dissertation). University of Washington, Seattle. Manomaivibool, Prapin (1976). Chinese and Thai: Are they related genetically? Computational Analyses of Asian & African Languages, 6, 1–31.
References 189
Manomaivibool, Prapin, & Tsou, Benjamin (1998). Early Sino-Thai linguistic links: Towards a semantic and quantitative study. In Benjamin Tsou, Tom Lai, Samuel Chan, William S.-Y. Wang (Eds.), Quantitative and computational studies on the Chinese language (pp. 253–282). Hong Kong: Language Information Sciences Research Centre, City University of Hong Kong. Margetts, Anna, & Austin, Peter K (2007). Three-participant events in the languages of the world: Towards a crosslinguistic typology. Linguistics, 45(3), 393–451. https://doi.org/10.1515/LING.2007.014 Matisoff, James (1973). The grammar of Lahu. Berkeley: University of California Press. Matisoff, James (1991). Areal and universal dimensions of grammatization in Lahu. In Elizabeth C. Traugott, Bernd Heine (Eds.), Approaches to grammaticalization Volume 2 (pp. 383–454). Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.19.2.19mat Matisoff, James (1992). Southeast Asian languages. In William Frawley & Bernard Comrie (Eds.), International encyclopedia of linguistics, Volume 4 (pp. 126–130). Oxford: Oxford University Press. Matisoff, James (2003). Lahu. In G. Thurgood & R. LaPolla (Eds.), The Sino-Tibetan languages (pp. 208–221). London: Routledge. Matthews, Stephen (2006). Cantonese grammar in areal perspective. In Alexandra Y. Aikhenvald & Robert M. W. Dixon (Eds.), Grammars in contact: A cross-linguistic typology (pp. 220– 236). Oxford: Oxford University Press. Matthews, Stephen, Xu, Huiling, & Yip, Virginia (2005). Passive and unaccusative in the Jieyang dialect of Chaozhou, Journal of East Asian Linguistics, 14(4), 267–298. https://doi.org/10.1007/s10831-005-3822-1 Matthews, Stephen, & Yip, Virginia (1994). Cantonese: A comprehensive grammar. 1st edition. London: Routledge. Matthews, Stephen, & Yip, Virginia (2001). Aspects of contemporary Cantonese grammar: The structure and stratification of relative clauses. In H. Chappell (Ed.), Sinitic grammar: Synchronic and diachronic perspectives (pp. 266–281). Oxford: Oxford University Press. Matthews, Stephen, & Yip, Virginia (2011). Cantonese: A comprehensive grammar. 2nd edition. London: Routledge. Mcdonogh, Gary, & Wong, Cindy (2005). Global Hong Kong. London: Routledge. McMahon, April (1996). Understanding language change. Cambridge: Cambridge University Press. Mei, Tsu-lin (2004). 介詞‘于’在甲骨文和漢藏語裏的起源 [The origin of the preposition yu in the oracle-bone inscriptions and Sino-Tibetan languages]. 中國語文 [Studies of the Chinese language], 4, 323–332. Mei, Tsu-lin (2005). 閩南話 Hoo ‘給予’的本字及其語法功能的來源 [The origin of hoo (give) and its syntactic function in Southern Min dialects]. In Dah-an Ho & Ovid J. L. Tzeng (Eds.), POLA FOREVER: Festschrift in honor of Professor William S-Y. Wang on his 70th birthday [永遠的POLA: 王士元先生七秩壽慶論文集] (pp. 163–173). Taipei: Institute of Linguistics, Academia Sinica. Meillet, Antoine (1958). L’evolution des formes grammaticales. In A. Meillet (Ed.), Linguistique historique et linguistique generale, volume 1 (pp. 130–148). Paris: Honore Champion. Melis, Ludo (1996). The dative in modern French. In William van Belle & Willy van Langendonck (Eds.), The dative: Descriptive studies, Volume 1 (pp. 39–72). Amsterdam: John Benjamins. https://doi.org/10.1075/cagral.2.05mel
190 Cantonese give and Double-Object Construction
Michaelis, Susanne, & Haspelmath, Martin (2003). Ditransitive constructions: Creole languages in a cross-linguistic perspective.” Creolica. Retrieved from http://www.creolica.net/ Ditransitive-constructions-Creoleon13/3/2021. McMahon, April (1996). Understanding language change. Cambridge: Cambridge University Press. Mu, Shihua (2003). 卡卓語研究 [A study of the Kazhuo language]. Beijing: Minzu chubanshe. Myers-Scotton, Carol (2002). Contact linguistics: Bilingual encounters and grammatical outcomes. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198299530.001.0001 Nakanishi, Hiroki. (2003). 畬語海豐方言基本詞匯集 [A collection of basic vocabulary of the She language in Haifeng]. Kyoto: Institute of Research in Humanities, Kyoto University. Newman, John. (1993). The semantics of giving in Mandarin. In Richard Geiger & Brygida Rudzka-Ostyn (Eds.), Conceptualizations and mental processing in language (pp. 433–486). Berlin: Mouton de Gruyter. Newman, John. (1996). Give: A cognitive linguistic study. Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110823714 Newman, John. (1998). The linguistics of giving. Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.36 Newman, John. (1999). Figurative giving. In Leon de Stadler & Christoph Eyrich (Eds.), Issues in cognitive linguistics: 1993 proceedings of the international cognitive linguistics conference (pp. 113–140). Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110811933.113 Newman, John. (2005). Three-place predicates: A cognitive-linguistic perspective. Language Sciences, 27(2), 145–163. https://doi.org/10.1016/j.langsci.2003.12.003 Nilsen, Don L. F (1973). The instrumental case in English. The Hague: Mouton. https://doi.org/10.1515/9783110815498 Norman, Jerry. (1969). The Kienyang dialect of Fukien (Unpublished doctoral dissertation). University of California, Berkeley, Berkeley. Norman, Jerry. (1982). Four notes on Chinese-Altaic linguistic contacts. The Tsing Hua Journal of Chinese Studies, 14(3), 243–247. Norman, Jerry. (1988). Chinese. Cambridge: Cambridge University Press. Norman, Jerry. (1991). The Min dialects in historical perspective. In William S.-Y. Wang (Ed.), Languages and dialects of China, Monograph Series No. 3 of Journal of Chinese Linguistics (pp. 323–358). Berkeley: Project on Linguistic Analysis. Norman, Jerry & Mei, Tsu-lin. (1976). Austroasiatics in ancient South China: Some lexical evidence. Monumenta Serica, 32, 274–301. https://doi.org/10.1080/02549948.1976.11731121 O’Melia, Thomas A (1941). First year Cantonese. Hong Kong: Maryknoll House. Oakley, R. H (1953) Rules for speaking Cantonese. Kuala Lumpur: Charles Grenier & Son Limited. Oehrle, Richard. (1976). The grammatical status of the English dative alternation (Unpublished doctoral dissertation). Massachusetts Institute of Technology, Cambridge. Ouyang, Jueya. (1998). 村語 [The Cun language]. Shanghai: Shanghai yuandong chubanshe. Peng, Chun-Yi. (2020). The placement of co-verb gěi in spoken Mandarin varieties: A study on regional influences. Chinese Language and Discourse, 11(2), 335–354. https://doi.org/10.1075/cld.18006.pen Peyraube, Alain. (1981). The dative construction in Cantonese. Computational Analyses of Asian & African Languages, 16, 29–65.
References 191 Peyraube, Alain. (1986). 雙賓語結構: 從漢代至唐代的歷史發展 [The double-object construction: Its historical development from the Han dynasty to the Tang dynasty].中國語文 [Studies of the Chinese language], 3, 204–216. Peyraube, Alain. (1988). Syntaxe diachronique du Chinois: Evolution des constructions datives: Du XIV siecle Av. J.-C. au XVIII siecle. Paris: College de France, Institut Des Hautes Etudes Chinoises. Phua, Chiew Pheng. (2007). 粵方言給予義雙賓語結構的來源 [The origin of the give-type double-object construction in the Yue dialects]. In Samuel Cheung, Song-Hing Chang & Hong-ken Chan (Eds.), 第十屆國際粵方言研究會論文集 [Proceedings of the tenth international conference on the Yue dialects] (pp. 214–229). Beijing: Chinese Academy of Social Sciences Press. Phua, Chiew Pheng. (2009). 從語義地圖看《廣東省土話字匯》中的“俾” [Studies of BEI in vocabulary of the Canton dialect with the semantic map theory]. In A. Chin, B.C. Kwok, P. Lee, & B. Tsou (Eds.), 第十三屆國際粵方言研究會論文集 [Proceedings of the thirteenth international conference on the Yue dialects] (pp. 367–382). Hong Kong: Language Information Sciences Research Centre, City University of Hong Kong. Phua, Chiew Pheng. (2013). 語義地圖看給予動詞的語法化: 兼論語義地圖和多項語法化的 關係 [Studies of the grammaticalization of give with the semantic map theory and on the relationship between semantic map and multiple grammaticalization]. In F. Wu, & X. Xing (Eds.) 語法化與語法研究(六)[Grammaticalization and syntactic studies: 6] (pp. 262– 307). Beijing: Commercial Press. Pierson, Herbert D (1998). Societal accommodation to English and Putonghua in Cantonese-speaking Hong Kong. In Martha C. Pennington (Ed.), Language in Hong Kong at century’s end (pp. 91–112). Hong Kong: Hong Kong University Press. Pineda, Anna, & Mateu, Jaume. (2020). Dative constructions across languages: An introduction. In Anna Pineda & Jaume Mateu (Eds.), Dative constructions in Romance and beyond (pp. iii–xiii). Berlin: Language Science Press. Pinker, Steven. (1989). Learnability and cognition: The acquisition of argument structure. Cambridge: MIT Press. Pulleyblank, Edwin. (1986). The locative particles yu 于, yu 於 and hu 乎. Journal of the American Oriental Society, 106(1), 1–12. https://doi.org/10.2307/602358 Qian, Nairong. (1992). 當代吳語研究 [A study of modern Wu dialects]. Shanghai: Shanghai jiaoyu chubanshe. Qian, Nairong. (1997). 吳語中的‘來’和‘來’字結構 [The syntactic constructions of lai in the Wu dialects]. 上海大學學報 [Journal of Shanghai University], 4(3), 102–108. Qian, Nairong. (2003). 上海語言發展史 [The historical development of the Shanghai dialect] Shanghai: Shanghai renmin chubanshe. Qian, Zengyi. (2005). 萊州方言志 [A description of the Laizhou dialect]. Ji’nan: Qilu shushe. Qin, Yuanxiong. (2005). 平話、粵語與壯語‘給’義的詞 [‘give’ in Pinghua, Yue dialects and Zhuang languages]. 民族語文 [Minority languages of China], 5, 57–62. Qiu, Xigui. (2010). 談談殷墟甲骨卜辭中的“于” [On yu in the oracle-bone inscriptions]. In W. South Coblin & Anne O. Yue (Eds.), Studies in Honor of Jerry Norman 羅杰瑞先生七秩 晉三壽慶論文集, pp. 421–451. Hong Kong: The Ng Tor-tai Chinese Language Research Centre, Institute of Chinese Studies, The Chinese University of Hong Kong. Qiu, Xueqiang. (2005). 軍話研究 [A study of the Jun dialects]. Beijing: Zhongguo shehui kexue chubanshe. CIT286
192 Cantonese give and Double-Object Construction
Ramsey, Robert. (1987). The languages of China. Princeton: Princeton University Press. Rice, Sally, & Kabata, Kaori. (2007). Crosslinguistic grammaticalization patterns of the ALLATIVE. Linguistic Typology, 11(3), 451–514. https://doi.org/10.1515/LINGTY.2007.031 Schuessler, Axel. (2003). Multiple origins of the Old Chinese lexicon. Journal of Chinese Linguistics, 31(1), 1–71. Shen, Ming. (2002). 太原話的‘給’字句 [The give construction in the Taiyuan dialect]. 方言 [Dialect], 2, 108–116. Shen, Ruoyun. (1999). 宜章土話研究 [A study of the Yizhang patois]. Hunan: Hunan jiaoyu chubanshe. Shi, Rujie. (2006a).《綴白裘》裏的‘拉’,‘來’以及相關結構 [Some syntactic structures related to la and lai in Zhui Bai Qiu]. In Rujie Shi (Ed.), 明清吳語和現代方言研究 [Studies of the Wu dialects of the Ming and Qing dynasties and modern dialects] (pp. 217–233). Shanghai: Shanghai cishu chubanshe. Shi, Rujie. (2006b).《綴白裘》的語言 [The language of Zhui Bai Qiu]. In Rujie Shi (Ed.), 明清 吳語和現代方言研究 [Studies of the Wu dialects of the Ming and Qing dynasties and modern dialects] (pp. 205–216). Shanghai: Shanghai cishu chubanshe. Shi, Rujie. (2006c). 《山歌》 “ 的語音和語法問題 [The issue of the phonology and the syntax in Shan Ge].” In Rujie Shi (Ed.), 明清吳語和現代方言研究 [Studies of the Wu dialects of the Ming and Qing dynasties and modern dialects] (pp. 164–181). Shanghai: Shanghai cishu chubanshe. Shi, Yuzhi. (2005). 漢語研究的類型學視野 [Linguistic typology and Chinese language]. Nanchang: Jiangxi jiaoyu chubanshe. Simmons, Richard VanNess. (1992). The Hangzhou dialect (Unpublished doctoral dissertation). University of Washington, Seattle. Sneddon, James Neil. (1996). Indonesian: A comprehensive grammar. London: Routledge. Snow, Don. (2004). Cantonese as written language: The growth of a written Chinese vernacular. Hong Kong: Hong Kong University Press. Standaert, Nicolas. (2001). Handbook of Christianity in China. Volume 1: 635–1800. Leiden: Brill. https://doi.org/10.1163/9789004391857 Stedman, Thomas Lathrop, & Lee, Kuei-pan. (1888). A Chinese and English phrase book in the Canton dialect. New York: William R. Jenkins. Stolz, Thomas. (2001). Comitatives vs. instrumentals vs. agents. In Walter Bisang (Ed.), Aspects of typology and universals (pp. 153–174). Berlin: Akademie Verlag. https://doi.org/10.1524/9783050078892.153 Sun, Hongkai, & Jiang, Di (2000). 漢藏語系研究歷史沿革 [An overview on the history of the study on the Sino-Tibetan language family]. In Pang-hsin Ting & Hongkai Sun (Eds.), 漢藏 語同源詞研究(一): 漢藏語研究的歷史回顧 [Cognate words in Sino-Tibetan languages 1: An overview of the history of the study on Sino-Tibetan languages] (pp. 1–116). Nanning: Guangxi minzu chubanshe. Sybesma, Rint. (2008). Zhuang: A Tai language with some Sinitic characteristics: Post-verbal ‘can’ in Zhuang, Cantonese, Vietnamese and Lao. In Pieter Muysken (Ed.), From linguistic areas to areal linguistics (pp. 221–274). Amsterdam: John Benjamins. https://doi.org/10.1075/slcs.90.06syb Szeto, Pui Yiu. (2019). Typological variation across Sinitic languages: contact and convergence (Unpublished doctoral dissertation). University of Hong Kong, Hong Kong.
References 193
Takashima, Ken-ichi, & Serruys, Paul L-M (2010). Studies of fascicle three of inscriptions from the Yin ruins. Volume 1: General notes, text and translations. Taipei: Institute of History and Philology, Academia Sinica. Takashima, Ken-ichi, & Yue, Anne (2000). Evidence of possible dialect mixture in oracle-bone inscriptions. In Anne Yue, Pang-Hsin Ting (Eds.), In memory of Li Fang-Kuei: Essays on linguistic change and the Chinese dialects (pp. 1–52). Seattle & Taipei: University of Washington & Institute of Linguistics, Academia Sinica. Tang, Sze-Wing. (1998). On the ‘inverted’ double object construction. In Stephen Matthews (Ed.), Studies in Cantonese linguistics (pp. 35–52).Hong Kong: The Linguistic Society of Hong Kong. Tang, Sze-Wing. (2003). 漢語方言語法的參數理論 [A parametric theory of Chinese dialectal grammar]. Beijing: Beijing daxue chubanshe. Tang, Ting-Chi. (1981). Double-object constructions in Chinese. In Ting-Chi Tang (Ed.), 國 語語法研究論集 [Collections of essays on the studies of Chinese grammar] (pp. 65–95). Taipei: Xuesheng shuju. Thepkanjana, Kingkarn, & Uehara, Satoshi (2008). The verb of giving in Thai and Mandarin Chinese as a case of polysemy: A comparative study. Language Sciences, 30(6), 621–651. https://doi.org/10.1016/j.langsci.2007.04.001 Thomason, Sarah Grey (2000). Linguistic areas and language history. In D. G. Gilbers, J. Nerbonne & J. Schaeken (Eds.), Languages in contact (pp. 311–327). Amsterdam: Rodpi. Thomason, Sarah Grey (2001). Language contact: An introduction. Edinburgh: Edinburgh University Press. Thompson, Laurence. (1965). A Vietnamese grammar. Seattle: University of Washington Press. Thompson, Sandra. (1995). The iconicity of ‘dative shift’ in English: Considerations from information flow in discourse. In Marge E. Landsberg (Ed.), Syntactic iconicity and linguistic freezes: The human dimension (pp. 155–175). Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110882926.155 Ting, Jen, & Chang, Miller (2004). The category of gei in Mandarin Chinese and grammaticalization. Taiwan Journal of Linguistics, 2(2), 45–74. Ting, Pang-hsin. (2005). 說‘五’道‘六’ [On numerals FIVE and SIX]. 民族語文 [Minority languages of China], 3, 1–6. Trask, Robert Lawrence (1993). A dictionary of grammatical terms in linguistics. London: Routledge. Traugott, Elizabeth Closs, & König, Ekkehard (1991). The semantics-pragmatics of grammaticalization revisited. In E. C. Traugott & B. Heine (Eds.), Approaches to Grammaticalization, Volume 1 (pp. 189–218). Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.19.1.10clo Tsou, Benjamin. (1981). 有關近代漢語順裁、逆裁結構演變的探究 [An investigation of forward deletion and backward deletion in pre-modern Chinese]. Paper presented at the Conference of the 1st Chinese Linguistics Association, Chengdu. Tsou, Benjamin. (1994a). A note on Cantonese Tone Sandhi (CTS) as a diffusional phenomenon. In Matthew Chen, & Ovid Tzeng (Eds.), In honor of William S-Y. Wang: Interdisciplinary studies on language and language change (pp. 539–548). Taipei: Pyramid Press. Tsou, Benjamin. (1994b). Language planning issues raised by English in Hong Kong: Pre- and post-1997. In Thiru Kandiah & John Kwan-Terry (Eds.), English and language planning: A Southeast Asian contribution (pp. 197–217). Singapore: Times Academic Press.
194 Cantonese give and Double-Object Construction Tsou, Benjamin. (1997).「三言」 、 「兩語」說香港 [Trilingualism and biliteracy in Hong Kong]. Journal of Chinese Linguistics, 25(2), 290–307. Tsou, Benjamin. (1999). Some remarks on entomological terms in Chinese and the Austroasiatic link revisited. In Alain Peyraube & Chaofen Sun (Eds.), In honor of Mei Tsu-Lin: Studies on Chinese historical syntax and morphology (pp. 203–221).Paris: École des Hautes Études en Sciences Sociales – Cahiers de Linguistique – Asie Orientale. Tsou, Benjamin. (2002). Some considerations for additive bilingualism: A tale of two cities (Singapore and Hong Kong). In Daniel W. C. So & Gary M. Jones (Eds.), Education and society in plurilingual contexts (pp.163–198). Brussels: Brussels University Press. Tsou, Benjamin, & Kwong, Olivia Oi Yee. (2015). LIVAC as a monitoring corpus for tracking trends beyond linguistics. In Benjamin Tsou & Olivia Kwong (Eds.), Linguistic corpus and corpus linguistics in the Chinese context (Journal of Chinese Linguistics Monograph Series No.25) (pp. 447–471). Hong Kong: The Chinese University Press. Tsou, Benjamin, Kwong, Olivia Oi Yee, Lu, Bin, & Tsoi, Wingfu (2011). 漢語共時語料庫與追蹤 語料庫: 語料庫語言學的新方向 [Chinese synchronous corpus and monitoring corpus: A new direction of corpus linguistics]. 中文信息學報: 慶祝中國中文信息學會成立三十周 年紀念論文集 [Journal of Chinese Information Processing], 25(6), 38–45. Tsou, Benjamin, Lin, Hing-Lung, Liu, Godfrey, Chan, Terence, Hu, Jerome, Chew, Ching-Hai. & Tse, John (1997). A synchronous Chinese language corpus from different speech communities: Construction and application. International Journal of Computational Linguistics and Chinese Language Processing, 2(1), 91–104. Tsou, Benjamin, & You, Rujie (2001). 漢語與華人社會 [The Chinese language and Chinese society]. Shanghai and Hong Kong: Fudan University Press and City University of Hong Kong Press. Tung, Tung-ho (1959). 四個閩南方言 [On four Southern Min dialects]. Bulletin of the Institute of History and Philology, 59(2), 729–1042. Wai, Puiman (2004). 從一九二八年說起:香港廣播七十五年專輯 [Since 1928: A monograph on the 75th anniversary of broadcasting in Hong Kong]. Hong Kong: Radio Television of Hong Kong. Wang, Chengyu (2004). 彝語仆拉話概況 [A description of the Pulai dialect of the Yi language]. 民族語文 [Minority languages of China], 6, 63–79. Wang, Huayun (2004). 鄂東方言研究 [A study of the dialects of the east of Hubei province]. Chengdu: Bashu shushe. Wang, Li (1996). 漢語史稿 [An outline of the history of the Chinese language]. Beijing: zhonghua shuju. Wang, Lixun, & Kirkpatrick, Andy (2019). Trilingual education in Hong Kong primary schools. Cham: Springer. https://doi.org/10.1007/978-3-030-11081-9 Wang, Pen-Ying, & Lien, Chinfa (2001). A-not-A question in Taiwan Southern Min. Journal of Chinese Linguistics, 29(2), 351–376. Wang, William S.-Y (1969). Competing changes as a cause of residue. Language, 45(1), 9–25. https://doi.org/10.2307/411748 Washio, Ryuichi (1993). When causatives mean passive: A cross-linguistic perspective. Journal of East Asian Linguistics, 2(1), 45–90. https://doi.org/10.1007/BF01440583 Wei, Gangqiang, & Chen, Changyi (1998). 南昌話音檔 [Sound File of the Nanchang dialect]. Shanghai: Shanghai jiaoyu chubanshe. Wells, Herbert Richmond (1931). Commercial conversations in Cantonese and English 英粵商業 雜話. Hong Kong: Kae Shean Printing Co.
References 195
Wells, Herbert Richmond (1941). Cantonese for everyone. Hong Kong: International Commercial Printing Press. Whitaker, Katherine. Po Kan. (1954). Cantonese sentence series. London: Arthur Probsthain. Whitaker, Katherine. Po Kan. (1959). Structure drill in Cantonese. London: Percy Lund, Humphries. Winford, Donald (2003). An introduction to contact linguistics. Oxford: Blackwell. Wisner, Otis Frank (1906a). Cantonese romanized 1. n.p. Wisner, Otis Frank (1906b). Cantonese romanized 2. n.p. Wisner, Otis Frank (1927). Beginning Cantonese 教話指南. Guangzhou: Guangzhou. Wong, Kwok-shing (2004). The acquisition of polysemous forms: The case of bei2 (‘give’) in Cantonese. In Olga Fischer, Muriel Norde & Harry Perridon (Eds.), Up and down the cline –the nature of grammaticalization (pp.325–343). Amsterdam: John Benjamins. https://doi.org/10.1075/tsl.59.17won Wu, Fuxiang (2005). 粵語能性述補結構“Neg-V得OC/CO”的來源. 方言 [Dialects], 4, 306–318. Wu, Hsiao-ling Iris (2021). Reverse double objects in Hakka ditransitive. Studia Linguistica. https://doi.org/10.1111/stul.12171 Wu, Rui-wen (2015). 從歷史語法的觀點論 「乞」 在閩語中的演變 [On the diachronic change of kʰi3 in Min dialect from the viewpoint of historical grammar]. Bulletin of Chinese Linguistics, 8(2), 245–266. https://doi.org/10.1163/2405478X-00802006 Wu, Qizhu (1998). 常寧方言研究 [A study of the Changning dialect]. Changsha: Hunan jiaoyu chubanshe. Wu, Yunji (2005). A synchronic and diachronic study of the grammar of the Chinese Xiang dialects. Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110927481 Wurm, Stephen Adolphe, Li, Rong, Baumann, Theo, & Lee, Mei W (1988). Language atlas of China: parts I and II [中國語言地圖集]. Hong Kong: Longman Group (Far East). Xia, Jianqin (1998). 瀏陽方言研究 [A study of the Liuyang dialect]. Hunan: Hunan jiaoyu chubanshe. Xiang, Mengbing (1997). 連城客家話語法研究 [A study of the grammar of the Hakka dialect of Liancheng]. Beijing: Yuwen chubanshe. Xu, Baohua, & Tang, Zhenzhu (1988). 上海市區方言志 [A description of the dialect of the Shanghai city]. Shanghai: Jiaoyu chubanshe. Xu, Dan (1994). The status of marker gei in Mandarin Chinese. Journal of Chinese Linguistics, 22(2), 363–394. Xu, Dan (2005). 某些具有[±給與]意義動詞的語法化 [Grammaticalization of some give-type verbs]. In Fuxiang Wu (Ed.), 漢語語法化研究 [Studies on grammaticalization in Chinese] (pp.245–260). Beijing: Commercial Press. Xu, Feng (2004). 漢語配價分析與實踐 – 現代漢語三價動詞探索 [Analysis and application of valency in Chinese: A study of tri-valent verbs]. Shanghai: Xuelin chubanshe. Xu, Guanghua (2000). 16至18世紀傳教士與漢語研究 [Missionaries and Chinese language studies between the 16th and 18th centuries]. 國際漢學 [International Sinology], 2, 456–490. Xu, Liejiong, & Peyraube, Alain (1997). On the double-object construction and the oblique construction in Cantonese. Studies in Language, 21(1), 105–127. https://doi.org/10.1075/sl.21.1.05lie Xu, Shixuan (2001). The Bisu language. Munchen: Lincom Europa. Xu, Yan (2015). 仡佬語與壯語給予類雙及物結構之比較 [The comparative study of giving-type ditransitive constructions in Geloa and Zhuang]. Journal of Sino-Tibetan Linguistics, 8, 64–76.
