The Routledge Encyclopedia of the Chinese Language

SAVE OFFLINE

The Routledge Encyclopedia of the Chinese Language Edited by Professor Chan Sin-wai Assisted by James Minett and Florence Li Wing Yee

The Routledge Encyclopedia of the Chinese Language will truly meet the academic, linguistic and pedagogical needs of those who are interested in the Chinese language in different capacities and for different reasons, such as Sinologists, Chinese linguists, and teachers and learners of Chinese as a second language. The Encyclopedia includes research on the changing landscape of the Chinese language by over 40 eminent academics in the field, from research on oracle bone and bronze inscriptions, to Chinese language acquisition, to the language of the mass media. This reference will be the most up-to-date and authoritative on the market; it will offer a guide to shifts over time in thinking about the Chinese language as well as providing an overview of contemporary themes, debates and research interests. The editors and contributors are assisted by an editorial board comprised of the best and most experienced Sinologists world-wide. The reference includes an introduction, written by the editor, which places the assembled texts in their historical and intellectual context. The Routledge Encyclopedia of the Chinese Language is destined to be valued by scholars and students as a vital research resource. Chan Sin-wai is Professor at the School of Humanities and Social Science, The Chinese University of Hong Kong (Shenzhen), China.

This page intentionally left blank

The Routledge Encyclopedia of the Chinese Language

Edited by Chan Sin-wai

Assisted by James Minett and Florence Li Wing Yee

First published 2016 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2016 selection and editorial matter, Chan Sin-wai; individual chapters, the contributors The right of Chan Sin-wai to be identified as author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data The Routledge encyclopedia of the Chinese language / edited by Chan Sin-wai; assisted by Florence Li Wing Yee and James Minett.   pages cm   Includes bibliographical references and index.   1.  Chinese language–Encyclopedias.  2.  Chinese philology–Encyclopedias.  3.  Chinese characters– Encyclopedias.  4.  Chinese language–Acquisition.  I.  Chan, Sin-wai, editor.  II.  Li Wing Yee, Florence, editor.  III.  Minett, James W., editor.   PL1031.R68 2015   495.1′3–dc23 2015017474 ISBN: 978-0-415-53970-8 (hbk) ISBN: 978-1-315-67554-1 (ebk) Typeset in Times New Roman by Graphicraft Limited, Hong Kong

Contents

List of Tables ix List of Figures xiii List of Consultant Editors xv List of Contributors xvi The Chinese Language: The Global, Historical, and Linguistic Aspects xxix Chan Sin-wai Acknowledgements xxxviii Routledge Encyclopedia of the Chinese Language   1 Ancient Chinese Alain Peyraube

1

 2 Cantonese Cheng Siu-Pong and Tang Sze-Wing

18

  3 Cantonese Romanization Cheng Siu-Pong and Tang Sze-Wing

35

  4 Chinese Characters John Jing-hua Yin

51

  5 Chinese Idioms Jiao Liwei

64

  6 Chinese Language and National Identity Andrew Simpson

90

  7 Chinese Language Education: Teacher Training Jane Orton v

104

Contents

  8 Chinese Language in a Global Context Liu Jin and Tao Hongyin

119

  9 Chinese Language Pedagogy Wu Weiping

137

10 Chinese Linguistics William S.-Y. Wang

152

11 Chinese Linguistics: Pragmatics Yan Jiang

184

12 Chinese Linguistics: Semantics Hsieh Shu-Kai

203

13 Chinese Morphology Jerome L. Packard

215

14 Chinese Neologisms: Word-formation Strategies in Chinese Antonella Ceccagno

227

15 Chinese: Parts of Speech Candice Chi-Hang Cheung

242

16 Chinese Proverbs and Popular Sayings Larry Herzberg

295

17 Chinese Psycholinguistics Jerome L. Packard

315

18 Chinese Rhetoric Andy Kirkpatrick

328

19 Chinese Slang Robert L. Moore

342

20 Chinese Syntax Walter Bisang

354

21 Chinese Taboo Amy He Yun

378

22 Chinese Xiehouyu 395 Grace Zhang 23 Classical Chinese Chris Wen-chao Li

408 vi

Contents

24 Computational Linguistics Xiaoheng Zhang

420

25 Corpus-based Study of Chinese Tony McEnery and Richard Xiao

438

26 Elastic Words in Chinese San Duanmu and Yan Dong

452

27 Hakka Dialect Lau Chun Fat

469

28 Hanyu Pinyin Lilly Lee Chen

484

29 Hong Kong Sign Language Gladys Tang

505

30 Lexicography Cheung Kam-Siu

531

31 Lexicography in the Contemporary Period Huang Chu-Ren, Li Lan, and Su Xinchun

545

32 Loanwords Miao Ruiqin

563

33 Mandarin Shi Dingxu

579

34 Mandarin and Other Sinitic Languages Hilary Chappell and Li Lan

605

35 Metaphor in Chinese: Cognition, Culture, and Society Zhuo Jing-Schmidt

629

36 Modern Chinese: Written Chinese Feng Shengli

645

37 Poetic Prosody Feng Shengli

664

38 Psycholinguistics: Reading Chinese Liang Tao and Alice F. Healy

685

39 Semantic Change in Chinese Janet Zhiqun Xing

706 vii

Contents

40 Standard Chinese Shi Dingxu

723

41 Syntax Li Yafei

736

42 Wade–Giles Romanization System Karen Steffen Chung

756

Index 777

viii

List of tables

1.1 37 Old Chinese initials 2 1.2 The six main vowels 2 Old Chinese codas 2 1.3 1.4 Middle Chinese initials 3 Middle Chinese main vowels 3 1.5 1.6 Middle Chinese codas 3 Initial consonants 47 3.1 3.2 Final consonants 47 48 3.3 Vowels 3.4 Tones 48 Ancient writing period 54 4.1 4.2 Modern writing period 54 Categories of Chinese character construction 56 4.3 4.4 Number of words and characters at each proficiency level 58 Chinese character strokes 60 4.5 5.1 Competitive idioms that mean ‘the difference/gap is large’ 83 7.1 ACTFL Program Standards for the Preparation of Foreign Language Teachers 108 9.1 ACTFL levels with sample linguistic functions 144 9.2 Summary of pragmatic points in the system of language use 146 10.1 Song of the Yue Boatman 156 10.2 Simplified chronology of Chinese history 161 10.3 Guan Ju, first poem of Shijing 165 10.4 Comparison of kanji, zhengti and jianti 171 172 10.5 Comparison of traditional and simplified sinograms 10.6 Comparison of three Romanization systems 173 174 10.7 Pinyin and some PTH obstruents 10.8 The vowel Spelled ‘i’ in Pinyin 175 175 10.9 Shapes of the Putonghua syllable 14.1 期 qī as a compound constituent and as a metacompound constituent 231 15.1 The list of prepositions that can be followed by -zhe 着 261 ix

List of tables

15.2 Prepositions functioning as prepositions and as verbs 263 15.3 Localizers in Chinese 280 403 22.1 Xiehouyu and others 24.1 Chinese font sizes in numbers, points, and mm 426 433 24.2 Rule-based English–Chinese machine translation with syntactic analysis 25.1 HSK graded lists and words and characters in Chinese 440 440 25.2 Coverage of top N words 25.3 A summary of the roles of corpora in lexicography in Chinese 442 452 26.1 Examples of elastic words in Chinese 26.2 Monosyllabic–trisyllabic pair 453 453 26.3 Either part of the long form can be the short form 26.4 More than one long form 453 453 26.5 An ordinary Chinese sentence and its English translation 26.6 Object required in a Chinese answer 453 454 26.7 Object not required in an English answer 26.8 Questions to be discussed with regard to elastic words 454 455 26.9 Semantic difference between short and long forms 26.10 Defining elastic words in Chinese 455 455 26.11 Three interchangeable expressions for ‘skilled worker’ 26.12 Mutual dictionary annotation of 煤 ‘coal’ and 煤炭 ‘coal-charcoal’ 456 456 26.13 Word pairs that do or do not form elastic words 26.14 Quantitative studies on the percentage of elastic words in Chinese 456 457 26.15 POS and length properties of Chinese morphemes 26.16 Example: monosyllabic word without elastic length 457 458 26.17 Morphologically and semantically related word pairs in English 26.18 Theories on why Chinese has so many elastic words 458 459 26.19 Predictions of the homophony theory and the prosody theory 26.20 Minimal word requirement in Chinese when there is no ambiguity 460 26.21 Homophone density and the percentage of elastic words among noun morphemes in Chinese. Nouns that are polysyllabic only are excluded. 462 A homophone density of 1 means a word has no homophone (but itself ) 26.22 Statistics of the data in Table 26.21, which show no correlation between 463 homophone density and the percentage of elastic words. 26.23 In [N N], 1+2 is disfavored 463 463 26.24 In [V O], 2+1 is disfavored 26.25 Phonological requirements (boldface indicates stress) 463 463 26.26 Analysis of [N N] 26.27 Analysis of [V O] 463 26.28 Word length patterns for [N N] in written Chinese (token frequencies) 464 26.29 Word length patterns for [V O] in written Chinese (token frequencies) 464 465 26.30 A comparison of how many things each theory can correctly account for 26.31 Elastic words created by truncation 465 466 26.32 Repetitive words (indicated by parentheses) in two Chinese sentences 26.33 A poem with extra positions to fill (three missing syllables per line) 466 466 26.34 Elastic words can fill extra positions 27.1 Monosyllabic Hakka words which can be traced back to Middle or even 473 Ancient Chinese 27.2 Onsets (18) 478 x

List of tables

27.3 Rimes (75) 478 27.4 Tones 479 480 27.5 Some lexical differences between Meixian and Hong Kong Hakka 27.6 Tones of the Bendihua group 480 Some examples of correspondence between the Gutian and Meixian rime 27.7  systems 482 482 27.8 The Gutian and Meixian tone systems compared 28.1 First-grade textbook 485 486 28.2 Xīzì Qíjì 28.3 Chinese syllable structure 491 492 28.4 Table of phonetic letters 28.5 Initial consonants 493 493 28.6 Finals 28.7 Tone marks 493 494 28.8 Tone mark placement 28.9 Vowel chart 494 494 28.10 Consonant chart 28.11 Four tones 495 497–498 28.12 Pinyin Chart 28.13 Comparison of retroflexs and alveolars in different systems 500 501 28.14 The use of the letter h in Wade–Giles and Pinyin 28.15 Mandarin homophonous words qishi and their correspondences 501 in Southern Min 28.16 Homophonous characters for qi [tɕ’i] 502 554 31.1 A list of Chinese corpora 31.2 The first Chinese–foreign language bilingual dictionaries 557 570 32.1 [Phonemic, semantic] hybrid loans 32.2 Mandarin substitution patterns of English plosives 572 573 32.3 Lack of stress-to-tone mapping 32.4 Native-likeness hierarchy of loanwords 575 575 32.5 Regional variations 33.1 Consonants of Mandarin 580 582 33.2 Glides of Mandarin 33.3 Vowels of Mandarin 582 584 33.4 Actual syllables with a glide as the onset 33.5 Initials of Pinyin 587 587 33.6 Finals of Pinyin 33.7 Mandarin pronouns 590 607 34.1 Sinitic languages of China 34.2 The eight dialect subgroups of Mandarin 609 610 34.3 Standard Mandarin tones 34.4 Pronouns of Standard Mandarin 610 34.5 Tone system in the Huojia dialect of Jin 获嘉方言 612 34.6 Pronoun paradigm for the Shanyin dialect of Jin (Shaanxi) 612 613 34.7 Changsha Xiang tones 34.8 Pronouns of Changsha Xiang 613 34.9 Pronominal paradigm: Longhui Xiang 隆回话 613 34.10 Tone system of Nanchang Gan 614 xi

List of tables

34.11 Pronouns of Nanchang Gan (Jiangxi province) 34.12 Singular personal pronouns in the Gan dialect of Qianshan 34.13 Tone system of the Jixi dialect of Hui (Anhui) 34.14 Jixi pronominal paradigm (Anhui) 34.15 Tone system of Shanghainese Wu 34.16 Pronouns of Shanghainese 34.17 Southern Min tone system (Taiwan) 34.18 Pronouns of Southern Min (Taiwan) 34.19 Tone system of Sung Him Tong Hakka (New Territories, Hong Kong) 34.20 Hakka pronominal paradigm (Bao’an, Huizhou subgroup) 34.21 Tone system of Hong Kong Cantonese 34.22 Examples of Cantonese tonal syllables 34.23 Cantonese pronominal paradigm (Hong Kong) 34.24 Tone system of the Tingzicun 亭子村 dialect of Southern Pinghua 34.25 Nanning pronominal paradigm 39.1 Summary of the polysemies of 23 lexemes 40.1 Shengmu 40.2 Yunmu 42.1 Some of Morrison’s spellings and descriptions of the phonetic values

xii

615 615 616 616 617 618 619 620 621 621 622 622 623 624 624 714 728 728 761

List of figures

Character component writing sequence 61–62 9.1 Major steps in CSL teaching syllabi focusing on language structure (LS) 138 9.2 Major steps in CSL teaching syllabi focusing on language use (LU) 143 9.3 Guiding structure in teaching materials preparation 147 Austronesian languages in Taiwan 157 10.1 10.2 The ‘initial China’ 160 Oracle bone inscription concerning Fuhao’s childbirth 162 10.3 10.4 A page from a rhyme book of the Song dynasty 168 Major dialects of Hanyu 170 10.5 12.1 The pyramid structure model 207 Syntagmatic relations from Chinese Word Sketch Engine 209 12.2 Chinese WordNet 210 12.3 12.4 Morpho-semantic linkage space 211 21.1 Pyramid hierarchy of words for death of people from different social classes in historical Chinese 386 24.1 Chinese character frequency of Hong Kong, mainland China, and Taiwan 431 A sample of a concordance for keyword ‘中文’ (Chinese) 432 24.2 25.1 Coverage of top N characters 441 Distribution and classification of Hakka 470 27.1 28.1 Beijing street sign 484 508 29.1a BEE (variation 1) 29.1b BEE (variation 2) 509 509 29.1c BEE (variation 3) 29.1d BEE (variation 4) 509 29.2a PERSON 人 510 29.2b FAIR 公平 510 29.2c INTRODUCE 介绍 510 29.3a BREAK_OFF_A_RELATION 511 29.3b INCOMPATIBLE 511 29.4a PIG 512 512 29.4b STUBBORN xiii

List of figures

29.5a BE_DETERMINED 512 29.5b SET_OFF 512 513 29.6 THANK_YOU 29.7a scold+CL_sem:human_entity 513 513 29.7b SCOLD 29.8a TEN 514 514 29.8b DIVIDE_EVENLY 29.8c DECREASE 514 515 29.8d INCREASE 29.9a FAR (X-plane) 515 516 29.9b CLOUD (Y-plane) 29.9c SIGN LANGUAGE (Z-plane) 516 516 29.9d FACE-PRETTY 29.9e HEALTHY 516 517 29.10 LOVE-SOMETHING-ARDENTLY 29.11a STOP 518 518 29.11b TRY 29.11c DOCTOR 518 518 29.11d SHARE 29.12a THINK and OLD (citation form) 519 519 29.12b THINK^OLD (assimilated thumb) 29.13 LEARN^HOUSE 520 520 29.14 PIG^MEAT 29.15 ROCKET 521 522 29.16a NOT_FEEL-ANNOYED 29.16b UNFAIR 523 29.17a stand+CL_sem:human 524 29.17b lie+CL_sem:human 524 Shuowen jiezi 532 30.1 30.2 Guangyun 536 Zihui 537 30.3 30.4 Kangxi zidian 538 The first tone 585 33.1 33.2 The second tone 585 The third tone 585 33.3 33.4 The fourth tone 585 Diagram of formal and informal Chinese 646 36.1 36.2 Modern formal Chinese 651 Tridimensional system of stylistic-register grammar 654 36.3 36.4 Interrelationships among elevated, formal and casual forms 655 Evidence for tridimensional register grammar 655 36.5 36.6 660 Stylistic wave (of formal Chinese) Visual differences between a Chinese and an English sentence 687 38.1

xiv

List of Consultant Editors

William S.-Y. Wang Department of Chinese and Bilingual Studies Hong Kong Polytechnic University, Hong Kong, China Huang Chu-Ren Department of Chinese and Bilingual Studies Hong Kong Polytechnic University, Hong Kong, China Feng Shengli Department of Chinese The Chinese University of Hong Kong, Hong Kong, China Cao Guangshun Institute of Linguistics Chinese Academy of Social Sciences, China

xv

List of contributors

Bisang, Walter Walter Bisang Dr. phil. I (Zürich) has been Professor of General and Comparative Linguistics at the University of Mainz (Germany) since 1992. He was the Director of a Collaborative Research Center on ‘Cultural and Linguistic Contact’ from 1999 to 2008 in Mainz. His research interests focus on linguistic typology, grammaticalization, language contact/areal typology, and the comparison of different theoretical approaches to language. His languages of interest are East and mainland Southeast Asian languages, Caucasian languages (Georgian and others), Austronesian languages (Bahasa Indonesia, Tagalog, Yabêm, Paiwan), and Yoruba (together with Remi Sonaiya). Ceccagno, Antonella Antonella Ceccagno is associate professor of Chinese Linguistics at the University of Bologna, Italy, where she also teaches Sociology of East Asia. In the field of Chinese Linguistics her interests revolve around Chinese morphology. More recently, she is also focusing analytical attention on the interactions between language and society. Her publications on Chinese linguistics and sociology of language include: ‘Farewell to tuhao, welcome to tuhao: Language and Society in China as they Emerge from the Buzzwords of the Last Decade’ 2014, in Magda Abbiati and Federico Greselin (eds) Il Liuto e i libri, Venezia, Cafoscari, 193–204; ‘Sino-Tibetan: Mandarin Chinese’ in Rochelle Lieber and Pal Stekauer (eds) The Oxford Book of Compounding, Oxford: Oxford University Press, 2009, 478–90, (with Bianca Basciano); ‘Metacompounds in Chinese’, Lingue e Linguaggio 8(2), 2009: 195–212; ‘The Chinese Language and Some Notions from Western Linguistics’, Lingue e Linguaggio, 8(1), 2009: 105–35 (with Bianca Basciano); Shuobuchulai, la formazione delle parole in cinese (‘Shuobuchulai (Word Formation in Chinese)’), Bologna: Serendipità, 2009 (with Bianca Basciano); ‘Compound Headedness in Chinese: An Analysis of Neologisms’, Morphology, n 17, 2007: 207–31 (with Bianca Basciano); ‘Classification, Structure and Headedness of Chinese Compounds’ in Lingue e linguaggio 2, 2006: 233–60 (with Sergio Scalise). Chan, Sin-wai Chan Sin-wai is Professor at the School of Humanities and Social Science, The Chinese University of Hong Kong, Shenzhen. His teaching and research interests lie mainly in the areas of translation studies, translation technology, and bilingual lexicography. He is the xvi

List of contributors

Chief Editor of Journal of Translation Technology, published by The Chinese University of Hong Kong. He has published more than 44 books in 54 volumes, mainly dictionaries and scholarly monographs, and translated works in different fields. He edited An Encyclopaedia of Translation and The Routledge Encyclopedia of Translation Technology (2015), revised Longman Dictionary of English Language and Culture (bilingual edition), and authored A Dictionary of Translation Technology and A Chinese–English Dictionary of the Human Body. His book translations from Chinese into English include An Exposition of Benevolence, Palaces of the Forbidden City, Letters of Prominent Figures in Modern China, Paintings and Calligraphy of Jao Tsung-I, Stories by Gao Yang, An Illustrated History of Printing in Ancient China, Famous Chinese Sayings Quoted by Wen Jiabao, and Selected Works of Cheng Siwei: Economic Reforms and Development in China, Vol. 2. He also translated My Son Yo Yo from English into Chinese. His most recent co-edited books include Style, Wit and Word-Play (2012) and The Dancer and the Dance (2013), both published by Cambridge Scholars Publishing. Chappell, Hilary Hilary Chappell is currently Chair Professor in Linguistic Typology of East Asian Languages at the Ecole des Hautes Etudes en Sciences Sociales in Paris, an appointment she took up in 2005 after teaching in the Linguistics Department at La Trobe University in Melbourne, Australia, for 18 years. She was originally awarded her doctoral degree in 1984 by the Australian National University in Canberra for her thesis entitled ‘A Semantic Analysis of Passive, Causative and Dative Constructions in Standard Chinese’ and has over 60 publications on Chinese linguistics and typology, including four edited books and a volume on Hakka, co-authored with Christine Lamarre. Her main research is on the typology of Sinitic languages (Chinese dialects) and the extent of their linguistic diversity, for which she directed an ERC Advanced Grant project from 2009 to 2013. She has also held two projects with Professor Alain Peyraube (CNRSEHESS) to investigate the diachronic syntax of Southern Min (Hokkien), using a corpus of late sixteenth and early seventeenth century materials. In 2010, she was elected member of the Academia Europaea. Chen, Lilly Lee Lilly Lee Chen received her doctoral and master’s degrees in linguistics from the University of Illinois, Urbana-Champaign, and her bachelor’s degree from the Department of Foreign Languages and Literatures at National Taiwan University. She taught for 30 years in the Department of Linguistics and the Center for the Study of Languages at Rice University. In the course of her teaching career at Rice she established a variety of language oriented new programs and taught linguistics and Chinese language and culture, including synchronic and diachronic syntax, structure and symbolism of the classical novel, Chinese cinema, and Taiwanese language, culture, and linguistics from a cognitive perspective. She has served as President of the Linguistic Association of Canada and the United States (2013), President of the Chinese American Educational Research and Development Association (2000), and Founding President of the Chinese Language Teacher Association of the South (2006). She was also the Founding Director of the Institute for Chinese Language Teacher Training (2007–2010). Cheng, Siu-Pong Cheng Siu-Pong is Research Assistant at T. T. Ng Chinese Language Research Centre of the Institute of Chinese Studies, The Chinese University of Hong Kong. His research interests focus on Cantonese discontinuous constructions, the cartographic approach, and empty categories. xvii

List of contributors

Cheung, Chi-Hang Candice Candice Chi-Hang Cheung is Research Assistant Professor in the Department of Chinese and Bilingual Studies at the Hong Kong Polytechnic University. Her research interests lie in formal syntax, the syntax-semantics interface, the syntax-information structure interface, and parametric syntax with special focus on Cantonese, Mandarin and other varieties of Chinese. She has published articles in a number of linguistic journals, such as the Journal of East Asian Linguistics and Natural Language and Linguistic Theory. Cheung, Kam-Siu Cheung Kam-Siu received his PhD in Chinese Language and Literature from The Chinese University of Hong Kong in 2007. He is Associate Professor in the Department of Chinese Language and Literature and Honorary Research Associate of D. C. Lau Research Centre for Chinese Ancient Texts at The Chinese University of Hong Kong. His research interests focus on Chinese philology, ancient Chinese texts studies, and lexicography. He has recently published two books on the research of the prominent Qing semanticist and textual critic Wang Niansun (1744–1832), including On Wang Niansun’s Collated Edition of Ancient Texts (Shanghai Ancient Books Publishing House, 2014), and Collected Annotations on Wang Niansun’s Collated Edition of the Hanfeizi (The Chinese University Press, 2014). He is also an editorial board member of Newsletter of Chinese Language. Chung, Karen Steffen Karen Steffen Chung (史嘉琳 Shǐ Jiālín), originally from St. Paul, Minnesota, USA, has taught English, linguistics, and phonetics in the Department of Foreign Languages and Literatures of National Taiwan University since 1990. She is currently an Associate Professor at National Taiwan University. She received her BA in East Asian Languages from the University of Minnesota in 1976, her MA in East Asian Studies from Princeton University in 1981, and obtained her PhD in Linguistics from Leiden University in 2004. The title of her doctoral dissertation is: ‘Mandarin Compound Verbs’. Dong, Yan Yan Dong received her MA in Linguistics and Applied Linguistics from Renmin University of China in 2008 and joined the PhD program in Linguistics at the University of Michigan in the same year. In July 2015 she recently defended her dissertation, which is entitled ‘The prosody and morphology of elastic words in Chinese: annotations and analyses’. Duanmu, San San Duanmu is Professor of Linguistics, University of Michigan. He received his PhD in Linguistics from MIT in 1990 and has held teaching posts at Fudan University, Shanghai (1981–1986) and the University of Michigan, Ann Arbor (1991–present). He is the author of The Phonology of Standard Chinese (2nd ed., Oxford, 2007) and Syllable Structure: The Limits of Variation (Oxford, 2008). Feng, Shengli Feng Shengli began his academic career as an undergraduate student in the History Department of Beijing Normal University in the class of 1977. He was accepted by the Graduate Program in the Department of Language and Literature at the Beijing Normal University in 1979, majoring in Ancient Chinese Exegesis under the supervision of Professor Lu Zongda. After completing his MA degree, he taught Classical Chinese and Language and Literature in the xviii

List of contributors

Department of Language and Literature at Beijing Normal University. He received his PhD in Linguistics from the University of Pennsylvania in 1995. He taught Chinese language, Chinese linguistics, and Chinese culture in the Department of East Asian Languages and Cultures at the University of Kansas from 1994, and in 1999 he was promoted to Associate Professor. He was appointed in 2003 as Professor of the Practice and Director of the Chinese Language Program at Harvard University and Director of Harvard-Beijing Academy. In 2005 he was appointed as Yangtze Scholar Adjunct Professor (长江学者讲座教授) at Beijing Language and Culture University. He is the author of The Prosody, Morphology, and Syntax of Chinese 漢語的韻律、詞法與句法 (1997), Prosodically Constrained Syntactic Changes in Early Archaic Chinese (1996), The Prosodic Syntax of Chinese (2002), and A Preliminary Theory of Chinese Poetic Prosody 漢語的韻律詩體學論稿 (2015). He has edited Written Chinese: the Present and the Past (2013), New Exploration on Chinese Prosodic Grammar (2015). He currently serves as associate editor for Language and Linguistics. He, Yun Amy Amy He Yun is a visiting research fellow in the Centre for Intercultural Politeness Research at the University of Huddersfield, the United Kingdom, where she has previously worked as a postdoctoral research associate. She obtained her PhD degree from Loughborough University for studying politeness in Mandarin Chinese, co-funded by an Overseas Research Studentship Award (Higher Education Funding Council for England), University Research Studentship, and Great Britain–China Educational Trust: Chinese Student Award. Her recent interests are in (im)politeness and face, but her research cuts across pragmatics, discourse analysis, sociolinguistics, intercultural communication, and translation studies. She has completed five research projects and has published widely in the aforementioned areas of research, including ‘Different generations, different face?’ (Journal of Politeness Research 8(1), 2012) and a chapter on identity in Pragmatic explorations in identity and communication (ed. Chen, X. Beijing: Higher Education Press, 2013). She is currently working on a book with the provisional title Generational Variation of Politeness in Contemporary Chinese. Healy, Alice F. Alice F. Healy is College Professor of Distinction and Director of the Center for Research on Training at the University of Colorado Boulder. She received her doctorate from the Rockefeller University in 1973, and she was on the faculty of Yale University from 1973 to 1981, when she joined the faculty of the University of Colorado. Dr. Healy has served as Editor of the journal Memory and Cognition, Chair of the Psychology Section of the American Association for the Advancement of Science, President of the Rocky Mountain Psychological Association, President of the Division of Experimental Psychology (Division 3) of the American Psychological Association (APA), and Chair of the Society of Experimental Psychologists. She has published over 250 journal articles and book chapters, is co-author of Cognitive Processes (Prentice Hall, 1986) and Train Your Mind for Peak Performance (APA, 2014), is editor of the volume Experimental Cognitive Psychology and its Applications (APA, 2005), and is senior editor of the two-volume Essays in Honor of William K. Estes (Erlbaum, 1992), Learning and Memory of Knowledge and Skills (Sage, 1995), Foreign Language Learning (Erlbaum, 1998), the Experimental Psychology volume of the Handbook of Psychology (Wiley, 2003 and 2013), and Training Cognition (Psychology Press, 2012). Herzberg, Larry Larry Herzberg has been a professor of Chinese at Calvin College in Grand Rapids, Michigan for the past 30 years, where he is also Director of Asian Studies. After studying Chinese for xix

List of contributors

five years at Vanderbilt University, he did his PhD work in Chinese language and literature at Indiana University. Together with his Chinese wife, Xue Qin, he has published three books, including The Chinese Survival Guide, Basic Patterns of Chinese Grammar, and Chinese Proverbs and Popular Sayings. He and his wife have also produced three films on China, including ‘China Today: Issues that Trouble Americans at the Start of the 21st Century’, ‘The China Threat: Perception versus Reality’, and ‘Chinese Dialects and Their Influence on the Pronunciation of Mandarin’. In 2011 he received the Presidential Award for Exemplary Teaching, the highest honor bestowed on faculty members by Calvin College. Hsieh, Shu-Kai Hsieh Shu-Kai is an associate professor in the Graduate Institute of Linguistics at National Taiwan University, Taiwan. He received his PhD in Computational Linguistics from the University of Tübingen in Germany. He has chaired and served on several program committees for linguistics at international conferences, and has given a number of invited talks on computational lexical semantics, corpus, and ontology at different universities. His primary research interests include computational lexical semantics and ontology. His publications include two books on computational treatment of Chinese characters and translation of the Generative Lexicon, and more than 30 refereed research papers published in journals, books, conferences, and workshops. He is a member of the editorial board and the linguistic section chair of IJCLCLP, an international journal of computational linguistics and Chinese language processing. He is a member of the council of the Linguistics Society of Taiwan and Global WordNet Association. Huang, Chu-Ren Huang Chu-Ren is a Chair Professor of Applied Chinese Language Studies at the Hong Kong Polytechnic University. He is Fellow and President of the Hong Kong Academy of the Humanities, a permanent member of the International Committee on Computational Linguistics, and a former President of the Asian Association of Lexicography. His main research areas are ontology, corpus and computational linguistics, lexical semantics, Chinese grammar, language resources, and digital humanities. The conference series in which he plays advisory and/or organizing roles include ALR, ASIALEX, CLSW, CogALex, COLING, IsCLL, LAW, OntoLex, PACLIC, ROCLING, and SIGHAN. His research output includes 12 licensable language resources, ten searchable online language databases, 20 books or edited volumes, over 130 journal articles or book chapters, and over 380 refereed conference papers. He is currently Chief Editor of the Journal Lingua Sinica, and the book series: Studies in Natural Language Processing and The Humanities in Asia. He received a Doctor honoris causa from Aix-Marseille University in 2013. Jiang, Yan Yan Jiang received his BA in English Language and Literature and his MA in modern English from Fudan University, Shanghai. He obtained his PhD in linguistics from London University. He is now a lecturer in linguistics and the languages of China at the Department of Linguistics, School of Oriental and African Studies, University of London. He has published studies on the Chinese adverb of scope ‘dou’, abduction and pragmatic inference, dexis and anaphora, Chinese counterfactuals, and quantification in Chinese. His book-length publications are listed below: Jiang Yan and Pan Haihua (1998, 2005) Introduction to Formal Semantics, China Social Science Press; Jiang Yan (2008) Translation of Dan Sperber and Deirdre Wilson: Relevance: Communication and Cognition, China Social Science xx

List of contributors

Press; and Jiang Yan (2011) (ed.) Approaching Formal Pragmatics, Shanghai Educational Publishing. Jiao, Liwei Jiao Liwei has a PhD in Chinese Linguistics from Nankai University, China. Since 1998 he has taught at Renmin University of China, the University of Durham, Princeton in Beijing, and the Duke Study in China Program. He is currently a Lecturer in Foreign Languages at the University of Pennsylvania. Among his many publications are 500 Common Chinese Idioms (co-authored, 2010), 500 Common Chinese Proverbs and Colloquial Expressions (co-authored, 2014), and A Cultural Dictionary of the Chinese Language: 500 Proverbs, Idioms and Maxims (forthcoming, 2017), all with Routledge, as well as Chinese Phonetics in the 20th Century (co-authored, 2004) with Shuhai Press, China. He is the Chinese translator of Professor Yuen Ren Chao’s oral history: Chinese Linguist, Phonologist, Composer and Author. His research fields include lexicology, relationships between language and culture, and Chinese phonetics. Kirkpatrick, Andy Andy Kirkpatrick is Professor in the Department of Languages and Linguistics, at Griffith University in Brisbane. He has worked in tertiary institutions across Asia including six years at the Hong Kong Institute of Education as Chair Professor. He has been a Visiting Scholar at Peking University and Bei Hang University in Beijing. His translation and commentary of Chen Kui’s Wenze won the International Society for the History of Rhetoric’s inaugural prize for best article. He is the editor-in-chief of the Journal of Multilingual Education and of a book series (with Springer) of the same name. His most recent book is Chinese Rhetoric and Writing: An Introduction for Teachers of Writing, co-authored with Xu Zhichang. Lau, Chun Fat Lau Chun Fat is Professor at the Department of Chinese Language and Literature at Xiamen University, China. He was born in the city but was brought up in a small Hakka village in the middle of the New Territories of Hong Kong. He studied Biology at The Chinese University of Hong Kong and received his Dr. rer. nat. degree at the Free University of Berlin. After spending more than a decade in the laboratory, he changed his career in the mid-1990s, when he found that practically no children were speaking Hakka around him. He received his master’s and PhD in Linguistics from the Polytechnic University of Hong Kong, and went to teach at Xiamen University in 2002. He has published more than 80 papers and authored eight books. His main interests are sociolinguistics, historical linguistics, Chinese dialectology, and documentation of endangered Chinese dialects around Hong Kong, as the whole area has shifted to Cantonese in the past three decades. He is also chairman of the Association of Conservation of Hong Kong Indigenous Languages. Li, Lan (Chapter 31) Li Lan is a Fellow of the Chartered Institute of Linguists, UK, with MPhil and PhD degrees in Applied Linguistics from the University of Exeter. She works as an Associate Professor at the Department of English, Hong Kong Polytechnic University. She is also a committee member of ASIALEX. She has been teaching and conducting research in semantics, lexicology, and bilingual lexicography. Her research interests and publications cover lexicology, lexicography, metaphor studies, computer-mediated professional communication, sociolinguistics, and corpus linguistics. xxi

List of contributors

Li, Lan (Chapter 34) After gaining his PhD in 1995 from the Chinese Academy of Social Sciences (CASS) in Beijing on the subject of ‘Variation in the Tones and Initials of Southwestern Mandarin Dialects’, Li Lan was offered a position at the CASS, where he continues to work as a Senior Research Fellow in the Institute of Linguistics and head of the Dialect Research Section. Apart from many articles on dialectology and typology of Sinitic languages, which have been published in the top-ranking linguistics journals in China such as Zhongguo Yuwen of the Chinese Language) and Fangyan《方言》(Dialects), Li Lan 《中国语文》(Studies   has also published two books on The Guiyang Dialect (1997) and on The Miao (Hmong) Language of Chengbu, Hunan (2004). He won the Wang Li prize at Peking University in 2005 for this latter monograph and has been the recipient of a large number of projects on linguistics funded by the Social Sciences Foundation of China. Li Lan has also undertaken research visits to the Hong Kong University of Science and Technology and to the Centre de Recherches Linguistiques sur l’Asie Orientale in Paris. Li, Chris Wen-chao Chris Wen-Chao Li is Professor of Linguistics at San Francisco State University, where he teaches linguistics, news writing, and translation-interpretation. He received his doctoral degree in General Linguistics and Comparative Philology from Oxford University, where his research focused on Mandarin sound change and Chinese phonology. He is the author of A Diachronically-Motivated Segmental Phonology of Mandarin Chinese (New York: Peter Lang, 1999) and Media Chinese (Taipei: Lucky Books, 2005). His translations and scholarship have appeared in Renditions, The Chinese Pen, Target, Language and Communication, and the Journal of the American Oriental Society. His current research interests include sound change, language contact, diglossia, standardization, phonological translation, Chinese Romanization systems, and the phonological description of Mandarin Chinese varieties. Li, Yafei Li Yafei is Professor of Linguistics at the University of Wisconsin-Madison, USA, and Siyuan Chair Professor (part-time) at Nanjing University, China. Born in Baoding, China, he received a BA in 1982 and an MA in 1985 from Shandong University and a PhD in Linguistics from MIT in 1990. Specializing in syntax and morpho-syntax, he has taught at Brandeis University, Cornell University, University of Wisconsin-Madison, and Nanjing University. His publications include articles (in Language, Linguistic Inquiry, Natural Language and Linguistic Theory, The Linguistic Review, Journal of East-Asian Linguistics,《中国语文》(Studies of the Chinese Language) and《当代语言学》(Contemporary Linguistics)) and two books, X 0: A Theory of the Morphology-Syntax Interface from MIT Press and the co-authored The Syntax of Chinese from Cambridge University Press. Currently, he is working on his third book, The Fact-first Principle, which evaluates the empirical foundation of the Chomskyan model of syntax and articulates novel analyses of facts from multiple languages that have not been accurately or adequately dealt with in the mainstream literature. Liu, Jin Liu Jin is Associate Professor of Chinese at Georgia Institute of Technology. She received her PhD in East Asian Literature and Culture from Cornell University and her MA and BA in Chinese Linguistics from Peking University. Her interdisciplinary research studies contemporary Chinese popular culture and media culture from the perspective of language, sound, voice, and music. She is the author of the book Signifying the Local: Media Productions xxii

List of contributors

Rendered in Local Languages in Mainland China in the New Millennium (Brill, 2013). She co-edited and contributed to the volume Chinese Under Globalization: Emerging Trends in Language Use in China (World Scientific, 2012). She has published articles in journals including Positions: Asia Critique, Modern Chinese Literature and Culture, Journal of Modern Literature in Chinese, Chinese Language and Discourse, and Harvard Asia Pacific Review. McEnery, Tony Tony McEnery is Distinguished Professor of English Language and Linguistics at Lancaster University. He has published a number of books on corpus linguistics including Corpus Linguistics (with Andrew Wilson, 1996), Corpus-Based Language Studies (with Richard Xiao and Yukio Tono, 2006), and Corpus Linguistics, Method Theory and Practice (with Andrew Hardie, 2011). A key feature of his research is that he has used the corpus approach to investigate a wide range of languages including, in collaboration with Richard Xiao, Mandarin Chinese. Tony is a Fellow of the Academy of Social Sciences and is currently Director of the ESRC Centre for Corpus Approaches to Social Science at Lancaster University. Previously he was Director of Research at the UK’s Arts and Humanities Research Council. Miao, Ruiqin Miao Ruiqin is currently Associate Professor of Linguistics in the School of Foreign Languages at Shanghai Jiao Tong University, China. She received her BA in English Language and Literature from Hunan University (1993), MA in English Linguistics and Applied Linguistics from the same university (1996), and PhD in Linguistics from the State University of New York at Stony Brook (2005). Her research focuses on second language acquisition, phonetics/ phonology, language contact, sociolinguistics, and language pedagogy. She has published in refereed journals and edited volumes including Journal of Asian Pacific Communication, ITL-International Journal of Applied Linguistics, and Developments in Applied Linguistics/ Series on Western Research in the Humanities and Social Sciences (Jianguo Ji and Nan Jiang, eds. Beijing: China Renmin University Press, 2007). She has written Foreign Language Listening Comprehension: Theory and Practice (co-authored with Liu Longgen, Beijing: Foreign Language Teaching and Research Press, 2011). She has also edited several English textbooks used in China and presented papers at various international conferences. Moore, Robert L. Robert L. Moore is a professor of anthropology at Rollins College and Director of Inter­ national Affairs for the college’s Hamilton Holt School. He has a BA in anthropology from Tulane University and a PhD from the University of California at Riverside. In 1993–1994 he was a visiting professor on the faculty of Qingdao University in China. He has undertaken research in Hong Kong, Qingdao, and Beijing and published a number of academic articles on linguistics, Chinese culture, and romantic love in non-Western societies. His current research focuses on contemporary youth cultures and family relationships in China. Orton, Jane Jane Orton, PhD, is Director of the Chinese Teacher Training Centre (CTTC) at the University of Melbourne, Australia, where she coordinated modern languages education for 20 years. Jane has an honours degree in Chinese language and literature from the University of Melbourne. She taught at Capital Normal University in Beijing from 1981 to 1983 and has been a frequent guest lecturer, conference speaker, and researcher in China over the ensuing 30 years. The CTTC undertakes research and professional development into Chinese language programs, teacher education, and innovative, research-grounded resource development. Jane’s xxiii

List of contributors

particular research interests are the learning demands of Chinese as a Second Language, especially in oral skill development. Jane is a board member of the association Chinese as a Second Language Research (CASLAR) and a member of the editorial board of the CASLAR Journal. Her recent publications include ‘Comparing teachers’ judgments of learners’ speech in Chinese as a foreign language’, Foreign Language Annals, 47(3), 2014: 507–26.〈澳大利亚 中小学汉语教学的复杂性〉(The Complexities of Teaching Chinese Language in Australian Primary and Secondary Schools)《云南师范大学学报(对外汉语教学与研究版)》(Journal of Yunnan Normal University (Teaching and Research of Chinese as a Foreign Language)) 4, 2014: 1–5; ‘Developing Chinese Oral Skills – A Research Base for Practice’, in Istvan Kesckes (ed.) Research in Chinese as a Second Language, Berlin: Mouton de Gruyter, 2013: 3–26. Packard, Jerome L. Jerome L. Packard is Professor of Chinese, Linguistics and Educational Psychology, and a faculty affiliate at the Beckman Institute for Advanced Science and Technology at the University of Illinois, USA. He received his BA in Chinese Language and Literature from the University of Massachusetts in 1976 and his PhD in Linguistics from Cornell University in 1984. He specializes in Chinese word structure, Chinese psycholinguistics, and Chinese language teaching and learning. He has taught Chinese at the University of Massachusetts, Cornell University, the University of Pennsylvania, and the University of Illinois, and was a Fulbright dissertation scholar in China. He has performed research on Chinese aphasia and dyslexia, the first language acquisition of Chinese, and the acquisition of Chinese as a second language. His current research interests include Chinese sentence processing, sentiment analysis of Chinese texts, and how children in China learn to read and write. He has received several grants and is the author of numerous publications relevant to Chinese language, including the books New Approaches to Chinese Word Formation, The Morphology of Chinese, Chinese Children’s Reading Acquisition (with W. Li and J. Gaffney), and Processing and Producing Head-final Structures (with H. Yamashita and Y. Hirose). Peyraube, Alain Alain Peyraube is currently emeritus Director of Research at the Centre National de la Recherche Scientifique (CNRS, Paris, France), Chair Professor of Chinese Linguistics at the Ecole des Hautes Etudes en Sciences Sociales (EHESS, Paris), and Director of the Institute of Advanced Studies ‘Collegium de Lyon’. He was Adjunct Professor at the University of Science and Technology of Hong Kong (2005–11) and Founding Member of the Scientific Council of the European Research Council (ERC) (2005–13), and he has been Honorary Professor at the Peking University since 2007, and Distinguished Member of the Chinese Academy of Social Sciences since 2009. He became a member of the European Academy (Academia Europaea) in 2006 and Chair of its Linguistics Section in 2013. As a specialist in Chinese historical syntax and linguistic typology of Sinitic languages, he has authored five books and more than a hundred articles on Chinese studies, mainly with respect to Chinese linguistics. His latest research has been done within a broadly functional and cognitive framework from a cross-linguistic perspective. He was conferred the title of Chevalier de la Légion d’honneur in 2008. Shi, Dingxu Shi Dingxu works in the Department of Chinese and Bilingual Studies of the Hong Kong Polytechnic University. He received his MA in applied linguistics from the University of xxiv

List of contributors

Pittsburgh, and MA and PhD in linguistics from the University of Southern California. His research interests are on syntax, typology, language change induced by contact, interface between syntax and semantics, and pedagogical grammar of Chinese. His publications cover a wide range of topics on Chinese in these fields. Simpson, Andrew Andrew Simpson is Professor of Linguistics and East Asian Languages and Cultures at the University of Southern California. His research focuses on the comparative grammar of East and Southeast Asian languages, and issues of language planning. He is the editor of Language and National Identity in Asia and Language and National Identity in Africa (Oxford University Press), and is the joint general editor of the Journal of East Asia Linguistics. Su, Xinchun Su Xinchun is the Head of the College of Humanities and Communications College of Jiageng College of Xiamen University, and also the Head of Language Education and Teaching Material Branch of National Language Resources Monitoring and Research Center (CNLR). His major teaching and research interests cover Chinese lexicology, ancient and modern Chinese vocabulary, cultural lexicology, metrology of lexicology, dictionary and corpus linguistics, terminology, and so on. He has published over 20 books and more than 100 articles on Chinese lexicology studies. Tang, Gladys Gladys Tang is Professor of Linguistics at the Department of Linguistics and Modern Languages, The Chinese University of Hong Kong. She is also Director of The Centre for Sign Linguistics and Deaf Studies. She has been researching sign language and how deaf children acquire language, signed and spoken. In recent years, she has also embarked on developing a model of education that benefits both deaf and hearing children. The sign bilingualism and co-enrollment approach promotes the early learning of signed language and spoken language to support children’s linguistic, cognitive and socio-emotional development. Tang, Sze-Wing Tang Sze-Wing received his BA and MPhil from The Chinese University of Hong Kong and completed his doctoral dissertation at the University of California, Irvine. He is Associate Professor and Vice-Chairman of the Department of Chinese Language and Literature, The Chinese University of Hong Kong, and concurrently serving as Director of T. T. Ng Chinese Language Research Centre of the Institute of Chinese Studies, The Chinese University of Hong Kong, and Chief Editor of Studies in Chinese Linguistics and Newsletter of Chinese Language. His research interests lie primarily in Chinese syntax, theoretical approaches to the study of Chinese dialects, and comparative grammar. Tao, Hongyin Tao Hongyin is Professor of Chinese Linguistics and Applied Linguistics at the University of California, Los Angeles. He has published a dozen books/edited volumes/special journal issues and over 90 articles in journals and edited volumes on Chinese discourse and grammar, applied linguistics, corpus linguistics, sociolinguistics, and English discourse grammar. Some of his books include Units in Mandarin Conversation: Prosody, Discourse, and Grammar (1996) and Current Trends in Sociolinguistics (1997/2004, co-edited). His most recent books are Working with Spoken Chinese (2011) and Chinese under Globalization: Emerging Trends xxv

List of contributors

in Language Use in China (2012, co-edited). In addition to being the president of the Chinese Language Teachers Association, USA (2014), he is on over a dozen editorial boards and is the Executive Editor of the journal Chinese Language and Discourse and its companion book series, Studies in Chinese Language and Discourse, both with John Benjamins. Tao, Liang Liang Tao is associate professor of linguistics and coordinator of the Chinese Language Program in the Department of Linguistics at Ohio University. She received her doctorate from the University of Colorado Boulder in 1993, and was a post-doctoral research associate in the Healy-Bourne Research Laboratory from 1993 to 1996, when she joined the faculty of Ohio University. Dr. Tao has broad research interests in the areas of theoretical linguistics, psycholinguistics, and applied linguistics, focusing on discourse-functional approaches in her research. She pioneered the study of repair in Chinese conversations, proposed the hierarchical structure of language processes against a popular iconic proposal of topic continuity, and proposed the interface of phonology and syntax (phono-syntactic conspiracy) in Mandarin grammaticalization. With Dr. Alice Healy she examined basic native and foreign language processes, using findings from experimental studies to propose, for instance, transfer of cognitive strategies in processing native and non-native languages, the developmental progression of the unitization effect, and the varied role of words in reading processes. She also conducted research on language development and cognitive approaches to Chinese pedagogy. Dr. Tao has published extensively in journal articles and book chapters, and has served as a co-editor of the volume Current Issues in Chinese Linguistics. Wang, William S.-Y. William S.-Y. Wang is Chair Professor at the Hong Kong Polytechnic University, based in its Department of Chinese and Bilingual Studies. He is also Professor Emeritus of the University of California at Berkeley (where he was Professor of Linguistics for 30 years), and Honorary Professor at Peking University and, Beijing Language and Culture University, and the Chinese University of Hong Kong. He is Editor of the Journal of Chinese Linguistics, which he founded in 1973. In 1992, he was elected President of the International Association of Chinese Linguistics at its formation. In the same year he was elected to the Academia Sinica in Taiwan. His central interest is in language within an evolutionary perspective. He has published some 200 papers and ten books in diverse areas of theoretical and applied linguistics. These have appeared in general magazines, such as American Scientist, Nature, Proceedings of the National Academy of Sciences (USA), Scientific American, etc., in specialized journals, such as Brain and Language, Diachronica, Language, Lingua, Language and Cognitive Processes, Neuropsychologia, Journal of Phonetics, etc., and in various encyclopedias. His writings have been translated into many languages. He has lectured widely in America, Asia, and Europe. Wu, Weiping Wu Weiping is Director of the Yale-China Chinese Language Center at The Chinese University of Hong Kong (CUHK) and a guest professor in four universities in China. He obtained his PhD in Linguistics from Georgetown University, and worked as a Research Associate and Director of the Chinese Language Testing Program at the Center for Applied Linguistics in Washington DC before moving to CUHK. His academic profile includes teaching, research, and publication in linguistics, teaching Chinese as a Second Language (CSL), and assessment. Among his contributions are articles in referred journals such as Forensic Linguistics, xxvi

List of contributors

Language Teaching and Research, TCSOL Studies, book and book chapters in CSL and sociolinguistics, the Computerized Oral Proficiency Assessment (COPA) in Putonghua and Cantonese for learners with background in English, Japanese, and Korean, and volumes of textbooks under the Pragmatic Framework. He is founding President of the CSL Teaching and Research Society (CSLTARS), a member of the editorial committee of several core journals in CSL, and series editor of Linguistics and Teaching Chinese as a Second Language. Xiao, Richard Richard Xiao is Reader in Corpus Linguistics and Chinese Linguistics in the Department of Linguistics and English Language at Lancaster University. His major research interests cover corpus linguistics, Chinese linguistics, contrastive linguistics, and translation studies. He has published numerous books in these research areas, including Aspect in Mandarin Chinese: A Corpus-based Study (2004), Corpus-based Language Studies: An Advanced Resource Book (2006), A Frequency Dictionary of Mandarin Chinese: Core Vocabulary for Learners (2009), Corpus-Based Contrastive Studies of English and Chinese (2010), Using Corpora in Contrastive and Translation Studies (2010), Corpus-Based Studies of Translational Chinese in English-Chinese Translation (2012), Translation and Contrastive Linguistic Studies at the Interface of English and Chinese (2014), and Corpus Pragmatics in Translation and Contrastive Studies (2015). He is a member of editorial boards of a range of international journals, such as Chinese Language and Discourse, Corpora, Corpus Linguistics, Foreign Language Learning: Theory and Practice, Glossa, Global Chinese, International Journal of Corpus Linguistics, and Languages in Contrast. Xing, Zhiqun Janet Janet Zhiqun Xing graduated from the University of Michigan in 1993, and is a Professor of Chinese and Linguistics at Western Washington University. Her research interests are in grammaticalization, semantic change, historical linguistics, discourse pragmatics, and Chinese language pedagogy. Her representative publications include two edited volumes – Newest Trends in the Study of Grammaticalization and Lexicalization (Mouton de Gruyter, 2012) and Studies of Chinese Linguistics – Functional Approaches (Hong Kong University Press, 2008); two monographs – A Contrastive Model of Teaching Chinese to Native English Speakers (Peking University Press, 2011) and Teaching and Learning Chinese as a Foreign Language: A Pedagogical Grammar (Hong Kong University Press, 2006); and several dozens of refereed journal articles related to grammaticalization, semantic change, and Chinese language pedagogy. Yin, Jing-hua John John Jing-hua Yin graduated from Beijing Capital Normal University in 1982, and he received his MEd and PhD in Foreign Language Education from the State University of New York at Buffalo in 1984 and 1995 respectively. He then began teaching Chinese at the Chinese Summer School at Middlebury College, the University of Oregon, and the College of William and Mary. Since 1997, he has been teaching Chinese language and literature at the University of Vermont. He is Professor of Chinese and Chair of the Department of Asian Languages and Literatures. His book Fundamentals of Chinese Characters was published by Yale University Press in 2006, and his book Practical Rhythmic Chinese was published by Foreign Language Teaching and Research Press in 2008. He has also published articles on his studies and the teaching of Chinese characters to college students in the USA. xxvii

List of contributors

Zhang, Grace Grace Zhang is an Associate Professor at Curtin University in Australia, where she teaches and supervises postgraduate students in Chinese linguistics. She has 30 years of tertiary teaching experience in Chinese language at various universities in Australia, China, New Zealand, and the United Kingdom. Her research interests include Chinese linguistics and pragmatics. She was awarded a PhD in linguistics by the University of Edinburgh in 1996, and has published extensively on linguistics. Her recently published books include: Using Chinese Synonyms (2010) by Cambridge University Press; Request Strategies: A Comparative Study in Mandarin Chinese and Korean (2008, co-author) by John Benjamins; Critical Perspectives on Language Education: Australia and Asia Pacific (2014, co-editor) by Springer; Elastic Languages: How and Why We Stretch Our Words (2015) by Cambridge University Press; Communicating through Vague Language: A Comparative Study of L1 and L2 Speakers (2015, co-author) by Palgrave Macmillan. Of specific relevance to this article is her co-authored journal article on the translation of Chinese xiehouyu: Liu, Chiung-wen and Grace Zhang (2006) ‘Translation of Chinese xiehouyu (sayings) and Relevance Theory’, Across Languages and Cultures 7(1): 49–76. Zhang, Xiaoheng Dr. Xiaoheng Zhang has a PhD degree in computer science from Oxford Brookes University, UK, an MSc degree in computer science from Hunan University, China, and a BA degree in English, Hunan University. His research interests include Modern Chinese writing systems, computational linguistics, computer-assisted language learning, and computer-assisted translation. He has published three books and over 60 research papers in these areas. Dr. Zhang is an assistant professor in the Department of Chinese and Bilingual Studies, Hong Kong Polytechnic University. The subjects he teaches at the undergraduate and post-graduate levels include Interface between Chinese and Information Technology, Computer-Assisted Chinese Language Teaching, Computer Tools for the Language Professionals, and Modern Chinese Characters and Information Technology. He is a member of the Editorial Committee of the Journal of Modernization of Chinese Language Education and a visiting professor of Hunan University. Zhuo, Jing-Schmidt Zhuo Jing-Schmidt received her BA and MA in German language and literature from Peking University, MA in Germanic Linguistics from the University of California, Los Angeles, and PhD in General Linguistics from the University of Cologne, Germany. She is associate professor of Chinese Linguistics at the University of Oregon. Her research is concerned with how language structure and its use are shaped by culture, society, and human psychology. The inquiry into the role played by human emotion in linguistic pragmatics constitutes an abiding focus of her work. Her corpus-based research methods have contributed to increased rigor in Chinese pragmatics and historical linguistics research. With scholarly experiences in three continents, she researches and publishes in English, Chinese, and German. She is the author of Dramatized Discourse: The Mandarin Chinese Ba-Construction (John Benjamins, 2005), and editor of Increased Empiricism: Recent Advances in Chinese Linguistics (John Benjamins, 2013), and has published numerous articles in leading linguistics journals including Cognitive Linguistics and Journal of Pragmatics.

xxviii

The Chinese language: The global, historical, and linguistic aspects Chan Sin-wai The Chinese University of Hong Kong, Hong Kong, China

Introduction This is an encyclopedia that will, we believe, meet the academic, linguistic, and pedagogical needs of millions of people in the world who are interested in the Chinese language in different capacities and for different reasons, such as Sinologists, Chinese linguists, and teachers and learners of Chinese as a second language. In the following, we will explain the contents of this encyclopedia of the Chinese language from a global, historical, and linguistic perspective.

The global aspects of the Chinese language Chinese, as we all know, is a language spoken by about one fifth of the world’s population (over a billion people) as China is a vast country with a population currently approaching 1.4 billion. According to statistics, the number of Chinese language speakers exceeds 1.4 billion, including 50 million people who have Chinese as their second language. As of 2014, there were 480 Confucius Institutes spread across the six continents, the number of Chinese language learners was over 40 million worldwide, and more than 3,000 higher education institutions in 109 countries offered Chinese language programs. All these figures show that the target users of this encyclopedia could be the countless language learners and linguists of Chinese worldwide, amateurs of Chinese literature and culture, and scholars in Chinese studies. It is not surprising then that the twenty-first century has sometimes been hailed as the century of the Chinese language. This encyclopedia contains a chapter on ‘Chinese Language in a Global Context’, co-authored by Liu Jin of Georgia Tech and Tao Hongyin of the University of California, Los Angeles. This chapter explores the relationship between globalization and the Chinese language, and discusses some major issues such as the contact of Chinese with other languages in historical times, the ongoing impact of English on Chinese and the perceived crisis involving Chinese, the rapidly ascending status of the Chinese language worldwide, and the impact of nationalization and modernization on regional varieties. Finally, a series of major changes in contemporary Chinese is analyzed, as is the emergence of notable new xxix

Chan Sin-wai

linguistic forms and their theoretical implications. Also related is the issue of the ‘Chinese Language and National Identity’, a chapter by Andrew Simpson of the University of Southern California. This chapter provides an overview of how Chinese has affected the development of national identity, and how the terms ‘nation’, ‘nationalism’, and ‘national identity’ apply in the context of China. It is shown that the relation of ‘Chinese’ to ‘national identity’ varies depending on the perspective of the Chinese nation that is adopted (ethnic vs. political), and that different emphases have been placed on the role of Chinese in the development of national identity at different times in modern history. The globalization of the Chinese language is closely related to the teaching of it as a second language and the training of Chinese teachers. Wu Weiping of The Chinese University of Hong Kong contributed a chapter on ‘Chinese Language Pedagogy’ in which he discusses the topic both diachronically and synchronically. While the former approach explores issues related to the major components of the language, namely phonology, vocabulary, grammar, and the writing system with characters; the latter focuses on the shift from language structure to language use in teaching activities. A systematic approach to implement the Pragmatic Framework in CSL teaching, covering assessment, curriculum design, teaching materials preparation, and teacher training, is also provided as part of this chapter. Jane Orton of the University of Melbourne focuses on ‘Chinese Language Education: Teacher Training’. She claims that certified schoolteachers of Chinese in the West generally graduate from generic courses in education studies and language teaching. They overwhelmingly comprise L1 users who need special training to manage the educational culture and learning styles of Western classrooms. The few dedicated Chinese teacher training programs available concentrate mainly on linguistic features. Analysis of Chinese as an object of learning, and the demands its very particular features make on speakers of European languages, are still to be incorporated in Chinese teacher training.

The historical aspects of the Chinese language The Chinese language is usually divided into Ancient (or Classical) Chinese and Modern Chinese, which are drastically different in several aspects. This encyclopedia includes a chapter on ‘Ancient Chinese’ written by Alain Peyraube of the Ecole des Hautes Etudes en Sciences Sociales (EHESS), Paris. Ancient Chinese, according to Peyraube, is simply defined by Wang Li as ‘the language of the documents of the past’. It covers a very long period, from the first Chinese inscriptions known to us, dated from the fourteenth century bce, until the nineteenth century ce. Three basic stages are usually distinguished for Ancient Chinese: (i) the Archaic period; (ii) the Medieval period; and (iii) the Modern period. It was during the Archaic period, between the fifth and second centuries bce, that what is known today as Classical Chinese par excellence was established, and it is this language, also called Late Archaic Chinese, that is discussed in this chapter. Covering the same period, but with a different approach, is the chapter on ‘Classical Chinese’ by Chris Li Wen-Chao of San Francisco State University. This chapter surveys the development of Classical Chinese from the language of early philosophical discourses to later imitations that have given rise to a written register distinct from the vernacular. Salient linguistic features of the classical language are explained, including the monosyllabicity of lexis, the fluidity of form class, and pronominal case grammar. The author also describes the shift in written language from Classical Chinese to the modern vernacular in the early twentieth century, viewed through the lens of diglossia. The language that we use today is known as ‘Standard Chinese’, about which Shi Dingxu of the Hong Kong Polytechnic University has written a chapter. According to Shi, Standard xxx

The Chinese language: The global, historical, and linguistic aspects

Spoken Chinese used to mean Guanhua ‘official language’ which was typically the dialect of the capital city. Guoyu ‘national language’ was created in the 1920s to take the role of the Standard Spoken Chinese and was succeeded by Putonghua ‘common language’ in the 1950s. The two versions were accompanied by Zhuyin Fuhao ‘phonetic symbols’ and Hanyu Pinyin ‘Chinese phonetic alphabets’ respectively. Another chapter on modern Chinese is written by Feng Shengli of The Chinese University of Hong Kong. This chapter, entitled ‘Modern Chinese: Written Chinese’, shows how written Chinese is different from spoken Chinese in modern times, how the formal register system of written Chinese has newly developed after the May Fourth Movement, and finally, what principles the formal register grammar must observe. The answer lies first on the argument that a formal style after the destruction of literary Chinese by the May Fourth Movement is established by creating a sense of expressive distance from everyday speech. That is, when linguistic expressions are used, the more distant the expressions are kept from everyday speech, the more formal sense they can create in faceto-face conversation. The demand for vernacular Chinese to function formally after the May Fourth Movement motivated and still motivates speakers and writers to use some classical forms to satisfy their urgent need for stylistic effect. This is the essential reason why the separation of writing (formal) from speaking (informal) has been resurrected in modern times. It is then argued that the formal features of written Chinese mainly consist of (i) Monosyllabic Words Used in Disyllabic Templates, (ii) Disyllabic Words Used in Disyllabic Copulates, and (iii) Formal Phrasal Patterns. Although these must be used following strict principles, such as Stylistic Coherence, Auditory Intelligibility (PAI), and Shaping by Prosody (PSP), they must also be mixed with some colloquial features in order to make the language natural. As a result, an amalgamation principle which modulates literary dictions with colloquial expressions is proposed here. During the analyses of the issues outlined above, the syllabic system of Prosodic Grammar and the Tripartite System of Stylistic-Register Grammar are introduced and their empirical applications are also discussed, such as the statistical measurement methods recently developed for degree of formality measurement, composition testing, readability scaling, style gradation, textbook compilation, L2 learning, literacy acquisition, etc. Finally, a discussion is held on what principles the formal register grammar must observe.

The linguistic aspects of the Chinese language The subjects under this section are numerous. They include characters, dialects, areas of linguistics, grammar, and Romanization systems. However, before diving into the chapters on these individual subjects, an overall review of the study of Chinese language, entitled ‘Chinese Linguistics’, has been written by William S. Y. Wang of The Chinese University of Hong Kong. His chapter gives a synoptic view on the languages and peoples of China from a multidisciplinary perspective. It describes the geographical setting of the country, as well as the prehistoric and historic background of the current multi-ethnic inhabitants. It draws upon knowledge gained in anthropology and genetics in addition to linguistics. While the main focus is on varieties of Mandarin, the dominant speech in China for over 1,000 years, the other major dialects are also discussed. This chapter serves as an introduction to other aspects of the Chinese language.

Chinese characters This encyclopedia has a chapter on ‘Chinese Characters’ by John Yin Jinghua of the University of Vermont, who introduces the history and development of Chinese characters from xxxi

Chan Sin-wai

ancient times to the present. It also explains the formation and structure of Chinese characters and presents the features that are unique to Chinese characters.

Dialects The Chinese language, which, as mentioned before, is spoken by the largest number of people on earth as their mother tongue has hundreds of different dialects. It is estimated that other than Mandarin, dialects in China are spoken by 1.2 billion people. The ten most popular dialects are: Mandarin, Min, Wu, Jin, Yue, Gan, Kejia (Hakka), Xiang, Pinghua, and Hui. This encyclopedia comprises a number of chapters that deal with dialects. ‘Mandarin and Other Sinitic Languages’, by Hilary Chappell of the Ecole des Hautes Etudes en Sciences Sociales in Paris and Li Lan of the Chinese Academy of Social Sciences in Beijing, provides a description of the ten main dialect groups which belong to the Sinitic or Chinese branch of the Sino-Tibetan language family. Beginning with the large Mandarin group which enjoys a wide geographical distribution in China, each branch of Sinitic is briefly characterized in turn in terms of its historical relation to Middle Chinese with respect to phonology (syllable structure), its tone system, and its personal pronoun paradigm, in addition to several interesting grammatical features with respect to a chosen representative dialect. Mandarin, as the most widely spoken dialect, is also discussed in detail in ‘Mandarin’ by Shi Dingxu of the Hong Kong Polytechnic University, who holds that it is the dialect spoken in Beijing in a narrow sense and the Northern dialect spoken in North China in a broader sense. The phonetic and phonologic system of Putonghua (‘common language’) is based on that of Mandarin. The description of Mandarin presented here covers its sound inventory, tonal system, words and morphology, phrases, clauses, sentences, aspect system, negation, comparison, and information package constructions. Two other dialects are also discussed in depth. First, ‘Cantonese’, a chapter written by Cheng Siu-Pong and Tang Sze-Wing of The Chinese University of Hong Kong; second, ‘The Hakka Dialect’ by Lau Chun Fat of Xiamen University.

Dictionaries Dictionaries are a necessary evil for language learners, translators, and language professionals, and lexicography has a particularly long history in China. According to Cheung Kam-Siu of The Chinese University of Hong Kong, who has contributed the chapter on ‘Lexicography’, this activity has evolved and developed in China for more than two millennia. This chapter gives an overview of various aspects of diachronic studies of Chinese lexicography, including dictionary compilation and theorization in China from the ancient time to the modern and contemporary periods, and the development of Chinese monolingual and bilingual dictionaries. With detailed illustrations, some of the classics in the history of Chinese lexicography are specifically introduced and analyzed. Lexicography has changed vastly from the ancient period to the present age. Huang Chu-Ren and Li Lan of the Hong Kong Polytechnic University, and Su Xinchun of Xiamen University, in ‘Lexicography in the Contemporary Period’, express the belief that the Chinese language enjoys a long history of dictionaries. The presentation of word meaning, pronunciation, grammar, and usage has developed in tandem with modern lexicography. The application of language mega-corpora has contributed to the compilation of Chinese dictionaries, either in paper or electronic form, for both native and non-native language learners and users. The dichotomy, however, is the definition of lexical units; there are no clear rules to segment words in data. Due to this reason, both character dictionaries and word dictionaries exist in the Chinese language. xxxii

The Chinese language: The global, historical, and linguistic aspects

Grammar Whether Chinese has the same parts of speech as the Indo-European languages has been the subject of much debate. This encyclopedia has a chapter on ‘Chinese: Parts of Speech’ by Candice Cheung Chi-Hang of the City University of Hong Kong that offers a comprehensive survey of the major parts of speech in Chinese, aiming to establish the parts of speech that are found both in Chinese and in the Indo-European languages, and those found only in Chinese. A significant implication of this study is that whereas some parts of speech may be viewed as universal, others are language specific.