196 Cantonese give and Double-Object Construction
Yan, Junrong (2005). ‘給’與‘V給’不對稱的實証研究 [‘A study on the asymmetry of gei and V-gei]. 語言研究 [Studies in language and linguistics], 25(1), 26–33. Yan, Sen (1993). 黎川方言研究 [A study of the Lichuan dialect]. Beijing: Shehui kexue wenxian chubanshe. Yan, Yiming (1994). 吳語槪說 [A description of the Wu dialects]. Shanghai: Huadong shifan daxue chubanshe. Yang, Guangyuan (2006). 傣語‘給’的用法 [Usage of ‘give’ in the Tai language]. 民族語文 [Minority languages of China], 6, 17–23. Yang, Hsiu-fang (1995). 臺灣閩南語語法稿 [The grammar of Taiwan Southern Min dialects]. Taipei: Da’an chubanshe. Yang, Jingyu (2006). 清末粵方言語法及其發展研究 [The Yue dialect in the late Qing period and its development]. Guangzhou: Guangzhou renmin chubanshe. Yang, Shifeng (1992). 台灣桃園客家方言 [The Hakka dialects in Taoyuan of Taiwan]. Taipei: Academia Sinica. Yang, Tongyin (2000). 莫話研究 [A study of the Mo language]. Beijing: Zhongyang minzu daxue chubanshe. Yang, Zaibiao (2004). 苗語東部方言土語比較 [A comparison of the eastern dialects of Miao]. Beijing: Minzu chubanshe. Yap, Foong-Ha, & Chen, Weirong (2018). Pathways to adversity and speaker affectedness: On the emergence of unaccusative ‘give’ constructions in Chinese. Linguistics, 56(1), 19–68. https://doi.org/10.1515/ling-2017-0038 Yap, Foong-Ha, & Iwasaki, Shoichi (2003). From causative to passive: A passage in some East and Southeast Asian languages. In Eugene Casad & Gary Palmer (Eds.), Cognitive linguistics and non-Indo-European languages (pp.419–446). Berlin: Mouton de Gruyter. https://doi.org/10.1515/9783110197150.10.419 Yates, Warren G., & Tryon, Absorn (1970). Thai: Basic course Volume 2. Washington: Foreign Service Institute, Department of State. Ye, Nong (2004). 明清時期來華歐洲傳教士中國語言學習活動勾沉 [Rediscovering the study of the Chinese language by European missionaries in the Ming and Qing dynasties]. Revista de Cultura, 53, 93–104. Yeh, Jui-chuan (2020). On the emergence of the inverted double object construction in Hakka. Journal of Chinese Linguistics, 48(2), 342–378. https://doi.org/10.1353/jcl.2020.0008 Yiu, Carine (2010). 早期粵語中的「畀」字句 [BEI-sentence in early Cantonese]. Journal of Chinese Linguistics Monograph Series 24, 162–185. You, Rujie (2002). 西洋傳教士漢語方言學著作書目考述 [A study of the publications on Chinese dialects by western missionaries]. Harbin: Heilongjiang jiaoyu chubanshe. Yuan, Jiahua (2001). 漢語方言概要 [An outline of Chinese dialects]. Beijing: Yuwen chubanshe. Yuan, Yulin (1998). 漢語動詞的配價研究 [A study of the valency of Chinese verbs]. Nanchang: Jiangxi jiaoyu chubanshe. Yuan, Zhongshu (1994). 黎語語法綱要 [A concise grammar of the Li language]. Beijing: Zhongyang minzu daxue chubanshe. Yue, Anne Oi-kan (1988). 漢語方言語法的比較研究 [Comparative studies on Chinese dialectal grammar]. Bulletin of the Institute of History and Philology, 59(1), 23–41. Yue, Anne Oi-kan (1992). 廣東開平方言的中性問句 [Neutral questions in the Kaiping dialect of Guangdong province]. 中國語文 [Studies of the Chinese language], 4, 279–286. Yue, Anne Oi-kan (1995). 粵語研究的當前課題 [Current issues in the study of the Yue dialects]. Journal of Chinese Linguistics, 23(1), 1–41.
References 197
Yue, Anne Oi-kan (1997). Syntactic change in progress – Part I: The comparative construction in Hong Kong Cantonese. In Anne Yue & M. Endo (Eds.), 橋本萬太郎紀念中國語學論集 [In Memory of Mantaro J. Hashimoto] (pp. 329–375). Tokyo: Uchiyama. Yue, Anne Oi-kan (2000). 粵語方言的歷史研究 – 讀《麥仕治廣州俗話《詩經》解義》[A historical study of the Yue dialect: The Book of Documents translated into colloquial Cantonese by Mo Shizhi]. 中國語文 [Studies of the Chinese language], 6, 497–507. Yue, Anne Oi-kan (2001). The historic role of the late Professor Y. R. Chao’s 1929 field materials. Language and Linguistics, 2(1), 197–228. Yue, Anne Oi-kan (2003). Chinese dialects: Grammar. In G. Thurgood & R. LaPolla (Eds.), The Sino-Tibetan languages (pp. 84–125). London: Routledge. Yue, Anne Oi-kan (2004). Materials for the diachronic study of the Yue dialects. In Feng Shi & Zhongwei Shen (Eds.), The joy of research: A festschrift in honor of Professor William S-Y. Wang on his seventieth birthday 樂在其中 – 王士元教授七十華誕慶祝文集 (pp. 246– 271). Nankai: Nankai University Press. Yue, Anne Oi-kan (2006a). Syntactic typology in Chinese (Part 1): The neutral question forms – V-not-V. Bulletin of Chinese Linguistics, 1(1), 201–245. https://doi.org/10.1163/2405478X-90000012 Yue, Anne Oi-kan (2006b). 粵音構擬之二:聲母 [Proto-Yue phonology 2: The initials]. In Dah-an Ho, H. Samuel Cheung, Wuyun Pan & Fuxiang Wu (Eds.), Linguistic studies in Chinese and neighboring languages: Festschrift in honor of Professor Pang-hsin Ting on his 70th birthday 山高水長:丁邦新先生七秩壽慶論文集 (pp. 75–170). Taipei: Academia Sinica. Yue, Jing (2004). 黃金鎮仫佬語概況 [A description of the Mulao language in Huangjin]. 民族 語文 [Minority languages of China], 4, 69–80. Yue-Hashimoto, Anne (1971). Mandarin syntactic structure. Unicorn, 8, 1–149. Princeton: Chinese Linguistics Project and Seminar, Princeton University. Yue-Hashimoto, Anne (1972). Studies in Yue Dialects 1: Phonology of Cantonese. Cambridge: Cambridge University Press. Yue-Hashimoto, Anne (1976). Southern Chinese dialects – The Tai connection. Computational Analysis of Asian and African Languages, 6, 1–9. Yue-Hashimoto, Anne (1979). The Teng-xian dialect of Chinese: Its phonology, lexicon and texts with grammatical notes. Monograph Series, No. 3 of Computational Analyses of Asian & African Languages. Tokyo: National Inter-University Research Institute of Asian & African Languages & Culture. Yue-Hashimoto, Anne (1985). The Suixi dialect of Leizhou: A study of its phonological, lexical and syntactic structure. Hong Kong: The Ng Tor-tai Chinese Language Research Centre, Institute of Chinese Studies, The Chinese University of Hong Kong. Yue-Hashimoto, Anne (1991a). Stratification in comparative dialectal grammar: A case in Southern Min. Journal of Chinese Linguistics, 19(2), 172–201. Yue-Hashimoto, Anne (1991b). The Yue dialects. In William S.-Y. Wang (Ed.), Languages and dialects of China. Journal of Chinese Linguistics Monograph No. 3 (pp. 294–324). Berkeley: Journal of Chinese Linguistics. Yue-Hashimoto, Anne (1993a). Comparative Chinese dialectal grammar: Handbook for investigators. Paris: École des Hautes Études en Sciences Sociales, Centre de Recherches Linguistiques sur l’Asie Orientale. Yue-Hashimoto, Anne (1993b). The lexicon in syntactic change: Lexical diffusion in Chinese syntax. Journal of Chinese Linguistics, 21(2), 213–254.
198 Cantonese give and Double-Object Construction
Yue-Hashimoto, Anne (1994). Development of the Xiamen neutral question forms. In Matthew Chen & Ovid Tzeng (Eds.), In honor of William S.-Y. Wang: Interdisciplinary studies on language and language change (pp. 593–610). Taipei: Pyramid Press. Yue-Hashimoto, Anne (1999). The Min translation of the Doctrina Christiana. In Pang-hsin Ting (Ed.), Contemporary studies on the Min dialects. Journal of Chinese Linguistics monograph no. 14 (pp. 42–76). Berkeley: Journal of Chinese Linguistics. Yue-Hashimoto, Anne (2005). The Dancun dialect of Taishan 台山淡村方言研究. Hong Kong: Language Information Sciences Research Center, City University of Hong Kong. Zee, Eric (2001). Chinese (Hong Kong Cantonese). In Handbook of the International Phonetic Association: A guide to the use of the International Phonetic Alphabet (pp. 58–60). Cambridge: Cambridge University Press. Zhan, Bohui (1981). 現代漢語方言 [Modern Chinese dialects]. Wuhan: Hubei renmin chubanshe. Zhan, Bohui (2004). 廣東粵方言槪要 [An outline of Yue dialects in Guangdong]. Guangzhou: Ji’nan chubanshe. Zhan, Bohui, & Cheung, Yat-shing (1987). 珠江三角洲方言調查報告 – 珠江三角洲方言字音 對照 [A survey of dialects in the Pearl River Delta. Volume 1: Comparative morpheme-syllabary]. Hong Kong: Xinshiji chubanshe. Zhan, Bohui, & Cheung, Yat-shing (1988). 珠江三角洲方言調查報告 – 珠江三角洲方言詞 匯對照 [A survey of dialects in the Pearl River Delta. Volume 2: Comparative Lexicon]. Guangzhou: People’s Publishing House of Guangdong. Zhang, Ansheng (2000). 同心方言研究 [A study of the Tongxin dialect]. Yinchuan: Ningxia renmin chubanshe. Zhang, Hongming (2005). 漢語‘被’的語法化 [Grammaticalization of ‘bei’ in Chinese]. In Jiaxuan Shen, Fuxiang Wu & Beijia Ma (Eds.), 語法化與語法研究(二) [Studies on grammaticalization and syntax (2)] (pp. 190–218). Beijing: Commercial Press. Zhang, Huiying (2002). 漢藏系語言和漢語方言比較研究 [A comparative study between Sino-Tibetan languages and Chinese dialects]. Beijing: Minzu chubanshe. Zhang, Junru (1980). 水語簡志 [A description of the Shui language]. Beijing: Minzu chubanshe. Zhang, Lili (2006). 漢語使役句表被動的語義發展 [The development of the semantics of causative and passive constructions in Chinese]. Language and Linguistics, 7(1), 139–174. Zhang, Meilan (2018). 施受關係之表達與南北類型特徵制約 [Expression of agent-recipient relationship and restriction of Southern and Northern type features – Talking from the Shanghai dialect and Cantonese versions of An Enchiridion of Mandarin]. 學術交流 [Academic exchange], 287(2), 139–149. Zhang, Meilan (2020). 文本比較視域下的清末民初粵語句式考察–以《粵音指南》 《訂正粵 音指南》為中心 [A text comparative study on the syntax of Cantonese in the late Qing dynasty and early Republic period: Focused on The Guide to Cantonese and The Revised Guide to Cantonese]. 漢語史學報 [Journal of the history of the Chinese language], 22, 29–41. Zhang, Min (2010). 漢語方言雙及物結構南北差異的成因: 類型學研究引發的新問題 [Revisiting the alignment typology of ditransitive constructions in Chinese dialects]. Bulletin of Chinese Linguistics, 4(2), 87–259. https://doi.org/10.1163/2405478X-90000063 Zhang, Min (2015). “語義地圖模型”: 原理, 操作及在漢語多功能語法形式研究中的運用 [The semantic map model: Its mechanism, operation and application for studying Chinese syntactic structures with multiple functions]. In X. Li, M. Zhang, & R. Guo (Eds.), 漢語多 功能語法形式的語義地圖研究 [Semantic map analysis of Chinese syntactic structures with multiple functions] (pp. 1–55). Beijing: Commercial Press.
References 199 Zhang, Xiaoqin (1999). 寧遠平話研究 [A study of the pinghua of Ningyuan]. Hunan: Hunan jiaoyu chubanshe. Zhang, Yizhou, Zhang, Qingyuan, & Deng, Yingshu (2001). 成都方言語法研究 [A study of the grammar of the Chengdu dialect]. Chengdu: Bashu shushe. Zhang, Zhenjiang (2009). 早期香港的社會和語言(1841–1884) [Language and society in early Hong Kong (1841–1884)]. Guangzhou: Zhongshan daxue chubanshe. Zhao, Minlan (2004). 柘山勉話概況 [A description of the Mien language in Eshan]. 民族語文 (Minority languages of China), 1, 70–81. Zhao, Weifang (2007). 香港電影史 1897–2006 [The history of Hong Kong movies: 1897–2006]. Beijing: China Radio and Television Publishing House. Zheng, Qingjun (1999). 常德方言研究 [A study of the Changde dialect]. Hunan: Hunan jiaoyu chubanshe. Zheng, Yiqing (1997). 回輝話研究 [A study of the Huihui language]. Shanghai: Shanghai yuandong chubanshe. Zhou, Changji, & Ouyang, Yiyun (1998). 廈門方言研究 [A study of the Xiamen dialect]. Fujian: Fujian renmin chubanshe. Zhou, Lei (2002). 烏魯木齊話‘給’字句研究 [A study of the give-type sentences in the Ürümqi dialect]. 方言 (Dialects), 1, 16–23. Zhou, Rijian (1990). 新豐方言志 [A description of the Xinfeng dialect]. Guangzhou: Guangdong gaodeng jiaoyu chubanshe. Zhu, Dexi (1979). 與動詞‘給’相關的句法問題 [Syntactic issues related to the verb give]. 方言 (Dialects), 2, 81–87. Zhu, Dexi (1983). 包含動詞‘給’的複雜句式 [Complex sentences involving the verb give]. 中 國語文 [Studies of the Chinese language], 3, 161–166. Zhu, Dexi (1990). A preliminary survey of the dialectal distribution of the interrogative sentence patterns V-NEG-VO and VO-NEG-V in Chinese. Journal of Chinese Linguistics, 18(2), 209–229. Zhuang, Chusheng, & Huang, Tingting (2014). 十九世紀香港新界的客家方言 [The Hakka dialect spoken in the New Territories of Hong Kong in the 19th century]. Guangdong: Guangdong remin chubanshe.
Appendix 1
Survey of the syntactic functions of GIVE in Chinese dialects and neighboring non-Sinitic languages
This appendix lists the four syntactic functions performed by the double-object verb GIVE in Chinese dialects and neighboring non-Sinitic languages. A, B, C and D represent the following four syntactic functions respectively: Double-object verb, IO marker, passive marker and causative verb. Tones are also indicated unless they are not provided in the materials. Notice that not all the four functions of GIVE are found in the materials we consulted. I. The Yue dialects
(1) The double-object verb 畀
(a) The Guangfu group (廣府片) Hong Kong 香港
Guangzhou 廣州 (H. Gao, 1980, pp. 358, 360, 366, 370)
A B C D
ŋͻ13 pei35 pun35 sy55 k’ɵy13 (我畀本書佢, I gave him/her a book) ŋͻ13 sʊŋ33 pun35 sy55 pei35 k’ɵy13 (我送本書畀佢, I gave him/her a book) ŋͻ13 pei35 jɐn11 ta35 (我畀人打, I was beaten by someone) ŋͻ13 pei35 k’ɵy13 ts’œŋ33 kͻ55 (我畀佢唱歌, I let him/her sing)
A
niŋ11 jyn35 pεi35 sa:m53 kͻ33 jyt2 jan11 kuŋ55 k’œy13 (寧願畀三個月 人工佢, I’d rather give him/her 3 months’ salary) tsœŋ53 syn33 sεi35 ts’ͻu35 kͻ35 fan22 jan11 jat5 ŋ13 jat5 sap2 kͻŋ35 pεi35 ma13 tsai35 t’εŋ53 (將算死草嗰份人一五一十講畀馬仔聽, Tell Ma-zai all the details of that misery guy ) pεi35 k’œy13 kik5 tͻu33 ŋa:n13 tat2 tat2 mͻu13 wa22 kͻŋ35 (畀佢激到 眼突突冇話講, I was provoked by him/her and made speechless)
CIT110
B
C Foshan 佛山 (B. Li & He, 1992, pp. 59, 71, 72)
A B C
Zengcheng 增城 (W. T. He, 1993, pp. 209, 211, 251)
A B
C
D
畀本書我 (give me a book) (你講畀佢知, you tell him) (因為自己嘅仔甩牙講話唔清, 畀人陷害, Because my son could not speak clearly, he was framed)
pei35 pœŋ35 ʃi55 ŋͻi13 (畀本書我, give me a book) tʃɐu22 k’ͻŋ11 iɐu55 tʃε33–35 pei35 iɐŋ11 m̩11–45 kεŋ33–35 ə tʃi55 mͻ55 (就 狂懮借畀人唔見□之麼, then s/he was so worried that it would get lost after lending to someone) hɐu13 ʃi11 lεŋ11–51 tai22 iɐŋ11 tou55–45 pei35 k’œ13 nɐŋ35–51 tou35 (有時 連大人都畀佢撚倒, sometimes, even adults would also be tricked by him/her) ŋͻi13 tou55–45 m̩11–45 hɐŋ35–51 pei35 kͻi33 tʃɐi35 kɐm33 ʃɐi33 tʃ’ɐk55 mœŋ11 (我都唔肯畀個仔咁細出門, I was not willing to let my son leave home as he was so young)
202 Cantonese give and Double-Object Construction 265266267268269
Xinyi 信宜 (K. Luo, A 1987, pp 221, 222, B 228) C Gaozhou Baoan xiang 高州保安 鄉265
A B C
Beihai 北海市 (X. Chen & Chen, 2005, pp. 377, 379, 380, 383) Beihai Nankang 北 海南康 (X. Chen & Chen, 2005, pp. 377, 379, 380, 383)
A B C
pei35 pun35 sy53 ŋͻ23 (畀本書我, give me a book) nei23 kͻŋ35 pei35 k’øi23 tei53 tsi53 (你講畀佢哋知, you tell them) mou35 pei35 sui35 t’ui53 tsɐu35 (無畀水推走, it was not driven away by water) ŋͻ13pei35pin22ɬᴀm55men55nei13 (我畀□三文你, I gave you three dollars first) tʃœŋ55 ɬᴀm55 kei33 kͻ33 ʒᴀ55 fʊŋ55 ɬɐn33 pei35 nei13 (張三寄一封信 畀你, Zhang San sent you a letter) tʃœŋ55 ɬᴀm55 pei35 yɐn11 tei22 t’ɐu55 tε22 jɐt5 kͻ33 ŋɐn11 pᴀu55 (張三 畀人哋偷□一個銀包, Zhang San got his wallet stolen)
D
pei35 k’ei13 jɐt5 pun35 si55 (畀渠一本書, give a book to him/her) ŋͻ13 kwͻŋ35 kwͻ33 pei35 k’ei13 t’iŋ55 (我講過畀渠聽, I told him/her) pei35 k’ei13 ku35 tsuŋ33 la33 (畀渠估中了, it was correctly guessed by him/her) pei35 ŋͻ13 t’ɐi35 t’ɐi35 (畀我睇睇, let me have a look)
A B C D
pei35 pun35 sy55 k’ei55 (畀本書渠, give him/her a book) ŋͻ13 kwͻŋ35 kwͻ22 pei35 k’ei55 t’εŋ55 (我講過畀渠聽, I told him/her) pei35 k’ei55 ku35 tsuŋ33 (畀渠估中, it was correctly guessed by him/her) pei35 ŋͻ13 t’ɐi35 t’ɐi35 (畀我睇睇, let me have a look)
A Nanning 南寧 (J. Chu, 2007, B pp. 37, 39, 90, 91)266 C
D
畀兩本書我 (give me two books) 介紹隻朋友畀佢 (introduce a friend to him/her) 張三挨公安局捉了 (Zhang San was caught by the public security bureau) 畀我開門 (let me open the door)
(b) The Northern Delta group (北三角洲片) Nanhai 南海267
A C
Shunde 順德268
Gaoyao 高要269
A A C
ŋᴀi13 pɐi53 tʃy33 sᴀm55 kᴀi44 ŋɐn44 t’in44 nei13 sin55 (我畀住三個銀 錢你先, I give you three dollars first) ŋᴀi13 hyt44 ε55 sy33 kœk4 pɐi55 hy55 hiak4 t’an55 (我脫[了]□腳畀佢 踢親, I, barefooted, was kicked and injured) nei13 pei53 tʃy22 sᴀm55 mɐn44 tʃ’in35 ŋoi13 tʃᴀ35 (你畀住三文錢我□, you give me three dollars first) ŋͻi13 pei35 ʃam55 mɐn55 dʒi13 ʃin55 lᴀ55 (我畀三文住先啦, I give three dollars first) ŋͻi13 t’yt4 ə55 kœk4 pei45 ky45 t’iak4 ʃœŋ55 a55 ŋͻi13 (我脫[了]腳畀佢 踢傷□我, I, barefooted, was kicked and injured)
265. Taken from A. Yue-Hashimoto’s unpublished field notes. 266. No transcriptions are given for these sentences. 267. Taken from A. Yue-Hashimoto’s unpublished field notes. 268. Taken from A. Yue-Hashimoto’s unpublished field notes. 269. Taken from A. Yue-Hashimoto’s unpublished field notes.
Appendix 1. Survey of the syntactic functions of give 203
Tengxian 藤縣 (Yue-Hashimoto, 1979, pp. 65, 245, 250, 255)
A B C
ɂbi55 toŋ33 ɬəi42 (畀東西, give something) jən42 vəi22 ŋͻ24 mau24 ɬiu33 ɬəm42 tə22 ʃy42 ɂbi35 ni24 (因為我無小心 就輸俾你, I lost to you just because I was not careful) hͻ33 ti22 kͻ35 tʃek5 kəi33 ɲi44 kam55 a33, təu22 ɂbi55 tʃek5 ta22 əŋ44 ɂdam44 a44 hy33 lͻk3 (好似嗰隻雞兒噉呀,就俾隻大鷹啣吖去咯, just like that chicken, which was taken away by the eagle with its beak)
(c) The Southern Delta group (南三角州片) Dongguan 東莞 (X. Chen, 1993, pp. 191, 198, 210)
A B C D
Huazhou Xinxu 化州新墟270
A C
k’ui13 pɐi35 pun35 sy324 ŋͻ13 (佢畀本書我, s/he gave me a book) t’uŋ31 ŋͻ13 tͻi42 tsi324 pɐt4 pɐi35 kͻ35 vͻi42 tai42 tsø35 tsͻi35 (同我遞枝 筆畀嗰位大姐仔, pass a pen to that lady for me) kͻ35 sy13 zau42 zɐn21 pɐi35 hei42 ts’ø324 tsͻŋ42 hɐi35 (嗰處有人畀汽 車撞起, someone was hit by a car there) nɐi13 tsͻu42 zø13 tsuŋ35 hͻi42 pɐi35 zɐn21 taŋ324 sɐm324 (你做嘢總係畀 人担心, you always make people worried when doing something) ɓei45 ɐt5 ɓun44 nei44 lᴀ55 (畀一半你啦, Give you a half) ɓei45 yɐn11 kᴀ53 p’ᴀ35 tʃͻ35 k’oy44 kͻ44 ŋɐn11 ɓᴀu53 (畀人家扒咗佢個 銀包, the wallet was stolen by someone)
(d) The Liang Yang group (兩陽片) Yangchun 陽春271
A C
ŋo21 pei21 tʃi54 ɬᴀm44 jiŋ43 nei21 ɬiŋ44 (我畀住三元你先, I give you three dollars first) k’ei42 tei54 pei21 jɐŋ42 ŋɐk35 tʃo21 (其哋畀人呃咗, they were cheated by someone)
(e) Overseas Kuala Lumpur 吉隆坡 (X. Chen, 2003, pp. 332–335)
A B C D
pei35 jɐt5 pun35 sy55 k’œy13 (畀一本書佢, give him a book) ŋͻ13 wa22 kʷͻ33 pei35 k’œy13 tsi55 (我話過畀佢知, I told him/her) pei35 k’œy13 ku35 tsuŋ33 tsͻ35 (畀佢估中咗, it was correctly guessed by him/her) pei35 ŋͻ13 t’ɐi35 ha13 (畀我睇下, let me have a look)
270271
(2) The double-object verb [ɂi] and [ei] Taishan Dancun 台山淡村 (Yue-Hashimoto, 2005, pp. 392, 394)
A B C
ɂi55 ɬam33 kə33 ŋan22 t’eiŋ35 ni33 ʃͻi55 kin55 (畀三個銀錢你使緊, give you three coins to use) k’ui33 kau33 tʃͻn55 nai55 t’eiŋ35 ɂi55 ni33 (佢交轉乃錢畀你, s/he passed my money to you) k’ui33 ɂi55 niεk55 hau33 ə33 k’ui33 kə33 hͻ22 pau35 (佢畀偌偷□佢個 荷包, his/her wallet was stolen by someone)