Idioms Idiomatic expressions in Chinese include idioms, end-clippers, proverbs, slangs, and taboos. All these topics are covered adequately in this encyclopedia. Jiao Liwei of the University of Pennsylvania focuses on ‘Idioms’. According to Jiao, most Chinese idioms (chengyu) consist of four characters, and typically their meaning is more than the sum of the parts. Based on idiomatic salience, Chinese idioms can be divided into five categories and 15 types. After Written Vernacular Chinese replaced Classical Chinese in the 1910s, idioms have become an elite part of the Chinese language and received tremendous attention from people from all walks of life ever since. The chapter on ‘Chinese Proverbs and Popular Sayings’ by Lawrence Herzberg of Calvin College in Michigan takes a different approach and provides an overview of the rich treasury of Chinese proverbs and popular sayings. Over 400 sayings are presented in the chapter, including both the ancient sayings of Confucius and Lao Zi as well as the homespun truths of everyday life. All of these pithy proverbs offer insights into the Chinese language and culture. ‘Chinese Slang’ is discussed by Robert Moore of Rollins College. Slang is a category of informal speech, characterized by attributes related to playfulness or identification with a specific social group. In Mandarin, the term liyu is commonly translated as ‘slang’, though there are some differences between Chinese liyu and English slang. The main difference is that liyu was traditionally thought to be linked to local dialects. However, a new version of liyu has emerged on the Internet which is not dialect-based but is closely associated with youth culture. Xiehouyu, or end-clippers, are covered in ‘Chinese Xiehouyu’ by Grace Zhang of Curtin University in Australia. This chapter provides a general review of Chinese xiehouyu, including its definition, origin, types, content, structure, features, functions, and translation. Xiehouyu refers to Chinese metaphorical folk sayings which follow a two-part pattern of a metaphor. These two parts are separated by a pause in speech or a comma or dash in writing, providing time for or drawing attention to the intended meaning and enhancing the effect of humour or satire. Finally, Amy He Yun of the University of Huddersfield in United Kingdom, on the other hand, contributes a chapter on ‘Chinese Taboo’. This chapter presents an introduction to Chinese taboo, focusing particularly on its distinctive characteristics such as the hierarchical nature of death taboos in ancient China. Starting by sketching the milestone stages of historical development, the chapter provides detailed analyses of taboo words/expressions and their euphemistic or dysphemistic substitutes in five categories. Apart from the motivations and functions of taboo and prosecution of transgressors, she also discusses the diachronic and synchronic variation as well as punishment avoidance strategies with illustrative examples. xxxiii

Chan Sin-wai

Linguistic areas This encyclopedia covers the main areas of linguistics, such as semantics and pragmatics; a number of sub-areas, such as corpus linguistics, psycholinguistics, computational linguistics, and sign language; as well as other related fields, such as rhetoric and syntax. Semantics is discussed in ‘Chinese Linguistics: Semantics’ by Hsieh Shu-Kai from the National Taiwan University in Taiwan. According to Hsieh, meaning is the key to the understanding of communication, human cognition, and culture. As a subfield of linguistics, contemporary studies of meaning are characterized by various formal modelling, abstract representation, and empirical evidence that can be verified or disproved in a scientific manner. This chapter aims to give a brief and non-technical introduction to modern linguistic semantics, giving priority to the core issues surrounding lexical meanings in Chinese, ranging from morpho-semantics to lexical semantics. Janet Xing of Western Washington University in the United States writes on ‘Semantic Change in Chinese’. Through analysis of diachronic data, she demonstrates how new meanings are developed at the lexical, sentential, and discourse levels. The result of her study suggests that three mechanisms, namely meta[phoricalization], meto[nymization], and semantic reanalysis, are commonly present in the process of semantic change in Chinese, rather than the mere two (meta and meto) recognized in Indo-European languages. Furthermore, she argues that Chinese lexemes tend to develop polysemies accretively due primarily to their isolating-analytical characteristics. Meanings carried by several types of words are indeed noteworthy. Zhuo Jing-Schmidt writes on metaphors under the chapter heading of ‘Metaphor in Chinese: Cognition, Culture, and Society’, which focuses on the sociopolitical function of metaphor in a sociohistorical context by providing a close examination of two Chinese metaphors that have shaped China’s collective unconsciousness. Neologisms are discussed in ‘Chinese Neologisms: Word-formation Strategies in Chinese’ by Antonella Ceccagno of the University of Bologna. This chapter presents some central word-formation strategies in Chinese by focusing on the word-formation patterns that emerge as a result of the crucial role played by compounding in Chinese. The word-formation strategies that are discussed include: metacompounding, reanalysis of syllables as morphemes, neo­logisms formed by selecting a new meaning for the constituents of existing compounds, and the creation of new morphemes or new meanings for existing morphemes. Most of these word-forming patterns have been singled out only recently, and their implications are explored. In the history of the Chinese language, words from various languages, including both languages of foreign countries and non-Chinese languages within China, have been borrowed. In ‘Loanwords’, Miao Ruiqin of Shanghai Jiao Tong University claims that loanwords are adapted into Chinese through one of the following four approaches: (i) phonemic transliteration, (ii) meaning translation, (iii) graphic borrowing, or (iv) the hybrid method. Linguistic processes to nativize loanwords into Chinese mainly involve phonetic, phonological, and orthographic transformations, whereas semantic, morphological, and grammatical changes are much less common. The use of loanwords in modern Chinese demonstrates certain sociolinguistic features in terms of stylistic and regional variations. In addition, existence of multiple adaptations present challenges for language standardization and language policy. Elastic words, whose length can vary from monosyllabic to disyllabic, without a change in meaning, are discussed by Duanmu San and Dong Yan, both of Michigan University. Elastic words have been known since Karlgren (1918), but some theoretical and empirical questions still remain. A precise definition of elastic words is provided, along with an estimate xxxiv

The Chinese language: The global, historical, and linguistic aspects

of the percentage of elastic words in modern Chinese. Finally, theories of why these words are created in Chinese are compared. Pragmatics is about the use of language in different contexts. Jiang Yan of the Hong Kong Polytechnic University writes on Chinese pragmatics. This chapter starts with a survey of some general concepts and their exemplifications in Chinese. It then moves on to a short summary of relevance theory, followed by reports on findings from three case studies in Chinese pragmatics. The first concerns the Chinese adverb bai, which triggers a presupposition that resembles the Principle of Relevance. The second studies ‘list gem’ as a unique figure of speech, which is thought to have a special role to play in the current debate between contextualism and semantic minimalism in the sense that each side needs to supply extra accounts to accommodate such a case. The third provides a new analysis of Chinese counterfactuals, which takes most Chinese counterfactual conditionals as falsifying contingent counterfactuals. Related to the core areas of linguistics there are some subfields, including corpus linguistics, psycholinguistics, computational linguistics, and sign language. ‘Corpus-based Study of Chinese’ by Tony McEnery and Richard Xiao, both of Lancaster University, explores the state of the art in using corpora in Chinese linguistic investigations by focusing on areas where corpus linguistics has made its greatest impact, namely lexical study, grammatical study, and Chinese interlanguage research. ‘Chinese Psycholinguistics’, on the other hand, is a chapter written by Jerome L. Packard of the University of Illinois. This chapter offers an overview of the field of Chinese psycholinguistics. The chapter begins by describing the psychology of Chinese script processing, including the role of strokes and radicals, followed by psycholinguistic aspects of Chinese text reading. The psycholinguistics of lexical access in Chinese is then discussed, including the issue of phonetic and semantic activation during word formation and retrieval. A summary of sentence processing is then presented, followed by a synopsis of Chinese speech perception and production. An additional chapter on psycholinguistics, entitled ‘Psycholinguistics: Reading Chinese’ is provided by Liang Tao of Ohio University and Alice F. Healy of the University of Colorado. This chapter presents findings from experimental studies on reading Chinese, including examinations of processes in English and Chinese, two typologically different languages in their respective orthography, syntactic structures, and discourse grammar. This cross-linguistic comparison illustrates reading processes that reflect (i) universal cognitive behavior and (ii) unique characteristics in Chinese reading. The theoretical issues supported in the former include the unitization hypothesis and the Stroop effect. In the latter, Chinese orthographic features invoke special cognitive processes in word identification and in discourse processes. Computational linguistics is discussed in a chapter written by Zhang Xiaoheng of the Hong Kong Polytechnic University. This chapter focuses on Chinese, and especially on its differences with other languages, such as English. Chinese computational linguistics is often referred to as Chinese information processing in China. This chapter introduces the important areas of Chinese character information processing, word segmentation, natural language understanding and generation, corpus linguistics, and machine translation. ‘Sign language’ forms another chapter in this area, introduced by Gladys Tang of The Chinese University of Hong Kong. According to Tang and her fellow researchers, linguistic study of Hong Kong Sign Language (HKSL) was begun in the 1990s by James Woodward (1993). Since then, research on the linguistic properties of HKSL at different linguistic levels has been flourishing. In this chapter, she first provides an account of the historical development of HKSL, and then focuses on the descriptions of HKSL at the phonological, morphological, and syntactical levels. xxxv

Chan Sin-wai

Other areas related to linguistics covered in this encyclopedia are morphology, syntax, and rhetoric. ‘Chinese Morphology’ is discussed by Jerome L. Packard of the University of Illinois. This chapter describes the properties of words in Mandarin Chinese. It begins with Chinese typological characteristics, and defines word as a syntactically free form. The four Chinese morpheme types (free content morpheme, free function morpheme, bound root morpheme, and affix) are then described, including how the processes compounding, composition, and affixation compose the word types compound, bound root word, derived word, and grammatical word. The chapter concludes with a discussion of borrowing, reduplication, abbreviation, and neologism in Chinese. ‘Syntax’ is covered in a chapter by Li Yafei of the University of Wisconsin-Madison. Chinese syntax during the past 30 years has uncovered a wide range of new facts, explored various novel analyses, and come up with a non-trivial number of previously unknown questions. It has benefited from the general theory of linguistics as well as contributed to it. Both the similarities and differences between Chinese and other well-studied languages are becoming increasingly better understood. A question is how to improve the ways in which Chinese syntax helps shape the theory of the human languages. Another chapter on syntax, entitled ‘Chinese Syntax’, is written by Walter Bisang. This chapter starts out from a critical discussion of the extent to which there is syntax in Chinese and the question of the inter­ action between syntax and pragmatics. For that purpose, it briefly highlights some claims concerning subject–object asymmetry in coordinate-clause constructions and in Chinese relative clauses. Later on, it addresses the argument structure, nominal expressions, verbal structures, anaphora, information structure (focus and topic), the formation of questions, and issues of word-order typology (SVO vs. SOV). The presentation of these phenomena includes well-known syntactic constructions of Chinese like the ba-construction, the bei-construction, and numeral classifiers. ‘Chinese Rhetoric’, on the other hand, is a chapter written by Andy Kirkpatrick of Griffith University in Brisbane, Australia. His chapter provides an overview of the development of Chinese rhetoric from the earliest times to the present. It argues that China needs to rediscover its rich rhetorical tradition so that it may be adapted for public discourse in contemporary China. It includes some comparison and contrast with developments in Western rhetoric. Related to this area is a chapter on ‘Poetic Prosody’ by Feng Shengli of The Chinese University of Hong Kong, which deals with poetic prosody in Classical Chinese. It is shown that Archaic Chinese prosody had undergone a change from moraic foot structure (a foot consisted with two moras) to syllabic foot structure (a foot consisted with two syllables). More specifically, disyllabic Foot Formation developed during the Spring and Autumn period (722–481 bc). Only later did trisyllabic super-foot structure develop, requiring the maturity of disyllabic Prosodic Word Formation. It took even longer for Prosodic Word Compounding in the form of four-syllable structures to fully develop. Observing the development of the prosodic system from Old Chinese up to Middle Chinese, it would be expected that the development of pentasyllabic lines would have to wait for the maturity of super-foot formation and that heptasyllabic lines would have to wait for the maturity of PrWd Compounding. The former is seen in the trisyllabic compound formation of the Eastern Han dynasty (25–220), while the latter appears in tetrasyllabic verb formation during the Southern and Northern dynasties (420–589). According to the theory and facts presented here, the prosodically conditioned poetic effects might provide a basis for motivating a theory of prosodic stylistics in future studies.

xxxvi

The Chinese language: The global, historical, and linguistic aspects

Romanization systems Romanization consists of the spelling of Mandarin in Roman letters. It is related to syllable initial and final in traditional Chinese phonology. There are over 20 Romanization systems for the Chinese language. This encyclopedia contains chapters on the Romanization of Cantonese, the Hanyu Pinyin system, and the Wade–Giles Romanization system. In the chapter on ‘Cantonese Romanization’, Cheng Siu-Pong and Tang Sze-Wing of The Chinese University of Hong Kong hold that the diversity of Cantonese Romanization is a reflection of different backgrounds and purposes in rendering Cantonese into Roman letters. Some forms of Romanization are marked for their historical significance and influence on later proposals; whereas others are note­ worthy for their current popularity, as they are often found in books and other materials. Individual frameworks are discussed in detail, and the concluding tables summarize these discussions. It follows that such diversity compensates for the lack of a predominant form of Cantonese Romanization. Hanyu Pinyin, which is the most popular Romanization system in China and possibly in the world, is discussed in detail by Lilly Chen of Rice University. This chapter begins with an introduction to the official system for spelling Mandarin in Roman letters, and how it emerged from among many competing spelling systems, the earliest of which were devised by foreign missionaries. The structure of Pinyin is described in detail, including also the concepts of syllable initial and final in traditional Chinese phonology. The historical development from a morpho-syllabic to a phonemic system of sound representation is discussed, along with a functional evaluation of Pinyin. Another system that is popular in Taiwan is the ‘Wade–Giles Romanization System’, explained by Karen Steffen Chung of National Taiwan University. The earliest efforts to Latinize Chinese began in the sixteenth century with the work of Italian Jesuit priests Michele Ruggieri and Matteo Ricci. Their system was modified over the centuries by other Western missionaries, most notably Robert Morrison (1782–1834). Sir Thomas Francis Wade (1818–1895) further refined Morrison’s scheme, and Herbert Allen Giles (1845–1935) later popularized Wade’s system mainly through his 1,415-page A Chinese–English Dictionary. The Wade–Giles system was soon widely adopted, until its eclipse in the 1970s by Hanyu Pinyin, which is now the undisputed worldwide standard.

Conclusion It is hoped that these chapters on the global, historical, and linguistic aspects of the Chinese language, written by 61 scholars and experts from different parts of the world, will provide a comprehensive background and an authoritative reference to learners and researchers of the Chinese language.

xxxvii

Acknowledgements

I would like to express my gratitude to my former colleague Professor David Pollard, who served as Chair Professor of Translation at The Chinese University of Hong Kong for some years, for offering me a chance to work on this very meaningful project, and to Andrea Hartill, Senior Publisher of Routledge, for initiating this encyclopedia. Without her support, the present volume would never have been completed and published. My thanks also go to Camille Burns, Editorial Assistant, and Geraldine Martin, production editor, both of Routledge, for their assistance in the editing and production of this volume. I am grateful to all the Consultant Editors for their advice and support, to all contributors for their time and effort in writing articles for this volume, to Florence and James, assistant editors of this encyclopedia, for their help and hard work, and to Florence and Wei Yuxiang for compiling the Index. Chan Sin-wai The editor and publishers would like to thank the following copyright holders for permission to reproduce the following material: Tables and material excerpted from ‘Fundamentals of Chinese Characters’. © 2006 John Jing-hua Yin. Reproduced with kind permission of Yale University Press. ‘ACTFL/CAEP 2012 Standards’. Reproduced with kind permission of American Council on the Teaching of Foreign Languages. Figure ‘Austronesian Languages in Taiwan’, courtesy of Professor Paul Li. Figure ‘235’ from ‘The Archaeology of Ancient China’ by Kwang-chih Chang. © 1987 Yale University Press. Reproduced with kind permission. Figures, Tables, and Excerpts from ‘Love and War in Ancient China: Voices from the shijing’ pp. xxii–xxxiii and p. 38 Figure 8b, © 2013 City University of Hong Kong. Reproduced with kind permission. Figures, Tables, and Excerpts from ‘On Modern Written Chinese’ 37: 145– 61, © 2009 Journal of Chinese Linguistics. Reproduced with kind permission. Table (1) from ‘Shoubuchulai, La formazione delle parole in cinese’ p. 128, © 2009 Serendipita Editrice. Reproduced with kind permission. Figures, Tables, and Excerpts from ‘A Prosodic Explanation for Chinese Poetic Evolution’ 2: 223–57, © 2011 Tsing Hua Journal of Chinese Studies. Reproduced with kind permission. Material excerpted from ‘Chinese Proverbs and Popular Sayings: With Observations on Culture and Language’. © 2012 Qin Xue Herzberg and Larry Herzberg. Reproduced with kind permission of Stone Bridge Press, Inc. While the publishers have made every effort to contact copyright holders of material used in this volume, they would be grateful to hear from any they were unable to contact. xxxviii

1 Ancient Chinese Alain Peyraube Centre National de la Recherche Scientifique, Ecole des Hautes Etudes en Sciences Sociales, Paris, France

1. Introduction ‘Chinese is only one of a very few languages whose history is documented in an unbroken tradition extending back to the second millennium bc’ (Norman 1988: ix). Chinese is usually divided into Ancient Chinese (gǔdài hànyǔ 古代汉语) and Contemporary Chinese (xiàndài hànyǔ 现代汉语). Ancient Chinese is simply defined as ‘the language of the writings of the past’ (Wang 1979: 1). It covers a very long period, from the oracle bone inscriptions (OBI, the first Chinese inscriptions known to us, dated from the fourteenth century bce) until the nineteenth century. Three basic stages are generally distinguished for Ancient Chinese: (i) the Archaic period (shànggǔ 上古), until the second century bce; (ii) the Middle or Medieval period (zhōnggǔ 中古), from the first century bce to the middle of the thirteenth century ce; (iii) the Modern period ( jìndài 近代), from the middle of the thirteenth century to the middle of the nineteenth century. It was during the Archaic period that what is known today as Classical Chinese (wényán 文言) was standardized. This language, playing a role like Latin in Europe, remained the main written language used in literary texts until the beginning of the twentieth century. The period for Classical Chinese par excellence refers more precisely to the language used by the philosophers and scholars of the Warring States period (475–221 bce) and it was probably not very different from the educated speech of the period. The gap between the written and the spoken language began to develop in Han times or Pre-Medieval Chinese (206 bce – ce 220) and increased considerably with time. It is essentially Classical Chinese, also called Late Archaic Chinese (hòu shànggǔ hànyǔ 后上古汉语), that will be discussed in this chapter. I will nevertheless make several digressions regarding the period prior to the fifth century bce, i.e. what is known as pre-Classical Chinese or Early Archaic Chinese (qián shànggǔ hànyǔ 前上古汉语), and about the Medieval period, especially in the phonological section.

2. Phonology The history of Chinese phonology is usually divided into four periods based on the two important rhyme dictionaries, the Qiēyùn《切韵》of 601 and the Zhōngyuán yīnyùn《中原 音韵》(Rhymes according to the pronunciation of the Central Plains) of 1324: Old Chinese, Middle Chinese, Old Mandarin, and Modern Chinese. Old Chinese refers to the period before 1

Alain Peyraube

the Qiēyùn and is the earliest period; the Qiēyùn represents the beginning of Middle Chinese; the Zhōngyuán yīnyùn represents the beginning of Old Mandarin; and Modern Mandarin is the period leading up to present time, with no fixed dates. See Peyraube and Shen (forthcoming).

2.1.  Old Chinese phonology The source materials for Old Chinese phonological reconstruction, mainly based on rhyming materials (largely drawn from the Shī jīng《诗经》(The Book of odes), eleventh–sixth centuries bce), do not allow us to clearly identify the reconstructed phonological forms in space and time. The syllable in Old Chinese is analyzed as being composed of an initial and a final. Following Baxter (1992: 7), whose inventory of phonetic segments is given below, (i) the initial contains a pre-initial (treated as a prefix in Baxter and Sagart 1998) and an initial (Table1.1), and (ii) the final contains a medial, a main vowel (Table 1.2), a coda, and a postcoda (Table 1.3). Table 1.1  37 Old Chinese initials (Baxter 1992: 177) p t

ph th

b d

m n

hm hn

ts k kw Ɂ Ɂw

tsh kh kwh x

dz g gw ɦ

ŋ ŋw

hŋ hŋw

w l r j z

hw hl hr hj s

Table 1.2  The six main vowels (Baxter 1992: 180) i e

ɨ ɑ

u o

Table 1.3  Old Chinese codas (Baxter 1992: 181) zero j w

k t wk p

ŋ n m

Three medial elements have been reconstructed: *-r-, *-j- (though the reconstruction of the medial *-j- has now been replaced by a contrast of vowel length), and, marginally, *-l-. The two post-codas are *-ʔ and *-s, which are the respective sources of the rising tone (shǎngshēng 上声) and of the departing tone (qùshēng 去声) in Middle Chinese. 2

Ancient Chinese

This reconstructed Old Chinese phonemic system is far from being universally accepted. In fact, several specialists continue to consider that we are unable to do more than to approach a reconstruction of Old Chinese.

2.2.  Middle Chinese phonology The reconstruction of Middle Chinese phonology, in a more accurate sense, means the reconstruction of phonetic values for the existing categories, which are usually assumed to be phonemic. Table 1.4  Middle Chinese initials (Baxter 1992: 45) Labials Dentals/ Alveolars Lateral Retroflex stops Dental sibilants Retroflex sibilants Palatals Velars Laryngeals

p t

ph th

b d

m n

tr ts tsr tsy k ʔ

trh tsh tsrh tsyh kh

dr dz dzr dzy g

nr ny ŋ

l s sr sy

z zr zy

x

h

y

The basic medials are the glides -j- and -w-. The reconstruction of main vowels has changed several times. Efforts have been made to reduce the number of main vowels and to make the vowel system more natural in relation to linguistic universals. Table 1.5  Middle Chinese main vowels (Baxter 1992: 61) i e ɛ æ

u o

These main vowels may be followed by commonly reconstructed codas, which are either stop consonants or approximants (Table 1.6). Table 1.6  Middle Chinese codas (Peyraube and Shen, forthcoming) zero ŋ k

w

m p

j n t

Middle Chinese has also a system of four tones which were first identified and named in the fifth century. These tones are called píng 平 ‘level’, shǎng 上 ‘rising’, qù 去 ‘departing or falling’, and rù 入 ‘entering and short’. Each Chinese syllable belongs to one of these four tonal categories. 3

Alain Peyraube

3. Morphology It is often said that Ancient Chinese, being a monosyllabic and monomorphemic language, is a language with an impoverished morphology, a language in which the grammatical processes are almost totally syntactic, thus a language of that type called analytic or isolating. However, Ancient Chinese did indeed possess morphological processes, although none of them was fully productive. They actually may represent a vestige of older stages in which such a process was considerably more productive. These word-formation processes are of the same type as those found in Contemporary Chinese: compounding, reduplication (total reduplication such as wēiwēi 巍巍 ‘tall and grand’, or partial reduplication such as tángláng 螳螂 ‘praying mantis’), and even affixation, which was not at all an unproductive process in Ancient Chinese. Having derivational and even inflectional affixes, Ancient Chinese could be thought of as more synthetic than later stages of the language. Several prefixes, suffixes, and infixes have now been reconstructed for Ancient Chinese. These are derivational morphemes changing the meaning or part of the speech of the words to which they are attached (the ensuing discussion closely follows the treatment of Baxter and Sagart 1998). A good example of a prefix that has been reconstructed is the prefix *N- (causing a following voiceless obstruent to become voiced in Medieval/Middle Chinese) which, when attached to a verb (or even a noun in some cases), seems to produce an intransitive verb or adjective: thus, kens 见 ‘to see’ > *N-kens ‘to appear’. Another prefix *k-, added to verbs, could refer to more concrete actions taking place in a limited time frame: *ljuk 鞠 ‘to nourish’ > *k-ljuk ‘to breast feed’. See Sagart (1999). Another prefix *s- derives causative verbs from non-causative verbs or even from nouns (denominative suffix): *m-lun-s 顺 ‘obedient’ > *s-lun ‘to make (a horse) obedient’. Concerning the suffixes, there is in Old Chinese a reconstructed suffix *-s, which is the source of the departing tone of Middle Chinese, which, when added to adjectives or verbs produces derived nouns: for example, *tsrek 责 ‘to demand payment’ > *tsrek-s ‘a debt’. See Mei (1989). Some gradable adjectives also have corresponding noun forms in which the suffix *-s functions like English -th, occurring in pairs such as ‘deep/depth’, ‘wide/width’ (see Downer 1959; Mei 1980). Two infixes have also been reconstructed in Old Chinese by Sagart (1993). The exact function of the first one, *-j-, is difficult to establish, but forms with and without *-j- do appear to be semantically related. The second infix, *-r-, is said to produce forms that are plural or collective in the case of nouns, and iterative, durative, or indicating effort in the case of verbs. There are also some bi-morphemic monosyllabic words in Classical Chinese, which result from the fusion of two morphemes. One of the most cited examples of this fusion phenomenon is the negative fú 弗, which is analyzed as a combination of the negative bù 不 ‘not’ and the third-person pronoun zhī 之 ‘him, her, it’. Among other contractions of this kind in Classical Chinese are to be found: zhī 之 ‘third personal pronoun’ + yú 于 ‘to, at’ > zhū 诸; wú 毋 ‘not’ + zhī 之 ‘third personal pronoun’ > wù 勿; yú 于 ‘at, to, from’ + zhī 之 ‘third personal pronoun’ > yān 焉; hú 胡 ‘why’ + 不 bù ‘not’ > hé 盍. Examples: (1) 乞诸其邻而与之(《论语》) Qi zhu qi lin er yu zhi ask-for 3sg+to his neighbour and give 3sg ‘(He) asked his neighbour for it and gave (it to) him’ (Confucian Analects, fifth century bce) 4

Ancient Chinese

(2) 得之则生,弗得则死(《孟子》) de zhi ze sheng fu de ze si neg+3sg get then die get 3sg then live ‘(If he) gets it (he will) live, (if he) does not get it, (he will) die’ (Mencius, fourth century bce) (3) 百亩之田,勿夺其时(《孟子》) bai mu zhi tian wu duo qi shi hundred mu lig field neg+3sg deprive their time ‘The hundred mu of fields, do not deprive them of their time (of cultivation)’ (Mencius) (4) 学焉而后臣之(《孟子》) xue yan er hou chen zhi learn from+him and later make-subject 3sg ‘(He) learned from him and later made him (his subject)’ (Mencius) In fact, starting in the Pre-Medieval (Han) period, ca. first century bce, all the above examples of fusion disappeared and a fission process came into being, which indicates undoubtedly a move from syntheticity to analyticity: zhū 诸 > zhī 之 + yú 于; fú 弗 > bù 不 + zhī 之; wù 勿 > wú 毋 + zhī 之; yān 焉 > yú 于 + zhī 之; hé 盍 > hú 胡 + bù 不.

4. Syntax In Ancient Chinese, the subject (when expressed, as many sentences are without any subject) precedes the predicate, the verb precedes its object, and the modifiers precede the words they modify. Thus, Ancient Chinese is an SVO language (where S = subject, V = verb, and O = object).

4.1.  Sentence types The predicate might be complex, i.e. composed of more than one verb. Such cases involve serial verb constructions of the type V1  .  .  .  V2  .  .  .  (V3).  .  .  .  The semantic relationship between verbs in series is varied. In the following example, it involves an implication of purpose, where yǐ ‘in order to’ links the two verb phrases: (5)  楚人伐宋以救郑(《左传》) Chu ren fa Song yi jiu Zheng Chu people raid Song in-order-to save Zheng ‘The people of Chu raided Song in order to save Zheng’ (The Tradition of Zuo, fifth century bce) Complex predicates may also involve a ‘pivotal construction’, in which the noun phrase object of the first verb is the subject of the second verb: (6) 请君讨之(《左传》) qing jun tao zhi ask Prince attack 3pl ‘[I] ask [you] the Prince to attack them’ (The Tradition of Zuo) 5

Alain Peyraube

An interesting characteristic of the subject–predicate construction in Classical Chinese is that the whole construction can be nominalized by inserting the particle (ligature) zhī 之 between the two constituents of the construction:  (7) 人之爱人求利之也(《左传》) ren zhi ai ren qiu li zhi ye person lig love person pursue profit 3sg fin-part ‘One person loving another person [would] pursue profit [for] him’ (The Tradition of Zuo) A transitive verb can take one object (usually a noun phrase) or two, an indirect object (IO) and a direct object (DO) in that order (V+IO+DO). The double-object construction is restricted to those verbs with the semantic feature [+ give], [+ say] or [+ teach]: Apart from the pattern V+IO+DO, example (8), two other common orders, involving the prepositions yǐ (DO marker) or yú (‘to’, introducing the IO), are possible for the dative construction: (i) yǐ + DO+V+IO (9), and V+DO+yú+IO (10). See Peyraube (1987).  (8) 公赐之食(《左传》) gong ci zhi shi prince offer 3sg food ‘The prince offered him food’ (The Tradition of Zuo)  (9) 尧以天下与舜(《孟子》) Yao yi tianxia yu Shun Yao om Empire give Shun ‘Yao gave the Empire to Shun’ (Mencius) (10) 尧让天下于许由(《庄子》) Yao rang tianxia yu Xu You Yao leave Empire to Xu You ‘Yao left the Empire to Xu You’ (Zhuangzi, fourth century

bce)

Verbs may be followed by complements (bǔyǔ 补语), a term used for adjuncts when they follow the verb. When adjuncts precede the verb, they are denoted as adverbials (zhuàngyǔ 状语). Complements are divided into two types, depending upon whether or not they are introduced by a prepositional marker which forms a prepositional phrase (PP). Notable examples of PPs are locative complements, which are usually introduced by the preposition yú 于/於 ‘at, to’, as in: (11) 北学于中国(《孟子》) bei xue yu zhong guo north learn at central state ‘He went to the north to learn [it] in the Central States’ (Mencius) The PPs can also be adverbials, especially when they are formed with the preposition yǐ 以, as in: (12) 以羊易牛(《孟子》) yi yang yi niu with sheep change ox ‘Change the ox for a sheep’ (Mencius) 6

Ancient Chinese

Sentences are also customarily divided into simple and complex types and can be further classified as declarative, interrogative, imperative, and exclamatory sentences. There are three basic types of interrogative sentences: (i) yes/no questions; (ii) WH-questions; and (iii) rhetorical questions. The first type is formed with final question particles (hū 乎, yú 欤, yé 耶), which essentially transform statements into questions. The second type (WH) contains a question word (one of the interrogative pronouns listed below), generally without any final particle. The third type is more complex. Some rhetorical questions are formed with a final particle (hū, yú, or yé), but also with an adverb of negation placed before the verb, implying an affirmative answer. Others are formed with the modal particle qǐ 岂 ‘how could’ requiring an affirmative answer. The final particles hū and zāi 哉 are also generally used, though they may be omitted: (13) 予岂好辩哉?(《孟子》) yu qi hao bian zai 1sg how-could like debate fin-part ‘How could I [be one who] loves debating?’ (Mencius) Imperatives are not syntactically marked as such in Classical Chinese. The subject is usually deleted, but this in itself is not a sufficient diagnostic of the imperative sentence. However, when the imperative is intended to be understood as a request, and not as an order or a prohibition, the verbs yuàn 愿 ‘wish’ or qǐng 请 ‘beg’ are used: (14) 王请度之(《孟子》) wang qing du zhi prince beg measure it ‘[My] Prince, please measure it’ (Mencius) The final particle zāi 哉 or sometimes yǐ 矣 is the usual marker of the exclamatory sentence. It can be added either to a declarative or to an interrogative. The subject–predicate order is usually inverted in exclamatory sentences. Example: (15) 死矣盆成括!(《孟子》) si yi Pencheng Kuo! dead fin-part Pencheng Kuo ‘He is dead, Pencheng Kuo!’ (Mencius) A special kind of sentence in Classical Chinese is the copular sentence. If one defines the copula as an overt word which, when used in equational sentences, links the subject to a nominal predicate, and expresses (i) an equivalence meaning or (ii) a property or classificatory meaning, then one can identify the presence of copulas in Classical Chinese, even if they are not strictly necessary. The most common way of creating copular sentences is to add the final particle yě 也 at the end of a sentence (see Peyraube and Wiebusch 1995 for a detailed account of the history of copulas in Ancient Chinese, and especially for a discussion of the status of yě as a copula): (16) 彼丈夫也,我丈夫也(《孟子》) bi zhangfu ye wo zhangfu ye that reliable-man fin-part 1sg reliable-man fin-part ‘They were reliable men, I am a reliable man [too]’ (Mencius) 7

Alain Peyraube

In addition to yě, other copulas are attested in Classical Chinese. Thus, the negative copula fēi 非 ‘to not be’ is required in all negative nominal predicate sentences. In affirmative copular sentences, the verb wéi 为, which also means ‘to do, to regulate, to act, to consider as’, and so forth also acts regularly as a copula, and finally the copular verb shì 是 ‘to be’, still used today, and which comes from the demonstrative pronoun shì ‘this’ through a process of grammaticalization, is already attested no later than the Qin dynasty (second century bce). Concerning the passive sentences, there are semantic passives in Classical Chinese, i.e. passives expressing passivity without any overt morphological marker: a transitive verb can be made passive by placing its object (the patient) in subject position. However, there are also passive structures marked with some formant, such as a preposition, or an auxiliary verb. The most common way to form a passive construction in Classical Chinese is to use the preposition yú 于/于 ‘by’ to introduce the agent (V + yú + Agent), as in: (17) 治于人者食人治人者食于人(《孟子》) zhi yu ren zhe si ren zhi ren zhe si yu ren rule by other the-one-who feed other rule other the-one-who feed by other ‘Those who are ruled by others feed others, those who rule are fed by others’ (Mencius) Four other passive structures are commonly attested: wéi 为 + V, jiàn 见 + V (where wéi and jiàn are best considered to be auxiliary verbs, and where no agent is expressed), and wéi + Agent + V and wéi + Agent + suǒ 所 + V, which became the most common passive forms from the second century bce on. Examples: (18) 盆成括见杀(《孟子》) Pencheng Kuo jian sha Pencheng Kuo passive kill ‘Pencheng Kuo was killed’ (Mencius) (19) 后则为人所治(《史记》) hou ze wei ren suo zhi late then passive other passive control ‘[If I react] late, [I] will then be controlled by others’ (Records of the Historian, first century bce) Yet another passive form appears at the end of the Classical period: bèi 被 + V, where bèi is a verb meaning ‘to suffer; to be affected’. It will first become an auxiliary verb expressing passivity, like wéi and jiàn without any agent, and several centuries later, around the Early Medieval period, a preposition introducing a noun phrase agent, after being grammaticalized. For a more detailed analysis of the passive forms in Ancient Chinese, see Peyraube (1989). Example of bèi being a passive preposition: (20) 亮子被苏峻害(《世说新语》) Liangzi bei Su Jun hai Liangzi passive Su Jun kill ‘Liangzi was killed by Su Jun’ (Shi shuo xin yu, fifth century 8

ce)

Ancient Chinese

Complex sentences are composed of two or more clauses joined through coordination or subordination. The joining of clauses can be accomplished without any overt marking, as in example (21) or with a connective marker ér 而 ‘and, but’ or yì 亦 ‘also’, as in (22): (21) 老者安之,朋友信之,少者怀之(《论语》) lao zhe an zhi pengyou xin zhi shao zhe huai zhi trust 3pl young 3pl care 3pl old the-one-who soothe 3pl friend ‘As for the old, soothe them; as for friends, trust them; as for the young, care for them’ (Confucian Analects) (22) 人民少而禽兽众(《韩非子》) renmin shao er qin shou zhong people few but bird beast numerous ‘People are few but [wild] animals are numerous’ (Han Feizi, third century

bce)

Subordination may be indicated by subordinating conjunctions or particles, which can occur in the first clause, in the second, or in both. In the case of conditional sentences, the conjunctions rú 如, ruò 若 or gǒu 苟 ‘if ’ may appear in the first clause (if-clause), and the markers zé 则 or sī 斯 ‘then’ in the main clause (for an exhaustive analysis of the conditionals in Classical Chinese, see Harbsmeier 1981: 229–87). In concessive sentences, the most commonly used conjunction of concession is suī 虽 ‘although, even if ’. In the main clause one often finds ér 而, which then has its adversative meaning ‘yet’. In sentences expressing cause, the ‘because’ clause may be introduced by the preposition yǐ 以, and the main clause may contain the connective gù 故 ‘so, therefore’: (23) 以其不争故天下莫能与之争(《老子》) yi qi bu zheng gu tian xia mo neng yu zhi zheng because 3sg neg compete therefore Heaven under nobody can with 3sg compete ‘Because he does not compete, nobody can compete with him under Heaven’ (Laozi, fourth century bce)

4.2.  Word order Saying that Classical Chinese is an SVO language (see above) is not uncontroversial. The problem of word order and word order change in Chinese has been much debated since the 1970s. The discussions have essentially surrounded the hypothesis of Li and Thompson (1974), according to which Archaic Chinese was a SOV language, which might have changed to SVO, before shifting back again to SOV, the last stage still being underway: a) SOV > SVO; b) SVO > SOV. The hypothesis b) SVO > SOV has been criticized, and it has been shown that, synchronically, Chinese still is and remains a SVO language and that the OV order is a marked word order coding contrastiveness. The first change a) SOV > SVO has been challenged less because the SOV order for Archaic Chinese seems plausible, insofar as this order is found in Classical Chinese under special conditions. Consequently, it has been claimed that ProtoChinese must have been SOV and, therefore, Proto-Sino-Tibetan too, since almost all the Tibeto-Burman languages have a verb-final order (see LaPolla 1993; Mei 1997). Contrary to this viewpoint, the first attested documents of Pre-Archaic Chinese, i.e. the Oracle Bone inscriptions (OBI) (fourteenth–eleventh centuries bce), show a regular order of 9

Alain Peyraube

SVO, and this stage of the language is indeed more SVO than later stages of Chinese, such as Early or Late Archaic. To claim, then, that in a more ancient stage, before the oracle bone inscriptions, the basic order could have been SOV, is a pure surmise, and cannot be empirically proved (see Shen 1992; Peyraube 1997a, 1997b; and Djamouri 2001), who gives the following figures: 93.8% of the sentences in the OBI are (S)VO vs. 6.2% (S)OV). There are also several clear indications for SVO order being the more basic and dominant word order than SOV in Early and Late Archaic, i.e. in Classical Chinese. When the object is a full lexical noun phrase (NP), the basic order is almost exceptionlessly VO, as in: (24) 君必矢国(《左传》) Jun bi shi guo Prince certainly lose state ‘The prince (will) certainly lose the State’ (The Tradition of Zuo) There are some cases in which the NP-object is in a preverbal position, but these cases are marginal and the OV order is then marked [+ contrastive]. The same situation applies when the noun object is followed by a preverbal marker, usually shì 是 or zhī 之, as in (6): (25) 今吴是惧(《左传》) jin Wu shi ju now Wu om afraid ‘Now [they] are afraid of [the state of] Wu’ (The Tradition of Zuo) However, there are also cases of OV order in Archaic Chinese, not found in Contemporary Chinese, in which the object is a pronoun, being either (i) an interrogative pronoun (26), (ii) the demonstrative pronoun shì 是 ‘this’ or (iii) a pronoun in a negative sentence (27): (26) 吾谁欺?欺天乎?(《论语》) wu shei qi? qi tian hu? I who deceive deceive Heaven int-part ‘Whom should I deceive? Should I deceive Heaven?’ (Confucian Analects) (27) 不吾知也(《论语》) Bu wu zhi ye neg 1sg understand fin-part ‘(You) don’t understand me’ (Confucian Analects) What can be said about these various OV orders involving pronouns? There are first some statistics that show that the OV order has always been non-dominant, since it is only wellattested in the case of pronoun objects. In a corpus of 2,767 VO or OV sentences drawn from the bronze inscriptions, 88.56% of objects (O) are nouns, only 3.3% are pronouns (see Guan 1981: 88). The ratio of pronoun objects is certainly higher in other later documents of Classical Chinese, but it never exceeded 15% of the entire body of VO and OV constructions. It is also known that in many languages, the position of pronouns is different from that of noun phrases, and that ‘unstressed constituents, such as clitic pronouns, are often, crosslinguistically, subject to special positioning rules only loosely, if at all, relating to their grammatical relation, so sentences with pronouns can be discounted in favor of those with full noun phrases’ (Comrie 1989: 89). 10

Ancient Chinese

4.3.  Full words and empty words Classical Chinese words are traditionally divided into two categories: shízì 实字 ‘full words’ and xūzì 虚字 ‘empty words’. The former are content words (which carry semantic content) and form an open class; included in this category are nouns, verbs, and adjectives. The latter are function words or grammatical words, used to express grammatical relationships. They include pronouns, adverbs, prepositions, conjunctions, and particles (see Peyraube 2004). Words can be used in functions customarily reserved for other words. This does not imply, however, as some scholars have assumed, that there are no parts of speech in Classical Chinese, and that words can be used indifferently in any grammatical category.

4.3.1.  Full words Chinese nouns typically function as subjects or objects. However, under certain conditions, they may function like verbs, such as the predicates in (28) and (29), or like adverbs, such as the adverbials in (30): (28) 君君臣臣父父子子(《论语》) jun jun chen chen fu fu zi zi ruler ruler minister minister father father son son ‘The ruler acts as a ruler, the minister as a minister, the father acts as a father, and the son as a son’ (Confucian Analects) (29) 物物而不物于物(《庄子》) wu wu er bu wu yu wu thing thing and neg thing passive thing ‘Treat things as things and do not be treated as things by things’ (Zhuangzi, fourth century bce) (30)

豕人立(《左传》) shi  ren   li pig man stand-up ‘The pig, like a man, stood up’ (The Tradition of Zuo)

Chinese verbs are fundamentally predicative in nature. Unlike nouns, which are negated by the adverb of negation fēi 非 ‘to not be’, verbs are negated by the simple adverb bù 不 ‘not’. One particular use of intransitive verbs in Classical Chinese is in a causative function: thus, huó 活 ‘to live’: ‘to make (people) live’; xíng 行 ‘to go’: ‘to put into motion’; yǐn 饮 ‘to give to drink’: ‘to give something to drink’; dòu 斗(鬭)  ‘to fight’: ‘to make (people) fight’. Example: (31) 不如先斗秦赵(《史记》) buru xian dou Qin Zhao better first (make-)fight Qin Zhao ‘It is better to make the Qin and Zhao fight first of all’ (Records of the Historian) Another verbal subclass is composed of modal auxiliary verbs (qíngtài zhùdòngcí 情态助动词), also sometimes called ‘can-wish verbs’ (néngyuàn dòngcí 能愿动词). They are verbs that take other verbs as their objects and express the modality of the following 11

Alain Peyraube

verb phrase. This modality (ability, possibility, probability, certainty, obligation, volition, etc.) can be characterized as epistemic, deontic, or dynamic. Auxiliary verbs form a closed list and can be classified into the following four semantic groups: (i) verbs expressing mainly possibility and permission, including kě 可, néng 能, zú 足, dé 得, huò 获, kěyǐ 可以, and zúyǐ 足以 (32); (ii) the four verbs of volition, gǎn 敢, kěn 肯, yù 欲, and yuàn 愿 (33); (iii) the two auxiliaries of necessity, obligation and requirement (certainty and obligation), yí 宜 and dāng 当; and (iv) the passive auxiliaries jiàn 见, wéi 为, and bèi 被. For a detailed analysis of mood and modality in Chinese, see Chappell and Peyraube (forthcoming). (32) 天子不能以天下与人(《孟子》) Tian zi bu neng yi tianxia yu ren om Empire give other Heaven son neg can ‘The Emperor cannot give the Empire (to) others’ (Mencius) (33) 子欲居九夷(《论语》) Zi yu ju Jiu Yi Master intend-to live Jiu Yi ‘The Master intends to live in Jiu Yi’ (Confucian Analects) Adjectives can also be considered as a subcategory of verbs. Indeed, they are intransitive verbs of quality, being negated by the adverb bù 不: (34) 名不正则言不顺(《论语》) ming bu zheng ze yan bu shun name neg correct then word not justified ‘If names are not correct, then words cannot be justified’ (Confucian Analects) Like intransitive verbs, adjectives can also have a causative use: (35) 王请大之(《孟子》) Wang qing da zhi King beg great it ‘Your Majesty, [I] beg [you] to make it great’ (Mencius) In addition, adjectives are also typically found as noun phrase modifiers, as in bái mǎ 白马 ‘white horse’, or as verb phrase modifiers, for instance jí zǒu 急走 (lit. rapid-run) ‘run rapidly’. Finally, one can consider that numerals constitute a subclass of the category of adjectives. They do indeed behave syntactically like adjectives; thus, they can form predicates and are negated by the adverb bù. Most commonly, however, they function as modifiers of nouns: (36) 年已七十矣(《孟子》) nian yi qi shi yi age already seven ten fin-part ‘[He] is already 70 years old’ (Mencius) (37) 吾何爱一牛?(《孟子》) wu he ai yi niu I why begrudge one ox ‘Why (should) I begrudge one ox?’ (Mencius) 12

Ancient Chinese

4.3.2.  Empty words Pronouns are usually analysed as empty words. Several types of pronouns can be identified: personal, demonstrative, interrogative, and indefinite. Personal pronouns characteristically occur in different forms. The most common ones, with no distinction being made between singular and plural are: (i) for the first person, wú 吾, wǒ 我, and yú 余(予); (ii) for the second person, rú 汝, ruò 若, and nǎi 乃; (iii) for the third person, zhī 之, jué 厥 and qí 其. Several scholars have tried to characterize their different usages according to case (nominative, accusative, or genitive), but dialectal variation also is a factor: in most instances, the different usages of the pronouns depend on the different texts in which they occur. The most common demonstrative pronouns are: (i) shì 是, cǐ 此, sī 斯, zhī 之, and zī 兹 ‘this, these, here’; and (ii) bǐ 彼, fū 夫, and qí 其 ‘that, those, there’. Here too, it is difficult to explain formal differences without considering dialectal variation. All of these demonstratives can be used as adjectivals (modifying the following nouns or noun phrases), or as subjects or objects. The interrogative pronouns are divided into two categories: (i) those that replace subjects or objects (which are usually nouns), shéi 谁 ‘who’, shú 孰 ‘which, who’, hé 何 ‘what’; and (ii) those that replace predicative verbs or adverbs, hú 胡 ‘why, how’, xī 奚 ‘why’, hé 曷 ‘how, why’, ān 安 ‘where, how’, yān 焉 ‘how, where’. See Peyraube and Wu (2005). The class of indefinite pronouns includes huò 或 ‘some, someone, something’, mò 莫 ‘none, no one, nothing’, and mǒu 某 ‘some, a certain one’. Adverbs are also considered as empty words. Usually positioned in preverbal position, they typically modify the predicate of the sentence. One can distinguish several types: (i) adverbs of degree ( jí 极 ‘extremely’, zuì 最 ‘most’, shǎo 少 ‘little’, shén 甚 ‘very’, etc.); (ii) adverbs of quantification and restriction ( jiē 皆 ‘all’, gě 各 ‘each’, dú 独 ‘only’, etc.); (iii) adverbs of time or aspect ( yǐ 已 ‘already’, cháng 尝 ‘once’, jiāng 将 ‘be going to’, nǎi 乃 ‘then’, fāng 方 ‘just’, etc.); (iv) adverbs of negation (bù, fú, fēi, wú, wèi 未, etc.). See above for the processes of fusion for some of these adverbs of negation. Chinese prepositions are all verbal in origin (i.e. they arise from verbs through a process of grammaticalization). There are two commonly occurring prepositions in Classical Chinese: (i) yú 于/於 ‘at, to, in, from, towards, than, by’, etc.; and (ii) yǐ 以 ‘with, by means of, in order to, because’, etc. The first of these, yú, can be used to form locative, ablative, dative, comparative, or passive constructions; the second, yǐ, primarily instrumental, also expresses purpose and several other grammatical relationships. An important characteristic of yǐ is that it can also introduce the direct object of a double-object construction (32). Additional prepositions are yòng 用 ‘with’, wéi 为 ‘for, on behalf of, for the sake of, because’, yǔ 与 ‘with’, zì 自 ‘from’, among still others. Generally, simple juxtaposition is sufficient to coordinate nouns or noun phrases, as in fù mǔ 父母 (lit. father mother) ‘father and mother’; or verbs and verb phrases in serial verb constructions. However, some coordinative conjunctions also occur, such as jí 及 and yǔ 与 ‘and’ for coordinating noun phrases, or ér 而 ‘and’ and qiě 且 ‘and, moreover’ for coordinating verb phrases or clauses. Liu and Peyraube (1994) have argued that the conjunctions jí and yǔ did not directly develop from verbs, but from prepositions, which were themselves derived from verbs. In other words, two processes of grammaticalization have occurred sequentially: verb > preposition > conjunction (Chinese conjunctions are thus more grammaticalized than prepositions). Finally, the category of particles (zhùcí 助词) is also obviously included in empty words. This category is usually divided into structural particles (zhī 之, suǒ 所, zhě 者) and modal particles that can occupy the initial, the medial, and, in most cases, the final position of 13

Alain Peyraube

a sentence. These final particles can be divided according to the sentence types in which they occur – declarative, interrogative, exclamatory, and so on. In declarative sentences, one often finds yǐ 矣 (a particle of the perfect aspect), yě 也 (transforming a statement into an assertion, a judgment), ěr 耳, and yān 焉. Hū 乎, yú 欤, yě 也, and sometimes zhě 者, are more typically used in interrogative sentences. Zāi 哉 occurs in exclamatory sentences.

5. Lexicon The overall lexicon of Ancient Chinese is quite different from that of Contemporary Chinese. The former is composed of: (i) words that are still attested in the contemporary language, like shān 山 ‘mountain’ or shuǐ 水 ‘water’; (ii) words that only exist in the ancient language and have disappeared from the modern language, such as yuē 曰 ‘say’; (iii) words that are still used today, but with different meanings, like zǒu 走 ‘run’ (Ancient Chinese) > ‘walk’ (Contemporary Chinese). Of the three types, the second are rare and the last are numerous (see He and Jiang 1980: 3).

5.1.  Historical development of the lexicon The Ancient Chinese lexicon has changed considerably since the Pre-Archaic period. From the vocabulary of everyday life (lexemes for food, clothing, housing), we find only 15 words in the oracle bone inscriptions (fourteenth–eleventh centuries bce) that are still used today, 71 in the bronze inscriptions (tenth–sixth centuries bce), and 297 in Shuō wén jiě zì 说文解字 (Explain the figures [single characters] and interpret the characters [compound characters], second century ce). On the varying richness of Chinese vocabulary in different periods, see He and Jiang (1980: 9). According to these two authors (1980: 136–7), Classical Chinese has an identifiable basic vocabulary of about 2,000 full words, of which 1,100 occur quite commonly. From four major works of the Late Archaic period, including Lún yǔ《论语》(Confucian Analects) and Mèngzǐ《孟子》(Mencius), they have isolated 4,466 distinct words, estimating that about half are proper nouns such as personal names or place names. There is no implication that the vocabulary of Classical Chinese is impoverished compared with that of Con­temporary Chinese – simply different. For example, there is only a single verb meaning ‘to wash’ in Contemporary Chinese (xǐ 洗), whereas there are five in Classical Chinese: mù 沐 ‘to wash (the hair)’, yù 浴 ‘to wash (the body)’; hui 沬 ‘to wash (the face)’; zǎo 澡 ‘to wash (the hands)’; xǐ 洗 ‘to wash (the feet)’. During the long history of Ancient Chinese, several different processes have led to changes in the lexicon. The major processes of internal development include: (i) compounding, a highly productive process beginning in the first century ce (Han period); (ii) semantic extension (e.g. zú 卒 ‘foot soldier’ > zú 卒 ‘all sorts of soldiers’); and (iii) semantic reduction (e.g. zǐ 子 ‘child’ (boy or girl) > zǐ 子 ‘son’). In addition, the Ancient Chinese lexicon was enlarged by borrowing words from other languages.

5.2.  Inherited elements and loanwords There has been a strong tendency in the past to view the Ancient Chinese lexicon as a monolithic linguistic entity, resistant to influences from all surrounding foreign languages. This is certainly a fallacy. Without going as far as Norman (1988: 17), who states: ‘the fact

14

Ancient Chinese

that only a relatively few Chinese words have been shown to be Sino-Tibetan may indicate that a considerable proportion of the Chinese lexicon is of foreign origin’, we can undoubtledly rightly assert that the Ancient Chinese lexicon contains numerous loanwords. Nevertheless, the identification of such words and their sources is often uncertain. Below we mention a few uncontroversial examples of loanwords. There are two common words for ‘dog’ in Ancient Chinese: quǎn 犬, which is probably the native Chinese word, and gǒu 狗, which appears at the end of the Warring States period. Gǒu 狗 is a loanword from a language ancestral to the Modern Miao-Yao languages (Hmong-Mien). The word 虎 hǔ ‘tiger’ might have been borrowed from an Austronesian language in prehistoric times (see Norman 1988: 17–20). Other words of non-Chinese origin are xiàng 象 ‘elephant’ (borrowed from a Tai language?), pútáo 葡萄 ‘grape’ (from Old-Iranian?), mòlì 茉莉 ‘jasmine’ (from Sanskrit), luòtuó 骆驼 ‘camel’ (possibly from an Altaic language).

6. Conclusion The above chapter on syntax and lexicon is concerned chiefly with the Classical Language, as represented by the Warring States period, i.e. the Late Archaic period (fifth–second centuries bce). From the time of the Early Medieval period (second–sixth centuries ce), one can consider that the vernacular language is actually distinct from the literary, deserving a separate description of its own. For example, certain important grammatical structures which did not exist in Classical Chinese are found to have developed in this later vernacular form, prior to the sixth century ce. For a detailed review of the developments in the language between these two stages, on the birth of the disposal construction (NP1-Agent + OM [object marker: qǔ 取, chí 持, zhuō 捉, jiāng 将, bǎ 把] + NP2-patient + VP), on the locative PPs, on the resultative construction, or on the birth and development of the classifier system, etc., see Peyraube (1996).

Acknowledgements This research is part of a project entitled ‘Typologie des processus synchronique et diachronique en Min-Sud (langue sinitique)’ funded by the French National Research Agency’s (ANR) ‘blue sky’ programme blanc: TYSOMIN n° 11-ISH2-001-01, 2012–14, projet bilatéral France (ANR) et Taiwan (National Research Council), 2013–15. I would also like to thank Hilary Chappell for her comments and questions.

List of abbreviations 1sg = 1st singular personal pronoun 3sg = 3rd singular personal pronoun 3pl = 3rd plural personal pronoun fin-part = final particle int-part = interrogative particle lig = ligature neg = negation om = object marker passive = passive marker

15

Alain Peyraube

References Baxter, William H. (1992) A Handbook of Old Chinese Phonology, Berlin: Mouton de Gruyter. Baxter, William H. and Laurent Sagart (1998) ‘Word Formation in Old Chinese’, in Jerome Packard (ed) New Approaches to Chinese Word Formation: Morphology, Phonology and the Lexicon in Modern and Ancient Chinese, Berlin: Mouton de Gruyter, 35–76. Chappell, Hilary and Alain Peyraube (forthcoming) ‘Mood and Modality in Sinitic Languages’, in J. van der Auwera and Jan Nuyts (eds) The Oxford Handbook of Mood and Modality, Oxford: Oxford University Press. Comrie, Bernard (1989) Language Universals and Linguistic Typology (2nd edition), Oxford: Basic Blackwell. Djamouri, Redouane (2001) ‘Markers of Predication in Shàng Bone Inscriptions’, in Hilary Chappell (ed) Sinitic Grammar – Synchronic and Diachronic Perspectives, Oxford: Oxford University Press, 143–71. Downer, Gordon (1959) ‘Derivation by Tone-change in Classical Chinese’, Bulletin of the School of Oriental and African Studies 22(2): 258–90. Guan, Xiechu (1981) Xi-Zhou jinwen yufa yanjiu《西周金文语法研究》(Research on the grammar of the bronzes of the Western Zhou), Beijing: Shangwu yinshuguan. Harbsmeier, Christoph (1981) Aspect of Classical Chinese Syntax, London: Curzon Press. He, Jiuying and Jiang Shaoyu (1980) Gu hanyu cihui jianghua《古汉语词汇讲话》(Talks on the lexicon of Ancient Chinese), Beijing: Beijing chubanshe. LaPolla, Randy (1993) ‘On the Change to Verb-Medial Word Order in Proto-Chinese: Evidence from Tibeto-Burman’, Paper presented at the 26th International Conference on Sino-Tibetan Languages and Linguistics, Osaka. Li, Charles N. and Sandra A. Thompson (1974) ‘An Explanation of Word Order Change: SVO > SOV’, Foundations of Language 12(2): 201–14. Liu, Jian and Alain Peyraube (1994) ‘History of Some Coordinative Conjunctions in Chinese’, Journal of Chinese Linguistics 22(2): 179–201. Mei, Tsu-lin (1980) ‘Sisheng bieyi zhong de shijian cengci’〈四声别义中的时间层次〉(‘Chronological Strata in Derivation by Tone-change’), Zhongguo yuwen《中国语文》(Studies of the Chinese Language) 6: 427–43. Mei, Tsu-lin (1989) ‘The Causative and Denominative Functions of the *s- Prefix in Old Chinese’, in Proceedings of the Second International Conference on Sinology, Taipei: Academia Sinica, 31–51. Mei, Tsu-lin (1997) ‘Typological Changes in the History of the Chinese language’, Chinese Languages and Linguistics 4: 85–104. Norman, Jerry (1988) Chinese, Cambridge: Cambridge University Press. Peyraube, Alain (1987) ‘The Double-object Construction in Lunyu and Mengzi’, in Wang Li Memorial Volumes, Hong Kong: Joint Publishing, 331–58. Peyraube, Alain (1989) ‘History of the Passive Construction in Chinese until the 10th Century’, Journal of Chinese Linguistics 17(2): 335–72. Peyraube, Alain (1996) ‘Recent Issues in Chinese Historical Syntax’, in James C. T. Huang and Audrey Y. Li (eds) New Horizons in Chinese Linguistics, Dordrecht: Kluwer Academic Press, 161–213. Peyraube Alain (1997a) ‘On Word Order and Word Order Change in Pre-Archaic Chinese’, Chinese Languages and Linguistics IV: 105–24. Peyraube Alain (1997b) ‘On Word Order in Archaic Chinese’, Cahiers de Linguistique – Asie Orientale 26(1): 3–20. Peyraube, Alain (2004) ‘Ancient Chinese’, in Roger D. Woodard (ed) The Cambridge Encyclopedia of the World’s Ancient Languages, Cambridge: Cambridge University Press, Chapter 41, 988–1,014. Peyraube, Alain and Shen Zhongwei (forthcoming) History of the Chinese Language, Cambridge: Cambridge University Press. Peyraube, Alain and Thekla Wiebusch (1995) ‘Problems Relating to the History of Different Copulas in Ancient Chinese’, in Matthew Y. Chen and Ovid J. L. Tzeng (eds) Linguistic Essays in Honor of William S.-Y. Wang, Taipei: Pyramid Press, 383–404. Peyraube, Alain and Wu Fuxiang (2005) ‘Origin and Evolution of Question-words in Archaic Chinese: A Cognitive Approach’, Cahiers de linguistique – Asie Orientale 34(1): 3–24.

16

Ancient Chinese Sagart, Laurent (1993) ‘L’infixe –r– en chinois archaique’, Bulletin de la Société Linguistique de Paris 88(1): 261–93. Sagart, Laurent (1999) The Roots of Old Chinese, Amsterdam and Philadelphia: John Benjamins Pubishing Company. Shen, Pei (1992) Jiagu buci yuxu yanjiu《甲骨卜辞语序研究》(Word order in the Oracle Bone Inscriptions), Taipei: Wenjin chubanshe. Wang, Li (1979) Gudai hanyu changshi《古代汉语常识》(Elements of Ancient Chinese), Beijing: Renmin jiaoyu chubanshe.

17

2 Cantonese Cheng Siu-Pong and Tang Sze-Wing The Chinese University of Hong Kong, Hong Kong, China

The old language–dialect debate is of particular relevance to Cantonese, or Yue, given the ambiguous status it holds in places like Hong Kong. This leads to an inquiry into its distinctive features. They might be connected to its origin, which relates to its divergence from other Chinese varieties. Whereas the subclassification of Yue reflects the intra-group variation, the comparison between Cantonese and Mandarin reveals not only their differences in sounds and words, but also in the lesser-known aspects of grammar.

1.  Cantonese: dialect or language? In English, the term ‘Cantonese’ either refers to the speech of Canton, modern-day Guangzhou city of Guangdong province, or the entire Yue group of similar tongues, or just any variety within the group. It is often used as a convenient term for the mainstream Cantonese spoken in Hong Kong. The native speakers of Cantonese (in the broad and narrow sense) are overwhelmingly of Han Chinese ethnicity.1 Depending on the communities, they refer to what they speak as gwongzauwaa / gwongfuwaa ‘Guangzhou speech’, gwongdungwaa ‘Guangdong speech’, jyutjyu ‘Yue language’, baakwaa ‘colloquial speech’, tongwaa ‘Tang (dynasty) speech’, or simply as zungman ‘Chinese language’.2 Like any other varieties of Chinese, the definition of Cantonese is tied to age-old disputes over whether it should be classified as a language or a dialect. Linguistically, it is not at all difficult to define ‘dialect’: a dialect is a subdivision of language. At issue is how the relationship between such terms as ‘Chinese’ and ‘Cantonese’ is incorporated into these notions. If ‘language’ is defined as a set of mutually intelligible varieties, many Chinese varieties, Cantonese included, can be conceived as separate languages. There is a very low degree of mutual intelligibility between Mandarin and Cantonese. In fact, the major ‘dialect groups’ of Chinese are all unintelligible to one another. This interpretation of ‘language’ implies that the term ‘Chinese’, when encompassing Cantonese (as well as Mandarin, etc.), should be treated as a language family rather than a single language. Thus, Cantonese is said to be a member of the Chinese or Sinitic languages. In most cases where Chinese is so defined, Cantonese is also deemed a language. None­ theless, there is a need for this Cantonese language to exclude a subset of in-group varieties. This is because the group of Yue varieties contains a number of subdivisions whose mutual intelligibility is at best marginal. In this regard, the kind of Cantonese spoken by the great 18

Cantonese

majority of Hong Kong people is a single language. The mainstream Cantonese as used in Guangzhou, Hong Kong, and Macao can also be grouped as one language, given their close similarities, but outside this subgroup which they form (to be discussed below), the inclusion of other varieties depends on one’s judgment on how similar they should be in order to be considered ‘intelligible’. The status of Cantonese also relates to the standard remark that ‘a language is a dialect with an army and a navy’, a quote attributed to Max Weinreich. Spoken in China, which now includes the two subnational entities of Hong Kong and Macao, Cantonese is also widely spoken by the Chinese diaspora in various parts of southeast Asia (Malaysia in particular) and the Western world (notably in Vancouver, Canada). Cantonese is not named as a national language in any of these countries, and no national or subnational governments explicate in their constitutional document that ‘Cantonese’ (the exact wording) serves as an official language. While the constitution of the People’s Republic of China does not touch on the issue of national or official languages, the Law on the Standard Spoken and Written Chinese Language, effective 2001, states that ‘the standard spoken and written Chinese language means Putonghua (a common speech with pronunciation based on the Beijing dialect) [.  .  .]’. Although Hong Kong is part of China, this law, together with the vast majority of other national laws, does not apply in the territory. Instead, the constitutional Basic Law states that ‘[i]n addition to the Chinese language, English may also be used as an official language by the executive authorities, legislature and judiciary of the Hong Kong Special Administrative Region’. A similar article is found in Macao’s Basic Law, with ‘Portuguese’ stated in place of ‘English’. The ambiguity of using ‘Chinese’ in the Basic Laws as well as other local laws (e.g. Hong Kong’s Official Languages Ordinance) raises questions over whether Cantonese per se is official. There is little doubt that in Hong Kong and Macao, Cantonese is the de facto official spoken variety of Chinese. Cantonese is spoken in courts and legislative council meetings almost whenever ‘Chinese’ is used there. Even so, the answer to the question of whether Cantonese by itself attains de jure status is not clear. There is one complicating factor. In formal contexts, government and business documents published in Chinese are given in standard written Chinese, which is based on Mandarin and comprehensible to all literate Chinese speakers worldwide. Written Cantonese, i.e. the written form of Cantonese expressed in Chinese characters, is mostly used in informal register. At first glance, this helps prove the subordinate status of Cantonese. However, since the Chinese script is a logographic writing system, the characters for standard written Chinese can be read with their Cantonese pronunciations. It is easier to come up with the view that it is rather the literary counterpart of the colloquial speech (known as shumianyu ‘literary language’), and its use does not undermine the status of Cantonese. In sociolinguistic terms, Cantonese is obviously an abstand language with respect to another Chinese variety, e.g. Mandarin; the variation between Cantonese and Mandarin in sounds and vocabulary is so great that they are on par with such European languages as French and Spanish. At issue is to what degree Cantonese is ‘autonomous’ enough to be an ausbau language. In mainland China, the answer is simple: Cantonese is considered by the Chinese government as a ‘local dialect’. In the best scenario, the official policies tolerate (but never promote) the use of Cantonese in broadcasting and, to a lesser degree, in schools. In Hong Kong and Macao, Cantonese is used in a variety of domains, ranging from educational to official settings. Nonetheless, given the unresolved issues surrounding the meaning of ‘Chinese’, the Cantonese variety spoken there should be viewed as more an abstand than an ausbau language. The Weinreichian notion of ‘language’ does not accord well with the reality of Cantonese regardless of where it is spoken. A final remark is on the term ‘dialect’ itself. While most Chinese-language publications indicate that 19

Cheng Siu-Pong and Tang Sze-Wing

Cantonese is a dialect, the original term they use in Chinese is fangyan, which literally means ‘regional speech’. It is construed as a notion relative to (and often subordinate to) gongtongyu ‘common language’. The possible semantic discrepancies between ‘dialect’ and fangyan had led DeFrancis (1984) to coin and adopt the term regionlect to capture the sense of the original word.3 This is another way to get over the dialect–language dichotomy. In the following sections, this dichotomy is set aside, and Yue and Cantonese are variably called ‘dialect’ or ‘language’ in a non-committal manner.

2.  Origin and distribution Cantonese, or Yue, belongs to the Sino-Tibetan language family. While all natural languages are descended from some proto-language, the discussions of the origin here focus on the emergence of Yue dialects in present-day southeast China where they are chiefly spoken. A vast area of southern China was known to be occupied by the Bai Yue ‘One Hundred Yue’ tribes. There is still no consensus as to which languages these people spoke in the absence of written records. They were speakers of Austroasiatic-, Kam-Tai-, Hmong-Mien-, or possibly Austronesian-related languages (Norman and Mei 1976; Blust 1984–1985; J. Li 1990; LaPolla 2001), or there was an assortment of speakers from some or all of these linguistic groups. Historical records show that Han Chinese migrated from their homeland in the north to this area starting from the third century bce. While most scholars agree that Modern Cantonese is descended from an earlier form of Chinese, it is still a matter of great debate as to what extent the Chinese variety first spoken in this area bears a relation to Modern Cantonese. The divergence of the earliest Yue variety from those of other parts of China is argued to have taken place from as early as the third century bce when substantial numbers of Han Chinese arrived (J. Li 1990) to as late as the thirteenth century ce (Lau 2000). In those accounts that posit a later divergence time, it is argued that another wave of Han Chinese migration was vital to its formation. Historical studies are also concerned about the influence of non-Han languages on the early Yue. Evidence based on some colloquial words has been provided to show that they form Tai or Hmong substrata (see Yue-Hashimoto 1976, 1991; Bauer 1987; J. Li 1990, etc.). The substratum influence of non-Han languages can be expressed in more neutral terms: Yue has experienced a prolonged period of language contact vis-à-vis non-Han languages. In modern times, the main area for the Yue dialect group is the central and western parts of Guangdong province and the southeastern part of Guangxi Zhuang Autonomous Region. This area also includes Hong Kong and Macao, which are now the Special Administrative Regions of the People’s Republic of China. The Yue dialects are spoken by around 4 to 5% of China’s 1.3 billion people. Language Atlas of China (2012) gives a figure of approximately 59.58 million speakers. In southeast Asia, there are considerable populations of Yue speakers in Malaysia, Vietnam, Singapore, and Indonesia. In particular, around 5% of Malaysia’s population are from the Cantonese speech group (Tan 2005), and an even larger proportion of Malaysians can speak Cantonese with varying degrees of fluency. Outside Asia, the Yue speakers amount to around 1% or more in Canada, the United States, Australia, and New Zealand. According to Language Atlas (1987), there were approximately 11 million overseas speakers in total during the time surveyed. Language Atlas (2012) states that there are over 68 million Yue speakers worldwide. Major works on Yue or Chinese dialects in general, such as Zhan (1991, 2002), Yuan et al. (1960/2001), X. Li (1994), and Language Atlas (1987, 2012), differ on the subclassification of Yue dialects. The following classification includes the four major Yue branches of: (i) Yuehai; (ii) Siyi; (iii) Gao-Yang; and (iv) Guinan; this 20

Cantonese

classification is ‘eclectic’ in nature given that these four branches are not subordinate to any others in the above works, while other branches, e.g. Goulou Branch, are subsumed elsewhere into one of them.4

2.1.  Yuehai Branch Also called the Guangfu Branch, it is mainly spoken in the Pearl River (Zhujiang) Delta. This area includes Guangzhou, the capital of Guangdong province, together with Hong Kong and Macao. For this reason, this branch is named after Guangzhou, otherwise known as Yuehai or Guangfu. Yuehai Branch is the most well-known and widely spoken branch of Yue dialects. Language Atlas (2012) gives a figure of about 28.34 million speakers (excluding the speakers of the Goulou Branch; to be discussed below). The same source indicates that approximately 20.72 million of them live in Guangdong. In addition, around 7 million people from Hong Kong and Macao can speak this variety. Depending on the extension of this subgroup, its area also covers the central and northern parts of Guangdong, and scattered parts of Guangxi such as downtown Hezhou and Wuzhou. As a result of migration from this core area, the Yuehai Branch is the predominant form of Yue spoken in overseas Chinese communities. In Malaysia, it is a major Chinese variety spoken in Kuala Lumpur, Ipoh, Kuantan, Seremban, and Sandakan. In most Chinatowns in North America, the Yuehai Branch has been the major Chinese variety (amid a growing number of Mandarin-speaking immigrants), often achieving its prominent status at the expense of the Siyi Branch. Particularly in Metro Vancouver, Canada, its use has been scaled up to the city level. The city of Richmond is an example par excellence. Accentual difference is found between the varieties spoken in downtown Guangzhou and the outskirt areas of the city, especially those districts newly (re-)attached to Guangzhou around this millennium, e.g. Panyu. A variety closely similar to the one used in downtown Guangzhou is also predominantly spoken in Hong Kong. Guangzhou and Hong Kong Cantonese are mutually intelligible to a very high degree, despite some lexical differences. However, it should be noted that the emergence and rise of Hong Kong Cantonese is resulted from the combined factors of immigration and language shift. Right before the beginning of the colonial time in the nineteenth century, a different Yue dialect, i.e. the Guan-Bao variety, was chiefly spoken. A gradual shift to Guangzhou-style Cantonese also took place in Macao. The rise of Cantonese to greater prominence in the latter half of the last century was mainly due to the social and economic development of Hong Kong. The extended influence of Hong Kong’s trade and entertainment industries has promoted the use of Cantonese in a greater area of Guangdong and overseas Chinese communities. As a result, many people in traditionally non-Cantonese-speaking regions, such as the eastern part of Guangdong, came to understand it and often speak it as a second language.

Guan-Bao Branch Zhan (1991) classifies it as a separate branch of Yue dialects. It is traditionally spoken in the eastern part of the Pearl River Delta. Precisely, it is the indigenous Yue variety of Hong Kong and Shenzhen, formerly parts of Xin’an (later named Bao’an), and of the neighboring Dongguan, hence the name ‘Guan-Bao’. In Hong Kong, it is known as Punti or waitauwaa. The number of its speakers is diminishing to the point that it has become endangered, and its use is now restricted to the older generation in the more rural parts of Hong Kong. A decline in use is also observed in the rapidly developing Shenzhen and Dongguan. The 21

Cheng Siu-Pong and Tang Sze-Wing

mutual intelligibility between Guan-Bao and Guangzhou Cantonese is less than ideal. Although downtown Guangzhou and Hong Kong are about 130 kilometers apart, the mainstream Cantonese in Hong Kong is much closer to Guangzhou Cantonese than it is to Guan-Bao.

Xiangshan Branch Another independent branch of Yue given in Zhan (1991), it is spoken in the western part of the Pearl River Delta, including Zhuhai and Zhongshan, which had previously been called Xiangshan before it was named after Sun Yat-sen (Sun Zhongshan). Formerly, the Xiangshan Branch was also spoken in Macao.

Goulou Branch First proposed around the time when the first edition of Language Atlas (1987) was published, it is spoken in the peripheral areas outside the Pearl River Delta, to the north and west of the heartland of Yuehai Branch. This area extends into Guangxi, and occupies the eastern portion of the region. It is named after the mountain range of Goulou in southeastern Guangxi bordering Guangdong. According to Language Atlas (2012), the number of Goulou speakers amounts to approximately 10.30 million.

2.2.  Siyi Branch Siyi ‘four counties’ refers to the region southwest of Guangzhou that lies to the west of the Pearl River Delta. The ‘four counties’ include Taishan, Kaiping, Enping, and Xinhui. In modern times, they are all under the jurisdiction of the city of Jiangmen, which also includes Heshan, so together they are called Wuyi ‘five counties’. The population of Siyi speakers in Guangdong is approximately 3.88 million, according to Language Atlas (2012). The main Siyi-speaking area includes modern-day Jiangmen and the western part of Zhuhai. Migration of Siyi speakers out of this region contributed to its prominence elsewhere. In Hong Kong after the Second World War and the subsequent Chinese Civil War, the influx of Siyi immigrants led to a time when the Siyi Branch was more spoken before it was largely displaced by Hong Kong Cantonese a few decades later. Besides, the Siyi Branch was once the most influential Chinese variety in the American Chinese communities. Up until half a century ago, it was spoken by the great majority of Chinese immigrants to the United States (Bauer and Benedict 1997). Within the Siyi Branch, the Taishan subdialect was the most prominent variety spoken overseas. Although its status was largely replaced by the Yuehai Branch and Mandarin, the use of Siyi remains significant in some American Chinatowns, especially among the earlier migrants.

2.3.  Gao-Yang Branch Also called the Gao-Lei Branch, it is the predominant Chinese variety spoken in the southwestern part of Guangdong, which includes Gaozhou and Yangjiang. The core areas of Gao-Yang, however, exclude the Leizhou Peninsula, together with some linguistic exclaves nearby, where the Gao-Yang Branch might be spoken but Hakka or Min dialects are the main speech. Language Atlas (2012) states that the Gao-Yang Branch is spoken by approximately 6.43 million people. The Wu-Hua Branch, considered a separate branch in Language Atlas, has about 1.28 million speakers. It is classified as part of the Gao-Yang-speaking region in the other accounts. 22

Cantonese

2.4.  Guinan Branch Guinan literally means ‘southern Guangxi’. The first edition of The Encyclopedia of China uses it as an umbrella term to cover all subdivisions of Yue dialects spoken in Guangxi. According to Language Atlas (2012), approximately 16.86 million Guangxi people speak the Yue dialects. The Yue varieties in Guangxi are further divided into the branches of: (i) Yuehai; (ii) Goulou; (iii) Yong-Xun; and (iv) Qin-Lian. The Goulou Branch is spoken in a continuous area across Guangdong and Guangxi, and the predominant use of Yuehai Branch in Guangxi is found in the linguistic enclaves, or ‘dialect islands’. Only the Yong-Xun and Qin-Lian Branches are specific to Guangxi.

Yong-Xun Branch It is used as the majority variety in scattered places along the rivers of Yongjiang and Xunjiang. Most of these places have served as inland river ports. Language Atlas (2012) gives a figure of approximately 4.69 million speakers.

Qin-Lian Branch It is spoken in the traditional areas of Qinzhou and Lianzhou (modern-day Hepu) along the southern (and the only) coast of Guangxi. Situated to the west of Guangdong coastal towns, this region was part of Guangdong until after the founding of the People’s Republic of China. According to Language Atlas (2012), the number of speakers of this branch amounts to approximately 3.90 million.

3.  Differences between Cantonese and Mandarin It has been well documented that Cantonese differs from Mandarin in terms of sounds and vocabulary. In general, the sound system of Yue dialects is comparatively more complicated by having more rhymes and tones. Bauer and Benedict (1997) identify 58 rhymes in Cantonese including two syllabic nasals. By contrast, a standard Mandarin rhyme table includes 35 rhymes. A major reason for the difference lies in the fact that in Cantonese /p, t, k, m/ are found in the syllable-final position, yet these codas are absent in Mandarin. Regardless of tones, a regular Mandarin dictionary contains a number of 410 syllables, which is less than the 584 Cantonese syllables given in Bauer and Benedict (1997) for the standard Chinese characters, not to mention others solely used for colloquial speech. The number of tonal categories in Yue dialects varies from six to ten (Zhan 1991). As for Hong Kong Cantonese (as well as other Yue varieties), the category of Entering tones (tones for syllables with a coda from /p, t, k/) might be subsumed into the other categories, depending on the categorization criteria, and the Entering tones are viewed as the short or extra-short variants of the other tones (Zee 1999). Even so, it contains a total of six tones. This contrasts with Mandarin, which has only four. Historically, tonal split and merger have given rise to varying numbers of tones across different Chinese varieties. As in other modern dialects, Yue experienced the split of tonal categories primarily as a result of the loss of voicing distinction of initials. Subsequent tone mergers, however, were more restrained in the case of Yue dialects. In modern-day Hong Kong Cantonese, there are 19 consonants, including plosives /p, ph, t, th, k, kh, kw, kwh/, nasals /m, n, ŋ/, affricates /ts, tsʰ/, fricatives /f, s, h/, and ‘semi-vowels’ as well as other approximants /j, w, l/. Vowels include both front and back, 23

Cheng Siu-Pong and Tang Sze-Wing

and high and low monothongs: /i, e, y, u, o, ɛ, ɔ, œ, ɵ, ɐ, a/. In addition to these monothongs, there are also ten gliding vowels (or diphthongs): /iw, ɵʮ, uʮ, ej, ɔj, ow, ɐj, ɐw, aj, aw/. The maximal syllable takes the form of ‘initial consonant + vowel + final consonant + tone’, or CiVCf /T (Bauer and Benedict 1997). In other words, a consonant cluster is not found within an indigenous syllable. Both the initial and final consonants might be absent, and a final consonant (from /m, n, ŋ, p, t, k/) is disallowed when the vowel is a gliding one. A syllable always contains a vowel, with the exception of two syllabic nasals: /m, ŋ/. Assuming that the three Entering tones, namely Upper Dark Entering /5/, Lower Dark Entering /3/, and Light Entering /2/, are respectively subsumed into /55/, /33/, and /22/, the modern-day Hong Kong variety of Cantonese contains a total of six tones, which are /55/, /25/, /33/, /21/, /23/, and /22/.5 In terms of vocabulary, Hou (2002) indicates that about two thirds of basic words in Yue dialects are common to both Yue and Standard Chinese, or, in his wording, the ‘common national language’. Out of the other one third or so of words that are different, over 90% of them were arisen as a result of the innovation of Yue speakers, whereas some others are old Chinese cognates that are no longer in use in modern Standard Chinese. Many scholars also note that some of the colloquial words owe their origin to non-Han languages including Tai or Hmong. The Yue dialects of major ports, notably Hong Kong Cantonese, stand out from other Chinese varieties by having more words borrowed from a European language, English in particular. Apart from English, Japanese is another major loan language. These words entered the Cantonese vocabulary in the form of semantic translation, phonetic borrowing, or in the case of most Japanese loanwords, the direct use of Japanese-derived words with the Chinese characters (kanji) read in Cantonese pronunciation. Structurally, the relative order of morphemes in a disyllabic word in Mandarin is sometimes reversed in Cantonese. Examples include ganjiu ‘important’ and siuje ‘night-time snack’. In Mandarin, the morpheme order is reversed and they are rendered respectively as yaojin and yexiao. Another well-quoted example is gaigung ‘rooster’. Although gunggai ‘male + chicken’ is also used, gaigung ‘chicken + male’ is indigenous. The morphemic ordering of this indigenous word in Yue and also in other Southern dialects has drawn speculation that the influence of non-Han languages was at work (see Yuan et al. 1960; Ouyang 1991, etc.). Other morphological variation includes the higher tendency for Cantonese, as compared to Mandarin, to have monosyllabic words, and the lower tendency for it to use certain suffixes, e.g. zi. It is often assumed that all Chinese varieties share the same grammar. Despite the general perception of close similarities in grammar among them, there are significant differences that also set them apart. Functional words are important contributors to the syntactic variation. The following parts are based on Tang and Cheng (2014) and focus on the distinctive grammatical features of Cantonese that are less salient or cannot be found at all in Mandarin. The variables involved include (i) structural particles; (ii) directional verbs; (iii) aspect markers; (iv) definiteness; (v) double-object construction; (vi) comparative sentences; (vii) passives; and (viii) post-verbal elements.

3.1.  Structural particles Homophonous particles in Mandarin might be audibly differentiated when translated into Cantonese. Zhu (1961) classifies three types of de in Mandarin: de1 follows an adverb to form an adverbial unit; de2 follows an adjective to form an adjectival phrase; de3 follows different phrases to form a nominal unit. 24

Cantonese

(1) Ta zhengtian buting-de shuo. he all-day non-stop-DE speak ‘He speaks non-stop all day’

de1-adverbial

honghong-de. (2) Zhe duo hua this Cl flower red-DE ‘This flower is a bit red’

de2-adjectival

(3) Xinxian-de ye you. fresh-DE also have ‘There is also something fresh’

de3-nominal

Zhu (1980) further suggests that the three subcategories of de correspond to gam, dei, and ge in Cantonese. The three sentences can be translated into Cantonese as follows. (4) Keoi sengjat mting-gam gong. he all-day non-stop-GAM speak ‘He speaks non-stop all day’

gam-adverbial

(5) Ni do faa hunghung-dei. this Cl flower red-DEI ‘This flower is a bit red’

dei-adjectival

(6) Sansin-ge dou jau. fresh-GE also have ‘There is also something fresh’

ge-nominal

In Cantonese, gam, dei, and ge are clearly separate morphemes as they differ in their pronunciation. With this in view, Zhu argues for the three subsets of de in Mandarin. As far as these structural particles are concerned, Cantonese is more ‘marked’ in their division of labor.

3.2.  Directional verbs Yuan et al. (1960) note that Mandarin tends to use shang ‘go up’ or dao ‘arrive’ as the main verbs, and qu ‘go’ and lai ‘come’ usually follow the object, as they do in (7) and (8). In Cantonese, on the other hand, heoi ‘go’ and lai ‘come’ are mostly used directly as the main verb, as in (9) and (10). (7) Wo shang Beijing qu. I go-up Beijing go ‘I go to Beijing’

(Mandarin)

(8) Wo jiejie shi zuotian dao Guangzhou lai de. I sister be yesterday arrive Guangzhou come DE ‘My elder sister came to Guangzhou yesterday’ (9) Ngo heoi Bakging. I go Beijing ‘I go to Beijing’

(Cantonese)

25

Cheng Siu-Pong and Tang Sze-Wing

(10) Ngo gaaze hai camjat lei Gwongzau ge. I sister be yesterday come Guangzhou GE ‘My elder sister came to Guangzhou yesterday’ According to Yuan et al., the use of qu ‘go’ and lai ‘come’ as the main verbs is also found in Mandarin, but this is obviously under the influence of the Southern dialects.6

3.3.  Aspect markers C. N. Li and Thompson (1989) identify four Mandarin aspect markers, le, zai, zhe, and guo: the perfective aspect is marked by le, the progressive aspect by zai, the durative aspect by zhe, and the experiential aspect by guo. Cheung (1972/2007) identifies six Cantonese aspect markers, in addition to a zero marker. They are perfective zo, experiential gwo, progressive gan, habitual hoi, durative zyu, and inchoative hei(soeng)lai. Matthews and Yip (1994/2011) and X. Li et al. (1995) added two more: delimitative haa and successive lokheoi. A linkage can be drawn between Mandarin and Cantonese perfective, experiential, and progressive / durative markers. Nonetheless, it is worth noting the distinction between Mandarin progressive / durative zai / zhe and Cantonese progressive / durative gan / zyu. Previous studies indicate that activity verbs take zai and stative verbs take zhe (Paris 1981; C. N. Li and Thompson 1989). Smith (1991/1997) also notes that in some Mandarin varieties the use of zhe has gradually become more extended and can convey both the progressive and durative aspect. In contrast, neither gan nor zyu has taken over the function of one another in Cantonese. While gan is a typical progressive maker, zyu has a static meaning and is used as a marker for durative aspect, as in (11) and (12). bo. (11) Zoengsaam  daa-gan Zoengsaam  play-Prog ball ‘Zoengsaam is playing ball’ (12) Zoengsaam  daai-zyu ngaangeng. Zoengsaam  wear-Dur eyeglasses ‘Zoengsaam is wearing eyeglasses’ While Mandarin zhe can be used in the equivalents of both sentences above, Cantonese progressive gan and durative zyu are unambiguously distinctive and cannot be used interchangeably.

3.4. Definiteness Like Mandarin classifiers, Cantonese classifiers precede nouns. In both varieties, the [classifier + noun] sequence follows a numeral, which itself might be preceded by a demonstrative, as in the following examples. (13) zhe san ben shu this three Cl book ‘these three books’

(Mandarin)

(14) ni saam bun syu this three Cl book ‘these three books’

(Cantonese)

26

Cantonese

The addition of demonstratives and numerals is not compulsory. The bare [classifier + noun] form is possible in both Mandarin and Cantonese. As indicated by Cheng and Sybesma (1999), the difference between Cantonese and Mandarin in this nominal phrase is a contrast in the interpretations. In Cantonese, a nominal phrase of this form can either have a definite or indefinite reading. In Mandarin, only the indefinite reading is available. Therefore, the [classifier + noun] form is found in the subject position of Cantonese, but not Mandarin, given that this position normally limits the interpretation of nominals to a definite one. (15) Wo xiang mai ben shu. I want buy Cl book ‘I want to buy a book’

(Mandarin)

(16) *Ben shu hen hou. *Cl book very thick (17) Ngo soeng maai bun syu. I want buy Cl book ‘I want to buy a book’

(Cantonese)

(18) Bun syu hou hau. Cl book very thick ‘The book is thick’ On the other hand, Mandarin bare nouns can receive definite, indefinite, or generic readings. Bare nouns in Cantonese can also have indefinite or generic readings, but the definite interpretation is largely prohibitive. For the definite reading, it is better for a bare noun to go with a classifier.7 (19) Gou yao guo malu. dog want cross road ‘The dog wants to cross the road’

(Mandarin)

(20) Zek gau jiu gwo maalou. Cl dog want cross road ‘The dog wants to cross the road’

(Cantonese)

3.5.  Double-object construction It has been noted that Cantonese has a double-object construction different from the one found in Mandarin (Yuan et al. 1960; Gao 1980). An animate object precedes an inanimate one in Mandarin, but in Cantonese the order appears to be in reverse. (21) Ta gei-le wo yi-ben shu. he give-Perf I one-Cl book ‘He gave me a book’

(Mandarin)

(22) Keoi bei-zo jat-bun syu ngo. he give-Perf one-Cl book I ‘He gave a book to me’

(Cantonese)

27

Cheng Siu-Pong and Tang Sze-Wing

In most cases, the inanimate and animate objects are respectively analyzed as the direct object (DO) and indirect object (IO). According to Cheung (1972), Cantonese places the DO before the IO most prevalently when the verb is bei ‘give’. Note that bei can be used in two different positions in the same sentence. The same is true for Mandarin gei. (23) Ta gei-le yi-ben shu gei wo. he give-Perf one-Cl book to I ‘He gave a book to me’

(Mandarin)

jat-bun syu bei ngo. (24) Keoi bei-zo he give-Perf one-Cl book to I ‘He gave a book to me’

(Cantonese)

While the first gei / bei ‘give’ is a verb, the second gei / bei ‘to’ is regarded as prepositional, as it often co-occurs with many other verbs to indicate IO.8 Traditionally, the form of doubleobject construction as illustrated in (22) is known to have been ‘inverted’, since the order found in Mandarin is reversed. On the other hand, more recent studies (Tang 1998a, 2003; Liu 2001) show that the ‘DO + IO’ order in Cantonese actually originated from the ‘DO + bei + IO’ construction through the deletion of bei ‘to’. That is to say, the original form of (22) is (24). Therefore, what distinguishes Mandarin and Cantonese is not the word order of objects, but the deletability of the preposition.9

3.6.  Comparative sentences Generally speaking, Cantonese and Mandarin use different comparison words, which results in different structures for comparative sentences. (25) Zhangsan  bi Lisi  gao. Zhangsan  compare Lisi  tall ‘Zhangsan is taller than Lisi’

(Mandarin)

(26) Zoengsaam  gou-gwo Leisei. Zoengsaam  tall-pass Leisei ‘Zoengsaam is taller than Leisei’

(Cantonese)

In the Cantonese example, the comparison word is gwo ‘pass’, which follows the adjectival predicate gou ‘tall’. The Cantonese comparative sentences are in the form of ‘X + predicate + comparison word + Y’. In the Mandarin example, the comparison word bi ‘compare’ precedes both Y and the adjective; the comparative sentences in Mandarin are in the form of ‘X + comparison word + Y + predicate’. This construction, however, is not impossible in Cantonese. As stated in Matthews and Yip (1994), there is an alternative maker in Cantonese, i.e. bei ‘compare’. It is based on the bi construction in Mandarin, and ‘has the formal flavor of Mandarin-based syntax’, as shown in their example (1994: 167). (27) Aawan  bei keoi muimui leng. Aawan compare she sister pretty ‘Aawan is prettier than her younger sister’ 28

Cantonese

It shows that the order of ‘X + comparison word + Y + predicate’ is also found in Cantonese, although the comparative construction with gwo ‘pass’ is arguably more common and ‘authentically’ Cantonese.10

3.7. Passives Following the distinction drawn by Ting (1995, 1998), passive sentences in Mandarin are classified into what are known as ‘long passives’ and ‘short passives’. A passive morpheme, bei, is used in both categories. Their difference is that an identified agent follows bei in the long passives, but such an agent is absent in the short ones. (28) Zhangsan  bei Lisi  piping-le. Zhangsan  BEI Lisi  criticize-Perf ‘Zhangsan was criticized by Lisi’

(Long Passive)

(29) Zhangsan  bei piping-le. Zhangsan  BEI criticize-Perf ‘Zhangsan was criticized’

(Short Passive)

Long and short passives are not equally acceptable in Cantonese. Unlike its counterpart in Mandarin, the indigenous passive marker bei2 (in high rising tone /25/ ) must contain an agent. Otherwise, the sentence will become ungrammatical. (30) Zoengsaam  bei2 *(ngo) naau. Zoengsaam  BEI I criticize ‘Zoengsaam was scolded (by me)’ Cantonese passives without an identified agent will be acceptable only if the passive morpheme is given in a different form, bei6, with the low level tone /22/. (31) Zoengsaam  bei6 zizaak. Zoengsaam  BEI censure ‘Zoengsaam was censured’ Passives with bei6 are normally used in formal speech and in literary contexts. While bei2 is the native form, bei6 is assumed to be borrowed from Mandarin. As a result, its usage is highly restricted. In everyday Cantonese, the use of long passives, rather than the short ones, is predominant. Even if the agent needs to be ‘hidden’ for some reason, it is more natural to use a long passive containing an unspecified agent, such as jan ‘person’, than to use a short passive with bei6.11

3.8.  Post-verbal elements Cantonese is rich in post-verbal elements in comparison to Mandarin. There are a large number of suffixes and sentence-final particles, and also a high degree of co-occurrence of post-verbal elements. Cheung (1972) identifies 15 predicative suffixes. Matthews and Yip (1994) include seven major aspect particles and 29 verbal particles. The number of sentencefinal particles is even higher. As noted by Fung (2000), previous studies show that it may range from 30 to as many as 90 (Kwok 1984; Ouyang 1990; Leung 2005). Furthermore, 29

Cheng Siu-Pong and Tang Sze-Wing

multiple particles often co-occur in the sentence-final position, as in the following example (Leung 2005: 87). (32)

Nei waa zeoido to do go leng jyut tim ge zaa aamaa? you say at-most delay more Cl more month TIM GE ZAA AAMAA ‘You said (I can) delay for one or more months at most, right?’

As for their usage, Kwok (1984) connects them with sentence functions. The particles are categorized as the signals of speaker’s attitude. Luke (1990) discusses three particles (laa, lo, and wo), which manifest different interactional goals. The particle laa is used to establish common grounds in conversation; lo is associated with interactional achievements; and wo is used to signal expectation and noteworthiness. Fung (2000) groups the majority of sentence-final particles into three main categories, i.e. Z-, L-, and G- families. Particles sharing the same initial belong to the same family, the distinctive core features of which are based on both their semantic and contextual meanings. Z-particles, including ze, zek, zaa, etc., have the core semantic feature of ‘restriction’; the core feature of L-particles, which include lei, laa, etc., is to signal temporal and aspectual meanings; the core function of G-particles, including ge and gaa, is to mark a situation given in a context.12 Cantonese is rich in both pre-verbal and post-verbal elements, and several pre-verbal and post-verbal elements share similar semantic properties. Take caa’mdo ‘almost’ and gamzai ‘almost’ as examples. They can co-occur in the same sentence. (33) Ngo caa’mdo gong-jyun gamzai. I almost speak-finish almost ‘I almost finished talking’ Tang (2006a, 2006b, 2007) suggests that post-verbal elements like gamzai ‘almost’ and pre-verbal ones like caa’mdo ‘almost’ form a constituent known as the ‘discontinuous construction’. He argues that in such a construction, the combination of pre-verbal and postverbal elements ‘appears to be redundant’. That means the apparent redundancy of the pre-verbal and post-verbal elements should be understood as their being close in meaning. So far, over 40 different pre- and post-elements have been identified to belong to this kind of construction. In comparison, the discontinuous constructions are rare in Mandarin.13

4. Implications Many scholars contend that Chinese varieties in essence only differ in sounds and vocabulary (Yuan et al. 1960; Chao 1968; Lü 1982). Since it is easier to observe the phonetic and lexical differences, many previous studies on Chinese dialects focus on these two aspects. Comparatively speaking, the outcomes of syntactic variation studies are less fruitful. On the surface, there are more similarities than differences in the grammar of different Chinese varieties. However, some ‘similarities’ are in fact more apparent than real. A number of obvious grammatical variables have been described, but there is still plenty of room for further research. Given the data from Cantonese as well as other dialects, we can discover many linguistic phenomena that are absent in Mandarin. This allows us to have a better understanding of Chinese as a whole. As shown in Zhu (1980), and Cheng and Sybesma (1999), etc., the comparative studies between Cantonese and Mandarin can help identify the grammatical subtleties of one another. 30

Cantonese

Notes   1 Many of the non-Han speakers of Cantonese belong to the Zhuang ethnic group, which mostly lives in the Guangxi Zhuang Autonomous Region.   2 The Romanization of Cantonese in this chapter is based on Standard Cantonese, and follows the Linguistic Society of Hong Kong (LSHK) Cantonese Romanization Scheme, also known as jyutping.   3 See also Mair (1991) for his proposal of using topolect as ‘an exact, neutral translation of fangyan’.   4 Those varieties whose status is contestable are excluded, e.g. Danzhouhua and Tuhua.   5 Tone numerals are used. The non-Entering tones of /55/, /25/, /33/, /21/, /23/, and /22/ are respectively known as Dark Flat, Dark Rising, Dark Departing, Light Flat, Light Rising, and Light Departing tones by their traditional names. /25/, /21/ and /23/ might be given as /35/, /11/ and /13/. For simplicity, the description above follows Bauer and Benedict (1997).   6 See Yiu (2005) for the spatial characteristics and semantic extension of Cantonese directional verbs.   7 Simpson (2010) observes that some native speakers of Cantonese seem to be able to interpret bare nouns as definite, and gives the following example to illustrate this. Coeng hai mai saan-zo? window be not-be closed-Perf ‘Is the window closed?’ See Y.-H. A. Li (1997), Cheng and Sybesma (1999), Au Yeung (2005), Sio (2006), Wu and Bodomo (2009), and Tang (2011) for the syntactic analysis of definiteness in Cantonese and Mandarin.   8 See Huang and Ahrens (1999) for the contrasting view that the post-DO gei in Mandarin is verbal.   9 See Zhang (2011) for typological evidence to support the deletion approach along these lines. 10 See Mok (1998) for the syntactic analysis of gwo comparatives. 11 See Chin (2009) for the historical development of bei that connects its functions in passives and double-object constructions. 12 See S.-P. Law (1990), Tang (1998b), Y.-K. A. Law (2004), and B. Li (2006) for the proposals on the syntactic positioning of sentence-final particles. 13 See Tang (2009) for the syntax and hierarchical order of discontinuous constructions.

References Australian Academy of the Humanities and Chinese Academy of Social Sciences (1987) Language Atlas of China, Hong Kong: Longman. Au Yeung, Wai-Hoo (2005) ‘An Interface Program for Parameterization of Classifiers in Chinese’, Doctoral Dissertation, Hong Kong University of Science and Technology. Bauer, Robert S. (1987) ‘Kadai Loanwords in Southern Chinese Dialects’, Transactions of International Conference of Orientalists in Japan 32: 95–111. Bauer, Robert S. and Paul K. Benedict (1997) Modern Cantonese Phonology, Berlin: Mouton de Gruyter. Blust, Robert (1984–1985) ‘The Austronesian homeland: a linguistic perspective’, Asian Perspectives 26(1): 45–67. Chao, Yuen-Ren (1968) A Grammar of Spoken Chinese, Berkeley and Los Angeles: University of California Press. Cheng, Lai-Shan Lisa and Rint Sybesma (1999) ‘Bare and Not-so-bare Nouns and the Structure of NP’, Linguistic Inquiry 30(4): 509–42. Cheung, Hung-Nin Samuel (1972/2007) Xianggang Yueyu Yufa de Yanjiu (A Grammar of Cantonese as Spoken in Hong Kong), Hong Kong: The Chinese University Press. Chin, Chi-On (2009) ‘The Verb GIVE and the Double-object Construction in Cantonese in Synchronic, Diachronic and Typological Perspectives’, Doctoral Dissertation, University of Washington. DeFrancis, John (1984) The Chinese Language: Fact and Fantasy, Honolulu: University of Hawaii Press. Fung, Suk-Yee Roxana (2000) ‘Final Particles in Standard Cantonese: Semantic Extension and Pragmatic Inference’, Doctoral Dissertation, Ohio State University. Gao, Huanian (1980) Guangzhou Fangyan Yanjiu (A Study of the Cantonese Dialect). Hong Kong: The Commercial Press. Hou, Jingyi (2002) Xiandai Hanyu Fangyan Gailun (An Outline of Modern Chinese Dialects), Shanghai: Shanghai Jiaoyu Chubanshe.

31

Cheng Siu-Pong and Tang Sze-Wing Huang, Chu-Ren and Kathleen Ahrens (1999) ‘The Function and Category of Gei in Mandarin Ditransitive Constructions’, Journal of Chinese Linguistics 27(2): 1–26. Institute of Linguistics of the Chinese Academy of Social Sciences, Institute of Ethnology and Anthropology of the Chinese Academy of Social Sciences, and Language Information Sciences Research Centre of the City University of Hong Kong (2012) Zhongguo Yuyan Dituji (Language Atlas of China), 2nd ed., Beijing: The Commercial Press. Kwok, Helen (1984) Sentence Particles in Cantonese, Hong Kong: Center of Asian Studies, University of Hong Kong. LaPolla, Randy J. (2001) ‘The Role of Migration and Language Contact in the Development of the Sino-Tibetan Language Family’, in Alexandra Y. Aikhenvald and R. M. W. Dixon (eds) Areal Diffusion and Genetic Inheritance: Problems in Comparative Linguistics, New York: Oxford University Press, 225–54. Lau, Chun-Fat (2000) ‘Xiandai Yueyu yuanyu Songmo yimin shuo’ (‘A Theory that Modern Cantonese Derived from Late Song Dynasty Immigrants’), in Proceedings of 7th International Conference on Yue Dialects, Beijing: The Commercial Press. Law, Sam-Po (1990) ‘The Syntax and Phonology of Cantonese Sentence-final Particles’, Doctoral Dissertation, Boston University. Law, Yan-Kei Ann (2004) ‘Sentence-final Focus Particles in Cantonese’, Doctoral Dissertation, University College London. Leung, Chung-Sum (2005) Dangdai Xianggang Yueyu Yuzhuci de Yanjiu (A Study of the Utterance Particles in Cantonese as Spoken in Hong Kong), Hong Kong: Language Information Sciences Research Centre, City University of Hong Kong. Li, Boya (2006) ‘Chinese Final Particles and the Syntax of the Periphery’, Doctoral Dissertation, Leiden University. Li, Charles N. and Sandra A. Thompson (1989) Mandarin Chinese: A Functional Reference Grammar, Berkeley and Los Angeles: University of California Press. Li, Jingzhong (1990) ‘Yueyu shi Hanyu zuqun zhong de duli yuyan’ (‘Cantonese as an Independent Language in the Family of Chinese Languages’), in Proceedings of 2nd Conference on Yue Dialects, Guangzhou: Jinan University Publishers. Li, Xinkui (1994) Guangdong de Fangyan (Dialects of Guangdong), Guangzhou: Guangdong Renmin Chubanshe. Li, Xinkui, Huang Jiajiao, Shi Qisheng, Mai Yun, and Chen Dingfang (1995) Guangzhou Fangyan Yanjiu (A Study of the Guangzhou Dialect), Guangzhou: Guangdong Renmin Chubanshe. Li, Yen-Hui Audrey (1997) ‘Structures and Interpretations of Nominal Expressions’, Ms., University of Southern California. Liu, Danqing (2001) ‘Hanyu geiyulei shuangjiwu jiegou de leixingxue kaocha’ (‘A Topological Study on the Ditransitive Construction of Chinese Give Verbs’), Zhongguo Yuwen《中国语文》(Studies of the Chinese Language) 5: 387–98. Lü, Shuxiang (ed) (1982) Yuwen Changtan (Elementary Knowledge of Language), Hong Kong: Joint Publishing. Luke, Kang-Kwong (1990) Utterance Particles in Cantonese Conversation, Amsterdam and Philadelphia: John Benjamins Publishing Company. Mair, Victor H. (1991) ‘What Is a Chinese ‘Dialect/Topolect’? Reflections on Some Key Sino-English Linguistic Terms’, in Sino-Platonic Papers 29: 1–31. Philadelphia: Department of East Asian Languages and Civilizations, University of Pennsylvania. Matthews, Stephen and Virginia Yip (1994/2011) Cantonese: A Comprehensive Grammar, London and New York: Routledge. Mok, Sui-Sang (1998) ‘Cantonese Exceed Comparatives’, Doctoral Dissertation, University of California, San Diego. Norman, Jerry and Mei Tsu-Lin (1976) ‘The Austroasiatics in Ancient South China: Some Lexical Evidence’, Monumenta Serica 32: 274–301. Ouyang, Jueya (1990) ‘Guangzhouhua de yuqi zhuci’ (‘The Modal Particles in Cantonese’), in Wang Li Xiansheng Jinian Lunwenji (Wang Li Memorial Volumes), Beijing: The Commercial Press, 464–76. Ouyang, Jueya (1991) ‘Yunyong diceng lilun yanjiu shaoshu minzu yuyan yu Hanyu de guanxi’ (‘Study of the Relationship between Minority Languages and Han Chinese with the Substratum Theory), Minzu Yuwen (Minority Languages of China) 6: 23–9.

32

Cantonese Paris, Marie-Claude (1981) Problèmes de Syntaxe et de Sémantique en Linguistique Chinoise, Paris: Collège de France. Rao, Bingcai (1988) ‘Yue fangyan’ (‘Yue Dialects’), in Zhongguo Dabaike Quanshu – Yuyan Wenzi (Encyclopedia of China – Language Volume), Beijing: Zhongguo Dabaike Quanshu Chubanshe. Simpson, Andrew (2010) ‘The Definite Bare Classifier Construction: A Cross-linguistic Study’, Paper presented at Workshop on Definiteness in Chinese, The Chinese University of Hong Kong. Sio, Ut-seong Joanna (2006) Modification and Reference in the Chinese Nominal, Utrecht: LOT. Smith, Carlota S. (1991/1997) The Parameter of Aspect, Dordrecht: Kluwer Academic Publishers. Tan, Chee-Beng (2005) ‘Chinese in Malaysia’, in Melvin Ember, Carol R. Ember, and Ian Skoggard (eds) Encyclopedia of Diasporas: Immigrant and Refugee Cultures Around the World, New York: Kluwer, 697–706. Tang, Sze-Wing (1998a) ‘On the “Inverted” Double Object Construction’, in Stephen Matthews (ed) Studies in Cantonese Linguistics, Hong Kong: The Linguistic Society of Hong Kong, 35–52. Tang, Sze-Wing (1998b) ‘Parametrization of Features in Syntax’, Doctoral Dissertation, University of California, Irvine. Tang, Sze-Wing (2003) Hanyu Fangyan Yufa de Canshu Lilun (A Parametric Theory of Chinese Dialectal Grammar), Beijing: Peking University Press. Tang, Sze-Wing (2006a) ‘Yueyu dakzai, matzai, gamzai shifou shuyu tongyige jiazu?’ (‘Do dakzai, matzai, and gamzai in Cantonese Belong to the Same Family?’), Studies in Chinese Linguistics 1: 1–11. Tang, Sze-Wing (2006b) ‘Yueyu yiwenju sin de jufa tedian’ (‘Syntactic Properties of Sin in Cantonese Interrogatives’), Zhongguo Yuwen《中国语文》(Studies of the Chinese Language) 3: 225–32. Tang, Sze-Wing (2007) ‘Yueyu kuangshi xuci de jubuxing he duochongxing’ (‘Locality and Multiplicity of Function Words in the Discontinuous Construction in Cantonese’), in Proceedings of the 10th International Conference on Yue Dialects, Beijing: Zhongguo Shehui Kexue Chubanshe. Tang, Sze-Wing (2009) ‘The Syntax of Two Approximatives in Cantonese: Discontinuous Constructions Formed with zai6’, Journal of Chinese Linguistics 37(2): 227–56. Tang, Sze-Wing (2011) ‘On Gerundive Nominalization in Mandarin and Cantonese’, in Foong-Ha Yap, Karen Grunow-Harsta, and Janick Wrona (eds) Nominalization in Asian Languages: Diachronic and Typological Perspectives, Amsterdam and Philadelphia: John Benjamins Publishing Company, 147–60. Tang, Sze-Wing and Cheng Siu-Pong (2014) ‘Aspects of Cantonese Grammar’, in C.-T. James Huang, Y.-H. Audrey Li, and Andrew Simpson (eds) Handbook of Chinese Linguistics, Hoboken, New Jersey: Wiley-Blackwell, 601–28. Ting, Jen (1995) ‘A Non-uniform Analysis of the Passive Construction in Mandarin Chinese’, Doctoral Dissertation, University of Rochester. Ting, Jen (1998) ‘Deriving the Bei-construction in Mandarin Chinese’, Journal of East Asian Linguistics 7(4): 319–54. Wu, Yicheng and Adams Bodomo (2009) ‘Classifiers ≠ Determiners’, Linguistic Inquiry 40(3): 487–503. Yiu, Yuk-Man Carine (2005) ‘Spatial Extension: Directional Verbs in Cantonese’, Doctoral Dissertation, The Hong Kong University of Science and Technology. Yuan, Jiahua, et al. (eds) (1960) Hanyu Fangyan Gaiyao (An Outline of the Chinese Dialects), Beijing: Wenzi Gaige Chubanshe. Yuan, Jiahua, et al. (eds) (2001) Hanyu Fangyan Gaiyao (An Outline of the Chinese Dialects), 2nd ed., Beijing: Yuwen Chubanshe. Yue-Hashimoto, Anne (1976) ‘South Chinese Dialects – the Tai Connection’, Computational Analyses of Asian and African Languages 6: 1–9. Yue-Hashimoto, Anne (1991) ‘The Yue Dialects’, in William Shi-Yuan Wang (ed) Languages and Dialects of China (Journal of Chinese Linguistics Monograph 3), Berkeley: Project on Linguistic Analysis, 237–93. Zee, Yun-Yang Eric (1999) ‘Chinese (Hong Kong Cantonese)’, in International Phonetic Association (ed) Handbook of the International Phonetic Association, Cambridge: Cambridge University Press, 58–60. Zhan, Bohui (ed) (1991) Hanyu Fangyan ji Fangyan Diaocha (Chinese Dialects and the Dialect Investigation), Wuhan: Hubei Jiaoyu Chubanshe.

33

Cheng Siu-Pong and Tang Sze-Wing Zhan, Bohui (ed) (2002) Guangdong Yue Fangyan Gaiyao (An Outline of Yue Dialects in Guangdong), Guangzhou: Jinan University Press. Zhang, Min (2011) ‘Hanyu fangyan shuangjiwu jiegou nan-bei chayi de chengyin: leixingxue yanjiu yinfa de xin wenti’ (‘Revisiting the Alignment Typology of Ditransitive Constructions in Chinese Dialects’), Bulletin of Chinese Linguistics 4(2): 87–270. Zhu, Dexi (1961) ‘Shuo de’ (‘On De’), Zhongguo Yuwen《中国语文》(Studies of the Chinese Language) 12: 1–15. Zhu, Dexi (1980) ‘Beijinghua, Guangzhouhua, Wenshuihua he Fuzhouhua li de de zi’ (‘De in the Beijing, Guangzhou, Wenshui, and Fuzhou Dialects’), Fangyan (Dialects) 3: 160–5.

34

3 Cantonese Romanizaton Cheng Siu-Pong and Tang Sze-Wing The Chinese University of Hong Kong, Hong Kong, China

The diversity of Cantonese Romanization is a reflection of different backgrounds and purposes in rendering Cantonese into Roman letters. Some forms of Romanization are marked for their historical significance and influence on later proposals; others are noteworthy for their current popularity, as they are often found in books and other materials. Individual frameworks are discussed in detail, and the concluding tables summarize these discussions. It follows that such diversity compensates for the lack of a predominant form of Cantonese Romanization.

1. Introduction Cantonese Romanization methods are as varied as they are numerous. Up to the end of the last century, over 30 varieties have existed (Wu 1997; Lam 2009). None of them, however, is in a predominant position as pinyin is to Mandarin Chinese. Lack of government-sponsored standardization programs is part of the reason. The general public in Cantonese-speaking regions also cannot reach a consensus on how a Cantonese word should be Romanized. Unlike Mandarin or Vietnamese, the use of Chinese characters to write Cantonese has not been subjected to challenge by the proponents of the Roman alphabet, and no serious attempt has ever been made to change the written form of Cantonese into an alphabetical system. The main function of Cantonese Romanization is to transcribe Cantonese for educational and research purposes. Another major function is to transliterate Cantonese terms into English or other European languages. Most Romanization systems aim at the former function, and different schemes have been introduced to cater for different target recipients. The following section first describes the sound system of Cantonese. After that, individual Romanization frameworks will be discussed. If looking for a shortcut, you can move to the Romanization tables in the concluding section.

2.  Cantonese phonology There are 19 consonants in modern-day standard Hong Kong Cantonese. These include plosives /p, ph, t, th, k, kh, kw, kwh/, nasals /m, n, ŋ/, affricates /ts, tsh/, fricatives /f, s, h/, and ‘semi-vowels’ as well as other approximants /j, w, l/. These sounds are considered phonemes 35

Cheng Siu-Pong and Tang Sze-Wing

because they change the meaning of a word when one consonant is replaced by another. Take the (un)aspirated bilabial stops as examples. Unlike English, aspirated /ph/ (like p in pie) and unaspirated /p/ (like p in spy) are two phonemes, and a minimal pair between them can be found, e.g. /pha55/ ‘crouch’ vs. /pa55/ ‘daddy’. Other plosives are the same and as a result there are two sets of plosives in Cantonese, one aspirated and another unaspirated. /ts, tsh/ may be realized as [tʃ, tʃ h], but they are either allophones (when the following vowel is rounded, especially when it is /y/) or simply free variants of /ts, tsh/, depending on the speaker’s accent. All of these consonants are syllable-initial. Nasals are also syllable-final, and syllable-final plosives /p, t, k/ are unreleased, expressed as [p2, t2, k2]. The 19 consonants reflect the standard pronunciation. Many native speakers do not articulate /kw, kwh, n, ŋ/ syllable-initially and /k, ŋ/ syllable-finally (except in rhymes /ek, ok, eŋ, oŋ/; see Bauer and Benedict 1997). This is of particular interest as far as Romanization is concerned, because many Cantonese dictionaries and textbooks consider the drop or replacement of these sounds to be non-standard or even ‘sloppy’ and retain the standard way of pronunciation.1 Vowels in Cantonese include both front and back, and high and low monothongs /i, e, y, u, o, ɛ, ɔ, œ, ɵ, ɐ, a/. Depending on which analysis is adopted, the use of the vowel length mark may be optional or conditional,2 and even the total number of vowels can be reduced as some of them occur in complementary phonetic contexts. Whether or not they are deemed phonemes depends on how these sounds are perceived. For example, /e/ may lose its phonemic status and be argued as an allophone of /i/, since they are not only similar but also complementary: /e/ before /k, ŋ/ and /i/ elsewhere. This is the same for the sound pair of /o, u/. In the standard, conservative accent, the sound pair of /œ, ɵ/ follows a similar distribution, although /œ/ can stand alone without a final consonant and monothong /ɵ/ is always followed by /t/ or /n/. This part being introductory, the academic debate associated is set aside. What concerns us is the fact that many Romanization schemes use a single letter for both sounds of the pair. In addition to these monothongs, there are also ten gliding vowels in Cantonese: /iw, ɵʮ, uʮ, ej, ɔj, ow, ɐj, ɐw, aj, aw/. In some analyses, the final glides are rendered as a vowel symbol. Notice that in the same syllable, a gliding vowel (or diphthong) is not followed by a consonant. Only a monothong precedes a consonant coda. For this reason, the final glide in a gliding vowel is often considered as a coda as well. The vowels above are all phonemic, with the apparent ‘exceptions’ of /ɵʮ, uʮ/. Minimal pairs can be found only when they are preceded by /k/, and also /kh/ if tones are disregarded. For the sound pairs of /ɐj, aj/ and /ɐw, aw/, they are clearly distinctive, and it is easy to find a minimal pair to differentiate the two members within. But given that they are similar in sound quality, these distinctive sounds, together with /ɵʮ, uʮ/, are sometimes represented by a single grapheme when they are rendered in Romanization. This will be further addressed in the following sections. In Cantonese, a syllable always contains a vowel, with the exception of two syllabic nasals /m, ŋ/. They can stand alone without a vowel to follow, in which case they are rendered as [m̩, ŋ̍]. Cantonese is a tonal language. There are six or nine tones in Cantonese, depending on one’s approach to categorize them. Traditionally, tones in Chinese are classified into Flat, Rising, Departing, and Entering. In Cantonese, the first three groups contain two tones (one Dark and one Light), and together with three Entering tones (Upper Dark, Lower Dark, Light), they amount to a total of nine tones. For the non-Entering tones, Cantonese has Dark Flat tone /55/, Light Flat tone /21/, Dark Rising tone /25/, Light Rising tone /23/, Dark Departing tone /33/, and Light Departing tone /22/.3 For the Dark Flat tone, there is also a variant which is high-falling /53/. The two Dark Flat tones are either not distinctive (Cheung 1986), or in fact, the falling variant might have been fading out altogether (Matthews and 36

Cantonese Romanizaton

Yip 1994). Therefore, the maximum number of tones in Cantonese is still said to be nine. For the Entering tones, they specifically refer to the tones of those syllables ended with an unreleased stop [p2, t2, k2], and the Upper Dark, Lower Dark, and Light Entering tones are given as /5/, /3/, and /2/, which respectively correspond to the Dark Flat, Dark Departing, and Light Departing tones. Given this, these Entering tones are often not viewed as separate tones, and as such, Cantonese is said to have six tones only. The Entering tones are only viewed as the short or extra-short variants of the other tones (Zee 1999). Different Romanization schemes variably adhere to the six-tone and nine-tone categorization, and this gives rise to a noticeable difference in how a tone is transcribed. It has also been observed that some native speakers merge some of these tones, especially the two Rising tones /25/ and /23/ (Bauer et al. 2003). This is again seldom reflected in dictionaries or textbooks with words transcribed in Cantonese Romanization. The term Cantonese itself, when used in the Hong Kong variety, is literally translated as ‘Yue language’ or ‘Guangdong speech’. Regardless of tone merger or any pronunciation deemed ‘non-standard’, the phonetic notation for the two words is given as /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ respectively.

3.  Various Romanization systems 3.1.  Earlier Romanizations The publication of Cantonese dictionaries in the early nineteenth century, intended for the Western readers, marked the beginning of a time when substantial work was done to Romanize Cantonese. Robert Morrison’s A Vocabulary of the Canton Dialect in 1828 was one of these earliest attempts to offer transcription for Cantonese words in Roman letters. Like many other dictionaries and glossaries that followed, every effort was made to ensure that the Romanization given resembled the native language of the target readers. For example, /i/ and /u/ are transcribed as ee and oo, written in a way in conformity with the English orthography. Letter u is used for /ɐ/, which only occurs in closed syllables. For instance, /sɐm/ ‘heart’ is rendered as sum. At this stage, there was no one-to-one mapping between letters and sounds. For the same letter u, when it is used in open syllables, it refers to /y/. For example, /tsy/ ‘pig’ is given as chu. But when /y/ occurs in a closed syllable, it is rendered as ue, as in uet /jyt/ ‘moon’. Besides, Morrison’s Romanization omits the indication of tones. Therefore, even though /thin55/ ‘sky’ and /thin21/ ‘field’ are in different tones, they are all Romanized in the same way as teen. Aspiration is not addressed either. For instance, unaspirated /t/ is indistinguishable from aspirated /th/, so teen may be used for /thin/ as well as /tin/, as in /tin22/ ‘lightning’. Readers cannot know whether it refers to sky or field or lightning simply through this Romanization. Several Cantonese dictionaries and textbooks emerged after Morrison’s work. These include E. C. Bridgman’s Chinese Chrestomathy in the Canton Dialect (1841), S. W. Williams’ A Tonic Dictionary of the Chinese Language in the Canton Dialect (1856), J. Chalmers’ An English and Cantonese Pocket-dictionary (1859), E. J. Eitel’s A Chinese Dictionary in the Cantonese Dialect (1877), and J. D. Ball’s Cantonese Made Easy (1883). Compared with Morrison’s Romanization, correspondence between letters and phonemes was improved in later schemes. In Bridgman’s dictionary, for example, distinction was made between aspirated and unaspirated stops and affricates by using inverted apostrophes. Therefore, /t/ and /th/ are represented by t and t‘, and the Romanized forms for /thin/ ‘sky’ and /tin/ ‘lightning’ came to be distinctive. Another feature is the use of semicircular marks to represent tones. This is not an invention by Western lexicographers, but adoption of the Chinese traditions. The Flat, 37

Cheng Siu-Pong and Tang Sze-Wing

Rising, Departing, and Entering tones are indicated by a semicircular mark placed at four different corners: lower left for Flat, upper left for Rising, upper right for Departing, and lower right for Entering. The Dark and Light tones are differentiated by marking the Light tones with a line underneath the semicircle. For instance, /thin55/ ‘sky’ is rendered as ˓t‘ín, whereas for /thin21/ ‘field’, the tone mark is further underlined. Ball’s Cantonese Made Easy followed the same methods to mark aspiration and tones, and by the same token also achieved a higher degree of phonemic distinction. B. F. Meyer and T. F. Wempe’s The Student’s Cantonese–English Dictionary (1934) also has a Romanization system similar to Eitel’s and Ball’s. Comparatively speaking, Y. R. Chao’s Cantonese Primer (1947) was a more recent work, yet his Romanization scheme is worth mentioning for historical purposes. Chao’s system is fully alphabetical in that tones are also represented by letters. Different sets of initial consonant letters are used for Dark and Light tones and, for each set, the representation of the same rhyme changes with different tones. To illustrate, the Dark /pa55/, /pa25/ and /pa33/ are represented as pa, pax, and pah, and their Light counterparts are written as ba, bax, and bah. Under Chao’s system, /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ are expressed as Yutyux and Kwoagtong-wah* respectively. Notice that Chao’s system is particular about whether the tone is the base tone or changed from another one. Since /wa/ here is originally in the Light Departing tone /22/, it is so expressed in its original tone as wah, rather than as uax, with an asterisk added to mark its change to a rising tone. Chao’s Romanization is historically significant as it is arguably the most complex scheme ever proposed for Cantonese Romanization. It has fallen into oblivion though.

3.2.  Government Romanization In Hong Kong, the Government Romanization is the most recognizable means to Romanize Cantonese. It is mainly used for local place names and personal names. To many Cantonese speakers in Hong Kong, this is the only Cantonese Romanization they can understand, and perhaps provide if they are asked to. As previously suggested, it would be dubious to call it a ‘system’ (Kataoka and Lee 2008); it is rather a hodgepodge of various ways of Romanization (Kataoka 2014). For obvious reasons, this is not a Romanization scheme – there is a low degree of correspondence between letters and sounds. The same letter is often used for different sounds. This is likened to the English orthography, in which letter c represents /s/ in cede, but /k/ in candy. In this form of Romanization, the four pairs of Cantonese plosives /p, ph/, /t, th/, /k, kh/, /kw, kwh/ are rendered the same way as p, t, k, and kw, regardless of whether they are aspirated. Given that these sounds are phonemes, distinctive words such as /tin55/ ‘insane’ and /thin55/ ‘sky’ would become indistinguishable when they are Romanized as tin. The same also applies to affricates /ts, tsh/, but there is one more complication: while the two consonants are Romanized the same way, they are variably rendered as ts or ch. Thus, although the first character for the place names of Chai Wan /tshaj21 wan55/ and Tsuen Wan /tshyn21 wan55/ share the same initial consonant, they are written in different initial letters.4 Nasals /m, n, ŋ/, fricatives /f, s, h/, and approximants /w, l/ are straightforwardly Romanized as m, n, ng, f, s, h, w, l. As in English, y is used for /j/. The Hong Kong Government Romanization is a defective script, and in fact it is defective in both directions – devoid of one-to-one correspondence from grapheme to phoneme, as well as from phoneme to grapheme. This resembles the English orthography more than, for example, the French orthography. The one-to-many mapping from letters to sounds is illustrated in the case of (un)aspirated consonants. And phoneme-to-grapheme inconsistency 38

Cantonese Romanizaton

is shown in the affricates and more often in the case of some vowels. Take /i/ as an example. It can be Romanized differently in the place names, as i for yi in Tsing Yi, as ee for see in Sai See Street, as z for tsz in Tsz Wan Shan, and as ze for sze in Sze Tei Shan. Personal names transliterated from Cantonese show the same inconsistency. /tsy55/, a common surname, is rendered as Chu, but /tsyn/ would likely become Chuen or Tsuen when transliterated. On the other hand, Chun (as in the name of Chief Executive Leung Chun-Ying) is reserved for /ts(h)ɐn/. Added to this is the fact that not all letter combinations are legitimate, even though they may represent a legitimate syllable. This is again likened to the English orthography, in which both letters s and c can refer to /s/, but only see, rather than cee, is accepted as a valid written unit. Therefore, /tsɐn/ is expressed as Chun, but Yun for /jɐn/ is unlikely. The adding up of these instances of inconsistency implies that many who know this Romanization have had to learn it by rote, no matter whether they are native Cantonese speakers. Only generalizations can be made here: (i) correspondence between monothongs and single vowel letters: /a, ɐ/ as a; /ɛ/ as e; /i, e/ as i; /ɔ/ as o; /y, u, o, ɵ/ as u; and (ii) correspondence between gliding vowels and double vowel letters: /iw/ as iu; /ɵʮ, uʮ/ as ui; /ej/ as ei; /ɔj/ as oi; /ow/ as ou; /ɐj, aj/ as ai; /ɐw, aw/ as au. These generalizations are far from comprehensive since a monothong can also be (and is often) expressed in more than one letter, and that /ow/ is more commonly expressed in a single letter, i.e. o, not to mention the use of consonant letters (plus y) and other irregularities in vowel representation (see Cheng 2014). The Government Romanization is not meant to be a teaching tool for Cantonese learners. Its purpose is to translate Cantonese names in the form of transliteration. Therefore, it almost never goes with the tonal indication. Unlike other ways of Romanization, tone is out of concern under most circumstances. Besides, not all place and personal names in Hong Kong are Romanized from Cantonese. Some place names in English are literally translated, e.g. Green Island for /tsheŋ55 tsɐw55/, or they are named with no regard to how they are called in Cantonese, e.g. Aberdeen for /hœŋ55 kɔŋ25 tsɐj25/. Furthermore, not all place names are transliterated from standard Hong Kong Cantonese. The most notable example is the name Hong Kong itself. Hong is based on the pronunciation of the boat people, or Tankas, during the time when the British arrived at the territory in the nineteenth century. If transliterated from the standard variety, it will most likely be rendered as Heung or Heong. Personal names are not always transliterated from Cantonese either. This is especially true for those not of Cantonese descent. Surnames vary even greater, since a native Cantonese speaker might have inherited a non-Cantonese family name. Language change is another complicating factor. A place name was Romanized and subsequent sound change made it deviant from the current pronunciation. All these mean that it is easy to draw erroneous generalizations when one attempts to learn this Romanization through a list of local place names or personal names. Though unruly, the Government Romanization is very often the only Romanization known to the general populace, not only used in the administration to transliterate proper names, but also used by the general public for various purposes. For instance, the Cantonese ‘chat alphabet’ used in informal communication such as instant messaging is loosely based on the government’s way of Romanizing Cantonese. When /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ ‘Cantonese’ are Romanized, if needed, they are given as Yuet Yu(e) and Kwong Tung Wa(h).5

3.3.  Yale Romanization The origin of Yale Romanization for Cantonese is commonly attributed to Parker Po-Fei Huang and Gerald Kok for their collaborated textbook Speak Cantonese. First published in 39

Cheng Siu-Pong and Tang Sze-Wing

1958, it has been reprinted several times over a period of four decades. In its sequel Book II, Huang states that the Romanization system was developed by Kok. Based on the year of its first publication, it is reasonable to assert that the scheme came into being no later than the late fifties. Huang’s later work Cantonese Dictionary: Cantonese–English, English– Cantonese also adopted this system. Cantonese is one of the languages Romanized under a system created at Yale University. Other Yale Romanizations involve East Asian languages such as Mandarin, Korean, and Japanese. Since its introduction, the Yale system has been extensively used in English-language textbooks, dictionaries, and glossaries made for Cantonese learners. These include some of the most popular titles currently in circulation. Comparatively speaking, the use of the Yale system is less prevalent in Chinese-language publications. While the native speakers sometimes do need to know how to pronounce a particular Chinese character, they are often referred to another homophonous character. Even when Romanization is consulted (not very often), the Yale system is not the one needed. So far, it is chiefly used to study Cantonese as a foreign language, rather than offer a pronunciation guide for the native speakers. Thus, Cantonese learners using this system might encounter a situation where they know this Romanization much better than the native speakers.6 Under the Yale system, the 19 consonants are expressed in a way that makes a distinction between aspirated and unaspirated consonants. The plosives /p, ph, t, th, k, kh, kw, kwh/ are respectively rendered as b, p, d, t, g, k, gw, and kw. This contrasts with the Government Romanization. For example, in the Government version, p represents both /p/ and /ph/, but Yale Romanization uses two separate letters b and p. In this case, the Yale system has oneto-one mapping between letters and phonemes. To achieve this goal, the Yale system uses letters traditionally reserved for voiced consonants to represent the voiceless ones. Affricates /ts, tsh/ are given as j and ch. Similarly this distinction is not observed in the Government Romanization, and they are both expressed as ch or ts. Nasals /m, n, ŋ/, fricatives /f, s, h/, and /j, w, l/ are straightforwardly rendered as m, n, ng, f, s, h, y, w, l. With /j/ Romanized as y, the Yale system resembles the English orthography. Phoneme-to-grapheme consistency is also achieved in the vowel representation, although a single letter may in separate situations be used to represent distinct sounds. Letter a, for example, is used for both /a/ and /ɐ/. Yet /a/ is expressed in single letter a only when it stands alone in the rhyme. Gliding vowels with /a/ are given in double letters aa, and aa is also used when /a/ precedes a consonant coda. Since /ɐ/ (always given as a) never occurs syllable-finally, a distinction can still be made between the two vowels. Letter i is used for /i, e/, which is common to most Romanization schemes as they are in complementary distribution. The same applies to u, which represents both monothongs /u/ and /o/. It should be noted, however, that u is also used in yu as a grapheme representing /y/. The y in yu does not represent the semi-vowel (approximant), except in the case where the initial consonant is /j/. To illustrate, /sy/ is expressed as syu, whereas /jy/ is given as yu, instead of the incorrect yyu. The ‘shorthand’ rule takes advantage of the fact that /ju/ does not exist as an indigenous syllable in Cantonese, but problems may arise when loanwords are involved. /ju/, for example, can mean ‘university’ in daily speech. Apparently, the Yale system does not assign a separate Romanized form for this syllable. Letter u is also used for other vowels when it is associated with e. Letter e as a single letter represents /ɛ/, but when combined with u to form eu, it refers to both /œ/ and /ɵ/: /lœk/ is given as leuk and /lɵt/ is given as leut. Letter o is used for monothong /ɔ/. The same principles are generally extended to the gliding vowels, so /iw, ej, ɔj, ɐj, ɐw, aj, aw/ are expressed as iu, ei, oi, ai, au, aai, aau, with u representing the rounded glide and i for the unrounded one. However, /ɵʮ, uʮ/ 40

Cantonese Romanizaton

are given as eui and ui even though /ʮ/ is considered rounded. /ow/ is given as ou, although monothong /o/ is Romanized as letter u. Thus, /sow/ is expressed as sou, but /sok/ is expressed as suk. The tonal representation is what gives the Yale system a distinctive hue, as it makes use of both diacritics and letter h. It adheres to the six-tone categorization. In its original version, a distinction was made between the high and high-falling tones of the Dark Flat tones, so that /a55, a53/ are expressed as ā and à, with the vowel differently marked by macron and grave accents. The Dark Rising and Departing vowels are respectively marked by acute and no accents. Thus, /a25, a33/ are given as á and a. With considerations of the sound change factor (discussed above), subsequent variants may use a macron accent for all Dark Flat tones, whereas a few publications nowadays still follow the original design and use a grave accent. The Light counterparts are marked with the same diacritic mark (with Light Flat tone corresponding to the high-falling Dark Flat tone), with an h additionally attached to the vowel. Therefore, /a21, a23, a22/ are given as àh, áh and ah. As it follows the six-tone classification, a separate representation for Entering tones is unnecessary, and the three Entering tones are treated in the same way as the non-Entering ones: /at5, at3, at2/ are rendered as àat, aat and aaht. The tone letter h follows the vowel but not the whole rhyme; while /sɵʮ21/ is expressed as sèuih, /sɵn21/ is expressed as sèuhn. Taken all together, when /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ ‘Cantonese’ are Romanized in the Yale system, they become yuht yúh and gwóng dūng wá. Regarding the tone letter h, an additional rule was stipulated in the original system, which states that when a syllable begins with m, n or ng, the h will not be used as an indicator. This is due to the fact that these syllables are seldom expressed in the Dark tones. /ŋɔ23/ ‘me’, for example, should be rendered as ngó rather than as ngóh. Most later applications disregard this rule, with the possible exceptions of the two syllabic nasals. The peculiarity as a result of using tone letter h causes some later versions to discard it altogether,7 and replace the diacritics with tone number: /55/ (or /53/, /5/), /25/, /33/ (or /3/), /21/, /23/, and /22/ (or /2/) are numbered from 1 to 6. In this ‘simplified’ version, yuht yúh and gwóng dūng wá are given as yut6 yu5 and gwong2 dung1 wa2.

3.4.  LSHK Romanization The LSHK Romanization Scheme, also known as Jyutping, was proposed by the Linguistic Society of Hong Kong (LSHK) in 1993. The LSHK is a non-governmental organization comprising linguistic academics and professionals based in Hong Kong. This Romanization scheme was collectively designed by scholars specializing in various subfields of language studies. A conference had been held earlier in 1992 to gather thoughts on the future scheme before the finalized version was released at the Fourth International Conference on the Yue Dialects in December 1993. The scheme aims to be multipurpose, systematic, inclusive, and user-friendly and takes care of both the earlier traditions and word-processing technology of the time (Cheung 1994). Since its introduction, the LSHK Romanization has gained a strong foothold in the academia. Up to this day, a majority of English and Chinese-language academic papers published in Hong Kong and also in the West adopt this system for Romanization, apart from the direct use of phonetic symbols. The Government of Hong Kong does not have a fixed position on which scheme should be the default option. Having said that, the post-handover Education Bureau, in collaboration with the tertiary scholars, has published materials that use this system as the pronunciation guide. They abandoned the earlier scheme (‘ILE Romanization’; to be discussed) used for another guidebook published by the Education Department in the early nineties. As the scheme was invented by tertiary 41

Cheng Siu-Pong and Tang Sze-Wing

scholars, it is taught in local university courses that touch upon Cantonese pronunciation. Besides, its promotion has occasionally been extended to primary and secondary schools. But still most Hong Kong people, including university students, have never been required to learn Cantonese sound notation at school. Also due to its fairly recent date of emergence, the LSHK Romanization is still on its way to reaching a wider spectrum of people outside the academic community. The representation of consonants has given this system some noticeably distinctive features. Same as the Yale Romanization, plosives /p, ph, t, th, k, kh, kw, kwh/, nasals /m, n, ŋ/, fricatives /f, s, h/, and approximants /w, l/ are conveniently given as b, p, d, t, g, k, gw, kw, m, n, ng, f, s, h, w, l. What makes the LSHK system spectacular is how it renders affricates /ts, tsh/, which are given as z and c. It conforms to Mandarin pinyin, but deviates from the Government traditions in which single graphemes z and c are never used to represent consonants. In the Government Romanization, c is only used in ch for /ts, tsh/, and z is never used for consonants and only used in few syllables like tsz /ts(h)i/ or sze /si/. Another deviation from the commonplace Romanization is on how it expresses /j/. Letter j is used for /j/ in the LSHK system, but in Hong Kong’s common practice, it is instead used for /ts/, as is often found in the name of local companies. Note that these peculiarities were made ‘on purpose’, so that it takes care of both the pinyin (for z and c) and IPA traditions (for j), and in so doing ‘enhances user-friendliness’ (Cheung 1994). Regarding the Romanization of vowels, the LSHK system is arguably the most precise one in representing the distinctive sounds. One-to-one mapping from phoneme to grapheme is achieved in most vowels. And each rhyme corresponds to exactly one written unit. Monothongs /i, e, y, u, o, ɛ, ɔ, œ, ɵ, ɐ, a/ are respectively rendered as i, i, yu, u, u, e, o, oe, eo, a, and aa. /i, e/ and /u, o/ are expressed in one letter only (i and u) given that these sounds are in complementary distribution. Except them, each monothong matches one grapheme. To achieve phonemic distinction, the LSHK scheme resorts to the use of four digraphs yu, oe, eo, and aa to represent monothongs, in which yu is somewhat conspicuous as letter y is more often understood in English as a consonant letter. Unlike the Yale system, however, no ‘shorthand’ rule is added ( yyu becomes yu for /jy/); /jy/ is directly expressed as jyu. Digraphs are specially designed so that only the basic Roman alphabet (26 letters) is used. This is of particular significance during the time when keying in a diacritic was a hassle. As for /ɐ, a/, the distinction made by using aa and a is an inheritance of the earlier traditions. Another important feature is that it distinguishes /œ, ɵ/. They have similar sound quality and have traditionally been expressed in only one grapheme (in the Yale system, for example, they are Romanized as eu). In so doing, it manages to render some ‘non-indigenous’ rhymes and syllables arising as a result of the recent sound change. In standard, conservative accent, /œ, ɵ/ are complementary, but recent merger of codas /t, k/ and /n, ŋ/ into /t/ and /n/ has created new rhymes such as /œn, œt/. Thus, in this new variety, /œ, ɵ/ are contrastive and can be differentiated with minimal pairs. Assigning two graphemes oe and eo separately for /œ/ and /ɵ/ allows the system to become ‘inclusive’ by capturing this recent contrastive distinction. The same conversion rules are extended to gliding vowels: /iw, ɵʮ, uʮ, ɔj, ɐj, ɐw, aj, aw/ are given as iu, eoi, ui, oi, ai, au, aai, aau, with letter u representing /w/ and i representing /j/ or /ʮ/. Since the latter two glides are not contrastive, they are expressed in one single letter – a common practice in Cantonese Romanization. /ej, ow/ are treated as if the vowel were less high: monothongs /e, o/ are given as i and u, but as an element in the gliding vowels, they are rendered as e and o, and /ej, ow/ are given as ei and ou. The paradigm of the LSHK system allows the inclusion of new rhymes or syllables traditionally non-existent but recently created due to foreign-language influences. These include letter 42

Cantonese Romanizaton

pronunciation such as /ɛw55/ ‘L’ (eu1), /ɛm55/ ‘M’ (em1) and /ɛn55/ ‘N’ (en1), as well as transliterated loanwords, such as /ɛp5/ ‘app’ (ep1). Many other systems, on the other hand, do not specify how, or if at all, these syllables should be Romanized. For example, the use of eu for /œ, ɵ/ in the Yale system implies that /ɛw/ does not exist; otherwise, the two vowels will share the same grapheme eu, which is implausible. Under the LSHK scheme, all these rhymes are regarded as existent so that it achieves ‘phonological inclusiveness’ (Cheung 1994). The scheme is systematic and highly inclusive, but it also risks producing some Romanized forms unintelligible to people who are only familiar with the Government Romanization. Examples of this include jeon (/jɵn/) and zyu (/tsy/). They are commonly Romanized as yun and chu (tsu), quite dissimilar from what they are given in the LSHK format. The LSHK system adheres to the six-tone categorization, meaning that there is no separate notation required for the Entering tones. Tone number is used to mark the tones: number 1 for /55/ (or /5/), 2 for /25/, 3 for /33/ (or /3/), 4 for /21/, 5 for /23/, and 6 for /22/ (or /2/). The LSHK working group decided that tone number 1 does not necessarily refer to /55/ (Cheung 1994). If tone /53/ exists in anyone’s speech, it is still marked by the same number. The use of numbering is desirable because numerals are in the basic ASCII character set. Since in computing Chinese characters are coded with at least one ‘non-basic’ ASCII character, like those with diacritic marks, inputting diacritics for tones might mistakenly produce Chinese characters in the earlier versions of Chinese word-processing software. The ordering also reflects a more common practice at the time. An alternative ordering puts the Flat-Rising-Departing-Entering sequence in the first order and the Dark–Light distinction in the second. To write the tone number, some earlier schemes stipulate that it be written in superscript style. This is not required in the LSHK scheme. Taken all together, when /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ ‘Cantonese’ are Romanized in the LSHK system, they become jyut6 jyu5 and gwong2 dung1 waa2. Often, tonal indication is omitted in translated terms, such as the term Jyutping itself.8 If tone number is omitted, jyut6 jyu5 and gwong2 dung1 waa2 will be expressed as Jyutjyu and Gwongdungwaa. It is assumed that the same capitalization rules of the target language apply to those terms transliterated in LSHK Romanization.

3.5.  Rao’s Romanization It refers to the Romanization scheme used in the dictionaries of which Rao Bingcai is one of the editors. The official name of the scheme is called the Scheme of the Cantonese Phonetic Alphabet. The system attributed to Rao is a modified version of the said scheme, which was officially announced in 1960 by the executive branch for education of Guangdong province. The original scheme was part of a larger project encompassing four different Romanization schemes for Guangzhou dialect (Cantonese), Kejia dialect (Hakka), Hainan dialect, and Chaozhou dialect (Swatow). In theory, this is a government standard for Cantonese Romanization. However, the official scheme has yet to gain widespread adoption. It should be best understood as a government invention without much standardization effort to follow it up. Rao’s modified version, on the other hand, has become predominant in mainland China, with its use extended from his own publications to a variety of Cantonese textbooks and dictionaries written by others. It has largely replaced the original scheme, which has since then been only occasionally referred to in the footnotes of the pronunciation guide. The changes made in the modified scheme include the replacement of some vowel graphemes with another one and also the addition of two syllabic nasals.9 They will be highlighted below. 43

Cheng Siu-Pong and Tang Sze-Wing

Under this scheme, a phoneme may be represented by more than one grapheme, depending on the phonetic and orthographic conditions. When used syllable-initially as an onset, plosives /p, ph, t, th, k, kh/ are given as b, p, d, t, g, k. This agrees with their representation in most other systems. However, when /p, t, k/ are used syllable-finally as a coda, expressed as unreleased [p2, t2, k2], they are given as b, d, g. This differs from all major forms of Romanization, including Government, Yale, and LSHK, in which they are rendered as p, t, k. Thus, /phit/ is expressed as pid in this scheme, but is expressed as pit in others. As for the other two plosives /kw, kwh/, they are given as gu and ku. Letter u, instead of w as in some other Romanizations, is used to represent the labialization of velar stops. The original scheme clarifies this by stating that gu and ku are syllables when they stand alone, i.e. in the case of /ku, khu/, whereas they are initials when they are not used independently in the phonetic notation; as in the case of /kwa, kwha/, they are given as gua and kua. Nasals /m, n, ŋ/, fricatives /f, h/, and approximants /w, l/ are as usual Romanized as m, n, ng, f, h, w, l, and /j/ is given as y. The most complex way of conversion occurs to /ts, tsh, s/. They are variably expressed as z, c, s or j, q, x depending on the following vowel letter: as j, q, x when they precede vowel letters i or ü (shorthanded as u; to be discussed), and as z, c, s elsewhere. To illustrate, /si/ is given as xi, while /sa/ is expressed as sa. Some previous literature suggests that it indicates a distinction between /ts, tsh, s/ (as z, c, s) and /tʃ, tʃ h, ʃ/ (as j, q, x). This is more apparent than real. First, whether the two sets of affricates are in complementary distribution is debatable. Even if they are so, the post-alveolar /tʃ, tʃ h, ʃ/ usually precede round vowels, but j, q, x apply to both rounded (/y/) and unrounded (/i, e/) monothongs, so do z, c, s. Besides, monothong /e/ and gliding vowel /ej/ are differently expressed as i and éi, and different sets of letters are used for them. More importantly, the scheme explicates that there is no distinction between these two sets of consonants letters. This is totally out of orthographic concerns and is irrelevant to any ‘natural classes’ that show distinctive phonetic features. The orthographic considerations are in turn taken to resemble the Mandarin pinyin system. Such an arrangement is more pedagogically relevant than phonetically necessary. As far as the consonants are concerned, Rao’s modification is equivalent to the original scheme, except it additionally specifies that the two nasals can also be syllabic. In Rao’s Romanization, monothongs /i, e, y, u, o, ɛ, ɔ, œ, ɵ, ɐ, a/ are respectively rendered as i, i, ü, u, u, é, o, ê, ê, e, a. Complementary /i, e/ and /u, o/ are expressed in one letter, i.e. i and u. /œ, ɵ/ are both expressed as ê, a plausible arrangement for the standard accent (see the description above). Note that diacritics, instead of digraphs, are employed to render these sounds. Therefore, a monothong is always expressed by a single letter. Besides, the use of ü for /y/ allows it to conform to the pinyin system. Meanwhile, the pinyin orthographic rules are transferred to the Cantonese system: ü is written as yu for yu /jy/, yun /jyn/, and yud /jyt/; the two dots are also removed after j, q, x, e.g. ju /tsy/, qu /tshy/, xu /sy/. These are essentially the pinyin rules despite that in Mandarin /j/ is deemed non-existent, and letter y is added but represents no consonant. The ‘shorthand’ rule is possible only because /u/ does not follow /ts, tsh, s, j/ in the indigenous Cantonese phonological system. Borrowed syllable /sut/ (for ‘suit’) might be expressed as sud, making use of two sets of homophonous letters, z, c, s vs. j, q, x, but /ju/ (for letter ‘U’ or ‘university’) cannot be represented without conflicting with the indigenous syllable /jy/. Conformity with the pinyin system is also exhibited in the case of /ɐ/. In other systems, the /ɐ, a/ distinction is often made by using single letter a and double letters aa. Here, e is used for /ɐ/ and a is reserved for /a/. The more basic e is reserved for /ɐ/, whereas é and ê are for /ɛ/ and /œ, ɵ/. This is possibly because rhymes like en /ɐn/ and eng /ɐŋ/ resemble Mandarin en /ən/ and eng /əŋ/, and syllables such as zé, cé, sé /tsɛ, tshɛ, sɛ/ do sound quite differently 44

Cantonese Romanizaton

from their near-homographs in Mandarin, i.e. ze, ce, se /tsɤ, tshɤ, sɤ/. Gliding vowels basically follow the same conversion pattern: the vowel element in /iw, ɵʮ, uʮ, ɔj, ɐj, ɐw, aj, aw/ follows the same rules and so they are converted into iu, êu, ui, oi, ei, eo, ai, ao respectively. Apart from this, /ej, ow/ are given as éi and ou, treated as if they were the gliding variants of /ɛ, ɔ/. In this regard, the Yale and LSHK systems also share the same treatment. Note that the glides in these vowels are variably represented. /w/ is given as u for iu /iw/ and ou /ow/, but as o for eo /ɐw/ and ao /aw/; /ʮ/ in êu /ɵʮ/ is expressed as u, but is expressed as i in ui /uʮ/. The description above is for Rao’s modified scheme. In the original scheme, the four rhymes /ɐw, ow, ɔŋ, oŋ/ are given as ou, ô, ông, ong, rather than as eo, ou, ong, ung. Phonetically speaking, these changes improve phoneme-to-grapheme consistency. Besides the obvious problem of using e for /ɐ/ but ou for /ɐw/, the use of ô also undermines the coherence of the scheme. Letter ô is the only non-digraph for gliding vowels. Even if the sound it represents is said to be monothongized as /o/, ông should better represent /oŋ/ than /ɔŋ/, not the other way around as it does in the original scheme. It appears that the original assignment of ong for /oŋ/ is deliberate because this rhyme is closer to Mandarin ong /ʊŋ/ than /ɔŋ/ is. Both the original and Rao’s modified schemes adhere to the six-tone categorization. As such, Dark Flat (Upper Dark Entering), Dark Rising, Dark Departing (Lower Dark Entering), Light Flat, Light Rising, and Light Departing (Light Entering) tones are numbered from 1 to 6 respectively. Cheung (2003) notes that the original 1960 scheme has established the now-common tonal notation traditions, in which Entering tones are no longer independent and the tonal order is first determined by the Dark–Light distinction rather than by the traditional tonal categories. Later systems like the LSHK scheme follow this ordering. Again, Tone 1 refers to either /55/ or /53/, a distinction prevalent in the older-day Guangzhou Cantonese for which the scheme was invented. Taken all together, /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ ‘Cantonese’ are Romanized in Rao’s system as yud 6 yu5 and guong2 dung1 wa2. Unlike the LSHK scheme, a superscript style for tone number is required.

3.6.  Other systems Other influential systems include those proposed by S. L. Wong, Sidney Lau, and the Institute of Language in Education. In some areas, S. L. Wong’s scheme is perhaps more influential than the Yale and LSHK ones. It originated from his book A Chinese Syllabary Pronounced according to the Dialect of Canton (1941) and forms the basis of the phonetic symbols used in a majority of the Standard Chinese dictionaries published in Hong Kong. Yet Wong’s system and its variants may not fit the strict definition of ‘Romanization’ because phonetic symbols, rather than ordinary Roman letters, are used for transcription. Most dictionaries simply term it ‘International Phonetic Alphabet’. It is often seen as a ‘broad’ phonetic transcription for Cantonese, and as a matter of fact, some conventionalized symbols should be understood in order to interpret them correctly. Most notably, it uses voiced consonant symbols for unaspirated plosives and affricates. Thus, unaspirated /p, t, k, kw, ts/ are rendered as /b, d, g, gw, dz/. Beginning learners with prior knowledge of IPA might risk viewing them as voiced consonants.10 Other features that make it a ‘broad’ transcription include the use of /œ/ for both /œ/ and /ɵ/, and /i, u, y/ in gliding vowels, so that /iw, ɵʮ, uʮ, ej, ɔj, ow, ɐj, ɐw, aj, aw/ are expressed as /iu, œy, ui, ei, ɔi, ou, ɐi, ɐu, ai, au/. Note that the vowel length mark is not used and /ʮ/ is variably expressed as /y/ or /i/, as in /œy, ui/. This system is sufficient for native speakers to look up the pronunciation of a Chinese character, but nonnative learners using it might enunciate wrongly if not knowing the associated phonetic details. 45

Cheng Siu-Pong and Tang Sze-Wing

The Institute of Language in Education (ILE) Scheme, on the other hand, is an exemplar of Cantonese Romanization. It was proposed by the said institute, which had then been a subsidiary of the colonial Education Department and was later incorporated into the Hong Kong Institute of Education (HKIEd). This scheme is a modification of the one proposed by P. C. Yu in his book Rhyming Glossary (1971). Same as the LSHK scheme, only 26 letters and Arabic numerals are used for Romanization. This system once acquired some degree of authority through the publication of List of Cantonese Pronunciation of Commonly-used Chinese Characters, printed by the Government Printer in 1990. Later publications by the administration, however, have changed to the LSHK system. A slightly modified version of the ILE scheme is used in The Dictionary of Standard Cantonese Pronunciation edited by Zhan Bohui (2002). In vowel representation, the ILE scheme renders /a/ as a in the open syllables and as aa in the closed ones (same as the Yale Romanization). Yu’s original scheme has the same treatment, except that the letter a for /ɐ/ is always italicized, whereas the one for /a/, as well as all other letters, is not. Zhan’s modification uses aa for /a/ in all phonetic contexts. Finally, Sidney Lau’s Romanization is associated with a series of Cantonese textbooks written by him and published by the Government Printer starting from 1965. Lau has spent years in the Hong Kong Government teaching Cantonese and was once the Principal of the Government Language School. One of the main characteristics of this system is that it bears some resemblance to the English orthography. In addition to the use of oo for /u/, it also resorts to the use of final h to close the syllable with /ɔ, œ/ when they are not followed by a final consonant or glide, i.e. goh for /kɔ/ but gong for /kɔŋ/. Lau stated that his scheme was an adaptation of the Meyer-Wempe system, which already included these features. The former two systems, Wong and ILE, adhere to the nine-tone categorization. The Upper Dark, Lower Dark, and Light Entering tones are numbered from 7 to 9, rather than being merged into the non-Entering ones for tonal notation. Sidney Lau’s Romanization, on the other hand, adheres to the six-tone categorization.

4. Conclusion In the practice of Cantonese Romanization, it is not uncommon for writers and academics to develop a new scheme or improve on existing ones – a reason for a substantial and growing number of systems. The Romanization forms discussed above are those that have historical significance and/or stimulated adoption by others who are not the inventors or definite proponents, or their associates. Tables 3.1–3.4 summarize the most popular schemes over the past half century.11 Together with the Government Romanization, these Romanization forms are employed in various occasions and reach different groups of target audience. For instance, to trans­ literate a Cantonese term for a Hong Kong-based English bulletin, without much regard to phonemic precision, the Government Romanization allows the largest number of locals to understand what the transliterated term means. English-speaking Cantonese learners are often exposed to the Yale Romanization and might prefer a transcription in this system. Meanwhile, Mandarin-speaking learners of Cantonese who have prior knowledge of pinyin might find Rao’s scheme most accommodating. In Hong Kong’s academia, the LSHK Romanization has been becoming the de facto standard and is expected to be used by the faculty and students alike in academic papers. With such a diversity of uses and users of many backgrounds, the apparent chaos of Cantonese Romanization may well be thought of as a ‘blessing in disguise’. 46

Cantonese Romanizaton Table 3.1  Initial consonants IPA

Wong

Yale

Rao

ILE

LSHK

Lau

p ph m f t th n l ts tsh s j k kh ŋ h kw kwh w

b p m f d t n l dz ts s j g k ŋ h gw kw w

b p m f d t n l j ch s y g k ng h gw kw w

b p m f d t n l z, j c, q s, x y g k ng h gu ku w

b p m f d t n l dz ts s j g k ng h gw kw w

b p m f d t n l z c s j g k ng h gw kw w

b p m f d t n l j ch s y g k ng h gw kw w

Note Wong: Variant of S.L. Wong’s Scheme Used in Hong Kong’s Dictionaries. Yale: The General Scheme of Yale Romanization. Rao: Rao Bingcai’s Modification of the Scheme of the Cantonese Phonetic Alphabet. ILE: Institute of Language in Education Romanization. LSHK: Linguistic Society of Hong Kong Romanization. Lau: The General Scheme of Sidney Lau’s Romanization.