270. Taken from A. Yue-Hashimoto’s unpublished field notes. 271. Taken from A. Yue-Hashimoto’s unpublished field notes.
204 Cantonese give and Double-Object Construction A Kaiping 開平 (Deng, 2000, pp. 85, B 91, 230) C D
ei33 vͻi33 ts’a11 ŋͻi33 (畀杯茶我, give me a cup of tea) tiε33 liεŋ55 vͻn55 si33 ei55 nei33 (借兩本書畀你, lend you two books) ŋͻi33 ei55 kui33 ŋak55 e33 lͻ55 (我畀佢呃誒咯, I was cheated by him/her) m̩11 ei55 jip32 (唔畀入, not allow to enter)
(3) The double-object verb 分
Yulin 玉林 (J. Chu, 2007, pp. 37, 39, 90, 91)
A B C D
Sanya Mai 三亞邁話 (D. Jiang et al., 2007)
A B C D
分本書我 (give me a book) 分蘋果分大家吃 (give the apples to everyone) 張三著公安局捉了 (Zhang San was caught by the public security bureau) 分我做工 (let me work)
ŋͻ33 vͻn13 aɂ5 bun33 θi13 vͻn13 ki55 (我分一本書分佢, I gave hime/ her a book) ŋͻ33 vͻn13 aɂ5 bun33 θi13 vͻn13 ki55 (我分一本書分佢, I gave hime/ her a book) hu33 vͻn13 nͻn55 tua13 ts’iaɂ1 liu33 (虎分人抓著了, the tiger was caught) vͻn33 ŋͻ33 hͻn13 aɂ5 ha21 (分我看一下, let me have a look)
(4) The double-object verb [hai]272 (a) The Inland sub-group of the Guangfu group (廣府內陸小片) Lianzhou 廉州272
A B C D
hɐi44 pun35 su45 ku44 (□本書渠, give him/her a book) ŋo35 kwεŋ35 kwo44 hɐi44 k’u44 t’ɐŋ44 lε21 (我講過□渠聽□, I told him/her) hɐi44 ku44 ts’ai45 tsuŋ44 lε21 (□渠猜中了, it was correctly guessed by him/her) hɐi44 ŋo35 hon44 hon44 (□我看看, let me have a look)
Beihai Wa dialect A 北海佤話 (X. Chen, C 2004, pp. 377, 379, 380, 383) D
hɐi11 ɐt5 pun11 su213 ku55 (□一本書渠, give him/her a book) hɐi11 ku55 ts’ai213 tsuŋ33 lε33 (□渠猜中了, it was correctly guessed by him/her) hɐi11 ŋa35 han11 han11 (□佤看看, let me have a look)
Beihai Haibian 北海海邊話 (X. Chen, 2004, pp. 377, 379, 380, 383)
A
hɐi11 ɐt5 pun11 sɐu213 hɐi11 kɐu55 (□一本書□□, give him/her a book)
B
ŋa55 kaŋ11 ku11 hɐi11 kau55 tεŋ213 (佤講過□□聽, I told him/her)
C
hɐi11 kɐu55 ts’ai213 tsik5 (□□猜著, it was correctly guessed by him/ her)
D
hɐi11 ŋa55–33 k’an55 k’an55 (□佤看看, let me have a look)
272. Taken from A. Yue-Hashimoto’s unpublished field notes.
Appendix 1. Survey of the syntactic functions of give 205
(b) The Northern Delta group (北三角洲片) Nanning Sitang 南寧四塘 (J. Chu, 2007, pp. 31, 32, 34, 35, 36, 84, 85, 87, 89)
A B C
Nanning Shibu 南寧石埠 (J. Chu, 2007, pp. 31, 32, 34, 35, 36, 84, 85, 87, 89)
A B C
Nanning Tianyang 南寧田陽 (J. Chu, 2007, pp. 31, 32, 34, 35, 36, 84, 85, 87 & 89)
A B C
D
D
D
許書許我 (give me a book) 許書許我 (give me a book) 張三挨/著公安局捉了 (Zhang San was caught by the public security bureau) 許我吃飯 (let me have rice)
許錢我 (give me some money) 介紹朋友給佢 (introduce some friends to him/her) 張三挨/著公安局抓去了 (Zhang San was caught by the public security bureau) 許我吃飯 (let me have rice)
許錢許我 (give me some money) 許錢許我 (give me some money) 張三挨公安局抓了 (Zhang San was caught by the public security bureau) 許我吃飯 (let me have rice)
II. The Hakka dialects
(1) The double-object verb 分
(a) Guangdong and Taiwan group (粵台片) (i) Guangdong province
Meixian 梅縣 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
A B C D
Dapu Fenglang 大埔楓朗 (Chiang, 2003, pp. 146, 149)
A C D
Dapu Gaopo 大埔高坡 (Chiang, 2003, pp. 146, 149)
Jiexi 揭西 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
A B C D A C D
pun44 it1 pun31 su44 pun44 ŋai11 (分一本書分亻厓, give me a book) pun44 it1 pun31 su44 pun44 ŋai11 (分一本書分亻厓, give me a book) pun44 ki11 t’ͻn44 ts’ͻk5 εi31 (分渠斷著哩, it was correctly guessed by him/her) pun44 ŋai11 k’ͻn52 it1 ha44 (分亻厓看一下, let me have a look) pun33 ŋai35 ʒit2 pun31 ʃu35 (分亻厓一本書, give me a book) pun113 ki33 mau35 tʃ’ok5 le31 (分佢□著咧, it was correctly guessed by him/her) pun35 ŋai35 k’on31 ha31 (分亻厓看下, let me have a look) pun35 pun31 ʃu33 pun35 ŋai13 (分本書分亻厓, give me a book) pun35 pun31 ʃu33 pun35 ŋai13 (分本書分亻厓, give me a book) pun52 ki113 ko31 te55 le31 (分佢講□咧, it was correctly guessed by him/her) pun35 ŋai113 k’on55 ha55 (分亻厓看下, let me have a look) pun53 pun21 ʃu53 ŋai24 (分本書亻厓, give me a book) pun53 ki24 t’ͻn53 tui42 liau24 (分渠斷對了, it was correctly guessed by him/her) pun53 ŋai24 k’ͻn42 ha21 lεi24 (分亻厓看下哩, let me have a look)
206 Cantonese give and Double-Object Construction Wengyuan 翁源 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454) Xinfeng 新豐 (R. Zhou, 1990, pp. 174, 175)
A B C D A B C
Dongguan 東莞 (X. Chen, 1993, pp. 341, 343) Qingxi 清溪 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454) Liannan 連南 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
A C A B C D A B C D
Gaozhou Xindong 高州新垌 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215) Huazhou Xin’an 化州新安 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215) Lianjiang Shijiao 廉江石角 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215)
A C D A C D A B C D
Lianjiang Qingping 廉江青平 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215)
A B C D
pun22 ŋai51 it2 pun21 sy22 (分亻厓一本書, give me a book) ŋai51 kͻŋ21 pun22 ti22 lεi0 (亻厓講分渠知哩, I told him/her) pun22 ki51 ts’ai22 tou21 lεi0 (分渠猜到哩, it was correctly guessed by him/her) pun22 ŋai51 k’on55 ha21 (分亻厓看下, let me have a look) pun44 ni24 sip5 tiau53 ts’ian24 (分你十吊錢, give you ten dollars) van24 liͻŋ31 tiau53 ts’ian24 pun44 ni24 (還兩吊錢分你, return two dollars to you) kuk1 pun44 sui31 tuk1 sip1 tet1 li44 (谷分水督濕得哩, the paddy was soaked by water) lau31 su33 pun33 k’i11 zit2 pun31 su33 (老師分佢一本書, the teacher gave him/her a book) lau31 t’ai31 pun33 k’i31 ta31 hͻi33 zit2 k’en11 (老弟分佢打開一拳, the younger brother was fisted by him/her) pin33 it22 pun21 su33 ŋai11 (分一本書亻厓, give me a book) ŋai11 kͻŋ21 kͻ42 pin33 k’i11 t’aŋ33 (亻厓講過分渠聽, I told him/her) pin33 ki11 ts’ai33 tui42 e0 (分渠猜對哩, it was correctly guessed by him/her) pin33 ŋai11 k’ͻn42 ha21 (分亻厓看下, let me have a look) pͻn44 it2 pͻn22 sy44 ŋai24 (分一本書亻厓, give me a book) ŋai24 kͻŋ22 kəu51 pͻn44 ki24 ti44 (亻厓講過分渠知, I told him/her) pͻn44 ki24 ku22 tʃoŋ51 εi0 (分渠估中哩, it was correctly guessed by him/her) pͻn44 ŋai24 k’uͻn51 ha51 tsi22 (分亻厓看下子, let me have a look) pən44 pən31 sy44 ŋai24 (分本書亻厓, give me a book) pən44 k’i24 ts’ai44 tsuŋ41 tε24 (分渠猜中□, it was correctly guessed by him/her) pən44 ŋai24 t’ε31 ha44 (分亻厓睇下, let me have a look) bun55 bun31 si55 ŋai35 (分本書亻厓, give me a book) bun55 ki35 ku34 ts’ͻk5 dε55 (分渠估著爹, it was correctly guessed by him/her) bun55 ŋai35 hͻn33 hͻn33 (分亻厓看看, let me have a look) pun45 pun31 su45 ŋai12 (分本書亻厓, give me a book) ŋai12 hͻk5 kͻ44 pun45 ki12 t’ɐn45 e45 (亻厓學過分渠聽欸, I told him/her) pun45 ki12 ku31 tsoŋ44 e45 (分渠估中欸, it was correctly guessed by him/her) pun45 ŋai12 hͻn44 ha44 (分亻厓看下, let me have a look) pun55 pun31 su55 ŋai24 (分本書亻厓, give me a book) ŋai24 kͻŋ31 kͻ33 (pun55) ki24 t’iaŋ55 (亻厓講過(分)渠聽, I told him/her) pun55 ki24 ku31 tͻ31 ε33 (分渠估倒欸, it was correctly guessed by him/her) pun55 ŋai24 hͻn33 hͻn33 (分亻厓看看, let me have a look)
Appendix 1. Survey of the syntactic functions of give 207
Dianbai Shalang 電白沙琅 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215)
A B C D
pən34 zət2 pən31 si34 ŋai213 (分一本書亻厓, give me a book) ŋai213–31 kͻŋ31 kͻ54 pən34 k’i21–31 t’εn34 (亻厓講過分渠聽, I told him/her) pən34 k’i213 ku31 ts’ͻk5 ε213 (分渠估著欸, it was correctly guessed by him/her) pən34 ŋai213 t’ε31 t’ε31 (分亻厓睇睇, let me have a look)
(ii) Taiwan Taizhong Dongshi 台中東勢 (Chiang, 2003, pp. 146, 149)
A C D
Hailu 海陸 (S. Yang, A 1992, pp. 113, 115, 117, 125, 134, 136, 154, 184) B C
D Sixian 四縣 (S. Yang, 1992, pp. 113, 115, 117, 125, 134, 136, 154, 184)
A C
D
pun35 ŋai113 ʒit2 pun31 ʃu33 (分亻厓一本書, give me a book) pun33 ki113 ton113 to31 le33 (分佢團著咧, it was correctly guessed by him/her) pun35 ŋai113 k’on55 ha52 (分亻厓看下, let me have a look) tsioŋ53 ȵiu55 tʃu53 kok5 ʒit5–32 tʃak5, tsiu13 lioŋ13 t’uŋ13 suŋ31 to31 ŋ13 ka53 pun53 ki55 (將牛豬各一隻,酒兩桶送到女家分佢, (someone) sent an ox, a pig, two barrels of wine to the bride’s home and gave these things to them) ts’iu22 tsioŋ53 lioŋ13 tʃak5–32 kok5 tsia31 pun53 liuŋ55 la0 (就將兩隻角 借分龍啦, somone then lent the two horns to the dragon) ka53 tʃuŋ53 ts’oŋ55 ʒu53 ʒit5–32 tʃak5 kim53 fu55 t’iap53 pun53 ȵin55 t’eu53 na53 hi0 (家中藏有一隻金蝴蝶分人偷拿去, the golden butterfly in the house was stolen by someone) ȵi55 ts’iu22 pun53 ŋai13 ts’oi53 t’eu55 ts’ian55 tseu13, ȵi55 ken53 heu31 poi31 (你就分我在頭前走,你跟後背, you let me walk in the front and you follow me) a24 pa24 pun24 ŋai31 kai0 tuŋ24 si0 tet2 si0 (阿爸分我個東西得是, father gave me something) lai55 ə0 kai0 sin55 miaŋ55 it2 t’in55 voi55 pun24 t’ian11 lo11 tsin24 t’ot5 hi0 la0 (孻子個性命一定會分田螺精奪去啦, the life of the youngest son would be taken away by the demon of the field snail) put2 tet2 put2 ts’iu55 iuŋ55 mo24 lam24 tsoŋ24 tʃ’ut2 loi31 pun24 t’ai55 ka24 k’on55 (不得不就用毛籃裝出來分大家看, they had to use a basket to hold it in order for everybody to see it)
(b) Fujian province (福建) Xiuzhuan 秀篆 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
A B C D
Ninghua 寧化 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
t’εi33 pun51 ʃy13 pun13 ŋai54 (提本書分亻厓, give me a book) ŋai54 kͻŋ51 tit24 ky54 ti13 lεu0 (亻厓講得渠知了, I told him/her) tit24 ky54 p’ͻi13 tͻu51 lεu0 (得渠□到了, it was correctly guessed by him/her) tit24 ŋai54 ŋiaŋ33 zit24 ŋiaŋ33 (得亻厓目英一目英, let me have a look)
A
pε̃i31 ŋa33 i5 pε̃ i31 su33 (分亻厓一本書, give me a book)
C
pε̃i33 kɤ33 ts’εi33 ts’o68 lεi0 (分渠猜著哩, it was correctly guessed by him/her)
208 Cantonese give and Double-Object Construction
(c) Guangxi (廣西) Fumian 福綿 (X. Chen, 2004, pp. 349, 350, 353)
Xingye 興業 (X. Chen, 2004, pp. 349, 350, 353) Beiliu 北流 (X. Chen, 2004, pp. 349, 350, 353) Rongxian 容縣 (X. Chen, 2004, pp. 349, 350, 353) Luchuan 陸川 (X. Chen, 2004, pp. 349, 350, 353) Bobai 博白 (X. Chen, 2004, pp. 349, 350, 353)
A B C D
pun44 pun31 su44 ki11 (分本書渠, give him/her a book) ŋai11 kͻŋ31 kuͻ52 pun44 ki11 t’aŋ52 (亻厓講過分渠聽, I told him/her) pun44 ki11 ku31 tsuŋ52 e31 (分渠估中□, it was correctly guessed by him/her) pun44 ŋai11 k’ͻn52 k’ͻn52 (分亻厓看看, let me have a look)
A B C D
pun33 pun21 su33 ki21 (分本書渠, give him/her a book) ŋai21 kͻŋ21 kͻ42 pun33 ki21 t’aŋ42 (亻厓講過分渠聽, I told him/her) pun33 ki21 ku21 tsuŋ42 (分渠估中, it was correctly guessed by him/her) pun33 ŋai213 k’ͻn42 k’ͻn42 (分亻厓看看, let me have a look)
A B C D
pun44 pun21 su44 ki21 (分本書渠, give him/her a book) ŋai11 kͻŋ21 kͻ52 pun44 ki21 t’aŋ52 (亻厓講過分渠聽, I told him/her) pun44 ki11 ku21 tsuŋ52 le21 (分渠估中了, it was correctly guessed by him/her) pun44 ŋai11 k’ͻn52 k’ͻn52 (分亻厓看看, let me have a look)
A B C D
pun33 pun21 ɬu33 ki11 (分本書渠, give him/her a book) ŋai11 kͻŋ21 kuͻ42 pun33 ki11 t’aŋ42 (亻厓講過分渠聽, I told him/her) pun33 ki11 ku21 tsuŋ42 (分渠估中, it was correctly guessed by him/her) pun33 ŋai11 k’ͻn42 k’ͻn42 (分亻厓看看, let me have a look)
A B C
pun44 pun21su44 ki21 (分本書渠, give him/her a book) ŋai23 kͻŋ21 kͻ42 pun44 ki23 t’aŋ42 (亻厓講過分渠聽, I told him/her) pun44 ki23 ku21 tsuŋ42 liau21 (分渠估中了, it was correctly guessed by him/her) pun44 ŋai23 k’ͻn42 k’ͻn42 (分亻厓看看, let me have a look) pun44 pun21 su44 k’i21 (分本書渠, give him/her a book) ŋai23 kͻŋ21 kͻ42 pun44 k’i23 t’aŋ42 (亻厓講過分渠聽, I told him/her) pun44 k’i23 ku21 tsuŋ42 liau21 (分渠估中了, it was correctly guessed by him/her) pun44 ŋai23 k’ͻn42 k’ͻn42 (分亻厓看看, let me have a look)
D A B C D
(d) Overseas
Tu’nai 土乃 (X. Chen, 2003, pp. 332 – 335)
A B C D
pun33 pun31 su33 ki21 (分本書佢, give him/her a book) ŋai21 kͻŋ31 kͻ42 pun33 ki21 t’aŋ42 (亻厓講過分佢聽, I told him/her) pun33 ki21 ts’ai33 tsuŋ42 hͻi33 (分佢猜中開, it was correctly guessed by him/her) pun33 ŋai21 k’ͻn42 ha42 (分亻厓看下, let me have a look)
(2) The double-object verb 畀 Guangdong province (廣東) Hong Kong 香港 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
A C D
pi34 pun41 su34 gai21 (畀本書亻厓, give me a book) pi34 gi34 ku41 tsuŋ52 ε0 (畀渠估中兒, it was correctly guessed by him/her) pi34 gai34 k’ͻn52 ts’in52 (畀亻厓看陣, let me have a look)
Appendix 1. Survey of the syntactic functions of give 209
Huizhou 惠州 (R. Liu, 1991, pp. 234, 238, 254)
A B
C Yangjiang Tangkou 陽江塘口 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215)
A C D
Yangjiang Sanjia A 陽江三甲 (R. L. Li, B 1999, pp. 208, 209, C 212, 213, 214, 215) D Xinyi Sihe 信宜思賀 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215)
A B C D
Xinyi Qianpai 信宜錢排 (R. L. Li, 1999, pp. 208, 209, 212, 213, 214, 215)
A B C D
k’y22 pi35–55 kͻn35–55 pit45 ŋͻi213 (佢畀桿筆我, s/he gave me a pen) a33 kͻ33 mut45–5 ȵia213 tu33 mu22 hun33 pi35–55 lau35–55 t’iε213 (阿哥乜 惹都無分畀老弟, the elder brother did not share anything with his younger brother) ŋͻi13 pi35–55 k’y22 hak45 it45–5 kiaŋ33 (我畀佢嚇一驚, I was scared by him/her) pi21 it2 pun21 su35 ŋai24 (畀一本書亻厓, give me a book) pi21 k’i24 ku21 tsuŋ52 lε0 (畀渠估中咧, it was correctly guessed by him/her) pi21 ŋai24 hͻn52 hͻn52 (畀亻厓看看, let me have a look) pi31 pun31 su35 ŋai12 (畀本書亻厓, give me a book) ŋai12 kͻŋ31 ko42 pi31 k’i12 t’ən35 (亻厓講過畀渠聽, I told him/her) pi31 k’i12 ku31 tsuŋ42 le0 (畀渠估中咧, it was correctly guessed by him/her) pi31 ŋai12 t’əi31 t’əi31 (畀亻厓睇睇, let me have a look) pi45 pun31 sy45 ŋai24 (畀本書亻厓, give me a book) ŋai24 kͻŋ31 kͻ51 pi45 ki24 t’ɐn45 (亻厓講過畀渠聽, I told him/her) pi45 ki24 ku31 tsoŋ51 ŋε33 (畀渠估中欸, it was correctly guessed by him/her) pi45 ŋai24 t’ɐi31 t’ɐi31 (畀亻厓睇睇, let me have a look) pi34 pun31 sy34 ŋai23 (畀本書亻厓, give me a book) ŋai23 kͻŋ31 kͻ52 pi34 ki23 t’ən34 (亻厓講過畀渠聽, I told him/her) pi34 ki23 ku31 tsoŋ52 lͻk2 (畀渠估中咯, it was correctly guessed by him/her) pi34 ŋai23 t’ei23 ha31 (畀亻厓睇下, let me have a look)
(3) The double-object verb 拿 (a) Jiangxi province (江西省)
Dayu 大余 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454) Nankang 南康 (L. Liu, 1999, pp. 462, 633) Anyuan 安遠 (L. Liu, 1999, pp. 462, 633)
Yudu 于都 (L. Liu, 1999, pp. 462, 633)
A C A B C A B C A B C
la5 ŋa42 i5 pε̃ 42 ɕy33 (拿亻厓一本書, give me a book) la5 tɕi42 ts’æ33 tͻ0 lͻ0 (拿渠猜到了, it was correctly guessed by him/her) na44 拿 (to give) na44 拿 na44 拿 na35 拿 (to give) na35 拿 na35 拿 na35 拿 (to give) na35 拿 na35 拿
210 Cantonese give and Double-Object Construction Longnan 龍南 (L. Liu, 1999, pp. 462, 633)
Quannan 全南 (L. Liu, 1999, pp. 462, 633) Dingnan 定南 (L. Liu, 1999, pp. 462, 633)
A B C A B C A B C
na24 拿 (to give) na24 拿 na24 拿 na24 拿 (to give) na24 拿 na24 拿 na35 拿 (to give) na35 拿 na35 拿
(b) Fujian province (福建省) Liancheng 連城 (Xiang, 1997, pp. 313, 361, 410)
A B C D
nu33 i35 pu51 ɕie3 tiɥə55 (拿一把扇佢, give him/her a fan) saŋ3 i35 paŋ51 ʂЧə33 nu33 tiЧə55 (送一本書拿佢, give him/her a book) sa33 nu33 fi51 t’iu33 tsie51 ə3 (衫拿水推走呃, the shirt was washed away) la5 ŋa42 k’ε̃ 24 i5 ha42 (拿亻厓看一下, let me have a look)
(4) The double-object verb 得 and 把
Changting 長汀 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
Heyuan 河源 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
B C D A B C D
na33 i24 peŋ42 ʃʉ33 te24 ŋai33 (拿一本書得亻厓, give me a book) te24 ke33 ts’ai33 tͻ54 le0 (得渠猜到哩, it was correctly guessed by him/her) te24 ŋai33 niaŋ54 ha21 tsɿ42 (得亻厓目英下子, let me have a look) pa24 puan24 sy33 ŋuai12 (把本書亻厓, give me a book) ŋuai12 kɐŋ24 pau0 pa24 k’i31 t’iaŋ33 lͻ0 (亻厓講□把渠聽囉, I told him/her) pa24 k’i31 ku24 tsoŋ12 pau0 (把渠估中□, it was correctly guessed by him/her) pa24 ŋuai12 t’iε24 ha55 lei0 (把亻厓睇下哩, let me have a look)
III. The Min dialects
(1) The double-object verb 乞
(a) The Eastern / Costal group (i) The Northereastern sub-group
Fuzhou 福州 (H. Lin, 2002, p. 127)
Changle 長樂 (H. Lin, 2002, p. 127)
A C D
tsy44 to53 so31 ßuoŋ31 k’øyɂ23 i44 (書馱蜀本乞伊, give me a book) k’i21 nøyŋ53 p’aɂ23 suoŋ44 ŋo0 (乞儂拍傷去, I was beaten by someone) k’i2 ŋuai31 k’ɑŋ213 nɑ242 (乞我看下, let me have a look)
A C
tsy44 to53 lo22 ßuoŋ22 k’yk3 i44 (書馱蜀本乞伊, give me a book) k’yk21 nøyŋ53 p’aɂ23 suoŋ44 ŋo0 (乞儂拍傷去, I was beaten by someone) k’y44 ŋuoi22 k’ɑŋ21 nɑ343 (乞我看下, let me have a look)
D
Appendix 1. Survey of the syntactic functions of give 211
Fuqing 福清 (H. Lin, 2002, p. 127)
Yongtai 永泰 (H. Lin, 2002, p. 127)
Ningde 寧德 (H. Lin, 2002, p. 127)
Gutian 古田 (H. Lin, 2002, p. 127)
Fuan 福安 (H. Lin, 2002, p. 127)
Shouning 壽寧 (H. Lin, 2002, p. 127)
Zhouning 周寧 (H. Lin, 2002, p. 127) Fuding 福鼎 (H. Lin, 2002, p. 127)
Cangnan 蒼南 (Akitani, 2005, pp. 245, 247, 251)
A C D
to44 syo44 puoŋ32 tsy53 k’yɂ12 i53 (馱蜀本書乞伊, give me a book) k’y5 nøŋ44 p’aɂ12 syoŋ53 ŋyͻ0 (乞儂拍傷去, I was beaten by someone) k’yɂ5 ŋua32 k’aŋ53 k’aŋ21 (乞我看看, let me have a look)
A C D
tsy44 to53 so44 ßuoŋ31 k’y21 i44 (書馱蜀本乞伊, give me a book) k’yɂ3 nͻyŋ453 p’aɂ3 suoŋ44 ŋo0 (乞儂拍傷去, I was beaten by someone) k’y3 ŋuoi31 k’aŋ212 a242 (乞我看下, let me have a look)
A C D
k’ik2 i44 θø44 ßoŋ42 tsy44 (乞伊蜀本書, give me a book) k’ik5 nœŋ22 p’aɂ2 θyøŋ44 lo0 (乞儂拍傷了, I was beaten by someone) k’ik2 ua42 k’aŋ35 k’aŋ35 (乞我看看, let me have a look)
A C D
tsy55 to33 syø21 ßuoŋ42 k’yk24 i44 (書馱蜀本乞伊, give me a book) k’yk33 nøyŋ55 p’aɂ2 syøŋ55 ŋyø21 (乞儂拍傷去, I was beaten by someone) k’øk5 ŋua42 ts’y21 la324 (乞我覷下, let me have a look)
A C D
tsøi44 tͻ11 seik44 puŋ42 k’øk5 ei44 (書馱蜀本乞伊, give me a book) k’øk5 nœŋ11 p’ak55 sioŋ44 tik5 (乞儂拍傷得, I was beaten by someone) k’øk5 ŋo42 ε35 a31 (乞我睨下, let me have a look)
A C D
tͻ11 siɂ33 puoŋ42 tsy33 k’iɂ5 i33 (馱蜀本書乞伊, give me a book) k’iɂ5 neŋ11 p’a24 syoŋ33 niɂ5 (乞儂拍傷得, I was beaten by someone) k’iɂ5 ua42 oŋ33 oŋ212 (乞我望望, let me have a look)
A C D
tͻ21 sik5 puan42 tsy44 k’yk5 i44 (馱蜀本書乞伊, give me a book) k’yk5 nœŋ21 p’aɂ5 θyəŋ44 le0 (乞儂拍傷得, I was beaten by someone) k’yk5 uo42 ε24 lia213 (乞我睨下, let me have a look)
A
to21 θiɂ23 puoŋ55 tsi44 tsak42 k’iɂ4 i44 (馱蜀本書冊乞伊, give me a book) k’iɂ4 neŋ21 p’aɂ4 θioŋ44 liɂ4 (乞儂拍傷得, I was beaten by someone)
C A B C
Taishun 泰順 (Akitani, 2005, pp. 245, 247, 51)
A C
ŋ35 ɕy44 k’əɂ5 (iəɂ3) paŋ35 k’əɂ5 o0 (你書乞(一)本乞我, you gave me a book) (pəɂ5) ɕiəɂ4 kəɂ5 məɂ3 za11 dau11 k’əɂ5 o35 ((把)許個物事馱乞我, take that thing and give it to me) ɕiəɂ5 (iəɂ5) kø332 vẽ45 k’əɂ5 i35 dza11 ioŋ11 m213 dy11 a0 la0 (許(一)個 碗乞伊在用唔著啊了, that bowl was not used by him/her)