Table 3.2  Final consonants IPA

Wong

Yale

Rao

ILE

LSHK

Lau

m n ŋ p t k

m n ŋ p t k

m n ng p t k

m n ng b d g

m n ng p t k

m n ng p t k

m n ng p t k

47

Cheng Siu-Pong and Tang Sze-Wing Table 3.3  Vowels IPA

Wong

Yale

Rao

ILE

LSHK

Lau

i e y u o ɛ ɔ œ ɵ ɐ a iw ɵʮ uʮ ej ɔj ow ɐj ɐw aj aw

i i y u u ɛ ɔ œ œ ɐ a iu œy ui ei ɔi ou ɐi ɐu ai au

i i yu u u e o eu eu a a(a) iu eui ui ei oi ou ai au aai aau

i i ü u u é o ê ê e a iu êu ui éi oi ou ei eo ai ao

i i y u u e o oe oe a a(a) iu oey ui ei oi ou ai au aai aau

i i yu u u e o oe eo a aa iu eoi ui ei oi ou ai au aai aau

i i ue oo u e o(h) eu(h) u a a(a) iu ui ooi ei oi o ai au aai aau

Table 3.4 Tones IPA

Wong

Yale

Rao

ILE

LSHK

Lau

55 / 53 (5) 25 33 (3) 21 23 22 (2)

a1 (a7) a2 a3 (a8) a4 a5 a6 (a9)

ā/à á a àh áh ah

a1 a2 a3 a4 a5 a6

a1 (a7) a2 a3 (a8) a4 a5 a6 (a9)

a1 a2 a3 a4 a5 a6

a1º / a1 (a1º) a2 a3 a4 a5 a6

Notes   1 Since consonants are not obligatory in the syllable-initial position, a zero consonant is often suggested as a phonological unit (as onset).   2 Zee (1999) indicates that /i, y, u, ɛ, ɔ, œ, a/ are long in open syllables and a third shorter in syllables closed by a plosive or nasal. The vowel length mark is not used here for the latter syllables.   3 Tone numerals are used. Dark Rising, Light Flat and Light Rising tones might also be given as /35/, /11/ and /13/. For simplicity, the description above follows Bauer and Benedict (1997).   4 According to Kataoka and Lee (2008), the use of ch and ts makes historical sense since Cantonese used to show distinction between /ts(h)/ and /tʃ (h)/. The merger into /ts(h)/ has now become the norm.   5 Neighbouring Macau, a former Portuguese colony, also has its own way to Romanize Cantonese. Precisely, this is not Romanization to English, but to Portuguese, although its Romanization is transferred directly into English contexts. Certain features differentiate it from the Hong Kong

48

Cantonese Romanizaton version, possibly designed to adapt for the Portuguese orthography. Major differences include its using v for /w/ (though w is also used) and i (or no letter at all) for /j/. /jyt2 jy23/ and /kwɔŋ25 toŋ55 wa25/ may possibly be expressed as Ut U (or Iut Iu) and Kuong Tong Va in the Macau variety (see Cheng 2014 for a comparative study of Government Romanizations in Hong Kong and Macau).   6 For some Cantonese speakers, the Yale Romanization might be noticed through its association with the Google Cantonese Input Method; although it accepts a wide range of Romanization variants, the ‘suggestion’ given in the input box is based on the Yale system (in a ‘simplified’ version; to be discussed below).   7 The use of letter h to mark the Light tones also creates a peculiar sorting order – in alphabetizing the dictionaries entries, for example, characters with the same phonetic segment but different tones are not put in consecutive order. /sɵn25/ shares the same rhyme with /sɵn23/ but not with /sɵʮ23/. But when the order is determined by the alphabetic sequence of their Yale Romanization, the three syllables will be in the sequence of ‘/sɵn23/-/sɵʮ23/-/sɵn25/’ as they are Romanized as séuhn, séuih, and séun. This differs from the normal practice in which characters having the same segment but different tones are sequenced consecutively.   8 This is reminiscent of how Mandarin pinyin is employed in foreign-language publications and scholarly works. In the pinyin system, an apostrophe is used to reduce ambiguity by marking the end of one syllable and the beginning of another, when the second one starts with a vowel letter. Scholars using the LSHK scheme often follow the same practice and put an apostrophe before syllables that begin with a vowel as well as syllabic nasals, i.e., co’ng for co3 ng6 ‘error’ and cong for cong1 ‘tumor’.   9 The changes are said to have been made ‘in response to public opinion’. It is not sure whether this may violate the specification of the original scheme which states that ‘in principle no more new letters will be invented and the letter pronunciation will not be casually changed’ (Guangdongsheng sizhong fangyan pinyin fang’an 1960). 10 Wong did mention that /b, d, g, dz/ are in fact devoiced [b̥, d̥, ɡ̊, d̥z̥ ] (/gw/ is treated as /g/ plus /w/). These phonetic details are normally omitted in dictionaries adopting Wong’s system. 11 For easy reference, the details regarding the orthographic conventions of the original Yale scheme, the ‘simplified’ Yale version, the predecessor/successor of the ILE scheme, and the original schemes of Rao’s and Wong’s systems are not shown in the tables. They are addressed in the preceding paragraphs.

References Ball, James Dyer (1883) Cantonese Made Easy, Hong Kong: China Mail Office. Bauer, Robert S. and Paul K. Benedict (1997) Modern Cantonese Phonology, Berlin: Mouton de Gruyter. Bauer, Robert S., Cheung Kwan-Hin, and Cheung Pak-Man (2003) ‘Variation and Merger of the Rising Tones in Hong Kong Cantonese’, Language Variation and Change 15(2): 211–25. Bridgman, Elijah Coleman (1841) Chinese Chrestomathy in the Canton Dialect, Macau: S. W. Williams. Chalmers, John (1859) An English and Cantonese Pocket-dictionary, for the Use of Those Who Wish to Learn the Spoken Language of Canton Province, Hong Kong: The London Missionary Society’s Press. Chao, Yuen-Ren (1947) Cantonese Primer, Cambridge, Mass.: Harvard University Press. Cheng, Siu-Pong (2014) ‘Gang-Ao liang di de zhengfu Yueyu pinyin’ (‘The Government Romanization of Cantonese in Hong Kong and Macao’), Zhongguo yuwen tongxun (Newsletter of Chinese Language) 93(1): 27–38. Cheung, Kwan-Hin (1986) ‘The Phonology of Present-day Cantonese’, Doctoral Dissertation, University College London. Cheung, Kwan-Hin (1994) ‘Xianggang Yuyanxue Xuehui Yueyu Pinyin Fang’an de yuanqi, sheji yuanze he tedian’ (‘The Genesis, Design Principles, and Characteristics of the LSHK Cantonese Romanization Scheme’), in Yueyu Pinyin Zibiao (Guide to LSHK Cantonese Romanization of Chinese Characters), Hong Kong: Linguistic Society of Hong Kong, ix–xix. Cheung, Kwan-Hin (2003) ‘Guangzhouhua Pinyin Fang’an xilie de wanshanhua guocheng’ (‘A Series of Refinements to the Cantonese Romanization Scheme 1960’), in Tang Keng-Pan (ed) Proceeding of 6th International Conference on Yue Dialects, Macau: Associação de Literatura Chinesa de Macau, 13–19.

49

Cheng Siu-Pong and Tang Sze-Wing Eitel, Ernst John (1877) A Chinese Dictionary in the Cantonese Dialect, London: Trübner and Company. Guangdongsheng sizhong fangyan pinyin fang’an (Four schemes of dialectal phonetic symbols of the Guangdong Province) (1960) Yuwen jianshe (Language Planning) 15: 21–5. Huang, Parker Po-Fei (1962) Speak Cantonese: Book II, New Haven: Institute of Far Eastern Languages, Yale University. Huang, Parker Po-Fei (1970) Cantonese Dictionary: Cantonese–English, English–Cantonese, New Haven: Yale University Press. Huang, Parker Po-Fei and Gerald P. Kok (1958) Speak Cantonese, New Haven: Institute of Far Eastern Languages, Yale University. Institute of Language in Education (1990) Changyongzi Guangzhouhua duyinbiao (List of Cantonese Pronunciation of Commonly-used Chinese Characters), Hong Kong: Government Printer. Kataoka, Shin (2014) ‘Xianggang zhengfu pinyin: yi ge luanzhongyouxu de xitong’ (‘Finding Order in Disorder: An Investigation into the Hong Kong Government Cantonese Romanization’), Zhongguo yuwen tongxun (Newsletter of Chinese Language) 93(1): 9–25. Kataoka, Shin and Cream Lee (2008) ‘A System without a System: Cantonese Romanization Used in Hong Kong Place and Personal Names’, Hong Kong Journal of Applied Linguistics 11(1): 79–98. Lam, Yan-Yan (2009) ‘Xianggang Yueyu biaoyin de xianzhuang’ (‘The Current Situation of Hong Kong Cantonese Sound Notation’), Zhongguo yuwen yanjiu (Studies in Chinese Linguistics) 27(1): 59–66. Lau, Sidney (1965) ‘Kwang Tung Wah’: Cantonese by Radio, Intermediate. Listeners’ Guides to Accompany Lessons 1 to 117, Hong Kong: Government Printer. Matthews, Stephen and Virginia Yip (1994) Cantonese: A Comprehensive Grammar, London and New York: Routledge. Meyer, Bernard F. and Theodore F. Wempe (1934) The Student’s Cantonese–English Dictionary, Hong Kong: St. Louis Industrial School Printing Press. Morrison, Robert (1828) A Vocabulary of the Canton Dialect, Macau: The Honorable East India Company’s Press. Williams, S. Wells (1856) A Tonic Dictionary of the Chinese Language in the Canton Dialect, Guangzhou: The Office of the Chinese Repository. Wong, S. L. (1941) Yueyin yunhui (A Chinese Syllabary Pronounced according to the Dialect of Canton), Shanghai: Zhonghua. Wu, Man-Ying (1997) ‘Yueyu Luoma pinyin fang’an de bijiao’ (‘Comparison of Cantonese Romanization Schemes’), Master Thesis, The Hong Kong Polytechnic University. Yu, Ping-Chiu (1971) Tongyin zihui (Rhyming Glossary), Hong Kong: Kwong Wah. Zee, Yun-Yang Eric (1999) ‘Chinese (Hong Kong Cantonese)’, in International Phonetic Association (ed) Handbook of the International Phonetic Association, Cambridge: Cambridge University Press, 58–60. Zhan, Bohui (ed) (2002) Guangzhouhua zhengyin zidian (The Dictionary of Standard Cantonese Pronunciation), Guangzhou: Guangdong Renmin Chubanshe.

50

4 Chinese characters John Jing-hua Yin University of Vermont, USA

The Chinese language, which is spoken by the largest number of people on the earth as their mother tongue, has many different dialects in addition to Mandarin Chinese. Despite the different accents of these dialects, all Chinese language speakers share one writing system enabling them to communicate with each other in writing when dialects are too difficult to understand by listening. The Chinese writing system, unlike the writing system of alphabetical languages, is formed with no letters or combination of letters to represent the sounds of the language. It is formed with Chinese characters, which are symbols constructed and used to convey meanings.

History and development of Chinese characters Although it is still not known how long Chinese characters have been in existence, we do know that Chinese characters have been a highly developed writing system for at least 3,300 years. The long history of the Chinese characters can be divided into two major periods: the ancient period and the modern period. Associated with the two periods, there are six major writing styles, starting from the earliest systematic form of Chinese characters inscribed on oracle bones. During the ancient writing period, characters we can still see today were inscribed on tortoise shells and animal bones. These characters are often referred to as ‘oracle-bone inscriptions,’ and they date to the Shang dynasty (1711 bce–1066 bce) in China. Later on, Chinese characters were cast or inscribed on bronze bells and vessels, and these characters are often called ‘bronze inscriptions.’ The bronze inscriptions were common in the Zhou dynasty (1066 bce–256 bce). While the characters’ size, positioning, complexity, and textual format in the oracle-bone inscriptions are inconsistent, the size and the textual format in the bronze inscriptions are more fixed. In bronze inscriptions, the characters are more symmetrical and the strokes are simpler. Toward the end of the Zhou dynasty, its Qin state began to use a new script both to write on bamboo strips and pieces of silk and to inscribe onto rocks or precious stones. This script is called the ‘seal script,’ which is still used for inscribing names on seals nowadays. After the Qin state conquered the other six states of the Warring States period (475 bce– 221 bce), unified China, and established the Qin dynasty (221 bce–206 bce), the seal script 51

John Jing-hua Yin

was decreed as the standard of writing for the whole country. In the seal script, the positioning of characters and complexity of forms became consistent, and all the characters were roughly square in shape. However, the character strokes became lengthened, curved, and complicated. Because a large of number of documents had to be written or copied as part of the administrative work after the unification of China, the seal script with its lengthened and curved strokes was quite time-consuming and cumbersome. Some local officials at the lower levels of the government who dealt with relatively unimportant documents developed a new script to save time. This new script continued to be developed in the Han dynasty (206 bce–ce 220) and became the officially approved formal way of writing. This script is now referred to as the ‘clerical script.’ The clerical script is very different from the seal script because there are four basic simplifications and modifications for convenience and speed of writing as well as for tidiness. First, the curved strokes in the seal script became somewhat straighter. Second, the overall number of strokes was reduced. Third, some different components were merged into one. Fourth, some components were modified and simplified. The change from the seal script to the clerical script is often referred to as the ‘clerical change.’ This change, which is the largest transformation of Chinese character structure in history, dropped the pictographic appearance of Chinese characters almost completely and established the foundation of the structures for the Modern Chinese characters. Therefore, in the history of Chinese character development, the clerical script is viewed as the turning point that divides the ancient writing period and the modern writing period. The ancient writing period, from the earliest known oracle-bone inscriptions in the late Shang dynasty to the development of the seal script, lasted about 1,160 years, and the modern writing period, from the time of the clerical change to the present, has lasted more than 2,200 years. This period of time in the history of Chinese character development is still considered modern because the structures of Chinese characters have remained the same from the clerical change until today. Although the structures of Chinese characters have not changed since the clerical change, the strokes of Chinese characters have undergone two major changes: regularization and normalization. Toward the end of the Han dynasty, strokes with an undulate end, which was a common feature of characters in the clerical script, became smooth and straight in the regular script, and some thick curvy lines seen in the clerical script changed to hooks in the regular script. This change is known as ‘regularization.’ As the strokes were relatively smoother, straighter, and clearer in the regular script, it was easier to read and write. Regular script quickly became widespread. The invention of printing helped the regular script, which was used as one of the major printing types, become widely accepted as the standard form of Chinese writing. Many calligraphers, such as Zhong Yao in the Three Kindoms period (220–280), Wang Xizhi and Wang Xianzhi in the Eastern Jin dynasty (317–420), Ouyang Xun, Yan Zhenqing, and Liu Gonquan in the Tang dynasty (618–907), Su Shi, Huang Tingjian, and Mi Fu in the Song dynasty (960–1279), and Zhao Mengfu in the Yuan dynasty (1279–1368), among many others, also contributed to making the regular script the standard for people in China to admire, imitate, learn, and master. The regular script has been the standard of Chinese writing for more than 1,800 years, with the semi-cursive and cursive styles developing concurrently. However, semi-cursive and cursive styles, as there are many variations of them, have never become standardized printing types. Instead, they have been used for everyday communication and appreciated as a form of art. 52

Chinese characters

Although the regular script changed the appearance of strokes in the clerical script, many characters continued to be complicated. Some characters have variations resulting from many centuries of use, during which time character variants with fewer strokes and simpler structures were created for convenience. Other characters have variations arising from a lack of uniformity among printing types after the invention of printing. Although these alternative forms of characters were popularly used for years, they were never officially recognized or accepted as the ‘standard forms of characters’ up to the 1950s. In the first three and a half decades of the twentieth century, attempts to accept these ‘non-standard forms of characters’ officially and to continue to simplify more Chinese characters were made, but they failed for one reason or another. However, these efforts laid the foundation for the language reform that took place in the second half of the twentieth century. After World War II and the civil war in China in the late 1940s, the normalization of Chinese characters, which comprises systemization, simplification, and standardization, was finally on the agenda of the Chinese government. A special government organization, first called the Committee for Chinese Language Reform and later the National Language Commission, was established in 1954 to direct and oversee the normalization of Chinese characters. In 1955, to systemize Chinese characters, the ‘List of the First Group of Standardized Forms of Variant Characters’ was officially published. Among the 810 sets of characters, with two to six variants in each set, 1,027 variant characters were abolished, and the remaining characters have been referred to as the ‘standard forms of characters’ since then. In 1956, after discussion, revision, and consultation, the ‘Scheme for Simplifying Chinese Characters’ was officially published. In 1964, the ‘Complete List of Simplified Characters’ was officially published, and it was republished in 1986. The list has 2,235 simplified characters in total, and 1,116 of them are frequently used today in daily life. In fact, among the 2,235 simplified characters, only 482 of them are basic. The remaining 1,753 characters in the list are all derivatives of these 482 basic simplified characters. Among the 482 basic ones, about 20% were newly created in the 1950s, while the other 80% were created as long as several thousand years ago. In 1964, in an effort to standardize the printed forms of Chinese characters, the ‘List of Chinese Character Forms for General Printing’ was officially published. The list has provided the standardized printed forms for 6,196 generally used characters. In 1988, the ‘List of Frequently Used Characters in Modern Chinese,’ which includes 3,500 characters, was officially published, and later in the same year, the ‘List of Generally Used Characters in Modern Chinese’ was also officially published. This second list contains 7,000 characters, which include the 3,500 frequently used ones contained in the first list. In 1997, ‘Standard Stroke Order for Generally Used Characters in Modern Chinese’ was officially published. After 50 years of simplifying, systemizing, and standardizing work on Chinese characters, they are considered normalized. The ‘normalized forms of Chinese characters’ include two major types: (i) the inherited characters, which are the Chinese characters that have kept their structure ever since the ‘clerical change’ time, and (ii) the simplified characters, which are the Chinese characters that have been simplified and officially recognized. The original forms of the characters, before being simplified or replaced by today’s simple forms, are often referred to as ‘complex characters’ by people in mainland China. These complex characters together with the inherited characters constitute the ‘traditional forms of Chinese characters.’ ‘Modern Chinese characters’ is another term referring to the normalized forms of Chinese characters. They are officially recognized and used in mainland China, Singapore, and 53

John Jing-hua Yin Table 4.1  Ancient writing period (c. 1711 Major forms

Ancient form

Oracle-bone inscription

Most popular time

Examples

Shang dynasty (1711 bce–1066

person

tree

water

fire

cart

horse

bird

cloud

Zhou dynasty (1066 bce–256

Seal script

Late Zhou dynasty and the Qin dynasty (221 bce–206 bce)

Major forms Traditional form

bce)

Bronze inscription

Table 4.2  Modern writing period (c. 206

Normalized form

bce–256 bce)

Clerical script

bce)

bce–present)

Most popular time

Examples

Qin dynasty and Han dynasty (206 bce–ce 220)

From the late Han dynasty to 1955 Regular script From 1955 to the present

Malaysia, as well as at the United Nations. The traditional forms of Chinese characters are still in use in Taiwan, Hong Kong, and Macao, as well as among some Chinese communities outside China. Modern Chinese characters as included in the ‘List of Generally Used Characters in Modern Chinese’ consist of about 31% simplified characters and about 69% inherited characters. Tables 4.1 and 4.2 provide examples of the historical development of Chinese characters. The development of Chinese characters in terms of the evolution of forms over the past several thousand years has demonstrated three general trends in the process. First, Chinese characters used to be drawings depicting objects and have slowly become more abstract writing symbols. Second, Chinese characters started with the pictographic method as the major way of character construction and have ended up using more and more abstract 54

Chinese characters

symbols developed from pictographs as phonetic symbols in new characters. Third, Chinese characters used to have a lot of pictographic symbols that were complicated to write, but they have now become logographs that are much easier to write.

Formation and structure of Chinese characters In the earliest known stages of written Chinese, there were six kinds of Chinese characters: pictographic characters, indicative characters, associative characters, picto-phonetic characters, explanative characters, and phonetic loan characters. The first five kinds (i.e. pictographic, indicative, associative, picto-phonetic, and explanative characters) are categorized according to the ways characters were created, and the last one (i.e. phonetic loan characters) complemented the other methods by borrowing existing characters to refer to things that no characters had been created for. These six kinds of Chinese characters, which were summarized and exemplified by Xu Shen in his An Analysis and Explanation of Characters at the beginning of the first century ce, reflect the earliest stages of the creation and use of Chinese characters. 1. The pictographic method is the first major way of creating Chinese characters that came into use. The human body or body parts as well as things that can be observed in nature were depicted in simple drawings based on their most conspicuous and differentiated traits. 2. However, an abstract concept is not expressed easily by drawing a picture, so the indicative method, in which a symbol was added to a drawing to indicate the concept, was adopted. 3. When a meaning could be sensed or deduced by combining two existing pictographs, a combined form was produced to save the trouble of creating a new pictograph. This method of combining two existing pictographs is called the associative method. 4. When a meaning could not be expressed by creating a character with the associative method, then the picto-phonetic method was deployed. The picto-phonetic method forms a character with a pictograph as one part to indicate the major category of meaning and another part to indicate the pronunciation of the whole character. 5. When a pictographic component representing both the sound and the meaning was viewed as unable to indicate the meaning clearly, another pictographic component would be added to clarify or differentiate the meaning of the whole character. This method is the explanative method. The explanative method was also deployed when a character was borrowed (but never returned) to mean something else simply because this borrowed character had the same pronunciation as the new idea, which as yet had no character to express it. 6. Those borrowed characters are called phonetic loan characters. Among the above six methods, the picto-phonetic method has been the most productive in creating characters. Because the explanative method has not been clearly understood for many years and because a lot of work to distinguish explanative characters from pictophonetic characters has still not been satisfactorily done, many explanative characters are confused with picto-phonetic characters. Nevertheless, it is widely accepted that over 90% of the characters used today are picto-phonetic in a broad sense. Of course, because of developments and changes in the language over the past few thousand years, in Modern Chinese no more than 30% of the picto-phonetic characters contain a phonetic component that can accurately represent the pronunciation of the whole character. 55

John Jing-hua Yin

Table 4.3 shows examples of the six categories of Chinese character construction: Table 4.3  Categories of Chinese character construction Categories 1.

2.

3.

4.

5.

6.

Examples

Pictographic characters 人 person

木 tree

上 upper

本 root

tree + tree = woods

person + tree = rest

water + (sound) = shower

water + (sound) = wash

tree + branch = branch (sound & meaning)

water + spill = spill (sound & meaning)

六 six (originally ‘hut’)

北 north (originally ‘back’)

Indicative characters

Associative characters

Picto-phonetic characters

Explanative characters

Phonetic loan characters

Regardless of how a character was created, when created it almost always had the trait of a pictograph one way or another. Over thousands of years, the pictographic essence slowly faded and is no longer visually obvious in Modern Chinese characters. Now, formed with various kinds of lines, dots, and hooks, Chinese characters are highly symbolic. They mostly appear as logographs rather than pictographs. In other words, they look more like symbols than drawings.

Features of Chinese characters 1.  One character, one phonetic syllable A Chinese character is an independent symbol that expresses a meaning. What makes the pronunciation of a character easy is the fact that the pronunciation of any character in Chinese is monosyllabic: one character represents one phonetic syllable. A phonetic syllable is a segment of pronunciation that contains one vowel sound with or without one or more consonant sounds. Vowel sounds are loud and clear. When they are pronounced, the airflow

56

Chinese characters

from the lungs vibrates the vocal chords, and the shape of the mouth and the position of the tongue make the different sounds. In Chinese, it does not matter what a character means, and it does not matter whether the character is a word by itself or a morpheme (part of word), it is always pronounced with one syllable.

2.  One syllable, often many different characters Although it is convenient that each character is pronounced in one syllable, the tones make it a bit complicated. What makes the relationship between characters and the Chinese sound system more complicated are the homophones. In the Chinese sound system, there are only about 400 different syllables. Even with as many as four different tones for each syllable, the total number of syllables with different tones in use in Chinese is 1,196. On the other hand, as collected and recorded in the Giant Dictionary of the Chinese Language, published in 1990 in China, the total number of Chinese characters that have been created since ancient times is 54,678. If we take the 2,500 characters that were found to be the most frequently used and divide by 1,196, which is the total number of different syllables in Mandarin Chinese, we still get about two characters for each syllable on average. In reality, some syllables are represented by more than two of these 2,500 most frequently used characters. Take the syllable yi as an example. The first-tone syllable yī has four characters; the second-tone syllable yí has six characters; the third-tone syllable yǐ has six characters; and the fourth-tone syllable yì has 16 characters. It would be impossible to understand what someone means when he or she says yi in any of the four tones out of the context.

3.  Not all characters are in active use The number of Chinese characters has grown over the last few thousand years, as indicated by the major dictionaries that have been compiled during that time. An Analysis and Explanation of Characters is the earliest dictionary we know of. It was put together by Xu Shen around 121 ce with a collection of 9,353 characters. The Kangxi Dictionary, compiled by Zhang Yushu, Chen Tingjing, and others in 1716 under the rule of Emperor Kangxi in the Qing dynasty (1644–1911), includes 47,043 characters. The Giant Dictionary of the Chinese Language, compiled by Xu Zhongshu and others and published in 1990, records 54,678 characters. However, the number of characters in general use in modern China is about 7,000. Among the 7,000 characters, the general public now frequently uses only about 3,500, as found in the ‘List of Frequently Used Chinese Characters’ published by the National Language Commission of China in 1988. Furthermore, a test done in 1987 by Shanxi University, under the request of the National Language Commission, on randomly selected samples of printed materials containing 2 million characters, shows that 1) the 2,500 most frequently used characters in the first part of the list can account for 97.97% of the characters that appeared in the data; 2) the next 1,000 characters in the second part of the list can cover 1.51%; and 3) the total coverage rate of the two groups (with 3,500 characters in all) is 99.48%. This means that if someone could recognize the first 2,500 characters in the list, he or she would run into only two unrecognized characters in every 100 characters that appear in Chinese printed materials similar to the tested data. Having learned the remaining 1,000 characters in the list, he or she would be able to reduce the number of unrecognized characters to one in every one hundred characters seen in the same kind of printed materials.

57

John Jing-hua Yin

4.  Not all characters are words It is encouraging to realize that knowing 3,500 of the most frequently used characters enables one to recognize 99% of the characters that appear in newspapers, magazines, and books for general readers, but that knowledge does not guarantee readers’ understanding because characters are not always words in Modern Chinese. One must know the relationship between the characters in a word and the meaning of a word formed by two or more characters. In Classical Chinese, most words are only one character. However, in Modern Chinese, most words contain two characters; as a result, many of the characters that would be words in Classical Chinese are now morphemes that form words in Modern Chinese. In designing the Chinese Language Proficiency Test for learners of Chinese as a foreign language in 1992, the Testing Center of the Office for the National Chinese Language Proficiency Test Committee developed the ‘General Outline of the Chinese Vocabulary Levels and Graded Chinese Characters.’ It is based on 16 different lists of Chinese words (not characters) and includes 8,822 carefully selected words at four levels (refer to Table 4.4). The developers predicted that 3,000 of the most frequently used words would cover 86% of the language materials for general readers, 5,000 would cover 91%, and 8,000 would cover 95%. If we look at the number of characters that the selected 8,822 words contain and compare them with the 2,500 most frequently used characters in the ‘List of Frequently Used Characters’ mentioned earlier, we see that the developers’ prediction about the vocabulary coverage rate is believable. The 8,822 words in the ‘General Outline’ have been formed with 2,905 characters, of which 2,485 characters overlap with the 2,500 characters that appear in the first part of the ‘List of Frequently Used Chinese Characters.’ Although not all characters are words, knowing the meaning of characters will help one learn the meaning of the words they form. If the best way to increase one’s English vocabulary is to learn prefixes, suffixes, and roots, then similarly the most efficient way to increase one’s Chinese vocabulary is to learn Chinese characters, especially those most frequently used characters.

5.  The smallest units forming a character are strokes Chinese characters originated as pictographs, but through thousands of years of development and changes, the pictographic essence slowly faded and is not visually obvious in Modern Chinese characters, which now are highly symbolic and constructed with various kinds of lines, dots, and hooks. These lines, dots, and hooks are called ‘strokes.’ A stroke is a mark made by a writing instrument, such as a pen, that is put down on a writing object, such as paper, and moved until it is lifted from the writing object. These strokes are arranged in a special way to form characters. Some characters have two or more components, and these components are also formed by strokes. Strokes are the smallest units used to form characters. In Modern Chinese characters, there are six basic strokes: the horizontal stroke, the vertical stroke, the left-falling curved stroke, the right-falling curved stroke, the dot, and the rising stroke. Table 4.4  Number of words and characters at each proficiency level

Words Characters

Level 1

Level 2

Level 3

Level 4

Total

1,033    800

2,018    804

2,202    601

3,569    700

8,822 2,905

58

Chinese characters

Apart from the six basic strokes, there are also three dependent strokes: the bend, the turn, and the hook. Dependent strokes cannot be used independently; they have to be attached to at least one other kind of stroke to form a complex stroke. The six basic strokes and the three dependent strokes together form all of the 27 compound strokes used in Modern Chinese characters. Table 4.5 is a list of all the strokes in Modern Chinese characters.

6.  There are stroke order rules to follow when writing characters If there are more than two strokes in a Chinese character, there are rules for conventional stroke order that help writers to balance Chinese characters. The knowledge of stroke order also helps people appreciate handwritten forms of Chinese. Rule 1: From top to bottom. The strokes at the top of a character should be written before those at the bottom. For example: 三 (three): Rule 2: From left to right. The strokes on the left side of a character should be written before those on the right. For example: 八 (eight): Rule 3: The horizontal before the vertical or the left-falling. The horizontal stroke should be written before the vertical stroke or the left-falling stroke if they cross each other. For example: 十 (ten): Rule 4: The enclosing strokes first, then the enclosed, and finally the sealing horizontal stroke. The enclosing strokes, which form the sides and top of the frame, should be written first. Strokes inside the frame should be written next, and then the closing stroke, which is usually a horizontal stroke at the bottom of the frame. For example: 四 (four): Rule 5: Left-falling before right-falling. When the left-falling stroke and the rightfalling stroke come together in a character, write the left-falling stroke first and then write the right-falling stroke. For example: 木 (tree): Rule 6: Vertical stroke in the middle before the strokes on both sides. When a vertical stroke or the stroke containing a vertical segment in the prominent middle position does not cross other strokes, it should be written first. For example: 水 (water): These rules are not absolute, but relative and mutually conditioned. For example, Rule 3 requires us to write the horizontal stroke before the vertical when they are together in a character; however, if the horizontal stroke does not cross the vertical stroke and is on its right side, then the vertical stroke goes before the horizontal stroke. Therefore, in the . Sometimes, after lower part of the character 足 (foot), the stroke order is years of writing, people find that one way of writing is easier than another, even though it does not comply with the stroke order rules. Therefore, there are exceptions to the above rules. For example, if you follow Rule 3, there is no exception in writing the character . However, writing the character 女 (female) is an exception because the 大 (big) as conventional way of writing 女 does not start with the horizontal stroke. It goes like this: . You may find it easier to write out a well-balanced 女 this way than having the horizontal stroke first. Another example is the character 子 (baby). The horizontal stroke is . drawn last: 59

Table 4.5  Chinese character strokes Stroke

Basic strokes

Name in Chinese Example

Basic stroke variants

Variant 1

Variant 2

Variant 3

héng

shù

ー

ー

piě

nà

horizontal

vertical

leftfalling

rightfalling

short horizontal

short vertical

vertical leftfalling

leveled rightfalling

pointed vertical

small leftfalling

Dependent strokes diǎn

tí

wān

zhé

gōu

ヽ dot

rising

bend

turn

hook

left dot

big right dot

leveled leftfalling

Two strokes combined

Compound strokes

ー ー

plus plus

plus wān plus (bending) wò plus (crouching) xié plus (slanting)

Three strokes combined

plus plus plus

Four strokes combined

plus plus plus plus plus

Five strokes combined

plus

plus

 

Chinese characters

7.  Regardless of the number of strokes or components a character may consist of, it should be written consistently in the same size square box as the others In English, a word that has more letters is longer than a word that has fewer letters. Longer words occupy more space than shorter words when spelled out. In Chinese, even though associative, explanative, and picto-phonetic characters are formed with two or more character components, they still have to be written in the same size as those single-element characters. Therefore, the number of strokes and components does not determine the size of a character. All characters should be written consistently in the same size. The formation structures are determined by the way character components are put together. These components are also written out in a conventionally fixed sequence. See the following. The numbers given in the boxes below indicate the character component writing sequence. 1

2

  1. Top–Bottom (T–B) structure: 男, 分

2

1

  2. Left–Right (L–R) structure: 明, 好 1

2

3

  3. T–B structure embedded with an L–R on the top: 哭, 想

1

2

  4. T–B structure embedded with an L–R at the bottom: 罚, 霜

3

2 1 3

  5. L–R structure embedded with a T–B on the right: 解, 语 1 3

  6. L–R structure embedded with a T–B on the left: 剖, 数 1

2

3

  7. Left–Middle–Right structure: 湖, 街 1 2 3

  8. Top–Middle–Bottom structure: 意, 爱 61

2

John Jing-hua Yin

1 2 3

  9. Complete-Wrap structure: 困, 园 1

2

10. Top-Wrap structure: 风, 同 1 2

11. Bottom-Left-Wrap structure: 道, 近

1

2

12. Top-Right-Wrap structure: 包, 氧 1

2

13. Top-Left-Wrap structure: 病, 庆 1 2

14. Left-Wrap structure: 匣, 匹

3

1

15. Bottom-Wrap structure: 画, 凶

2

8.  Characters contain recurring side components, by which characters are arranged in character dictionaries and with which people learn to recognize more characters Multicomponent characters have many recurring side components, which are commonly referred to as ‘radicals.’ The entries in a dictionary are traditionally categorized by these radicals. People also learn to recognize more characters by knowing the meanings of these radicals first. Here are the characters commonly used as radicals1: 厂卜八人儿几刀力厶又士土工寸大弋小口山巾广门(門)尸己弓子女马(馬)幺王 韦(韋)木犬歹车(車)戈比瓦止日曰贝(貝)水见(見)牛手毛气片斤爪父月欠风(風) 殳文方火斗户心毋示石龙(龍)业(業)目田皿矢禾白瓜用鸟(鳥)立穴疋皮矛耒老耳 臣页(頁)虫缶舌竹臼自血舟衣羊米艮羽糸麦(麥)走赤豆酉辰豕卤(鹵)里足身采谷 豸角言辛青其雨齿(齒)黾(黽)隹金鱼(魚)革骨鬼食音鬥髟麻鹿黑鼠鼻 62

Chinese characters

There are also radicals that cannot be used alone as characters since they are either strokes or dependent components, including the reduced forms of characters: 勹匚冂廴廾刂亻亠冫冖讠(訁)卩  阝 2 阝 3 凵彐巛扌艹(艹)尢囗彳彡犭夕夂饣(飠) 丬(爿)氵忄宀辶纟(糹)屮攵攴灬礻肀罒钅(釒)疒衤覀虍

Notes 1 Those simplified characters used as radicals are followed by their traditional counterparts in parentheses. 2 A reduced form of the character 阜 (mound) used as a left side component in a character. 3 A reduced form of the character 邑 (town) used as a right side component in a character.

Bibliography Guojia Duiwai Hanyu Jiaoxue Lingdao Xiaozu Bangongshi Hanyu Shuiping Kaoshibu 国家对外汉语 教学领导小组办公室汉语水平考试部 ‘Chinese Language Proficiency Level Test Division of the National Office of the Leading Group for Teaching Chinese to Speakers of Other Languages’ (1992) Hanyu Shuiping Cihui Yu Hanzi Dengji Dagang《汉语水平词汇与汉字等级大纲》‘General Outline of the Chinese Vocabulary Levels and Graded Chinese Characters’, Beijing: Beijing Yuyan Xueyuan Chubanshe. Hanyu Dazidian Bianji Weiyuanhui 汉语大字典编辑委员会 ‘Editorial Committee of the Chinese Giant Dictionary’ (1995) Hanyu Dazidian (Suoyiben)《汉语大字典(缩印本)》‘Chinese Giant Dictionary (Compact Edition)’, Wuhan: Hubei Cishu Chubanshe and Chengdu: Sichuan Cishu Chubanshe. Li, Dasui 李大遂 (1993) Jianming Shiyong Hanzixue《简明实用汉字学》‘Concise and Practical Study of Chinese Characters’, Beijing: Beijing Daxue Chubanshe. Su, Peicheng 苏培成 (1994) Xiandai Hanzixue Gangyao《现代汉字学纲要》‘Essentials of the Study of Modern Chinese Characters’, Beijing: Beijing Daxue Chubanshe. Xinhua Dictionary with English Translation《英汉双解新华字典》(2000) Beijing: The Commercial Press. Xu, Shen 许慎 (1963) Shuo Wen Jie Zi《说文解字》‘An Analysis and Explanation of Characters’, Beijing: Zhonghua Shuju. Yin, Bingyong 尹斌庸 and J. S. Rohsenow (1994) Modern Chinese Characters《现代汉字》, Beijing: Sinolingua. Yin, John Jing-hua 印京华 (2006) Fundamentals of Chinese Characters, New Haven: Yale University Press. Yuyan Chubanshe 语言出版社 ‘Language Press’ (1997) Yuyan Wenzi Guifan Shouce《语言文字规范 手册》‘Language and Character Standardization Handbook’, Beijing: Yuyan Chubanshe. Zhang, Shuyan, Tiekun Wang, Qingmei Li, and Ning An 张书岩、王铁昆、李青梅、安宁 (eds) (1997) Jianhuazi Suyuan《简化字溯源》‘Origins of Simplified Characters’, Beijing: Yuwen Chubanshe. Zhao, Yuanren 赵元任 (1980) Yuyan Wenti《语言问题》‘Language Issues’, Beijing: The Commercial Press.

63

5 Chinese Idioms Jiao Liwei University of Pennsylvania, USA

1.  A brief introduction and three examples: 守株待兔 (shouzhu-daitu), 千方百计 (qianfang-baiji), and 当务之急 (dangwuzhiji) Chinese idioms (成语, chengyu) are one of four types of formulaic expressions (熟语, shuyu), which also include collocations (惯用语, guanyongyu), two-part allegorical sayings (歇后语, xiehouyu), and proverbs (谚语, yanyu; or 俗语, suyu) (D. Wen 2005: Ch. 2, Sec. 3). Conventionally, the term chengyu is translated into English as ‘idiom’, ‘set phrase’, or ‘four-character expression’. More than 90% of Chinese idioms consist of four characters, which can be parsed into groups of two when recited at normal speed. Typically, the meaning of idioms is more than the sum of the parts. In sentences, Chinese idioms function mainly as predicates, or often as attributives or adverbials, but seldom as subjects or objects. Chinese idioms make writing or speech more concise and more vivid, conveying a sharper impression to readers and listeners. Chinese idioms are an important marker of written language, but they also contribute to increasing the variety of colloquial expressions (L. Wang 1987: i). Below are three examples: ● 守株待兔 (shouzhu-daitu, guard-tree stump-wait for-hare, i.e. to wait passively for a windfall, wait for gains without pains, trust chance and windfalls). This idiom makes reference to a story in the Wu Du (five vermin) section in the Han Feizi (before 233 bc): ‘There was once a farmer who was ploughing land in the State of Song. One day a hare was dashing rapidly; all of a sudden, it ran into a wooden post in the ground, snapped its neck and died. The farmer was happy and therefore abandoned his plough and waited next to the wooden post, hoping to pick up hares again. But he did not get any more hares, and was mocked by the people in the State of Song.’ If one did not know the allusion, one could hardly understand the figurative meaning of this idiom, which goes far beyond the literal meaning of the four characters. ● 千方百计 (qianfang-baiji, thousand-method-hundred-plan, i.e. by every possible means, by hook or by crook). There are three grammatical or semantic linkages in this idiom: 千 (thousand) with 百 (hundred), 方 (method) with 计 (plan), 千方 (thousand methods) with 百计 (hundred plans), all of which contribute to the overall meaning. ● 当务之急 (dangwuzhiji, should-pursue-’s-urgency, i.e. a matter of great urgency). This idiom comes from a line in Mencius which nevertheless did not contribute to its qualification as an idiom. However, the nature of the four compositional characters in the written language is the prerequisite for this phrase to become an idiom. 64

Chinese idioms

2.  From a historical perspective: Song–Yuan dynasties, 1910s, and 1958 Nowadays idioms are highly esteemed by people from almost all walks of life in China, from middle school students to marketing managers of large pharmaceutical companies. As of July 15, 2013, a search of ‘成语’ (Chinese idioms) in book titles yielded 3,605 results in the Chinese literature collection of the National Library of China. It is extremely common to find Chinese idioms in newspapers. But it was not until the twentieth century that people began collecting and categorizing them (Fang 1943; Z. Zhou 1955; Ma 1962). The term 成语 was first used in the Tang dynasty. There were some collections of idioms in the Ming and Qing dynasties, such as《幼学须知》or《幼学琼林》(The Children’s Knowledge Treasury), compiled in the Ming dynasty,《陔余丛考》(The Notebook during Home Leisure Period, 1791),《恒言录》(The Record of Eternal Words, 1805),《常语寻源》(The Search for Origins of Common Sayings, 1876), or even as early as《释常谈》(Annotated Common Sayings) in the Song dynasty (see S. Wen and Wen 2009: 4–18; J. Zhou 2004). However, the entries, mixed with many other non-idioms, were intended mainly to remind the people that these expressions were quotations from earlier writers. The first true dictionary of idioms was《国语成语大全》(The Comprehensive Dictionary of Chinese Idioms), compiled by Houjue and published by Zhonghua Bookstore in 1926. It contains approximately 3,200 entries. Archiving and academic research on idioms emerged late, but the extensive use of idioms started early. The online Dictionary of Chinese Idioms《成语典》compiled by the Taiwan Ministry of Education has a nautilus fossil on its homepage to symbolize the fact that idioms are linguistic artifacts from the past – language fossils. However, though they are rooted in history, they are also widely used in modern times. Studies of the origin of idioms vary in their statistics. Xiao (1987), for example, claims that 3,128 of 4,600 idioms – roughly 68% – originated from before the Han dynasty, while only 8% dated from after the Song dynasty. Sun (1989: 90) has 15% of idioms dating from after the Yuan, while T. Zhang (1999) claims that of 6,593 idioms, 1,120 (14.52%) originate from before the Han dynasty, and 4,036 (55.56%) from after the Song dynasty. These differences reflect different ways of determining the date of an idiom (Sun 1989: 88–90). For example, 结草衔环 (jiecao-xianhuan) has two allusions: 结草 is from a story attested in 594 bc; 衔环 is from a story of a man who lived from ad 59 to 124. But the full form, 结草衔环, was not recorded until the Yuan dynasty (1271–1368). Almost all idioms consist of four characters, and more than 40.82% of idioms are symmetrical in structure (Y. Liu et al. 2010). According to Lu Zhiwei (1956), extensive use of juxtaposed quadrasyllabic expressions began as late as in the Yuan dynasty. From as early as the Song dynasty through to the beginning of the twentieth century, there was a major division between the formal/written language and the informal/vernacular language. That is to say, people used vernacular language when speaking, but Classical Chinese when writing, which is often considered a monosyllabic language. Poetry, prose, fiction, and drama, written mainly in Classical Chinese, were highly valued by the educated elite in history. Written Vernacular Chinese – modern written Mandarin – finally replaced Classical Chinese after the Written Vernacular Chinese Movement in the mid 1910s. Based on the major treebanks of Modern Chinese at University of Pennsylvania, Tsinghua University, Peking University, and Academia Sinica1 (see Wang Yuelong and Ji 2009), we estimate the average number of characters in a Chinese simple sentence is around 16. If a quadrasyllabic idiom appears in a 16-character sentence, it is salient in length, among other 65

Jiao Liwei

distinguished features including rhetorical and cultural ones. Since the average length of all non-idiomatical words in Chinese is 1.9 syllables/characters (H. Wang 2009), idioms of four syllables naturally leave a stronger impression upon readers and listeners. According to statistics by Chen (2011) and Zhang Jie (2007), there are 3,679 quadrasyllabic idioms of a total of 5,648 quadrasyllabic words and expressions in the authoritative Contemporary Chinese Dictionary (5th ed.); that is to say, about two thirds of all quadrasyllabic words or expressions are idioms. Idioms are strongly associated with quadrasyllables, which are longer than the Chinese standard foot of two or three syllables (Feng 1998). The above evidence shows that the rising status of Chinese idioms in language is a result of natural selection in language evolution and reform, and that the nature of idioms preserved certain forms of Classical Chinese and much content of Chinese culture. However, the dictionary《汉语成语小词典》(A Concise Dictionary of Chinese Idioms, 1958) itself changed the course of the spread of Chinese idioms among the people. On February 11, 1958, the Chinese Central Government promulgated the Scheme of the Chinese Phonetic Alphabet. In order to present a gift to the National Day on October 1, 17 linguistics students in the Department of Chinese at Peking University, under the supervision of the renowned linguists Wei Jiangong and Zhou Zumo, spent 20 days compiling a dictionary containing more than 2,700 Chinese idioms. This dictionary was the first of its kind to arrange entries according to the Latin alphabet, and was intended to target middle school students, cadres, and general readers with a middle school education. The dictionary became extremely popular. The first printing in 1958 was of 300,000 copies, and the first printing of the first revised edition in 1959 was of 800,000 copies. By 2012 this dictionary had had six editions and 165 printings, and had greatly popularized knowledge of idioms in mainland China. With the outburst of reality talent shows in China, two programs on idioms were broadcast on TV. The first one, named ‘Idioms Hero’ (成语英雄), was hosted by Henan TV and first aired in 2013. In the second season, the producers recruited a prominent writer, the former Minister of Culture, Wang Meng, to be guest commentator. The other program, ‘Chinese Idioms Conference’ (中国成语大会) was first broadcast in 2014 at prime time on China Central Television, and won the highest audience ratings among Sunday evening shows in its first season. Examinations play a crucial role in the Chinese formal education system and from October 1954, Chinese idioms have been included in the Chinese language guidelines of elementary schools. Chinese Language and Literature is compulsory part of the annual College Entrance Exams, which are vital to Chinese students’ futures. Since idioms are always tested, teachers and students pay great attention to them. Education institutions in Taiwan value Chinese idioms just as highly as those in mainland China. The Ministry of Education of Taiwan publishes an online dictionary of idioms, named 成语典 http://dict.idioms.moe.edu.tw/cydic/ index.htm. 成语典 and 成语词典 differ by one character only. 典 means ‘classic’ while 词典 means ‘dictionary’. The importance of idioms is stressed in the book title. At the turn of the twenty-first century and along with rise of the Internet and social media, the power to create new words and expressions shifted from the pens of famous writers to the keyboards and mouths of celebrities and occasionally to ordinary people. New words or expressions such as 神马都是浮云 (shenma doushi fuyun, celestial-horses-all-arepassing-clouds, i.e. nothing is eternal) seemingly prevail over idioms like 万事皆空 (wanshijiekong, all-things-are-empty, i.e. nothing is eternal). However, Chinese idioms will continue to be very important in the realm of Chinese words and expressions, owing to the fact that the use of idioms has remained stable throughout history and their prestige in the education of Chinese language and literature to young people. 66

Chinese idioms

3.  Quantity of Chinese idioms: 50,000, 10,000, and 500 In English there are about 25,000 idioms. As for Chinese idioms, some comprehensive dictionaries catalogue more than 20,000 entries. There are, for instance, 28,505 entries in 《成语典》(Dictionary   of Chinese Idioms) of the Ministry of Education of Taiwan, 45,000 in《汉大成语大词典》(‘Han Da’ Comprehensive Dictionary of Chinese Idioms), and 50,000 in《汉语成语源流大辞典》(Etymological Comprehensive Dictionary of Chinese Idioms) by Liu Jiexiu. For compact dictionaries,《汉语成语小词典》(A Portable Dictionary of Chinese Idioms) by the Commercial Press collects 4,600 common idioms, and the authoritative Contemporary Chinese Dictionary) has 3,400 entries. According to 《现代汉语词典》(The   one estimate, there are about 10,000 Chinese idioms excluding variants (T. Zhang 1999). Liu and Qin (2007) analyzed statistics from a corpus of 426 million Chinese characters collected from archives of 15 major newspapers in China over the course of a year and found a total of 8,637 different four-character idioms. Jiao et al. (2011) collected statistics from 60 million characters based on five corpora of different types and found 8,844 four-character idioms. It is a reasonable guess that the number of Chinese idioms in actual use is about 8,000 to 10,000. The frequency of occurrence of idioms naturally varies greatly. According to C. Liu and Qin (2007), cumulative occurrences of the 405 most common idioms account for 41.5% of occurrences of all idioms. Another statistic is the cumulative occurrence of the 500 most common idioms, which make up about 50% of the occurrences of all idioms (Jiao et al. 2011: vii). Therefore, we can infer that there are 400 to 500 very common idioms in the Chinese language. A total of 755 idioms appear in all the 24 volumes of Chinese textbooks in use in mainland China in 2009 (Jiao et al. 2011). As for single volume books, Dream of the Red Chamber (《红楼梦》), the greatest novel in the history of Chinese literature, uses 1,429 idioms in its total 730,000 characters. Among its 1,429 idioms, 1,044 are colloquial idioms which include proverbs and collocations (see J. Zhou and Shi 2012). This being said, there are only about 400 written-style idioms. In the most famous historical novel, Romance of the Three Kingdoms (《三国演义》), about 500 idioms were used. As for individuals in modern history, Mao Zedong was well known for his language skills, but there are only about 600 idioms and allusions in the five-volume Selected Works of Mao Zedong. There are 481 idioms in the three-volume Selected Works of Deng Xiaoping. One can conclude, therefore, that the number of Chinese idioms known to the average well-educated person is about 500.

4.  Definition of Chinese idioms: quadrasyllabic, conventional, and semantic duality On November 3, 2005, the name of the famous Western fairy tale The Three Little Pigs was included as an entry in the reference materials section in the online Dictionary of Chinese Idioms of Taiwan. In July of 2007 this matter was much hyped by the media, with the Minister of Education at that time asserting that 三只小猪 (san zhi xiaozhu, ‘three little pigs’) was, in fact, now a Chinese idiom. His opinion was ridiculed by the media and by linguistic professionals, both in Taiwan and in mainland China.2 But why was the phrase 三只小猪, consisting of four characters and with its apparent duality of meanings, not accepted as an idiom? What is the definition of a Chinese idiom? There are many definitions of Chinese idioms to date (see Xu 1997; Zeng 2011: 3–6). Some major features of Chinese idioms are listed below: 67

Jiao Liwei

Quadrasyllabic (四字格) Almost all idioms consist of four characters. (Z. Zhou 1955). Some researchers include other lengths, from disyllabic to as long as 16 syllables. For instance, 推敲 (tuiqiao, push-knock, to ponder intently) is derived from a story of a Tang dynasty monk and poet, Jia Dao, who tried to find the best word for a scenario. 各人自扫门前雪,莫管他人瓦上霜 (individualpeople-himself-sweep-door-front-snow, don’t-care-other-people-tile-top-frost, i.e. to mind one’s own business, to hoe one’s own row) is a maxim from the Song dynasty. The longest idiom might be 不是东风压倒西风,就是西风压倒东风 (if-not-east-wind-prevail-over-westwind, then-be-west-wind-prevail-over-east-wind), which is from a line in Dream of the Red Chamber.

2+2 (二二相承/骈体性) The four characters in an idiom can be parsed into two groups of two (Lü 1989; Sun 1989: 75; D. Wen 2005: 70, 291). On the phonetic level, all idioms, except 作壁上观 (read as zuo-bishangguan), 成一家言 (read as cheng-yijiayan), 居大不易 (read as ju-dabuyi) and possibly a few more, can be parsed into two groups of two, regardless of the internal grammatical structure, for example, the internal grammatical structures of 妙不可言 (miaobu­ keyan), 衣食住行 (yishizhuxing), and 一衣带水 (yiyidaishui) are 妙|不可言, 衣|食|住|行, and 一衣带|水.

Conventional and fixed (约定俗成的, 定型的) Generally, the compositional characters of an idiom cannot be changed to other characters. For example, the 伪 (wei) in 去伪存真 (quwei-cunzhen, eliminate the false and retrain the true) cannot be changed to 假 (jia) even though 伪 and 假 are synonyms (Z. Zhou 1955). Similarly, the 贫 (pin, impoverished) in 一贫如洗 (yipinruxi, in utter destitution) cannot be changed to 穷 (qiong, poor).

Integral (整体性) Regardless of the internal structure, most idioms usually function as a word in a sentence (Z. Zhou 1955; Ma 1962: 33). For example, the syntax of 愚公移山 (yugong-yishan, foolish-old man-move-mountain, freely translated as ‘where there’s a will, there’s a way’) is NP1 (愚公) + VP (V (移) + NP2 (山)), however, this idiom functions mainly as attributive, often followed by 精神 ( jingshen, spirit).

Semantic duality (表意的双层性) An idiom’s actual meaning is different from the sum of the literal meanings of each individual character; that is to say, an idiom has overtones or hidden meaning between the lines (Li and Liu 1975: 101; S. Liu 1982; Shi 1979: 12; Dictionary of Chinese Idioms Taiwan 2010). For example, 胸有成竹 (xiongyouchengzhu, chest-have-fully developed-bamboo) does not mean one has a bamboo in his chest, but ‘to have a well thought out plan, have a card up one’s sleeve’. Psychological experiments show that meaning integration does not depend on the intactness of structural information (Y. Liu et al. 2010). It is true that while a large number of idioms are used in a metaphorical (比喻), exaggerative/descriptive 68

Chinese idioms

(形容), referral (引申), or denotive (借代) way which could be easily identified in the annotation part of an idiom in any authentic Chinese dictionary of idioms, there are many idioms whose overall meaning is just the summarization of the four compositional characters, or else a very close approximation. For example, 全心全意 (quanxin-quanyi, complete-heartcomplete-intention), means ‘wholeheartedly’.

Classicality (经典性/历史性) Zhou Jian (1997) claims ‘Idioms have the features of primitive simplicity and elegance, i.e. classicality, and this characteristic of idioms was acquired in history’. Ma (1962: 15) and the Dictionary of Chinese Idioms (Taiwan) (2010) hold similar criterion. Needless to say, the idioms Zhou categorizes are accepted by all, professionals and nonprofessionals. However, as Shi (1979: 27) points out, many idioms taken from what were esteemed as classic works by later times were actually common sayings at the time. For instance, 亡羊补牢 (wangyang-bulao, lost-sheep-mend-fold, i.e. it’s never too late, better late than never) is from Stratagems of the Warring States. ‘The proverb goes, it’s not late to call a hunting dog after you have seen a hare; and it is not slow to mend the sheepfold after you have lost sheep.’

5.  Sources of Chinese idioms: from allusions, quotations, and folk sayings There are two primary sources of Chinese idioms: written literacy and folk sayings. Written literacy includes fables, historical events, and quotations (Z. Zhou 1955). The Dictionary of Chinese Idioms (Taiwan) (2010) collects idioms from three resources: 典出 (condensed from allusions and stories), 语出 (from direct quotations), and 语本 (adapted from a phrase or a line). For example: 狐假虎威 (hujia-huwei, fox-borrow-tiger-strength, i.e. to rely on powerful connection with one person to bully another person) is a fable from vol. 1 of Stratagems of Chu in Stratagems of the Warring States. 草木皆兵 (caomu-jiebing, grass-tree-all-soldier, i.e. frightened, panic-stricken) derives from a historical event in 383, the Battle of Fei River, in which the emperor of Former Qin was defeated and as he fled he thought all the bushes and trees on the mountain looked like enemy soldiers. 曲突徙薪 (qutu-xixin) and 焦头烂额 (jiaotou-lan’e) are from the line 曲突徙薪亡恩泽,焦头烂额为上客 (the man who suggested that the homeowner bend the chimney and remove the firewood from near the hearth to prevent a possible fire was not well treated, but the man who was badly hurt in fighting the fire was well treated) in the Biography of Huo Guang in the History of the Former Han. 怨天尤人 (yuantian-youren, complain-heaven-blame-others, i.e. blame everybody but oneself) is from a line in the Analects, 不怨天,不尤人 (no-complain-heaven, no-blame-others). Most researchers regard idioms that derive from literary resources highly, but Shi (1979: 30) claims the primary source of idioms is from common folk sayings. He even predicts that the next hundred years will bring a vast number of new idioms to the language (Shi 1979: 540). Recently statistics on idioms in specific books are increasingly rich. Sun (1989: 91–3) collects basic statistics of idioms distributed in the Confucian classics and major individual works of history, philosophy, and literature. For example, 101 idioms originate from Mencius, 91 from Zhuangzi, and 69 from works of Su Shi (1037–1101). The Dictionary of Chinese Idioms, Taiwan (2010) has more detailed statistics. The number of graduate and PhD theses on the subject has increased over the last 20 years. For example, the title of Zeng Xiangling’s 69

Jiao Liwei

PhD thesis is〈《诗经》成语研究〉(‘A Study of Chinese Idioms in the Book of Songs’) (Zeng 2011). Other theses examine idioms in The New Text Book of Documents, The Classic of Filial Piety, The Chronicle of Zuo, Stratagems of the Warring States, Records of the Great Historian, Book of Han, The Records of the Three Kingdoms, The Works of Yang Xiong, etc. Other books focus on the subject matter of idioms, such as the military, plant names, architecture, religion, etc. For example, a book named Buddhism Idioms has about 400 entries (R. Zhu 2006).

6.  Chinese idioms and rhetoric: figure of speech, idiomaticity, and idiomatic salience The use of idioms can serve to make speech or writing concise and vivid (Z. Zhou 1955; Ma 1962: 40–5). Shi (1979:124–5) describes 25 major figures of speech exploited in idioms. Harmony and symmetry in sound, structure, prosody, and semantics are extensively studied (Wu 1995; Z. Liu and Xing 2000, 2003; Z. Liu 2004). 非常聪明 (feichang congming, very smart) and 花钱浪费 (huaqian langfei, spend money lavishly) do not qualify as idioms, but 冰雪聪明 (bingxue-congming, ice-snow-smart, i.e. extremely intelligent) and 一掷千金 (yizhi-qianjin, one-throw-thousand-gold, i.e. to spend money recklessly or extravagantly) do qualify since 冰雪 is used metaphorically and 一掷千金 in a manner of exaggeration. Therefore, Chinese idioms are essentially figures of speech. Virtually no idioms appear in laws and regulations, nor on product manuals or specifications, because idioms cannot convey accurate meanings objectively. Eloquent lawyers, however, tend to have great command of idioms and proverbs. There is also an annoyingly large number of advertisements which abuse the use of Chinese idioms in Chinese media, in an attempt to impress their audiences. Once an idiom is used in a sentence, others are likely to follow.3 For example, after Steve Jobs, CEO of Apple Inc., died in October 2011, Al Gore, former vice president of the USA commended him in a TV interview. In the Chinese translation, there are five idioms in merely 200 characters, the first four of which convey praise for Jobs, 独一无二 (duyi-wu’er, unique, the only one of its kind), 举世无双 (jushi-wushuang, unrivalled in the whole world), 绝无 仅有 (juewu-jinyou, one and only), 非同凡响 (feitongfanxiang, uncommon).4 千变万化 (qianbian-wanhua) is a very common idiom which expresses the notion that changes are numerous. However, almost no one except lexicographers knows that this idiom is from Liezi and was later used in Zhuangzi and Records of the Grand Historian. When this idiom is mentioned, no one associates it with the above-mentioned famous works, although some people might associate it with the Monkey King. What qualifies this phrase as an idiom then? What is meant by the idiomaticity of Chinese idioms? The meaning of the above-mentioned idiom, 守株待兔 (shouzhu-daitu, waiting passively for a windfall), was enhanced by the fable 守株待兔, which describes the stupid peasant who abandoned farming to spend each day waiting for hares to crash into the same tree stump. The structure ‘千X万X’ intensifies the degree of ‘change’ which is embodied on 变 and 化 in the idiom 千变万化. The 冰雪 (ice and snow) is a metaphor to describe the smartness of a girl or woman. The meaning of ‘uniqueness’ of 独一无二 is intensified by the juxtaposition of two similar segments, 独一 (only one) and 无二 (no any other). The idiomaticity of 为所欲为 (weisuoyuwei) is embodied in three aspects, first, the two 为 are written Chinese and its corresponding colloquial word is 做; second, 所欲为 (what one wants to do) follows the structure of a Classical Chinese noun phrase; third, the phonetic structure 为所||欲为 differs from the grammatical structure 为||所欲为. Another example, 扬长 70

Chinese idioms

(yangchang), probably an erroneous form of 佯长, is a word of assonance, and it vividly describes the manner of 去 (to leave) in the idiom 扬长而去. The 而 in this idiom is a Classical Chinese function word. All the examples mentioned above show some structural, semantic, or other linguistic salience. Below are categories and types of Chinese idioms, and how their unique forms contribute to their idiomaticity.

7.  Types of Chinese idioms: five categories and 15 types Based on idiomatic salience, Chinese idioms can be divided into five categories and 15 types.

Category 1: Externally salient, for example, 狐假虎威 (hujia-huwei), 围魏救赵 (weiWei-jiuZhao), and 螳螂捕蝉,黄雀在后 (tanglang-buchan, huangque-zaihou) Type 1: originated from fables and myths, for example 狐假虎威 (hujia-huwei), 黔驴技穷 (Qianlü-jiqiong), and 八仙过海 (Baxian-guohai) Chinese idioms originated from fables are unanimously accepted by all people, professionals and common folk alike. Even though the literature of Chinese fables is not particularly rich, there are a considerable number of fables scattered in the literature of ancient philosophers, politicians, historians, and Buddhist masters. Idioms from fables in the pre-Qin period are most famous. Below are some samples with the works from which they originated. 愚公移山 (Yugong-yishan), Liezi; 揠苗助长 (yamiao-zhuzhang), Mencius; 朝三暮四 (zhaosan-musi), Zhuangzi; 刻舟求剑 (kezhou-qiujian), Lüshi Chunqiu; 滥竽充数 (lanyu-chongshu), Han Feizi; 塞翁失马 (Saiweng-shima), Huainanzi; 画蛇添足 (huashe-tianzu), Stratagems of the Warring States; 杯弓蛇影 (beigong-sheying), Fengsu Tongyi; 黔驴技穷 (Qianlü-jiqiong), Three Admonitions by Liu Zongyuan. In ancient times, culture was undeveloped and channels of information were largely obstructed, so fables became an important tool for analogical analysis. Idioms originated from fables are mainly from between the time of Mencius and Zhuangzi and the time of Stratagems of the Warring States. Later this kind of idiom became scarce, the last famous one perhaps being 黔驴技穷 of the Tang dynasty. This kind of idiom gives people a striking impression since they are associated with long and iconic stories. The four characters of an idiom cannot provide its true meaning since it comes from the associated fable. This is even truer for learners of Chinese. For example, the literal meaning of 黔驴技穷 is ‘Guizhou Province-donkey-skill-exhausted’, while its true meaning in English is ‘to exhaust one’s (limited) bag of tricks’. The English translation does not have strong negative connotations, but the idiom is highly derogatory in Chinese. The negative connotation does not come from 技穷 (tricks exhausted) but from the story of the whole fable. Textbook compilers in China highly value this kind of idiom. They have been incorporated in textbooks, and are sure to be passed down from generation to generation. Many books and dictionaries include almost all four-character fairy tales or myths as idioms, while actually only very few of them are qualified as such. Myths such as 女娲补天 (Nüwa butian), 夸父追日 (Kuafu zhuri), 后羿射日 (Houyi sheri), and 嫦娥奔月 (Chang’e benyue) are not idioms, while 精卫填海 (Jingwei-tianhai), 牛郎织女 (Niulang-Zhinü), 八仙过海 (Baxian-guohai), and a few dozen others are among the true idioms originating from fairy tales. The fairy tale of 嫦娥奔月 first appeared in Huainanzi. The author, Duke of Huainan (179–122 bc), used this myth to mock those politicians, such as Shen Buhai (385–337 bc), 71

Jiao Liwei

Shang Yang (395?–338 bc), and Han Fei (281–233 poorly. The myth goes like this:

bc)

who administered their countries

Yi pleaded and received the elixir of life from Xi Wangmu (Queen Mother of the West), but his wife, Chang E stole the elixir and used it for herself. Then she flew to the moon but once there, regretted the fact that she had no more elixir to fly back to the earth. She did not know how and where to obtain the elixir. In this sense, borrowing fire is not as good as learning how to drill wood to make fire, and getting water from your neighbor is not as good as digging a well. This fairy tale had first taken on the metaphorical meaning of a similar idiom, 舍本逐末 (sheben-zhumo, to attend to trifles/surface but neglect essentials). However, due to the fact that Chinese people worship the moon and its beauty, and Chinese writers have inexplicable zest for the sense of loneliness, the tale later evolved to be sympathetic to a lonely beauty and lost its original implication completely. Now 嫦娥奔月 is used only as a proper noun, not as a general word or an idiom.

Type 2: originated from historical events or stories, for example 负荆请罪 (fujing-qingzui), 胯下之辱 (kuaxiazhiru), and 望梅止渴 (wangmei-zhike) China has a long and rich history from which many important historical events and stories were later coined as idioms, such as 负荆请罪 (fujing-qingzui), 围魏救赵 (weiWei-jiuZhao), 纸上谈兵 (zhishangtanbing), 鸡鸣狗盗 (jiming-goudao), 四面楚歌 (simian-Chuge), 破釜沉舟 (pofu-chenzhou), 毛遂自荐 (Maosui-zijian), 乐不思蜀 (lebusiShu), 东山再起 (Dongshan-zaiqi), 草木皆兵 (caomu-jiebing), 请君入瓮 (qingjunruweng), and 东窗事发 (Dongchuang-shifa). These idioms are associated with certain historical figures and events in the minds of educated Chinese. For example, Chinese associate the idiom 负荆请罪 with two historical figures: General Lian Po (327–243 bc) and a politician named Lin Xiangru (329–259 bc), their conflicts and later reconciliation, and maybe the image of a powerful general, with twigs of chaste tree on his back, sincerely pleading guilty to a slim-figured politician. Many idioms originated from historical stories. Among historical figures, Han Xin, Xiang Yu, Zhuge Liang, Cao Cao, and Cao Zhi are associated with many idioms. Han Xin (231–196 bc) was a marshal and strategist. He suffered many humiliations before becoming famous, and his later life was full of twists and turns; eventually he came to an unexpected end. Idioms associated with Han Xin are quite popular among the Chinese people. Examples include: 胯下之辱 (kuaxiazhiru), 一饭千金 (yifan-qianjin), 暗渡陈仓 (andu-Chencang) and 多多益善 (duoduoyishan). The story of 胯下之辱 (hip-under-’s-humiliation) in Records of the Grand Historian goes like this: In the Qin dynasty a butcher in Huaiyin county insulted Han Xin by saying, ‘Even though you are tall and like to carry a sword with you, you are actually a coward.’ He publicly humiliated Han Xin and said further, ‘If you have guts, then stab me; if you don’t, crawl under my legs.’ Han Xin stared at him for a long while, then knelt down and crawled under the butcher’s legs. The crowd all laughed at Han Xin, judging him to be timid. Thus, the meaning of 胯下之辱, extreme and unbearable humiliation, is embodied in this shocking story. Some idioms stem from anecdotes about historical figures, such as 高山流水 (gaoshan-liushui), 望梅止渴 (wangmei-zhike), 三顾茅庐 (sangumaolu), 画龙点睛 (hualong-dianjing), 妙笔生花 (miaobi-shenghua), 凿壁偷光 (zaobi-touguang), 黄袍 加身 (huapao-jiashen), 屡败屡战 (lübai-lüzhan), etc. 72

Chinese idioms

Type 3: Eight-character idioms, for example: 螳螂捕蝉, 黄雀在后 (tanglang-buchan, huangque-zaihou); 近朱者赤, 近墨者黑 (jinzhuzhechi, jinmozhehei); and 翻手为云, 覆手为雨 (fanshou-weiyun, fushou-weiyu) There are scores of Chinese idioms composed of eight characters, such as 螳螂捕蝉,黄雀 在后; 鹬蚌相争,渔翁得利; 欲加之罪,何患无辞; 智者千虑,必有一失; 仁者见仁,智者 见智; and 一人得道,鸡犬升天, etc. All eight-character idioms can be parsed into two groups of four. The meanings of the two groups may be similar, for example, 无源之水,无本之木 (wuyuanzhishui, wubenzhimu, without-source-’s-water, without-root-’s-tree, i.e. things without solid foundation); or successive, for example, 一人得道,鸡犬升天 (yirendedao, jiquanshengtian, one-person-attain-Tao, chicken-dog-ascend-heaven, i.e. when a man gets to the top, all his relations and friends get there with him); or complementary, for example, 仁者见仁,智者见智 (renzhejianren, zhizhejianzhi, benevolent-man-see-benevolence, wise-man-see-wisdom, i.e. different people have different views); or contrastive, for example, 道高一尺,魔高一丈 (daogaoyichi, mogao­ yizhang, virtue-rise-one-foot, vice-rise-one-zhang (10 feet), i.e. the force of evil always manages to beat the force of law). The meaning of the whole idiom is enhanced by the semantic relations of its two parts. Sometimes only one part of the eight-character idioms is mentioned, such as 近朱者赤(近墨者黑) (jinzhuzhechi, jinmozhehei, approximate tocinnabar-person-red, approximate to-ink-person-black, i.e. different environments provide different influences), and 兼听则明(偏信则暗) (jiantingzeming, pianxinze’an, both-listenthen-enlightened, biased-heed-then-benighted, i.e. a clear head comes from an open mind). Furthermore, some eight-character idioms can be shortened into four characters, such as 见仁见智 rather than 仁者见仁,智者见智, and 翻云覆雨 rather than 翻手为云,覆手为雨 (turn up-hand-make-cloud, turn down-hand-make-rain, i.e. to play tricks). The three types of Chinese idioms described above can create a strong impression on learners, either because of their vivid semantics or because of their length and rhythm. Learners have to learn such idioms one by one, but once they are acquired, they are hard to forget.

Category 2: Structural salient or overall symmetrical, for example: 酸甜苦辣 (suantiankula), 千变万化 (qianbian-wanhua), and 根深柢固 (genshen-digu) Type 4: Juxtaposition of four characters, like 酸甜苦辣 (suantiankula), 之乎者也 (zhihuzheye), and 青红皂白 (qinghongzaobai) There are approximately two dozen idioms that juxtapose four examplars of a generic notion such as taste or color to create metaphorical meaning. Thus, the four tastes, 酸 (suan, sour), 甜 (tian, sweet), 苦 (ku, bitter), and 辣 (la, spicy) combine to form an idiom that signifies the ‘joys and sorrows of life’. Another example: 声 (sheng, sound), 色 (se, beauty), 犬 (quan, dog), and 马 (ma, horse) imply, individually, ‘going to the theatre, visiting brothels, keeping dogs, and riding horses’. Together, they form a phrase that signifies ‘sensual pleasure’. Another idiom consists of four common Classical Chinese particles: 之 (zhi, a possessive marker), 乎 (hu, an interrogative particle), 者 (zhe, a noun marker), and 也 (ye, an affirmative particle.) Together they signify ‘obscure literary jargon’. Other idioms of this kind include 悲欢离合, 博大精深, 古今中外, 生老病死, 喜怒哀乐, 青红皂白, 老弱病残, 光明正大, 是非曲直, etc. Juxtaposition of four similar, contrastive, or related words enhances their rhetorical effect. 73

Jiao Liwei

Type 5: With crossly corresponding structures, for example: 千变万化 (qianbian-wanhua), 惊天动地 (jingtian-dongdi), and 七嘴八舌 (qizui-bashe) In an idiom where the four compositional characters are represented as ABCD, characters AC or BD form a pair of synonyms, antonyms or related words, such as 变/化 in 千变万化 (thousand-transform-ten thousand-change, ever-changing), 惊/动 in 惊天动地 (shock-heavenshake-earth, earthshaking), 嘴/舌 in 七嘴八舌 (seven-mouth-eight-tongue, all talking in confusion). One pair serves as the foundation of the basic or transferred meaning of an idiom, while the other pair, for example, 千/万, 天/地, or 七/八, modifies the pair representing the basic meaning of the idiom. Take the example of 千变万化 which means ‘numerous changes’, 变 and 化 signify ‘changes’; ‘千X万X’ (literally ‘thousand X and ten thousand X’) signifies ‘a large number of’. There are about 60 idioms with the structure of ‘千X万X’, and almost all these idioms have the implication of ‘numerous in quality’ or ‘high in degree’. ‘天X地X’, combining with ‘X天X地’, is the most prolific structure for Chinese idioms. There are more than 180 idioms with this structure, and at least three dozen of them are very common. For example, 冰天雪地 (bingtian-xuedi, ice-heaven-snow-earth, i.e. bitterly cold), 欢天喜地 (huantian-xidi, merry-heaven-happy-earth, i.e. completely overjoyed), 天崩地裂 (tianbeng-dilie, heaven-fall-earth-rend, i.e. deafening (sound)), 天长地久 (tianchang-dijiu, heaven-long-earth-forever, i.e. everlasting), etc. Almost all of the idioms have implications of ‘extremely’ or ‘to a high degree’. Other structures have inscrutable connotations, for instance, ‘七X八X’ and ‘X七X八’ suggest some kind of ‘chaos’. There are a great number of fixed structures like ‘天X地X’ or ‘X七X八’ in Chinese. Zhang Weiguo (1989) describes about 70 structures. Jiao et al. (2011: 298–300) and Jiao (2012) describe 21 and 37 of the most common structures. Based on statistics of texts totaling 65 million characters, Jiao (2012) finds 8,844 four-character idioms among which 2,086 structures are based only on either the first and third characters, or the second and fourth characters. In the 8,844 idioms, 4,949 (55.92%) of them have at least one shared structure. The top 107 most common structures generate 1,291 idioms, about 14.46% of all idioms. This shows that Chinese idioms have a strong tendency to be paradigmatic. Most of the structures have semantic implications. Further observation, description, and explanation of the semantic features of these structures are highly valuable (Z. Zhou 1955). Combinations of numerals in idioms were thoroughly accounted (Han 1984). A distinguishing feature of fixed structures is that the two characters in a structure are either synonyms, near synonyms, antonyms, or closely related, and most form a word, for example, 天地 (heaven-earth), 千万 (thousand-ten thousand), 东西 (east-west), 头尾 (headtail), 云雾 (cloud-fog); meanwhile, the other two characters of an idiom are usually synonyms or antonyms, for example, 天翻地覆 (heaven-upturn-earth-overturn, extremely upheaval), 千军万马 (thousand-soldier-ten thousand-horse, a powerful army), 东奔西走 (east-dashwest-run, bustle about), 街头巷尾 (street-head-lane-tail, streets and lanes), and 腾云驾雾 (rise-cloud-ride-fog, travel through space, feel giddy). In this way, the two non-structural characters form the fundamental meaning of the idiom, while the structure modifies the fundamental meaning. For example, the fundamental meaning of 街头巷尾 derives from 街 (street) and 巷 (lane), and since the structure of 头 (head)/尾 (tail) represents ‘whole’ or ‘complete’, the whole idiom means ‘entire streets’ or ‘all streets and lanes’.

74

Chinese idioms

Type 6: With antithetical structure, for example: 根深柢固 (genshen-digu), 居安思危 (ju’an-siwei), and 应有尽有 (yingyou-jinyou) The defining characteristic of this structure is that characters A and B are either contrastive or synonymous to characters C and D. For example, in 根深柢固 (genshen-digu), 根深 literally means ‘secondary roots are deep’ and 柢固 literally means ‘primary root is stable’. The two parts are complementary, thus the meaning of the whole idiom is ‘the foundation is firm and unshakable’. Another example, in 居安思危 ( ju’an-siwei), 居安 means ‘to stay at a safe place’ and 思危 means ‘be vigilant against danger’. The two parts are contrastive and conjunctive, meaning ‘even if you stayed at a safe place, you still need to keep vigilant against danger’ when combined. Other idioms of this type include 聚精会神, 发号施令, 方兴未艾, 专心致志, 众叛亲离, 卖官鬻爵, 出类拔萃, 落井下石, and 感恩戴德 etc. There is a special form in this type of idiom: character A and C, or B and D are the same. For example, 我行我素 (woxing-wosu, I-do-I-usually do, i.e. stick to one’s own way of doing things), 顺风顺水 (shunfeng-shunshui, down-wind-down-stream, i.e. very smooth), 应有尽有 (should-have-all-have, i.e. lacking nothing, complete), and 心服口服 (xinfu-koufu, heart-convinced-mouth-convinced, i.e. be sincerely convinced), etc.

Category 3: Partially symmetric, for example: 林林总总 (linlin-zongzong), 比比皆是 (bibi-jieshi), and 未雨绸缪 (weiyu-choumou) The defining characteristic of idioms in this category is one part of the idiom is either phonetically or semantically salient.

Type 7: AABB idioms, for example: 林林总总 (linlin-zongzong), 风风雨雨 (fengfeng-yuyu), and 洋洋洒洒 (yangyang-sasa) As can be easily seen, this type of idiom consists of two parts, each of which repeats a character, for example, 林林||总总 (linlin-zongzong) and 风风||雨雨 (fengfeng-yuyu), etc. Some idioms of this type are the result of reduplication of a word AB. For example, 风风 雨雨 is the reduplication of 风雨, 浩浩荡荡 of 浩荡, 三三两两 of 三两, 朝朝暮暮 of 朝暮, 鬼鬼祟祟 of 鬼祟, etc. The basic meaning of 风风雨雨 is from 风雨 (wind-rain), which is a metaphor of ‘hardship’, and the meaning of the whole idiom is ‘many hardships’. Other idioms of this type do not result from reduplication of a word, but the combination of two words, AA and BB. For example, the meaning of 林林总总 is a combination of 林林 (numerous) and 总总 (numerous and chaos, large in quantity and various in category). This subtype of idiom also includes 浑浑噩噩, 洋洋洒洒, 熙熙攘攘, 兢兢业业, 战战兢兢, 卿卿我我, etc.

Type 8: AABC or ABCC idioms, for example: 比比皆是 (bibi-jieshi), 大名鼎鼎 (daming-dingding), and 蒸蒸日上 (zhengzheng-rishang) The defining characteristic of this type of idiom is that the first two or last two characters are the same, and are usually adjectives or adverbials. For example, in the idiom 比比皆是 (bibi-jieshi), 皆是 means ‘all are’, while 比比 means ‘one after another’, which enhances the meaning of 皆是. Another example, in 大名鼎鼎 (daming-dingding), 大名 means ‘big 75

Jiao Liwei

name’, while 鼎鼎 means ‘splendid, celebrated’, which enhances 大名. Other AABC idioms include 彬彬有礼, 沾沾自喜, and 蒸蒸日上, etc. Other ABCC idioms include 气喘吁吁, 千里迢迢, 风尘仆仆, and 生气勃勃, etc. Some ABCC idioms can be reversed to CCAB because character B is usually a noun. For example, 大名鼎鼎 and 鼎鼎大名 coexist, but ABCC is the more common form. AABC cannot be reversed as BCAA because AA is usually an adjective or adverbial which modifies the nucleus BC.

Type 9: Alliteration, assonance/rhyming, or juxtaposition of synonyms or antonyms, for example: 淋漓尽致 (linli-jinzhi), 未雨绸缪 (weiyu-choumou), 旗鼓相当 (qigu-xiangdang), and 名副其实 (mingfu-qishi) The defining characteristic of this type of idiom is that a part of the idiom is alliterative or assonant in sound, or synonymous or antonymous in meaning, so as to enhance the rhetorical result. For example, Alliteration: 踌躇满志 (chouchu), 光明磊落 (leiluo), 金碧辉煌 (huihuang) Assonance: 精神抖擞 (dousou), 五彩缤纷 (binfen), 逍遥法外 (xiaoyao) Juxtaposition of synonyms or antonyms: 德才兼备 (integrity-competence-both-possess), 不见经传 (not-seen-classic-commentary), 哭笑不得 (cry-laugh-not-appropriate), 乐极生悲 (joy-extreme-beget-sorrow). Alliteration or assonance has an immediate effect on the impression of the sound of an idiom, because the structure of Chinese syllables is relatively simple, consisting of only two parts, namely initial and final. If two consecutive syllables have half of their constituents alliterated or rhymed, the repetitive sound leaves a strong impression upon the hearer. Antithesis was a universal feature of Ancient Chinese verse, and remains prominent to this day. The coordination of two antithetical characters falls mainly into three groups. The first way is the use of antonyms, such as 朝夕 (morning-night) in 朝夕相处, and 名 (name) and 实 (fact) in 名副其实. The second way is the use of synonyms or near synonyms, such as 脍炙 (sliced meat-barbequed meat) in 脍炙人口. The third way is to use specific terms (co-hyponyms) to represent generic ones (hypernyms), such as 鸡犬 (chicken-dog) in 鸡犬 不宁, where 鸡 and 犬 are hyponyms of ‘livestock’. This subtype of idiom is categorized based on semantics, however, there are some functional words in the place of the third character of an idiom (XX[ ]X) which can serve as a marker of this type of idiom, for example: 相 (each other) in 旗鼓相当 (banner-drum) and 萍水相逢 (duckweed-water); 交 (simultaneously) in 饥寒交迫 (hungry-cold) and 心力交瘁 (heart-strength); 双 (both) in 名利双收 (fame-wealth); 两 (double) in 进退两难 (advance-retreat); 俱 (all) in 玉石俱焚 ( jade-stone); 皆 (all) in 啼笑皆非 (cry-laugh); 兼 (both) in 软硬兼施 (soft-hard); 并 (same) in 声情并茂 (voice-expression); 全 (all) in 文武全才 (literary-military); 共 (together) in 雅俗共赏 (elegant-vulgar); 齐 (together) in 金鼓齐鸣 (bell-drum). 都 is the most common adverb, meaning ‘both/all’, however, it does appear in this situation. It is worth noting that as the Chinese language evolved with time, its pronunciation probably underwent the greatest change. Some characters rhymed or alliterated in ancient times, but do not any more, and vice versa. Take 沆瀣 (hang-xie) in 沆瀣一气 for instance; they alliterated in ancient times, but do not in Modern Chinese. Another example, 身心 (shen-xin) in 身心交瘁 rhymes in Modern Chinese, but not in Ancient Chinese because 身 ended with ‘-n’ while 心 ended with ‘-m’. 76

Chinese idioms

Category 4: Partially marked/locally salient, for example: 视死如归 (shisi-rugui), 恍然大悟 (huangran-dawu), 刻不容缓 (keburonghuan), and 言而无信 (yan’erwuxin) Idioms in this category do not have salient allusions as in Category 1, salient parallel structures as in Category 2, or salient phonetic forms as in Category 3, but some locally salient words distinguish them from other idioms.

Type 10: Idioms with simile marker 犹, 若, 如, 似, for example: 记忆犹新 (jiyi-youxin), 呆若木鸡 (dairuomuji), and 如花似玉 (ruhua-siyu) Many idioms have figurative meanings, however, no idioms contain the most common simile marker 象/像 (xiang) ‘like, similar to’, which has been used to indicate similarity since the Ming dynasty (1368–). The most common simile markers in Chinese idioms include 犹 (you), 若 (ruo), 如 (ru), and 似 (si). Idioms with 如 include 如日中天 (like-sun-middle-sky, at the peak of one’s power or career), 易如反掌 (easy-as-turn-palm, piece of cake), and 视死如归 (view-death-as-returning home, face death with equanimity), etc. Idioms with 若 are not few in number, for example 呆若木鸡 (stunned-like-wood-chicken, dumbstruck), 口若悬河 (mouth-like-hanging-river, eloquent). Few idioms use 似: 繁花似锦 (flourishing-flower-like-brocade) and 归心似箭 (returning-heart-like-arrow) are the two most commonly used. Figurative meaning is vivid in this kind of idiom. If there are two simile markers in one idiom, the pattern is most likely to be ‘如X似X’, such as 如花似玉 (like-flower-like-jade), 如饥似渴 (as-hungrily-as-thirstily). ‘似X如X’ is scarce. There might only be one example, 似醉如痴 (as-drunk-as-obsessed); however, it takes a more common form in 如醉如痴. No idioms with the structures 若/似 or 如/若 have been found. Zhu Guanming (2000) describes all of the major monosyllabic simile markers in Ancient Chinese. The simile marker 犹 appeared early and disappeared early too, so there are few idioms that include it, for example 过犹不及 (excessive-as-not-enough, going beyond is as wrong as falling short), 记忆犹新 (memory-like-fresh, remember vividly), and 虽死犹生 (although-dead-like-live, live on in spirit). Since 如 and 似 coexisted for the most time, and 如 appeared earlier than 似, there are many idioms with the structure ‘如X似X’. These phenomena show that idioms were created roughly at the same pace that the Chinese language evolved in history.

Type 11: Idioms with 然, for example: 恍然大悟 (huangran-dawu), 井然有序 (jingran-youxu), and 春意盎然 (chunyi-angran) 然 can function as a suffix of certain adjectives and adverbs indicating a state of affairs. The character 然 and its preceding adjective or adverb make an idiom vibrant and lively. One form of this type is ‘X然XX’, for example, 恍然大悟 (huangran-dawu, sudden--ly-greatlyrealized, suddenly see the light) and 井然有序 (jingran-youxu, neat--ly-have-order, in good order). The third and fourth characters of this subtype of idiom are usually an adjectival phrase or verbal phrase, so the first character plus ‘然’ serves as attributive or adverbial. The other form of this type is ‘XXX然’, for example, 一目了然 (one-glanceunderstandable--ly, to understand fully at one glance) and 春意盎然 (spring-sign-abundant-ly, full of spring atmosphere). The first two characters of this subtype of idiom are most often nouns, and the second part of this subtype of idiom is a description of the nouns. 77

Jiao Liwei

Type 12: ‘X不XX’ and ‘X而XX’ idioms, for example: 刻不容缓 (keburonghuan), 言而无信 (yan’erwuxin), and 取而代之 (qu’erdaizhi) A large number of idioms belong to this category. The defining characteristic of this type of idiom is the discordance of 不 or 而 on syntactical and phonological levels. Syntactically the structure is ‘X||不XX’, ‘X||而XX’ or ‘X而X||X’; however, phonologically the structure is ‘X不||XX’ or ‘X而||XX’. This discordance makes such idioms very special, thus enhancing the effect. For example, 刻||不容缓 (moment-not-allow-delay), 马||不停蹄 (horse-not-stopgallop), 言||而无信 (speak-but-no-credit), 锲||而不舍 (carve-but-not-give up), and 取而代||之 (take-and-replace-it). Due to the special function of 不 and 而 in the formation of idioms, all expressions with the structure ‘不X而X’ tend to be classed as idioms. For example 不辞而别 (not-farewellbut-leave), 不寒而栗 (not-cold-but-shiver), 不约而同 (not-arrange-but-same), etc.

Category 5: Miscellaneous, for example: 不亦乐乎 (buyilehu), 惟利是图 (weilishitu), and 自告奋勇 (zigao-fenyong) Some idioms do not have any obvious marker, but there are unique reasons for their existence.

Type 13: From famous quotations, like 不亦乐乎 (buyilehu), 一尘不染 (yichen-buran), and 水落石出 (shuiluo-shichu) Zhou Jian (1997) claims that only idioms that originate from famous writers or famous works of literature are true idioms. To put it another way, they possess ‘classicality’. This is mostly true since in ancient times, only those famous works got much readership due to limited printing medium. The website of Chinese Idiom in Taiwan has statistics on the origins of idioms from major literary works, such as from the Confucian classics, history, philosophy, and literature. For example, 46 idioms originated from the Book of Odes, 113 from Records of the Grant Historian, 53 from Zhuangzi, and 24 from the works of Su Shi.5 Take the Analects for example. The Chinese Imperial Examinations existed for almost 1,300 years, from 607 to 1905. These examinations focus on the Confucian classics. The Analects is the first and foremost volume to be tested, so it was naturally the most popular of the classics among intellectuals and literati. Lines from the Analects were particularly prone to be coined as idioms. For example, a phrase in Confucius’s comment on Kong Wenzi, 不耻下问 (don’t feel ashamed to ask and learn from one’s subordinate) became an idiom. Confucius’s own ideology, such as 见义勇为 (to stand up for what is right), 当仁不让 (to not shirk responsibilities), and 过犹不及 (going beyond is as wrong as falling short) were revered as idioms. Even some very common expressions at the time, such as 不亦乐乎 (isn’t it a great pleasure?), had become revered idioms with a new meaning of ‘extremely, awfully’. Buddhism is a very influential religion in China, especially in secular culture. There are hundreds of idioms that originate from Buddhist works, for example 一尘不染 (one-dustnot-polluted, maintain one’s original pure character), 三生有幸 (three-incarnation-have-luck, stroke of luck), 天花乱坠 (sky-flower-chaos-drop, extravagantly colorful description). Buddhists believe that ‘color, sound, smell, taste, touch, object’ pollute people’s ‘eye, ear, nose, tongue, body, and mind’, so they are called ‘the six dusts’. 一尘不染 means ‘free from any speck of dust’, ‘spotless’ or ‘to maintain one’s original pure character’. 78

Chinese idioms

Ouyang Xiu (1007–1072) and Su Shi (1037–1101) were prominent writers in the Song dynasty, and their works were very popular at the time. Since 水落而石出者 and 水落石出 appeared in their prose An Account of the Old Toper’s Pavilion and Second Ode to The Red Cliff in 1046 and 1082 respectively, 水落石出 was coined as an idiom meaning ‘the truth is fully revealed’. It is worth noting that all the idioms mentioned above in this section have notable origins which helped them gain fixed form and popularity, but later generations, however, forgot the glorious origins of these idioms, and even transformed their meanings to be more close to literal meanings or new ones. Ordinary people do not know that 一尘不染 originates from Buddhism at all, or that 水落石出 was first written by Su Shi. They even changed the meaning of 空空如也 from ‘shallow (used in a self-deprecating way by Confucius)’ to the more literal ‘all empty’.

Type 14: With residue of Classical Chinese grammar or lexis, for example 唯利是图 (weilishitu), 不刊之论 (bukanzhilun), and 无能为力 (wunengweili) Some idioms have preserved the peculiar grammar of Classical Chinese, for example, the fronting of an object like 利 (profit) by putting it in front of the verb 图 (to seek) in the idiom 唯利是图 (only-profit-do-seek, to seek nothing but profit). Other idioms preserved flexible usage. For example, in 有口皆碑 (be universally acclaimed) 碑 is used as the verb ‘to commend’ rather than its original meaning of ‘stone tablet’. Some idioms preserved the etymological meaning of a character, for example, the original meaning of 刊 (to delete) is preserved in the idiom 不刊之论 (unalterable truth), while in Modern Chinese 刊 usually means ‘to publish’. Some characters have preserved their original meaning in idioms. For example, 别 is used mainly as the adjective ‘another’ (别的) or modal verb ‘don’t’ (不要) in Modern Chinese; however, its original meaning is ‘to differentiate, to dissect’, and it is partially preserved as ‘otherwise, differently’ in idioms such as 别出心裁 (come out with a different plan or concept), 别具一格 (have a unique or distinctive style), 别开生面 (start something new, break new ground), and 别有用心 (harbor ulterior motives). Some functional words are common in idioms, and these words help idioms maintain characteristics of Classical Chinese. For example, 莫名其妙 (its), 无能为力 (do, exert), 众望所归 (noun phrase marker), 一孔之见 (’s), 同归于尽 (at), 以卵击石 (with).

Type 15: Miscellaneous Fourteen types of idioms are elaborated above to show how their uniqueness in linguistic form or semantic connotation contributes to their idiomaticity or idiomatic salience. These types cover almost all common Chinese idioms, however, some idioms do not fit any of those patterns, so we categorize them as ‘miscellaneous’. For example: 引人入胜 (yinrenrusheng, lead-people-enter-wonderful, fascinating), 脚踏实地 (jiaota-shidi, foot-step on-solidground, conscientious and dependable, with honesty and dedication), 自告奋勇 (zigaofenyong, self-claim-effort-courage, offer to undertake a difficult or dangerous task), etc. It is obvious that some idioms belong to two or even more groups. For example, 南辕北辙 (nanyuan-beizhe, southbound-shaft-northbound-rut, i.e. to act in a way that defeats one’s purpose) belongs to both Type 1, as its original fable ‘南辕北辙’ appeared in vol. 4 of Stratagems of Wei in Stratagems of the Warring States, and Type 5, with its parallel structure ‘南X北X’. 79

Jiao Liwei

8.  Classification of Chinese idioms: by internal grammatical relationships, degree of lexicalization, and other linguistic forms Chinese idioms are counted by thousands, so it is necessary to classify them for further observation and analysis. Liu Shuxin (1990: §4.2) drew on his expertise in lexicology and divides idioms into eight types, namely those with (i) juxtaposed structures, e.g. 龙飞凤舞 (dragon-fly-phoenix-dance, elegant handwriting); (ii) modified structures, e.g. 众矢之的 (all-arrow-’s-target, a target of public criticism); (iii) complementary structures, e.g. 寄人篱下 (live-other-roof-under, depend on other people for a living); (iv) nominative structures, e.g. 拾人牙慧 (pick-others-toothdebris, plagiarize); (v) descriptive structures, e.g. 怒发冲冠 (furious-hair-spring-crown, bristle with anger); (vi) special structures, e.g. 开门见山 (open-door-see-mountain, come straight to the point), 引火烧身 (draw-fire-burn-oneself, ask for trouble); (vii) condensed structures, e.g. 种瓜得瓜 (sow-melon-get-melon, reap what one sows); and (viii) parataxis structures, e.g. 杯水车薪 (cup-water-cart-firewood, utterly inadequate). Sun (1989: 161–8) divides Chinese idioms into three types according to the degree of semantic fusion. Type 1: fused (融合性) idioms, e.g. 洛阳纸贵 (Luoyang-paper-expensive, become a bestseller) and 低三下四 (low-three-down-four, servile, humble). The meaning of this type of idiom is highly integral, and grammatical analysis is virtually impossible. Type 2: comprehensive (综合性) idioms, e.g. 按图索骥 (by-picture-seek-fine horse, do something mechanically) and 焦头烂额 (burned-head-scorched-forehead, in terrible shape, in trouble). The actual meaning of this type of idiom is extended to metaphorical interpenetration of the literal meaning. Type 3: compositional (组合性) idioms, e.g. 不期而遇 (not-expected-butmet, run into). The actual meaning of this type of idiom is basically the summarization of its constituents. This approach was developed so that five types were differentiated according to their degree of lexicalization (Qian and Yu 2003). For example, a) phrase-like idioms, e.g. 抱头鼠窜 (cover-head-mouse-flee, run helter-skelter); b) metaphorical idioms, e.g. 半途而废 (half-way-then-stop, give up halfway); c) asematic idioms, e.g. 班门弄斧 (Lu Ban-gate-play-axe, display one’s slight skill before an expert); d) agrammatical idioms, e.g. 白云苍狗 (white-cloud-black-dog, things change in an unpredictable manner); e) completely lexicalized idioms, e.g. 别风淮雨 (‘other’-wind-‘Huai’-rain, 别 and 淮 are ‘misspellings’ of 列 and 淫 which mean ‘fierce’ and ‘heavy’ respectively. The meaning of this idiom is to be ‘full of wrongly written characters’.) Tsou (2012) analyzes the internal linguistic features of idioms and enumerated eight types: a) hypernymy: 三五成群 (three-five-become-crowd, in small groups); b) classical language usage: 三年五载 (three-year-five-year, in a few years); c) culture bound: 三生有幸 (threeincarnation-have-luck, forever indebted); d) locus classicus: 朝三暮四 (morning-three-eveningfour, indecision); e) synonymy 说三道四 (say-three-call-four, mumbling insignificant things); f) word morphology 三长两短 (three-long-two-short, accident); g) homonymy (phonetic/ semantic replication or rhyme) 三三五五 (three-three-five-five, in small groups); h) antonymy 朝三暮四 (morning-three-evening-four, indecision). Most of the above classifications take into account linguistic structures; however, they are not exhaustive. The linguistic forms of idioms occur in obvious patterns. More than half of all idioms are either symmetric in semantics or harmonic in sound (Z. Liu 2004; Z. Liu and Xing 2000, 2003; C. Zhou 1988). Psychological experiments prove this. Evidence from eye-movements and event-related potentials (EPRs) show that gaze duration, total reading time, and the number of fixations of the third and fourth characters of an idiom are much shorter or less than those of the first two characters. The number of fixations of the last two 80

Chinese idioms

characters is only about 40% of that of the first two characters (Wang Yiwen et al. 2012). This proves that idioms are highly integral in semantics. Focus is not evenly distributed on the four constituent characters. This experiment also shows that gaze duration and total reading time of the second character are much longer that those of the other three characters. It shows that comprehension of an idiom is complete by the end of the second character. In other words, the first two characters lay the foundation or provide a strong hint as to the essential meaning of the whole. Supposing an educated native speaker of Chinese saw or heard 崇山 (towering-mountain), it is natural to expect the following part will be 峻岭 (lofty-ridge). The same for 夫唱 (husband-sing) and 妇随 (wife-follow), which in combination mean ‘wife being her husband’s echo’; 落井 (fall into-well) and 下石 (throw-stone) which in combination mean ‘to hit someone when he is down’; 天经 (heaven-normal) and 地义 (earth-righteousness) which in combination mean ‘entirely justified’. In rare cases where it is not possible to successfully guess the complete form of an idiom even after the second character is known, the third character will disclose everything. Statistics in the above paper show that the number of fixations of the fourth character is only about one sixth to one fifth of that of the first or the second character. During the first season of 中国成语大会 (China Idioms Conference) in 2014, the competitors usually did not need a hint of more than two characters to guess an idiom successfully. Any prerequisites for an idiom are not at all related to meaning. For instance, 惊天动地 (shock-heaven-shake-earth, earthshaking) is an idiom, so is 鸡毛蒜皮 (chicken-feathergarlic-skin, trivial); 凤毛麟角 (phoenix-feather-unicorn-horn) describes someone or something (precious and) rare, while 獐头鼠目 (river deer-head-mouse-eye) describes a repulsively ugly person. What matters most for an idiom is its rhetorical effect, which comes mainly from the salience reflected in linguistic forms. Therefore, classification of Chinese idioms should be based mainly on linguistic form, especially fixed structures or patterns.

9.  Evolution of idioms The evolution of idioms will be discussed here from four aspects, namely the shift of commentary connotation between complimentary and derogatory, competition among synonymous idioms, emergence of new idioms, and the spread of Chinese idioms to neighboring countries.

9.1.  Shift of commentary connotation between complimentary and derogatory Many idioms have commentary connotations. For example, 脱颖而出 (tuoying’erchu, talent revealing itself) has a complimentary connotation, whereas 别有用心 (bieyouyongxin, have an ulterior motive) has a derogatory connotation, and others are neutral, for example, 众所周知 (zhongsuozhouzhi, as everyone knows). Psychological experiments show that if a complimentary idiom is shown in the upper part of a screen, the mental processing time of the idiom is faster than when the idiom is shown in the lower part of the screen, and vice versa (Zhang Jijia et al. 2011). Several dozen common idioms have changed register over the time; however, the number is very small compared to the total number of idioms. Shi (1979: 222) points out that the main trend in the change of register was from complimentary to derogatory. The reason for this is that later generations were inclined to comprehend an idiom from its literal meaning without consulting its original context. For example, the origin of 难兄难弟 (nanxiong-nandi) is from Chapter 1 of Morality of Shishuo Xinyu (A New Account of the Tales of the World). 81

Jiao Liwei

A person commented on two brothers and said ‘Yuanfang 难为兄, and Jifang 难为弟’ which means the two brothers are equally excellent. The 难 here is obviously with the yangping tone (light level) and means ‘hard to be’. However, later generations take the character 难 (nan4, misfortune) at face value, and so the idiom is used to mean ‘fellow sufferers, birds of a kind’. Other examples include 浑浑噩噩, 老气横秋, 天花乱坠, 尔虞我诈, and 空空如也, etc.

9.2.  Competition among synonymous idioms Most idioms have close synonyms. If viewed from a historical perspective, this phenomenon is more obvious. Below are three pairs and one group of synonyms. In Modern Chinese, 千欢万喜 (qianhuan-wanxi, overjoyed) is not sayable though it could be well understood. However, in the PKUCCL (Peking University Center for Chinese Linguistics) corpus as of June 2013, there were 36 cases of 千欢万喜, and 74 cases of 欢天喜地 (huantian-xidi) in Ming dynasty documents. But there were only two cases of 千欢万喜 and 106 cases of 欢天喜地 in Qing dynasty documents. In the Ming dynasty short story series San Yan Er Pai (《三言二拍》), these two idioms coexisted. So this pair is an instance of what once coexisted but where one later prevailed and the other was eliminated. 班门弄斧 (Banmen-nongfu) means ‘to display one’s slight skill before an expert’ since Lu Ban was revered as the Grand Master of carpenters. This idiom originates from Liu Zongyuan’s line and Ouyang Xiu used it in its fixed form for the first time. However, there is a similar idiom 布鼓雷门 (bugu-Leimen, cloth-drum-thunder-gate, don’t hold a drum to pass by Gate Lei) which appeared in a much earlier document, from Biography of Wang Zun in the Book of Han. 雷门 is in Shaoxing, Zhejiang province. It is said that there was a giant drum on the city gate and when it was struck, the sound traveled as far as Luo Yang in Henan Province, 1,000km away. It seems that the figurative meaning of 布鼓雷门 was typical enough to qualify as an idiom. However, there was only one usage recorded in Chinese documents, which was in a drama by Wu Changling in the Yuan dynasty. Therefore, 班门弄斧 and 布鼓雷门 have the closest meaning but one was never the other’s match in prevalence. The Former Ode to the Red Cliff《前赤壁赋》, written by Su Shi (1037–1101), is one of the most famous pieces of prose in the literary history of China and all intellectuals could cite some parts of it since it includes many famous lines. One line is 寄蜉蝣于天地, 渺沧 海之一粟. 蜉蝣 means ‘mayfly’, which is an insect with a very short life span, ranging from several minutes to several days. 蜉蝣天地 (fuyou tiandi) is a perfect expression of ‘life is short’. However, there were only two usages of this idiom in ancient times, and no usage in Modern Chinese, while there emerged an idiom, 沧海一粟 (canghai-yisu) from the phrase directly next to 蜉蝣天地. 粟 means ‘millet’ which is clearly very small in comparison with the deep sea (沧海). It is a vivid expression to describe ‘how small human beings are compared to the universe’. We can see that even though two possible idioms were placed adjacent to one another, their fate was totally different. One became an idiom, and the other was kept in its original form. Besides the three pairs of competitive idioms mentioned above, we want to discuss a group of competitive idioms. There are a quite large number of idioms that mean ‘the difference/gap is large’ as seen from Table 5.1. Data is from PKU Chinese corpora. There are three idioms in the top five on the list of most common idioms in Ancient and Modern Chinese: 迥然不同, 大相径庭, and 千差万别. Two common idioms in Ancient Chinese, 天渊之别 and 不啻天渊, were replaced by 天壤之别 and 截然不同 in Modern 82

Chinese idioms Table 5.1  Competitive idioms that mean ‘the difference/gap is large’ Character to character translation 天冠地屦 天悬地隔 天差地远 天差地别 不啻天渊 天壤之判 天壤之别 天渊之别 霄壤之别 云泥之别 迥然不同 截然不同 千差万别 大相径庭

Number of cases in Ancient Chinese

heaven-hat-earth-shoe heaven-hang-earth-separate heaven-different-earth-apart heaven-different-earth-distinct not-different-heaven-lake heaven-earth-’s-distinguish heaven-earth-’s-distinct heaven-lake-’s-distinct sky-earth-’s-distinct cloud-mud-’s-distinct far--like-not-same clear cut-like-not-same thousand-different-ten thousand-distinct great-mutual-path-yard

 2  4  6  1 19  0 11 23  7  2 23  9 13 15

Number of cases in Modern Chinese 0 1 13 16 0 1 241 47 14 5 458 1,019 419 455

Chinese. The change was probably not because the character 渊 (swirling deep water) was rare and complex, since it survived in idioms such as 临渊羡鱼 and 为渊驱鱼. The structure 天/地 was probably popular at the time, so the contrast between 天 and 渊 was weakened. 啻 is more complex, so the elimination of 不啻天渊 was the result of natural selection. Although 天渊之别 was no longer common, a similar idiom 天壤之别 replaced it, so its sense is somewhat compensated. The 然 in 截然不同 helped the whole idiom deserve its idiomaticity. In brief, if an idiom, such as 蜉蝣天地, is defeated in competition against synonymous idioms in history, it is very likely that it cannot stage a comeback. Even some common Chinese idioms are facing challenges from newly emerged proverbs, for example, 趋炎附势 (quyan-fushi, to curry favor with powerful people) has been seriously challenged by 抱大腿 (bao datui, hug-big-thigh) in spoken Chinese.

9.3.  Emergence of new idioms New idioms emerge very slowly, mainly for the following two reasons. First, although new concepts emerge quickly due to the rapid development of science and technology, the need for new words focuses mainly on nouns. Chinese idioms are a kind of rhetorical device, and they are mainly applied to express Chinese people’s reflection of objects, not the objects themselves, so the emergence of new idioms lags far behind the emergence of proper nouns. For example, if one compares all the entries of the newest and the oldest editions of Cihai (辞海, lit. ‘sea of words’), one will find that there are many more entries starting with 电 (electricity or electronic) in newer editions; however, until now all the uses of 电 in idioms mean ‘lightning’, for example, in 电闪雷鸣, 风驰电掣, and 电光火石, not a single instance points to ‘electricity or electronic’. Second, there are already a large number of idioms, so demand for new idioms is weak. Given the limited memory of human beings, 8,000 general words may be able to satisfy 83

Jiao Liwei

ordinary people’s linguistic needs (词涯八千, Cheng 1998). As for the demand for idioms, perhaps a few hundred idioms would suffice. Now many idiomatic dictionaries are apt to collect more than 10,000 idioms, so these idioms have already been sufficient to satisfy the needs of ordinary people. But new idioms do emerge, albeit very slowly. Works of literature and film by famous writers and directors are the main inspiration behind new idioms. For example, 华山论剑 (Huashan-lunjian, Mount-Hua-compete-sword, discussion lead by top attendees) emerged from Louis Cha Jin Yong’s martial arts novel The Legend of the Condor Heroes (《射雕 英雄传》, 1957–9). This idiom is now widely used in summit meetings in many fields such as business, culture, and technology. 尘埃落定 (chen’ai-luoding, dust-fall-settled, to end up with) comes from the writer 阿来’s novel of the same name. 与狼共舞 (yulang-gongwu, with-wolf-together-dance, to venture) is from the famous film Dances with Wolves directed by Kevin Costner. 横空出世 (hengkong-chushi, across-sky-born, emerge out of the blue) is from a line in Mao Zedong’s poem Niannujiao Kunlun (念奴娇•昆仑) in 1935. There are many restrictions to the emergence of new idioms. Basically two conditions must be satisfied. The first is that the phrase must possess characteristics of idioms, and the second is there must be a large number of uses in a short period of time. In 2007 the incident of Zhou Zhenglong, who claimed to have taken a picture of a South China tiger but was later imprisoned for two years for swindling, was hyped by Chinese media, and many people predicted that 正龙拍虎 (Zhenglong-photograph-tiger) would become an idiom; however, it proved not to pass the test of time. 人艰不拆 (renjian-buchai, people/life-harddon’t-expose, i.e. life is already hard, don’t spoil the fun) appeared on the Internet from around 2013, and its suitability as an idiom is currently being tested.

9.4.  Borrowing and spread of Chinese idioms Idioms do not refer to objects, but are a way to creatively describe them. Simply put, idioms are a rhetorical device. Languages borrow far more ordinary words than idiomatic phrases from other languages. Chinese, for example, has borrowed hundreds of words from English and Japanese, but only a few of them are idioms. The few idioms borrowed from other languages include 以牙还牙, 以眼还眼 (yiyahuanya, yiyanhuanyan, a tooth for a tooth, an eye for an eye) and 象牙之塔 (xiangyazhita, ivory tower) from English, 火中取栗 (huozhongquli, fire-inside-pull-chestnut, i.e. be a cat’s paw, ‘tirer les marrons du feu’) from French, 一期一会 (yiqi-yihui, one-lifetime-one-chance, i.e. once in a lifetime) from Japanese. In return, English has borrowed about a few dozen words from Chinese, but none of them are idioms. Since ancient China had great influence over its neighboring countries in politics and culture throughout history, hundreds or perhaps even more than a thousand Chinese idioms were borrowed into Vietnamese, Japanese, and Korean, respectively. Vietnamese has 4,174 idioms, among them 1,760 entries are related to Chinese idioms, and 301 Chinese idioms were borrowed intact, for example ‘danh chinh ngon thuan’ (名正言顺, mingzheng-yanshun, name-proper-word-justified, i.e. perfectly justifiable) (Cai 2011). Japanese idioms are called ‘four-character compounds’, and the idiomatic youjijukugo were borrowed from Chinese, for example 単刀直入 たんとうちょくにゅう (单刀直入, dandao-zhiru, short hand-knifedirect-stab, i.e. come straight to the point). According to Moon (2005), Korean has 1,271 idioms borrowed from Chinese intact, for example 격화소양 (隔靴搔痒, gexue-saoyang, outside-boot-scratch-itch, i.e. attempt an ineffective solution). W. S.-Y. Wang (1991) studied language prefabs and habitual thought of the peoples in East Asia. 84

Chinese idioms

10.  Application of Chinese idioms Since Chinese idioms are such an important aspect of the Chinese language, they are naturally applied in many walks of lives. Below are some of their major applications.

10.1.  Idiom games The most common language game with idioms is 成语接龙 (idiom relay). The rule is a player gives an idiom, and the next player must come up with an idiom whose first character is same as the last character in the previous idiom. Then the first character of the third idiom must be the same as the last character of the second idiom, and so on. If the rules of this game are strictly enforced, the overlapping characters must be exactly the same; but if the rules are lax, the overlapping characters can be homonyms. The longest recorded idiom relay contains 1,792 idioms, starting from 胸有成竹 (xiongyouchengzhu, chest-have-conceivedbamboo, i.e. to have a well thought-out strategy) followed by 竹报平安 (zhubaoping’an, family letter-assure-safe and sound)  .  .  .  and ending with 成竹在胸 (chengzhu-zaixiong, conceived-bamboo-in-chest). However, relays with more idioms appear time and again. This game can be played without noting down the idioms and many people can join in, so it is very popular in China. Other idiom games can be found in Mo (2001: 242–55).

10.2.  Mottos of Chinese colleges When Chinese colleges were first established from the end of the nineteenth century, it happened that the status of idioms in Chinese language was rising at the same time. Therefore the mottos of many of the oldest colleges in China contain idioms which are also naturally inspiring. For example, one of the earliest mottos of Peking University was 兼容并包 (all-embrace-both-include, all-embracing); the motto of Tsinghua University contains 自强 不息 (self-improve-not-cease, to strive unremittingly); Nankai University 日新月异 (daynew-month-different, to make rapid progress); Tianjin University (originally ‘Imperial Tientsin University/Peiyang University’, the oldest university in China, founded in 1895) 实事求是 (to seek truth from facts). Many colleges chose idioms which are in accordance with the missions of their schools. For example, many teachers’ colleges set 为人师表 (to serve as a model) as their mottos. The motto of the Ocean University of China contains 海纳百川 (ocean-admit-hundred-river, the sea refuses no river), China University of Geosciences 艰苦朴素 (to live simply and work hard), China Agricultural University 博大精深 (abundantbig-refined-profound, broad and deep), China Youth University of Political Sciences 朝气蓬勃 (to be full of youthful energy). Other unique mottos include 德才兼备 (virtue-talent-bothpossess) of Beihang University and 饮水思源 (drink-water-think-source, do not forget where your happiness comes from) of Shanghai Jiao Tong University.

10.3.  Media titles Idioms are widely used in media such as titles for novels, films, TV series, albums, bands, and blogs. Not to mention those idioms with positive connotations such as 皆大欢喜 (jiedahuanxi, all-very-happy, i.e. to everyone’s delight and satisfaction), which has been used as the title of a Chinese movie, TV series, album, Pingju (a form of local Chinese opera), Chinese translation of Shakespeare’s play ‘As you like it’ and one of its film adaptations. Even the idiom 对牛弹琴 (duiniu-tanqin, to-cow-play-thither, i.e. to cast pearls before swine), 85

Jiao Liwei

which has a negative connotation, has been used as the title of two albums and a novel on the Internet. The second most common Chinese idiom 千方百计 (by hook or by crook) has been used as the title of a song, an album, and a TV series. From 1905 to 2000, there were 6,273 domestic and imported films shown in China, and among them 128 titles were idioms. For example, 一鸣惊人 (set the world on fire, 1954), 龙凤呈祥 (excellent good fortune, 1964), 归心似箭 (wish to wing one’s way home, 1979), 顾此失彼 (unable to manage two or more things at once, 1981), 时来运转 (have a change of luck, every dog has his day, 1985), 大惊小怪 (make a fuss about nothing, 2000). It may seem that 128 titles is only small fraction of the total number of titles over nearly 100 years; however, there were only about 30 titles in idioms before 1981, that is to say, from 1981 to 2000 there were nearly 100 movie titles that were Chinese idioms, five titles per year. The sensational martial arts movie 卧虎藏龙 (Crouching Tiger Hidden Dragon, 2000) and the title of its original novel were both Chinese idioms.

10.4.  Business and economic worlds In the business world Chinese idioms are most extensively used in advertisements in the form of homonyms. For example, a cough medicine company used 咳不容缓 which literally means ‘cough-not-allow-delay’ and is from 刻不容缓 (keburonghuan) which means ‘secondnot-allow-delay, be of great urgency’. 咳 and 刻 are homonyms of ‘ke’ but with different tones. Since these kinds of advertisements are so common on TV and in other media, parents and educators worry that this language phenomenon will have a negative impact on students’ learning. This had aroused such heated debate that some local governments in China passed regulations prohibiting the alteration of fixed forms of idioms, with a fine of as high as tens of thousands of RMB.6 Chinese idioms are used not only in business, but also in economics. There are about 315 Chinese idioms in the translation of Paul Samuelson’s Economy (18th ed.), occurring a total of 450 times. A Japanese economist, Atsushi Kajii published a book named『故事成语でわかる経済学のキーワード』which uses 30 idioms to exemplify 30 economic phenomena.

10.5.  Educational purposes Chinese idioms are a refined part of the language. They are not easy but quite necessary to grasp for all native speakers and learners of Chinese, including children. The very first Chinese idiom that children in mainland China learn is probably 五颜六色 (wuyan-liuse, five-color-six-color, i.e. be riotous with color). There are about 80 idioms in one of the most popular series of children’s books Les P’Tites Poules. In 2001, Sun Wenhua wrote a concise book about Chinese history and culture. The book was written totally in four-character phrases, more than 70% of which are Chinese idioms. According to a news report of October 2012, 27.8 billion RMB would be invested in a project named 中华成语文化园 (cultural park of Chinese idioms) at Handan, Heibei Province.7 Since 2001, a panel of economists, columnists, and intellectuals in South Korea have voted on one idiom each year as the idiom of the year. In 2001 it was 五里雾中 (utterly mystified) and in 2013 it was 倒行逆施 (go against the trend of the times). In April 2012, the Jiangsu Province Research Association of Chinese Idioms was founded.

86

Chinese idioms

Notes 1 An introduction and basic statistics of those Chinese treebanks are available at http://www.cis.upenn. edu/~chinese/ctb.html, http://cslt.riit.tsinghua.edu.cn/~qzhou/chs/Resources.htm, http://ccl.pku.edu. cn:8080/WebTreebank/WebTreebank_Readme.html, http://rocling.iis.sinica.edu.tw/CKIP/treebank.htm. 2 http://zh.wikipedia.org/zh-tw/三隻小豬. 3 But some linguists advocate not using idioms excessively. For instance, a prominent Chinese linguist Lü Shuxiang once made the joke: 成语词典害死人 (Dictionaries of Chinese idioms kill writers). See China Press and Publishing Journal 5, 2009: 8 http://www.chinaxwcb.com/xwcbpaper/html/20096/05/content_52748.htm. 4 See http://cn.wsj.com/gb/20111021/BOG144431.asp. 5 These statistics are different from those by Sun (1989). 6 See http://epaper.bjnews.com.cn/html/2011-03/18/content_212028.htm?div=-1 and http://www.jxnews. com.cn/jxrb/system/2010/12/07/011536612.shtml (Article 18). 7 http://news.xinhuanet.com/local/2012-10/06/c_113283723.htm.

References Cai, Xinjiao (2011)〈越汉成语对比研究〉(‘A Comparative Study of Vietnamese and Chinese Idioms’), Master Thesis, East China Normal University. Chen, Wen (2011)〈《现代汉语词典》中定型四字格的语义分析〉(‘Semantic Analysis of the Fixed Quadrasyllabics in the Contemporary Chinese Dictionary’), Master Thesis, Nanjing Normal University. Cheng, Chin-Chuan (1998)〈从计量理解语言认知〉(‘Quantification for Understanding Language Cognition’), in Benjamin K. T’sou, Tom B. Y. Lai, Samuel W. K. Chan, and William S.-Y. Wang (eds)《汉语计量与计算研究》(Quantitative and Computational Studies on the Chinese Language), Hong Kong: City University of Hong Kong, 15–30. Fang, Huisheng (1943)〈成语和成语的运用〉(‘Chinese Idioms and their Use’), Guowen Zazhi (Journal of National Language) 2(3): 21–6. Feng, Shengli (1998)〈论汉语的自然音步〉(‘On Natural Foot in Modern Chinese’), Zhongguo Yuwen 《中国语文》(Studies of the Chinese Language) 1: 40–7. Han, Chenqi (1984)〈论汉语成语中的数词〉(‘On Numerals in Chinese Idioms’), Jiaoxue yu Yanjiu (Teaching and Research) 3: 64–9. Jiao, Liwei, Cornelius C. Kubler, and Zhang Weiguo (2011) 500 Common Chinese Idioms: A Frequency Dictionary, London and New York: Routledge. Jiao, Liwei (2012)〈从固定结构的角度增进对汉语成语的理解和使用〉(‘Comprehension and Production of Chinese Idioms: A Pattern Approach’), in Xu Weimin and He Wenchao (eds)《国际汉语教学的 理念与教学实践研究》(Research on Concepts and Practice of Developing Chinese Language Teaching Materials), Hangzhou: Zhejiang University Press, 88–93. Li, Xingjian and Liu Shuxin (1975)《怎样使用词语: 汉语词汇学习》(How to Use Words: The Learning of Chinese Words), Tianjin: Tianjin People’s Press. Liu, Changzheng and Qin Peng (2007)〈基于中国主流报纸动态流通语料库(DCC)的成语使用情况 调查〉(‘An Investigation of Idioms in Chinese Newspapers Based on the Dynamic Circulation Corpus (DCC)’), Yuyan Wenzi Yingyong (Applied Linguistics) 3: 78–86. Liu, Jiexiu (1985)《成语》(Chinese Idioms), Beijing: The Commercial Press. Liu, Shuxin (1982)〈固定语及其类别〉(‘Fixed Expressions and Their Categories’), in Liu Shuxin (ed) Yuyan Yanjiu Luncong (Journal of Linguistic Studies) 2: 156–79. Liu, Shuxin (1990)《汉语描写词汇学》(Descriptive Lexicology of Chinese), Beijing: The Commercial Press. Liu, Youyi, Li Ping, Shu Hua, Zhang Qirui, and Chen Lang (2010) ‘Structure and Meaning in Chinese: An ERP Study of Idioms’, Journal of Neurolinguistics 23: 615–30. Liu, Zhenqian (2004)〈汉语四字格成语平仄搭配的对称性与认知〉(‘Symmetrical Features of Ping-ze Arrangement in Four-character Chinese Idioms and Cognition’), Shandong Daxue Xuebao (Journal of Shandong University) 4: 44–51. Liu, Zhenqian and Xing Meiping (2000)〈汉语四字格成语语义结构的对称性与认知〉(‘The Semantic Symmetrical Features of Four-character Idioms in Chinese and Their Effects on Cognition’), Shijie Hanyu Jiaoxue (Chinese Teaching in the World) 1: 77–81.

87

Jiao Liwei Liu, Zhenqian and Xing Meiping (2003)〈四字格成语的音韵对称与认知〉(‘Symmetry in Phonological Patterns in Four-character Idioms and Cognition’), Yuyan Jiaoxue yu Yanjiu (Language Teaching and Linguistics Studies) 3: 48–57. Lu, Zhiwei (1956)〈汉语的并立四字格〉(Chinese Quadrasyllabic Expressions), Yuyan Yanjiu (Studies in Language and Linguistics) 1: 45–82. Lü, Shuxiang (1989)〈《中国俗语大辞典》序〉(‘Preface to A Comprehensive Dictionary of Chinese Proverbs’), Shanghai: Shanghai Lexicographical Publishing House. Ma, Guofan (1962)《成语简论》(A Brief Study of Chinese Idioms), Shenyang: Liaoning People’s Press. Ministry of Education, R. O. C. (2010)《成语典》(Dictionary of Chinese Idioms), online dictionary, http://dict.idioms.moe.edu.tw/cydic/index.htm. Mo, Pengling (2001)《汉语成语与汉文化》(Chinese Idioms and Culture), Nanjing: Jiangsu Education Press. Moon, Mi-jin (2005)〈从语言结构看汉语成语对韩语与其成语的影响〉(‘A View of the Influence of Chinese Idioms on Korean Language and Korean Idioms from Language Contact’), Journal of Nanjing University of Aeronautics and Astronautics (Social Sciences), 7(4): 59–62. Qian, Yun and Yu Ge (2003)〈现代汉语四字格成语的词汇化研究〉(‘A Research in the Lexicalization of Four-character Idioms in Modern Chinese’), Yuyan Kexue (Linguistic Sciences) 2(6): 86–96. Shi, Shi (1979)《汉语成语研究》(Studies of Chinese Idioms), Chengdu: Sichuan People’s Press. Sun, Weizhang (1989)《汉语熟语学》(Chinese Phraseology), Changchun: Jilin Education Press. Tsou, Benjamin K. (2012) ‘Idiomaticity and Classical Traditions in Some East Asian Languages’, in Proceeding of the 26th Pacific Asia Conference on Language, Information and Computation, Bali, 8–10 November 2012, 39–55. Wang, Hui (2009)〈词义, 词长, 词频〉(‘Polysemous Words: Meaning, Length, and Frequency), Zhongguo Yuwen《中国语文》(Studies of the Chinese Language) 2: 120–30. Wang, Li (1987)〈序言〉(‘Preface’), in Wang Tao et al. (eds)《中国成语大辞典》(A Comprehensive Dictionary of Chinese Idioms), Shanghai: Shanghai Lexicographical Publishing House. Wang, Tao et al. (eds) (1988)《中国成语大辞典》(A Comprehensive Dictionary of Chinese Idioms), Shanghai: Shanghai Lexicographical Publishing House. Wang, William S.-Y. (1991) ‘Language Prefabs and Habitual Thought’, in William S-Y. Wang Explorations in Language, Taipei: Pyramid Press, 397–412. Wang, Yiwen, Zheng Yuwei, Shen Deli, Cui Lei, Yan Guoli (2012)〈解读成语中的他人心理与互动 心理: 来自眼动和 EPR 的证据〉(Reading Another’s Mind and Reading the Interactive Mind from Chinese Idioms: Evidences from Eye-movements and ERPs), Acta Psychologica Sinica 44(1): 100–11. Wang, Yuelong and Ji Donghong (2009)〈汉语树库综述〉(A Review of Chinese Treebanks), Dangdai Yuyanxue (Contemporary Linguistics) 11(1): 47–55. Wen, Duanzheng (2005)《汉语语汇学》(Chinese Phraseology), Beijing: The Commercial Press. Wen, Shuobin and Wen Duanzheng (2009)《汉语语汇学研究》(A Study of the History of Chinese Phraseology), Beijing: The Commercial Press. Wu, Huiying (1995)〈四字格中的结构美〉(‘Structural Harmony in Quadrasyllabics’), Xiuci Xuexi (Rhetoric Learning) 1: 21–2. Xiao, Zhusheng (1987)〈四言成语的两项小统计〉(Two Statistics of Four-character Idioms), Zhongguo Yuwen Tiandi (Chinese Language World ) 5: 37–8. Xu, Yaomin (1997)〈成语的划界, 定型和释义问题〉(Chinese Idioms: Extension, Formation, and Annotation), Zhongguo Yuwen《中国语文》(Studies of the Chinese Language) 1: 11–17. Zeng, Xiangling (2011)〈《诗经》成语研究〉(‘A Study of Chinese Idioms in the Book of Songs’), Doctoral thesis, National Taiwan Normal University. Zhang, Jie (2007)〈《现代汉语词典》与《现代汉语规范词典》成语比较研究〉(‘A Comparative Study of The Contemporary Chinese Dictionary and Standard Dictionary of Modern Chinese’), Master Thesis, Shandong Normal University. Zhang, Jijia, He Benxuan, and Chen Xuqian (2011)〈‘上下意象图示’对汉语成语感情色彩加工的 影响〉(‘The Up-down Image Schema Affects the Emotion Process of Chinese Idioms’), Xinlixue Xintan (Psychological Exploration) 31(2): 144–9. Zhang, Tiewen (1999)〈成语的数量及产生年代〉(Number of Idioms and the Time They Originated), Yuwen Jianshe (Language Planning) 5: 23–4. Zhang, Weiguo (1989)《四字语型及其应用》(Quadrasyllabic Patterns and Their Application), Beijing: China Logistics Publishing House.

88

Chinese idioms Zhou, Changji (1988)〈汉语成语的音律美〉(‘Harmonic Prosody of Chinese Idioms’), Xiucixue Yanjiu (Rhetoric Studies) 4: 311–19. Zhou, Jian (1997)〈论成语的经典性〉(‘On the Classicality of Chinese Idioms’), Naikai Xuebao (Journal of Nankai University) 2: 29–35, 51. Zhou, Jian (2004)〈四字组合论〉(On Four-character Combinations), Hanyu Xuebao (Chinese Linguistics) 1: 40–6. Zhou, Jinguo and Shi Min (2012)〈《红楼梦》成语的文化透视〉(‘Cultural Observation of the Idioms in the Dream of the Red Chamber), Changzhou Gongxueyuan Xuebao (Journal of Changzhou Institute of Technology, Social Sciences Edition) 6: 67–71. Zhou, Zumo (1955)〈谈‘成语’〉(‘On Chinese Idioms’), Yuwen Xuexi (Chinese Learning) 1: 31–4. Zhu, Guanming (2000)〈比喻词的历史更替〉(‘Evolution of Simile Words’), Xiuci Xuexi (Rhetoric Learning) 5–6: 74–5. Zhu, Ruiwen (2006)《佛教成语》(Buddhism Idioms), Shanghai: Publishing House of the Unabridged Chinese Dictionary.

89

6 Chinese language and national identity Andrew Simpson University of Southern California, USA

1.  Nations, nationalism, national identity, and Chinese This chapter provides an overview of how Chinese language has affected, and continues to affect, the development and maintenance of national identity, drawing on a body of works which have investigated related topics. In order to embark on this task, it is first necessary to discuss how the important terms ‘nation’, ‘nationalism’, and ‘national identity’ may be applied in the context of China both synchronically and over the course of the last 150 years, and also what the language label ‘Chinese’ is regularly used to refer to. It will be seen that the relation of ‘Chinese’ to ‘national identity’ in fact varies depending on the perspective of the Chinese nation that is adopted (ethnic vs. political), and that different emphases have been placed on the role of Chinese in the development of national identity at different times in modern history. Having discussed these key terms in section 1, section 2 describes the place of Chinese in the modernization of China and the development of the People’s Republic of China through to the end of the twentieth century, examining how the growth of Mandarin Chinese has affected the emergence of national identity among different Han Chinese groups. Section 3 turns to consider language and national identity issues specifically among the nonHan minorities in the People’s Republic of China, and section 4 looks at changing attitudes towards language, national identity, and nationalism in very recent times.

1.1.  Nations, nation-states, and nationalisms The term ‘nation’ is often used in two rather different ways in discussions relating to national identity. A majority of analysts of nationalism hold that the development of nations is a comparatively recent phenomenon, resulting from a change in the way that populations have organized themselves in modern, industrial times. Nations are prototypically suggested to have a number of properties and consist of populations which are made up of a single ethnic group with a common ancestry, history, culture, and language, living in a distinct territory, where the national population benefits from a uniform mode of political organization, citizenship, and equal rights. A second way that the term ‘nation’ is often applied is in reference to all politically independent states, regardless of whether such territories contain populations

90

Chinese language and national identity

connected by a common ancestry, language, history, and culture. In the literature focusing on nationalism, such potentially heterogeneous states are sometimes referred to as ‘official nations’ or ‘territorial nations’, whereas the term ‘ethnic nation’ is reserved for populations with a shared lineage and culture etc., and ‘nation-state’ for those ethnic nations which achieve political independence. Considering China, much twentieth-century political activity aimed at establishing China as a strong modern nation has been focused on building an official nation which includes not only the ethnic nation of the Han Chinese, but also other non-Han minority peoples living within the borders of China who do not share obvious ancestry and language with the Han majority. Safran (1998b) and others point out that the sizeable presence of such minority groups in China makes it extremely difficult to conceive of China as a nation-state populated by a single ethnic nation. Earlier, clear reference to properties of an ethnic Han nation were however prominent in the speech of Dr. Sun Yat-sen, who Guo (2004: 160) notes as having described the Chinese nation as potentially being defined by ‘blood kinship, common language, common livelihood, common religion, and common customs’. In assessing the relation of Chinese to national identity, it is therefore important to consider what kind of ‘Chinese nation’ is being conceived of in any characterization of national identity – a Han Chinese ethnic nation excluding minority groups, or a more inclusive but heterogeneous official nation made up of both Han and non-Han groups. The bifurcation between ethnic and official nations as (potentially) different kinds of population groupings also feeds into the occurrence of different types of nationalism – ‘cultural nationalism’ and ‘political nationalism’ (Hsiau 2000; Gladney 2004; Guo 2004; Wei 2002). The former type of nationalism is a movement focused on the maintenance and strengthening of a shared historical culture among a people with the properties of an ethnic nation, whereas the latter is a political movement which attempts to achieve and maintain independence and autonomy for a population as an official nation. As will be noted later in the chapter, both forms of nationalism have occurred in China at different times, with differing repercussions on issues relating to language and national identity.

1.2.  Language and national identity ‘National identity’ is commonly viewed as the self-conscious belief and subjective awareness that people have of belonging to a nation with certain objectively describable properties, such as a common history, language, ancestry, and culture, in ethnic nations, and membership of a territorially autonomous state with equal rights and a common future to invest in official nations, the latter occasionally also being referred to as ‘state identity’. In the cultivation of feelings of national identity, language has regularly been assumed to have an important role to play, as a symbolic marker of group identity and boundary device which can separate adjacent populations with different languages, and create sentiments of group self-interest and solidarity among those who speak the same language. Furthermore, the promotion of a standardized, common ‘national’ language throughout a population has the potential to even out socio-economic inequalities and stimulate the unification of a nation through the provision of better opportunities for advancement and future prosperity (Simpson 2007). With regards to China, Chinese language has consistently been seen to have served a major binding role among the Chinese nation, as it has emerged in modern times, as part of Chinese national identity. This will be discussed at some length in future sections. The primary

91

Andrew Simpson

components of this general (Han) Chinese national identity are suggested in Meissner (2006) to include and be predicated on the following: (i) the long history of China, (ii) the shared identity of the Han people as descendants of the Yellow Emperor, (iii) the notion of a Chinese Empire continuing through different dynasties, (iv) the uniqueness of the Chinese language, (v) shared traditions of religion and philosophy, (vi) Chinese literature, art and music, and (vii) Chinese achievements in the area of medicine, ceramics, and the invention of printing, papermaking, gunpowder, and the compass. Meissner (2006) and Smits (2013) both suggest that while the ready ingredients for a Chinese ‘national’ identity were indeed long present in previous centuries, clear feelings of nationalism and national identity only emerged and grew from the mid-nineteenth century onwards, when foreign incursion into China and military defeat triggered the beginnings of a new collective national consciousness and modernization, amid resistance to outside Western and Japanese forces. Such a national identity then developed further and in different ways following the course of major political changes in China during the twentieth century and the economic booming of the country in the early twenty-first century.

1.3. ‘Chinese’ The final, complex term which needs some clarification here before we consider the inter­ action of ‘Chinese’ and ‘national identity’ is the language label ‘Chinese’ itself, often discussed and disputed and of major importance to the notion of Chinese national identity. The critical question is whether there is any single language that can actually be referred to as ‘Chinese’, or whether there are many distinct languages which might all be classed as Chinese languages, as was suggested by Bloomfield (1933) in his statement that: ‘Chinese is not a single language but a family of languages made up of a variety of mutually unintelligible languages’. As noted by Bloomfield and many others, the problem is that, descriptively, there are many regional varieties of ‘Chinese’ that vary so significantly in their pronunciation that they cannot be understood by speakers of ‘Chinese’ from other parts of China. If the criteria of mutual intelligibility is applied as a linguistic means of distinguishing languages from dialects, the conclusion should be that northern/Mandarin Chinese and Min, Hakka, Xiang, Gan, Wu, Yue, Jin, and Hui varieties all constitute different languages. However, officially, all such regional forms have been classified as dialects of a single Chinese language, and subjectively, such a classification has largely been accepted as reasonable within China and has not been seriously challenged. What has significantly helped buttress the impression of a single Chinese language with regional dialect forms is the existence of two sets of connections among such forms. First, it has been noted that although the ‘dialects’ are pronounced in ways that are often extremely different from each other, they do nevertheless share much of a basic lexicon and grammar. Second, since perhaps the time of the Qin (221–206 bce) and Wei–Jin (ce 220–420) dynasties, the writing of Chinese has almost exclusively been carried out in all Chinese-speaking regions in a uniform way, making use of a parallel style, vocabulary, and orthographic means of representation, which for much of Chinese history has been substantively different from common forms of spoken Chinese. This special linking function of the written language, coupled with strong underlying lexical and grammatical similarities among regional forms of Chinese has been enough to maintain the image of a single Chinese language (with regional dialect forms), and allow it to serve as one of the several pillar components of Chinese cultural, ethnic, and later national identity, with common standardized forms of both written and spoken Chinese becoming ever more important during the course of the twentieth century, as will shortly be discussed. 92

Chinese language and national identity

2.  Language and the development of national identity in modern China 2.1.  The emergence of China as a nation, political nationalism, and language reform There is some divergence in opinion as to whether a Chinese ‘nation’ existed in pre-modern times. Certain scholars have suggested that China was indeed a nation from early times on, perhaps from as early as the Qin–Han unification (Townsend 1996: 25) with a population sharing a single culture, ancestry, and history, governed by a powerful, centrally administered state. However, the majority opinion is that nationalism, and the emergence of China as a nation, only occurred from the late nineteenth century onwards (Xu 2002; Guo 2004). What preceded this is viewed at most as being possibly a ‘collective consciousness’ (Guo 2004) of belonging to a larger state with a widespread, similar culture and historical continuity. Smits (2013) suggests that such a consciousness of being Chinese was furthermore largely confined to the elites of society, and that the vast majority of the population in the empire had no clear feelings of being Chinese. He adds that national consciousness in China remained minimal among non-elites through until the time of the first Sino-Japanese war in 1894–5. Ten years afterwards, however, it had started to become strong in all those areas in which foreigners had come to live and carry out business in China, and later on, in the 1920s, propaganda distributed by the Guomindang nationalists helped stimulate the wider growth of Chinese national identity among all sections of society. Subsequently, from 1949 onwards until the present, the Chinese Communist Party fueled this further into a much stronger sense of national consciousness among ordinary people, so that ‘Today, national consciousness in China is intense’ (Smits 2013). Considering language in pre-modern, pre-nationalist times, written and spoken forms of Chinese underwent significant divergence from the end of the Han dynasty (206 bce – ce 220) onwards, with regional, spoken forms of Chinese diverging both from each other and from the written language of classical Chinese, which itself remained unchanged and fully dominated the creation of all official written Chinese and high literature right until the early twentieth century. However, the acquisition of literacy skills in classical Chinese was both very costly and highly time-consuming and beyond reach for all but the advantaged classes in China. Consequently, when China entered the twentieth century, its general population still had no common spoken form of Chinese that could be widely used and understood, and little widespread knowledge of written Chinese, which had long become disconnected from all spoken forms of Chinese. In pre-modern China, language thus neither served as an effective linking device among most of the large population, nor helped facilitate the acquisition of literacy because of the differences between written and spoken Chinese and the effort needed to learn classical Chinese. At the beginning of the twentieth century, it was felt among emerging nationalist leaders that some kind of language change was necessary as part of the broader drive to modernize the country and stimulate a stronger sense of collective national identity among the people. It was argued that only when the population in general could communicate effectively, have easier linguistic access to education, and feel united as a single people would China be able to strengthen itself and successfully defend itself against foreign incursion and exploitation. The modernization that took place was influenced by changing orientations towards the value of Chinese traditions and culture in comparison to Western systems of knowledge, and had important consequences for the kinds of language reforms that were advocated and adopted. The language changes that occurred in turn reflected a particular view of the ways 93

Andrew Simpson

that language should best support the development of national identity and modernization in China. Meissner (2006) charts three distinct phases of attempts made at modernization prior to the founding of the People’s Republic of China in 1949, in which differing amounts of Western technology and culture were absorbed into China. In the first period, from the mid-nineteenth century to 1895, there was a selective adoption of Western scientific learning for purely practical purposes, and the Confucian system and other forms of Chinese traditional learning were maintained to regulate other social and cultural areas of life. In the second period, 1895–1911, it was claimed by certain intellectuals that the adoption of Western technology alone would not strengthen China sufficiently, and other aspects of Western learning were imported for use alongside traditional Chinese philosophies. In the third, most important period of modernization, 1911–49, four intellectual groups vied for dominance in the development of post-imperial China, with significantly different outlooks. The Confucians strongly defended Chinese traditions, and campaigned fervently against the increased adoption of Western ideas. The nationalists, led by Sun Yat-sen, attempted to blend Chinese traditional culture with Western political models. A vocal body of ‘liberal thinkers’ argued that only the full-scale import of Western learning would save China from the advances of its enemies. Finally, the Marxists, led by Mao Zedong, set out to destroy Chinese ‘feudal’ culture and Confucianism and replace this with a new socialist identity inspired by the writings of Marx and Lenin (Meissner 2006: 45–6). These different attitudes towards the optimal way forward for China competed with each other and affected the shape and success of proposals intended to ‘modernize’ Chinese and make literacy and a national language accessible to the masses. What was broadly agreed upon was the need for significant improvement in the ability of common people to read, write, and potentially communicate with other Chinese throughout the country. What caused dispute was the way this should practically be achieved, particularly in the area of the orthographic representation of Chinese either via traditional characters or some alternative mode of writing. Prior to the vigorous debates over orthography, however, important progress was made relatively swiftly in reforming the language used to write down Chinese. While the norms of classical Chinese had been used in all formal writing and high literature for two millennia, and had grown increasingly separated from spoken language, at the end of the nineteenth century a movement grew to fully replace classical Chinese style with a written style more closely connected to the vernacular. A form of vernacular writing known as baihua had in fact come into use from the time of the Tang dynasty (ce 618–907) onwards, but was highly restricted in its use, being reserved for certain ‘lower’ forms of literature and the writing of unofficial notes. As pressure for more effective written communication asserted itself at the turn of the century, a form of baihua based on northern varieties of Chinese came to be increasingly used in writing and in print, and was officially sanctioned by the government in place of the older written style. Chen (1999: 71) notes that by 1911 there were dozens of new newspapers and magazines published in baihua, over 1,500 baihua novels, and that rising generations soon came to learn to read and write via baihua following a 1920 decree of the Ministry of Education that baihua be used as the language of literacy in schools. The replacement of classical Chinese with a written style much closer to contemporary spoken forms of Chinese had the effect that the acquisition of literacy was made easier to achieve for the common population, a clearly positive result. However, it was argued that the use of thousands of complex Chinese characters to represent the language still created a major impediment to the rapid learning of written Chinese, and was likely to hold back progress towards mass literacy. Radical suggestions were subsequently made by many 94

Chinese language and national identity

leading liberal intellectuals to actually replace the traditional system of characters with an alphabetic system as a way to speed up the acquisition of written Chinese further, and though such proposals clearly threatened to eliminate one of the central symbolic components of Chinese tradition and identity, quite remarkably the motion to use a system of Romanization to write Chinese received very wide support as a desirable goal to work towards, being opposed only by the Confucianists. Chen (2007) highlights how surprising it may appear to be that major liberal thinkers and nationalists would advocate such a potentially drastic change to the distinctive writing system of Chinese, which had maintained a link among different Chinese dialect groups and functioned as a highly visible symbol of the Chinese throughout their long imperial history. Chen notes that prominent intellectuals of the time not only emphasized the simple practical value of introducing a less complicated alphabetic representation of Chinese for the learning of reading and writing, they also vigorously and emotionally depicted the traditional system of characters in strongly negative terms, referring to it as ‘backward’, ‘clumsy’, ‘abominable’, ‘primitive’, ‘uncivilized’, ‘coarse’, and having other highly undesirable qualities. Whereas aspects of inherited linguistic tradition have elsewhere often been made use of as important symbols of national unity by groups seeking to ignite popular nationalist sentiments (e.g. in Germany, Poland, France, Croatia), precisely the reverse appeared to happen with regard to written Chinese during the modernization of China in the twentieth century, and not only the classical Chinese style of writing but also the long-revered system of characters were targeted as negative inheritances from the past, responsible for impeding progress towards the attainment of a united, modern nation. Chen suggests that such an anti-traditionalist position successfully dominated discussions of language and modernization among early Chinese nationalists due to the special circumstances of China’s history and its language. He notes that China had long evolved as an independent polity, and Chinese nationalists therefore had no need to incite the population to rise up and establish China as a new territory. The potentially symbolic value of the Chinese language consequently did not need to be made use of as a force to trigger nationalist sentiments, unlike certain nation-building situations elsewhere in which a focus on inherited, shared language has been pivotal in helping amalgamate a previously un-united population in a new territorial space. Rather, what was needed in China in the view of nationalists and intellectuals was a purely utilitarian approach to language and the adaptation of spoken and written forms of Chinese in whatever way would best help the nation achieve its modern­ ization goals. If this meant the discarding of traditional forms in order to make progress and strengthen the country, so be it. China should not dwell on its past traditions as sacred in any way, but look to a new, modernized future in which China would secure its position and be able to defend itself against foreign agression. As things happened, a switch from the use of characters to an alphabetic representation of Chinese ultimately did not come about. Internal discord in China involving conflict between nationalist and the communist armies, followed by Japanese invasion, interrupted efforts at language reform, including plans to devise an alphabetic representation of Chinese, and a second major language project, the ‘creation’ of a spoken national language, guoyu, that could be used as a means of communication by all the population and unite the nation. When these two projects were resumed in the 1950s, the drive to replace characters with a form of Romanization was superseded by a different modernization strategy aimed at simplifying the learning of written Chinese, and the national language project also underwent change. These events played out as China was led in a new direction under a quite different nationalist philosophy, as the People’s Republic of China came into existence. 95

Andrew Simpson

2.2.  The People’s Republic of China, simplified characters, putonghua, and regional forms of Chinese The plans to reform spoken and written Chinese noted in 2.1 were made and effected during a time when the dominant approach to nation-building was utilitarian rather than symbolic and also significantly one of political nationalism, aimed at a strengthening of the Chinese state in general rather than being guided by ideas of any Han-centric ethnic nationalism. Certainly, it was northern Chinese that was earmarked to become promoted as the base of the national language, but under the circumstances this was simply a natural, practical choice given the very large numbers of speakers of mutually intelligible northern dialects. It was not supposed to symbolize any belief that Han Chinese in the north somehow represented the critical ‘essence’ of the nation, hence that their way of speaking should specially be adopted as the nation’s language. The selection of northern Chinese as the future expanded link language in China was purely motivated by the practicality of developing a successful lingua franca for the modernizing state. As the founding of the People’s Republic of China ushered in a fourth phase of attempted modernization up until the late 1970s (Meissner 2006), this new period in China’s history saw a significant rejection of both Western and Chinese ‘feudal’ culture, particularly during the Cultural Revolution, and the attempt to shape China according to Marxist–Leninist socialist principles. Concerning language and pre-People’s Republic of China plans to fully replace characters with an alphabetic representation of Chinese, this potentially dramatic change was not made by the new communist regime, and efforts at developing a Romanization schemata for Chinese were restricted to the adoption of the pinyin alphabet system as a supplementary way to help in the early learning of characters and writing. Chen (2007) suggests that the primary reason why characters were ultimately not abandoned was in fact a practical, linguistic one, specifically that the existence of large numbers of homophonous words in Chinese makes the use of an alphabetic system impractical for the writing of Chinese and would have resulted in much confusing ambiguity in the written language. A second interesting reason why the Romanization efforts may have been significantly reduced in the 1950s, hinted at in Chen (1999), is the fact that this might have opened up the way for the expanded application of alphabetic/phonetic writing to other, non-Mandarin dialects of Chinese in a politically undesired way. From the time of the Qing dynasty onwards through the nationalist era, there were common warnings that the development of ways to successfully write down southern dialects of Chinese was harmful to the unity of Chinese and the unity of the country, and similar concerns may have had an impact on Romanization language planning in the early years of the People’s Republic of China, when the leadership was heavily focused on establishing and maintaining a unified state. If the common character-based writing system were to be replaced and allow for multiple regional alphabets in different dialect areas, the unifying role of written Chinese would be lost, facilitating political division along regional lines and the possibility that China might even split into different linguistic ‘nations’. What did happen to the script, instead, is that it underwent a thorough simplification process in which many of the complex, traditional characters had their shapes converted into new simplified characters requiring fewer strokes, over 2,500 simplified characters being officially promulgated for use between 1956 and 1964. As this alteration of traditional characters into modern, new forms was mostly effected at a time that ‘feudal’ traditional culture was under common criticism, it did not cause the kind of defensive reaction from nationalists that was directed at similar simplification attempts in Japan during the 1930s, where it was argued that tampering with traditional characters was a direct attack on the 96

Chinese language and national identity

national identity of the country, heavily embodied in the writing system. Indeed, as Guo (2004) notes, the communist leadership in the People’s Republic of China aimed at revolutionizing China, and in the area of language this naturally translated into a jettisoning of the traditional language. The new simplified characters then came to be associated with modernization and the eradication of China’s ‘backward’ feudal past. As for spoken Chinese, the continuation of the goal to create a form of Chinese that could be learned and used by all of the population as a link language was re-initiated in the early years of the People’s Republic of China, taking the Beijing dialect as the base for most pronunciation norms and northern varieties of Chinese as the source of grammar and lexicon. Significantly, there was a renaming of this form of Chinese from the pre-People’s Republic of China term guoyu ‘national language’ to putonghua ‘common language’. In Chen (1999: 25), it is suggested that the adoption of the new designation ‘common language’ occurred so as to avoid the perception that northern Chinese was being presented as more representative of the nation than other varieties of Chinese, and signals a clear move away from overtly emphasizing any links between language and national identity as communicated by the term guoyu. In such a way, the non-political promotion of a national lingua franca in the People’s Republic of China contrasts with the symbolic use of language by ethnic nationalist movements in other states during periods of modernization and the strengthening of independence. The promotion of putonghua from the mid-1950s onwards was very successful, and putonghua is now very widely understood, used in education, the media, and also socially in many areas. Liu (2011) notes that the 2001 ‘Law of the People’s Republic of China on the Use of Chinese Language and Chinese Characters’ prescribes that putonghua and standard characters are to be the principal language and means of writing in broadcast media, movies, school education, and administration, and discourages the use of dialects in such domains. With the great success of putonghua, it can be said that though the initial motivation for spreading this form of Chinese was certainly functional, putonghua has now over time become a clear part of Chinese national identity for much of the population at least, in the sense of being a major part of everyday life in China for much of the population, speaking and ‘consuming’ putonghua in the media and in written Chinese. The regular daily use of putonghua therefore embeds a national link among the population and connects up people from around the very large area of the People’s Republic of China, instilling feelings of belonging to a single national entity. Although the government has strongly promoted the learning of putonghua as a shared means of communication in China, it does not forbid or even discourage the use of local dialects in non-official domains of everyday life (e.g. at home and with friends etc.). In the pre-People’s Republic of China nationalist times, there were also no initiatives to suppress dialect use, and this attitude has been largely maintained ever since. Dialects are seen as perfectly appropriate for use in the more informal domains of life, and at no point has the Chinese leadership pursued an agenda to fully replace other varieties of Chinese with putonghua. While there has thus not been any explicit suppression of non-Mandarin varieties of Chinese in domestic and social domains, the successful spread of putonghua has nevertheless had the effect that knowledge and use of regional forms of Chinese has decreased, significantly in the case of certain varieties. As putonghua has increasingly become associated with modernity and higher levels of education and socio-economic status, it has come to be used more frequently not just in those domains where it is officially promoted (education, media, and government administration) but also in informal interaction, and ever more so by urban younger generations. Despite the observation of such a general downward trend in use of 97

Andrew Simpson

regional varieties of Chinese as putonghua has strengthened, there are also some recent signs that certain dialects may now be enjoying new prestige among younger-generation speakers, and, for example, the use of Shanghai dialect is now viewed as ‘hip’ and trendy among the young, distinguishing them from their parents’ generation who may largely use just putonghua. This use of dialect is not confined to oral communication but has been making its way onto the Internet among the young, with dialect writing being used in online forums such as shanghaining.com where there is no strict emphasis on standard language. As Liu (2011: 62) observes, ‘Dialect writing on the Internet celebrates creativity, flexibility, multiplicity, heterogeneity, unpredictability, informality, freedom, and rebellion’. There is also an increased occurrence of the use of dialects on television in talk shows, docudramas, dialect-dubbed films, dialect television series, and sitcoms, despite the 2001 Chinese Language Law (Liu 2011). Yet in the absence of any well-accepted, standardized way of writing regional forms of Chinese, it is hard to see such forms posing a real challenge to putonghua and potentially replacing it in commerce, education, and officialdom, even if the government were to allow this. Hence the cohesion facilitated by putonghua in both its spoken and written modes is unlikely to be threatened by the existence of popular regional forms of Chinese. It may also be hoped by the government that the overarching Chinese state identity reinforced by daily use of a common language will serve as a counterbalance to new emphases on regional cultural differences among Han Chinese groups. Gladney (2004) remarks that with the rapid economic growth in the south of China, southerners such as the Cantonese, and others, have begun to assert their cultural differences. As part of such a process, there has been a reinterpretation of Chinese history, and southern scholars have argued that the early southern Kingdom of Chu spread its culture northwards to significantly influence the development of northern Chinese civilization, rather than the other way round as is commonly believed; hence there is a local re-evaluation of the importance of the south to China’s past and presently inherited identity. Furthermore, relating to the existence of an ethnic Han Chinese national identity, many southern groups are noted to have traditionally regarded themselves not as Han people descended from the Han Yellow Emperor, but as historical descendants of the Tang dynasty. The potential for cracks in an assumed all-encompassing Han identity is therefore present, with possible consequences for the broader unity of Chinese ‘state’ national identity. The connections strengthened throughout the People’s Republic of China from the increased use of putonghua may consequently be viewed as beneficial for national unity in shoring up signs of fragmentation caused by other factors.