ŋ455 hεɂ5 ŋa455 sιɂ3 puͻi35 ɕy213 (你乞我蜀本書, you gave me a book) ɕy455 kͻi35 υaŋ455 hεɂ5 i455 p’a53 tai0 la0 (許個碗乞伊拍逮了, that bowl was broken by him/her)
(ii) The Xinghua sub-group
Putian 莆田 (Z. T. Chen & Li, 1991, 95, 116)
A C
ts’a11 ɬœɂ35 pue453 k’œɂ42 kua453 (冊蜀本乞我, give me a book) kεɂ11 naŋ13 tiaɂ11 kia13 (乞農扌弱行, someone was caught)
212 Cantonese give and Double-Object Construction
(iii) The Southern Min sub-group
Datian 大田 (Z. T. Chen & Li, 1991, 95, 116)
Longxi 龍溪 (Z. T. Chen & Li, 1991, 95, 116)
Jieyang 揭陽 (Tung, 1959, pp. 798, 802, 808, 910, 911, 912, 915)
A C
tsi33 tseɂ5 pueŋ53 k’e31 bua53 (書蜀本乞我, give me a book) k’eɂ3 laŋ24 lia24 k’i0 (乞農扌弱去, someone was caught)
A C
tsy33 k’ia24 ɕie33 pũ55 ŋua55 (書乞蜀本我, give me a book) k’a24 nɤŋ12 na33 k’y53 (乞農扌弱去, someone was caught)
A
ua53 tsai53 k’eɂ44 lə53 gou11 tsap11 ŋeŋ55 (我再給你五十銀, I gave you another 50 taels) tsi35 kai11 pok11 dzeŋ55 tsiu11 tsiaŋ33 i33 tsu35 dzeŋ55 kai0 tsua35 tiau55 saŋ52 k’eɂ44 he11 hou11 lai55 t’oi53 (這個僕人就將他主人的紙條送 給夏狐狸看, this servant then gave the paper slip of his master to the fox to read) i44 kai0 kaŋ33 tsĩ55 tsiu11 k’eɂ44 tsi35 kai11 pu51 naŋ55 siõ11 tsiaɂ44 k’ə0 (他的工錢就給這個人富人吞沒去, his salary was then devoured by this rich man) haĩ11 ua35 lai11 tiaŋ35 kai11 pŋ35 nia53 k’eɂ44 lə35 t’oi53 (讓我來施展 個本領給你看, let me show you my calibre)
B
C
D A Suixi 遂溪 (Yue-Hashimoto, 1985, pp. 358 – 360) B C Leizhou 雷州 (L. Lin, 2006, pp. 228, 232)
mba53 sieŋ13 k’i55 lu53 sa13 kai33 ŋieŋ22 (我先給汝三個銀, I gave you three coins first) ɂi13 kia21 ɂkuεɂ21 ʒaɂ33 paŋ13 sieŋ21 k’i55 lu53 (伊寄過一封信給汝, I sent you a letter) ɂi13 k’i53 naŋ22 t’au13 ʒaɂ3 kai33 ŋieŋ22 tæ13 (伊給農偷一個銀袋, his/ her wallet was stolen by someone)
A B C
ba52 k’i55 pui52 tsu213 k’i55 i213 (我乞本書乞伊, I gave him/her a book) i213 tsio55 pui52 tsu213 k’i55 ba52 (伊借本書乞我, s/he lent me a book) a55 bu11 k’i55 naŋ11 t’ai11 la55 (□牛乞人治啦, the ox was butchered by someone)
A C D
k’oiɂ2 i33 tsek5–2 puŋ53–33 ts’eɂ2 (給伊蜀本冊, give him/her a book) k’eɂ2–4 i33 aɂ2 tioɂ2 (乞伊押著, someone was caught by him/her) k’eɂ2 ua53–24 t’oi53 tse11 (乞我睇□, let me have a look)
(iv) Overseas
Xinshan 新山 (X. Chen, 2003, pp. 332–335)
(2) The double-object verb 互 or 與 The Southern Min sub-group A Xiamen 廈門 (C. Zhou & Ouyang, 1998, pp. 332 – 335) B
C D
gua53 hͻ11–21 i55–11 tsit55–31 paɂ31–53 k’ͻ55 (我互伊一百簵, I gave him/her 100 dollars) li53 tsu53 e35–11 miɂ5–21 kiã11 lͻŋ53–55 si11–21 p’aŋ35–11 hͻ11–21 pat5–21 laŋ35 tsiaɂ55 (你煮的東西全部捧給別人食, Give what you have cooked to people to eat) gua53 hͻ11–21 gun53–55 lau11–31 bu53 me11 (我互阮老母罵, I was scolded by my mother) kan53–55 hͻ11–21 gun53–55 tan53–55 kaɂ31–5 kui55–11 puã31–53 lit5 (敢互 阮等甲規半日, you dare let me wait for you for half day)
Appendix 1. Survey of the syntactic functions of give 213
Zhangzhou 漳州 (Z. T. Chen & Li, 1991, pp. 95, 116)
Taiwan 台灣 (H. Yang, 1995, pp. 289, 290)
A C
ts’eɂ32 tsit121 pun53 hͻ22 gua12 (冊蜀本與我, give me a book) hͻ22 laŋ12 liaɂ22 k’i (與農扌弱去, someone was caught)
A B
gua53 ho22 i44 lŋ44 liã53 sã44 (我與伊兩領衫, I gave you two shirts) kia31 ho22 a44 iŋ44 tsit33 pun53 ts’εɂ22 (寄與阿英蜀本冊, send a book to Ah Ying) i44 ho22 a44 iŋ44 kiã44 tioɂ33 (伊與阿英驚著, s/he was scared by Ah Ying)
C
(3) The double-object verb 欠 The Central Min sub-group Yongan 永安 (Z. T. Chen & Li, 1991, pp. 95, 116)
Shaxian 沙縣 (Z. T. Chen & Li, 1991, pp. 95, 116)
A C
k’εiŋ35 ŋuɒ52 kuɒ21 puã21 ʃy52 (欠我寡本書, give me a book) k’εiŋ35 lɑ̃33 lɒ54 k’ա35 (欠人拿去, someone was caught)
A C
ʃy33 nͻ33 ka24 puĩ21 k’iŋ24 gua33 (書拿個本欠我, give me a book) k’iŋ24 lεiŋ31 nɑ̃53 k’o24 (欠人拿去, someone was caught)
(4) The double-object verb 納
The Northwestern sub-group Jian’ou 建甌 (Z. T. Chen & Li, 1991, pp. 95, 116)
A C
na42 tsi42 pͻŋ21 sy54 na24 uε42 (拿蜀本書納我, give me a book) na24 neiŋ33 na42 k’ͻ33 (納人拿去, someone was caught)
(5) The double-object verb [t’ͻ] Jinjiang 晉江 (Tung, 1959, pp. 798, 802, 808, 910, 911, 912, 915)
A
B C
D
k’uã55 kĩ55 tsit5 ge11 hak11 sŋ44 t’ͻ11 i44 nŋ11 pa53 niu35 gun24 (看見 這個學生給他兩百兩銀, someone saw this student give him 200 silver taels) gua55 ia35 e11 ki55 tsit11 nŋ11 haŋ33 səɂ55 t’ͻ11 lə35 t’iã44 (我還能記得 一兩項說給你聽, I still remember one or two items to tell you) i44 e0 kə35 tsi55 e11 ts’ut55 t’ͻ55 t’ͻ55 laŋ11 ts’io31 (他的舉止會露馬腳 給人家笑, his behavior would betray himself and be laughed at by others) t’ͻ11 tsit55 ge11 am11 kã35 sai31 kuã35 kin55 gueɂ53 (使這個傻女壻趕 緊夾, (something) made this silly son-in-law clip tightly)
(6) The double-object verb 分
Longyan 龍岩 (Tung, 1959, pp. 798, 802, 808, 910, 911, 912, 915)
A C
tsi45 ts’eɂ55 pun55 gua tsit pun (書冊分我蜀本, give me a book) pun55 laŋ21 nie52 k’i (分農扌弱去, someone was caught)
214 Cantonese give and Double-Object Construction
IV. The Gan dialects
(1) The double-object verb 把 or 擺
(a) The Changqing sub-group A Nanchang 南昌 (Wei & Chen, 1998, C p. 90)
ko4 pı̵ n213 ɕy42 paɂ5 tεɂ5 n̩213 (箇本書paɂ5得你, give you that book) t’iεn21 ɕiεn45 paɂ5 fʊŋ21 ts’ui t’ɵn21 liεu213 (電線paɂ5風吹斷了, the electric wire was broken by the wind)
(b) The Fuguang sub-group A Dongxiang 東鄉 (L. Liu, 1999, pp. 462, B 633) CIT223
Linchuan 臨川 (L. Liu, 1999, pp. 462, 633)
A B
Nanfeng 南豐 (L. Liu, A 1999, pp. 462, 633) B Yihuang 宜黃 (L. Liu, A 1999, pp. 462, 633) B A Lichuan 黎川 (S. Yan, 1993, p. 80) B CIT223
pai353 (擺, to give) pai353 (擺, as 給 in 借給他) pai35 (擺, to give) pai35 (擺, as 給 in 借給他) pai11 (擺, to give) pai11 (擺, as 給 in 借給他) pai232 (擺, to give) pai232 (擺, as 給 in 借給他) pai44 pεn44 ɕiε22 pai44 ŋo44 (把本書把我, give me a book) nε44 ua13 pai44 tɕiε44 hiaŋ22 (汝話把渠聽, you told him/her)
(c) The Jicha sub-group
Jishui 吉水 (R. L. Li & A Chang, 1992, pp. 444, C 451, 452, 454) CIT207
pa31 pən31 fʉ35 tau33 ŋͻ25 (把本書到我, give me a book) pa31 kε25 ts’ai25 tau25 li0 (把渠猜到哩, it was correctly guessed by him/her)
(d) The Yingyi sub-group Poyang 鄱陽 (L. Liu, 1999, pp. 462, 633)
A B
Leping 樂平 (L. Liu, A 1999, pp. 462, 633) B (e) The Yiliu sub-group Pingxiang 萍鄉 (L. Liu, 1999, pp. 462, 633)
Liuyang 瀏陽 (Xia, 1998, pp. 196, 242, 274)
A B C A B
C D
pa42 (把, to give) pa42 (把, as 給 in 借給他) pa213 (把, to give) pa213 (把, as 給 in 借給他) pa35 (把, to give) pa35 (把, as 給 in 借給他) pa35 (把, as 被) pa24 pən24 ɕy33 ŋo11 (把本書我, give me a book) fan11 tan33 pa24 cin33 ɲin55 fən33 k’ai33 i44 ɕia33 pa24 lin55 ʂa11 (范丹 把金銀分開一些把鄰舍, Fan Dan divided the gold and silver to the neighbor) t’a33 ɲia24 ciau24 ŋa44 liau24 i44 c’iau24 (他惹狗嚙了一口, he was bitten by a dog) pa24 t’a33 ci44 sɿ11 ta0 (把他急死咑, someone made him so anxious)
Appendix 1. Survey of the syntactic functions of give 215
(2) The double-object verb 得 (a) Jiangxi province Chaling 茶陵 (R. L. Li & Chang, 1992, pp. 444, 451, 452, 454)
B C
ma53 pẽ53 ɕy353 te21 ŋo353 (□本書得我, give me a book) te21 tɕi53 ku53 tɐ33 lo0 (得渠估到了, it was correctly guessed by him/her)
A
45 44 xa11 to45 te33 ŋo33 liͻ44 ̃ ȶͻ̃ lo (還多得我兩張囉, still have to give me two more pieces) 33 24 45 24 11 11 tã13 te33 p’ie33 zẽ11 xã44 tso33 lͻ44 t’ͻ11 ̃ , ȶ’o sɿ sã sɿ sɿ ȵiæ̃ tɕ’iæ̃11 ke33 sɿ13 tͻ33 (但得別人喊作老唐,欲是三四十年前格事到, it was almost 30 or 40 years ago that I was addressed as Mr. Tang)
(b) Hunan province Changning 常寧 (Q. Wu, 1998, pp. 298, 301)
C
(3) The double-object verb 畀 or 拿 (a) The Changqing sub-group Xiushui 修水 (R. L. Li & Chang, 1992, pp. 444, 451, 454)
A C D
laɂ32 pɤn31 su24 tau5 ŋͻ31 (拿本書到我, give me a book) laɂ32 təɂ0 hε22 dz’ai24 tau5 təɂ0 (拿得渠猜到嘚, it was correctly guessed by him/her) laɂ32 təɂ0 ŋͻ31 hͻn45 it32 ha0 (拿得我看一下, let me have a look)
(b) The Fuguang sub-group Jianning 建寧 (R. L. Li & Chang, 1992, pp. 444, 451, 454)
A B C D
pei55 pən55 sə34 pei55 ŋa55 (畀本書畀阿, give me a book) pei55 pən55 sə34 pei55 ŋa55 (畀本書畀阿, give me a book) pei55 sə55 t’ai34 t’ͻk5 ki0 (畀使猜著□, it was correctly guessed by him/her) pei55 ŋa55 k’ͻn33 ha45 i0 (畀我看下兒, let me have a look)
V. The Xiang dialects
(1) The double-object verb 把
(a) The Changyi sub-group Changsha 長沙 (Y. M. Li, 1991, p. 533)
Yiyang 益陽 (Cui, 1998, pp. 284, 286, 287)
A B A B D
li42 pu24 pa42 põ55 ko55 kən33 tɕiẽ13 pa42 ŋo42 (你不把半個工錢把 我, you didn’t give me the salary) fən33 tiẽ42 xən13 li21 pa42 ŋo52 sa21 (分點紅利把我唦, share some profits with me) pa41 pən41 ɕy33 ŋo41 (把本書我, give me a book) ko33 ko33 liəu13 pa41 xa33 i55 pi55 liẽ13 (哥哥留把他一筆錢, his elder brother left a sum of money to him) 33 33 11 13 ni41 iəu11 pa41 xa33 ma11 tɕi41 tɕy13 sə̃55 ta0, ȵiͻ11 ̃ xa ɕiau a tɕ’i (你就把他罵幾句算噠,讓他消下氣, you just let him scold you so that he can be mollified)
216 Cantonese give and Double-Object Construction
(b) The Loushao sub-group Shaoyang 邵陽 (Z. Chu, 1998, p. 193)
Qiyang 祁陽 (W. Q. Li, 1998, pp. 120, 201)
A B
pa42 pən42 ɕy55 pa42 ŋo42 (把本書把我, give a book to me) suŋ24 niã42 pa42 tɕ’iã55 pa42 ni42 mən0 (送兩把槍把你們, give you two guns)
A B C
pa53 pen53 ɕy55 pa0 ŋo53 (把本書把我, give me a book) suŋ35 pen53 ɕy55 pa0 ȵi53 (送本書把你, give you a book) ŋo53 ʑaŋ35 pa53 tʃʅ53 ta53 liau53 i33 ts’an55 (我讓把己打了一餐, I was beaten by him/her)
(c) The Jipu sub-group Xupu 漵浦 (K. He, 1999, pp. 239, 255)
A C D
pɒ23 ŋʊ23 kʊ35 pə̃23 ʂʅ44 (把我過本書, give me a book) uɑ̃13 kei23 ts’ʊ35 zə̃13 sɒ13 kua0 liɑʌ0 (黃狗著人殺呱了, the dog was killed by someone) pɒ23 ȵi23 uei13 lε̃ 13 liɑʌ0 (把你為難了, (something) made you embarrassed)
(2) The double-object verb 得 The Changyi sub-group Hengyang 衡陽 (Y. M. Li, 1986, pp. 425, 475, 476)
A C
D Anren 安仁 (M. Chen, 1995, pp. 172, 173, 241, 245)
A B C D
ni33 min11 ɕi11 mai33 lai11, ŋo33 tɕiu213 te2 tɕien11 ni33 (你明日買來, 我就得錢你, When you come tomorrow, I will give you money) tai213 ɕy213 ko2 ken35 tɕin33 tɕin33 ko2 tsua35 tɕy213 ni11 pa35, pu2 te2 ɕy33 pa33 ni11 pa35 ts’əŋ35 tsəu33 (大樹咯根緊緊咯抓住泥巴,不得 水把泥巴沖走, the root of the tree held the soil tightly so that it would not be washed away) tɕi11 nin11 ko2 fa213, te2 ɕian35 li33 ɕin11 tian35 tau35 ta2 (其伲咯話,得 鄉里人聽到噠, s/he let the villagers hear his/her conversation )
tʃi313 pu313 te313 n̩31, n̩31 a51 u35 pan31 fa313 (佢不得你, 你也無辦法, you can do nothing if s/he does not give you) ku31 si44 tĩ35 kͻ44 te313 n̩31, n̩31 k’o51 tsu31 te313 ke313 (固些田交得你, 你可做得, can you manage them if I pass those paddy fields to you?) n̩31 te313 po35 lioŋ35 ha313 tͻ31 ta313 (你得婆娘嚇到噠?, were you scared by the woman?) le31 tʃa313 i313 tsɿ51 n̩31 te313 ŋo51 k’e31 (那只日子你得我去, you have to let me go on that day)
Appendix 1. Survey of the syntactic functions of give 217
VI. The Wu dialects
(1) The double-object verb 撥, 不 or 八
(a) The Taihu sub-group (i) Jiangsu province
Shanghai 上海 (B. Xu & Tang, 1988, pp. 487, 480)
A B C
D Nanhui Zhoupu A 南匯周浦 (N. Qian, B 1992, pp. 1048, C 1049, 1050, 1051)
pəɂ55 tsã53 tsɿ34 dɤ23 pəɂ55 ŋu53 (撥張紙頭撥我, give me a piece of paper) tã34 tsᴀɂ5 di23 ɦuᴀ23 pəɂ55 noŋ23 (打隻電話撥儂, give you a call) vəɂ23 iͻ34 uᴀ53 lᴀ53 uᴀ53 lᴀ53 tɕiͻ34, pəɂ55 zən23 tɕiᴀ53 t’in53 tɕi34 nᴇ23 t’in53 (勿要哇啦哇啦叫,撥人家聽見難聽口伐, do not scream, otherwise, it will be embarrassing if other people hear it) ŋu53 pəɂ55 i53 tɕ’iιɂ55 dɑ̃23 (我撥伊吃糖, I let him/her eat candies) ɦu ɂbəɂ noŋ ɂbəŋ ɕy (吾撥儂本書, I gave you a book) ɦu ɕiɑ ɕιŋ bəɂ ɦi (吾寫信撥伊, I wrote a letter to him) ɦi ɂbəɂ ɦu ɂbᴀ̃ zɿ liᴀ̃ tɕi (伊撥吾打是兩記, s/he was beaten by me for a couple of times)
Baoshan Luodian A 寶山羅店 (N. Qian, B 1992, pp. 1048, C 1049, 1050, 1051)
ŋ̩ pəɂ noŋ ɂiιɂ pε̃ n sɿ (吾撥儂一本書, I gave you a book) ŋ̍ siɑ sιn pɐɂ ɦi (吾寫信撥伊, I wrote a letter to him) ɂi pəɂ ŋ̍ tã tɐɂ liã tɕi (伊撥吾打得兩記, s/he was beaten by me for a couple of times)
Baoshan Shuangcao A Dun 寶山霜草 B 墩 (N. Qian, 1992, C pp. 1048, 1049, 1050, 1051)
n̩ pəɂ noŋ ɂiιɂ pε̃ sɿ (吾撥儂一本書, I gave you a book) n̩ siɑ sι̃ pəɂ ɦi (吾寫信撥伊, I wrote a letter to him) ɂi pəɂ n̩ tã dəɂ liã tɕi (伊撥吾打特兩記, s/he was beaten by me for a couple of times)
A B C
nu pəɂ zͻ ɂiιɂ pəɲ ɕy (奴撥造一本書, I gave you a book) nu kəɂ ɦi ɕiɑ ɕιŋ (奴克伊寫信, I wrote a letter to him) ɦi pəɂ nu k’ͻ zɿ liε̃ tɕi (伊撥我敲是兩記, s/he was beaten by me for a couple of times)
A B C
ŋəu pəɂ nən ɂiιɂ pən sʮ (我撥能一本書, I gave you a book) ŋəu t’əɂ ɦi siɑ sin (我脫伊寫信, I wrote a letter to him) ɦi pəɂ ŋəu k’ͻ zɿ liã tɕi (伊撥我敲是兩記, s/he was beaten by me for a couple of times)
A B C
ŋա pᴇɂ nε̃ ɲ ɂιɂ pε̃ ɲ ʂʯ (我撥能一本書, I gave you a book) ŋա siɑ sι̃ ɲ pᴇɂ gᴇ (我寫撥佢, I wrote a letter to him) gᴇ pᴇɂ ŋա tᴀ̃ tsᴇɂ liᴀ̃ tɕi (佢撥我打則兩記, s/he was beaten by me for a couple of times)
A B C
ŋɜu pəɂ pən sʮ nᴇ (我撥書倷, I gave you a book) ŋɜu siɒ siιn pəɂ ɂlij (我寫信撥俚, I wrote a letter to him) ɂlij pəɂ ŋɜu k’æ/ tᴀ̃ tsɿ liᴀ̃ tɕij (俚撥我敲/打仔兩記, s/he was beaten by me for a couple of times)
Songjiang 松江 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051) Kunshan 昆山 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Changshou 常熟 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051) Suzhou 蘇州 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
218 Cantonese give and Double-Object Construction Wujiang Lili 吳江黎 A 里 (N. Qian, 1992, B pp. 1048, 1049, C 1050, 1051)
ŋ̍ pəɂ nɒ ɂiəɂ pəɲ sʮ (吾撥那一本書, I gave you a book) ŋ̍ siɒ pəɂ ɦij ɂiəɂ foŋ siəɲ (吾寫撥伊一封信, I wrote a letter to him) ɂij pəɂ ləɂ/zɿ ŋ̍ k’ᴀͻ zɿ liã tɕij (伊撥勒/是吾敲是兩記, s/he was beaten by me for a couple of times)
Wujiang Shengze A 吳江盛澤 (N. Qian, B 1992, pp. 1048, C 1049, 1050, 1051)
ɦu pəɂ nəɂ ɂiɐɂ pəɲ sʮ (吾撥吶一本書, I gave you a book) ɦu ɕiɑ ɕιɲ pəɂ ɂij (吾寫信撥伊, I wrote a letter to him) ɂij pəɂ ɦu k’ᴀɑ t’əɂ liæ̃ tɕ’ij (伊撥吾敲脫兩記, s/he was beaten by me for a couple of times)
Wuxi 無錫 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Changzhou 常州 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051) Yixing 宜興 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Liyang 溧陽 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Jiangyin 江陰 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Jintan 金壇 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
A B C
ŋʌɤ pəɂ ȵi ɂiəɂ pən sʮ (我撥你一本書, I gave you a book) ŋʌɤ təɂ dʌɤ siɑ sin (我得他寫信, I wrote a letter to him) dʌɤ pəɂ ŋʌɤ tã ləɂ liã tɕi (他撥我打勒兩記, s/he was beaten by me for a couple of times)
A B C
ŋʌա pəɂ ȵij ɂiιɂ pəɲ ɕʮ (我撥你一本書, I gave you a book) ŋʌա ɕiɑ ɕiɲ pəɂ dɑ (我寫信撥他, I wrote a letter to him) dɑ pəɂ ŋʌա tᴀɲ tsəɂ liᴀɲ tɕij (他撥我打則兩記, s/he was beaten by me for a couple of times)
A B C
ŋu pəɂ pəŋ ɕy ȵij (我撥本書你, I gave you a book) ŋu ɕiᴀ ɕiŋ pəɂ t’o (我寫信撥他, I wrote a letter to him) t’o pəɂ ŋu tᴀŋ tɑɤ liᴀŋ tɕij (他撥我打到兩記, s/he was beaten by me for a couple of times)
A B C
ŋʌա pəɂ ȵi ɂiιɂ pən ɕy (我撥你一本書, I gave you a book) ŋʌա ɕio ɕin pəɂ t’o (我寫信撥他, I wrote a letter to him) t’o pəɂ ŋʌա to tsəɂ lie tɕiz (他撥我打則兩記, s/he was beaten by me for a couple of times)
A C
ŋɜɤ pɑɂ ȵij ɂiɘɂ pᴇɲ ɕy (我八你一本書, I gave you a book) t’ɑ pɑɂ ŋɜɤ tɑ dɑɂ ɂliᴀŋ tɕij (他八我打著兩記, s/he was beaten by me for a couple of times)
A B C
o pɑ ieɂ pəŋ səu ni (我叭一本書你, I gave you a book) o ɕiɑ ɕiŋ pɑ t’ɑ (我寫信叭他, I wrote a letter to him) t’ɑ pei o tɑ ɦɑ liɑŋ xɑ tsɿ (他被我打啊兩下子, s/he was beaten by me for a couple of times)
(ii) Zhejiang province
Hangzhou 杭州 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051) Jiaxing 嘉興 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
A B C
ɂŋou pəɂ ni ɂiιɂ pən sʯ (我不你一本書, I gave you a book) ɂŋou ɕi ɕιn pəɂ t’ɑ (我寫信不他, I wrote a letter to him) t’ɑ pəɂ ŋou k’ͻ ləɂ ɂliᴀŋ tɕi (他不我敲勒兩記, s/he was beaten by me for a couple of times)
A B C
ŋ̍ pəɂ ne ɂiəɂ pən sɿ (吾撥內一本書, I gave you a book) ŋ̍ pəɂ ɂi ɕiɑ foŋ ɕin tɕ’i (吾撥伊寫封信去, I wrote a letter to him) ɂi nɑ pəɂ ŋ̍ k’ͻ ləɂ liᴀ̃ tɕi (伊奴撥吾敲勒兩記, s/he was beaten by me for a couple of times)
Appendix 1. Survey of the syntactic functions of give 219
Huzhou Shuanglin A 湖州雙林 (N. Qian, B 1992, pp. 1048, C 1049, 1050, 1051) Ningbo 寧波 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Shaoxing 紹興 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
Zhoushan 舟山 (Fang, 1993, pp. 78, 81, 82)
ɂŋ̍ pəɂ ne pən sɿ (吾撥倷本書, I gave you a book) ɂŋ̍ ɕiɑ ɕιn pəɂ dʑiz (吾寫信撥其, I wrote a letter to him) dʑiz pəɂ ŋ̍ tã dəɂ liã tɕieɂ (其撥吾打突兩級, s/he was beaten by me for a couple of times)
A B C
ŋo pɐɂ nəυ paŋ sʮ (我撥奴本書, I gave you a book) ŋo ɕia ɕιŋ pɐɂ dʑi (我寫信撥其, I wrote a letter to him) dʑi pɐɂ ŋo tã lɐɂ liã tɕi (其撥我打勒兩記, s/he was beaten by me for a couple of times)
A B C
ŋo pιɂ noɂ ɂιɂ pι̃ ɕy (我不你一本書, I gave you a book) ŋo ɕia ɕιŋ pιɂ ɦi (我寫信不伊, I wrote a letter to him) ɦi pəɂ ŋo k’ɑɒ ləɂ liaŋ tɕi (伊不我敲勒兩記, s/he was beaten by me for a couple of times)
A B
pɐɂ55–33 ŋau24 iəɂ55–33 pɐŋ24–33 sʮ53 (撥伢一本書, give me a book) tɕia44 tiŋ35–33 ɦy24–13 sε44–0 pɐɂ55–33 ŋau24 (借頂雨傘撥伢, lend me an umbrella) dzo22 pai53–44 pɐɂ55–33 dʑi22–31 k’ͻ53–33 sai44–53 lai22–0 (茶杯撥其敲碎 唻, the cup was broken by him/her) pɐɂ55–33 ŋau24–13 ta44 ta44 ts’oɂ55–0 ɦiu13 le22 (撥伢帶帶出就來, let him/her bring it out)
C D (b) The Taizhou sub-group Tiantai 天台 (Z. Dai, 2003, p. 155)
Huangyan 黃岩 (N. Qian, 1992, pp. 1048, 1049, 1050, 1051)
A
B
A B C
na214 niaŋ24–41 pøɂ5 iιɂ5–1 paɂ5 k’uei55–11 æɂ5–1 ɕy55–33 diᴇ24 pøɂ5–1 ɕiau325–31 niŋ24–51 (乃娘撥一百塊壓歲鈿撥小人, your mother gave 100-dollar red-packetmoney to the children) na214 niaŋ24–41 pøɂ5 iιɂ5–1 paɂ5 k’uei55–11 æɂ5–1 ɕy55–33 diᴇ24 pøɂ5–1 ɕiau325–31 niŋ24–51 (乃娘撥一百塊壓歲鈿撥小人, your mother gave 100-dollar red-packet money to the children) ŋo pɐɂ n̩ ɂieɂ pəŋ ɕy (我不你一本書, I gave you a book) ŋo ɕiᴀ ɕiιŋ pɐɂ ge (我寫信不佢, I wrote a letter to him) ge pɐɂ ŋo ɂliã tɕij tã ɦɒ (佢撥我兩記打號, s/he was beaten by me for a couple of times)
VII. The Mandarin dialects
(1) The double-object verb 給
(a) The Northern sub-group Beijing 北京 (Hou, A 1998, pp. 35, 40, 89) C
D
kei214–35 uo214–35 pər214 ʂu55 (給我本兒書, give me a book) tuŋ55 ɕi2 ʐɑŋ51 lɑu214–21 tʂɑŋ55 kei214–21 ʂɑu55 tɕ’y2 lə1 (東西讓老張 給捎去了, the thing was taken away by Zhang) ʂei35 nəŋ35 ɕian55 tɕiɑu51 tʂei51 kə1 tsou214 tɑur51 tə1 pai214 t’a55 tə2 xou51 ta51 i55 t’uo55 ɕia2 lai1 (誰能先叫這個走道兒的把他的厚大 衣脫下來? who can first make this old man on the road take off his thick coat?)