3.  Non-Han minorities, language, and national cohesion Approximately 8% of the population in the People’s Republic of China is made up of non-Han minority groups, recognized as belonging to 55 different ‘nationalities’. All such minority nationalities are officially held to be Chinese, the People’s Republic of China proclaiming itself to be ‘a unitary multi-ethnic Chinese nation’ (Dwyer 2005, quoting a People’s Republic of China Embassy statement). The Chinese leadership, at least as far back as Mao and his slogan ‘nationalities unite’ (minzu tuanjie, quoted in Safran 1998b), therefore envisage non-Han groups as belonging to the Chinese nation along with the Han. A natural question is whether the minority nationalities themselves feel that they are genuine parts of such an announced multi-ethnic Chinese nation. To some considerable extent, this depends on how effectively a state national identity is stimulated among the minority nationalities through measures of inclusion, including equitable access to socio-economic resources, the stimulation of feelings of being welcome partners in the future of the nation, the enabling 98

Chinese language and national identity

of increased communication between Han and non-Han groups, and the non-devaluation of minority language and culture. With regards to language issues, government policy towards minority languages has gone through a number of different stages. In the first two decades of the People’s Republic of China, there was an assimilative drive to spread putonghua at the expense of the maintenance of minority languages. However, the 1970s and 1980s ushered in a new era of linguistic and cultural accommodation which has essentially been maintained until the present. The government initiated work to help develop minority languages, and the 1975 constitution confirmed the right of all nationalities to use their spoken and written languages (Dreyer 2003a: 369). Such rights were recognized again in the 1999 constitution (Article 4, quoted in Chen 2007: 162), which states, ‘The people of all nationalities have the freedom to use and develop their own spoken and written languages, and to preserve or reform their own ways and customs’. Since the 1970s, then, there have been no attempts to eliminate or repress the use of minority languages (and culture), and certain government support for the development of these languages (the provision of writing systems, educational materials, etc.) has been made available. Minority nationalities are relatively free to make use of their languages, and Han Chinese living in minority areas are also encouraged to learn local minority languages, though in practice inter-ethnic communication tends to be carried out in putonghua or other strong regional forms of Chinese, due to the economic pressures minorities feel to learn Chinese in order to access better employment. Partially as a result of such pressures and the ‘marketvalue’ of competency in putonghua, many minority groups have experienced a decline in the knowledge and use of their languages in the communities where they live, in a way similar to the increasing shift to putonghua among Han Chinese from non-Mandarin dialect groups, and paralleling global patterns of language shift to economically more powerful and prestigious languages. Rather more positively, however, putonghua does provide the means for minority nationalities to participate fully in the economic, political, and social activities of the People’s Republic of China, and so in theory develop feelings of attachment and belonging to the Chinese state where other negative conditions do not undermine the growth of such a broader, non-ethnic, ‘national’ identity. In other non-linguistic domains, the People’s Republic of China has indeed attempted to alleviate the poorer living conditions present in many minority areas, and introduced special preferential policies for minority nationalities in family planning, school and university admissions, hiring and promotion, the financing and taxation of businesses, and regional infrastructural support (Sautman 1998: 87). The government also provides considerable financial support for large regions inhabited by minority groups, 50% of Xinjiang’s annual budget being subsidized by the government, and much higher investment being made in Tibet (Wang 2002). In this connection, Gladney (2004: 18) notes that various of the minority nationalities who live near China’s borders are aware that they are economically better off in the People’s Republic of China than their ethnic kin living across the border in states such as Kazakhstan and Kyrgyzstan. Where there have been ethnic problems between minority nationalities and Han Chinese, these have typically not been caused by language issues, but by other factors (e.g. the broader issue of increased autonomy and religion in Tibet and Xinjiang province). Hence in sum it can be said that the non-repressive and generally inclusive language policies implemented in China since the 1970s have laid the groundwork for minorities to feel more included in the nation and not negatively discriminated against, potentially allowing for a sense of national identity to develop over time with new generations, as has occurred in other ethnically mixed states where an official state language has been spread as a lingua franca without any accompanying suppression of minority languages; for example, Indonesia and Tanzania. 99

Andrew Simpson

Finally, concerning the general issue of whether it may be possible to construct bona fide ‘nations’ that are ethnically heterogeneous with multiple minority groups, the mainstream conception of nations necessarily requiring a shared ethnic origin, language, history, and culture has been challenged by a number of scholars and political scientists who advocate an alternative model of genuinely multi-ethnic nations which potentially lends itself to the situation in the People’s Republic of China. The nineteenth-century French philosopher Ernest Renan, for example, has argued that modern nationhood is not based on a shared language, culture or ancestry, but on a shared sense of destiny and the desire of a population to live together (Dreyer 2003b: 393). Within China, Guo (2004: 13) notes that such thinking has been echoed by a prominent professor of philosophy at Zhongshan University, Li Zonggui, who argues that what is indispensable to a nation is a sense of belonging together, not necessarily a common language, culture or historical bloodlines. In such conceptions, then, it is perfectly feasible for the minority nationalities to become part of a modern Chinese nation oriented towards the future, and the adoption of a Chinese national identity perceived in such a way should not require the abandonment of other ethnic language and culture. A second, recent approach to conceiving of the Chinese nation in a way that is inclusive of minority nationalities has been referred to as ‘racial nationalism’ in Meissner (2006), and is quite different to Renan’s view of multi-ethnic nations, essentially denying that fundamental ethnic differences exist among minorities in the People’s Republic of China. Meissner notes that since the watershed events of Tiananmen in 1989, when the government saw it as increasingly important to stress the collective identity of the Chinese, official documents have attempted to emphasize a common ancestry of both the Han and the minorities. Han and Tibetans (and other minorities) are presented as parts of a common race, and the early cultures of such peoples are regarded as identical. In such a view, Tibetans (and quite possibly other minority groups) may belong to the Chinese nation through a genuine blood relationship, and there is no need to imagine the People’s Republic of China as a multiethnic nation. While the historical accuracy of such declarations may be questionable, it is nevertheless indicative of the leadership’s desire to integrate the non-Han into a closer national identity and may function to adjust traditional perceptions of such groups as being unconnected to the Han.

4.  Chinese and national identity in modern times Perhaps the most striking phenomenon relating to Chinese national identity in very recent times in the People’s Republic of China has been strident calls from many Han intellectuals for a return to Chinese traditions and Confucian values as a way to take the nation forward and save it from the spiritual decay perceived to be growing in China in the wake of an increasing loss of faith in Marxism and other aspects of Western thought which had fuelled China’s modernization during the twentieth century. For the new ‘cultural nationalists’, it is paramount that China reclaim its national identity from traditional sources and Confucianism to combat the social chaos (crime, corruption, materialism, gambling, prostitution, etc.) which has arisen since the 1980s (Guo 2004), and calls for a return to tradition as a means to reenergize nationalism have significantly not been censured by the political leadership, which is keen to see loyalty to the state stimulated at a time when its Marxist platform of support may be less inspiring than in the past. The Chinese national identity which is being brought into focus by the new movement of the cultural nationalists is one which is considerably Han-centric, based on aspects of Han Chinese tradition and culture, and seems to be primarily 100

Chinese language and national identity

aimed at strengthening and uniting the Han core of the People’s Republic of China. It is therefore an ethnic nationalism which is not preoccupied with the integration of non-Han minorities within the state as nation, and instead makes its first concern the rejuvenation of the morally-threatened Han population from its traditional roots. Organized into prominent bodies such as the International Confucius Foundation, the Chinese Yellow Emperor Association, the Chinese Culture Society at Beijing University, and the Beijing International Chinese Character Research Association, Guo (2004: 18) notes that the cultural nationalists take the Chinese language to particularize the Chinese way of thinking and enshrine the ‘spirit’ of the nation, and it is bemoaned that modern Chinese thought has become weakened and confused by the intrusion of Western influences. In new works of literature created by the cultural nationalists, Guo (1998: 172) reports that writers have accordingly set out to restore the purity of the national language corrupted by Western cultural influences and bring back ‘the Chinese essence’ through writing about Chinese traditions. Extending the idea that Chinese embodies the ‘spirit’ of the nation to include its special mode of orthography, a major linguistic focus of the cultural nationalists’ activities has been the defense of traditional characters, which in much of the twentieth century had been negatively portrayed as the direct cause of low levels of literacy among common people. The cultural nationalists argue that characters, and in particular traditional characters, should instead be recognized as instantiating ‘the root of Chinese culture, a symbol of the Chinese nation,  .  .  .  the single most important agent for national cohesiveness, the transmitter of China’s national spirit’ (Guo 2004: 105). They thus suggest that it is crucial for rising generations to learn traditional characters, as a means to access and appreciate the full meaning of Chinese literature and philosophy written in previous centuries, and also warn against any further attempts at character simplification or increased use of Romanization, which would cause more harm towards the national identity. Going on the offensive, the cultural nationalists have additionally made claims that, far from being detrimental to learning, Chinese characters actually help very considerably in the acquisition of literacy, and should therefore be valued and used as a possible mechanism to promote literacy in other languages of the world. Such an assertive defense of characters has led on to other eulogies of a range of ‘superior’ properties of the Chinese language which are taken to potentially qualify it for the role of a new world language (Lu 2008). These properties are well described in Guo (2004: 96) and include suggestions that: ● Chinese is the clearest and the most concise language. ● Chinese is the world’s easiest language to learn. ● The ideographs (characters) are ‘expressions of meanings’ and ‘symbols of ideas’ and store more information than linear writing. ● Chinese grammar is the closest to the grammar of mathematics, musical notation, and chemical symbols. Supporters of the potential spread of Chinese as an international language take heart from the growth in government-sponsored centers around the world (now over 200; Lu 2008: 269) which promote the international learning of Chinese language and culture; for example, the Confucius Institute in the USA. In this atmosphere of energetic assertion of the positive values/possible superiority of Chinese, it is also relevant to note that ‘Chinese’ is now increasingly being referred to as Hanyu, the language of the Hans, in everyday speech in China, in place of the ethnically more neutral putonghua ‘common language’ (Dwyer 1998: 80). 101

Andrew Simpson

This further points to the emergence of a Han ethnic nationalism at the heart of the resurgence in language pride and a perception of Mandarin Chinese as being principally anchored in the Han nationality. While such a movement to return to and re-exploit Chinese tradition has been strong amongst intellectuals in recent years, Guo (2004: 102–8) points out that there are also opponents to any increased use or learning of traditional characters, and it is maintained that the simplified script also has an important symbolic value for national identity, being the specially distinctive script of the People’s Republic of China. Guo notes that such proPeople’s Republic of China critics of cultural nationalism, termed ‘political state nationalists’, emphasize that ‘modern Chinese, simplified characters, and pinyin are the three pillars of the People’s Republic of China’s socialist culture’ (Guo 2004: 103), hence should not be displaced by the use of traditional characters, which are seen as symbols of capitalism, due to their continued use among overseas populations of Chinese living in non-socialist states. Guo concludes that this political symbolism of the simplified script will make it difficult for the government to allow for the reintroduction of traditional characters in any official way, though there has been frequent unofficial use of traditional characters in public signs such as restaurant and store names for several decades now, despite the imposition of fines for such displays (Chen 1999: 191). It therefore seems that it is unrealistic for the cultural nationalists to expect that simplified characters or pinyin will disappear and be replaced by full-form characters. However, in their campaigning for a return to the traditional past, the cultural nationalists have really put Chinese (Han) language back into focus as central to Chinese (Han) national identity, in a way that has not occurred so virulently before, during the twentieth century. It will be interesting to see how this engagement with national identity and the tensions between ethnic and political/state nationalism and language unfold further in the People’s Republic of China as China and its complex population confront new challenges through the twenty-first century.

References Bloomfield, Leonard (1933) Language, New York: Henry Holt. Chen, Ping (1999) Modern Chinese: History and Sociolinguistics, Cambridge: Cambridge University Press. Chen, Ping (2007) ‘China’, in Andrew Simpson (ed) Language and National Identity in Asia, Oxford: Oxford University Press, 141–67. Dreyer, June (2003a) ‘The Evolution of Language Policies in China’, in Michael Brown and Sumit Ganguly (eds) Fighting Words, Boston: MIT Press, 353–84. Dreyer, June (2003b) ‘The Evolution of Language Policies and National Identity in Taiwan’, in Michael Brown and Sumit Ganguly (eds) Fighting Words, Boston: MIT Press, 385–409. Dwyer, Arienne (1998) ‘The Texture of Tongues: Languages and Power in China’, in Safran (1998a), 68–85. Dwyer, Arienne (2005) The Xinjiang Conflict: Uyghur Identity, Language Policy, and Political Discourse, Washington: East-West Center. Gladney, Dru (2004) Dislocating China, Chicago: University of Chicago Press. Guo, Yingjie (1998) ‘Patriotic Villains and Patriotic Heroes: Chinese Literary Nationalism in the 1990s’, in Safran (1998a), 163–88. Guo, Yingjie (2004) Cultural Nationalism in Contemporary China, London and New York: Routledge. Hsiao, A-Chin (2000) Contemporary Taiwanese Cultural Nationalism, London: Routledge. Liu, Jin (2011) ‘Deviant Writing and Youth Identity: Representation of Dialects with Chinese Characters on the Internet’, Chinese Language and Discourse 2(1): 58–79. Lu, Dan (2008) ‘Pre-imperial Chinese: Its Hurdles towards Becoming a World Language’, Journal of Asian Pacific Communication 18(2): 268–79.

102

Chinese language and national identity Meissner, Werner (2006) ‘China’s Search for Cultural and National Identity from the Nineteenth Century to the Present’, China Perspectives 68: 41–54. Safran, William (ed) (1998a) Nationalism and Ethnoregional Identities in China, Portland: Frank Cass. Safran, William (1998b) ‘Introduction: Nation, Ethnie, Region, and Religion as Markers of Identity’, in William Safran (ed) (1998) Nationalism and Ethnoregional Identities in China, Portland: Frank Cass, 1–7. Sautman, Barry (1998) ‘Preferential Policies for Ethnic Minorities in China: The Case of Xinjiang’, in William Safran (ed) (1998) Nationalism and Ethnoregional Identities in China, Portland: Frank Cass, 86–118. Simpson, Andrew (2007) ‘Language and National Identity in Asia: a Thematic Introduction’, in Andrew Simpson (ed) Language and National Identity in Asia, Oxford: Oxford University Press, 1–29. Smits, Gregory (2013) ‘Topics in Modern East Asian History’, Online textbook at: http://www.personal. psu.edu/faculty/g/j/gjs4/textbooks/175/index.htm. Townsend, James (1996) Chinese Nationalism, New York: M. E. Sharpe. Wang, Lixiong (2002) ‘The “Tibetan Question”: Nation and Religion’, in Wei C. X. George and Liu Xiaoyuan (eds) (2002) Exploring Nationalisms of China: Themes and Conflicts, London: Greenwood Press, 151–72. Wei, C. X. George (2002) ‘A Cultural Search for National Identity: The Evolution of the Nationalism of Taiwan’, in Wei C. X. George and Liu Xiaoyuan (eds) (2002) Exploring Nationalisms of China: Themes and Conflicts, London: Greenwood Press, 123–50. Wei, C. X. George and Liu Xiaoyuan (eds) (2002) Exploring Nationalisms of China: Themes and Conflicts, London: Greenwood Press. Xu, Jilin (2002) ‘Intellectual Currents behind Contemporary Chinese Nationalism’, in Wei C. X. George and Liu Xiaoyuan (eds) (2002) Exploring Nationalisms of China: Themes and Conflicts, London: Greenwood Press, 27–39.

103

7 Chinese language education: Teacher training Jane Orton University of Melbourne, Australia

Provision Training paths The increasing global importance of China in the twenty-first century has led to growth in tertiary Chinese language programs and the quite widespread introduction of Chinese programs around the world in primary and secondary schools, and even kindergartens. This growth has brought a matching demand for teachers of Chinese at all levels of the educational system, as well as in commercial language programs. In Western countries, no formal training in teaching is required of those who teach languages at tertiary level, although some institutions offer professional development programs in teaching, and a few even make such programs compulsory. Primary and secondary school teachers, however, must be trained to teach their specific learning area, and the continuing demand for graduates trained to teach Chinese in schools has put considerable pressure on teacher education institutions. In the English-speaking world, and in many other countries, candidates obtain certification to teach school children by undertaking formal training as set out by the state, at national or provincial level. In many of these countries, certification is required of those who would teach in government schools, and most private schools also stipulate that their staff be certified. Typical routes to certification are a one-year Education course after obtaining a Bachelor’s degree, or a four-year Bachelor’s degree combining undergraduate studies and Education. In either case, for language teachers the undergraduate program will include major studies in the target language, literature, and culture, and the Education component will include studies of educational theories and systems, the nature of language and learning, curriculum design and assessment, pedagogy, and supervised classroom practice. There are less standard means of entering school teaching, including obtaining partial certification or provisional licensure. In Chinese language teaching these paths have opened over the past decade in some places due to the pressure to get the language going in the school system, and to the general shortage of qualified teachers to fill positions that are being 104

Chinese language education: Teacher training

created (Stewart and Wang 2005). For these reasons, too, some schools have asked first language (native) or heritage (Chinese background) speakers who have no training in language teaching, but are trained to teach other subjects in the curriculum, to shift into language teaching positions. In yet other cases, shortcuts to certification have come from innovative teacher training programs intentionally fast-tracking undergraduates into the classroom. These candidates receive some initial training and then follow their teacher education courses as they work, attending classes provided in an intensified format, often online.

Candidate profiles Prospective teachers of Chinese may be first language (L1) speakers of Chinese, who have grown up in a Chinese society, heritage speakers (HS), who have learned the language as a home language while living outside a Chinese society, or second language (L2) speakers, who have learned Chinese as a foreign language in a classroom. Where the student is already proficient in the language, they may be directed to follow undergraduate major studies largely in Chinese literature and culture. If they have already taken an undergraduate degree in the medium of the target language (in China, for example), students seeking to become teachers may even be exempted from the undergraduate language study requirement altogether, provided they pass a test showing that their culture and language knowledge and skills are at least equivalent to those of a graduating major in the language. Language studies offered to undergraduates in Arts faculties expect students to acquire proficiency in oracy and literacy, as well as a grounding in ancient and contemporary philosophy, history, and literature. These cultural studies make an excellent and, indeed, essential base for using the language well by providing deep, connected networks of meaning to the modern language being learned. Due to time pressures and the difficulty of catering to the very wide range of future areas of work of their diverse student body, undergraduate programs in language departments rarely offer professional language study as an option. An exception is the occasional document-reading course for law or business. As a result, while often able to recite classical poetry by heart, graduates of language departments who want to be teachers of Chinese have usually not acquired the vocabulary and phrases useful for running classroom activities and managing students of their own. Most tertiary teachers of Chinese are first language speakers and, like L1 speakers of all languages, without training in learning and teaching they often have little knowledge or understanding of the challenges their language poses for students. As well, having been educated in a Chinese system where students did not ask questions, and lessons of all kind consisted of teacher talk and home learning of the book by heart, the staff of most tertiary Chinese programs to date provide a narrow range of teaching methods and learning activities for their students, and have few strategies to pass on beyond being diligent and persevering. The American Foreign Service Institute in Washington DC estimates that it takes English speakers about 3.5 times longer to master Chinese than a European language, yet Chinese courses in schools and universities are scheduled in parallel with those of other languages, and all face restricted time limits. Given these factors, and the inherited devaluing of language development in favour of literature and social studies in Western universities, it is not surprising that L2 students of Chinese generally graduate with comparatively low proficiency in the language. Unless sustained by a lengthy time spent in country, spoken proficiency, in particular, is rarely of a standard desired as a model in teaching, or sufficient for running a classroom in the target language. There are some exceptions to this situation, but they are not yet the norm. 105

Jane Orton

Training programs Until very recently, almost all training in the teaching of Chinese as a foreign or second language in English-speaking countries was provided in a generic course on methods of teaching modern languages taught in a Faculty of Education as part of a broader pre-service secondary teacher education program. In some cases, casual staff were employed to add Chinese-specific matters to content originally designed for and staffed by teachers of European languages. In other cases, it was only when students went out for professional experience that they had contact with experienced teachers of their language. A dedicated method course in teaching Chinese language has been rare. By 2015 a number of Master degrees in the teaching of Chinese have appeared in Western countries. However, most such courses are taught in a department of Applied Linguistics and their content is predominantly, if not exclusively, centred on the nature of Chinese language. Few include pedagogy, and almost none combine linguistic knowledge with wider studies of learning and educational thinking or classroom management, and most have no period of supervised professional practice. Another strand of Chinese language teacher training comprises the Master of Chinese as a Foreign Language (MCFL) (对外汉语课) courses which have mushroomed in universities throughout China in recent years. The courses are offered by teacher education institutions such as normal universities, as well as by Applied Linguistics departments in faculties of Arts. In either case, they are strongly oriented to detailed studies of the language, with only some study of basic pedagogy, and no studies in educational thought. With at best only a handful of local institutions teaching Chinese as a Second Language available for their students to observe actual teaching of a class, usually at tertiary level, most graduates of these programs have not been exposed to learners of any age, let alone taught the language, and many have not even heard Chinese spoken by an L2 user. In an effort to strengthen their experience, the Chinese government’s Office of International Chinese, the Hanban, annually sends groups of MCFL students overseas for a year in one of a variety of countries, where they work as assistants to teachers of Chinese language. The training of primary teachers of Chinese is very varied in form. In some cases, L1 speakers who are not teachers are invited to conduct activities such as singing, counting, and storytelling with children while being overseen by a classroom teacher; in other cases, trained generalist primary teachers who have the language are invited to run such a program, or even to teach the language more formally; and, lastly, there are also certified primary teachers of the language, graduates of pre-service teacher education courses, who may exclusively teach Chinese, going from class to class in one or more schools, or who include language teaching as part of their general load. In addition to their Chinese language classes, generalist teachers of the language may largely teach curriculum in the local L1 language (e.g. English), or they may be fully occupied teaching in Chinese as part of a bilingual, immersion or intensive program of second language and content learning. There is little or no kindergarten or pre-school training for teaching Chinese as yet, and certified staff working with those age groups outside China have virtually all trained as early childhood educators in a Chinese society. As part of their training they will have undertaken child language development studies, although not studies in the development of Chinese as a foreign language.

Learner levels English-speaking countries have accepted migration for centuries, right up until the present. Chinese speakers have been a very large group among these immigrants, usually settling in 106

Chinese language education: Teacher training

concentrations within large cities. The children of these families speak their parents’ language, which may be a dialect, and many attend weekend classes to develop their literacy. With very mixed proficiency in standard Chinese, and very particular needs in terms of language development and cultural identity formation, their presence in classes in regular schools in numbers ranging from 30% to just one or two adds a challenging second level to L2 classes. As well, their proficiency is often a disincentive for L2 classroom learners to continue, especially in the competitive senior years. In the last years in some high schools there may also be international students from China preparing for university entrance. These are L1 speakers, who add a third level of Chinese curriculum, possibly to the one class, in which they follow studies in Chinese language and literature appropriate to their capabilities. There is little or no special teacher preparation for the two levels of background learners, and curricula and resources for them are oriented to the norms of education in a Chinese society, despite their being – or becoming – bilinguals outside Chinese society, with quite special needs.

Chinese teacher preparation Standards Teaching is a highly complex social practice that involves a practitioner drawing on integrated, articulated layers of different forms of knowledge in a great many areas. Like their colleagues in other professions, teacher educators continue to grapple with the tension between passing on the wisdom of experience and developing a new teacher’s own capacities. Programs are still commonly structured along a theory-to-practice line of progression, with candidates first instructed in formal knowledge of their subject matter and how to teach it, and then sent out into schools to apply this knowledge in real-life settings. Other programs, however, are based on perception of the task as a more iterative process, one in which the novice enters a cycle of knowledge and skill gathering, guided practice which tests the meaning of concepts and values comprehended so far and extends knowledge and understanding, reflection on practice, and the results of reflection fed back to inform and expand knowledge, understanding, and practice. While recognising that there are routines which can usefully be mastered by student teachers, professional educators aim to develop artistry in their novices: the ability to perceive and analyse situations which will inevitably arise that are novel, unexpected, or uncertain, and to design, test, and assess on the spot quite new action in response to them. It is this capacity that they nominate as the core attribute of a practitioner of high quality and an essential if a teacher’s best efforts are to be fruitful. Despite variations in perspective, three matters all teacher educators agree on are that ● deep knowledge of their subject is the absolute essential for a teacher to survive, to develop, and to innovate ● while learning to teach is a lifetime project, a certain breadth of factors must be involved in training from the start ● a practicum is an essential component of initial training. These elements are strongly represented in the professional standards for school teachers that have been developed in recent years in all English-speaking countries (e.g. United Kingdom Government Department for Education 2011). In the United States and Australia, dedicated standards for language teachers have also been established (American Council 107

Jane Orton

on the Teaching of Foreign Languages (ACTFL) 2002; Australian Federation of Modern Language Teachers Associations 2005), and the Hanban have published their own standards specifically for teachers of Chinese as a foreign language (2007). While each has its particularities, the lists all include ● ● ● ● ● ●

strong proficiency in the target language knowledge of the nature of language and culture knowledge of learning and learner development a repertoire of instructional strategies the capacity to plan a course of learning and to assess it the ability to reflect and improve their own practice.

Proficiency in the target language is the first area of knowledge required by all, and insistence on this reflects a reality in which the regulatory standard for admission to teaching candidature everywhere has most often simply been a pass in language courses. This has allowed for a very wide range of proficiency among those presenting as teacher candidates. Since publication of the standards, regulation testing of prospective language teachers to ensure they meet a high minimum spoken and written proficiency level in the standard form of the language has been increasing. Furthermore, as the teacher preparation standards in Table 7.1 show, development of this proficiency is seen as the shared responsibility of language departments and the Faculty of Education. In realising any of the sets of standards, thorough exploration of three educational factors present in the teaching of all language is especially critical in the sound preparation of Table 7.1  ACTFL Program Standards for the Preparation of Foreign Language Teachers The preparation of foreign language teachers is the joint responsibility of the faculty in foreign languages and education. In order for foreign language teacher candidates to attain the knowledge, skills, and dispositions described in the ACTFL Program Standards for the Preparation of Foreign Language Teachers, programs of foreign language teacher preparation must demonstrate that they include the components and characteristics described below. 1. 2. 3. 4. 5. 6. 7. 8.

The development of candidates’ foreign language proficiency in all areas of communication, with special emphasis on developing oral proficiency, in all language courses. Upper-level courses should be taught in the foreign language. An ongoing assessment of candidates’ oral proficiency and provision of diagnostic feedback to candidates concerning their progress in meeting required levels of proficiency. Language, linguistics, culture, and literature components. A methods course that deals specifically with the teaching of foreign languages, and that is taught by a qualified faculty member whose expertise is foreign language education and who is knowledgeable about current instructional approaches and issues. Field experiences prior to student teaching that include experiences in foreign language classrooms. Field experiences, including student teaching, that are supervised by a qualified foreign language educator who is knowledgeable about current instructional approaches and issues in the field of foreign language education. Opportunities for candidates to experience technology-enhanced instruction and to use technology in their own teaching. Opportunities for candidates to participate in a structured study-abroad program and/or intensive immersion experience in target language communities.

108

Chinese language education: Teacher training

Chinese teachers. The first is that, in addition to knowing the nature of language and student learning, and target language particularities, teachers need to understand the target language as an object of learning, that is, to perceive the demands that learning it will make on learners who are speakers of their particular language. This is the crucial link between knowledge of the teaching content and design of methods and resources which will successfully allow learners to master it. In the teaching of Chinese to speakers of Western languages, fine analysis of the learning task involved in the various aspects of the language identified as needing to be taught has been insufficient to date, and this is a major factor in the comparatively high failure rate in Chinese learning. A second factor that teachers must be taught to appreciate is the more general development that is achieved through language learning, what might be called its educational affordances. While this most normally is of central concern in school education, it is nonetheless always present and no less a value for college students or adult learners in later life. The third factor teachers in preparation must know and take account of is the culture of education in which the teaching will take place.

Educational factors Chinese as an object of learning There are a number of core features of Chinese language which make demands on learners from European language backgrounds that the expertise they developed in acquiring their first language is insufficient to meet, and hence these features require further foundations be laid. The demands lie principally in the fundamentals of the phonological and orthographic systems, and the features of Chinese vocabulary. There are also some very specific matters in the grammar and the culture which are a challenge to European learners. These very central aspects of Chinese need to be known by Chinese teachers, who must be prepared to use resources creatively if they are to assist learners to successfully meet and master them.

Oracy The most salient feature of the Chinese phonological system is that syllables are tonal, that is, they vary in pitch and contour and this variation changes the meaning. To obtain meaning, these shifts in pitch must be heard with the segmental sounds, so that mà and má are heard as being as different as ‘pan’ and ‘pen’ in English, that is, as different words, with different meanings. To perceive tone, L1 speakers take their cue from both the entry pitch of a syllable and the contour. Tone is a very new use of voice for speakers of non-tonal languages and most find tones difficult to perceive, and both entry pitch and contour hard to control. One fundamental reason for their difficulties in tone perception is that they simply forget to listen for it. Instead, as they are accustomed to doing in their own language, speakers of non-tonal languages naturally attend only to the segmental sounds. Thus they will say they heard, for example, the syllable shi and then wonder what tone ‘to put on it’. But there is no shi in Chinese, what they heard was either shī, shí, shǐ, or shì. A second difficulty is that tones are volatile, with only syllables that are being stressed in an utterance carrying full tonal value. Indeed, in modern rapid spoken Chinese, it is estimated that as much as 47% of syllables may be atonal (Triskova 2008). This causes perception difficulties, and in production, having diligently learned their vocabulary of tonal syllables, 109

Jane Orton

many learners have difficulty reducing tone on certain syllables when speaking, making their own speech overstressed and hard to understand. To speak intelligibly, learners need to master flows of syllables with varying tones and varying tone quality. The basic skill of judging pitch entry and the length of the contour takes work, and students can be unaware for quite some time that they are not achieving the sound they think they are making. On a positive note, studies show that work on tone perception and production are mutually beneficial and, over time, if the learning weaknesses of listening for tone and controlling pitch entry, contour, and tone quality are targeted, exercises can be quite effective in developing mastery (Miracle 1989; Wang et al. 2006; Orton 2013). Prosodically, rapid, intimate spoken Chinese is generally considered to be stress-timed, like English (but not French, for example), but Chinese word stress more often falls on the second syllable of a word, not, as in English, on the first. English speakers are thus often competent at chunking their utterances in Chinese but misplace stress. More difficult than rhythm and phrasing for these learners is restraining from using strong English intonational patterns to express attitude, which if used, very often interfere with the accurate production of tone. Use of the Chinese modal particles to carry some of the load intonation carries in their own language is a late acquisition for most learners of Chinese.

Literacy The densely compacted square of information that constitutes a Chinese character is a very different form of writing down human language from the linear strings of alphabetical combinations used in European and some other languages. Perceiving and obtaining meaning from characters demands learners of Chinese organise their reading eye in a quite new way, perceiving the square and noticing its components from left side to right side, and from top to bottom (Lin and Childs 2010). Although characters continue to be taught as a series of single strokes, the most efficient way to read a character is to perceive the, usually multistroke, components it is composed of, just as the most efficient way to read the English word London is to perceive the two starting strokes not separately as I and _ but together as L. Another significant difference involved in mastering reading in Chinese is that characters are written evenly spaced across the line, so that those representing the syllables of a multisyllabic word need to be gathered in one sweep of the eye, while other characters will be a single word in their own right. Characters do not represent sounds in the way that letters of an alphabet do, although some links with sound in about 40% of characters can be perceived once they are known. Thus learning a character requires learning both its sound and its meaning. As a way of recording the sound of a character, students are usually taught Pinyin, the orthodox form of Romanisation. Pinyin is a spelling system for the spoken language and, like all such systems, its sign-sound combinations are regular only to a point. Exceptions and anomalies exist and must be learned. Nonetheless, if mastered, Pinyin is the reliable way to know the accurate sound of a character or item of vocabulary. The most basic capacities in Chinese – to hear and utter, to read and write – require foreign learners to extend very basic motor skills and to change long ingrained habits in perception and production of spoken and written language. The skill base developed in learning their first language to a high degree of expertise and automaticity is simply insufficient to support European language speakers’ perception and production of the most basic Chinese. Their first attempts at writing characters also show starkly how true this is. To write an even passable hand requires the same laborious work gone through in elementary school. Some 110

Chinese language education: Teacher training

try to avoid the problem by using a word processor, but while the results look more pleasing than the childish script of the novice, studies are yet to be undertaken that could show whether the consequent reliance on the eye alone to identify a character, with no tactile support, and the reduction of time on basic task, will permit a student to achieve the same degree of reading competence and, especially, writing competence, once the volume of characters involved begins to mount.

Vocabulary Unlike any European language or even Japanese, Chinese vocabulary is special as an object of learning for speakers of European languages, first and foremost, for the sheer number of new words to be learned. There are virtually no cognates, so there can be no shortcuts as exist in the thousands of Latin- and Greek-derived words common to most European languages, such as education, commence, photograph, and in the everyday words shared among languages with a common history, such as the German and English Finger, Hand, Arm, Schulter, and finger, hand, arm, shoulder, or the many borrowings from English in modern Japanese, for example, pikunikku teiburu for ‘picnic table’. Chinese dictionaries list characters, which represent meaningful syllables, and are often also independent words. Ending only in a vowel or vowel + n or ng, to the European ear, Chinese syllables provide a restricted range of features to distinguish them and so are not easily perceived and retained. It is impossible to remember the learning process by which as babies we sorted out the sounds of our own language being made around us from all the possible sounds we could utter, but research shows we did all have to learn what variations matter and to recognise and disregard changes in sound that were due simply to phonic environment, or individual voice quality, or the emotion of a specific utterance (Gopnik et al. 2000). When confronted with very new sets of sounds that we are not in the habit of attending to and storing, we can perceive our bias towards those combinations found in our own language. Chinese presents the English-speaking learner with a large number of such sets. For example, the ratio of syllables starting with the letter ‘x’ in Chinese compared to English is in the order of 85 : 1, while the total of syllables starting with ‘x’, ‘y’, and ‘z’ make up some 22.4% of Chinese, but only about 0.6% of an English vocabulary (Orton 2014). The difficulty this presents becomes evident when students try to read characters they have only recently met: ‘Oh that’, they say, ‘that’s xu  .  .  .  xi  .  .  .  xue – something like that’. Even within one tone there are a great many homophones. The sound yì, for example, can be written in 50 different characters, that is, as 50 different syllables with 50 different meanings. While this is an extreme, there is considerable homophone load on a large number of syllables that appear commonly in everyday speech. The lack of cognates, the newness of many sounds, the limited range of sound combinations, and the heavy load of homophones make high demands on European learners of Chinese to develop their perceptive skills in ways never before required of them, perceptive skills they have largely left undisturbed since infancy and pre-school. At the same time, due to the high volume of new words that must learned, the burden on memory of learning Chinese vocabulary is substantially greater than that imposed in learning another European language. And this demand is made of a usually underdeveloped memory power, already overtaxed by having to remember a new character for every syllable. Unless they are assisted to extend their capacity to hear and have frequent opportunity to engage with vocabulary and characters in vital ways that involve mind and affect, learners will find new Chinese words difficult to internalise and have available for spontaneous use. 111

Jane Orton

Grammar Like all new languages, Chinese also presents its learners with new aspects of grammar, however most do not prove more challenging to master than, for example, those posed for a German speaker learning French, or a Spanish speaker learning English. Although a generally new class of word for learners, measure words and classifiers are not difficult to understand, but finding and remembering the correct one for each noun is a chore, as is linking verbs and their complements correctly. Having no articles or tenses, and no word for ‘Yes’ takes adjustment of perceptive and productive sensitivity to other terms being used for these functions, while the reverse clause order of complex sentences using de (的) segments and, especially, of a series of embedded de-segments, demands considerable cognitive rerouting in the learner if the relationships between clauses are to be correctly linked. Even when heard, the communicative significance of particles used to express aspect and mood takes a long time to impact the ear.

Culture A final aspect of Chinese as an object of learning is that of culture. Chinese are rightfully proud of their long and illustrious cultural heritage, and it is aspects of the traditional heritage which are taken to represent Chinese culture in most language textbooks and courses. Feast days such as Spring Festival and the Mid-Autumn Festival, historical events such as the founding of the People’s Republic, and world discoveries such as paper making and the compass, are described in detail, while lessons are provided in calligraphy and various folk arts and crafts such as paper cuts and seal making. In resources from China especially, there has been little about modern lived culture, nor any systematic introduction to the network of worldview and beliefs that still give meaning to much of the language in use today. Thus simply to say that Chinese has separate words for ‘older brother’ and ‘young brother’ is to leave the learner ignorant of the critical meaning of this: that age is the most significant relegating factor in Chinese society, one which people orient by constantly; that identity is realised in relationships organised by age; and that age relations will underpin expectations of the learner in social interactions with Chinese. At the same time, learners need to know that in the modern world, even such cultural basics as deference to age may be contested at times by the young owning new social power, such as money, technological knowledge, or fluency in a foreign language. While the exotic exemplified by dragons and grave sweeping can be enticing to young learners, their natural egocentrism may mean they find introduction to the new culture easier through the perspective of their own lives – what people eat for breakfast, what young people learn at school, how texting is done in Chinese – and only gradually develop interest in parts of the Chinese world different from their own. In any case, the essential in teaching about cultural matters is to help students see them as parts of a web of social meaning, linked to language, not just a set of singular, exotic customs.

Resources Knowledge and appreciation of the deep demands of Chinese as an object of learning set out above are still rare, and even the facts of the language and its culture are often not well known or understood by those who create resources, or the teachers who use them. Furthermore, to date few materials for teaching Chinese at any level are comprehensive in terms of 112

Chinese language education: Teacher training

the scope of interrelated factors they engage the learner with, and even within the content that is presented, many still repeat what has been standard information which modern research is widely known to have shown is not actually so. A case in point is the continual representation of a 3rd tone using a V-shaped mark, when most 3rd tones are actually pronounced with a ‘tub-shaped’ contour which has a very short drop, a lengthened, low, flat middle and a short rising finish (Cao 2002: 94). Despite bright colours and varied layout, a great many resources, including many produced on the Internet, remain not much more than compilations of exemplary grammatical patterns with vocabulary and character lists. There is often no development of the learners as language users, so succeeding chapters are identical in length and demand, with little to engage the mind in reflection on the nature of human language, or strategies for coping with the learning demands on memory. This gap between what is available and what is needed poses a challenge when seeking to help new teachers perceive the learning demands and build a repertoire of content and strategies that will enable their students to meet them. The gap becomes acute when considering teaching for real-life use, such as in intensive programs that use content and language integrated learning. One avenue offering remarkable potential in resource creation is information and communication technology (ICT). From flexibly scaffolded linguistic puzzles to trans-media adventure activities, to real-life contact on Skype, ICT can offer great support in dealing with the learning demands of Chinese. It is essential that beginning teachers steadily develop technical and pedagogical skills in using it effectively.

Educational affordances In most of Western culture, life is seen as ‘.  .  .  a series of situations in which one has to learn to think for oneself, to solve problems on one’s own, and even to discover new problems for which creative solutions are wanted’ (Gardner 1989: 5). Thus the education system in Western countries stresses individual identity and will, and aims to develop students to their fullest potential, able to live successfully in a society where the core value is to stand on one’s own feet. Within such a framework, language learning is valued for having particular potential to develop students cognitively and affectively (Adesope et al. 2010). As well, becoming bilingual has been shown to develop a tolerance for ambiguity and a flexible approach to problem solving, while accepting new ways of speaking and behaving allows appreciation of diversity and difference in thinking, valuing, and organising life. Learning a second language has proven to be influential in development of generic language skills, which can transfer to learners’ use of their first language (Demont 2001; Fortune 2012). This is especially the case in classrooms where metalinguistic knowledge is made available for discussion. Likewise, drawing learners’ awareness to similarities and contrasts between their two languages and cultures makes evident the nature of language and human society. From understanding cultural, historical, and geographical factors involved in producing the social phenomena they are studying, and examination of their own natural, emotional ethnocentricity, language learners can be assisted to develop a more rational acknowledgement of difference, and an inclusive, ethnorelative acceptance of it. These affordances are present in all language learning, but they have a particular significance in school learning, where the aim explicitly is to develop students as persons, not simply train them in a skill. In this regard, learning a language such as Chinese, which demands new development of the self cognitively and affectively, as well as presenting some challenging new ways of looking at the world, has particular potential value. 113

Jane Orton

The culture of education Learner-centred teaching Education is the quintessential cultural process by which a society passes on to the new generation its heritage of beliefs, values, and practices, and at the same time attempts to open these up to modification so students will have a viable life in an inevitably changed future society. The strong Western belief in an individual who has responsibility for his or her life leads to teaching–learning practices and ways of relating in the classroom aimed at fostering students’ ownership of their study and their development of well-reasoned personal opinions, their own stance on social and aesthetic matters, and the capacity to take initiative in responding creatively to novel practical situations. There is an underlying belief in the students’ capacity to discover knowledge for themselves and, indeed, this and the power to critically assess what they are discovering, are what schools set out to develop. Formal learning is undertaken by doing while in conversation with peers as well as teachers, and errors are seen as a natural outcome of experimenting, and treated as useful. From early childhood, living is presented as a series of causally linked intentional actions, a process in which the child is, or can become, an autonomous agent. Problem behaviour is handled by asking the child to reflect on the consequences of his or her actions, then left to choose a productive alternative. While generally accepted views are handed down, students are also expected to challenge propositions, and raising questions is encouraged, even admired. In contemporary education, the capacity to memorise matter is not highly regarded, with value instead placed on students being able to find knowledge when needed. Not having been required to learn much by heart beyond some spelling and the occasional poem, students grow up with very low powers of memorisation. They find the chore of learning the vocabulary and irregular grammatical patterns of any new language both arduous and irksome, and many find the burden of committing to memory the extensive vocabulary and characters of Chinese overwhelming, unless memory is assisted by frequent involvement with the items in mind engaging, puzzle-like activities. In teacher preparation, study of the philosophy and psychology of education is usually confined to specific courses, and in teaching method courses it is the learner-centred principles of practice which derive from these disciplines that are presented as rationale for the proposed teaching activities and learning strategies. There are some potential contradictions inherent in this approach when applied to modern languages education, and in the teaching of Chinese these are usually critical.

Contradictions in approach From the start of any language course, teacher and students are faced with the ambiguity of cultural reference point in their classroom: do they locate themselves in the local society of the school and only ‘visit’ the target language society through the learning resources, or do they locate themselves in a virtual target language society within their classroom? Or do they move between the two? Their position on this is expressed, primarily, in their choice of language for relating to one another and running the lesson: Do they begin with a greeting in English or the target language? Does the teacher become Ma Laoshi or remain Ms McIntyre even when speaking Chinese? Is there any type of activity, or point in the lesson, or place in the room, where only the target language may be used? These days there is not a great deal of difference in the culture of education and teaching between the various European 114

Chinese language education: Teacher training

and English-speaking societies, so that even where the classroom is transformed into a target language-only environment, the actual behaviour of those in the European language room is unlikely to be greatly modified from that being used in the mathematics class next door being run in, for example, English. In the Chinese language classroom this matter is more problematical. This is because the teacher is almost always a first language speaker of the language, who will have been educated in a culture of education and teaching which is different from that of Western societies, in several places to the point of contradiction. Furthermore, it will not only be in regard to relating to one another and organising activities in the room that these differences will appear, but also in their perception of the very nature of their task and the values they ascribe to the various means of working on it (e.g. Haley and Ferro 2011; Grant et al. 2013; Li 2014). Among key beliefs and values found commonly in teachers from China and Taiwan are that learning will be achieved by teachers displaying their own mastery of the subject matter through presenting aloud the information and explanations in the textbook; that language is to be first learned – by which is meant memorised – and only then used; the view that mistakes are bad and errors in language to be corrected at once by the teacher saying the right thing, while the error in the learner’s thinking or skills that led to the mistake is left to the learner to sort out; and the expectation that no questions will be raised, nor will assertions in the book or by the teacher be challenged. A heavy homework load is considered normal and learning is expected to come from committing chunks of language to memory using personally developed strategies and very high repetition of same and similar exercises. Inside and outside the classroom there is little personal contact between teacher and students beyond greetings or reprimands. Unless called on, students are expected to be silent, attentive, obedient, and diligent. Most first language speaker teacher candidates of Chinese approach their teaching in English-speaking countries not only armed with an experience of education throughout their life as described above, but also with the belief that, as the target is Chinese language, it is proper to expect students to learn it in a classroom that reflects Chinese educational culture. Most consider it natural to draw on their own learning experience as a child in which acquiring characters was paramount – 500 per year throughout primary school – and achieved through constant repetitive tracing and copying by strokes done in the correct order, and identifying individual characters on flashcards. As they were taught to as a child, many scorn the Pinyin writing system as babyish, and most have long forgotten it. Like L1 speakers everywhere, without special training, Chinese have difficulty identifying features of their own competently used language. Many cannot identify the tone of syllables they use and can say little that is useful to learners about the system of verb particles. To do so would require linguistic knowledge, deep understanding of the demands it makes on learners, and exceptionally good spoken command of the students’ own first language for explaining and discussing both. Knowledge of the features of the language can be taught and even where it is not part of formal study, most student teachers pick some of it up through textbooks and exposure to experienced teachers during their practicum. But even those who have graduated from Chinese as a Second Language courses in China, and arrive in classrooms overseas with knowledge of the language and its problem spots for foreign learners, have low understanding of the learning challenge as perceived by the learners themselves. Like their compatriots already abroad, most are also quite unprepared for relating personally to students in a firm yet nonauthoritarian way and for presenting their special knowledge so that it interests students and results in solid learning. In schools where Chinese is obligatory, they have few strategies for 115

Jane Orton

dealing with reluctant learners beyond bribing with treats or threatening with tests, and many give up, claiming the students are lazy and undisciplined. Even in primary school programs of only 30 minutes a week there is a relentless slide towards drilling the characters, which most young children eventually find tedious. Unused to the practice of speaking to students about their learning, and often fearful of being misunderstood or mocked for errors in their English, many miss out on the help in solving learning and behaviour problems that students can provide if asked. Outside the classroom, older teachers from China often find their perspectives not shared by local colleagues, so they retreat, participating in school life only where required, and rarely contributing in meetings except over formal matters. Younger, more recently educated L1 speakers are often more able to join in activities and relate to students with some warmth. However, with little prior experience of such an environment, these teachers often find themselves quickly losing control, unable to perceive the quite subtle boundaries for work and behaviour that their local colleagues set successfully. Although outwardly more inter­ active and varied in their teaching than their older compatriots, their view of the fundamental ends and means often remains traditional. As their unrestrained Chinese is not understood by students, most provide an immediate translation of anything they say, and many run their classrooms almost totally in the local first language, for example, English. Second language user teachers are commonly more confident in the social aspects of teaching and are experienced in a culture of education and learning that includes stimulating intellectual curiosity and independent discovery. Unlike their counterparts teaching a European L2 language, however, few have the language proficiency they need to teach in the target language, so while their approach might be advanced, their lessons also are often run largely in their own and the students’ first language.

The goal Teacher preparation in Chinese needs to assist L1 and L2 speaker teacher candidates to become aware of their deeply held beliefs, to perceive the differences and contradictions between them, and through dialogue to forge a platform for practice which they can both espouse. Doing this will mean recognising that educational policy in China has been considerably reformed in the past decade and that, in a break with tradition, it now promulgates an understanding of children’s capacity to learn for themselves and an appreciation that to be successful in the future, students must develop the intellectual curiosity, independence, and creativity cited by Gardner above. It will also mean understanding that the goal for students of the language outside China is not ‘to become Chinese’ but ‘becoming Sinophone’ – international users of the language with a developing bicultural identity (McDonald 2011). From these considerations, teacher educators and candidates will need to decide the question of which traditional Chinese teaching and learning practices are now obsolete, superseded by more creative modern strategies; and which remain cultural constants and hence essential learning experiences for students. Growth in demand for trained teachers of Chinese has strained existing providers and led in places to shortcut courses, a practice very much at odds with internationally accepted professional standards for language teachers. This is all the more risky given that teaching Chinese successfully requires teachers to develop extensive expertise to match the unique demands the language makes on speakers of European languages to develop quite new perceptive and productive capabilities, and manage a severe burden on memory. As well, to date the predominantly first language speaker cohort of teacher candidates and practising 116

Chinese language education: Teacher training

teachers of Chinese approach their task largely on the basis of beliefs and values drawn from their own experience in a very different educational environment from that in which they are teaching. In particular, their fundamental orientation to learning and student–teacher relationships contradict much proposed by Western educational institutions, while even those who would change lack experience of what it might mean. Second language speaker teachers rarely have sufficient language proficiency to teach as they might wish. To be effective, language teacher educators need a deep understanding of the challenges of the Chinese language and the professional skills to assist L1 and L2 teacher candidates to reflect on their understandings and develop dialogue over differences from which they may create a shared, bicultural perspective on their task and its goals, and jointly create the means to achieve them.

References Adesope, O. O., Lavin, T., Thompson, T., and Ungerleider, C. (2010) A systematic review and metaanalysis of the cognitive correlates of bilingualism. Review of Educational Research 80(2): 207–45. American Council on the Teaching of Foreign Languages (2002) Program Standards for the Preparation of Foreign Language Teachers. Alexandria, VA. Australian Federation of Modern Language Teachers Associations (2005) Professional Standards for Accomplished Teaching of Languages and Cultures. Retrieved 31 March, 2015, from http://pspl. afmlta.asn.au/doclib/Professional-Standards-for-Accomplished-Teaching-of-Languages-and-Cultures. pdf Cao, Wen (2002) 汉语语音教程 Hanyu yuyin jiaocheng (A course in Chinese pronunciation). Beijing: Beijing Yuyan Wenhua Daxue Chubanshe. Demont, E. (2001) Contribution of early 2nd-language learning to the development of linguistic awareness and learning to read/Contribution de l’apprentissage précoce d’une deuxième langue au développment de la conscience lingustique et à l’apprentissage de la lecture. International Journal of Psychology, 36(4): 274–85. Fortune, T. W. (2012) What the research says about immersion. In Asia Society (ed) Chinese language learning in the early grades: A handbook of resources and best practices for Mandarin immersion, 9–13. Retrieved 29 March, 2015 from http://asiasociety.org/education/chinese-language-initiatives/ chinese-language-learning-early-grades Gardner, H. (1989) To Open Minds. New York: Basic Books. Gopnik, A., Meltzoff, A., and Kuhl, P. (2000) How Babies Think: The Science of Childhood. London: Weidenfeld & Nicholson. Grant, L., Strong, J., Xu, X. X., and Popp, P. (2013) West Meets East: Best practices from expert teachers in the U.S. and China. Alexandria, VA: ASCD. Haley, M. and Ferro, M. (2011) Understanding the perceptions of Arabic and Chinese teachers toward transitioning into U.S. schools. Foreign Language Annals 44(2): 289–307. Hanban (2007) Standards for Teachers of Chinese to Speakers of Other Languages. Beijing: Hanban. Retrieved 31 March, 2015, from http://english.hanban.org/node_9906.htm Li, X. F. (2014) Reflection on pre-service science teacher education in China and North America. In Wen Ma (ed) East Meets West in Teacher Preparation: Crossing Chinese and American borders. New York and London: Teachers College Press. Lin, P. and Childs, R. (2010) An analysis of orthographic processing: non-Chinese and Chinese readers’ visual-spatial concept. In J. Chen, W. Wang, and J. Cai (eds) Teaching and Learning Chinese: Issues and Perspectives. Charlotte, NC: Information Age Publishing, 257–78. McDonald, E. (2011) The 中国通 or the ‘Sinophone’? China Heritage Quarterly 25. Retrieved 17 May, 2013, from http://www.chinaheritagequarterly.org/tien-hsia.php?issue.025 Miracle, W. C. (1989) Tone production of American students of Chinese: A preliminary acoustic study. Journal of Chinese Teachers Association 24(3): 49–65. Orton, J. (2013) Developing Chinese oral skills – a research base for practice. In I. Kecskes (ed) Research in Chinese as a Second Language. Mouton de Gruyter: Berlin, 9–31. Orton, J. (2014) Chinese as an Object of Learning. Paper presented at CASLAR’s Third International Conference on Chinese as a Second Language Research, 28–30 August 2014, Parma, Italy.

117

Jane Orton Stewart, V. and Wang, S. C. (2005) Take Both Short- and Long-Term Approaches to Create a Supply of Qualified Chinese-Language Teachers. Expanding Chinese Language Capacity in the United States. Washington: Asia Society. Retrieved 29 March, 2015, from http://www.asiasociety.org/files/ expandingchinese.pdf Triskova, Hana (2008) The sounds of Chinese and how to teach them. Archiv Orientalni 76(4): 509–44. Retrieved 29 March, 2015 from http://www.orient.cas.cz/miranda2/export/sitesavcr/data.avcr.cz/humansci/ orient/kontakty/pracovnici/publikace/Triskova/sounds.pdf United Kingdom Government Department for Education (2011) Teachers’ Standards. Retrieved 30 March, 2015, from http://webarchive.nationalarchives.gov.uk/20111218081624/http://tda.gov.uk/ teacher/developing-career/professional-standards-guidance/downloads.aspx Wang, Y., Jongman, A., and Sereno, J. (2006) L2 acquisition and processing of Mandarin tone. In P. Li, L. Tan, E. Bates, and O. J. L. Tzeng (eds) The Handbook of Asian Psycholinguistics. Cambridge: Cambridge University Press, 250–6.

118

8 Chinese language in a global context Liu Jin Georgia Institute of Technology, USA

Tao Hongyin University of California, Los Angeles, USA

1. Introduction As a language spoken by about one fifth of the world’s population (over a billion people), Chinese has always been in contact with other languages and undergoing changes. There is perhaps no better time than the beginning of the twenty-first century to witness the interlocking relation between Chinese and other languages, as the process unfolds in front of our eyes. In this chapter, we undertake the task of describing Chinese in a global context. We concentrate particularly on how globalization has impacted Chinese and how Chinese participates in the process of globalization. After a brief historical account of the contact of Chinese with other languages, we focus on the ongoing impact of English as a global language on Chinese and the perceived crisis involving Chinese. We then summarize a paradoxical phenomenon, namely the rapidly ascending status of the Chinese language worldwide. In the context of globalization, English is often perceived as the global hegemonic language, suppressing and marginalizing other national languages, including standard Mandarin, as local languages. Yet, we will show that the dynamics between the global and the local are complicated by the tension between the national language and the subnational, local dialects in this multilayered linguistic hierarchy. Next we explore the dominance of standard Mandarin and the impact of nationalization and modernization on the Chinese regional varieties. From this perspective, we show that the function of the nation-state seems more and more aligned with globalization and its concomitant homogenization and centralization. Next we illustrate that a variety of diverse, dynamic forces, including the impact of the English language, the increasing popularity of Internet-mediated communication, and the tremendous economic, social, and cultural changes China experiences in its modernization and integration with the world, all give impetus to a series of radical changes in the Chinese language. Finally, we document the emergence of the new linguistic forms and explore their theoretical implications along the axes of standardization and deviation, stability and mobility, and normativity and creativity. 119

Liu Jin and Tao Hongyin

2.  A historical perspective If history is of any guidance, the Chinese language is elastic and adaptive. In the long history of the Chinese language, there have been numerous times where rapid changes took place in the forms of expansion of vocabulary, development of new grammatical markers and syntactic structures, appearance of new phonological features, and emergence of new writing styles. Typically, these rapid and sometimes radical changes correlate with heightened interaction between indigenous and non-Chinese cultural elements. Thus, during the late Han period (second century ce) through the Tang dynasty (tenth century ce), the introduction of Hindu Buddhism brought to China not only Buddhist doctrines and practices, but also an expansion in vocabulary (such as loanwords from Persian, Sogdian, Sanskrit, Mongolian, Manchu, and other Asian languages, see M. Gao and Liu 1958, esp. Ch. 2) and novel syntactic features (as exemplified by the disposal construction, see Cao and Yu 2000). The best-known example, however, belongs to the period between the late nineteenth century and the early twentieth century, which led to the baihua vernacular movement in the May Fourth period. During this period, severance of ties with traditional Chinese culture (as represented by the Confucian tradition) facilitated by major influxes of Western ideas and thoughts resulted in a near total reconfiguration of the landscape of the Chinese language. Not only were writing styles changed, literary forms invented, and syntax increasingly Europeanized, even some basic vocabulary forms and a large number of neologisms were added (especially SinoJapanese-European loanwords) (Gunn 1991; L. Liu 1995). Into Mao Zedong’s era, partly following the former Soviet Union’s model in symbolic and ideological control, the Chinese Communist Party, after it won power in 1949, launched a massive revolutionary program of linguistic engineering, formalization, and orthodoxization, which culminated in the Cultural Revolution (Schoenhals 1992; Wagner 1999; Ji 2004; B. Wang 2011; and Link 2013). This seems an especially apt piece of evidence for Bourdieu’s prediction: Political crises are conducive to verbal explosion (Bourdieu 1977: 663). The beginning of the twenty-first century is remarkable in many ways: Now we are at the start of yet another period of historical significance. As China experiences an unprecedented level of economic and social transformation since its reform and opening-up policy in the late 1970s, and as the world becomes increasingly globalized and interconnected through the use of technology, freer flow of information, and the spread of cultural artifacts, language use in China has also undergone dazzling changes in the past few decades. At the global level is the clash between the ubiquitous global English and the Chinese national language. Within China, the struggle between the national language and regional dialects and ethnicminority languages at multiple levels plays out partly as a matter of nationalization versus the preservation of local identities and heritages. There is also the emergence of the Internet language and other new linguistic forms. Many of them have become creative sources for constructing a distinct identity for the Chinese youth. Moreover, it transcends locality and redefines key theoretical notions in sociocultural linguistics such as the speech community (Gumperz 1968) and linguistic landscape (Landry and Bourhis 1997).

3.  The global and the local: The English craze and the Chinese crisis While there are many major languages in the world with considerable political, cultural, as well geographical clout, English remains the only global language in every sense of the word (Crystal 2012). It is not surprising, then, that English has exhibited profound impact on the Chinese language and the society in which it is used. 120

Chinese language in a global context

Under the reform and opening-up policy, China witnessed a national obsession with learning English, a language denounced as bourgeois and imperialist yet manipulated to perpetuate Maoist worship during the Cultural Revolution. Today, English is viewed as a necessary tool in China’s modernization and integration with the world. For an extended period of time, the subject of English was listed as indispensable and mandatory by college and postgraduate entrance exams. Most elementary schools in urban cities have started to offer English classes. The English as a Second Language (ESL) industry has boomed in the past few decades. According to sources from the National Education Development Statistical Bulletin, the English-training industry in China had an estimated market value of 15 billion yuan (US$2.5 billion) in 2006, and the value doubled to 30 billion yuan (US$5 billion) in 2010 (He 2010). More and more affluent parents can afford to send their children to study abroad for college and increasingly for high school. According to the Institute of International Education, more than 194,000 Chinese students were enrolled in US colleges and universities in the 2011–12 academic year, far exceeding any other country outside the USA. The New Oriental School (Xindongfang xuexiao), which began in 1993 as a program to help Chinese students with the US-based TOEFL and GRE tests, has grown into the largest private language education enterprise in China. The school’s success even became the base storyline for a 2013 domestic hit film, Zhongguo hehuoren (American Dreams in China). The spread of the English language is both a resource for modernization and a source of contention. Chinese people, like others across the globe, have criticized the global dominance of English for suppressing other languages and monopolizing the world’s information. Chinese intellectuals, such as the writer Li Rui (2000), have expressed consternation that the Internet would encourage the global dominance of English and marginalize all other languages. The notion of a ‘Chinese Crisis’ even became the title of a book in 2008, in which Pan Wenguo, a linguist at East China Normal University, outlined five symptoms of Chinese in crisis, the top one being the overwhelming use of English on the Internet and the accelerated cultural globalization mediated in English (Pan 2008; see also Zhu 2005). Although the hegemony of English online has gradually weakened with the proliferation of Chinese-based sites and user bases, the debate between these academics and their opponents remains a hot topic in the media and public discourse. For example, Zhang Shuhua, a dean at the Chinese Academy of Social Sciences and a member of China’s top political advisor body, criticized that the excessive craze for learning English has wasted huge education resources, distracted students from their subject studies, jeopardized educational quality, and weakened students’ grasp of the Chinese language and culture. As he observed, many applicants have been denied admission to postgraduate programs solely due to their failed English tests; English is also an obstacle for working professionals to overcome if they are to get a promotion or salary raise, even though they barely use the language in their workplace. His remarks at a top national political meeting sparked heated debate among the public and triggered nearly 90,000 online comments within just two days (Xinhuanet 2013). Largely due to this resentment, the Chinese government has recently decided to downgrade English scores on the highly competitive national college entrance examination. Of course, this is not to suggest that the Chinese public is going to boycott English any time soon. On the contrary, many shrewd businessmen take the English frenzy as a window of opportunity and act upon it in clever ways. Li Yang, an engineer by training, cashed in on Chinese nationalism in his ‘Crazy English’ English education business, with his slogan ‘conquer English to make China stronger!’ (Woodward 2008, 2012). Other symbolic business changes abound. For example, it has become a common practice for landmark Chinese business entities to adopt an English identity, with one of the more famous cases being the 121

Liu Jin and Tao Hongyin

Tianjin-based restaurant chain that specializes in steamed stuffed buns, Gou Bu Li (lit., ‘dogs don’t bother’), officially adopting ‘Go Believe’ as the store’s English brand name (Meiri xinbao 2008).

4.  The global spread of Chinese While English is the undisputed world language, whether or not Chinese is rising to the status of a global language can be debated. However, there is no denying that the rising profile of China, backed by considerable economic, military, and political power, has popularized the Chinese language across the globe. As Lo Bianco (2012) puts it, Chinese has become a ‘gigantic up-and-comer’. This is of course in stark contrast with the perceived Chinese language crisis that was much discussed in the early 2000s. The numbers are quite striking. It is estimated that there are currently 30 million people around the world learning Chinese as a second language. According to statistics released by the Chinese Ministry of Education, the number of foreign students in China reached a record high of more than 260,000 in 2010 (Chen 2011). Based on the annual report by the US-based Institute of International Education (IIE), the number of Americans studying in China increased by 25% in the 2006–7 academic year over the previous year. Advanced Placement (AP) Chinese exams for US high school students were offered for the first time in 2007. The US government sees Chinese as one of its ‘critical languages’ for security reasons and has thus dedicated enormous resources to Chinese language learning. The Language Flagship Program is one of the federal government’s initiatives to produce professionally proficient speakers of languages critical to the national future. Of the 26 centers and programs, there are 12 Chinese programs, including three pilot Reserve Officers’ Training Corps (ROTC) Chinese programs launched in 2011. Accordingly, the enrollment in the Chinese programs (403 students) accounted for nearly half of the total enrollment in all of the flagship programs (932 students) in 2010. Similarly, the Chinese government sees the surge of interest in Chinese as a unique opportunity to promote Chinese culture and identity across the globe. In 1987, the Chinese government established the Office of Chinese Language Council International (Hanban), which is charged with the task of establishing Confucius Institutes and supporting Chinese language and cultural programs around the world. The institutes are modeled after governmentbacked programs such as France’s Alliance Française, the British Council, and Germany’s GöetheInstitut. While the first Confucius Institute opened in Seoul, South Korea, in 2004, as of December 2014, 475 Confucius Institutes had been established in 93 countries and regions. The rapidly ascending status of the Chinese language has brought nationalistic pride to many Chinese people. With the establishment of large numbers of nondegree and degree programs in teaching Chinese as a second language at Chinese universities and the mushrooming development of language-teacher training programs and private Chinese languagerelated enterprises, the twenty-first century has sometimes been hailed the century of the Chinese language. One of the most popular Chinese songs in 2007 was ‘Zhongguohua’ (‘The Chinese Language’), sung by the pop trio S. H. E. As the lyrics go: ‘The whole world is now learning Chinese; the language of Confucius is going global’. Nevertheless, in her critical reflection on Chineseness as a monolithic, culturally essentialized ethnic supplement to Western hegemony, Chow (2013a: 48) relates the problematic claim of a homogeneously unified, univocal China to the myth of ‘standard Chinese’, which has been affirmed in the pedagogical dissemination of the Chinese language(s) in the West. As she observes, the learning of the Chinese language has often been reduced to the learning of standard Mandarin in the United States, and those professionals who are non-Beijing 122

Chinese language in a global context

Mandarin native speakers, particularly Southern-dialect speakers, have often been discriminated against in the screening of Chinese teachers. The next section will take the multiple Chinese dialects and regional varieties into consideration. We explore the tension between the local and the national and the issues of Chinese language(s), Chineseness, and China in the context of globalization.

5.  Impacts of nationalization and standardization on dialects In linguistic discourse on contemporary mainland China, a distinction is often made between standard Mandarin, or Putonghua (lit. ‘common speech’), and nonstandard local languages and dialects, or fangyan (lit. ‘regional speech’). The classification of these local dialects is still debated, but seven major fangyan groups are conventionally recognized: Mandarin (guanhua or beifanghua), Wu, Min, Cantonese ( yue), Gan, Xiang, and Hakka (kejia); each can be further divided into various subvarieties. Chinese local languages and dialects have long been a fundamental feature of Chinese everyday life and popular culture. The essence of fangyan has been variously identified and characterized as the living, vernacular, or oral language; regional speech; one’s mother tongue; folk language; rural patois; the speech of the uneducated; vulgar slang; unofficial subcultural lingo, and so on. During the twentieth century, they were associated with and simultaneously dissociated from the historical project of building the Chinese nation-state. Although dialects were valued and promoted at various historical moments, building a unified, modern, national language remained the paramount and overarching concern for China in its quest for modernity. The central government has promoted standard Putonghua as the official national language and the principal language for mass media and school education. Consequently, the local varieties have been marginalized and subordinated during this ongoing process of building a modern nation-state, a national culture, and a national language. On October 31, 2000, the Beijing government promulgated its first law on language and writing, the Law of the People’s Republic of China on the Use of Chinese Languages and Chinese Characters (Zhonghua renmin gongheguo guojia tongyong yuyan wenzifa), effective as of January 1, 2001. The law prescribes Putonghua as the principal language for broadcast media, movies, school education, and administration, and it strongly discourages the public use of all other local languages and dialects (except in the non-Han Chinese regions). One of Bourdieu’s (1991) main theses is that the educational system as an institution plays a decisive role in the standardization, legitimization, and imposition of an official language. Kipnis (2012) also argues that educational standardization, including the school use of standard Mandarin by both teachers and students, is an important part of the nationbuilding in contemporary China. Indeed, through formal education, the mass media, and increasingly through administrative measures, the promotion of standard Mandarin has been largely successful. According to an official national survey result on language use in China released in 2004 (Zhongguo yuyan wenzi shiyong qingkuang diaocha), approximately 53% of the population is able to communicate in Putonghua, with urban areas more successful in implementation than rural areas (urban penetration is 66% and rural penetration is 45%) (Yuyan kexue 2005: 112). Even in Hong Kong, where Cantonese has remained predominant, according to the results of the population census in 2011 (14 years after the handover in 1997), about 46.5% of the population claimed to be able to speak Putonghua. While official documents attest to the importance of protecting regional linguistic and cultural heritages, the promotion of Putonghua has led to a sharp decline in the use of dialects, particularly among the urban, school-educated youth. For example, in a survey on 123

Liu Jin and Tao Hongyin

the use of the Suzhou Wu dialect among students in grades two through eleven – 90% of whose parents are native Suzhounese – more than 70% of the students claimed more fluency in Putonghua than in Wu (P. Wang 2003). In another survey on the use of the Zhejiang Jinhua Wu dialect among six- to 14-year-olds, as many as 52% said they could not speak their native dialect, though virtually all could speak Putonghua (jhnews.com 2003). In Shanghai, only about 60% of elementary and middle-school students who were born in Shanghai could speak Shanghai Wu in a survey conducted by the Shanghai Academy of Social Sciences in 2012 (Tian 2012). Forty years ago, this would have been unthinkable. Clearly, the longstanding tradition of diglossia, the use of more than one linguistic variety in a community, has tilted toward the dominance of one single language. For many children born in the 1990s and later, it is no longer their (grand)parents’ dialect but Putonghua which has become their mother tongue. Yet, as is the case with the reaction to the predominance of English, the national domination of Putonghua has not been imposed without challenge in China. Despite the 2001 language law, local media and literary genres produced in regional dialects have flourished in recent years. In television, dialect-based news talk shows that draw on the arts of local traditional story-telling are burgeoning; popular shows have also been localized, as in 2005 when the cartoon series Tom and Jerry was dubbed into around 20 dialects nationwide. In film, internationally renowned director Jia Zhangke’s Hometown Trilogy, largely in Shanxi Mandarin, ushered in a wave of underground and independent films employing dialects to represent the aesthetics of ‘the marginal and the unassimilated’. On the Internet, one can find rap music in dialects, tests on dialect competence mimicking the formats of official English exams, dialect texts parodying Chinese characters and the writing system, blogs employing local slang and expressions, and downloadable mobile ringtones recorded in dialects. In fiction, a number of established writers who previously adhered to standard Mandarin, such as Mo Yan, Jia Pingwa, Yan Lianke, and Zhang Wei, have begun to experiment with writing novels in their native dialects. In the post-socialist reform period, as the aesthetic, entertainment, and commercial value of regional dialects is (re)discovered, the tension between the state, capital, and art is intensified. The rise of regional television shows rendered in dialects, the proliferation of dialects on the Internet, and the commercial success of dialect cultural productions in the local market all attest to the urgency of reimagining a distinct local community that cannot be adequately represented by a single national language (J. Liu 2013). As the nation-state becomes increasingly homogenized and centralized, the use of dialects in mass media and in public spheres asserts the values of pluralism and diversity, and defies the characterization of China as a unified and homogeneous nation-state. The multiplicity of the Chinese dialects is directly involved in the recent academic trend of engagement with and rethinking of the issues of Chinese language, Chineseness, and China. The Chinese language is not a homogeneous entity. Numerous local dialects and varieties are subsumed under this single name and exhibit various degrees of diversity in different areas. While the speakers of the northern Mandarin varieties can largely communicate with each other despite the tonal and lexical differences, linguistic diversity (especially in phonology and lexicon) is much more pronounced in the south, especially in the Min-speaking areas. The multiplicity of Chinese local languages has led to much discussion over how the term fangyan should be translated into English. DeFrancis (1984) discusses the terminological dilemma created by the unique linguistic situation in China. He thinks neither ‘dialect’ nor ‘language’ can justifiably convey the ambiguities and obscurities attached to fangyan. Mair (1991) further examines the problem of the prevalent translation of fangyan as ‘dialect’, an issue he deems 124

Chinese language in a global context

as ‘extraordinarily sensitive’. He develops DeFrancis’s ‘regionalect’ and proposes the term ‘topolect’ as a literal translation of fangyan. Claiming a departure from the umbrella term ‘world/overseas literature in Chinese’ (shijie/haiwai huawen wenxue), the new discipline of Sinophone studies disputes the common practice of equating the Chinese language(s) with Mandarin (or more precisely, standard Putonghua/Guoyu/Mandarin) and defies overgeneralization and essentialism in the assessment of China and Chineseness. Taking Mair’s 1991 article as a major reference, Shih (2007: 189) claims that those southern fangyans that are most frequently the subject of Sinophone studies, including Cantonese, Hakka, and the Southern Min varieties such as Amoy (Xiamen) Min/Taiwanese, Zhangzhou Min, Quanzhou Min, and Teochew (Chaoshan) Min, are not dialects but ‘clearly separate languages from Putonghua’. In fact, the controversy over the relationship between dialect and language is a global and often politicized problem (Hobsbawm 1992: 51–63, 93–100; Haugen 1966). Linguistically speaking, the differences among Chinese fangyans are often conceived to be analogous to those that distinguish the Romance languages in Europe. However, the identification of the former as dialects and the latter as languages is never a purely linguistic issue, and cannot be separated from politics, nation, culture, ethnicity, and other nonlinguistic factors. At the grassroots level, efforts to promote local dialects and to construct a distinct local linguistic and cultural identity have emerged across regions in various forms. For instance, dialect training workshops in the major Wu-dialect metropolises such as Shanghai, Hangzhou, Ningbo, and Suzhou have all been well received. Local newspapers run articles on strategies to preserve readers’ mother tongues against the ‘over-dissemination’ of Putonghua. In Shanghai, a supplementary textbook designed for elementary students was revised in 2005 to include a reading which briefly introduces the history of Shanghai Wu and its vocabulary (Y. Wang 2005). In 2013, roughly 100 elementary and middle schools in Shanghai started to offer Shanghai Wu classes (Y. Liu 2013). Efforts such as these have spurred local legislators in various cities to launch municipal-level initiatives to ‘rescue and protect’ local dialects as ‘endangered cultural heritages’. In the summer of 2010, the pro-Cantonese protests and demonstrations against restrictions on Cantonese in the media, first in Guangzhou and soon echoed in Hong Kong, have been viewed not simply as a linguistic action (Waldron 2010). According to Eng (2010), ‘the size and fervor of the pro-Cantonese movement were derived from well-founded fears that Cantonese was losing ground, and also from legitimate sources of popular dissatisfaction beyond the realms of language policy: collective concerns about the progressive disappearance of Cantonese culture and social dislocations resulting from massive urban renewal and construction’. Just as the title of a popular rap song during the protest suggests, ‘Everything Is Being Dismantled; Cantonese Must Not Be Dismantled’ 乜都拆, 广州话唔可以拆 (‘Mat dou caak, gwongzauwaam hoji caak’, 2010). Nevertheless, the reaction to the protection and revival of regional dialects has been decidedly mixed. In contrast to Cantonese-, Min-, and Wu-speaking areas, which are largely well-developed coastal regions, localities in less-developed regions, which are mostly within or near the Mandarin-speaking sphere, are lukewarm to, if not totally detached from, the dialect crisis. Furthermore, the public use of dialect has always been controversial. A frequently debated topic is how to deal with the relationship between the use of dialect in the media and the overriding state policy of promoting a single national language. For instance, although Lu Chuan, the director of the hit studio film Xun Qiang (Missing Gun, 2002) that employs Guizhou Mandarin, highlights the aesthetic effect of dialect that can capture unique acoustic texture and modulation, a government official accused Jiang Wen, the leading actor of the film, of countering the national language policy and bringing to the 125

Liu Jin and Tao Hongyin

big screen the ‘dregs’ of the dialects (Yin 2002). In addition, the concern for education also stirs controversy concerning the use of dialect in public. From time to time, parents and teachers complain that the media use of ‘vulgar and uncultured’ dialects would be counterproductive to children’s learning Putonghua (in other words, speaking Putonghua is equated with being civilized). However, the dialect promoters often contend that it is equally important to pass on local cultural heritage transmitted in dialect down to the younger generation. The profusion of local languages in mass media has caused the authorities considerable concern. With increasing frequency, the State Administration of Radio, Film, and Television (SARFT) has issued or reiterated regulations to contain the media’s use of dialect and to promote a pure, normative Mandarin. However, this agenda has been continuously challenged by tides of new vocabulary and linguistic innovations that have swept through Chinese society, as the next section shows.

6.  Emergence of new linguistic forms 6.1.  Internet language and youth language Ever since its introduction to China in the early 1990s, the Internet has been developing very rapidly. As of January 2010, according to a state-sponsored study by the China Internet Network Information Center (CNNIC), there were 384 million netizens in mainland China. Various computer-mediated communication formats include emails, e-chatrooms, short message service (SMS) text messaging, BBS forums, social networking websites and smartphone-aided social media (such as renren, weibo [microblog], weixin [wechat]), and other electronic genres. The Internet is primarily a youth culture in China, as in many other countries. According to the same CNNIC study, more than 70% of the Internet users were young people under 35 years of age, and the age group between 18 and 24 years old had consistently accounted for a much higher portion (usually 35–42% between 2000 and 2009) of users than any other age group. Regarding their level of education, approximately half of China’s netizens had a college or associate degree. Roughly one third of China’s netizens were currently students. The rapid development of Internet communication has created a considerable impact on the Chinese language. Researchers have documented that computer-mediated communication (CMC) has helped give rise to a new variety of Chinese language, which generally terms as Chinese Internet language, also rather amusingly known as huoxingwen (language from Mars) (L. Gao 2008, 2012). Young people are the most active creators and trendsetters of the Internet language. The creativity of the Chinese Internet language is mainly manifested at the lexical level. Some common methods to coin new words and expressions include near homophone or phonological approximation (e.g. 神马 shenma for 什么 shenme ‘what’, 有木有 youmuyou for 有没有 youmeiyou ‘do you have  .  .  .  ?’), compounding (e.g. 给力 geili, lit. ‘give power’, now usually ‘cool, awesome, exciting’), phonological fusion (e.g. 酱紫 jiangzi from zheyangzi 这样子 ‘like this, this way’), semantic shift (e.g. 腐败 fubai originally ‘corruption’ and now ‘dine out’), emoticons (e.g. 囧 jiong for ‘embarrassing, in a bad mood’, an esoteric character originally meaning ‘bright [light]’), acronyms (e.g. LP for ‘wife’), script-mixing or script-switching (e.g. I服了U I fule U ‘You’re awesome’), letter and/or number homophones (e.g. 886 ‘bye-bye le’), word decomposition (e.g. 蓝介 lanjie from 尴尬 ganga for ‘awkward, embarrassing’, 竹本犬者 zhubenquanzhe from 笨猪 benzhu for

126

Chinese language in a global context

‘stupid [pig]’), regionalisms (e.g. 俺 an for the first-person pronoun ‘I’, from Shandong and Henan dialects), translation of English words into Chinese characters (e.g. 粉丝 fensi from ‘fans’, 闹太套 naotaitao from ‘not at all’), coinage of English translation for the newly coined Chinese words (e.g. ‘gelivable’ for geili, ‘shitizen’ for pimin [fart people, the common ordinary people]), and so forth. Similar to the Internet language elsewhere, the styles of Chinese Internet language are marked by: (i) a written–spoken hybrid style, where many sociolinguists regard the computermediated discourse as a hybrid register that integrates features of both oral and written discourse (Frehner 2008: 26, 31–5), or as the third genre on the spoken–written continuum (Crystal 2006: 52; Gibbon and Kul 2010: 76); (ii) brevity, which adheres to the Principle of Clarity and Principle of Least Effort, two major principles explaining the dominant economic strategies adopted in digital communication (Gibbon and Kul 2010: 80); (iii) bilingual or multilingual code, which is a distinctive feature of Internet communication; and (iv) a joking style, ‘which is usually achieved through using overly terse and straightforward language, unusual homophones, enigmatic folk similes, and other metaphorical devices’ (L. Gao 2012: 20). The playfulness can be subversive. The Chinese youth challenge and subvert the Chinese characters and the writing system by playing with the characters in transcribing their native dialects online. For example, in using 椅弯撵 yǐwānniǎn for the Shaanxi Mandarin version of 一万年 yíwànnián in Zhou Xingchi’s popular line in his film The Chinese Odyssey, the Chinese character is deprived of its semantic reference function and reduced to a phonetic symbol (J. Liu 2011). It is also noted that sometimes the youth intentionally write incorrect characters as a way to signify a distinct youth identity. Researchers show how misspellings or ‘spelling errors’ are employed symbolically to construct a rebellious youth identity against the mainstream and adult culture, and yield a tension between creativity and prescriptivism (Sebba 2003; Leung 2009). For example, the Chinese Internet youth deliberately write the word 悲剧 (bēijù ‘tragedy’) using the wrong characters 杯具 (bēijù ‘cup set’) in the widely circulated line 人生像茶几,上面摆满了杯具 (Life is like a tea table, all covered in cups [punning as ‘tragedies’]). Internet censorship in China, represented by the Great Firewall of China (or Jindun gongcheng, literally ‘the Golden Shield Project’), has fueled a surge of coded language. In 2009, as an impish protest against the anti-vulgarity campaign which the government launched to tighten political control of the entertainment industry and the Internet, the Chinese netizens coined an innocent, mythical animal ‘grass-mud horse’ (cǎonímǎ, a near homophone of càonǐmā, lit. ‘fuck your mother’), whose lives are threatened by the invasion of river crabs (héxiē, a near homophone of héxié ‘harmony’, a term directly derived from former President Hu Jintao’s regular exhortations for Chinese citizens to create a harmonious society) (Wines 2009). This highly subversive neologism has ever since become an icon of resistance to Internet censorship, and in 2010 the China Digital Times (CDT) launched a project called ‘Grass-Mud Horse Lexicon’ to compile the online resistance discourse. Young netizens’ subversion of the official discourse is very nuanced, packaged, and symbolic. The coded language and metaphors allow them to criticize and ridicule the government, vent out their resentment, and yet avoid outright censorship. Xiao Qiang and Perry Link (2013) argue that the coded language has powerful implications and significance. Taking guiguo (your [honorable] state) as an example, they argue that the netizens’ making distance between themselves and ‘your state’ entails a big question: ‘What do they identify with at the national level? What is it to be Chinese?’

127

Liu Jin and Tao Hongyin

6.2.  English’s impact on the Chinese language The English craze in the Chinese educational institution, the rapid development of the Internet and other high-tech electronic communications, as well as China’s pursuit of modernization and globalization, all contribute to the increasing impact of English, as the global language and the global lingua franca, on the Chinese language. Here are some new linguistic varieties exemplifying its impacts: ● zimuci (letter-word): foreign words (primarily English) and their abbreviations, and pinyin alphabet, written in alphabetic letters and not in Chinese characters, such as ATM, 三G手机 (third-generation mobile phone), and HSK (abbreviation of Hanyu shuiping kaoshi ‘Chinese language standard test’). Letter-words are arguably different from those loanwords that are transcribed into Chinese characters, such as 可口可乐 kekoukele ‘Coco-Cola’. The interference of the two topologically different languages, Chinese and English, is a central topic of interest for linguists. More specifically, the questions of how alphabetic letters integrate with the phonetics and syntax of the Chinese language; how to classify letter-words; what are the graphic and orthographic effects of letter-words on the ideographic Chinese writing; etc. (Kozha 2012: 110). At the discourse level, here are a couple of distinctive examples currently in vogue among the Internet-savvy youth: a) 你out了 ni out le ‘You’re left behind the times’; b) hold 住 hold zhu: ‘hold’ indicates ‘hold fast’, ‘hold your ground’, where the ending sound ‘d’ of ‘hold’ sounds and functions as the Chinese de in the potential complement construction, 住 means ‘live’, ‘stay’, and can be used as a result complement in the de construction, and the entire newly coined construction signifies perseverance, taking charge, being in command of a situation with self-confidence and conviction (Baidu Beat 2012); c) 我吃饭/大哭/ 开会/恋爱 ing wo chifan/daku/kaihui/lian’ai ing (the progressive aspect marker in English), usually in the present tense and meaning ‘I’m eating/crying/having a meeting/ in love’ (Kozha 2012: 118). ● Chinglish: Chinglish is usually defined as a depreciative usage of English by native speakers of Chinese. The increasing presence of public signs with English translations strongly suggests China’s eager engagement with the global tendency of advocating bilingual signs as a means of communication and a sign of social sophistication and cosmopolitanism (Radtke 2012: 151). However, due to the translators’ insufficient English competence, and increasingly due to the use of word-by-word machine translation programs, many public signs are ridiculously translated, and the errors range from simple typos to total gibberish. For example, 财务处处长 (caiwuchuchuzhang ‘chief financial officer’) was translated as ‘Financial affairs is everywhere long’; 残疾人厕所 (canjirencesuo ‘toilet for the handicapped’) was translated as ‘deformed man toilet’ (Hwang 2012: 7, 24). Chinglish may enjoy a slightly Orientalist reputation for the native speakers of English as an entertainment source. A 2011 Broadway play is entitled Chinglish, which was inspired by the playwright David Henry Hwang’s experience of visiting China. However, on the Chinese side, the official Commission for the Management of Language Use finds such mangled English humiliating and feels compelled to clean up English signage, especially prior to large-scale international events, such as the 2008 Beijing Olympic Games or the 2010 Shanghai EXPO (Jacobs, 2010). Global Times, a government organ, ran an article in 2010 titled ‘Chinglish Sentenced to Death’ (Fang 2010). Yet as Radtke (2012: 158–9) argues, for the majority of Chinese within China, the English words placed on billboards, road signs, stores, restaurants, menus, and 128

Chinese language in a global context

product labels may not be intended as a means of communication, but as a decoration for appearing bilingual to attract domestic consumers. In other words, these Chinglish signs are meant for ornamentation, not for information. A more intellectually informed analysis of Chinglish from the perspective of cultural studies can be found in Chow (2013b).

6.3.  Nonstandard Mandarin styles and varieties With the rise of the middle class and new elites in urban cities, the penetration of commercial and consumer-driven culture, and the influence of Hong Kong and Taiwanese popular culture on mainland China, innovative, nonstandard Mandarin linguistic styles and varieties have emerged to respond to the dramatic social and cultural changes and to mark distinctions. Zhang (2005) examines four linguistic variables used by the Beijing yuppies, those cultured cosmopolitans and enterprising professionals working for foreign businesses: rhotacization of the syllable rhyme, interdental realization of (ts), lenition of (sh), and full-tone realization of a neutral tone. She associates the first three local Beijing Mandarin variables with two well-known Beijing character types: the ‘Beijing smooth operator’ and the ‘alley saunterer’. The yuppies try to construct a distinctive cosmopolitan style from their counterparts in the state-owned enterprises, where professionals’ Mandarin carries a prominent Beijing accent. Zhang (2012) explores media hosts’ linguistic styles in consumer-oriented programming in the media, particularly the so-called shishang (trendy) television programs on fashion, consumption, and lifestyles. The hosts draw from linguistic sources, including sounds, words, discourse strategies, regional dialects, English, and other semiotic resources, including makeup, hairstyle, accessories, and clothing, to create a trendy and cosmopolitan persona. She argues that this new linguistic style plays a crucial role in the formation and dissemination of a new consumer culture and the education of the Chinese public to become consumers. Also on media language, Han (2012) examines metaphors used in contemporary enter­ tainment news. She identifies both the common source domains of metaphors (war, martial arts, fire, wind, food, etc.) and target domains (competition, conflict, celebrity, etc.). For example, the metaphor of ‘competition is war’ is illustrated by a news title about the TV entertainment show Happy Boy, quite similar to the American Idol show in format: ‘7 out of 9 candidates will go into the next round of Happy Boy. Today, the smoke of gunpowder is rising again’ (Kuaile Nansheng jiu jin qi xiaoyan jin zaiqi). She points out that metaphors serve multiple purposes: They cannot only embody abstract concepts but also serve as an ideological tool for describing and evaluating people and situations in discourse and influence the viewers’ perception of the world. Metaphors in use can thus serve as a window into the changing media ecology in contemporary China. Television, popular songs, and other media have played a key role in disseminating Hong Kong and Taiwanese popular culture to mainland China. The Cantonese- and Taiwaneseaccented Mandarin (gangtaiyin) that have penetrated the mainland through popular culture since the late 1970s are viewed as ‘cool’ and ‘trendy’ and are avidly imitated by young people in mainland China. Tang (2001) details the words and phrases from Taiwan and Hong Kong appearing in print in mainland Chinese publications. It demonstrates that during the 1980s and 1990s the vast majority of traffic was one-way, from Hong Kong and Taiwan to mainland China, and not vice versa. The soft- and tender-sounding gangtai Putonghua, perceived to be a perfect language for romance, has become the predominant speech style to imitate in mainland drama productions featuring youth idols. 129

Liu Jin and Tao Hongyin

6.4.  Socially embedded language Rapid social changes and network technologies have prompted new words, new expressions, and new structures to emerge literally every day to reflect the dramatically changing social settings in which the language is used in contemporary China. These changes can be outlined as follows. ● Semantic shift/extension: A considerable number of existing words have acquired new meanings. For example, both the addressing terms xiaojie (Ms.) and tongzhi (revolutionary comrade) became sensitive words in post-socialist China, with the former referring to prostitutes and the latter to homosexuals. In a similar vein, laoshi (teacher) has extended its function to now serve as a respectful marker after a person’s (sur)name, regardless of whether or not the person is a teaching professional. ● New ‘social words’: Neologisms are created to indicate social phenomena and events and to signify changing values, identities, and social meanings. To give a few examples, there are an increasing number of accomplished women who are well educated, well employed, but still single at a certain age, and they are called shengnü (leftover women). Fu’erdai and guan’erdai are terms designated for those young people who are the second generation of the rich and powerful. The frequent use of the terms, which are often associated with many negative stereotypes, indicates the growing gap between the rich and the poor and the growing disdain or even hatred of the super-rich and the powerful. Reflecting the social issue of forced property demolitions, the word dingzihu (nail resident) was coined to refer to those who refuse to move mainly because of the unsatisfactory compensation. It is so common to see the character of chai (demolish) painted on walls across the country, prompting some netizens to give the English word ‘China’ a new literal translation as chai na 拆那. ● Grammatical construction: Grammar is usually the least prone to change. So far the best example is the expanded use of the bei construction. bei is a preverbal passive marker for verbs that usually indicate suffering, misfortune, or adversity, and the construction conveys a usually negative, inflictive connotation. For example, wo che bei zhuang le (my car got crashed/got an accident). However, more and more nonpassive verbs, intransitive verbs, and even adjectives appear in this bei construction in contemporary Chinese. For instance, bei daibiao ([one’s opinions] being represented), bei hexie (being harmonized/censored), bei juankuan (being forced to donate), bei zengzhang ([income] being increased in statistics), bei jiuye (exaggerated report about graduates’ employment), bei zisha (a lawsuit settled by claiming to be a suicide case), and bei xingfu (being happiness-ified) are all commonly found on the Internet. Drawing on the construction’s ‘involuntarily passive, and compulsory’ implication, these expressions serve to reflect the absence of subjectivity and civil rights in contemporary China on the one hand and to express the frustration of being underprivileged and powerless in a very much top-down system found in China on the other.

7.  Theoretical implications of the new forms 7.1.  A reaction to the single, standard Putonghua The Beijing novelist Wang Shuo (S. Wang and Lao 2000: 209) points out that, in the contemporary post-socialist period, standard Putonghua in mainland China has evolved into an overly politicized language, featuring a hollow, exaggerated emotionalism, a harsh 130

Chinese language in a global context

and aggressive rhetoric, and a standard, authoritative ‘broadcasting intonation’. The linguist S. Robert Ramsey (1987: 47) aptly describes this ‘broadcasting intonation’ as: ‘All radio Peking announcers, both men and women, broadcast in a pitch range noticeably higher than that of their normal speaking voices. Each sentence begins high and shrill. Then pitch falls gradually, reaching a lower key by the end of the sentence. Pauses are exaggerated, and the normal rise of a nonconcluding clause becomes longer and more drawn out. The devices of this strident intonation may well be borrowed, in part, from traditional Chinese drama and opera; but their use in the media today seems intended to arouse in the audience an impression of struggle and determination.’ Similarly, the Yunnan poet Yu Jian (2004: 137) finds that when he speaks in Putonghua, he becomes a different person who ‘has no sense of humor and is self-abased, nervous, stuttering, and pretentiously serious’. He argues that from the perspective of sociolinguistics, Putonghua is not just a tool for nationwide communication, but rather a ‘social dialect’ that is highly politically charged. This ‘social dialect’ is best suited to mass mobilization; metaphysical spirituality; abstraction; central state-sanctioned ideological and literary orthodoxy; propagandistic eulogy in the public sphere; grandiose, heroic, and utopian narrative; formal diction and power; and revolutionary discourse. In this sense, the emergence of diverse linguistic styles, registers, nonstandard accents, dialects, and features outlined in the previous sections can be seen as a reaction to the single, official, standard Putonghua.

7.2.  The tension between mobility vs. stability and creativity vs. normativity Similar to the tension between the expanded use of dialect in the media and the tightened SARFT regulations, the emerging linguistic forms participate in a linguistic battle with the authorities and formal institutions as well. The trend of the popularity of gangtai accents among the mainland youth led to a SARFT regulation in 2005, stipulating that broadcasting hosts should always use standard Mandarin and should stop adopting Hong Kong or Taiwanese slang and accents. The language censors are also worried that the expanded use of English words and Roman letters is sullying the purity of the mother language, Chinese. In 2010, the General Administration of Press and Publication (GAPP), China’s regulator of news, print media, and Internet publications, stipulated that standardized Chinese should be the norm: The press should avoid foreign abbreviations and acronyms, as well as Chinglish. In his influential article entitled ‘Questioning the “standardization of modern Chinese language”,’ the linguist Qian Nairong (2004) argues about the constantly changing nature of language and the impossibility of language standardization and purification. The accusation against nonstandard usages represents a romanticist view of language purity which cannot be upheld in modern society and would make Putonghua a stagnant, minimalist, artificial, and dead language. He calls for an open and liberal mind to approach the emergent new linguistic varieties and to foster linguistic multiplicity and diversity. Zhang (forthcoming) incisively contends that underlying such regulations is a strategy of de-legitimization and ‘othering’ by the government and official institution, positioning ‘nonstandard’ accents and writings as inauthentic and thus ‘wrong’. In this way, Putonghua, based on Beijing Mandarin in the north, is construed as the center of the Chinese linguistic hierarchy, enjoying hegemony and prestige. So fundamentally this is a dynamic struggle between normativity and creativity, standardization and deviation, the center and the periphery, the official mainstream discourse and the unofficial popular discourse. The linguistic battle over enregisterment 131

Liu Jin and Tao Hongyin

reveals that what is really at stake is the location of the cultural center of modern Chineseness. The state’s project of a modern, globalizing China firmly locates the center on the mainland, whereas alternative visions of modern Chinese value transregionality, hybridity, and diversity. Nevertheless, language is in constant flux. Some of the new usages may be transient, while some may evolve into the core of the Chinese language and signify diachronic linguistic change in progress. As noted, at least three Internet words have been included in the 5th edition of Xiandai hanyu cidian (Dictionary of Modern Chinese, 2005), one of the most authoritative dictionaries on the modern Chinese language: yimeier ‘email’, ku ‘cool, awesome’, and heike ‘hacker’. The Internet buzzword in 2010, geili (lit. ‘giving power,’ meaning ‘cool, awesome, exciting, cooperative’), was granted the ‘official seal of approval’ by appearing in People’s Daily, the official paper of the Communist Party (Shanghai Daily 2010). In addition, the latest 2010 edition of the Oxford Chinese–English, English–Chinese Dictionary, the most widely used Chinese–English reference book, included trendy lingo such as shanzhai (‘cheap copy’, like a knock-off mobile phone), fangnu (mortgage slave), and shengnü (old maid) (The Independent 2010). In terms of grammar, the use of nouns as adjectives has been documented in Xiandai hanyu cidian (2005), for example, yangguang ‘sun’ as ‘healthy and optimistic’. In summary, most linguistic innovations can, at least partially, be attributed to mobile and network technology. While technology is not the only factor giving rise to the large volumes of linguistic innovation, it plays a vital and integral role in the process. A strong corollary of mobility is creativity: New technologies create new space and new communication modes which were not possible in the past and which foster user participation and fuel creativity. On the other hand, there is always the tension between creativity and normativity, where the forces to preserve regularity and tradition are at work. The interaction among these parameters, mobility and stability, creativity and normativity, and deviation and standardization, has resulted in a number of features characterizing the language of the new millennium and at the same time has created significant tensions involving linguistic identity, social commentary, and cultural politics.

8. Conclusions When the Chinese language is viewed in the context of globalization, we are presented with an extremely dynamic and complex landscape of language use and change. Out of the complexity, however, some major patterns emerge: 1. Language contact has been a constant phenomenon in the history of Chinese. What is really remarkable about contemporary Chinese under globalization is the extent of the contact with other languages (chiefly English) and the degree of impact that has been incurred on Chinese. 2. While Chinese often plays a receiver role in the global stage of language interaction, it simultaneously spreads its own influence behind a rising economic, political, and increasingly military superpower. This is just another reflection of the multitude of world powers that characterize the world scene in this age of globalization. 3. Language as a symbolic system is always shaped by factors both inside and outside itself. In the case of globalization, such extralinguistic factors as social–cultural identity, tradition and progress, conservation and creativity, technological advancement, and so forth, all prove to be key in defining what to change and how changes take place. 132

Chinese language in a global context

4. While we can perhaps never predict the end result, if there is ever such a thing as an end result, we can be certain that the Chinese language will continue to evolve as it interacts with the rest of the world languages and negotiates status among its subnational varieties. The key is to identify the underlying factors and understand how they impact the way the Chinese language evolves.

Note This chapter synthesizes our work on this topic with additional materials and observations. Most of the previous work can be found in our co-edited book, Chinese under Globalization: Emerging Trends in Language Use in China (J. Liu and Tao 2012). More specifically, Section 2 is from pages 1–2 and Sections 3–5 is the revised and updated version of pages 204–10 in the volume. We wish to thank Like Li for her meticulous editorial assistance with a draft copy of this chapter. However, we are solely responsible for the final shape of the chapter.

References Baidu Beat (2012) ‘Baidu’s Top 10 New Internet Phrases of 2011’, available at: http://beat.baidu. com/?p=3515. Bourdieu, Pierre (1977) ‘The Economics of Linguistic Exchanges’, Social Science Information 16(6): 645–68. Bourdieu, Pierre (1991) Language and Symbolic Power, edited by J. B. Thompson, Cambridge: Polity Press. Cao, Guangshun and Yu Hsiao-jung (2000) ‘The Influence of Translated Later Han Buddhist Sutras on the Development of the Chinese Disposal Construction’, Cahiers de Linguistique Asie Orientale 29(2): 151–77. Chen, Jia (2011) ‘Expat Student Numbers Rise’, China Daily, March 4. Chow, Rey (2013a) ‘On Chineseness as a Theoretical Problem’, in Shih Shu-mei, Tsai Chien-Hsin, and Brian Bernards (eds) Sinophone Studies: A Critical Reader, New York: Columbia University Press, 43–56. Chow, Rey (2013b) ‘The Subject of Chinglish’, Testo a Fronte: teoria e pratica della traduzione letteraria 48 (1 semestre 2013): 155–62, in Tim Parks and Edoardo Zuccato (eds) Special issue: Towards a Global Literature/Verso una Letteratura Globalizzata. Crystal, David (2006) Language and the Internet, 2nd ed., Cambridge: Cambridge University Press. Crystal, David (2012) English as a Global Language, 2nd ed., Cambridge: Cambridge University Press. DeFrancis, John (1984) The Chinese Language: Fact and Fantasy, Honolulu: University of Hawaii Press. Eng, Bob (2010) ‘Is Cantonese in Danger of Extinction? The Politics and Culture of Language Policy in China’, available at http://chinamusictech.blogspot.com/2010/08/is-cantonese-in-danger-ofextinction.html. Fang, Yunyu (2010) ‘Chinglish Sentenced to Death’, Global Times September 29. Frehner, Carmen (2008) Email, SMS, MMS: The Linguistic Creativity of Asynchronous Discourse in the New Media Age, Bern and New York: Peter Lang. Gao, Liwei (2008) Chinese Internet Language: A Study of Identity Constructions, Munich: Lincom Europa. Gao, Liwei (2012) ‘Synchronic Variation or Diachronic Change: A Sociolinguistic Study of Chinese Internet Language’, in Liu Jin and Tao Hongyin (eds) Chinese under Globalization: Emerging Trends in Language Use in China, Singapore: World Scientific, 7–28. Gao, Mingkai 高明凯 and Liu Zhengtan 刘正埮 (1958) Xiandai hanyu wailaici yanjiu《现代汉语外 来词研究》(Studies of Loanwords in Modern Chinese), Beijing: Wenzi gaige chubanshe. Gibbon, Dafydd and Małgorzata Kul (2010) ‘Economy Strategies in Restricted Communication Channels: A Study of Polish Short Text Messages’, in J. Normann Jørgensen (ed) Vallah, Gurkensalat 4U and me! Current Perspectives in the Study of Youth Language, Frankfurt am Main: Peter Lang, 75–98. Gumperz, John (1968) ‘The Speech Community’, in David Sills (ed) International Encyclopedia of the Social Sciences, New York: Macmillan, 381–6.

133

Liu Jin and Tao Hongyin Gunn, Edward (1991) Rewriting Chinese: Style and Innovation in Twentieth-century Chinese Prose, Stanford: Stanford University Press. Han, Chong (2012) ‘The Metaphorical World of Chinese Online Entertainment News’, in Liu Jin and Tao Hongyin (eds) Chinese under Globalization: Emerging Trends in Language Use in China, Singapore: World Scientific, 29–57. Haugen, Einar (1966) ‘Dialect, Language and Nation’, American Anthropologist 68(4): 922–35. He, Na (2010) ‘Rush to Learn English Fuels Quality Issues’, China Daily August 5. Hobsbawm, Eric J. (1992) Nations and Nationalism Since 1780: Programme, Myth, Reality, Cambridge: Cambridge University Press. Hwang, David Henry (2012) Chinglish: A Play, New York: Theatre Communications Group. Jacobs, Andrew (2010) ‘Shanghai is Trying to Untangle the Mangled English of Chinglish’, New York Times May 2. Jhnews.com (2003) ‘Jinhua fangyan, ping shenme bei paoqi’〈金华方言，凭什么被抛弃〉(‘Why the Jinhua Wu Dialect Has Been Abandoned’), available at: http://www.jhnews.com.cn/gb/content/ 2003-04/18/content_172194.htm. Ji, Fengyuan (2004) Linguistic Engineering: Language and Politics in Mao’s China, Honolulu: University of Hawaii Press. Kipnis, Andrew B. (2012) ‘Constructing Commonality: Standardization and Modernization in Chinese Nation-Building’, The Journal of Asian Studies 71(3): 731–55. Kozha, Ksenia (2012) ‘Chinese via English: A Case Study of “Lettered-Words” as a Way of Integration into Global Communication’, in Liu Jin and Tao Hongyin (eds) Chinese under Globalization: Emerging Trends in Language Use in China, Singapore: World Scientific, 105–25. Landry, Rodrigue and Richard Bourhis (1997) ‘Linguistic Landscape and Ethnolinguistic Vitality: An Empirical Study’, Journal of Language and Social Psychology 16(1): 23–49. Leung, Genevieve Y. (2009) ‘Hong Kong University Students’ Language Use in Blogs: Tensions between Creativity and Prescriptivism (“omg my grammar is rubbish”)’, Sino-Platonic Papers 189: 30–62. Link, Perry (2013) An Anatomy of Chinese: Rhythm, Metaphor, Politics, Cambridge, MA: Harvard University Press. Li, Rui 李锐 (2000) ‘Wangluo shidai de fangyan’〈网络时代的方言〉(‘Dialect in the Age of the Internet’), Dushu《读书》(Reading) 4: 42–7. Liu, Jin (2011) ‘Deviant Writing and Youth Identity: Representation of Dialects with Chinese Characters on the Internet’, Chinese Language and Discourse 2(1): 58–79. Liu, Jin (2013) Signifying the Local: Media Productions Rendered in Local Languages in Mainland China in the New Millennium, Leiden: Brill. Liu, Jin and Tao Hongyin (eds) (2012) Chinese under Globalization: Emerging Trends in Language Use in China, Singapore: World Scientific. Liu, Lydia H. (1995) Translingual Practice: Literature, National Culture, and Translated Modernity – China, 1900–1937, Stanford: Stanford University Press. Liu, Yilin 刘轶琳 (2013) ‘Benshi yiyou 100 duosuo zhongxiaoxue kaishe “Shanghaihua” ke’〈本市已有 100多所中小学开设《上海话》课〉(‘More Than 100 Elementary and Middle Schools in Shanghai Have Offered “Shanghai Wu” Classes’), available at: http://sh.eastday.com/m/20130524/u1a7412644.html. Lo Bianco, Joseph (2012) ‘Chinese: The Gigantic Up-and-Comer’, in Linda Tsung and Ken Cruickshank (eds) Teaching and Learning Chinese in Global Contexts, London and New York: Continuum, xiii–xiv. Mair, Victor (1991) ‘What is a Chinese “Dialect/Topolect”? Reflections on Some Key Sino-English Linguistic Terms’, Sino-Platonic Papers 29: 1–31. Meiri xinbao (2008) ‘Tianjin goubuli xuangua yingwen zhaopai Go Believe’〈天津狗不理悬挂英文 招牌 Go Believe〉(Tianjin Famous Chain Store Goubuli Puts on the English Brand Name Go Believe)’, available at: http://news.sina.com.cn/s/2008-01-10/180214714025.shtml. Pan, Wenguo 潘文国 (2008) Weijixia de Zhongwen《危机下的中文》(Chinese in Crisis). Shenyang: Liaoning Renmin Chubanshe. Qian, Nairong 钱乃荣 (2004) ‘Zhiyi “xiandai hanyu guifanhua” ’〈质疑《现代汉语规范化》〉(‘Questioning the “Standardization of Modern Chinese Language” ’), Shanghai wenxue《上海文学》(Shanghai Literature) 4: 78–83. Radtke, Oliver (2012) ‘More Than Errors and Embarrassment: New Approaches to Chinglish’, in Liu Jin and Tao Hongyin (eds) Chinese under Globalization: Emerging Trends in Language Use in China, Singapore: World Scientific, 145–70.

134

Chinese language in a global context Ramsey, S. Robert (1987) The Languages of China, Princeton: Princeton University Press. Schoenhals, Michael (1992) Doing Things with Words in Chinese Politics, Center for Chinese Studies Research Monographs 41, Berkeley: Institute of East Asian Studies, University of California. Sebba, Mark (2003) ‘Spelling Rebellion’, in Jannis Androutsopoulos and Alexandra Georgakopoulou (eds) Discourse Constructions of Youth Identities, Amsterdam and Philadelphia: John Benjamins Publishing Company, 151–72. Shanghai Daily (2010) ‘The Rise of Net Buzzword “Gelivable” ’, available at: http://www.china.org. cn/china/2010-11/12/content_21327924.htm. Shih, Shu-mei (2007) Visuality and Identity: Sinophone Articulations across the Pacific, Berkeley: University of California Press. Tang, Zhixiang 汤志祥 (2001) Dangdai hanyu ciyu de gongshi zhuangkuang jiqi shanbian: 90 niandai Zhongguo dalu, Xianggang, Taiwan hanyu ciyu xianzhuang yanjiu《当代汉语词语的共时状况 及其嬗变: 90年代中国大陆,香港,台湾汉语词语现状研究》(The Synchronic state of contemporary Chinese vocabulary and its transformation: Studies in the situation of vocabulary in mainland China, Hong Kong, and Taiwan during the 90s), Shanghai: Fudan University Press. Tian, Xiaoling 田晓玲 (2012) ‘Zhongxiao xuesheng chengzhang qingkuang diaocha baogao fabu’ 〈中小学生成长情况调查报告发布〉   (‘The Survey of the Elementary and Middle School Students in Shanghai Is Released’), Wenhui bao May 11. The Independent (2010) ‘Dictionary Adds Chat-room Chinese Words That Are Simply “Niu” (awesome)’, available at: http://www.independent.co.uk/arts-entertainment/books/dictionary-adds-chatroomchinese-words-that-are-simply-niu-awesome-2074467.html Wagner, Rudolf 瓦格纳•鲁多夫 (1999) ‘Zhonggong 1940–1953 nian jianli zhengyu zhengwen de zhengce dalüe’〈中共 1940–1953 年建立正语、正文的政策大略〉(‘An Overview of CCP Policies in Establishing an Orthodoxical Language and Discourse in 1940–1953’), in Peng Xiaoyan 彭小妍 (ed) Wenyi lilun yu tongsu wenhua《文艺理论与通俗文化》(Literary Theory and Popular Culture), Taipei: Zhongyang yanjiuyuan Zhongguo wenzhe yanjiusuo choubeichu, 11–38. Waldron, Arthur (2010) ‘Will Linguistic Centralization Work? Protesters Demonstrate against Restrictions on Cantonese’, Jamestown Foundation, China Brief, 10(16), August 5. Wang, Ban (ed) (2011) Words and Their Stories: Essays on the Language of the Chinese Revolution, Leiden: Brill. Wang, Ping 汪平 (2003) ‘Putonghua he Suzhou Hua zai Suzhou de xiaozhang yanjiu’〈普通话和 苏州话在苏州的消长研究〉(‘A Study of the Rise of the Putonghua and the Decline of Suzhou Wu Dialect in Suzhou’), Yuyan jiaoxue yu yanjiu 1: 29–36. Wang, Shuo 王朔 and Lao Xia 老霞 (2000) Meiren zeng wo menghanyao《美人赠我蒙汗药》(A Beauty Presents Me with a Sleeping Potion), Wuhan: Changjiang wenyi chubanshe. Wang, Youjia 王有佳 (2005) ‘Shanghai: fangyan jieshao wenzhang “Nong xiaode fa” jin keben re zhengyi’〈上海:方言介绍文章《侬晓得伐》进课本惹争议〉(‘Shanghai: An Article Introducing Shanghai Wu ‘Do you know’ Evoked Controversy’), People’s Daily, September 6. Wines, Michael (2009) ‘A Dirty Pun Tweaks China’s Online Censors’, The New York Times March 12. Woodward, Amber R. (2008) ‘A Survey of Li Yang Crazy English’, Sino-Platonic Papers 180. Woodward, Amber R. (2012) ‘Learning Chinese to Promote Chinese: A Study of Li Yang’s Crazy English’, in Liu Jin and Tao Hongyin (eds) Chinese under Globalization: Emerging Trends in Language Use in China, Singapore: World Scientific, 127–43. Xiao, Qiang and Perry Link (2013) ‘In China’s Cyberspace, Dissent Speaks Code’, Wall Street Journal January 4. Xinhuanet (2013) ‘Chinese Craze for English Debated’, available at: http://news.xinhuanet.com/english/ china/2013-03/14/c_124460602.htm. Yin, Tingting 尹婷婷 (2002) ‘Guojia yuwei fuzeren piping Jiang Wen jiang yuyan ‘zaopo’ banshang yinmu’〈国家语委负责人批评姜文将语言《糟粕》搬上银幕〉(‘The Official from the State Language Commission Criticizes Jiang Wen for Bringing Linguistic Dregs to the Silver Screen’), Chengdu ribao《成都日报》(Chengdu Daily) September 2. Yu, Jian 于坚 (2004) ‘Shige zhi she de ying yu ran: shige yanjiu cao’an: guanyu dangdai shige de lianglei yuyan xiangdu’〈诗歌之舌的硬与软——诗歌研究草案:关于当代诗歌的两类语言向度〉 (‘The Hard and Soft of the Tongue of Poetry – A Draft of Poetry Study: On Two Different Directions in the Language of Contemporary Poetry’), in Yu Jian Jujue yinyu: Zongpi shouji, pinglun, fangtan 《拒绝隐喻:棕皮手记,评论,访谈》(Refusal   of Metaphor: Brown Notebook, Criticism, and Interviews), Kunming: Yunnan renmin chubanshe, 137–51.

135

Liu Jin and Tao Hongyin Yuyan kexue 语言科学 (Language Science) (2005) ‘Zhongguo yuyan wenzi shiyong qingkuang diaocha jieguo zai Beijing gongbu’〈中国语言文字使用情况调查结果在北京公布〉(‘The Result of the National Survey on Language Use in China Released in Beijing’), 3: 112. Zhang, Qing (2005) ‘A Chinese Yuppie in Beijing: Phonological Variation and the Construction of a New Professional Identity’, Language in Society 34(3): 431–66. Zhang, Qing (2012) ‘ “Carry Shopping Through to the End!”: Linguistic Innovation in a Chinese Television Program’, in Juan Manuel Hernandez-Campoy and Juan Antonio Cutillas-Espinosa (eds) Style-shifting in Public: New Perspectives on Stylistic Variation, Amsterdam and Philadelphia: John Benjamins Publishing Company, 205–24. Zhang, Qing (forthcoming) Sociolinguistic Variation in China: Undoing Commonness through Cosmopolitan Mandarin, New York and London: Routledge. Zhu Jing (ed) 朱竞 (2005) Hanyu de weiji《汉语的危机》(The Chinese Language Crisis), Beijing: Wenhua Yishu Chubanshe.

136

9 Chinese language pedagogy Wu Weiping Hong Kong polytechnic University, Hong Kong

Although Chinese language pedagogy (CLP) can be viewed from various perspectives (Xing 2006; Y. Wu 2012), a diachronic approach is adopted in this chapter so that the current situation in this field can be presented. Synchronically, some key stages of CLP will also be mentioned briefly to provide a vertical dimension for the current picture. Within the diachronic approach, the discussion here will center around two major trends of practices in the science and art of Chinese language teaching: one emphasizing language structure, while the other language use. Obviously, pedagogy will differ depending on whether we view Chinese as a Native Language (CNL) or Chinese as a Second Language (CSL). For discussion purposes, the assumption here is CSL if not otherwise noted, since this is a major area in most research and publications. There are of course similarities in pedagogy no matter whether it is for native speakers or for second language learners, such as the importance attached to certain salient features of the language (e.g. tone as a meaning-indicating feature in Chinese), but comparison between CNL and CSL approaches is not part of this chapter. In the course of the discussion, key issues related to the two trends in terms of the purpose, mode, scope, focus and assessment of CSL teaching will be explored.

1.  Chinese language pedagogy focusing on language structure (LS) Language teaching used to be viewed as any other teaching activities over the years, with the primary purpose of knowledge transmission from the teacher to the students. Learning a language, therefore, involves the study of the structure of that language. All pedagogical issues under such an approach naturally boil down to the three major components of a language in terms of its structure: namely, phonology, semantics and syntax. A syllabus focusing on language structure in teaching CSL may have a variety of forms and shapes, but Figure 9.1 is typical of most syllabi under such an approach. In Figure 9.1, we could further describe each step: 1. Presenting knowledge of language structure: Examples intending to show basic rules of the language in terms of its phonological, semantic and syntactic structure; 2. Explaining language structure illustrated by the examples, whether inductive or deductive in various attempts to provide explanations by teachers, with rules governing the structure of what we see and hear in communication by language as their final destination; 137

Wu Weiping Presentation of LS (be it sound, vocabulary or grammar) →Explanation of basic units in LS →Practices on selected units based on LS →Feedback on LS →Test on what is learned as part of LS

Figure 9.1  Major steps in CSL teaching syllabi focusing on language structure (LS)

3. Practicing by students, both in the classroom or in the form of homework, focusing on comprehension and correct use of language structure (e.g. sentence making with given vocabulary or key words in sentence patterns); 4. Checking, by teachers or their designated representatives who are assumed to possess the knowledge being transmitted, on the correctness of the language forms produced by the students; 5. Finding out what the students cannot do in an achievement test based on what has been taught, pointing out errors spotted and (occasionally) reminding students of contexts in language use. In step 1, basic units of the phonological features of the language (i.e. initials, finals and tones in Chinese), common vocabulary items (either based on structure of the characters that make up the words, or on various kinds of frequency counts) of the characters used and fundamental grammar rules (e.g. from simple to complex sentences in Chinese) are presented. By step 5, most people following such a structure-based syllabus, both teachers and students, would consider the job done because examinations are held and results obtained (W. Wu 2008).

1.1.  Language as a system of systems Nobody can talk about the way of doing something if that ‘something’ is not well-defined or clearly understood. While exploring various ways to teach Chinese focusing on the structure of the language, we tend to consider language as a system incorporating three subsystems. The sound system in Chinese, for example, consists of a definite number of initials, finals and tones (Norman 1988). This is a closed system from a diachronic perspective, meaning all the components of this system are relatively stable at any given time. The vocabulary, as a system to express meaning by words, is at the other end of the spectrum in terms of stability. While the majority of the words may have little change over a significant period of time, some words will become obsolete and eventually die and new words will pop up from time to time. Thus this system is viewed as an open system, in the sense that it will keep changing regardless of any conscious efforts from teachers, researchers or linguists. Grammar, or the rules for putting words together to make meaningful expressions, is something in between: not as stable as the sounds of language but with less change when compared with vocabulary. Typically, teachers deal with each of the subsystems in different ways pedagogically.

1.2.  Pedagogy in phonology Methods and skills in teaching the Chinese sounds may share many similarities with those in teaching other languages, but no pedagogical approach would be complete without considering tones as an integrated part of every syllable, a unique feature of the Chinese language. 138

Chinese language pedagogy

Chinese is often considered as a syllabic language in which each syllable can be decomposed into three components: initial, final and tone. Pictographically, it is usually represented by one character in the Chinese writing system, though some of the syllables used orally may not have their character representation. There are all together 23 initials, 34 finals and four tones (or five if ‘light tone’ is counted). Teaching the sound system is a multilayer task. In terms of comprehension, teachers need to make sure that students understand the rules of combining initials and finals together, the sound pitch and contour associated with each tone, as well as possible tone sandhi for certain combinations in connected speech. They also need to train their ears so that they can distinguish the sometimes subtle differences among certain minimum pairs, such as the retroflex /sh/ versus the non-retroflex initial /s/. In terms of production, teachers would try their best to cultivate accurate pronunciation among their students, usually by a combination of efforts including trial and error, imitation, error analysis and a formidable amount of exercises. Some of the more effective pedagogical tools are listed below: ● Use of visual aids such as charts and symbols, referring to articulating organs ● Audio-video assistance making use of modern technology ● Listen and imitate, the most basic and effective approach, may still be the best way in the words of many students (Xing 2006) To most learners, it is hard enough to distinguish the four tones and their numerical values (level 55, rising 25, dipping 214 and falling 52) (Chao 1968), let alone understanding the subtle and yet important differences between each minimum pair. Moreover, the step following comprehension and production, the ability to use them in connected speech for communication purposes, takes much more effort. Even though each and every student knows the value of the four tones, trying to get them right while using a random combination of them often proves to be too much.

1.3.  Various teaching approaches to vocabulary Unlike the sound system, vocabulary is an open system and only a selected number of items in this system are taught in any language program. Compared with the other two subsystems (sound and grammar), the teaching of vocabulary turns out to be most complicated because of many disputed concepts. To start with, people disagree on the line between a character and a word, which is largely responsible for the on-going debate over the character-based, or zibenwei (Bellassen 2002), versus word-based, or cibenwei, pedagogy (Liu 2010). The former insists that characters are the very basic units of all teaching activities, while the latter argues that words, which are often combinations of characters, should be such basic units. No matter which line they stick to, however, the common pedagogical approach is to teach a set of commonly used characters (or words) by the 3P model: Presentation of sample items by teachers, Practice of various kinds (individual, paired, group, teacher–students, student–students, etc.) and Production by students, which usually consists of using the newly learned vocabulary items in sentence-making exercises to show that they can use these words ‘correctly’ in a given language structure. As always, there are people who take the middle of the road and attempt to combine the strength from both sides. Then they argue about which set of vocabulary is most common, less common or uncommon, and by what criteria they are so labeled. The reason behind such argument is the belief 139

Wu Weiping

that the most common ones should be taught first. The problem is, there are quite a few vocabulary lists based on some kind of frequency count. Since the same item may occupy a different place in different lists, which list should be used becomes a hard choice in teaching-material preparation and in teaching-activity design for pedagogical purposes. To make matters worse, such frequency counts often come from written data, so they are far from satisfactory when it comes to courses and materials focusing on listening and speaking (Tao 2011). A large-scale corpus for spoken language in the natural setting has yet to be built and, without the support of such a corpus, frequency counts for spoken language are impossible. Obviously, what is frequently used in written materials may not be frequent at all in the spoken form. Thus relying on the ‘most frequent’ vocabulary in making pedagogical decisions at this stage can be rather misleading. Finally, scholars and practitioners, under the influence of different theories they believe in, tend to disagree on how the vocabulary should be presented when teaching them. Among the factors that affect pedagogical decisions is the major issue of the role of Romanization in vocabulary learning. Romanization is not a natural part of the Chinese language for native speakers, but an artificial tool created to facilitate the learning process. As such, it is often regarded as a crutch in most pedagogical approaches. Many teachers would start with it to help the students master the pronunciation of the vocabulary and make the link between sound and meaning, and let it fade out as the number of words remembered by the learner has increased to a considerable size (which again differs from program to program). The period in which such a crutch is used varies, from one or two weeks to several months. This explains why we often encounter the question: ‘When will characters be introduced in the program?’ At one extreme end of the spectrum, there are also programs in which Romanization is just pushed aside and the vocabulary learning process starts with making the link between meaning and shapes of the characters used to represent the words.

1.4.  Teaching grammar Grammar may mean totally different things in the eyes of a research scholar and a teacher. The former tries to find a system that can explain all linguistic phenomena of the language (research grammar), while the latter focuses on some basic rules that can guide the learners (pedagogical grammar) in their efforts to put words together to express meaning (Lu 2006). Teaching grammar from a pedagogical perspective, therefore, involves only the task of identifying a set of basic rules for teaching and how to make the students understand these rules. This may sound simple but is again rather complicated in reality. It is not hard at all to imagine that reaching an agreement in identifying the set of rules is mission impossible to start with. All language programs are built upon certain concepts about language and each individual teacher will always have his or her own view about grammar. A basic rule considered by Teacher A as part of the set for teaching may or may not be recognized by Teacher B in the same way. Unlike vocabulary where everybody agrees what a vocabulary item should be. The major debate usually centers on where such an item should stand in the frequency account and when is the best time to introduce it to the learners. The world of grammar is much more complicated, with parts of speech, prosodic features, normal SVO patterns, special syntactic structures like the ba-pattern (Feng 2013). Thus trying to select certain features to be included in teaching is already a formidable task in itself. When it comes to pedagogical approach in teaching grammar in CSL programs, the general agreement is to downplay issues related to parts of speech and tense, and stress issues related 140

Chinese language pedagogy

to basic and special sentence patterns and aspects. In addition to the introduction of general concepts such as the SVO order in sentences, including simple and some complex ones in which conjunctions are used, most programs will pay particular attention to special sentence patterns in Chinese, such as the ones with ba, le (or double-le), bei, the ‘left-branching’ structure with -de between the modifying phrases and clauses and the head noun, the differences in word-order between Chinese and Indo-European languages when time and place adverb or adverbial phrases are used, and other features such as the use of a designated modifier before most countable nouns. All these are proven to be difficult spots for most learners who are not familiar with the Chinese language and culture and tend to attract more attention pedagogically from teachers. Similar to the concept of character-based and word-based, there is also pedagogical approach advocating ‘sentence-based’, or jubenwei. Scholars and teachers holding this view argue that a sentence is the most basic unit in language because only a complete sentence can express meaning, however short it may be (Zhang 2006). In short, the ideal way to teach grammar is to have a ready set of rules recognized by all teachers so that every teacher is clear about what to teach. Such a set of rules, again ideally, will be used in all teaching materials. Moreover, the order of appearance for rules in this set will represent the accepted order of acquisition by most learners based on sound research in CSL acquisition. The reality in teaching CSL, however, is far from ideal due to the complexity of various issues related to Chinese grammar.

1.5.  Characters and reading in Chinese If the three subsystems discussed above, sound, vocabulary and grammar, are common to all languages, the writing system in Chinese is something unique and therefore requires a unique approach in teaching. In essence, the primary task in all language learning is to establish the link between sound and meaning in speaking and listening, and between the writing system and meaning in reading and writing. What makes CSL special is the lack of relationship between sound and the writing system for most learners. The huge number of Chinese characters, which developed from the tradition of meaning-expressing pictographs, is indeed overwhelming for both teachers and students. The learning process is greatly hindered in most regular programs because of the high threshold: around 3,000 individual characters to be memorized, recognized and used in reading and writing. No matter what pedagogical tricks experienced teachers can play – by meaning-expressing radicals, by soundexpressing components or by association, with flash cards, low-tech writing brushes to motivate interest, or a high-tech IT platform to woo the attention of younger learners – the amount of time that can be given in any program is much less than needed to master this basic set of characters, let alone the combination of individual characters in this basic set to form a much greater number of words and phrases. Apart from the built-in challenge of learning such a writing system, the linguistic and cultural backgrounds of the learners may also lead to differences in pedagogical approach in teaching. For Asian learners whose languages and cultures are somehow related to the use of characters, including Japan, Korea, Vietnam and Singapore (Cui and Chen 2000), the path of learning can be steep and the speed of learning can be fast, with reading and writing exercises at a very early stage of the program. For learners without any background in a writing system relying on ideographic and pictographic symbols, they need a much gentler approach, with a very limited number of characters in each lesson, a well-structured reappearance interval and a much slower speed in progress (An 1988; Jiang 2007). 141

Wu Weiping

As for the differences in the two varieties of some commonly used characters, namely the simplified and the traditional form, most discussions and approaches over the years are tinted by politics, as well as cultural constraints to a certain degree. Without going into any details, and therefore debates, we can simply state that the norm now accepted by many is to teach simplified characters in mainland China and to teach traditional ones in Taiwan. Where outside forces in politics and culture are not homogenous, powerful or dominant enough and teachers and programs can put academic choice first, such as in North America or Hong Kong, they do have the choice but with it comes the dilemma: meeting the needs of one group means running the risk of neglecting the wish of the other. It was under such circumstances that the pedagogical approach to ‘recognize the traditional but use the simplified’ came about, and practiced with varying degrees of success in many programs in these areas. Before we move to describing a different approach based on language use instead of language structure, we should at least mention the relationship between the approaches in teaching the four basic components in CSL as outlined above, and the training of four basic skills (listening, speaking, reading and writing) among learners of the language. Such a relationship is illustrated below: Phonology: courses focusing on listening and speaking Vocabulary: courses related to all the four basic skills Grammar: courses related to all the four skills Characters: all courses focusing on reading skills The realization of such a link is important, especially with increasing attention among scholars and practitioners to ‘pedagogical models’ in CSL teaching, as indicated in more and more discussions by key players in the field (Liu 2014).

2.  Chinese language pedagogy focusing on language use (LU) Language structure and language use are two different matters and, therefore, each requires a different pedagogical approach. Needless to say, one cannot have French fries if there is no potato. No matter how many potatoes you may have, on the other hand, they cannot become French fries if you don’t go through the required cooking process. Such a process also exists between the language structure you have and the language you use. You need to put the language structure you have learned in the appropriate setting so that the right meaning can be sent and received.

2.1.  The concept of Pragmatic Framework in CSL pedagogy If we really follow the pedagogical approach in which language use is the goal, we need to understand the process that makes what we hear and see meaningful. That means the significance of all the contextual factors in language use, namely the people who use the language, the context in which we use the language, and the timing or purpose related to the language being used (Bates 1976; Hymes 1972; Levinson 1983; Omaggio 1986). All these factors form what we call ‘Pragmatic Framework’ (W. Wu 2008), which should also be included as part of the pedagogical activities and be taught or acquired by language learners, in addition to the teaching of language structure that we are all familiar with. The importance of such a framework should prevent us from treating it as an add-on that we remind our students of at the end of the teaching process, and instead ensure that it is an integrated part of a pedagogical approach throughout the whole learning process, as explained below. 142

Chinese language pedagogy Identify factors involved in language use (starting point of the learning process) →Presentation of pragmatic points as the basic units in LU →Explanation of both structural and contextual factors in LU →Practices and feedback based on LU →Test focusing on LU

Figure 9.2  Major steps in CSL teaching syllabi focusing on language use (LU)

First of all, the presentation unit will be based on language use instead of language structure. Instead of the phonological features covering the sound system, selected semantic units as represented by words, and a representative set of sentence patterns based on the syntax, the unit representing the system of language use is what we call a ‘pragmatic point’ (W. Wu 2006), which is defined as a linguistic function in context. What we want the students to understand is that such a function includes all the structural information (sound, word and grammar), as well as the contextual information (who to whom, where and when) Then, the major stages of learning, as indicated in Figure 9.2, can each be handled pedagogically based on research findings in the CSL field. Each of the steps in Figure 9.2 can be described as follows: 1. Recognizing key factors in language use, in addition to language structure, especially subtleties related to who, where, when and how in a communication event, all of which are major components of the pragmatic framework; 2. Presenting authentic or semi-authentic materials in language use; 3. Explaining the components for the basic unit of language use, which consists of linguistic knowledge and pragmatic knowledge; 4. Practice in classroom, in the form of homework or in real-life contexts as part of the curriculum design, focusing on both language structure and language use, with frequent feedback on student performance, aiming at both appropriateness and correctness; 5. Finding out what the students can do by performance-based assessment tools. Now that the major steps in the teaching process are established, we can explore the pedagogical approach in each stage within the process. Such a process is here described as ‘counter-clockwise’ because assessment is put at the starting point of the learning process, rather than at the end as normally found in most curricula.

2.2.  Establish the link between purposes of teaching and assessment The approach of respecting and making use of the power of tests (Shohamy 2001) tries to establish the link between the purpose of learning and the assessment procedure. If we need to evaluate the actual ability of communication, then we need a direct test to gather information about the actual ability of the learner (Bachman and Palmer 1996; Byram 1997). In other words, ability is the aim of testing for ability-oriented programs, and knowledge is the aim of testing for knowledge-based approaches. So, a proficiency-oriented performance test would be the natural choice for programs focusing on language use, rather than an indirect test of language structure and, based on the result of such a test, ‘estimate’ the ability of the learner. One common practice in language programs is to start with teaching and end with a test to see if what has been taught has been learned. With the general tendency to avoid ‘teaching for testing’, most teachers would avoid placing too much emphasis on testing. Such 143

Wu Weiping

a precaution is quite understandable given the fact that most of the programs and teaching materials are structure-based. With a direct test aiming at the actual ability to use language (here we use spoken language and oral proficiency tests as examples for discussion), a task bank with real-life situations can be created. Since each task in the test is created as a real life, or similar to a real life task, the ability to perform such a task is deemed as close, or very close, to the real-life ability of the learner (Candlin 1987; Chi 1996). It is with such an assumption that a properly designed assessment tool becomes the start point rather than the end point of the learning process. That does not mean the first step for all learners is to take a test. A proficiency test, by definition, can be taken any time. The purpose of putting together a sizeable task bank is to use it as a guide for stages to follow. Just like a selected number of grammatical features are treated as a representation of the grammar system in terms of language structure, a group of real-life tasks, also known as pragmatic points, are treated as a representation of language use. Here we must reiterate the fact that, for many learners, they are learning in order to use the language, and the practice of presenting units of language use for learning purposes, instead of language structure, is not hard to accept. This group of tasks, each of them associated with a linguistic function, is then divided into categories and grouped into proficiency levels according to their relative degrees of difficulties. Each level serves as a subsystem of language use, similar in relation to the situation in which phonology is a subsystem of language structure. Borrowing the concept and the terminology used in the Speaking Proficiency Guidelines by the American Council on the Teaching of Foreign Languages (ACTFL 2012), each pragmatic point in the database belongs to one of the three proficiency levels: Intermediate, Advanced or Superior. A sample of linguistic functions commonly used in our daily life is provided in Table 9.1. In general, learners with higher proficiency in the language they learn will perform better than those with lower proficiency. The logic behind an assessment tool built on linguistic functions is the belief that: (i) using a linguistic function to perform a task in daily life will inevitably involve language structure, including sounds, utterances and discourse if we speak, and words, sentences and paragraphs when we write; and (ii) linguistic functions found in high proficiency levels (Superior and Advanced) tend to involve a larger variety of grammatical structures, vocabulary items and rhetorical devices, while those at the lower end of the proficiency levels will involve a smaller number of structural varieties, which also tend to be simpler in nature from the perspective of most users (Ellis 2003; Willis and Willis 2007).

Table 9.1  ACTFL levels with sample linguistic functions Level

Linguistic function

Context

Sample content

Novice

listing counting

Informal setting

Specific, memorized material

Intermediate

giving information asking questions

Informal to semi-formal

Specific, personal content ‘me’

Advanced

narrating a story summarizing an event

Semi-formal/ formal

Specific/abstract, concrete world

Superior

supporting an opinion making a speech in public

Formal and official

Abstract concepts, controversial issues

144

Chinese language pedagogy

Each task in the data bank, regardless of the level in which it is placed, has some specific factors from the pragmatic framework related to the linguistic function, including: 1. People/Interlocutor: factors such as age, gender, social status, one to one, one to many, one for one or one for many (on behalf of ) and so on; 2. Place/Setting: formal, informal, semi-formal, official, friendly, public, private, and so on; 3. Timing/Purpose: right or wrong moments for the interlocutor with their identity and social status, intended or not intended occasion for what is being discussed, and so on. With all these complex factors, as opposed to pure language structural ones, a lot more needs to be planned pedagogically in each of the following stages: curriculum design, teaching-materials preparation and teaching activities.

2.3.  Curriculum design based on expected outcome In terms of curriculum design, the focus is on the structure of the whole program and the variety of courses that will serve the ultimate goals of learning. Like all other language programs, knowledge of language is the foundation of any proficiency level that we expect our students to achieve (Krashen 1987; W. Wu 1993). That means learners will have to spend a reasonable amount of time trying to learn about rules and basic facts in pronunciation, vocabulary and grammar. The key difference in a program that claims to focus on language use is the emphasis on the context in which language is used. To better serve the needs in cultivating the communicative ability of the students, courses focusing on language use should be created and treated as an indispensable part of the curriculum. Since each and every task created for the data bank in assessment mentioned in the last section already has its built-in pragmatic clues in language use, the same database is used as the basis in designing the curriculum with all the courses. In other words, all the tasks, as summarized in Table 9.2, would somehow become the major focus of certain courses in the curriculum. Each category of linguistic functions in Table 9.2 has a certain number of specific tasks (between 5 and 50). Many of the tasks are used in this stage to form the skeleton of different courses at different levels in the curriculum. In the next stage, the content area found in each task will be used as a starting point in compiling teaching materials. While a detailed discussion of the whole curriculum is not feasible here, a brief description of one new course can be cited as an example to show how language use can become the focus in the learning process. In light of the need to remind students of their major goals in learning, a special course, to be conducted in real-life situations rather than the classroom, can be created. One feature of such a course is that students are required to participate in learning activities located in a Chinese speaking environment (e.g. cities like Beijing or Dalian where standard Chinese is widely spoken), with field trips and presentations, rather than classroom instructions and explanations. All the key elements of the course, from location of class and teaching materials to requirements for the students, are put together to give one message: learning is for using and ways to use can become ways to learn. The emphasis of the course is on cultivating learners’ sensitivity to contextual factors, which are always available in actual communication. Their sensitivity to appropriateness according to the situation in which they use the language, and their awareness that the learning process will not be complete until they are able to use the language in real life are also developed in the learning process. 145

Wu Weiping Table 9.2  Summary table of pragmatic points in the system of language use PL

Task categories

Tasks

Notes

I

 I1.  I2.  I3.  I4.  I5.  I6.  I7.  I8.  I9. I10.

50 10 10 10 40 10 15 15 25 10

All the 70 tasks under categories I1 to I3 are picture-based. Categories I4 to I10, with a total of 125 tasks, are I-level linguistic functions with basically informal contexts in various content areas.

 A1. 投诉/complaining  A2. 分析/analyzing  A3. 比较/comparing and contrasting  A4. 阐释/expounding  A5. 陈述/narrating  A6. 建议/suggesting  A7. 拒绝/refusing  A8. 推销/promoting  A9. 致歉/apologizing A10. 批评/criticizing A11. 介绍(半正式)/introduction (semi-formal)

15 50 50 35 20 30 20 10 10 15  5

All the tasks at the A-level categories, with 11 linguistic functions covering 260 scenarios with various content areas, are created with settings of a semi-formal nature, such as business or social encounters of a business nature.

  S1.   S2.   S3.   S4.   S5.   S6.   S7.   S8.   S9. S10.

35 40 10 10 10 10 10 10  5  5

Based on the 10 linguistic functions at the S-level, 145 tasks are created to cover the needs for language use in the formal setting, including some representative speech events of a public nature.

中 级

A 高 级

S 特 级

描述/describing 说明/explaining (picture-based) 指路/providing directions 留言/leaving messages 介绍(日常生活)/introduction (daily life) 感谢/expressing thanks 说明/explaining (word-based) 解释/illustrating 讨论/discussing 提供信息/offering information

发表见解/expressing an opinion 会议发言/speaking at meetings 公开致谢/expressing thanks in public 劝说/persuading 反驳/rebutting 祝贺/congratulating 演讲/delivering a speech 辩护/defending 号召/calling (to action) 介绍(正式)/introduction (formal)

(PL: Proficiency Level; I: Intermediate; A: Advanced; S: Superior)

2.4.  Teaching-materials preparation according to curriculum In a language program that emphasizes learner ability in oral communication, students would find it absurd to have their final grade based on a written test. Similarly, a program with focus on language use may not fare well with teaching materials built on language structure alone (Li 2004; Zhao 2007). Thus the principle in teaching-materials preparation designed to be used with such a pedagogical approach is to have the pragmatic framework as the top layer, with representative pragmatic points as the anchor tasks in the system of language use. Components of language structure, including phonology, vocabulary and grammar, are then used as subsystems attached to the top layer, as indicated in Figure 9.3. As seen from the structure in Figure 9.3, the top layer is the system of language use with pragmatic points as units of presentation; structural knowledge commonly found in traditional 146

Chinese language pedagogy

System of language use: as represented by real-life tasks

Phonology

Vocabulary

Grammar

Figure 9.3  Guiding structure in teaching materials preparation

textbooks are then organized as the second layer. The phonological component, as mentioned at the beginning of this discussion, is a closed system that includes all the sounds and basic tones of the Chinese language. All of them can be distributed in the initial lessons of the textbook in a contextualized situation. The semantic and grammatical components, on the other hand, consist respectively of a selected list based on frequency of usage, or structural complexity if applicable.

2.5.  Teaching training and teaching activities All pedagogical approaches are realized by teachers, even if we follow the student-centered approach. For it needs the teachers to make sure that the class is student-centered. It is from this perspective that we say teacher training plays a crucial role in all pedagogical activities. Teaching CSL has been around for many years, with ‘new’ theories replacing old ones from time to time. New pedagogical approaches would naturally come with new theories, but actual teaching practices are always a little bit behind. This is unavoidable because it takes time for teachers to first understand the theories, then persuade themselves to familiarize new types of activities and exercises with new pedagogical approaches, and finally spend time to either modify their current practice to implement new tactics, or start afresh, to learn new tricks and trial them out in their teaching practice. Due to the need for such a ‘transforming’ process, it takes time for most CSL teachers to cross the gap in pedagogy from the language structure side to the language use side. A systematic approach in teacher training to narrow that gap and make the crossing easier is, therefore, something we could not neglect at this stage (W. Wu 1995, 2014). Key issues in such a training program may be many, as illustrated by examples from one of the in-service training models outlined below. Current practices in teacher training for language instructors tend to give priority to what we call ‘microlevel training’, especially with basic skills and techniques used in the classroom. There is no guarantee, however, that a teacher who performs well in the classroom can also do well if learning happens beyond the classroom, which becomes more and more common in programs focusing on language use. Familiarization and explanation of fundamental concepts associated with new theories should form a major part of any training program. If teachers are expected to willingly conduct all pedagogical activities in programs focusing on language use, then they need to understand the theories and concepts behind such an approach. They need to re-examine their views on the nature of language, the nature of teaching and learning activities, and the aim of language learning for most students. 147

Wu Weiping

e.g. Macrolevel training focusing on fundamental concepts ● ● ● ● ● ●

Teaching Chinese as a Second Language: Concepts, trends and directions Speaking Chinese: A pragmatic approach Teacher training focusing on assessment Teacher training focusing on curriculum design Teacher training focusing on teaching-materials preparation Teacher training focusing on teaching activities

Together with macrolevel training in which the pedagogical approach is designed to improve the ability in language use, issues related to specific skills are still part of the training. In general, this kind of microlevel training can be divided into two categories. One will focus on the pedagogy related to the three language components (sound, word and grammar) and four skills (listening, speaking, reading and writing), as illustrated by the examples below. e.g. Microlevel workshops focusing on language components and language skills ● ● ● ● ● ● ●

Phonology: How to teach sounds and tones Vocabulary: Inter-language as indicated by vocabulary Grammar: How to handle difficulties in grammar under the Pragmatic Framework Listening: Varieties of listening comprehension in the IT era Speaking: Extended discourse for advanced learners Reading: Reading for speaking Writing and script: Stories and learning strategies for Chinese characters

The other category includes some basic techniques in teaching, which has been a major component in most training programs and is always welcome by new teachers. In addition to this, however, some sizeable program or institution also has specialized training related to their own program, known as Program Specific Teacher Training (PSTT). Without such a component, it would not be easy to maintain the characteristics, and therefore effectiveness and reputation of their program (Ji 2006). That is why quite a few of the long-standing and reputable summer CSL programs (e.g. Middlebury, Princeton in Beijing) conduct their own PSTT each year. e.g. Microlevel training focusing on basic techniques and program/course-specific features ● ● ● ● ● ● ● ●

Techniques: How to handle Q and A in class Techniques: Assignments, quizzes and tests Techniques: Student-centered classroom activities Techniques: Internet as a teaching platform PSTT: Characteristics of program focusing on language use PSTT: Special course for individualized training PSTT: Sector-specific language training PSTT: Summary on feedback from class auditing (as a means of quality control)

3.  Concluding remarks The on-going discussion of pedagogical models in the CSL field has witnessed many controversies, doubts and heated debates from time to time because of the lack of consensus 148

Chinese language pedagogy

about what a model should be in the first place (Brown 2001; Ma 2004). Some define a model from the learning environment, some from the procedure and framework for activities, while others would focus on key elements of such a model, thus leading to various models with four, five or six elements (Y. Wu 2012). Needless to say, each and every model may have its own strengths and weaknesses and, depending on the needs of the learning community, are either recommended or rejected by certain programs and institutions. Thus the first concluding remark we can draw is: There is no ideal pedagogical approach for all language programs. CSL in China in its native linguistic and cultural environment, for example, cannot rely on the same approach as CSL in places where Chinese is a foreign language. A CSL program designed to train scholars in Chinese linguistics, on the other hand, will not adopt any pedagogical model for programs focusing on cultivating the speaking ability of diplomats. A close look at various models also reveals the fact that certain features are indeed shared by almost all models no matter which school of theory they come from. So the second concluding remark we can draw is: There are key elements shared by many different approaches, although each model may have its own priority for these elements and each of the elements in any given model may differ in its relative importance within the model. With these two concluding remarks in mind, we will see that any search for the ideal pedagogical model will most likely be in vain. We may also realize that the best and most appropriate model for any CSL program will only become available once we clearly understand the specific traditions and objectives of our own program. Based on such understanding, the best option for us is to create our own model or to modify existing ones to meet our needs. Scholars in assessment once claimed that all tests are specific tests (Douglas 2000); we can also claim here that all good CSL programs are the ones with specific objectives and pedagogical approaches that serve such objectives. With such realization, our attention may then shift to two key questions in our quest for the best pedagogical model: 1. What are the shared elements in CSL pedagogy? 2. What are the factors that affect both the priority and relative importance of these elements in any given model? Based on what has been discussed, either in this chapter or beyond, among the shared elements in any CSL pedagogical model are issues related to what to teach (the materials, either in the form of a traditional textbook or an electronic platform), how to teach (teaching activities and methodology, including teacher training), as well as the criteria and procedures by which the effectiveness and results of teaching and learning are assessed. As for the second question, we can identify at least three factors that will affect what we teach, how we teach and why we assess our teaching and learning the way we do. The first is the primary purpose of the program. The second is the resources available, including time and availability of learning platform demanded by the pedagogical model. The third is not easy to pinpoint, but can be described as the degree to which the professional team can perform as expected, including the knowledge structure, expertise and dedication of the teachers in the team. Finally, we must say that various answers to these two questions can also be found in all the changes we have seen over some key stages in the development of CSL pedagogy. Regarding the mode of teaching, we saw the change from 3Ps focusing on language knowledge to task-based activities. In the scope of teaching, we witnessed the move from structure-only to structureplus contexts. We watched the focus of teaching shift from teacher-centered to 149

Wu Weiping

student-centered, and we have realized that, in assessment, we now rely more and more on information from proficiency tests, rather than just clues from course-based achievement tests. Instead of predicting the trends of development in our endless quest for the best pedagogical model, we can just stick to the hope that all CSL programs will always find the most appropriate model, as long as key players in the program are clear about what they want to do.