220 Cantonese give and Double-Object Construction Changli 昌黎 (Hebeisheng, 1984, pp. 124, 132, 134, 135, 272)
A B C D
Zhengzhou 鄭州 (Lu et al., 1998, pp. 44, 46, 110)
A B C
Luoyang 洛陽 (W. He, 1993, pp. 110, 111, 116)
A B D
Jinxiang 金鄉 (F. Ma, 2000, pp. 144, 211)
A C D
Laizhou 萊州 (Z. Qian, 2005, pp. 226, 227, 324, 372, 373)
A
B
C Wenshui 文水 (Hu, 1984, pp. 65, 78)
A B D
Lingchuan 陵川 (M. Jin, 1983, pp. 23, 24)
A B C
ni213 iɑu53 tuo44 ʂɑu0 tɕ’ian24, uo213 tʂou53 kei213 tuo44 ʂɑu0 tɕ’ia24 (你要多少錢,我就給多少錢, I can give you whatever you want) tɑ53 ti0 tɕiɑ53 kei0 i24 kə0 tʂuo53 kuaա44 ti0 (大的嫁給一個做官的, the eldest was married to an official) ts’uɑŋ24 xuo0 ʐɑŋ55/tɕiɑu55/tɕiɑu55 fəŋ24 kuɑ42 k’ai42 liə0 (窗戶 讓/教/叫風刮開咧, the window was blown open by the wind) xai24 tei213 ʐɑŋ53 ʂan44 ku0 ie0 i44 tɕ’iəա24 lai24 tʂ’ʅ23 fan53 (還得讓三 姑爺一齊兒來吃飯, (someone has to) let the husband of the third daughter come to have dinner) kei24 uo53 i24 pər53 ʂu24 (給我一本書, give me a book) kei24 no312–31 tuŋ24.si na42 kei24 uo53 (給那個東西拿給我, take that thing and give it to me) tɕiau312 kou53 iau53 i24 k’ou53 (叫狗咬一口, (someone) was bitten by a dog) kuə33 pəա53 ʂu33 (給我本兒書, give me a book) ni53 ku33 na412 kuə33 i33 ɕiɐա412 (你箍那給我一下兒, you took that hoop to me) t’a53 tsəu412 ku33 t’a53 tiε33 ts’iaŋ53 tͻ0 ti421–13 in412 tsə0 li0 (他就給他 爹藏到地窨子裏, he then hid his father in the basement) ȵi55–42 sͻ55 kei312–31 uə0 la0 pε0 (你少給我啦唄, you gave too little to me) lͻ55 tsei42 tsɿ55 tɕiͻ312 t’a213 mε312 la (老宅子叫他賣啦, the old apartment was sold by him) ɣã55 ȵiaŋ213 tɕiͻ312 uə55 tsou55 tɕ’in42 tɕ’in42 tɕ’y312 (俺娘叫我走親 親去, my mother let me go)
kə55 kuə213 lε0 tʂəŋ213 su42–55 ləu0 kei55 tʂͻ42 k’uɑŋ213–45 iẽ213 tʂ’ʅ55 (擱鍋來蒸熟了給趙匡胤吃, take a pot to steam it for Zhao Kuangyin to eat) iəu55 tsəŋ42–55 mə0 kə0 ɕiɑ55 xuɑr42, uə55 ʂuə55–45 kei55 nẽ55 t’iŋ213 t’iŋ0 (有這麼個瞎話兒,我說給恁聽聽, there is a yatter; I am telling you) mẽ213 kei55 fəŋ213 ts’uei213 k’ε213 liͻ55 (門給風吹開了, the door was blown open by the wind) kue423–42 ŋəi423 iəɂ2–3 pəŋ423–21 su22 (給我一本書, give me a book) n̩423 kəi24 kue423–21 t’a22 (你告給他, you told him) xəi423–42 ts’uəɂ2–23 iəɂ2 kəɂ312 taɂ2 lai22 kue423 xe22–33 məɂ2 k’ən35 (荷 出一圪瘩來給孩們看, show the scar to the children)
kɤɂ32 uʌ213 ieɂ32 pə̃213–211 ʂu33 (給我一本書, give me a book) ni213–211 ʂuʌɂ32 kɤɂ32–35 t’ᴀ0 (你說給他, you told him) yə̃53 tʂ’ᴀi213–211 tɕiɑo22 fəŋ33 lᴀi35 tʂ’uei33 p’ɑo53 lᴀ35 (云彩叫風來吹 跑了, the cloud was blown away by the wind)
Appendix 1. Survey of the syntactic functions of give 221
(b) The Southwestern sub-group Zhongxiang 鍾祥 (Chao, 1939, pp. 141, 158, 160)
A B D
Chengdu 成都 (Y. Zhang, et al., 2001, pp. 321, 322)
A B C
kə53 (給, give) ni53 k’o53 i53 tʂ’ui24 kə53 o53 t’in214 t’in0 k’an214 (你可以吹給我聽聽 看, you can play it for me) kə53 o53 nai31 suan213 i31 suan214 pa0 (給我來算一算罷, let me figure it out) iau ʂï ʐaŋ ni ɕien nai pa, iu t’ai xau sï ta ni (要是讓你先來罷, 又太好事達你, if i let you go first, it may give you an advantage) 這個東西, 你給給他嘛 (as for this thing, you give it to him) 這個東西,你給給他嘛 (as for this thing, you give it to him) 豆芽給你炒鹹了 (the bean sprout was made too salty by you)
(2) The double-object verb 把 The Jianghuai sub-group Hangzhou 杭州 (Simmons, 1992, pp. 158, 159, 161)
A B C
Huanggang 黃崗 (H. Wang, 2004, pp. 177, 217)
A B C
pᴀ52 t’ᴀ33 iəɂ21-tin52 mͻ33 tæ45 (把他一頂帽戴, give him a cap) ŋo52 soŋ45 iəɂ21 pən52 sʮ33 pᴀ52 t’ᴀ33 (我送一本書得他, I gave him a book) ŋo52 pᴀ52 t’ᴀ33 tᴀ52 liͻ52 iəɂ21 tən45 (我把他打了一頓, I was beaten by him) 我不把錢得你 (I did not give you money) 你送一本書得他 (you gave him a book) 二哥把魚吃了 (the second brother was eaten by a fish)
VIII. Jiushi zhenghua 舊時正話 in Dianbai 電白 Dianbai zhenghua A 電白正話 (Y. Chen, B 2006, pp. 27, 89, 95) C
pa31 pɐn31 si33 ŋo31 (把本書我, give me a book) ŋo31 kaŋ31 ko55 pa33 t’a33 t’iŋ33 (我講過把他聽, I told him) won31 pa31 t’a33 ta31 lan55 e33 (碗把他打爛誒, the bowl was broken by him)
IX. The Jun dialects 軍話 (i) Guangdong province Huizhou Pinghai A 惠州平海 (X. Q. Qiu, C 2005, pp. 47, 48)
pi55 a33 pən31 ʃy33 ŋo31 (俾啊本書我, give me a book) t’a33 pi55 kəu31 ŋau31 tau31 (他俾狗咬到, s/he was bitten by a dog)
Lufeng Kanshitan 陸豐坎石潭 (X. Q. Qiu, 2005, pp. 47, 48)
pa51 a55 pan51 ʃu33 pa51 t’a33 (把一本書把他, give him/her a book) pa51 a55 pan51 ʃu33 pa51 t’a33 (把一本書把他, give him/her a book) t’a33 pa51 kou51 ŋau51 a33 a55 k’ou51 (他把狗咬阿一口, s/he was bitten by a dog)
A B C
222 Cantonese give and Double-Object Construction
Lufeng Qingtang A 陸豐青塘 (X. Q. Qiu, C 2005, pp. 47, 48)
pan33 a55 pan13 ʃi33 ki33 (分一本書佢, give him/her a book to) ki33 pan33 kou13 ŋau13 tau13 (佢分狗咬到, s/he was bitten by a dog)
(ii) Guangxi
Yongan 永安 (X. Q. A Qiu, 2005, pp. 47, 48) C
k’i35 ia55 puon55 ʃɿ33 vai31 (乞一本書我, give me a book) i33 k’i55 a55 k’en31 ka24 a31 ki33 na35 (伊乞阿犬咬啊□哪, s/he was bitten by a dog)
(iii) Hainan province
A Basuo 八所 (X. Q. Qiu, 2005, pp. 47, 48) B Zhonghe 中和 (X. Q. Qiu, 2005, pp. 47, 48)
A B C
Yacheng 崖城 (X. Q. A Qiu, 2005, pp. 47, C 48)
kei51 ʒi33 pən51 ʃy33 kei51 ŋo51 (給一本書給我, give me a book) kei51 ʒi33 pən51 ʃy33 kei51 ŋo51 (給一本書給我, give me a book) kei51 ʒi11 pɐn51 ʃi33 kei51 ŋo51 (給一本書給我, give me a book) kei51 ʒi11 pɐn51 ʃi33 kei51 ŋo51 (給一本書給我, give me a book) a33 ɂbei55 kɐu51 ʒau51 tʃo11 (□俾狗咬著, s/he was bitten by a dog) kei11 ŋo11 ʒi214 pɐn11 ʃi33 (給我一本書, give me a book) he33 kei11 kɐu11 ʒau11 liau11 ʒi214 kɐu11 (□給狗咬了一口, s/he was bitten by a dog)
X. The patois土話273 (a) Lianzhou patois (連州土話) in Northern Guangdong Lianzhou 連州 (Chang & Wan, 2004, pp. 355, 358, 360, 366)
Xingzi 星子 (Chang & Wan, 2004, pp. 355, 358, 360, 366) Baoan 保安 (Chang & Wan, 2004, pp. 355, 358, 360, 366)
A B C D A B C D A B C D
piᴀu11 pᴀn53 sy22 ki24 (俵本書渠, give him/her a book) piᴀu11 俵 in 借給我 (lend (it) to me) piᴀu11 ki24 ku53 tse11 lᴀ (俵渠估中了, it was correctly guessed by him/her) piᴀu11 ŋᴀu24 mei33 hᴀu33 (俵我望下, let me have a look) pieu53 ɑ55 paĩ55 ʃy33 hɑ33 (俵一本書□, give him/her a book) pieu53 俵 in 借給我 (lend (it) to me)273 pieu53 hɑ55 ku55 tʃʌŋ53 lə0 (俵□估中了, it was correctly guessed by him/her) pieu53 hɑŋ21 miaŋ21 ɑ33 miaŋ21 (俵幸望一望, let me have a look) pεu53 i33 pʌn45 ʃy33 ki33 (俵一本書渠, give him/her a book) pεu53 俵 in 借給我 (lend (it) to me) pεu53 ki33 ku45 tʃͻŋ53 lʌ (俵渠估中了, it was correctly guessed by him/her) pεu53 ha33 mεi11 hͻu11 (俵□望下, let me have a look)
273. Chang and Wan (2004) did not give a complete sentence for the use of GIVE as an IO marker. They only have an entry (借)給(他) (p. 355).
Appendix 1. Survey of the syntactic functions of give 223
Xi’an 西岸 (Chang & Wan, 2004, pp. 355, 358, 360, 366)
A B C
Fengyang 豐陽 (Chang & Wan, 2004, pp. 355, 358, 360, 366)
A B C
D
D
tɐi11 əi11 pɐn22 ʃy55 ke31 (得一本書渠, give him/her a book) tɐi11 得 in 借給我 (lend (it) to me) tɐi11 ke22 tʃ’o31 təա53 lɐ (得渠猜對了, it was correctly guessed by him/her) tɐi11 hɐ22 me22 me22 (得□望望, let me have a look) nͻŋ21 εi34 pεn55 ʃi34 kεi34 (□渠一本書, give him/her a book) nͻŋ21 as 給 in 借給我 (lend (it) to me) nͻŋ21 kεi34 ku55 tʃͻŋ53 liε0 (□渠估中了, it was correctly guessed by him/her) nͻŋ21 ha21 miε21 hɵ21 tsai55 (□□望下仔, let me have a look)
(b) Lechang patois (樂昌土話) in Northern Guangdong Changlai 長來 (Chang, 2000, pp. 352, 353, 358)
Beixiang 北鄉 (Chang, 2000, pp. 352, 353, 358) Huangpu 黃圃 (Chang, 2000, pp. 352, 353, 358) Guitang 皈塘 (Chang, 2000, pp. 352, 353, 358) Sanxi 三溪 (Chang, 2000, pp. 352, 353, 358)
A C D A C D A B C D A B C D A B C D
a12 ji11 pεŋ24 ʃy31 ki33 (□一本書渠, give him/her a book) pæi33 i33 ku24 tʃœŋ33 lei0 (被渠估中了, it was correctly guessed by him/her) pei33 ŋu33 t’i24 ji11 t’i24 (俾我睇一睇, let me have a look) a212 i55 pou55 ʃy212 ki33 (□一本書渠, give him/her a book) a212 ki33 ku55 tʃͻi33 lə0 (□渠估中了, it was correctly guessed by him/her) a212 ŋa33 t’εi55 t’εi55 (□我睇睇, let me have a look) no24 εi33 pιŋ42 ʃy24 kεi33 (拿一本書渠, give him/her a book) kaŋ42 kəw21 no24 kεi33 t’aĩ21 (講過拿佢聽, tell him/her) no24 kεi33 ku42 tʃoŋ21 tεi0 (拿渠估中了, it was correctly guessed by him/her no24 ŋɤա33 k’ɤŋ21 εi33 ho21 (拿我看一下, let me have a look) a13 kei33 ei33 pεŋ33 ʃy13 (□渠一本書, give him/her a book) kͻu33 a13 kei33 t’ai21 (講□渠聽, tell him/her) a13 kei33 kuə33 tʃεu21 kei0 (□渠估中了, it was correctly guessed by him/her) a13 ŋͻ33 mͻu21 ei31 hͻ21 (□我望一下, let me have a look) piͻu52 kei33 ei52 peŋ35 ʃy31 (□渠一本書, give him/her a book) ua33 ko52 piͻu52 kei33 t’ei52 (話過□渠聽, tell him/her) piͻu52 kei33 ku35 tʃ’o33 tei0 (□渠估著了, it was correctly guessed by him/her) piͻu52 ŋa33 maŋ33 hͻ33 (□我望下, let me have a look)
(c) Hunan patois (湖南土話) A Dongan 東安 (Bao, 1998, pp. 230, 254) B
C
tso42 tu55 ɕyẽ33 xo33 du13 me35 tai35 (摘朵鮮花□妹戴, (someone) picked a flower and gave it to the girl to put on) ȵi35 ɣau24 k’uai35 tɕ’ie33 li55 tɕin35, tɕio35 du35 ŋo55 ta55 luei13 yŋ24 i42 ɣo24 (你有塊千里鏡,借□我打雷用一下, As you have a thousand-mile mirror, lend me to use for a while) e33 tɕio42 dzo13 pei33 du13 ŋ35 ta55 kua0 li0 (□隻茶杯□他打呱哩, that cup was broken by him)
224 Cantonese give and Double-Object Construction
Ningyuan寧遠 (X. Zhang, 1999, pp. 267, 271, 296)
A B
C Yizhang 宜章 (R. Shen, 1999, p. 214)
A B D
ie53 piaŋ33 ɕie435–53 io53 (與本書我, give a book to me) xəu53 xe33 va21 sɿ21 tsaŋ33 sɿ53 liəu213 tie0 ie53 ɕi53 nͻ213 tə0 i21 (有好 物事總是留倒與小兒子食, any good food was left for the young child to eat) səu33 ie53 pie435–53 mie213 tɕ’ie33 kə213 xə435–53 kua0 lie0 (手與斑毛 草割開了, the hand was cut by the grass) no44 (ei33) pεi53 ɕy13 no44 ŋəա53 (拿(一)本書拿我, give me a book) tɕye21 sei33 k’ua21 tɕ’ie44 no44 ŋəա53 (借十塊錢拿我, lend me 10 dollars) sεi21 kəu33 kei0 tsɿ53 hau31 to53 tie31 ts’aŋ44 p’u21, no44 sεi21 ai13 kei0 ɕie21 kau21 (姓谷格只好打點床舖,拿姓歐格睡覺, What Gu could do was to make the bed for Ou to sleep)
XI. Non-Sinitic languages (a) The Tibeto-Burman group (i) The Loloish branch
Bisu 畢蘇語 (S. Xu, A 2001, pp. 83, 87) B Lahu 拉祜語 (Matisoff, 1991) Pulai 仆拉話 (C. Wang, 2004)
D
ʑa31-maŋ31 ga33 na33 ni31 sɿ31 pi31 (The elderly person gave me two pieces of fruit) ga33 tɕ’i31 pi31 ʑoŋ33 na33 sum55-pak31 kin55 ko33 (I lent them 300 catties of grain) ʑaŋ33 na33 p’ja31 le33 pi31 a55 (let him crawl up there)
A B D
phu pî (to give money) yͻ̃ àɂ chi qhe qôɂ pι̃ (told that to him) ce pι̃ (to make it fall)
A
kɤ33 ʑi21-dʑι22 ta21 vι22 vι22 ŋa21 kɤ22 gι22 a22 (他給我一挑水, he gave me a bucket of water ) ŋa22 xɤ21 nɤ22 lι33 gι33 (我拿給你, I passed it to you) ŋa2 ts’ɤ22-ε33 kι33 sι31 ta33 a33 (我被人拉著了, I was held back by someone) ŋa22 lɤ21 lι22 sͻ55 gι22 lɤ33 (讓我來學吧, let me learn it)
B C Kazhuo 卡卓語 (Mu, 2003, pp. 96, 114, 116, 120, 333)
D A B C D
Lisu 傈僳語 (Bradley, 2003, pp. 222–235)
A D
ŋɑ33 tɕ’i24 si24 k’ua31 nε33 kա31 (我給你三塊錢, I gave you three dollars) ʑi33 m31 tε31 mɑ323 ŋɑ33 ŋɑ55 kա31 (他借一匹馬給我, he lent me a horse) ʑi33 ŋɑ33 kε33 tɕ’i55 sɑ24 (他被我踢傷, he was injured by my kick) tɕi33 wɑ53 mε33 fv33 mɑ33 mε33 k’v33 zɿ55 fv33 mɑ33 ʑi33 ko55 ȵɑ31 t’ε323 kա31 (金豬使湖水豐厚, the golden pig made the lake full) ŋwɑ33 lɑ33 su44 tε55 t’o21ɣա21 gա21 o44 (I gave a book to the person who was coming) gա21 dzɑ21 (meaing: cause to eat)
Appendix 1. Survey of the syntactic functions of give 225
(ii) The Loloish branch Langsu 浪速語 (Q. Dai, 2005, pp. 41, 50)
A B
31 55 31/51 pjik55/31 nε55 (你要給哪位, to whom do you nͻ31 ̃ k’ə jauk ʒε want to give?) 31 31 55 55 31 nͻ31 ̃ ŋͻ ʒε mͻ pjik laɂ (你教給我, you taught me)
(b) The Kam-Tai family274 (i) The Zhuang-Dai branch A Boai 剝隘 (F. Li, 1988, pp. 12, 18, 20) B
C
D Dai 傣語 (G. Yang, 2006)
A B C D
Wuming 武鳴 (F. Li, 1956, pp. 25, 30, 41)
A B C
D
haï44 tee24 šiin24 šaaŋ55 ŋan55 (n)ee31 taau22 pai24 šiiŋ33 mee31 čee22 tee24 (給他一千兩銀子,回去養他老母親, give him one thousald taels so that he can go home and provide living to his old mother) pin22 čhεε55 ɂook22 ɂan31 fan24 hïn55 haï44 tee24, čoo24 maai22 tee24 kaa44 taai24 juu33 maau22 (並且托一個夢給他,謝謝他殺死姦夫, then (someone) gave him a dream and thanked him for killing his adulterer ) laŋ24 poo24 mii5 tuu55 juuŋ55 ee31 ŋaai31 kuk55 hap44 taai24 (背後山 有隻羊捱虎咬死, a sheep was bitten to death by a tiger in the back of the mountain) tuu55 juuŋ55 nii44 koo44 šin22 šiiu55 haï4 mïŋ55 tïï55 taau22 pai24 kïn24 (這隻羊也就給你拿回去吃, take this sheep home to eat) me33 haա31 so35 kau33 pa:k11 ləŋ33 (媽媽給我一百塊, mother gave me 100 dollars) man55 ku31 xau31 pək11 haա31 kau33 ta:ŋ33 ləŋ33 (他借給我一籮谷 子, he lent me a basket of paddy) haա31 pən33 la11 ɂen31 (給人家罵, scolded by someone) lաt35 hau53 haա31 xau35 ka11 lͻn55 lai31 hau53 (夜深了, 讓他們去睡, it is very late; let them sleep) koŋ33 ta33 ɕi51 taŋ33 jiɑn21 ɂbɑu55 hɑա55 a55 (岳父就當然不給了, the father-in-law of course did not give) tan13 hi55 kiŋ33 ha13 lաk12 θau33 hɑա55 te33 liɑu51 (已經嫁子女給他 了, (someone) has married the daughter to him) tɑk21 lաk12 me13 ne55 wai31 ɕi24 ne55 θɑt12 tաk12 tu31 kuk45 kam31 hɑu55 ɂdɑա33 ɂdoŋ33 poi33 ɂi55 lo33 (那婦人韋氏的兒子呢實被 一隻老虎銜入山裏去了, the son of lady Wei was taken to the mountain by the tiger) ɕi51 hɑա55 te33 ɂɑu33 køn24 (給他取先, let him take it first)
274. The passive sentence is taken from J. Zhang (1980), p. 52.
226 Cantonese give and Double-Object Construction Longzhou 龍州 (F. Li, 1993, pp. 44, 46, 47, 70)
A B
C
D Lingao 臨高 (Liang & Zhang, 1997, pp. 98, 304, 305)
A B C
D
kau33 mi31 miŋ11 liŋ11 hա24 maա31 (我有命令給你, I have an order to give you) mən31 tɕau11 kia:ŋ24, tɕuŋ31 tau31 ɂit5 ji11 kia:ŋ24 hա24 po11 kwa:n33 nai24 tiŋ11 a33 (他就說,從頭一二說給這個官聽了, he then told everything to the official) pin24 tu33 ma33 lə:ŋ3 ne55 tɕau11 hա24 ɂo33-ko55 mən31 fa:t31 ha:i33 pai33 a55 (於是隻黃狗呢,就給他哥打死去了, the yellow dog was therefore beaten to death by his brother) naա33 kam11 ŋo33 ha:k5 mi31 fa:p5 tɕi24 ɂo:k5 ma31 hա24 ni33 mi55 ɕai24 k’o24 ti33 (早晚我自己有法子出來給你不用窮的, one day I will come up with an idea which will make you rich) hau55 se213 sek33 bun33 kə55 (我給本書他, I gave him a book) ŋan11 haŋ11 nai33 vͻ55 bek33 ŋͻn55 se213 kə55 (銀行借兩百銀給他, the bank lent him 200 dollars) hau55 lai33 xu11 nə33 bi55 van11 xi213 man55, xim11 kə33 ja33 bi55 van11 ka33 dai213 vͻi33 (我有一個弟弟被人家欺負,弟媳也被人家殺死 了, I had a younger brother who was bullied by others. His wife was also killed by others ) tan55 ti11 lan55 ho55 en213 fuŋ213 lua55 jͻu33, mən55 se213 lua55 kua33, fͻu33 kə55 lai33 miŋ55 liŋ55 dͻŋ213 nau11 tsun33 xͻi213 lua55 (但胡家仍 然封船不准過海, 要有他的命令才准開船, but the Wu family still kept the ships from crossing the sea unless he gave a command)
(ii) The Kam-Shui branch Mo 莫話 (T. Yang 2000, pp. 110, 134, 153, 284)
A B C D
A Shui 水話 (F. Li, 2005, pp. 124, 260, 172)274 B C D
pin31 toŋ11-si11 näi11 ha:i11 ŋ31 vi53-ljeu53 (這些東西完全給你, I gave you all these things) sin11 pa:i31 van11 zun53 to31 deu11 ha:i11 tə-sɿ33 (真的每天送一隻給 獅子, it was true that someone sent one to the lion every day ) ai33-ljak55 ŋäi31 din11 kuŋ31 tjap31 (小偷被很多人打, the thief was beaten by many people) sa31 ŋ31 kau35 du13 man11, ŋ31 ȶəu11 me31 ha:i11 man11 ljit31 la53 (如 果你看見他,你就別給他跑了, if you see him, don’t let him run away) nu45 taŋ45 ɂi33 hai11 ni54 sau33 ɤan11 (遞一椅給我主婦, pass a chiar to my wife) man55 ta45 loŋ45 hai11 qoŋ45 tsau11 ja54 (留下中間給公與婆, leave the middle to the old lady and gentleman) man11 ŋa:i54 la:u23-sɿ33 phai33-phin54 ljeu33 (他挨老師批評了, he was criticized by the teacher) ljaŋ55 mjen11 tau11 hai11 ho33 ɂnam33 ɂan45 (兩面到使我想老, the two sides made me think about the old times)
Appendix 1. Survey of the syntactic functions of give 227
Mulao 仫佬語 (J. Yue, 2004)
A B
Kam 侗語 (Long, 2003, pp. 100, 139, 145)
D A B C
A Maonan 毛難語 (Liang, 1980, pp. 59, B 79, 80) C
heu232 haŋ55 mͻ214 na:u55 pan55 lε232 (我給他一本書, I gave him a book) heu232 haŋ55 na:u55 pan55 lε232 haŋ55 mͻ214 (我給一本書給他, I gave him a book) Jͻ:ŋ232 haŋ55 mͻ214 p’a:u55 ljeu0 (別讓它跑了, don’t let it run away) mau33 sai35 jau212 ta33 (他給了我, he gave (something) to me) wən11 ʑaŋ44 ti35 sai11 noŋ53 (分一半梨子給弟, share half of the pears with the younger brother) ȶaŋ33 məm31 kit31 (被老虎咬, bitten by a tiger) ȶiu211 mɐi31 ȶa53 tɐu33 nɐm31 lau31 k’uk13 wai33 jaŋ31 (那木橋被大水沖壞了, that wooden bridge was damaged by the flood) man231 ɂna:k44 lε42 ɦe231 (他給我書, he gave me a book) man231 tsi51 zu:n51 tha:u44 kuk55 mai44 dεu231 ɂna:k44 tsu231 man231 (他就送一套新衣服給他的舅父, he then gave a new set of clothes to his uncle) man231 ti:ŋ51 zu:i231 cit23 ljeu213 (他被蛇咬了, he was bitten by a snake)
(iii) The Li branch Li 黎語 (Z. Yuan, 1994, pp. 109, 110, 142, 210)
A B
C Pochun 坡春話 (Fu, 2005)
hou11 göm33 na11 t’äu31 ɂbe11 (我被他打了, I was beaten by him) A B D
Cun 村語 (Ouyang, 1998, pp. 128, 140, 144)
na11 tաŋ31 ve:ŋ33 tաŋ31 hou33 (他給衣服給我, he gave me some clothes) gai33 meա11-ta11 ɂwai31 tաŋ31 hou11, ɬau33 meա11-ta11 ɂdա33 ra33 (請你們告訴給我哥哥,你們在哪, please tell my elder brother where you are) na11 ɂja11 pa11 ka:n33 he11 (他被狗咬了, he was bitten by a dog)
A B C
ɂbui34 la34 kou21 hoɂ33 puŋ33 puŋ22 (拿鑼給我敲敲, bring me the gong) na21 soŋ34 tsա21 viaŋ34 kuɂ33 kou21 hͻɂ33 ɂe22 (他送一把鋤頭給我 了, he gave me a hoe) ziaŋ51 tsա21 vit55 ku:n55, kou21 f ’ə21 ku21 (讓一點路,給我們走, give way for us to pass through) kə tsաəŋ buən sͻi na (我給本書他, I gave him a book) kə tsաəŋ tsi buən sͻi tsաəŋ na (我給一本書給他, I gave him a book) a-muŋ tsաəŋ a iu lai (別給人看見, don’t let people see it)
(iv) The Geyang branch Mulao 木佬語 (Bo, 2003, pp. 91, 93, 112)
A C D
ko53 na31 seŋ24 pe24 zə53 li24 (他拿錢給我了, he took some money and gave it to me) lai53-la33 pe24 to53-lo24 qui24 li24 (孩子被老師打了, the child was beaten by the teacher) na31 seŋ24 pe24 mo31 lə24 (拿錢給你看, I showed you some money
228 Cantonese give and Double-Object Construction Laji 拉基語 (Y. B. Li, 2000, pp. 132, 133, 147, 196)
A B C D
Gelao 仡佬語 (J. He, 1983, pp. 27, 37, 41, 43) Buxing 布興語 (Y. Gao, 2004, pp. 123, 127, 130, 241) Buyang 布央語 (J. F. Li, 2003)
A B C D A B C D A B C D
ʑa55 ki55 m55laŋ55 (給我五個, give me five) k’o55-ʑua55-tin55 lja13 ʑa55 kje31 su55 qei55 laŋ55 tin55 (銀行借給他二 百塊錢, the bank lent him 200 dollars) kje31 tɕo31 li44-ma44 tja55 su55 lei44 (他被狗咬兩口, he was bitten by a dog) A44-ku44 ʑa55 kje31 puŋ13 ljo31 (別讓他跑了, do not let him run away) hau ni i tsu-tsu (拿給我看看, take it and give me to have a look) zu ko mu təա xɒ, i tsəա sen ni mu xɒ (如果你想吃,我就買給你吃, if you want to eat, I will then buy some for you) u ŋkai ŋkau tau kəա (他被蛇咬了, he was bitten by a snake) ni i tsu si tau (給我看一眼, let me have a look) ʒͻ dε mi tʃi-naŋ sam ʒian (我給你錢三塊, I gave you three dollars) nai-ɂua dak tʃi-naŋ, ʒaŋ sai dε ʒͻ (王猴子拿銀子金子還給我, the monkey king repaid me with gold and silver) ɂai-pian pɤ ma tiɂ tsɤl-nͻk liau (岩扁被馬踢傷了, Yanbian was kicked and injured by the horse) mi dε ŋa tuɂ tε (你給他吃吧, you let him eat) ku33–322 na:k11 mə31 tu33–322 ða:i55 ma55-lu322 ja33 (我先給你三元錢, I give you 3 dollars first) ko322 ta45 θa322 də33 qai322 na:k11 mə31 (姑姑送兩隻雞給你, aunt gave you two chickens) wan33-ni55 la:k11-wa:u33 mə31 na:k11 lum24 ma55-lu322 (今天你兒子 被罰錢, today your son was fined) kə55 ko55 tat55 tək33 na:k11 nu45 (她也砍,讓樹倒下, she also logged and let the tree fall down)
(c) Miao-Yao family (i) The Miao branch Miao 苗語 (Z. Yang, 2004, pp. 181, 342, 369)
A C D
(ii) The She branch She 畬語 (Nakanishi, 2003, pp. 353, 354, 358)
A B C
kaŋ11 (under the entry 給 of the vocabulary list) ʈo11 (under the entry of 被 of the vocabulary list) kaŋ11 (for the meaning of 讓 and 叫) vaŋ54 ka31 kɤ35 mɤŋ31 pa22 laŋ22 ŋin31 (我先給你三塊錢, I give you three dollars first) tsͻŋ44-san44 ki31 kua11 i35 fɤŋ44 sin31 kɤ35 mɤŋ31 (張三寄過一封信 給你, Zhang San sent you a letter) lɤŋ54-le31 kɤ35 le31 ki22 ͻ44 (他們被人騙了, they were cheated by someone)
Appendix 1. Survey of the syntactic functions of give 229
(iii) The Yao branch Mian 勉話 (M. Zhao, 2004)
A C
tie35 pun33 ie33 liŋ331 / tie35 pun33 liŋ331 ie33 (爸分我田 / 爸分田我, father divided the land and gave some to me) ni331 pun33 koŋ33 an33 ky311 tsͻ55 kye35 i33 t’ͻŋ35 (他分公安局拿過 兩趟, he was caught by the public security bureau twice)
(d) The Austroasiatic family The Mon-Khmer branch Kemu 克木語 (G. Chen, 2002, pp. 177, 198)
Kemie 克蔑語 (G. Chen, 2005, pp. 110, 129, 228)
A D A B D
Hu 戶語 (J. F. Li, 2004, pp. 69–81)
joŋ ɂa ɂoɂ ɂmit ȵεɂ moi dͻ (父親給我一個小刀, father gave me a small knife) maɂ mͻk ɂa ɂiɂ vεk (母親叫給我們回去, mother said she would let us go back) ͻɂ53 lεɂ53 bͻp53 t’o35 meɂ53 ma13 bͻp53 (我拿書給你一本, I took some books and gave one to you) mai35 ɤn35 up33 t’o35 ͻɂ53 k’aɂ53-ŋaŋ35 (他哥哥講述給我聽, his elder brother told me something) jam31-ε31 iau31-k’un35 kͻ31 t’o35 ɤn35 vai31 (當時雞公就給他借, at that time, the rooster was borrowed by him)
A
ən55 wam̩55-mul55 k’a31-ͻɂ55 (他給錢我, he gave me money)
B
a-kiɂ55 a-ka:p31 ɤəɂ55 ka:ŋ33 ͻɂ55, leɂ55 wa55 k’a31 meɂ55 (那鴨是我家 送給你的, that duck was given to you by our family) ən55 ɕͻɂ55 kɒk55 tɕap55 (他被狗咬, he was bitten by a dog)
C
(e) The Austronesian family Huihui 回輝話 (Y. Zheng, 1997, pp. 73, 85, 93, 95)
A B C D
ha33 haŋ11 pu33 haŋ11 p’a43 (你肯不肯給, are you willing to give?) kau33 p’a33 ta11 puən11 k’än21 p’a33 nau33 (我給一本書給他, I gave him a book) na11-ku55 pi33 miau33 ma24 p’i55 (老鼠被貓捉了, the mouse was caught by a cat) tsi11 zäu33 p’a33 näu32 hiok24, kau33 zit24-tiŋ33 p’ia43-pian33 (只要讓 去學, 我一定努力, if you let me learn, I will certainly work hard)
Appendix 2
List of pre-modern Cantonese dialect materials Anonymous. (1877). 散語四十章 [Forty chapters of prose]. Hong Kong: St. Paul’s College. Ball, D (1888). Cantonese made easy (2nd ed.). Hong Kong: China Mail Office. Ball, D (1894). Readings in Cantonese colloquial. Hong Kong: Kelly & Walsh Limited. Ball, D (1912). How to speak Cantonese (4th ed.). Hong Kong: Kelly & Walsh Limited. Ball, D (1924). Cantonese made easy (4th ed.). Hong Kong: Kelly & Walsh Limited. Bonney, S. W (1853) Phrases in the Canton colloquial dialect. Guangzhou: Lingnan University. Bridgman, E. C (1841). Chinese chrestomathy in the Canton dialect. Macao: S. Wells Williams. Brouner, W. B., & Fung, Y.-M. (1935). Chinese made easy. Leiden: E. J. Brill. Bruce, R (1954). Cantonese lessons for Malayan students. Kuala Lumpur: Charles & Son Ltd. Bunyan, J (1871). The pilgrim’s progress: From this world to that which is to come [天路歷程] (G. Piercy, Trans.). Guangzhou: South China Religious Tract Society. Caysac, G (1926). Introduction a l’etude du dialecte cantonais. Hong Kong: Imprimerie de Nazareth. Chan, Y. K (1955). Everybody’s Cantonese. Hong Kong: The Man Sang Printers. Chao, Y.-R (1947). Cantonese primer. New York: Greenwood Press. https://doi.org/10.4159/harvard.9780674732438 Chapman, T (1973). A practical guide to Cantonese conversation. Hong Kong: Hong Kong. Chiang, K.-C.蔣克秋 (1949). Cantonese for beginners, book 2 [粵語易解卷二]. Singapore: The Chin Fen Book Store. Chiang K.-C. 蔣克秋 (1951). Cantonese for beginners, book 3 [粵語易解卷三]. Singapore: The Chin Fen Book Store. Cowles, R. T (1920). Inductive course in Cantonese (2nd ed.). Hong Kong: Kelly & Walsh Limited. Dennys, N. B (1874). A handbook of the Canton vernacular Chinese language. London: Trubner & Co. Devan, T. T (1847). The beginner’s first book in the Chinese language (Canton vernacular). Hong Kong: China Mail Office. Fulton, A. A (1888) Progressive and idiomatic sentences in Cantonese colloquial. Hong Kong: Kelly & Walsh Limited. Gospel of Luke [路加傳福音書]. (1873). Shanghai: Meihua shuguan. Gospel of Luke. (1931). New and old testaments (in Cantonese) [(廣東話)新舊約全書]. Hong Kong: Hong Kong Bible Society. Gospel of Luke. (1997). The holy bible – New Cantonese bible [聖經 – 新廣東話]. Hong Kong: Hong Kong Bible Society. Harper, L (1874). That Sweet Story of Old [悅耳真言]. Guangzhou: Xiguan tongde dajie fuyintang. Hobson, Benjamin. (1850). Dialogues in the Canton vernacular. Guangzhou: n. p. Hoh, F. T., & Belt, W (1936). The revised and enlarged edition of a pocket guide to Cantonese. Guangzhou: Lingnan University.
232 Cantonese give and Double-Object Construction
Lau, S (1972a). Elementary Cantonese: Volumes I and II. Hong Kong: Government Training Division. Lau, S (1972b). Intermediate Cantonese: Volumes I and II. Hong Kong: Government Training Division. Leblanc, C (1910). Cours de langue chinoise parlee dialecte cantonnais. Honoi-Haiphong: Imprimerie d’ Extreme-Orient. Mai, S.麥仕治. (1893a). An annotation of the Book of Odes in colloquial Guangzhou dialect [廣 州俗話詩經解義]. Mai, S.麥仕治. (1893b). An annotation of the Book of Documents in colloquial Guangzhou dialect [廣州俗話書經解義]. Morrison, R (2001). Vocabulary of the Canton dialect. London: Ganesha. (Original work published 1828). O’Melia, T. A (1941). First year Cantonese. Hong Kong: Maryknoll House. Oakley, R. H (1953). Rules for speaking Cantonese. Kuala Lumpur: Charles Grenier & Son Ltd. Stedman, T. L., & Lee, K. P (1888). A Chinese and English phrase book in the Canton dialect. New York: William R. Jenkins. Wells, H. R. (1931). Commercial conversations in Cantonese and English [英粵商業雜話]. Hong Kong: Kae Shean Printing Co. Wells, H. R (1941). Cantonese for everyone. Hong Kong: International Commercial Printing Press. Whitaker, K. P. K (1954). Cantonese sentence series. London: Arthur Probsthain. Whitaker, K. P. K (1959). Structure drill in Cantonese. London: Percy Lund, Humphries. Wisner, O. F (1906a). Cantonese Romanized 1. [n.p.] Wisner, O. F (1906b). Cantonese Romanized 2. [n.p.] Wisner, O. F (1927). Beginning Cantonese [教話指南]. Guangzhou: Guangzhou. Wu, T. C (1960). Daily Cantonese [日用粵語]. Hong Kong: Too Hung Engraving & Printing Co. Yuan, Y. C (1960). A guide to Cantonese. Hong Kong: Caslon Printers Limited.
Appendix 3
Frequency distribution of double-object patterns in Wang Shuo’s corpus
Verbs that have the highest frequency for the same syntactic pattern are grouped together. Verb 遞 ‘to pass’ 教 ‘to teach’ 分 ‘to share’ 賜 ‘to bestow’ 分配 ‘to distribute’ 許 ‘to marry off a daughter’ 打電話 ‘to give a call’ 講 ‘to tell’ 介紹 ‘to introduce’ 發 ‘to send out’ 寫信 ‘to write a letter’ 提供 ‘to provide’ 讓 ‘to give’ 派 ‘to distribute’ 說 ‘to tell’ 回信 ‘to write in return’ 寄 ‘to send’ 拍電報 ‘to telegram’ 餵 ‘to feed’ 提意見 ‘to bring up some ideas’ 寫紙條 ‘to write a note’ 找 ‘to give change’ 寫情書 ‘to write a love letter’ 打手勢 ‘to gesticulate’ 打電報 ‘to telegram’ 補 ‘to make up for’ 塞 ‘to tuck’ 回電話 ‘to call back’
IO DO
V給 IO DO 12 3 2 2 1 1 0 0 0 1 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
給 IO V DO 1 0 0 0 0 0 85 30 19 12 12 8 5 6 6 4 4 3 3 3 3 2 2 2 2 2 1 1
DO IO V IO DO
V DO給 IO
1 2 1 0 0 0 0 0 0 1 0 0 2 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0
0 0 0 0 0 0 4 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
234 Cantonese give and Double-Object Construction
Verb 來信 ‘to write a letter’ 來電話 ‘to give a call’ 傳話 ‘to pass on a message’ 匯報 ‘to report’ 敬煙 ‘to present cigarettes’ 輸 ‘to lose’ 賠償 ‘to compensate’ 解釋 ‘to explain’ 給 ‘to give’ 送 ‘to give’ 借 ‘to lend’ 還 ‘to return’ 給予 ‘to give’ 賠 ‘to compensate’ 交 ‘to handover’ 獎 ‘to award’ 傳授 ‘to teach’ 透風 ‘to divulge a secret’ Total
IO DO
V給 IO DO 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 28
給 IO V DO 1 1 1 1 1 1 1 1 0 8 0 1 0 0 0 0 0 0 233
DO IO V IO DO
V DO給 IO
0 0 0 0 0 0 0 0 261 19 7 3 2 2 1 1 0 0 305
0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 8
Appendix 4
Double-object sentences with the IO DO pattern found in pre-modern Cantonese dialect materials
Unless stated otherwise, the translations of the examples are taken from the original sources.275276 Morrison, R. (1828/1997) Vocabulary of the Canton Dialect 1. 俾我個本書 ‘Give me that book’ 2. 我俾你個半 ‘I give you one dollar and fifty cents’
Bridgman, E. C. (1841) Chinese Chrestomathy in the Canton Dialect 1. 算起嚟至多俾你一個一毫六 (p. 235) ‘the very utmost I can give you is one dollar and sixteen cents’ 2. 至多俾你七個二毫半 (p. 239) ‘the very utmost I can give is seven dollars and twenty-five cents’ 3. 我俾你五個四毫半 (p. 247) ‘I will give you $5.45’ Bonney, S. W. (1853) Phrases in the Canton Colloquial Dialect 1. 明日俾你寔音 (p. 67) ‘Tomorrow I will give you a decided answer’ 2. 我俾你一兩五錢啫 (p. 79) ‘I will give you one tael and five mace’
Bunyan, J. (1873) The Pilgrim’s Progress 1. 又俾過行天路之人快樂 [My translation: Also give happiness to those people who cross the road to Heaven] Stedman, T. L. & Lee, K. P. (1888) A Chinese and English Phrase Book in the Canton Dialect 1. 多煩俾我三箇寄信士担 (p. 41) ‘Please give me three postage stamps’ 2. 多煩俾我一張打電報嘅紙 (p. 45) ‘Please give me a telegraph blank’ 3. 我俾你兩年批 (p. 71) ‘I will give you a two year lease’ 4. 俾我三罐牛奶及一的咸鮚魚 (p. 117) ‘Give me three cans of condensed milk, and some salt codfish’ 5. 俾我五磅羊排骨 (p. 121) ‘Please give me five pounds of lamb chops’ 6. 我俾你一塊嫩呢 (p. 121) ‘We will give you a tender piece this time’ 7. 我俾你兩文 (p. 131) ‘I will give you two dollars’ 8. 俾我一札信封添囉 (p. 159) ‘You may give me a package of envelopes, too’ 9. 俾我一磅繩仔及兩磅蠟 (p. 167) ‘You may give me one pound of twine and two [pounds] of Japan wax’ 10. 我俾你六箇銀錢一百磅 (p. 167) ‘I will give you six dollars for a hundred pounds’
275. The translations of the sentences from Mai’s two works are given by myself. 276. The English translation of the examples quoted from the Gospel of Luke is based on Carroll & Prickett (1997).
236 Cantonese give and Double-Object Construction Mai Shizhi (1893a) 廣州俗話詩經解義275 1. 就傳命來賞賜我一杯酒 (Then send the order to award me a glass of wine) 2. 兼之又賜爾樣樣都係盡善盡美、咁至好嘅福澤 (And also grant you all the best blessings which make everything smooth and perfect) 3. 且爾又受到天恩賜爾享到百樣嘅天祿 (And you will receive a wide variety of blessings bestowed on you from the Heaven) 4. 共又賜爾得到長長衿享嘅福祿 (And also grant you the blessings for your long-lasting enjoyment) 5. 一定就又賜爾多多福樂 (The Heaven must also grant you a lot of fortune and joy) 6. 就唔著賜佢得有咁大官位、當個的謹要大事嘅責任咯 (Then it is not deserving to grant him such a high official position with huge responsibility over all important matters) 7. 點解個陣時、就賜我遇得咁唔好嘅時辰呢 (Why, at that time, let me encounter such a bad timing?) 8. 兼又賜爾得有好景遇、共好世界添咯 (Also grant you a good prospect and a good world) 9. 求庇蔭來增賜我地後來得有好大的福 (To ask for protection and grant us to have big blessings in future) 10. 又賜我地得有大福 (Also grant us to have big blessings) 11. 兼又賜我地亦又得長命 (Also grant us to have longevity) 12. 報答賜佢地得受大福 (To reward them so that they have big blessings) 13. 咁就祝福賜爾地有一百樣嘅福澤 (Then wish to grant you 100 kinds of blessings) 14. 賜爾地又有個的極之至好的福澤 (Grant you to have those extremely good blessings) 15. 咁就望神人庇蔭賜我地得長命 (Then wish the Heaven and our ancestors to protect us and grant us to have longevity) 16. 受天庇蔭賜我地得天咁大的福澤 (To receive protection from the Heaven which will grant us to have blessings as big as the sky) 17. 咁佢就庇蔭賜我地得有大福 (Then he protects us and grants us to have big blessings) 18. 兼又賜我地亦又得長命 (Also grant us to also have longevity) 19. 兼賜我的民人、共各人的老婆、個個都得豐衣足食的好處 (Also grant all my people and their wives to have the benefits of ample food and clothes) 20. 咁就望神恩庇蔭、賜我地各人都得有大福 (Then wish the God will protect us and grant each of us to have big blessings) 21. 兼又賜我地亦又得長命 (Also grant us to have longevity) 22. 咁求望庇蔭加賜我地得的好福 (Then wish to protect us and grant us to have good blessings) 23. 我地賜爾有多多大福 (We grant you to have a lot of big blessings) 24. 共又個上帝賜佢有天下呢一個命令 (And the God also gave the order for him to own the world) 25. 個天就賜佢配合得一個匹配的女子 (The Heaven granted him a lady who perfectly matched him)
275. The translations of the sentences from Mai’s two works are given by myself.
Appendix 4. Double-object sentences with the IO DO pattern 237
26. 共又佢亦賜佢得受天命來登天子的帝位 (Also grant him to have the mandate to take the position as the emperor) 27. 求神明庇蔭來增賜佢後來得有好景的大福 (Wish the God to protect him and grant him to have big blessings with good prospects) 28. 咁就先先贈賜佢得大的國份 (Then first grant him to have a big state) 29. 咁就先先贈賜佢得大的國份 (Then first grant him to have a big state) 30. 咁就先賜個大王得有岐山腳下周呢一處地方、來落藉開國長住 (Then first grant the emperor to have this piece of land under Mountain Qi to develop the state for long-term settlement) 31. 個天又故意賜佢有一個賢妃、大姜、來內中咁幫助佢 (The Heaven also deliberately granted him to have a virtuous concubine and Jiang to assist him from within) 32. 個上帝、又賜佢個心得靈敏、有知識來个謀度事理 (The Heaven also granted him to have a sensitive mind and knowledge to consider and deal with matters) 33. 賜佢得位做王帝、呢一叚的大福 (Granted him to have a big blessing to take the position as the emperor) 34. 兼又得受到天降賜佢地有福澤添咯 (Also the Heaven granted them to have the blessings) 35. 個文王受天命賜立佢得位做王帝之時 (At the time Emperor Wen received the mandate which grants him to take the position as the emperor) 36. 咁就求望庇蔭增賜各人後、來得有好景的大福 (Then wish to protect and grant everyone to have big blessings with good prospects) 37. 增賜爾有光明正亮的大智識 (Grant you to have bright and upright big wisdom) 38. 咁降賜爾添子 (Then grant you to have sons) 39. 個上天就加賜爾有福有祿 (The Heaven then granted you to have fortune and blessings) 40. 咁就賜爾享福、享得萬萬年咁多的福祿 (Then grant you to have the fortune and blessings to enjoy for million years) 41. 個天、就賜爾配娶到一個有本事英雄的女子、來做爾的老婆 (The Heaven granted you to marry a capable and hero-like lady to be your wife) 42. 咁個天、賜爾配娶到呢一個有本事英雄女子、來做爾的老婆 (Then the Heaven granted you to marry a capable and hero-like lady to be your wife) 43. 爾曾經得受到天命、賜爾做皇帝呢一個帝位 (You were granted by the Heaven the mandate to take the position as the emperor) 44. 個上天生賜出佢地呢的咁惡行的人 (The Heaven granted them to have these badly-behaved people) 45. 個個王帝、又贈賜個個申伯、四匹強強壯的馬 (That emperor also granted Shen Bo four strong horses) 46. 我也曾相量過、封爾住的地方 (I have also given a thought on the place I should grant you for settlement) 47. 我今賜爾呢一個圭玉的大玉印 (I now grant you this big jade seal) 48. 日後得來作為封賜爾守國功名的大印信 (This official seal will be granted to you in future for your merit in defending the state) 49. 咁爾莫廢去我現下封賜爾呢一個責任、呢一條命令 (Then you do not abandon this responsibility and this command that I now give you)
238 Cantonese give and Double-Object Construction 50. 個王帝就御賜個韓侯、一枝精精緻、裝飾得多多翠毛、的龍旗 (The emperor then granted the marquis of Han a fine and elaborated dragon flag with a lot of green feathers as decoration) 51. 我現下賜爾一隻拜祭奠酒用的大圭玉的玉酒盞 (I now grant you a big jade wine vessel made of Daigui Jade for offering wine to ancestors) 52. 我現下又封賜爾多的山川共田土 (Now I also grant you a lot of mountains, rivers, fields and lands) 53. 咁佢就封賜爾地、世世代代相傳、都照嗣次序來承繼爾地個國位的爵位咯 (Then he granted you the land, to be passed on to your subsequent generations, who would succeed the lordship of the state in accordance with their seniority in the family) 54. 賜我地咁多人有呢的谷米糧食 (Grant us to have all these grains and food) 55. 更爾又賜我地有呢的大麥、共小麥、呢兩樣麥種 (You also grant us to have malt and wheat, two kinds of wheat) 56. 佢就賜我地得有豐熟好收成的時年 (He then granted us to have a year of good harvest) 57. 又增賜我地後來得有好大的福 (Again grant us to have big blessings in future) 58. 就賜我今生就得有長壽咁的壽眉 (Then grant me to have such long eyebrows representing longevity) 59. 咁又賜我有多多個的咁嘅大福祉 (Then grant me to have a lot of fortune and big blessings) 60. 賜佢得享長壽 (Grant him to enjoy longevity) 61. 咁感格得神明來賜我得受咁多呢的福 (To thank god for granting us to have a lot of fortune and blessings) 62. 現下個天子王帝、都封賜佢有咁大爵位 (Now the emperor grants him to have such a big lordship) 63. 又賜佢得受封有一個國 (Again grant him to have a state) 64. 就封賜佢有呢的山、共河海 (Then grant him to have these mountains, seas and rivers) 65. 俾爾長命、兼富貴 (Give you longevity and wealth) 66. 共又望亦庇佑賜爾有壽、兼又常常都精精壯 (Also wish to protect you and grant you to have longevity and always a strong body) 67. 個上天就賜我地呢一個公王、有大大的福 (The Heaven granted us this prince with very big blessings) 68. 咁佢就祝福賜我地得有長壽咁的壽眉 (Then he blessed and granted us to have long eyebrows representing longevity) 69. 賜佢得有天下做皇帝、呢個帝位呢條誡命 (Grant him to have this commandment to own the world and take the position as emperor) 70. 個上帝的天意、想賜佢得有天下做皇帝呢一個恩典 (The intent of the Heaven was to grant him to have such a grace to own and world and become the emperor) 71. 唔賜俾佢商朝 (Do not grant him the Shang dynasty) 72. 又賜佢來得大建立成佢中興天下呢一的福澤 (Also grant him to have the blessings to rebuild a strong and prosperous state) 73. 亦又常常降賜俾爾有福、兼又添子 (Also often grant you to have blessings and sons) 74. 因此是以佢就得受到上天賜俾佢做皇帝、享天子至尊呢一個帝位、呢一樣的福祿 (Thus, he received from the Heaven the mandate and the blessings to become the emperor and to enjoy such a unique position)
Appendix 4. Double-object sentences with the IO DO pattern 239
75. 咁現下個王帝、又照來封賜俾呢一個韓侯、追共貊個兩處地方 (Now the emperor granted the Marquis of Han these two places, Zhuigong and Moge) 76. 來賜俾佢得有呢的咁大的福 (To grant him to have such a big blessing) 77. 就又降賜俾佢得有一位大賢臣 (Then grant him to have a major virtuous minister)
Mai Shizhi (1893b) 廣州俗話書經解義 1. 就封賜佢個官銜,叫做暘谷官 (Then grant him an official title, Yanggu) 2. 就封賜佢個官銜,叫做明都官 (Then grant him an official title, Mingdu) 3. 就封賜佢個官銜,叫做昧谷官 (Then grant him an official title, Meigu) 4. 就封賜佢個官銜,叫做幽都官 (Then grant him an official title, Youdu) 5. 咁上天賜爾個帝位個的天祿, 就會永遠絕僥無咯 (Then the Heaven will grant you the position as the emperor and blessings which will never end) 6. 咁光明來受個上帝賜爾有恩澤 (As it is so bright to receive the blessings granted to you by the Heaven) 7. 又重新再賜爾得受天命來享受帝位 (Then again grant you the mandate to enjoy the position as the emperor) 8. 共又賜爾得有大多福樂 (Again to grant you to have all the big blessings and happiness) 9. 爾就據實佢地所辦事的功績來獎賞賜佢地功名車服 (You then, according to their actual merits and performance, grant them titles, clothes and chariots) 10. 是以個上天, 就生出賜我地個國王 (Thus the Heaven grants us the birth of our emperor) 11. 個上天的天意就賜我呢一個人 (The intent of the Heaven is to grant me this man) 12. 個天意賜人受享天下呢一條誡命 (The Heaven granted the people a commandment to have the world) 13. 等佢來賜佢受得天命享坐帝位 (To let him grant him to receive the mandate to take the position as the emperor) 14. 就一定賜爾地有處安樂長久好住的地方 (Then (someone) must grant you a place where you can live happily and comfortably for a long time) 15. 就夢見上天個一位上帝, 賜我得有一個忠良的臣來幫輔我, 秉理朝政 (I dreamt that the God from the Heaven granted me to have a virtuous official to assist me in administering the ruling of my country) 16. 就賜佢長壽短壽 (Then grant him a long or a short life) 17. 個天命賜我得呢一個天下 (The Heaven granted me the mandate to have this world) 18. 就來繳賜夏朝享受天下個一條諭令 (Then to grant the Xia Dynasty a commandment to enjoy the world) 19. 佢自己有恃話佢得有上天賜佢自己做皇帝咁的天命 (He always said with self-assurance that the Heaven granted him the mandate to make himself the emperor) 20. 於是就大受皇天賜佢得受天下做皇帝呢一條天命 (Then (someone) was granted the mandate by the Heaven to have the world and become the emperor) 21. 咁賜佢有此興旺發達處 (Then to grant him to have this wealthy and prosperous place) 22. 兼又封賜佢地得享爵位 (Also grant them the lordships to enjoy) 23. 佢就賜佢得有天下做皇帝呢一條帝命 (He then granted him the mandate to have the world and to become the emperor) 24. 個上天賜佢有天下做皇帝呢一條帝命 (The Heaven granted him the mandate to have the world and become the emperor)
240 Cantonese give and Double-Object Construction 25. 惟是個天賜佢有天下做皇帝呢一條帝命 (Thus the Heaven granted him the mandate to have the world and become the emperor) 26. 因此現下就將賜佢有天下做皇帝呢一條天命 (Therefore, the Heaven is now going to grant him the mandate to have the world and become the emperor) 27. 皇帝就全享受得個上天所賜佢有天下做皇帝呢一條帝命 (The emperor now fully enjoys the mandate that the Heaven granted him to have the world and become the emperor) 28. 得受上天賜佢有天下個一條帝命咁得到天下做皇帝 (He was granted by the Heaven the mandate to have the world, and he had the world and become the emperor) 29. 上天賜佢有天下做皇帝個一條帝命 (The Heaven granted him the mandate to have the world and become the emperor) 30. 得受上天賜佢有天下個一條帝命 (The Heaven granted him the mandate to have the world) 31. 上天賜夏朝商朝個兩朝代的皇帝, 有天下做皇帝個一條咁的天命 (The Heaven granted the emperors of the Xia and Shang Dynasties the mandates to have the world and become the emperors) 32. 咁現下個上天或有賜我地個王個條命生有賢才智識 (The Heaven now may grant our emperor a destiny with virtues, wisdom and knowledge) 33. 共又或賜佢條命來享天下享得多年 (Also, or to grant him the destiny to enjoy the world for many years) 34. 求得上天長保賜爾得有天下來享呢一條恩典咁的帝命 (Wish the Heaven to forever grant you the graceful mandate to have the world) 35. 上天賜我周家有天下, 呢一段咁好機會的恩澤好處 (The Heaven granted us, the Zhou family, such a good opportunity and graceful benefit to have the world) 36. 咁失去上帝眷顧佢, 賜佢有天下個一條帝命咁的恩典 (Then he lost the grace and mandate of having the world that the Heaven specially granted to him) 37. 上天賜佢得受享有天下個一條帝命 (The Heaven granted him the mandate to have the world) 38. 然後正得來受到上天, 賜有天下做皇帝, 咁呢一條帝命 (Then he received from the Heaven the mandate to have the world and become the emperor) 39. 賜佢有天下做皇帝, 個一條咁的恩典帝命咯 (The Heaven granted him the graceful mandate to have the world and become the emperor) 40. 受到上天賜佢有天下做皇帝, 個一條恩典咁的帝命 (He was granted by the Heaven the graceful mandate to have the world and become the emperor) 41. 賜佢有天下做皇帝, 咁呢一條帝命 (The Heaven granted him the mandate to have the world and become the emperor) 42. 就賜佢商朝天下, 就得有大多人民 (Then granted him the world, or the Shang Dynasty, which had a lot of subjects) 43. 佢就賜佢地得到有個的良臣 (He then granted them to have a virtuous official) 44. 咁我地然後正得上天賜得我地周家, 得有天下來做皇帝, 呢一條恩典的帝命 (We, the Zhou family, then were granted by the Heaven the graceful mandate to have the world and become the emperor)
Appendix 4. Double-object sentences with the IO DO pattern 241
45. 賜佢有天下呢一條恩典的大帝命 ((The Heaven) granted him the big and graceful mandate to have the world) 46. 賜佢有天下做皇帝, 咁呢一條恩典的帝命 (The Heaven granted him the graceful mandate to have the world and become the emperor) 47. 就賜佢得有呢一班實在有善德的能人 (Then to grant him to have these really capable and virtuous men) 48. 上天賜我地有天下來享呢一件事 (The Heaven granted us one thing which is to enjoy the world) 49. 咁一樣照來受個上天, 賜佢有天下來享, 咁呢一條帝命 (Then, the Heaven also granted him the mandate to have the world) 50. 凡辦事辦得有功, 就來獎賞, 陞賜佢官銜 (If one completes one’s tasks with merits, one may come forth for a reward in terms of a promotion or an official title) 51. 我現下賞賜爾一埕香花黑糯米浸的香酒 (I now grant you a pot of wine brewed with black sticky rice and fragrant flowers) 52. 佢就來封賜俾佢做大官 (He then came to grant him a big official title) 53. 佢就降賜俾佢有百樣好的福澤 (Then he granted him to have hundred kinds of blessings) 54. 佢就降賜俾佢來受百樣的災難 (He then made him suffer hundred kinds of disasters) 55. 又封賜分俾佢地的地方 (Also grant and divide some lands among each of them) 56. 於是個上天就賜俾過大禹得有呢一條九疇治天下的大法子 (Then the Heaven granted Emperor Yu to have a magnificent way to make use of the laws of the nature to rule the world) 57. 現下上天將賜過大國殷家有天下做皇帝, 呢一條天恩的帝命 (Now the Heaven grants the Yin family to have the graceful mandate to have the world and become the emperor) LeBlanc, Charles (1910) Cours de langue Chinoise Parlee Dialecte Cantonnais 1. 佢俾我一幅大地理地圖 (p. 54) ‘il m’a donne une grande carte’ [my translation : He gave me a big map] 2. 你寫字之後我俾你野食 (p. 119) ‘quand tu auras ecrit, je te donnerai quelque chose a manger’ [my translation : After you finish writing, I will give you something to eat] 3. 佢來我處嘅俾我個張紙 (p. 154) ‘il est venu chez moi me donner cette feuille de papier’ [my translation : He came to my place and gave me that piece of paper] 4. 我俾你十個銀錢一個月 (p. 157) ‘je te donne dix piastries par mois’ [my translation : I give you ten dollars per month] Caysac, Georges (1926) Introduction a l’Etude du Dialecte Cantonais 1. 俾我一隻芭蕉 (p. 31) ‘donne-moi une banane’ [My translation : Give me a banana] 2. 佢俾我毋夠十文銀一個月 (p. 78) ‘il me donne moins de dix piastres par mois’ [My translation : He gives me less than ten dollars per month] 3. 俾佢兩文錢 (p. 84) ‘donne-lui sapeques’ [My translation : Give him/her two dollars] 4. 俾我一隻隨便邊隻 (p. 100) ‘donnez-m’en un, n’importe lequel’ [My translation: Give me any one of those] 5. 俾佢飯食 (p. 107) ‘donne-lui du riz a manger’ [My translation : Give him/her some rice to eat] 6. 俾佢兩文錢 (p. 107) ‘donne-lui sapeques’ [My translation : Give him/her two dollars]
242 Cantonese give and Double-Object Construction
Gospel of Luke (1931)276 78. 耶穌叫十二門徒埋來, 賜佢哋能力權柄 (p. 84) ‘Then he [Jesus] called his twelve disciples together, and gave them power and authority’ 79. 我已經賜你哋權柄 (p. 88) ‘I give unto you power’ 80. 若係訛詐過人嘅, 就還返佢四倍 (p. 102) ‘and if I have taken any thing from any man by false accusation, I restore him fourfold’ 81. 我將賜你口才智慧 (p. 105) ‘For I will give you a mouth and wisdom’ Chiang, K. C. (1951) Cantonese for Beginners 58. 俾佢一寸, 佢要一尺 (p. 49) ‘Give him an inch and he’ll take an ell’ Oakley, R. H. (1953) Rules for Speaking Cantonese 1. 俾每個人十文 (p. 89) ‘Give each man $10’ 2. 送個個人一部書 (p. 89) ‘Give each man a book’ 3. 俾個老年人呢件野 (p. 89) ‘Give that old man this’ 4. 你俾我10文, 我至去 (p. 107) ‘Only if you give me $10 I will go’ Lau, Sidney (1972) Intermediate Cantonese 1. 我要加佢人工 (p. 777) ‘I must increase his wages’
Gospel of Luke (1997) 1. 佢嘅鄰居同親戚聽見主賜佢咁大嘅恩慈 (p. 65) ‘And her neighbours and her cousins heard how the Lord had shewed great mercy upon her’ 2. 我已經賜你哋權柄 (p. 80) ‘I give unto you power’ 3. 賜俾我哋每日需要嘅飲食 (p. 81) ‘Give us day by day our daily bread’ 4. 你俾過我乜嘢呢 (p. 88) ‘thou never gavest me a kid’ 5. 請俾我哋更大嘅信心 (p. 89) ‘Increase our faith’ 6. 我就還佢四倍 (p. 92) ‘I restore him fourfold’ 7. 但凡有嘅, 要俾佢更多 (p. 93) ‘That unto every one which hath shall be given’ 8. 係邊個俾你呢啲權呢 (p. 93) ‘Who is he that gave thee this authority?’ 9. 因為我會賜你哋口才同智慧 (p. 95) ‘For I will give you a mouth and wisdom’
276. The English translation of the examples quoted from the Gospel of Luke is based on Carroll & Prickett (1997).
Appendix 5
Sentences used in the production task
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Verb
Subject
畀 請你 Please give me a chance
DO
個機會
IO 我
畀 我 對筷子 I gave you a pair of chopsticks
你
畀 (你) Give me a sweet potato
畀 我會 一個交代 I will give you all an explanation
大家
條番薯
我
畀 (你) You gave him/her a cap
頂帽
佢
畀 你 Give him some money
啲錢
畀 公司 呢個獎 The company gave me this award
我
畀 我 呢本書 I gave this book to the friend that I just bumped into.
求先我撞到嗰個朋友
佢
畀 (你) 兩斤白菜 Give me two catties of vegetables
我
借 你 咁多錢 You lent him/her so much money
佢
幾文
我
借 (你) Please lend a pen to this lady
枝筆
嗰位小姐
借 你 You lent him/her three pots
三個煲
畀 樓下個看更 一疊用花紙包好嘅嘢 對面大廈門口個阿婆 The watchman downstairs gave a package in nice wrapping paper to the old lady at the entrance of the opposite building
借 你 Please lend me some money
借 你 幾十萬 王先生兩公婆 You lent several hundred thousand dollars to the Wong couple 佢
244 Cantonese give and Double-Object Construction
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Verb
Subject
DO
借 你 對手襪 You lent him/her a pair of gloves
IO 佢
借 你 三份報紙 You lent him/her three newspapers
佢
十文
我
借 你(求其) Lend me whatever you have
啲嘢
我
遞 你 Please pass a pen to my mum
枝筆
借 佢 He lent me 10 dollars
借 你(千祈) 任何嘢 You must not lend anything to this guy
呢個人 我阿媽
遞 你 啱啱點著嘅蚊香 睏喺細房個阿婆 Please pass the mosquito-repellent incense just lighted to the old lady who is sleeping in the smaller room 遞 黃經理 一份最新嘅公司報告 董事長 Mr. Wong, the manager, passed the latest company report to the chairman 遞 個爸爸 張紙巾 The father passed a napkin to his son 遞
個仔
坐喺佢隔籬戴太陽眼鏡個 男人 He passed some black stuff to the man with sunglasses sitting next to him (佢)
舊黑“麻麻”嘅嘢
遞 消防員 樽水 想跳樓個女人 The fireman passed a bottle of water to the woman who wanted to jump off from the building 遞 個大作家 幾本佢寫嘅書 今日來訪問佢嘅記者 The writer passed some of the books he wrote to the reporters who interviewed him today 遞
個賊
一張寫有「打劫兩萬」 個櫃檯職員 嘅紙仔 The robber passed a piece of paper with “20 thousand dollars” on it to the teller behind the counter 交
陳生
幾件貨辦
隔籬街永新工廠大廈個 老闆 Mr Chan passed some samples to the manager of Wing Sun factory building of the next block 交 你 呢封信同埋呢件衫tdtd 你哋嘅黃經理 Please give this letter and this piece of garment to your manager, Mr Wong
Appendix 5. Sentences used in the production task 245
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
Verb
Subject
DO
IO
交 政府 立法草案 立法會 The government gave the draft bill to the Legislative Council
交 你 二千五百文入會費 我哋個會計部 Give the admission fee of 2500 dollars to our accounting department 交 你 今個月嘅租 Have you given the rent of this month to the landlord?
包租婆未呀?
交 你(仲未) 報稅表 You still haven’t sent the tax return to the tax bureau?
稅局
交 個銀行經理 收返來嘅假銀紙 差人 The bank manager gave the fake bank notes he received to the police
分 我 I share a half with you
一半
你
分 你 呢啲贜物 幾多個人 To how many people did you distribute these stolen goods? 分 個有錢佬 自己副身家 屋企人 That rich man divided his property among his family members 分 你 咁少嘢 You only share so little with me
我
分 佢 我辛辛苦苦賺返來啲錢 佢啲仔女同埋細佬妹 S/he divided the money that I earned among his/her children and brothers and sisters 分 志願團體 啲糧食同埋災款 啲難民 The voluntary organization allocated the food and donations to the victims 退 個老闆 收多咗嘅錢 啲人客 The boss refunded the over-charged amounts to the customers 退 我 呢幾件衫同埋架車 人哋 I am going to return these clothes and this car to someone 退 你哋 啲錢同埋車馬費 Please give me back the money and traveling expenses
我
退 你 錢 Can you return the money to me?
我
退 政府 六成嘅稅 低收入嘅家庭 The government refunded 60% of the tax income to those low-income families
退 (你) 呢件穿咗窿嘅冷衫 Return this defective sweater to your boss 還
我而家
你老闆
十幾年前同你借落嘅 你 本錢 I am returning you the capital I borrowed from you 10 years ago
246 Cantonese give and Double-Object Construction
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
Verb
Subject
DO
還 佢 啲書 He returned the books to the library
IO
圖書館
還 佢 筆錢 Has he returned the money yet?
人 (未呀?)
還 我 所有嘢 I have returned everything to you
你
輸 你竟然 我層樓 How could you lose my house to the dealer?
莊家
輸 老黃 盤棋 Wong lost a chess game to Lee
老李
還 阿老李 個電視機 Lee returned the TV set to you
還
你
黃生
一個花樽,一個花盆,兩 我哋 個槌仔 Mr Wong returned a vase, a flower pot and two hammers to us
輸 佢 兩盤棋 He lost two chess games to me
我
輸 佢 兩盤棋 To whom did he lose two chess games?
邊個
輸 老劉 廿萬 Lau lost 200 thousand dollars to my brother
我細佬
補 我哋 一堂 We will give a make-up class to all of you
大家
補 你 Supply the documents to us
我哋
輸 有個經理 幾萬文同埋架車 人 A manager lost several thousand dollars and his car to someone
補 佢哋 兩百文 They compensated me and my wife with 200 dollars
我兩公婆
補 我(要) 乜嘢文件 你哋公司 What kind of documents should I supply to your company? 啲文件
補 我哋公司 兩個月嘅雙糧 你 Our company will compensate you with two-months’ salary
補 你 張卡同埋一扎花 你女朋友 You’d better make up and send a card and a bunch of flowers to your girl friend 租
樓上嗰兩公婆 間屋
啱啱喺外國讀完書返來嗰 對兄妹 The couple upstairs rented out their apartment to the brother & sister who just returned from their overseas studies
Appendix 5. Sentences used in the production task 247
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84
Verb
Subject
租
我
DO
租 你今次 幾多樣嘢 How many things did you rent to your friend this time?
IO
你個好朋友
呢間樓連天台/棚,游泳 佢 池同埋球場 I rented him/her this house, with the balcony, swimming pool and the field for ball games 租 你 架車 Did you rent the car to someone?
人
租 你(竟然) 隻船 How can you rent the boat to someone else?!
第二啲人
租 你 個單位 To whom did you rent the apartment?
乜嘢人?
獎 你 What will you award to me?
乜嘢
我
獎 爸爸 乜嘢 What will Daddy award to me?
我
獎 你(今次) 兩件恤衫 This time, award two shirts to your brother
細佬
獎 媽咪 杯士多啤梨雪糕 Mother awarded a straw-berry ice-cream to you
你
獎 你(應該) 啲有意義嘅嘢 You’d better award something meaningful to him
佢
獎 上頭 四十萬現金 今次有份破案嘅同事 The supervisor awarded 400 thousand dollars cash to those colleagues who solved the case 撥 政府 兩幅地皮 承建商 The government gave two new pieces of lands to the contractor
撥 公司 二千萬 我哋個部門 The company allocated 20 million dollars to our department
撥 政府 五億四千萬 香港啲大學 The government allocated 540 million dollars to the universities in Hong Kong 撥 政府 好多資源 中小學 The government allocated many resources to primary and secondary schools
撥 社會福利署 一筆緊急救濟金 山泥傾瀉嘅災民 The Social Welfare department allocated some emergence funds to the victims of landslide 賣 你想 呢間舖 To whom do you want to sell this shop? 賣 我 I sold the car to the dealer
架車
邊個 車行
248 Cantonese give and Double-Object Construction
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Verb
Subject
DO
賣 你想 層樓 Do you want to sell this apartment to Miss Lee? 賣 嗰個人 間餐館 To whom did that man sell the restaurant?
IO
李小姐 邊個
賣 我 間屋 I sold the house to the man who came yesterday
尋日來嗰個人
送 我 I gave him/her a book
本書
佢
一張枱
佢
送
隔籬嗰阿叔
兩本書,三個書架同埋 我同我細佬 其他嘢 The gentleman next door gave 2 books, 3 bookshelves and other things to my younger brother and me
送 我 I gave him/her a desk
送 你 樽豉油 To whom did you give the bottle of soy sauce?
邊個
找(錢) 個夥計 三百文 The staff changed 300 dollars extra to the customers
啲客
送 阿媽 你頭先啱啱整好個蛋糕 樓上陳師奶嗰仔 Mum gave the cake you just made to the son of Mrs Chan upstairs
找(錢) 你不如 幾個五文同埋十文紙 我 How about giving change to me with some 5-dollar coins and 10-dollar notes? 派 班義工 冷衫手襪頸巾 老婆婆 The voluntary workers delivered sweaters, gloves and scarves to the old ladies 派 先生 啲試卷 The teacher distributed the exam paper to us
大家
教 佢 兩度散手 He taught some tricks to my son
我個仔
教 佢 好多英文生字 He taught a lot of English words to his son
個仔
嫁 佢老豆 個女 His father married the daughter to the debtor
債主
100 嫁 佢阿媽竟然 個女 His mother married the daughter to that poor fellow
嗰個窮鬼
102 讓 你 三步棋 Let me have three moves first [in a chess game]
我
101 讓 請你 個位 Could you please give your seat to me and my children?
我哋幾仔乸
Appendix 5. Sentences used in the production task 249
Verb
Subject
DO
103 賠 佢 一千幾百 S/he compensated around 1000 dollars to us
IO
我哋
104 寄 阿媽 相 To whom did mother send the pictures?
邊個
106 寄 黃師奶 一個包裹 Mrs. Wong sent a parcel to her son
個仔
108 寄 你 呢件貨辦 Send this sample to the manufacturer
廠家
110 寄 阿媽 幾條短褲 The mother sent some shorts to her son
個仔
112 介紹 舅父 個女仔 Uncle introduced a girl to you
你
114 介紹 你 陳小姐 To whom did you introduce Miss Chan?
邊個
116 介紹 佢 幾個朋友 He introduced some friends to me
我
118 寫 佢 封信 He wrote a letter to the gentleman you met yesterday
你尋日識嗰位先生
120 打(電話) 個姑娘 電話 The nurse gave a call to the doctor
醫生
105 寄 你哋 乜嘢生日禮物 What kind of birthday gift are you going to send to Dad?
老豆
107 寄 阿媽 啲錢 Mum sent some money to grandma
婆婆
109 寄 阿媽 把遮 To whom did Mum send the umbrella?
邊個
111 寄 王師奶 封信 去咗英國讀書個姪仔 Mrs. Wong mailed a letter to her nephew who studied in the UK
113 介紹 佢 個男仔 He introduced a man to his aunt
佢姑媽
115 介紹 陳經理 新客仔 我哋公司 Manager Chen introduced some new clients to our company
117 寫 佢 封信 He wrote a letter to Miss Chan
陳小姐
119 寫 你 You wrote a note to them
佢哋
張字條
121 打(電話) 個孫女 個電話 The granddaughter gave her grandmother a call
嫲嫲
Appendix 6
Sentences used in the perception task 1. 唔該畀我三個郵票 (Could you please give me 3 stamps?) 1a. 唔該畀三個郵票我 1b. 唔該畀三個郵票畀我 2. 我諗如果你繼續畀佢拔蘭地,佢重會飲得落 (I think if you continue to give him brandy, he can still drink it) 2a. 我諗如果你繼續畀拔蘭地佢,佢重會飲得落 2b. 我諗如果你繼續畀拔蘭地畀佢,佢重會飲得落 3. 請你遞畀老闆呢份嘢 (Could you please pass this thing to the boss?) 3a. 請你遞呢份嘢畀老闆 3b. 請你遞老闆呢份嘢 3c. 請你畀老闆遞呢份嘢 4. 你找返我兩文先啱 (It is only correct if you change me 2 dollars) 4a. 你找返兩文我先啱 4b. 你找返畀我兩文先啱 4c. 你找返兩文畀我先啱 5. 我送畀你一本書 (I give you a book) 5a. 我送你一本書 5b. 我送一本書畀你 5c. 我送一本書你 5d. 你想送佢乜嘢? (what do you want to give him?) 6. 我畀咗佢兩萬一千五百文 (I gave him 21,500 dollars) 6a. 我畀咗兩萬一千五百文畀佢 6b. 我畀咗兩萬一千五百文佢 6c. 你畀咗兩萬一千五百文邊個? (To whom did you give 21,500 dollars?) 6d. 你畀咗兩萬一千五百文畀邊個? 7. 你分啲蛋糕畀啲小朋友啦 (Divide the cake among the kids) 7a. 你分畀啲小朋友啲蛋糕啦 7b. 你分啲小朋友啲蛋糕啦 7c. 你分啲蛋糕過啲小朋友啦 8. 你交咗啲功課畀邊個呀? (To whom did you submit your assignment?) 8a. 你交咗畀邊個啲功課呀? 8b. 你交咗邊個啲功課呀? 8c. 你畀邊個交咗啲功課呀? 9. 我送你呢一枝筆 (I give this pen to you) 9a. 我送畀你呢一枝筆
252 Cantonese give and Double-Object Construction 9b. 我送呢一枝筆你 9c. 我送呢一枝筆畀你 9d. 我畀你送呢一枝筆 10. 我還返上個禮拜整爛咗個藍色公仔畀佢 (I returned the blue doll that I damaged last week to him/her) 10a. 我還返畀佢上個禮拜整爛咗個藍色公仔 10b. 我還返佢上個禮拜整爛咗個藍色公仔 10c. 我還返上個禮拜整爛咗個藍色公仔佢 11. 我借咗佢三本書 (I borrowed 3 books FROM him/her OR I lent three books TO him/her) 12. 我租咗佢兩間屋 (I rented 2 houses FROM him/her OR I rented two houses TO him/her) 13. 我無送杯畀佢 (I did not give him a cup) 13a. 我無送佢杯 13b. 我無送畀佢杯 14. 你寄份禮物邊個? (To whom did you send a gift?) 14a. 你寄份禮物畀佢 (You sent a gift to him/her) 14b. 你寄畀佢一本書 (You sent him/her a book) 14c. 你寄份禮物我細佬 (You sent a gift to my brother) 14d. 你寄佢一本好貴嘅書 (You sent him a very expensive book) 15. 我會畀大家一個交代 (I will give you all an explanation) 16. 你不如畀佢多一次機會啦 (Why don’t you give him/her another chance?) 17. 我好多謝公司畀我呢個大獎 (I would like to thank the company for giving me this top prize)
Appendix 7
Information sheet for the fieldwork of the double-object construction in Hong Kong Cantonese
Ref. No: ________ Background Information Sheet for the Investigation of the Double-object Construction in Hong Kong Cantonese277 1. Age: 15 below ☐ 16–20 ☐ 21–25 ☐ 26–30 ☐ 31–35 ☐ 36–40 ☐ 41–45 ☐ 46–50 ☐ 51–55 ☐ 56–60 ☐ 60 above ☐ 2. Gender: Male ☐ Female ☐ 3. Native place: ________________ 4. Place of birth: ________________ No ☐ 5. Raised in Hong Kong: Yes ☐ If not, then length of stay in Hong Kong: ____________ Other place(s) stayed and length of stay: _____________ Other: ______ 6. Native dialect: Hong Kong Cantonese ☐ 7. Educational level: Undergraduate level ☐ a. If still in school: Graduate level ☐ Matriculated level ☐ Secondary level ☐ Primary level ☐ Received in Hong Kong: Yes ☐ No ☐ Major/main stream: ________________________________ Undergraduate level ☐ b. If not in school: Graduate level ☐ Matriculated level ☐ Secondary level ☐ Primary level ☐ Not applicable ☐ Received in Hong Kong: Yes ☐ No ☐ Major/main stream: ________________________________ 8. Occupation (or before retirement/unemployment): _______ 9. Language/dialect mainly used with colleagues/classmates: _______________ No ☐ 10. Knowledge of Putonghua/Modern Standard Chinese: Yes ☐ Ability: Reading ________ Listening ________ Writing ________ Speaking ________ Watching/listening to programs in Mandarin: with subtitles/scripts? Yes ☐ No ☐
277. The information on this sheet were elicited orally instead of asking the interviewees to fill out the form.
254 Cantonese give and Double-Object Construction
11. Knowledge of English: Yes ☐ No ☐ Ability: Reading ________ Listening ________ Writing _________ Speaking ________ Watching / listening to programs in English: with subtitles/scripts? Yes ☐ No ☐ 12. Knowledge of other dialects/languages: _______________ Ability: Reading _________ Listening _________ Writing _________ Speaking _________ 13. Language/dialect used with family: __________________ 14. No. of family members living together: _______________ 15. Overall usage of different languages (in a daily basis) 31–50% ☐ 51–70% ☐ 70% above ☐ Cantonese: below 30% ☐ English: below 30% ☐ 31–50% ☐ 51–70% ☐ 70% above ☐ Putonghua: below 30% ☐ 31–50% ☐ 51–70% ☐ 70% above ☐ (*including working task, such as report writing, emails, or online chatting, etc.) 16. Background of FATHER: Age: 36–40 ☐ 41–45 ☐ 46–50 ☐ 51–55 ☐ 56–60 ☐ 60 above ☐ Native place: _______________________ Native dialect: ______________________ Born in Hong Kong: Yes ☐ No ☐ Raised in Hong Kong: Yes ☐ No ☐ Length of stay in Hong Kong: __________________________ Language mainly used with parents/relatives / siblings / friends: _________________ Language mainly used at work: _________________________ Education: Graduate ☐ Undergraduate ☐ Secondary ☐ Primary ☐ Not Applicable ☐ Major of study: ____________________________ Occupation (or before retirement/unemployment): _________ 17. Background of MOTHER: 41–45 ☐ 46–50 ☐ Age: 36–40 ☐ 51–55 ☐ 56–60 ☐ 60 above ☐ Native place: _______________________ Native dialect: ______________________ Born in Hong Kong: Yes ☐ No ☐ Raised in Hong Kong: Yes ☐ No ☐ Length of stay in Hong Kong: __________________________ Language mainly used with parents/relatives/siblings/friends: ___________________ Language mainly used at work: _________________________ Education: Graduate ☐ Undergraduate ☐ Secondary ☐ Primary ☐ Not Applicable ☐ Major of study: ____________________________ Occupation (or before retirement/unemployment): ______ 18. Any caretakers/nannies at home? Yes ☐ No ☐ Length of service with your family: ____________________ Nationality of the caretakers/nannies: _______________ Language backgrounds of the caretakers/nanny: _________ Language used with the caretakers/nanny: ______________
Appendix 7. Information sheet for the fieldwork of the double-object construction 255
19. Any siblings? Yes ☐ (how many? ___) No ☐ Relationship and background: _________ 20. Living with other relatives (such as grandparents, uncles, aunts, cousins…): Yes ☐ No ☐ Relationship: ________________________________________ Their linguistic background: ____________________________ 21. Novels/literary works/magazines that you have read / always read: Modern Standard Chinese: _____________________________ English: ____________________________________________ Cantonese: __________________________________________ 22. Favorite TV/Radio programs/movies that you always watch/listen: Putonghua:__________________________________________ English: ____________________________________________ Cantonese: __________________________________________ 23. Favorite songs that you like or always sing/listen: Putonghua:__________________________________________ English: ____________________________________________ Cantonese: __________________________________________ 24. Time spent on newspaper reading (on a daily basis): 30 min ☐ 1–2 hrs ☐ 2 hrs above ☐ Chinese: No ☐ English: No ☐ 30 min ☐ 1–2 hrs ☐ 2 hrs above ☐ Name(s) of newspapers: _______________________________ Section(s) read: ______________________________________ 25. Time spent on watching TV or listening to radio programs (on a daily basis): Cantonese: No ☐ 30 min ☐ 1–2 hrs ☐ 2 hrs above ☐ Mandarin: No ☐ 30 min ☐ 1–2 hrs ☐ 2 hrs above ☐ English: No ☐ 30 min ☐ 1–2 hrs ☐ 2 hrs above ☐ Name(s) of the program(s): _____________________________ 26. Usage of ICQ/MSN/Chatroom/Message Board on Internet Use or visit the above? Yes ☐ No ☐ If yes, how much time do you spend on them (on a daily basis):__________________ What language(s) do you often use in online chatting, posting web messages? Cantonese ☐ English ☐ Modern Standard Chinese ☐ 《 END OF SURVEY》
Appendix 8
Non-native double-object sentence patterns used by the informants in the production task
The numbers represent the sentence numbers of the production task. Informant
[+GIVE] verb V IO DO
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
50,89,94,98 3,36,39,48,76,97,98,102 3,97,98,102 3,55,58,93,102 3,52,97,98,102 1, 2,3,9,60,64,72,73,77,97,98 3,36,52,72,74,90,97,98,102,116 2, 3,4,5,59,98,102 1,3,4,9,75,89,102 3,36,72,76,97,98,102 3,102 3,9,39,48,52,55,72,73,74,75,76,90,97,98,102,103,105 3,48,52,58,61,73,74,98,102 97,98,102 3,97,102 3,9,36,58,97,98,102 3,10,36,48,61,72,88,97,98 3,72,97,98 36,72,98,102 3,10,49,73,97,102 3,23,60,64,72,74,97,98 3,97,98,102 1,3,10,64,72,74,76,90,97,102 3,9,48,61,64,72,73,76,93,94,97,102,103,106 3,97,98 3,60,74,97,98,102 3,9,74,75,96,98,102 36,72,73,97,98,102 3,10,98,102 3,31,36,39,53,55,61,64,73,74,76,78,94,97,98,102,116
V 畀 IO DO
0 0 56 0 0 48 0 55,103 48 23,48,53,55,57,103 0 0 37,55,76,94,112 0 0 0 23,77,105 0 0 0 48,78,51,52 0 39 0 48 32 0 39,48,53,64 45,81
258 Cantonese give and Double-Object Construction
Informant
[+GIVE] verb V IO DO
31 32 33 34 35 36 37 38 39 40
3,36,97,98,102 3,55,61,72,74,89,94,97,98,102 3,98 64,89,90,97,102 3,98 3,74,97,98,102 3 3 3 3
V 畀 IO DO 0 0 0 0 0 0 0 0 0 0
Index
A Acceptability 133–134, 146, 160–167, 169–171 Agent 44–47, 50–52, 92–94, 99–100 Allative 88–89 Allative case 88–89 Altaic languages 18, 33, 37 Ambiguity 98–100, 158, 167, 171 Apparent time approach 55–56 Areal feature 17, 37, 81, 122, 174 Areal features 19, 173 Areal linguistics 37–38 Areal-typological perspective 17 Austroasiatic 17–20, 233 Austronesian 17, 20, 37, 233 B Beijing Mandarin 18, 107, 110, 112, 114 Beneficiary 8, 34–36, 41–43, 51, 84–86, 114–118, 157, 174 Beneficiary construction 12, 41–43, 82, 86, 112 Beneficiary constructions 42 Beneficiary marker 34–36, 41–43, 51, 82, 84–85, 174 Bilingualism 173 Biliteracy and Trilingualism 140 Borrowing 18, 38, 133, 173 C Cantonese ix, 10, 14–17, 39–49, 55–60, 62, 73–77, 81–82, 85–86, 89–94, 96, 99–101, 119–121, 123, 125–128, 131–147, 150, 154, 158–160, 162, 165, 167, 171–175, 235–236, 239 Cantonese corpora 175 Causation 127
Causative ix, 5, 33–38, 43–49, 51, 73, 97–101, 158, 169–171, 173, 205 Causative construction ix, 5, 44, 46–48, 98–99, 101, 158, 169, 171 Causative verb 33–37, 43–44, 46, 48, 51, 73, 99–100, 173, 205 Causativity 5, 51 Caused event 44 Causee 44, 51–52 Causer 44, 51–52 Chaozhou people 137 Chinese dialects 10, 14–17, 19–21, 28–30, 32–33, 37, 86, 88–89, 105, 107, 109, 118–119, 122–123, 173, 205 Chronological development ix, 40, 99–101 Chronology 53, 55, 173 Code-switching 174 Cognitive linguistics approach 134 Comparative construction 106–107 Conghua dialect 88–89 Contact linguistics 174 Contact-induced grammaticalization 38, 173 Content word 15, 37, 39 Content words 39 Corpus ix, 4, 58, 76, 96, 112– 115, 118, 122, 131, 175, 237 Corpus data ix, 112, 115, 118, 122, 175 D Dative 7–9, 51, 88–89, 99, 101 Dative alternation 126 Dative construction 8, 89 Dative marker 8, 99 Dative shift 126–127, 129–130
Deep structure 126 Degree of acceptability 146, 160, 163, 166, 170 Degrees of acceptability 163 Degree of unacceptability 165 Deprivation 154, 167–168 Deprive type verbs 133 Desirable action 47 Diachronic ix, 39–41, 55–56, 99–100, 173–174 Direct object 5–6, 8 Directional verbs 82, 85, 173 Discourse approach 129–131, 133 Distal demonstrative 100 Ditransitive construction 7–8 Ditransitive sentence 8, 10 Ditransitive verb 7–8, 10 Ditransitive verbs 7 DO IO pattern 10, 14, 108, 110, 112–114, 120–122, 125–127, 129–131, 133, 135, 144 DO IO word order 107, 119, 127 Dominance 129 Double-object construction ix, 5–14, 41–44, 50, 52–53, 77, 79, 82, 85, 89, 105, 107–110, 112–114, 118–120, 122–123, 125–128, 130–131, 133, 136–137, 144, 147, 150, 171, 173–174, 257 Double-object sentence pattern 164, 170, 261 Double-object sentences 58, 77–80, 85, 90, 110, 113–114, 129, 133, 150, 158–159, 171, 239 Double-object verb 3–4, 6–7, 14–17, 21, 23, 25, 27–30, 32–33, 35–37, 40, 42–45, 50–51, 61, 82, 86–90, 99, 111, 158, 169, 172–174, 205, 207–209, 212– 214, 216–221, 223, 225
260 Cantonese give and Double-Object Construction
Double-object verbs 5–6, 12, 22–23, 25, 28–30, 32, 34, 37, 40, 61–62, 77, 79, 86, 110–111, 128–129, 133, 144–145, 147, 150, 157–158, 160, 163–164, 166, 173 E Early dialectal materials 56, 58, 61 Ethnolinguistic vitality index 138 Extra-linguistic factors 125, 144 F Fieldwork ix, 3, 88, 115, 117–118, 122, 135–137, 143–147, 149–150, 171, 173–174, 257 Function word 15, 37, 39 G Gan dialects 25, 109, 218 Genetic relationship 18 give ix, 3–12, 14–17, 20–21, 28–53, 55, 60–74, 76–77, 80–82, 158, 163, 173–175, 205 Giver 5, 44, 52, 85–86, 89 Give-type double-object construction 8, 10, 107, 112, 119, 122, 150 Give-type double-object verbs 5–6, 14 Give-type IO marker 86, 89 Giving 3–5, 7, 12, 43–44, 49, 51–52, 85, 167–168 Gospel of Luke 76, 93–94 Go-type IO marker 86–89, 174 Grammatical functions 14–15, 39, 51, 53, 99–100 Grammaticalization ix, 11, 15–16, 19, 34, 38–41, 51, 55, 88, 99–101, 173, 175 H Hakka dialects 22–23, 28–30, 109, 209 Hakka people 137 Haplology 81, 90, 161 Heavy End Shift 127 Heavy Noun Phrase Shift 163
High language 140 Hmong Mien 20 Hoklos people 137 Home language 138–140 Homonymy 37 Homonymies 40 Hong Kong Cantonese 10, 14–17, 42–49, 52, 76–77, 101, 105–106, 123, 126, 133, 135, 137, 139, 143–147, 158–159, 171–173, 257 I Indirect object ix, 5–6, 8 Indirect object marker ix, 12, 41 Instrument ix, 15, 49–52, 96–99, 174 Instrument marker ix, 96–97 Instruments 15, 49, 95–100, 174 Involuntary permission 48 IO DO pattern 10, 11, 77, 107–112, 114, 120–122, 125–128, 130–135, 137, 146, 174, 239 IO DO word order 10, 107, 120, 127, 154, 173 IO marker 12, 14–15, 33–36, 41, 43, 51–52, 74–79, 81–82, 84–90, 99, 101, 160–163, 173–174, 205 J Jianghuai Mandarin dialects 108 K Kam-Tai 18–20, 29–30, 229 L Language attitude 143 Language change 51, 174 Language contact 10, 28, 32, 37–38, 107, 125, 133, 135–136, 171, 174 Language families 17, 20 Language shift 139 Lingua franca 10, 137, 140 Linguistic area 20, 34, 37, 88, 173–174 Linguistic areas 20
Linguistic change 11, 55–56, 77, 136, 145, 173 Linguistic contact 33–34, 143 Loan feature 35, 135 Loan form 107, 125 Low languages 140 M Manchu 33–34 Mandarin dialects 27–29, 33, 105, 107–108, 110, 114–115, 117, 223 Marked double-object construction 11 Meaning extension 37, 40 Miao-Yao 18, 20, 29, 232 Min dialects 23, 29, 88, 99, 108, 214 Modern Standard Chinese ix, 136–137, 140–141, 143–147, 149, 154–155, 157, 159, 167–168, 171–174 Mon Khmer 20 Monotransitive construction 13 Mother tongue 136, 138, 141, 143 Mother tongue education 141 N Native pattern 10, 112, 119, 125–126, 134–135, 144, 147, 150, 153, 155, 158, 160–161, 163, 173 Native patterns 106, 122, 125, 128, 135, 147, 150, 157–159, 163, 170–172 Neutral question 56, 174 Neutral questions 56 New information 129, 153 Non-native double-object patterns 158–159, 171–172 Non-native patterns 135, 147, 157–159, 163, 170–172 Non-native structures 136 Non-Sinitic languages 15–18, 20, 28–29, 33, 35, 37–38, 122, 173, 205, 228 Northern dialects 10, 18, 29, 33, 37, 106–107, 110, 114, 118–119, 128, 144, 173
Index 261
O Official language 10, 128, 139–140 Old information 129 On-going linguistic changes 129 On-going syntactic change 11, 159 Oracle-bone inscriptions 87, 89, 173 Overt IO marker 81, 160–161 P Passive construction ix, 35, 46–48, 52, 92, 99, 101 Passive marker ix, 14–15, 33–38, 45, 48, 51, 90–95, 97–101, 174, 205 Passive morphology 48 Perception task 146, 160, 168–172, 255 Permission 44, 47–48, 52, 100, 170 Permissive 15, 34–35, 44, 47–48, 51–52, 73, 97, 101, 158 Permissiveness 15, 44 Polysemy 37, 52 Polysemies 40 Possession 5, 44, 52, 86, 88–89, 126 Possessive construction 166 Possessive noun phrase 165–166, 171 Possessive noun phrases 165–166 Post-verbal gei-phrase 123 Pre-modern Cantonese dialect materials 58–59, 92, 235, 239 Pre-modern dialect materials 58, 60, 78, 92, 135 Pre-verbal gei-phrase 123 Production task 145–146, 149, 158, 160, 164, 168–169, 171–172, 247, 261 Putonghua ix, 10, 110, 122, 125, 128, 135–147, 149, 154–155, 157–159, 167–168, 171–174
R Rate of change 129, 172 Real time approach 56 Reanalysis 51, 100 Receive 49, 51–52 Recipient 5–10, 12–14, 41–44, 52, 85–87, 89–90, 100, 115–119, 157, 174 Recipient language 174 Relative chronology 55 S Semantic attributes 51, 89 Semantic constraint 127 Semantic content 40, 101 Semantic extension ix, 86–87, 89 Semantic feature 6, 12, 42, 44, 82, 118, 126–127, 144 Semantic map 52–53 Semantic role 9, 41–42, 46, 49, 114–118, 174 Semantic-pragmatic tendencies 100 Sentence processing 100, 131 Serial verb construction 11–13, 43 Serial verb languages 41 Sino-Tibetan language family 18, 20 Social variables 143–145 Sociolinguistic alignment 138, 141 Source 6, 116 Source language 174 Southeast Asian languages 17, 19–20, 34, 36, 38 Southeast Asian linguistic area 20, 34, 37, 88, 173–174 Southern Chinese dialects 19, 37 Southern dialects ix, 10, 18, 34– 35, 105–110, 112, 114, 119, 122, 135 Southwestern Mandarin dialects 108 Spatial transfer 86, 89 Standard Mandarin 11–12, 14, 33, 41–42, 45, 48–49, 110, 119, 122, 125–126, 128, 130, 135–136, 144, 165
Stratification 105–107, 109–110 Sub-categorization 12 Sub-categorize 6 Supreme language 139–140 Surface structure 14, 42, 46, 51, 82, 97–98, 101, 158, 169 Synchronic ix, 53 Syntactic change 82, 98, 100, 133, 144, 159, 171 Syntactic strata 123 Syntactic stratification 105, 109–110 T Tai-Kadai 17, 20 Tanka people 137 Theme 7–8, 89 Thing 5–8, 10, 13–14, 41, 44, 52, 85, 134–135 Three-argument structure ix, 3 Three-participant clauses 44 Three-participant event 50 Three-place predicate 7–8, 47 Three-place predicates 7 Tibeto-Burman 17–20, 29, 34–35, 228 Transfer of possession 86, 88–89 Transformational approach 126 Triglossic hierarchy 139 Triglossic model 139–140 Typological contrast ix, 122 Typological differences 18, 107, 119, 144, 174 Typological patterns 109, 118, 173 Typological significance 86, 112 Typologically distinct patterns 106, 125 Typology 10, 48, 105–107, 119 U Underlying form 126 Underlying structure 82, 126 Unmarked double-object construction 11 Unwilling permission 47
262 Cantonese give and Double-Object Construction
V V DO IO pattern 77, 78, 81, 127, 131, 133, 162, 167 V DO 畀 IO pattern 77, 78, 81, 146, 154, 160, 161, 165, 167 V IO DO pattern 77, 111, 114, 131, 133–135, 146, 150–158, 163–168, 171, 172 V 畀 IO DO pattern 133–135, 156–159, 163, 164, 167, 171
W Wang Shuo 112–115, 131, 237 Word order 10–11, 14, 105, 107, 110, 112–113, 115, 118–122, 125, 127, 129, 131–133, 144, 154, 157, 166–167, 173 Word order change 105, 137 Word order typology 10, 105 Wu dialects 26, 108, 221 X Xiang dialects 25, 28, 109–110, 219
Y Yue dialects 19, 21, 28–30, 32, 42, 50, 56–58, 119–120, 135, 144, 205 Z Zhongyuan Mandarin dialects 107 Zhuang 19–20, 30–32, 122 畀 畀 IO V DO pattern 127, 128, 133, 157, 158, 169–171
GIVE is a versatile morpheme in many languages. While there have been extensive studies on the interplay between the syntax and semantics of GIVE in many languages, not much has been done in a similar manner on Cantonese, a member of the Yue dialect group of the Chinese language family. This monograph reports on the study of GIVE and its associated functions and syntactic constructions in Cantonese from diachronic, synchronic, and typological perspectives. Drawing on cross-linguistic data, and 19th century Cantonese dialect materials, this study first traces the chronological development of the various functions played by GIVE in Cantonese. It then examines the double-object construction. Besides the typological features of this construction in Cantonese, this study investigates the use of the northern pattern in Cantonese as a result of the increasing influence of Putonghua and Modern Standard Chinese by means of a sociolinguistic survey with 40 native speakers of Cantonese.
isbn 978 90 272 1104 0
John Benjamins Publishing Company