References ACTFL (2012) ‘ACTFL Proficiency Guidelines’. Available at http://actflproficiencyguidelines2012.org/ speaking. An, Z. (1988) ‘Chinese Characters Revisited’, in Collection of Papers from the 2nd International Conference on Chinese Language Teaching, Beijing: Beijing Language College Press. Bachman, L. and A. Palmer (1996) Language Testing in Practice, Oxford: Oxford University Press. Bates, E. (1976) Language and Context: The Acquisition of Pragmatics, New York: Academic. Bellassen, J. (ed) (2002) On Characters and Words, Beijing: Peking University Press. Brown, H. D. (2001) Teaching by Principles: An Interactive Approach to Language Pedagogy, Beijing: Foreign Language Teaching and Research Press. Byram, M. (1997) Teaching and Assessing Intercultural Communicative Competence, Clevedon: Mulitlingual Matters. Candlin, C. (1987) ‘Towards Task-based Language Learning’, in C. Candlin and D. Murphy (eds) Language Learning Tasks, Englewood Cliffs, NJ: Prentice Hall International, 5–21. Chao, Yuen-Ren (1968) A Grammar of Spoken Chinese, Berkeley, CA: University of California Press. Chi, T. R. (1996) ‘Toward a Communicative Model for Teaching and Learning Chinese as a Foreign Language: Exploring Some New Possibilities’, in S. McGinnis (ed) Chinese Pedagogy: An Emerging Field, Columbus, OH: Ohio State University Foreign Language Publications, 1–27. Cui, Y. and Chen X. (2000) ‘Analysis of Factors Affecting the Study of Chinese Characters by Learners without Character Background’, in Zhao L. and Huang G. (eds) The Application and Spread of Chinese Characters, Huayu Jiaoxue Press. Douglas, D. (2000) Assessing Languages for Specific Purposes, Cambridge: Cambridge University Press. Ellis, R. (2003) Task-based Language Learning and Teaching, Oxford: Oxford University Press. Feng, S. (2013) Chinese Prosodic Grammar, Beijing: The Commercial Press. Hymes, D. (1972) ‘On Communicative Competence’, in J. Pride and J. Holmes (eds) Sociolinguistics, Harmondsworth: Penguin, 269–93. Ji, C. (2006) ‘On the Construction of CSL Pedagogical Models: With Reference from the Teaching Practice in Middlebury’, Hanyuxuexi 4. Jiang, X. (2007) ‘Chinese Characters Teaching for Western Learners: Separation of Recognition and Writing, Read More and Write Less’, in A. Gu et al. (eds) Chinese Character Cognition and Teaching: Collection of Papers from International Conference on Character Learning by Western Learners, Beijing: Beijing Language and Culture University Press. Krashen, S. D. (1987) Second Language Acquisition and Second Language Learning, Oxford: Pergamon. Levinson, S. (1983) Pragmatics, Cambridge: Cambridge University Press. Li, Q. (2004) ‘CFL Teaching Materials: For Whom?’ Shijie Hanyu jiaoxue 2. Liu, S. (2010) ‘On the Relationship between Character-based and Word-based Pedagogy’, Huawen jiaoxue yu yanjiu 1. Liu, S. (2014) ‘Topic Studies on TCSOL Models’, TCSOL Studies 2. Lu, Q. (2006) A Practical Pedagogical Grammar in Teaching Chinese as a Foreign Language, Beijing: Peking University Press. Ma, J. (2004) ‘Research in Chinese Pedagogical Models’, Yuyan jiaoxue yu yanjiu 1. Norman, Jerry (1988) Chinese, Cambridge: Cambridge University Press. Omaggio, A. C. (1986) Teaching Language in Context: Proficiency-oriented Instruction, 2nd ed., Boston: Heinle and Heinle Publishers, Inc. and Oxford University Press. Shohamy, E. (2001) The Power of Tests: A Critical Perspective on the Uses of Language Tests, London: Pearson Education.

150

Chinese language pedagogy Tao, H. (2011) Working with Spoken Chinese, State College, PA: Center for Advanced Language Proficiency Education and Research (CALPER) Publications. Willis, D. and J. Willis (2007) Doing Task-based Teaching, Oxford: Oxford University Press. Wu, W. (1993) ‘Towards a Theory of Teaching Chinese as a Second Language’, Springfield, VA: ERIC Document Reproduction Service, ED 366 216, Paper based on Presentation at Chinese Language Teachers Association (CLTA) Annual Meeting, San Antonio, Texas. Wu, Weiping (1995) ‘Education of Second Language Teachers: The Link between Linguistic Theory and Teaching Practice’, in Proceedings of the Georgetown University Round Table on Language and Linguistics, 480–97. Wu, Weiping (2006) ‘Pragmatic Points in Teaching Chinese: A Practical Approach’, Chinese Teaching in the World 1: 91–6. Wu, Weiping (2008) ‘Pragmatic Framework and its Role in Language Learning: With Special Reference to Chinese’, in W. Chan, K. Chin, M. Nagami, and T. Suthiwan (eds) Processes and Processorientation in Foreign Language Teaching and Learning, Germany: De Gruyter Mouton. (Reprinted 2011). Wu, Weiping (2014) ‘A Pragmatic Approach in Teacher Training: Ideals and Reality’, in Lee, S., C. Xie, and W. Wu (eds) Linguistics and CSL Teaching: Theories and Practices in Cultivating Pragmatic Ability, Hong Kong: The Commercial Press. Wu, Y. (ed) (2012) Pedagogy in Teaching Chinese as a Foreign Language, Hong Kong: The Commercial Press. Xing, J. (2006) Teaching and Learning Chinese as a Foreign Language: A Pedagogical Grammar, Hong Kong: Hong Kong University Press. Zhang, P. (2006) ‘ “Sentence” Is the Basic Unit of Language’, Hanzi Wenhua 3. Zhao, J. (2007) ‘New Pedagogical Model and Teaching Materials Preparation’, in Collection of Papers from the 8th International Conference on Teaching Chinese.

151

10 Chinese linguistics William S.-Y. Wang Hong Kong polytechnic University, Hong Kong, china

Preamble ‘All the world’s a stage,’ Shakespeare noted in As You Like It, ‘and all the men and women merely players; they have their exits and their entrances.’ The theme of this chapter is Chinese linguistics. We will need a stage, upon which the players are to speak their various languages. Therefore we begin our discussion with a description of the stage, which is the physical geography of China with its tall mountains and great rivers. We will then see that the earliest players to have entered the stage from Africa over 500,000 years ago, Homo erectus, were creatures quite distinct from the ones we see now. The modern players, Homo sapiens sapiens, also from Africa, had two major entrances, by a coastal route from the south, and by a northern route across the mountains of West Asia. This dual origin of the diverse Chinese peoples from ancient times is clearly reflected in the genes they carry and the many languages they speak. We have continuous specimens of the languages of China that reach back in time over 3,000 years, which is unique among the languages of the world. Concern with language may be traced to the philosophical observations of Kongzi1 (Confucius 孔子, 551–479 bce), who emphasized the pragmatic importance of words with the term 正名 ‘correct names’. In contrast to Shakespeare’s Juliet, who said famously, ‘What’s in a name? that which we call a rose by any other name would smell as sweet’, Kongzi saw words as the all-important foundation of correct behavior, a view he expressed in the following injunction: 名不正则言不顺; 言不顺,则事不成; ……2 ‘When names are not correct, what is said will not sound reasonable; when what is said does not sound reasonable, affairs will not culminate in success;  .  .  .’ There are many modern examples of Kongzi’s concern with using the right word. During a period of racial tension in the USA, it was critically important to distinguish the words ‘Negro’, ‘Black’, ‘Afro-American’. Sensitivity to women’s rights prompted the introduction of the title of ‘Ms.’ in preference to the traditional ‘Mrs.’ and ‘Miss’, which necessarily reveal marital status. Similarly, two new sinograms with the female radical, 妳 and 她, to 152

Chinese linguistics

complement the traditional 你 and 他, were invented in early twentieth-century China to encourage gender awareness, even though no new pronunciations were proposed. Kongzi’s discussion was a starting point for Xunzi’s (荀子, b. 313 bce) view that names, or words, are created by social conventions. It is a remarkable coincidence that Xunzi stated this view at approximately the same time as Plato (429–347 bce) did in Ancient Greece.3 Here are his words, expressed in three pithy sentences which appear in Section 22 of his anthology, titled〈正名篇〉to recall Kongzi’s injunction: 名无固宜,约之以命; 约定俗成谓之宜,异于约则谓之不宜。 名无固实,约之以命实,约定俗成,谓之实名。 名有固善,径易而不拂,谓至善名。 Roughly translated, they mean: Words have no intrinsic correctness; their correctness is established by convention. Words have no intrinsic content; their content is given by convention. Words do have intrinsic appropriateness; those which are direct and not misleading   are appropriate words. Since these words are the first expression of a fundamental property of words, and therefore of language, they are often taken to be the beginning of linguistics in China. The language of the ethnic group that has been dominant for well over two millennia is called Hanyu, named after the Han dynasty that flourished around the same time as the Roman Empire. The bulk of this chapter will be on Hanyu, its history and structure. But for now let us set the stage upon which the languages and peoples of China are to act out their destinies.

Setting the stage Some 50 million years ago, the continental plate on which India is now situated drifted northward and collided with the much larger continental plate we now call Eurasia. This collision forced an upward projection of the landmass, resulting in a mighty range of mountains, the Himalayas, a word of Sanskrit origin with a stem meaning of ‘snow’. The tallest peak in the range was marked on the maps compiled during the long reign of Emperor Kangxi (1661–1722) as Zhumulangmafeng 珠穆朗玛峰, based on the Tibetan word, Qomolangma,4 which may be translated as ‘Goddess-Mother’. The peak is almost 9,000m above sea level, and continues to rise. The physical stage upon which the peoples and languages of China play out their destinies was thus set in major outline a long time ago. It is marked by an ultra dry western part of high plateaus, since the moist air currents from the oceans down south are blocked by tall mountains. At the same time, the land has a pronounced west–east tilt, and the two great rivers which originate from these mountains, the Huanghe 黄河 ‘Yellow River’ and the Changjiang5 长江 ‘Long River’, flow eastward to the Pacific, giving life to the land in between. The Last Glacial Maximum peaked during 25,000 and 20,000 bp. Much of the Earth’s water was locked in huge glaciers; sea levels were low, exposing many passages which are now under water. The northeastern corner of Asia was joined by land to the Americas, thus providing passage for Asians to cross over Beringia and colonize the new continents. Recent 153

William S.-Y. Wang

investigations by anthropologists indicate that the Asians may have stopped for quite a while at Beringia before entering the New World and beginning the long trek down to the tip of South America.6 Linguists have begun exploring the genetic relations between Chinese and far-flung languages across continents.7 As sea levels were lowered by the Last Glacial Maximum, Taiwan was connected to the Asian mainland by land, allowing ancestral Austronesian speakers to reach Taiwan on foot. After sea levels rose again, many of these peoples sailed out to circle the earth: east to distant islands in the South Pacific, such as New Zealand and Samoa, and west to Madagascar in the Indian Ocean. This scenario is supported by the fact that Taiwan had the greatest diversity of Austronesian languages until recent decades, when many of these languages became extinct. Taiwan was the launching pad, so to speak, for the great diaspora of this great family of Austronesian languages.8 Similarly, the land bridges allowed the early speakers of Altaic languages to cross over from Korea to Japan, while leaving descendants in northern China who spoke languages that have evolved into modern Uighur, Mongolian, Manchu, Evenki, etc. These Altaic peoples have played important roles in China’s cultural history. They were especially prominent in two great dynasties: the Yuan9 元 (1206–1368) ruled by the Mongols, and the Qing 清 (1616–1911) ruled by the Manchus. A recent census (2010) lists their populations to be over 5 million for the Mongols and over 10 million for the Manchus. Altaic languages are distinguished from most of the languages in China in many ways. Syntactically the verb occurs at the end of the sentence, rather than in the middle. Morphologically the structure is agglutinative in having strings of suffixes after the verb stem. Phonetically Altaic languages are distinguished by vowel harmony, whereby there are restrictions on which vowels may co-occur in the same word. However, this feature is no longer so obvious in the Altaic languages spoken in China due to extensive lexical borrowings from neighboring languages. Returning to the present geography, the yellow, muddy color of the Huanghe is caused by the fact it flows out of the Loess Plateau, carrying with it large amounts of wind-borne clay dust. This sedimentation causes the river bed to build up quickly, leading to massive flooding when not properly controlled. Shortly after it descends from the Loess Plateau, the Huanghe makes three right-angle turns, first northeastward, shortly after Lanzhou 兰州, then eastward above 40 degrees north, then southward at around 110 degrees east, enclosing in this giant loop the Ordos region of Inner Mongolia. In its southward course, it separates two provinces with nearly homophonous names: Shaanxi 陕西 to its west and Shanxi 山西 to its east. Lastly, at around 35 degrees north, it turns sharply to resume its eastward journey, this time separating Shanxi to its north and Henan 河南 to its south. At this last turn, the Huanghe is joined by the Weishui 渭水, forming the bottom leg of an immense rectangle. This general region surrounding where the Weishui flows into the Huanghe and turns east is often referred to as Zhongyuan 中原 ‘Central Plains’. It is a region whose significance has been often stressed by historians with terms like ‘cradle of China’. Two cultural centers in this region have played special roles during the two millennia 1000 bce–1000 ce: Xi’an 西安 in the west and Luoyang 洛阳 in the east, though they had different names at different historical times. However, recognizing the importance of the Central Plains should not obscure the fact that the Chinese civilization of today is the result of at least 6,000 years of ethnic and linguistic amalgamation, with some sources quite distant from this region. From its headwaters in the western highlands of Qinghai 青海 and Xizang 西藏 (Tibet), the Changjiang flows southward in parallel with the Nujiang 怒江 and the Lancangjiang 澜沧江. Whereas the latter two rivers flow into Southeast Asia, to become the Salween and 154

Chinese linguistics

the Mekong respectively, the Changjiang is blocked in its path by some massive granite boulders in the vicinity of Shigu 石鼓 in Yunnan 云南. It makes some small hairpin curves and settles on an eastward journey, staying quite close to 30 degrees north. This latitude of 30 degrees north is of special interest in the cultural geography of China, being the location of several major cities, from Lhasa in Xizang, to Chongqing 重庆 in Sichuan 四川, to Shanghai on the Pacific coast. In addition to the two great rivers Huanghe and Changjiang, another important waterway should be mentioned: the Jinghang Da Yunhe 京杭大运河, also called the ‘Grand Canal’.10 Construction of this artificial river began in 605 ce. At over 1,700km, it joined together the regions of the two great rivers, Beijing in the north and Hangzhou 杭州 in the south. In its heyday, the Yunhe was the longest canal in the world, and won admiration from many international travelers, including Matteo Ricci11 (1552–1610) from Italy. Nowadays, however, only its southern half remains consistently navigable. Wide, fast-flowing rivers often created boundaries to separate peoples in early times, when they lacked the ability to cross them. At the same time, they also provided routes to travel along, whether by primitive rafts or boats or simply by foot along their banks, since water is essential for life. It has been speculated that many ethnic groups could have started from the western highlands and migrated along the two great rivers through the millennia. Such a scenario seems particularly relevant for the Austric hypothesis, which includes two major branches – Austronesian and Austro-Asiatic. While many scholars have argued for the historic unity of these two branches of languages, their modern forms are quite different. Austronesian languages typically begin their sentences with verbs, and have morphemes which are polysyllabic. The morphemes in Austro-Asiatic languages, on the other hand, are often monosyllabic and distinguished by lexical tones, much as in modern Chinese. In a pioneering paper of 1976, Norman and Mei proposed some lexical evidence for Austro-Asiatic in South China, particularly with respect to the names of the two great rivers. Although some of their evidence has been called into question recently (Zhang 1998), the reconstruction of language contact in prehistoric China is of obvious importance and should continue in a multidisciplinary perspective, in collaboration with archeology and genetics. This issue on Chinese river names is further discussed by Takashima (2012). The Bai Yue 百越 peoples mentioned in traditional history books were presumably largely Austro-Asiatic in their composition; they covered large regions of South and Southeastern China, extending into northern Vietnam. The Zhuangzu 壮族 of Guangxi 广西 descended from one of the Yue 粤 peoples. With a population over 16 million in 2010, it is the largest ethnic minority in China today, second in population size only to the Hans. There are some precious data on an early Austro-Asiatic language, preserved in the Yue Ren Ge 越人歌 ‘Song of the Yue Boatman’, transliterated with Chinese writing accompanied by a translation. The song was composed in the form of Chuci 楚辞, a genre of poetry from the first millennium bce. The Chinese linguist Zhengzhang Shangfang 郑张尚芳 (1991) analyzed this song with the phonology of Old Chinese (OC), which he reconstructed, and compared the verses with Written Thai; written language is typically more conservative than varieties of the spoken language. To give an idea of this work, let us consider the first line in the Yue Ren Ge, which appeared in the classical text as follows: 滥兮抃草滥. As Table 10.1 shows, pronouncing this line in Putonghua (PTH), the line would be: làn xī biàn cǎo làn. Neither the string of sinograms nor the Putonghua pronunciation makes any sense. However, Zhengzhang replaced the Putonghua pronunciations with his Old Chinese reconstructions for these sinograms, and got: OC *ɦgraams ɦee brons ts’uuɁ ɦgraams. It is 155

William S.-Y. Wang Table 10.1  Song of the Yue Boatman〈越人歌〉 Line 1 of Yue Ren Ge Putonghua Old Chinese Translation via Written Thai

滥 làn *ɦgraams evening

兮 xī ɦee particle

抃 biàn brons joy

草 cǎo ts’uuɁ meet

滥 làn ɦgraams evening

interesting to note from Table 10.1 that PTH làn and biàn both have a falling tone, which derives from historical tone III (Qu sheng 去声), and PTH cǎo has a low tone, which derives from historical tone II (Shang sheng 上声). Zhengzhang reconstructs tone III with a final *-s and tone II with a final *-Ɂ; both reconstructions are consistent with the views of other scholars in the field of historical Chinese phonology, such as those discussed in Baxter (1992). In addition to providing the OC reconstruction here, Zhengzhang made the contribution of comparing the reconstructions with phonetically similar syllables in Written Thai, which is ‘the most anciently attested form of Thai and other languages of the Tai group’ (1991: 160). By using these methods, he was able to meet the challenge presented by the classical text, verify that the song was in an Austro-Asiatic language, and translate the first line literally as: ‘evening, particle, joyful, to meet, evening’. It expressed the joy of the boatman to meet a Han prince that evening. The classical text indicated that the boatman was from the southern region of Chu 楚. This is reflected in the use of the particle xī as the second syllable in the line; this particle was abundantly used in Chuci, which are famously represented by the writings of Qu Yuan 屈原 (340–278). Zhengzhang concluded his investigation with this remarkable observation: ‘Although the words of the Yue People’s Song may be compared with Siamese, the verses would be easily understood by modem Thai speakers’ (1991: 167). Unfortunately, we have no ancient text for Austronesian comparable to the Yue Ren Ge for Austro-Asiatic. Many Austronesians must have settled in Taiwan many millennia ago, before the island became separated from mainland China by so wide a strait of water. The antiquity of their settlement is in part evidenced by how deeply divergent the Austronesian languages have become in Taiwan; a map of the historical distribution of these languages is shown in Figure 10.1.12 Many experts now believe that Taiwan was the last stop before the Austronesians went on to colonize far flung islands of the world, from Southeast Asia all the way eastward to Hawaii, and all the way westward to Madagascar. They were indeed the finest sailors of the ancient world.

Early ancestors in China There is general consensus among scientists in several disciplines that we, Homo sapiens sapiens (anatomically modern humans, AMH), are the only living representatives of the biological genus Homo; other species in this genus, such as the Neanderthals in Europe and Western Asia, the Denisovans in Siberia, and Homo floresiensis in Indonesia, all became extinct long ago. Such knowledge is based on the integration of a variety of evidence offered by many disciplines, including Comparative Anatomy, Physical Anthropology, Archeology, and especially Molecular Genetics. We also know that our species, Homo sapiens sapiens, emerged in Africa. Some members of this species left the homeland to colonize the world 156

Chinese linguistics

Basay

Luilang

Kulon

Northern

Qauqaut

Ts’ole’&Sqoleq

Southern

Ts’ole’

Ts’ole’

Atayal

Ts’ole’

Papora

Kavalan Trobiawan

Sqoleq

Saisiyat Taokas

Ketangalan

Pazih

Seediq Ts’ole’&Sqoleq

Babuza

Sakizaya

Ts’ole’

Thao

Northern

Northern

Tavalong-Vata’an

Hoanya

Duhtu Tfuea

Amis

Tsou

Tapangu

Central

Takupulan

Kanakanavu Taivoran Saaroa

Central

Formosan Languages Amis Atayal Babuza Basay Bunun Hoanya Kanakanavu Kavalan Kulon Paiwan Papora Pazih Puyuma Qauqaut Rukai Saaroa Saisiyat Siraya Taokas Thao Tsou Batanic(Philippines) Language Yami

Bunun

Southern

Southern

Siraya

Puyuma Rukai

Nanwang

Makatau

Lütao

Southern Amis

Paiwan Southern Amis Southern Amis

Yami Lanyü

Southern Amis

Figure 10.1  Austronesian languages in Taiwan, courtesy of Prof. Paul Li

over 100,000 years ago. Recent successes with ancient DNA research indicate there was a good deal of interbreeding between AMH and other ancient populations,13 though so far ancient fossils from China have not yet been analyzed for prehistoric mixing. The success of these early peoples to colonize the world is due to the emergence of language, which qualitatively enhanced their ability to think, to communicate, and to cooperate. The evolution of languages is always intricately interwoven with the evolution of the peoples who speak them, each enhancing the development of the other. Among the extinct species is Homo erectus, including the famous Peking Man 北京猿人, found in the caves near the village Zhoukoudian 周口店, on the southwest outskirts of 157

William S.-Y. Wang

Beijing. Zhoukoudian represents an extremely important archeological site in the study of human evolution because of ‘repeated human visitations, through 11 major depositional units that span a period of some 300,000 years, its abundance of archeological and faunal residues, and, of course, its unique sample of Homo erectus remains’.14 Peking Man lived some 600,000 years ago (Shen et al. 2009), and was among the earliest users of fire – a trait not found in any other animal. Since the discovery at Zhoukoudian in the 1920s, human fossils, stone tools, and bone artifacts have been unearthed at numerous other sites in China, reaching back well over a million years (Wu and Olsen 1985). A notable recent find is at the Zhirendong 智人洞 in Guangxi of a human mandible, which is the oldest fossil from Homo sapiens sapiens outside of Africa (Liu et al. 2010). Also worthy of mention here is the remarkable discovery of a set of flutes, made from crane’s leg bones, at Jiahu 贾湖 in Henan (Zhang et al. 1999). The authors call these ‘the oldest playable musical instruments’, and the audio file of a Chinese song played on these flutes can be accessed on the Internet via their publication. Also discovered at Jiahu are some potsherds with inscriptions on them which are highly suggestive of writing.15 These discoveries give us a glimpse of Neolithic culture in China, nearly 9,000 years ago, which was unexpectedly rich. According to current consensus, AMH emigrated from Africa in many waves, beginning over 100,000 years ago, settling first in Asia. From Asia, the human diaspora spread successively westward to Europe, southward to Australia, and across the Beringia to the Americas. For most of these 100,000 years, our ancestors lived as roving bands of hunters and gatherers, their movements driven by climatic conditions and the availability of fauna and flora for food. The little we know of their world is based on the material traces they left behind, primarily in the form of fossils and stone tools.16 Contrasting with the Out-of-Africa hypothesis, in which the emigrants from Africa completely replaced the earlier archaic Homo species, some scholars prefer an alternative scenario in which there was some interbreeding; cf. Abi-Rached et al. (2011). These latter voices favor a Multi-Regional hypothesis, in which archaic humans contributed to the gene pool alongside the AMH, who left Africa 100,000 years ago. Many decades back, Franz Weidenreich, who oversaw some of the first Zhoukoudian excavations in the early twentieth century, lectured on the anatomical continuities between ‘Peking Man’ and the modern Chinese.17 Wu Xinzhi 吴新智, a leading authority on Zhoukoudian fossils, continues to explore this line of thought.18 Interestingly, recent excavations in Zhoukoudian, at a site called Tianyuandong 田园洞, yielded fossils which are more compatible with the Multi-Regional hypothesis.19 Currently the evidence is accumulating that although our modern genes are mostly inherited from ancestors who left Africa 100,000 years ago, there have been admixtures with other archaic species within the Homo genus before they became extinct. Archeologically, of great interest is the very recent discovery of pottery remains in Jiangxi 江西20 which date back some 20,000 years, many millennia before agriculture was invented. After the onset of the Holocene period, the human condition changed dramatically with the advent of agriculture. With the raising of crops and domestication of animals around 10,000 years ago, ancient peoples settled down. With the systematic production of food, they became more numerous, and villages eventually grew into towns and cities. They started making pottery for keeping and transporting food and water, and began to mark the pottery with symbols to identify the maker or the owner. These marks are the early precursors of writing, which emerged several millennia later. There were numerous Neolithic villages all over China which developed relatively independently for many millennia.21 These include Dawenkou 大汶口 in Shandong 山东 and 158

Chinese linguistics

Zengpiyan 曾皮岩 in Guangxi, and many others. The best known of these many sites is the Banpo 半坡 village situated near modern Xi’an, where an on-site museum has been built to preserve and exhibit its remains. Banpo, dating back some 7,000 years, is an example of the Yangshao 仰韶 culture, known for its painted pottery, termed Caitao Wenhua 彩陶文化 ‘colored pottery culture’ in Chinese. ‘Yangshao’ is the name of a village in Henan, where the archeological site was first discovered in 1921 by the Swedish archaeologist/geologist Johan Gunnar Andersson22 (1874–1960). This discovery is regarded as the beginning of archeology in China. Shortly after the Yangshao discovery, Neolithic pottery of several millennia later was unearthed in Shandong, where the pottery was more delicate and painted black. This became known as the Longshan 龙山 culture, called Heitao Wenhua 黑陶文化 ‘black pottery culture’ in Chinese. It is interesting that an average linkage analysis of the cranial indices reported on nine fossil skulls unearthed at these sites revealed a major separation of north and south.23 This would suggest that the north/south division among the peoples of China has a very early origin. This division is supported by an extensive immunological study reported by Zhao and Lee (1989). Together with recent advances whereby DNA can be extracted from fossils to examine their genetic affinity,24 our understanding of the phylogeny of Asian populations has deepened significantly, thanks to the contributions from molecular genetics. As the scope of these villages increased over time, they met and interacted with greater regularity and frequency, eventually connecting together into one integrated cultural sphere some 6,000 years ago, when there began sustained contact and sharing of cultural innovations. The cultural sphere formed in this way was dubbed the ‘initial China’, by the late archeologist Chang Kwang-chih 张光直, whose authoritative volume The Archeology of Ancient China in its several editions guided the field and trained generations of students for several decades. Figure 10.2 is reproduced from p. 235 of the 4th edition of Chang’s classic book, published in 1986.25 Chang marked off nine Neolithic regions for 6,000 bp, and drew arrows to highlight the interaction among them. The identified sites going down the Pacific coast are: Hongshan 红山, Tuzhu 土珠, Dawenkou, Majiabang 马家浜, Hemudu 河姆渡, Tanshishan 昙石山, Fengbitou 凤鼻头, and Shixia 石峡. The remaining three sites are Shanbei 山背, Daxi 大溪, and Yangshao. Yangshao is presumably the best known of these cultures, as mentioned earlier, in part because it was the earliest discovered. In the words of Chang, it was 6,000 years ago26 that ‘these cultures became closely linked, and they share common archeological elements that bring them into a vast network within which the cultural similarities are quantitatively greater than without. By this time we see why these cultures are described together: not just because they are located within the borders of present-day China, but because they were the initial China’ (1986: 234; emphasis added). Later on, on p. 410, Chang stresses the organic whole of the Chinese culture thus united by the charming metaphor: ‘When the Weishui River valley sneezed, as it were, the Lake Taihu region caught cold’.

Shells and bones The English word ‘China’ derives from the name of the first imperial dynasty in China, Qin 秦, short-lived as it was. The man who accomplished the formidable task of uniting the warring kingdoms of the third century before the Common Era into one empire is called Qin Shi Huangdi 秦始皇帝 ‘Qin First Emperor’. The world knows of him through the fabulous terracotta army unearthed in his tomb near modern Xi’an. The name ‘China’, in its 159

William S.-Y. Wang

Hung-shan

Tu-zhu

Ta-wen-k’ou Yang-shao

Ma-chia-pang Ta-hsi

Ho-mu-tu Shan-pei

T’an-shih-shan Shih-hsia Feng-pi-t’ou −400

0 100 200 300 400 500 600 km

Figure 10.2  The ‘initial China’

various phonetic variants, is used in most European languages. A notable exception is the name for ‘China’ in Russian, Kitai, which derives from a people that ruled northern China during the Liao dynasty. Their name is now spelled Qidan 契丹 in China; also Khitan in older literature. Table 10.2 gives a thumbnail sketch of the history of China over some four millennia. There is still some controversy among specialists on the nature of the Xia dynasty, its location and its material remains. However, there is abundant archeological and textual evidence on the Shang dynasty, dating back over three millennia. Whereas books by earlier historians tell of this dynasty, the major breakthrough toward understanding it came only at the beginning of the twentieth century. It came when a scholar in Beijing named Wang Yirong 王懿荣 160

Chinese linguistics Table 10.2  Simplified chronology of Chinese history Xia Shang Zhou, Western _____ , Eastern Qin Han, Western _____ , Eastern Three Kingdoms Jin Nanbei Chao Sui Tang Wudai Shiguo Song, Northern _____ , Southern Liao Xixia (Western Xia) Jin Yuan Ming Qing Republic of China People’s Republic of China

21st c. to 16th c. bce 16th c. to 1046 bce 1046 to 771 bce 770 to 221 bce 221 to 206 bce 206 bce to 25 ce 25 to 220 220 to 280 265 to 420 420 to 581 581 to 618 618 to 907 907 to 979 960 to 1127 1127 to 1279 907 to 1125 1038 to 1227 1115 to 1234 1206 to 1368 1368 to 1644 1616 to 1911 1912 to 1949 1949 to present

夏 商 西周 东周 秦 西汉 东汉 三国 晋 南北朝 隋 唐 五代十国 北宋 南宋 辽 西夏 金 元 明 清 中华民国 中华人民共和国

bought some ‘dragon bones’ from local stores for medicinal purposes. These bones were primarily turtle shells and ox scapula, and some bore graphs on them. From his knowledge of ancient sinograms inscribed on bronze ware, he recognized that many of these graphs were ancient sinograms as well. For the study of the languages and peoples of China, this was a momentous event. Inscriptions on bronze ware, mostly on ritual vessels and weapons of war, are called Jinwen 金文; the ‘jin’ in the name means ‘gold’ in its narrow sense, but ‘metal’ in its broad sense. Inscriptions on shells and bones are called Jiaguwen 甲骨文, where the ‘jia’ refers to the turtle shells and the ‘gu’ refers to various bones. After Wang Yirong made the discovery, numerous specimen of Jiaguwen have been collected, mostly from archeological excavations in the city of Anyang 安阳 in the province of Henan. Anyang turned out to be the last capital of the Shang dynasty. Decades of research on Jiaguwen has given us much first-hand information on the languages and peoples during the Shang dynasty, which are highly valuable for understanding early societies in human civilization. The shells and bones were inscribed for the purposes of divination, not unlike the Delphic oracles in Ancient Greece. For this reason, Jiaguwen is also called Oracle Bone Inscription (OBI) in English. The inscriptions contained questions to be divined, regarding weather, warfare, childbirth, etc., and often recorded related events. From these inscriptions we get to know about the various ethnic groups that were active in the Central Plains area. We are also able to reconstruct a complete lineage of the many kings of the Shang dynasty, as well as of some of the important personages. Of the latter, particular mention should be made of a remarkable woman warrior, named Fuhao 妇好, consort of Wuding 武丁, one of Shang’s most successful kings. 161

William S.-Y. Wang

Figure 10.3  Oracle bone inscription concerning Fuhao’s childbirth, reproduced with permission from City University of Hong Kong Press

Space limitations do not permit much discussion of Jiaguwen here. But it is useful to see a sample rubbing in Figure 10.3, which is a divination on Fuhao’s childbirth.27 The sinograms were inscribed on a turtle plastron in this case, prepared beforehand for divination. After the question was inscribed, heat was applied to the plastron to induce cracks. The diviner, sometimes the king himself, would interpret these cracks as answers from heaven. The question here began in the rightmost column, and the columns move leftward in this case. Of special interest is the third sinogram down in the rightmost column, which looks like , but is now written 卜, pronounced bu in Putonghua but buk in Cantonese. This sinogram actually means ‘to divine’; it has the shape of cracks made on the shell, and a pronunciation that resembles the sound the shell makes when it cracks. Another sinogram to note here is the second one down in the leftmost column, which is made up of two parts. The left part resembles a stick figure kneeling, facing to the right; this part is now written 女, meaning ‘woman’. The right part of this sinogram is now written 子, meaning ‘child’. The modern sinogram made up of these two parts, ‘woman’ and ‘child’ is 好, meaning ‘good’, and is pronounced hao in Putonghua. In fact, it is the second sinogram in the name of the woman Fuhao, with whose childbirth the king is concerned in this divination. The main lesson from Jiaguwen for Chinese linguistics is that the sinograms and their principle of formation as well as the grammar of the Shang inscriptions can be clearly seen to be ancestral to modern Hanyu. Since these inscriptions represent the earliest texts in the language, the language itself is called Old Chinese,28 or Shanggu Hanyu 上古汉语.

Poems and reconstructions There are two sets of cultural treasures which shed light on the roots of Chinese civilization. One is Jiaguwen, which lay buried underground for three millennia before it was discovered 162

Chinese linguistics

accidentally. The other set is Shijing 诗经, a collection of songs, poems and odes, which also dates back to the Shang and Zhou dynasties. The collection is sometimes said to have been organized by Kongzi, though there is little evidence for this belief. On the other hand, he spoke of it often in his teachings, sometimes with reverence. For instance, he proclaimed: 《诗》   三百,一言以蔽之,曰: 思无邪 ‘The 300 poems in the Shijing can be covered by one phrase: have no improper thoughts.’ Here Kongzi referred to the Shijing with just the sinogram 诗 shi ‘poem’. The term now commonly used is《诗经》, which includes the sinogram jing 经 to show its high cultural status. This sinogram is also used in a religious context; for instance, the Bible is called Shengjing《圣经》, and the Koran is called Gulanjing《古兰经》. There are actually 305 poems in the commonly used version, ranging in length from just a few lines of a lover’s lament to several pages of a historical chronicle. These poems provide us with a precious window on the diverse cultures of Ancient China, not just from the viewpoint of affairs of state, as with Jiaguwen, but often with the everyday voices of common folk. The particularly valuable aspect of the Shijing for Chinese Linguistics is the information it provides us indirectly on how the language may have sounded 3,000 years ago. It does this by allowing us to pool together sinograms which rhyme with each other. We may assume that most of the time sinograms that rhyme were pronounced with a common tone and a common syllable final, i.e. the nuclear vowel of the syllable plus any consonant or consonants that may follow it. The systematic investigation of the pronunciation of the Shijing from a scientific viewpoint was pioneered by a remarkable scholar named Chen Di 陈第 (1540–1620). Whereas earlier scholars were at a loss to explain why the Shijing poems no longer rhymed in their pronunciation, Chen clarified the situation with these famous words: 盖时有古今,地有南北,字有更革,音有转移,亦势所必至。 Time, there is past and present. Space, there is south and north. Words change, pronunciations shift. All are in the nature of things. Chen went on to group the sinograms in the Shijing in terms of their phonetic relations, mindful of the type of evidence that is relevant according to the decisions made – whether it is internal to the Shijing itself, or external to it from other considerations. In blazing out a trail for generations of Chinese linguists to follow, the role he played in China is similar to that of William Jones (1746–94) in the West, who established the very successful paradigm of Indo-European linguistics, basing itself on the comparative method. It was Jones’s contribution, encapsulated in his Calcutta lecture of 1786, that launched the modern science of linguistics. The most significant limitation in Chen’s scholarship, and that of the brilliant Qing dynasty scholars who followed him, was restricting themselves to only textual data from ancient times, instead of availing themselves additionally of the live data of the numerous Hanyu dialects and minority languages which can shed light on the Shijing. Nonetheless, their contributions laid the foundation in discovering the abstract rhyme categories for later researchers to link up to modern data. One scholar who early brought insights from Western linguistics to the study of the Chinese language was Bernhard Karlgren (Gao Benhan 高本汉, 1889–1978) from Sweden, a younger colleague of the J. G. Andersson mentioned earlier. His PhD dissertation includes a pronouncing dictionary of many dialects, providing valuable data for the reconstruction of 163

William S.-Y. Wang

Old Chinese, and illustrating the importance of fieldwork. His early work Études sur la phonologie chinoise was translated into Chinese under the title《中国音韵学研究》by a trio of Chinese linguists, who were to exert a formative influence on the development of Chinese linguistics in the twentieth century: Y. R. Chao 赵元任 (1892–1982), F. K. Li 李方桂 (1902–87), and C. P. Luo 罗常培 (1899–1958). Other Western scholars who have contributed significantly to the reconstruction of Old Chinese and Middle Chinese include André-Georges Haudricourt from France and E. G. Pulleyblank from Canada. Chinese linguists who combine the training of traditional Chinese linguistics with the comparative method to work in this area include Zhengzhang Shangfang, mentioned earlier in connection with Yue Ren Ge, Tung Tungho 董同龢, Mei Tsulin 梅祖麟, Ting Pang-hsin 丁邦新, and Gong Hwangcherng 龚煌城, who was an outstanding scholar on the extinct Xixia language. US linguist William Baxter (Bai Yiping 白一平) has done the field of Old Chinese phonology a great service by publishing in 1992 a comprehensive handbook, putting the various contributions in a balanced historical perspective. However, Old Chinese phonology remains a dynamic area of scholarly explorations. Ho has provided an incise critique of the central issues in his review article of 2015. In the sample from the Shijing we will look at below, we will use Baxter’s reconstructions. The best known of the 305 poems in the Shijing is probably the first one, called Guan Ju〈关雎〉 . In his Analects VIII: 15, Kongzi praised this poem in these words: 师挚之始,〈关雎〉之乱,洋洋乎盈耳哉！ which has been translated as ‘When Zhi the Master Musician begins to play and when Guan Ju comes to its end, how the sound fills the ear!’29 The poem has a very simple theme, of a young man seized by the pangs of unrequited love. The ‘guan’ in the title is a syllable imitating the sound of bird cry; the ‘ju’ is the first syllable of the name of the bird jujiu 雎鸠. The setting is an idyllic one: an islet in a river, with water plants floating around it. Table 10.3 gives two authoritative translations. Arthur Waley was an English sinologist based in London who translated numerous classics from Chinese and Japanese into English. Yang was a professional translator of many Chinese classics into English; he and his collaborators were based in Beijing. The translations differ, as is to be expected when poetry is involved, but not by too much for this poem. On the other hand, there are poems for which experts differ significantly in their interpretations. Even though the literature on the Shijing is already extensive, there are numerous issues on it for which there is no consensus, simply because we do not know enough about Ancient China as yet. As we can see from Table 10.3, the poem is divided into five verses, each verse containing four lines, and each line containing four sinograms. The rhyme scheme in verses 1 and 3 is AAxA, with the rhyme on the last syllable. The rhyme scheme in verses 2, 4, and 5 is xAxA; however, since the last syllable in these lines is a grammatical particle, i.e. zhi, which probably was unstressed, the rhyme falls on the next-to-last syllable. I have italicized these unstressed syllables in the table to make clear this point. The rhyming syllables are presented in Table 10.3 in their Old Chinese pronunciations, preceded by *–, as these are reconstructed in Baxter’s Handbook. The Old Chinese reconstructions do not indicate lexical tones; experts are not agreed on whether Old Chinese made use of lexical tones. On the other hand, these syllables are marked in the Putonghua readings, according to the tone the sinogram had in Middle Chinese, the language reflected in a rhyme dictionary compiled around 601 ce during the Sui dynasty. 164

关关雎鸠 在河之洲 窈窕淑女 君子好逑

参差荇菜 左右流之 窈窕淑女 寤寐求之

求之不得 寤寐思服 悠哉悠哉 辗转反侧

参差荇菜 左右采之 窈窕淑女 琴瑟友之

参差荇菜 左右芼之 窈窕淑女 钟鼓乐之

1.1 1.2 1.3 1.4

2.1 2.2 2.3 2.4

3.1 3.2 3.3 3.4

4.1 4.2 4.3 4.4

5.1 5.2 5.3 5.4

〈关雎〉   Guan Ju

cen ci xing cai zuo you maoIII zhi *mawks yao tiao shu nv zhong gu leIII zhi *grawk

In patches grows the water mallow To left and right one must choose it Shy is this noble lady With gongs and drums we will gladden her

In patches grows the water mallow To left and right one must gather it Shy is this noble lady With great zither and little we hearten her

Sought her and could not get her Day and night he grieved Long thoughts, oh, long unhappy thoughts Now on his back, now tossing to his side

qiu zhi bu deIV *tik wu mei si fuIV *bjik you zai you zai zhan zhuan fan ceIV *tsrjik

cen ci xing cai zuo you caiII zhi *sriɁ yao tiao shu nv qin se youII zhi *wjiɁ

In patches grows the water mallow To left and right one must seek it Shy was this noble lady Day and night he sought her

‘Fair, fair’ cry the ospreys On the island in the river Lovely is this noble lady Fit bride for our lord

Translation by Arthur Waley

cen ci xing cai zuo you liuIv zhi *Crju yao tiao shu nv wu mei qiuIv zhi *grju

guan guan ju jiuIu *krju zai he zhi zhouIu *tju yao tiao shu nv jun zi hao qiuIv *grju

Table 10.3  Guan Ju, first poem of Shijing

Short and long the floating water plants Left and right you may collect them Gentle and graceful is the girl He’d like to marry her, bells and drums beating

Short and long the floating water plants Left and right you may gather them Gentle and graceful is the girl He’d like to wed her, the qin and se playing

When the courtship has failed Awake he thinks of her and in his dreams Filled with sorrowful thoughts He tosses about unable to sleep

Short and long the floating water plants Left and right you may pluck them Gentle and graceful is the girl Awake he longs for her and in his dreams

Merrily the ospreys cry On the islet in the stream Gentle and graceful is the girl A fit wife for the gentleman

Translation by Yang (1983)

William S.-Y. Wang

We will simply note here that the Iu and Iv tones in verses 1 and 2, have developed into tones 1 and 2 in Putonghua, pronounced high level and high rising respectively. Middle Chinese tone IV, as found in verse 3, occurred on syllables which end in –p, –t, or –k, as we see in the Old Chinese reconstructions. These endings are still preserved in some modern dialects, such as Cantonese, as in the pronunciation of the integers sap 十 ‘ten’, cat 七 ‘seven’, luk 六 ‘six’; but in Putonghua they have all become lost. The syllables which have lost these endings have merged into other tones; in this poem, de and fu have merged with Iv while ce has merged with III. Finally, note that the tone II in verse 4 has developed into Putonghua tone 3, a low tone, and tone III in verse 5 into Putonghua tone 4, a high-falling tone. As for the final –Ɂ in the Old Chinese form in verse 4, and the final –s in verse 5, these are both consistent with Zhengzhang’s reconstructions we saw earlier in Table 10.1. As one last observation on how Old Chinese was pronounced, we note the presence of consonant clusters in the reconstructed forms in this poem. In addition to the ending *–ks in line 5.2, we find initial consonant clusters which include the liquid *–r– in several lines. This *–r–, in fact, later develops into an l– initial. These Old Chinese consonant clusters have all essentially disappeared in modern dialects. Nonetheless, the fact is that liquid consonants, such as [l] and [r], frequently form initial clusters with obstruents in languages of the world, as in English play, clay, pry, cry, frill, shrill, slay, etc.

Phonograms and rhyme dictionaries The primary line of evidence for the existence of consonant clusters in Old Chinese comes from phonograms in the written language. Phonograms are constructed from two parts, typically with a radical indicating its meaning on one side and a phonetic on the other side. It turns out many sets of phonograms can be found in modern dialects which share a phonetic, but which differ in their initial consonant; furthermore, one of these initial consonants is an obstruent while the other is the liquid [l]. Thus the phonograms 剥 and 禄 share a phonetic, but the former is pronounced bo while the latter is pronounced lu. Similarly, 果 is pronounced guo while 裸 is luo; and 使 is pronounced shi while 吏 is li. As Old Chinese changed into Middle Chinese, its clusters reduced to single consonants in different ways in different phonograms, some keeping the obstruent and losing the liquid, while others kept the liquid and lost the obstruent. As can be seen from Table 10.2, starting with the Zhou, several major dynasties are divided into two periods according to when their capitals were moved, usually forced by military invasions. The two philosophers we mentioned earlier, Kongzi and Xunzi, were both active during Eastern Zhou. Also, note that while the Zhou and the Han moved their capitals eastward to avoid invaders from the west, the Song moved its capital southward due to military pressures from the north. Census figures from historical studies show it was shortly after the Song dynasty moved its capital to Lin’an 临安 (modern Hangzhou) that the population in South China came up approximately to the same level as that in the north. Previously, the bulk of the population centered around the Central Plains. It was during the Han dynasty that the first empirical studies in Chinese linguistics began; particularly the pioneering works of Yang Xiong 扬雄 on linguistic geography, and of Xu Shen 许慎 on lexicography. Yang compiled a precious list of words in a volume commonly called the Fangyan 方言, as these were pronounced in different regions of the Han empire. Xu produced the first comprehensive dictionary of Chinese sinograms, called the Shuowen Jiezi 说文解字; in the post-face to this dictionary, he proposed a scheme of classifying sinograms into six types, a scheme that continues to be used in traditional scholarship today. 166

Chinese linguistics

It is also a legacy of this great dynasty that the dominant ethnic group call themselves Hanzu 汉族, and call their language Hanyu 汉语. Another name also used by Chinese to refer to themselves comes from another great dynasty, the Tang dynasty. Thus the name Tangren 唐人, and many Chinatowns in Western countries are called Tangren Jie 唐人街. Yet another term of similar meaning is Hua 华, whose etymology is more obscure. In Singapore, for instance, the Chinese refer to themselves as Huaren 华人 and the language as Huayu 华语. While terms like Han, Tang, and Hua suggest influence from the dominant ethnic group, we should keep in mind that these ethnic labels did not acquire their current meaning until the Common Era. Previously, there were numerous ethnic groups, in the forms of minor kingdoms and nomadic tribes, which do not connect in simple ways with modern populations because of the extensive mixings over the many centuries. Furthermore, even in the simplified chronology of Table 10.2, there are many polities which were not Han and which did not speak Hanyu. These include periods when China was highly divided, such as during the Wudai Shiguo and Nanbei Chao, and when the rulers were obviously not Han, such as the Xixia, who were Tibeto-Burman, and Liao, Jin, Yuan, and Qing, when the rules were Altaic. So the peoples of China, their cultures and their languages, are truly heterogeneous, once we get past the superficial similarities.30 During the latter part of the Han dynasty, Buddhism began to enter China. With the immense task of translating its sutras into Chinese, China became aware of the Sanskrit language, its script, and its associated linguistic scholarship. Largely as a result of this strong influence, two major advances were soon made toward understanding how Chinese was pronounced. One was the explicit awareness of the nature of lexical tones. Four tones were recognized; they were named: Ping 平 ‘level’, Shang 上 ‘rising’, Qu 去 ‘departing’, and Ru 入 ‘entering’; we have represented these respectively as I, II, III, and IV. The other advancement was in the representation of pronunciations by a method called fanqie 反切 ‘reverse cut’. Before this method was introduced, the only way to help a reader with how a sinogram X is pronounced was by means of another sinogram Y, with the formula ‘Read X as Y’. However, if the reader does not know the pronunciation of Y either, then he receives no help. The fanqie method, on the other hand, generalizes the formula to ‘Read X as A+B’, with the understanding that X has the same initial consonant as A, and that it has the same tone and final as B. These two advancements paved the way to the structure of rhyme dictionaries and rhyme charts, and led to deeper knowledge in phonetics. The Sui dynasty, though short lived, is important because it was then that a rhyme dictionary of critical importance was compiled. This dictionary, the Qie Yun 切韵 ‘cut rhyme’, and its descendent dictionaries form the basis for projecting backward in time to reconstruct Old Chinese, and for projecting forward in time for connections to the modern dialects. The language inferred from this dictionary is aptly called Middle Chinese. This dictionary must have had an important influence on the greatest poets China ever produced, who composed their brilliant literature in rhymes during the great Tang dynasty which directly followed the Sui. By Qie Yun’s time, it was clear that each sinogram corresponds to a single syllable, which has a three-part phonetic structure: tone, initial, and final, respectively called shengdiao 声调 ‘tone’, shengmu 声母 ‘initial’, and yunmu 韵母 ‘final’. The rhyme dictionary was organized by first grouping the sinograms by the four tones, Ping, Shang, Qu, and Ru. Then it was next organized by groups of finals. Lastly, it was organized by grouping the sinograms by their initials. The sinograms which share all three parts of phonetic structure are homophones. The first sinogram, or the main sinogram, of each group of homophones is distinguished two ways: a little circle is placed above it to mark the beginning of the group, and a fanqie is provided at its end to notate the pronunciation. 167

William S.-Y. Wang

Figure 10.4  A page from a rhyme book of the Song dynasty, adapted from Wikipedia: http://en. wikipedia.org/wiki/Rime_dictionary#/media/File:Guangyun_Dong_Rhyme_1.jpg, under CC-BY-SA 3,0 (free to share in any format for any purpose, even commercially)

The Qie Yun itself, unfortunately, is no longer available. But many later versions based on the same organizing principles are available. Current knowledge about ancient pronunciations are largely based on these later versions. Figure 10.4 is a page taken from the Guang Yun 广韵, compiled in the Song dynasty as a successor to the Qie Yun. In the fourth column, counting from the right, we see the sinogram 同 tong, with a little circle above it to mark the beginning of a group of homophonous sinograms. 同 is then followed by various commentaries on this sinogram, written in vertical lines which are half-column wide. These commentaries end in the next column to the left in a string of sinograms, 徒红切四十五, before the next homophonous sinogram begins, which is 仝 tong. The third sinogram in the string, 切 qie, tells us that its two preceding sinograms are the fanqie of the main sinogram. It reminds us to use the formula, read X as A+B. In this case, the A sinogram is 徒 tu, with the initial t–, and the B sinogram is 红 hong, with the final and tone –óng. Putting A and B together, we get the pronunciation of tóng, which is correct for 同, even though the fanqie was proposed in the rhyme book Qie Yun well over a millennium ago. Another important rhyme dictionary was compiled during the Yuan dynasty, the Zhongyuan Yinyun 中原音韵, which reflects the pronunciation of yet another period of the language, usually called Early Mandarin. The English translation ‘Mandarin’ is based on the Chinese term Guanhua 官话 ‘Official Speech’. However, recent research by Shen Zhongwei31 shows 168

Chinese linguistics

that Mandarin with its defining features actually emerged during the Liao dynasty, several centuries earlier than the Zhongyuan Yinyun. This discovery was made by careful examination of phonetic information contained in texts written in Khitan 契丹 and Jurchen 女真 scripts.

Many peoples, many voices We began this chapter by setting the stage of what is now China, though the political boundaries of the nation have changed many times since the earliest dynasty, the Shang, which is attested both archeologically and textually. The home of a nation is marked by its mountains and rivers. We have discussed the actors on this stage, the Chinese peoples, who are heterogeneous genetically, culturally, and linguistically. The linguistic roots of China may be traced to the dual heritage of the Shang dynasty – Jiaguwen and the Shijing, expressed in the language reconstructed as Old Chinese. The early emergence of a complete writing system, in the form of sinograms, well over 3,000 years ago, is of inestimable importance in binding the various peoples together into one over-arching literate culture over great expanses of space and time. Another landmark in the evolution of the language is the rhyme dictionary of the Sui dynasty, reconstructed as Middle Chinese. The language reflected in the Qie Yun served as a critical midpoint for the backward reconstruction of Old Chinese on one hand, and for the projection to modern dialects on the other. By the time of the Zhongyuan Yinyun, another rhyme dictionary compiled in the Yuan dynasty, Mandarin has already taken root in Beijing, which has served as the nation’s capital for most of the past millennium. More than any other dialect, many features of Mandarin have spread far and wide across the land, from the northeast to the southwest. Similarly, the other major dialects have also been shaped from continuous southward migrations from the Central Plains in the north, two major waves taking place when the Han dynasty collapsed, and when the Song dynasty moved its capital to modern Hangzhou. The migrants mixed their Hanyu with the languages of the earlier settlers in China’s southeast, which were mostly Austric. The dialects which resulted have evolved sufficiently apart from each other over many centuries that they qualify as separate languages. Nonetheless, the term ‘dialect’ is the preferred usage because of a highly interlocked mosaic of cultures and because of the shared writing system. Figure 10.5 gives an approximate geography of the major dialects of Hanyu. Over the past half century, the government in Beijing has advocated the use of the variety of Mandarin that is called Putonghua 普通话 ‘common speech’, which is primarily based on the pronunciation of the educated in Beijing and which draws its vocabulary and phrases from other regions as well. Besides Mandarin, which is also simply called Beifanghua 北方话 ‘northern speech’, there are six major dialects which are well recognized, three in Central China: Xiang 湘, Gan 赣, and Wu 吴, and three along the southern coast: Min 闽, Kejia 客家, and Yue 粤. Kejia ‘guest family’ is labeled Hakka on the map, which is a southern pronunciation; whereas the other five major dialects are named with province names, the Kejia came south later and were thus referred to as ‘guest families’. The map also shows Jin in the north, Hui in Central China, and Ping in the south; these have not yet been investigated with as much depth as the six major ones.

Sinograms and alphabets We will now return for more discussion on the writing system. Alphabetic spelling,32 of which Hanyu Pinyin is the latest version, came to China only in the sixteenth century. The indigenous script that has long been in use and continues to be used in China consisting of thousands of graphic units is popularly called ‘Chinese characters’, which we here call 169

William S.-Y. Wang

Hohhot Beijing

JIN

丨

Ta iy u an

Lanzhou

.Xi'an

Naniinq Shanghai

MANDARIN W uhan

Chenadu

Hangzhou

HUI

Chongqing

XIANG

WU

Nanchang

GAN

Ihangsha

Fuzhou

HAKKA

K u nm in g

YUE

Nanning

PING

MIN

Taipei

Guangzhou Ho ng Kong

Figure 10.5  Major dialects of Hanyu, adapted from Wikipedia: http://en.wikipedia.org/wiki/List_of_ varieties_of_Chinese#/media/File:Map_of_sinitic_dialect_-_English_version.svg, under CC-BY 3,0 (free to share in any format for any purpose, even commercially)

‘sinograms’. In China, these units are called Hanzi 汉字, where the ‘Han’ again refers to the largest ethnic group and ‘zi’ refers to a unit of writing. The earliest specimens of sinograms currently available date back only some 3,000+ years. However, given the maturity of these specimens, as we saw in the inscriptions in Figure 10.3, it is almost certain that the earliest forms of this script date much further back; presumably such ancient specimens are buried deep somewhere, waiting to be discovered in some future excavation. The first sinograms invented were in all likelihood iconic in nature, in that their shapes are suggestive of some physical properties of the object represented – such as the head of a sheep 羊 yang, or the branching of a tree 木 mu, or the two sides of a door 門 men, now simplified to 门. A particularly interesting iconic sinogram was 且 qie, which is based on a phallic symbol, showing early awareness of the important role males play in reproduction, 170

Chinese linguistics

and of the father–child bond in addition to the mother–child bond.33 The emergence of this sinogram may be related to the early transition from matrilineal to patrilineal society in Ancient China. Before long, these simple sinograms were combined into more complex ones, such as putting two or three trees together 林 lin, 森 sen to mean ‘forest’, or combining 日 ri ‘sun’ and 月 yue ‘moon’ to mean ‘brightness’ 明 ming.34 且 qie became combined with another sinogram, 示 shi, which has to do with matters of the spirit; the complex sinogram which results is 祖 zu, meaning ‘ancestor’, a core concept in Chinese civilization. The 礻on the left side of 祖 is a compressed version of 示; such compression commonly occurs when simple sinograms are combined to form complex ones. However, by far the most productive principle in constructing sinograms is to combine a component having to do with sense with another component having to do with sound. This can be exemplified by 材 cai, where the left component is the ‘tree’ shown above, or more generally, ‘wood’; here it gives the approximate sense that the sinogram refers to the nature of the material some object is made from. The right component 才 is another simple sinogram, pronounced cai. Thus the complex sinogram is pronounced cai. Such sinograms are called ‘phonograms’ in English, or xingsheng zi 形声字 in Chinese, literally, ‘shape–sound writing’. We have already seen several phonograms in our earlier discussion of Old Chinese. As Chinese civilization progressed over the millennia, the number of sinograms accumulated, so that large dictionaries list over 50,000 of them, including some whose shapes are extremely complex in terms of the number of strokes they contain. No one of course knows all of these sinograms, let alone uses them on a daily basis. For ordinary usage, such as reading newspapers, some 3,000 or 4,000 would suffice. There have been several occasions when government stepped in to systematize the sinograms, to standardize usage, and to redesign some of them. An early occasion was during the time of the First Emperor, when the Qin conquered the other kingdoms and unified China into a single empire. The most recent occasion of standardization for writing Chinese was in the 1960s, when Hanyu Pinyin was officially adopted, and the first batch of simplified sinograms were introduced. This resulted in two systems of sinograms: the traditional sinograms called either zhengti zi 正体字 or fanti zi 繁体字, and the simplified sinograms called jianti zi 简体字. Mainland China and Singapore now use jianti zi; on the other hand, Hong Kong, Macao, and Taiwan continue to use zhengti zi. The sinograms in this chapter are typically in jianti zi. The sinograms used in Japan, called kanji, which is the Japanese pronunciation of the Chinese word Hanzi, underwent some changes after they were adopted, some of them well over a thousand years ago. Many of the kanji have the same shape as both the zhengti and the jianti, as illustrated in line 1 of Table 10.4. Or, all three sinograms may be different, as illustrated in line 2. In line 3, the kanji is the same as the zhengti, while in line 4, the kanji is the same as the jianti. Table 10.4  Comparison of kanji, zhengti and jianti

1. 2. 3. 4.

Kanji

Zhengti

Jianti

母 読 車 体

母 讀 車 體

母 读 车 体

171

meaning meaning meaning meaning

‘mother’ ‘to read’ ‘vehicle’ ‘body’

William S.-Y. Wang Table 10.5  Comparison of traditional and simplified sinograms Traditional and simplified sinograms Trad.

Simp.

Trad.

Simp.

Trad.

Simp.

十 土 水 我 的 家 金 笑

十 土 水 我 的 家 金 笑

滅 龍 夢 廣 護 農 葉 塵

灭 龙 梦 广 护 农 叶 尘

詳 讀 銀 鐵 紅 線 財 帳

详 读 银 铁 红 线 财 帐

All four possibilities occur. Nonetheless, Chinese and Japanese readers can usually read texts in any of the three systems of sinograms, with varying degrees of difficulty. Table 10.5 illustrates the differences between the zhengti zi and the jianti zi in terms of three categories. As can be seen by comparing the leftmost pair of columns, many of the high-frequency sinograms have not been affected – they are the same in both systems. In the second category in the middle pair of columns, however, the sinograms have been simplified quite a bit. The top sinogram 滅, for example, has 13 strokes while its simplied form 灭 has only five strokes. The reduction in the number of strokes in these sinograms is the most significant. For the third category, illustrated in the last pair of columns in Table 10.5, the number of sinograms affected is the greatest. This is because the simplification is applied to the radicals, and each simplified radical affects dozens of sinograms which contain that radical. For example, the radical 言 in the top sinogram 详 has seven strokes, while the simplified form of this radical in 详 has only two strokes. Thus this saving of five strokes applies to all sinograms which contain this radical. There are occasional exceptions to this generalization; examples are the 5th sinograms in the middle pair of columns. Here, the traditional form 護 ‘to protect’ with the 言 ‘speech’ radical has been simplified to 护, changing it into a phonogram with a new ‘hand’ radical 扌 derived from 手, and a phonetic 户, pronounced hu. Moreover, to make the discussion here more complete, note also that in the even numbered sinograms in this rightmost pair of columns, the right component of the sinograms have been additionally simplified as well. Sustained Western influence entered China around 1580 in the form of Christian missionaries; an early one who had a deep impact was the Italian Jesuit Matteo Ricci.35 Among Ricci’s contributions was his translation of Euclid’s Elements, which was published in 1607. Over the next centuries many others came, bringing with them pieces of Western knowledge in various disciplines, including astronomy, mathematics, and medicine. Ricci’s work on the Chinese language laid the foundation for a successor, the Flemish Jesuit Nicolas Trigault,36 to build upon and devise the first Romanization of Chinese, published in 1626. Such works provide valuable data for investigating the history of the language. But the system of Romanization for Mandarin that had the longest influence in Chinese studies was one that was devised by two Englishmen; started by Thomas Wade and completed by Herbert Giles at the end of the nineteenth century. The Wade-Giles system was by far the most used Romanization for Mandarin until 1958, when Hanyu Pinyin became officially 172

Chinese linguistics

adopted by the National People’s Congress in Beijing. One can often learn something about a Chinese from the way his surname is Romanized. For instance, the common surnames 张, 江, and 徐 would be spelled Zhāng, Jiāng, and Xú according to Hanyu Pinyin, but Chang1, Chiang1, and Hsu2 according to the Wade-Giles system. If a person’s surname is spelled according to Wade-Giles, it means either that the surname was Romanized before Hanyu Pinyin became official, or that he is from Taiwan, where Hanyu Pinyin was not officially adopted until 2009. In connection with Romanization, mention should also be made of an ingenious system mostly developed by the linguist Y. R. Chao called Gwoyeu Romatzyh 国语罗马字 or Guóyǔ Luómǎ zì in Hanyu Pinyin. In this system, tone distinctions do not require superscript numerals as used by Wade-Giles or diacritic marks as used by Hanyu Pinyin. Instead, the distinctions are spelled into the letters for vowels and consonants. For instance, the syllable guo would be written guo, gwo, guoo, and guoh, four different spellings for the four tones. Table 10.6 gives a quick comparison to illustrate the three systems we have just mentioned. As we saw earlier, Pinyin uses diacritics to mark tones while Wade-Giles uses superscript numerals. For unvoiced, unaspirated stops, Pinyin uses ‘b’ to distinguish from unvoiced aspirated ‘p’. For this distinction, Wade-Giles uses an apostrophe instead, which is closer to the notation of the International Phonetic Alphabet. Gwoyeu Romatzyh is unique in distinguishing the four tones of Putonghua by changing the vowels and consonants in the spelling. For instance, the vowel is doubled in rows 1, 2, 5, and 6 to indicate the longer syllables associated with the PTH tone 3. Similarly, it uses consonants like ‘q’ and ‘h’ to indicate the shorter syllables associated with the PTH tone 4. While Gwoyeu Romatzyh has some loyal users, who stress its greater effectiveness in learning and remembering tones, and Wade-Giles continues to be used in some contexts, Hanyu Pinyin has emerged as the one standard system; the advocacy of this system has been summarized in a document published in 2001 in the Journal of Chinese Linguistics jointly by Cheng Chin-Chuan 郑锦全, Ting Pang-hsin, William S.-Y. Wang 王士元, and Mei Tsulin. We will consider briefly some aspects of pronunciation that an English speaker should note about the Pinyin system. For a fuller discussion as well as a CD to illustrate the sounds, the reader may consult Lin (2007). While our focus is on Putonghua, many of these aspects pertain to other Chinese dialects as well. Following the traditional order, we will begin with tones, proceed to consonants, to vowels, and to syllable structure. Chinese is a tone language in the sense that the words are constructed not only with consonants and vowels, as in all the languages of the world, but also with tones, realized by distinctive pitch contours. Although we cannot be sure at present, it is quite likely that tones have been in the Chinese sound system since ancient times.37 In Putonghua these tones Table 10.6  Comparison of three Romanization systems

1. 2. 3. 4. 5. 6. 7.

Hanyu Pinyin

Wade–Giles

Gwoyeu Romatzyh

bang pang zai cai xi jin qiao

pang p’ang tsai ts’ai hsi chin ch’iao

bang pang tzai tsai shi jin chiau

173

barng parng tzair tsair shyi jyn chyau

baang paang tzae tsae shii jiin cheau

banq panq tzay tsay shih jinn chiaw

William S.-Y. Wang

are typically numbered as 1 = high, 2 = rising, 3 = low, and 4 = falling. Other dialects may differ in the number of tones they have and in how their tones are numbered. The tones in Putonghua may be illustrated with a set of words that differ only in tone, such as: 1 2 3 4

= = = =

mā, má, mǎ, mà,

妈 麻 马 骂

meaning meaning meaning meaning

‘mother’, or ‘hemp’, ‘horse’, ‘to scold’,

shī, shí, shǐ, shì,

诗 时 史 市

meaning meaning meaning meaning

‘poetry’ ‘time’ ‘history’ ‘city’

Oftentimes, the diacritic in tone 1 is left out, and a syllable not marked with tone diacritics is understood to carry tone 1.38 Since tone 1 occurs with the highest frequency among the four tones, not marking it gives the page a cleaner look. However, when tone diacritics are left out altogether, many ambiguities arise. An inconvenient ambiguity that arises in connection with Ancient China is between Zhōu 周, the name of a dynasty, and Zhòu 纣, the name of the last king of the Shang dynasty. Another ambiguity is between two names of provinces, Shānxi 山西 and Shǎnxi 陕西. The conventional solution here is to write them respectively as Shanxi and Shaanxi.39 There are cases of true homophony as well, such as the names of two ethnic groups in China, both pronounced Yí. They are written distinctly of course: 夷 versus 彝. In general the letters for consonants in Pinyin have similar values as they have in English. The letter ‘g’ before ‘e’, is always pronounced like in English ‘get’ and never like English ‘gem’. The English ‘r’ sound is always pronounced with the lips rounded, like in whistling. In contrast, the Chinese ‘r’ is pronounced with the lips rounded only before rounded vowels like ‘u’, like in ru 入 ‘enter’; otherwise the lips are not rounded, like in re 热 ‘hot’. The differences between Pinyin and English become greater in the case of the consonants arranged in Table 10.7 according to their phonetic values. After each Pinyin letter, I have also provided its closest English equivalent. The relevant letters in the English words are underlined. The consonants in the top two rows are called affricates; the aspirated ones are pronounced with a puff of air immediately following, while the unaspirated ones are not followed by such a puff of air. In the dental column, the sounds represented by Pinyin ‘z’ and ‘c’ do not occur as initial consonants in English. A good way to start practicing them is to put a word after them, such as ‘bids are’ and ‘bits are’, and then suppress the ‘bi’, much as one removes scaffolding after it has served its purpose. The consonants in the retroflex column are perhaps the most unfamiliar and challenging to speakers of other languages. Their Pinyin spelling is unusual in that a sequence of two letters is used to spell single sounds, like the use in English of ‘th’ for spelling the word ‘the’. They are called retroflex because they are pronounced with the tip of the tongue tilted slightly upward, as one can verify from x-ray photographs. We will see these consonants a lot in the pages to follow, since the name of both the Shang dynasty and the Zhou dynasty begin with retroflex consonants in Putonghua. Table 10.7  Pinyin and some PTH obstruents

Unaspirated Aspirated Fricative

Dental

Palatal

Retroflex

z = bids c = bits s = sew

j = gin q = chin x = shin

zh = drew ch = true sh = shrew

174

Chinese linguistics Table 10.8  The vowel spelled ‘i’ in Pinyin

Unaspirated Aspirated Fricative

Dental

Palatal

Retroflex

zi = 字 ‘character’ ci = 次 ‘second’ si = 四 ‘four’

ji = 记 ‘record’ qi = 气 ‘air’ xi = 细 ‘thin’

zhi = 志 ‘ambition’ chi = 翅 ‘wing’ shi = 市 ‘city’

The consonants clusters in the English words in Table 10.7 are similar in acoustic effect; nonetheless they are but an approximation to the Chinese retroflex consonants. However, it may be a comfort to some readers who are especially challenged here to know that many Chinese speakers coming from some dialect regions often speak Putonghua without these retroflexes. They are simply pronounced as dental consonants instead. It is useful to compare the Pinyin vowels with those in European languages where the spelling is more regular and simple, like Italian or Spanish. The situation with English is much more complicated because every vowel letter has several different pronunciations, depending on the word; for example the ‘i’ in ‘five’ and ‘fifth’; or the ‘e’ in ‘me’ and ‘met’. The only Pinyin vowel that the reader may find challenging is ‘i’. To discuss this vowel, in Table 10.8 we return to the consonants we have met, except now they are followed by the Pinyin vowel ‘i’. For the palatal column, there should be no problem: Pinyin ‘ji’ is like English ‘gee’, Pinyin ‘qi’ is like English ‘chee’, and Pinyin ‘xi’ is like English ‘she’, all pronounced with a falling tone. However, in the dental column, Pinyin ‘i’ is pronounced as an apical vowel; this is a sound most readers unfamiliar with Chinese would not have encountered. The sound has been described as a continuation of the preceding consonant but without the frication. The same description may be offered for pronouncing ‘i’ after the retroflex consonants, except there should be ‘r’-coloring added as well. Of course, no amount of paper discussion on pronunciation can substitute for the benefits of just listening to some examples by a friendly native speaker. After all, the eye is no substitute for the ear. The nine words in Table 10.8 are selected to illustrate the phonetics just discussed, specifically the apical vowels and retroflex consonants. As if to make up for the problems that the tones, the retroflexes, and the apical vowel may cause the learner of the language, the syllable structure in modern Chinese is elegantly simple. A syllable begins with at most one consonant, which may or may not be followed by a glide. A syllable may end with at most one consonant. In the Pinyin system, a glide which begins a syllable is spelled ‘w’ or ‘y’. However, when the glide is preceded by a consonant, it is spelled ‘u’ or ‘i’. A syllable without consonants at the end is called ‘open’; otherwise it is ‘closed’. All eight possible patterns are shown in Table 10.9, with illustration for each pattern, all in Table 10.9  Shapes of the Putonghua syllable Open syllable 1. 2. 3. 4.

V GV CV CGV

a wa ha hua

阿 蛙 哈 花

Closed syllable

‘prefix’ ‘frog’ ‘interjection’ ‘flower’

5. 6. 7. 8.

175

VC GVC CVC CGVC

an wan han huan

安 弯 憨 欢

‘peace’ ‘curve’ ‘honest’ ‘happy’

William S.-Y. Wang

tone 1. These eight patterns may be summarized by a simple formula, i.e. (C)(G)V(C), where C is consonant, G is glide, and V is vowel, using parentheses to indicate optional elements.40 Ambiguities may arise with certain sequences of consonants and vowels with respect to syllable boundaries. For such cases we use an apostrophe to indicate a syllable boundary. For instance, we write Xi’an for the disyllabic name of the famous city in the province of Shaanxi. Written without the apostrophe, xian is a single syllable 县 that may mean ‘county’. Xi’an means ‘western peace’. In olden times, as in the Tang dynasty, the city was called Chang’an 长安, meaning ‘eternal peace’. Note, however, without the apostrophe to mark the syllable boundary, the sequence could represent the two syllables chan and gan, each representing several sinograms. One final comment on the scope of the syllable in Chinese. As mentioned earlier, there are three dimensions in language that may be called the three S’s: namely, Sound, Script, and Sense. The syllable is a unit of Sound, but in Chinese it always corresponds to a unit in Script, which is the zi 字, or Chinese character. In turn, the Chinese character almost always corresponds to a unit in Sense, which is the morpheme.41 Thus in Chinese there is the simple equation relating sound to script to sense: One syllable = One sinogram = One morpheme Such an equation would not hold for an alphabetic language like English, where the unit in Script is the word, which may contain two or more syllables as well as two or more morphemes. In Chinese, words are not separated by spaces as in English. Here is an example to illustrate the difference. The Chinese sentence has four syllables, four characters, and four morphemes; the relations among the three dimensions are straightforward. In contrast, the corresponding English sentence with the same meaning has four syllables, the first syllable containing two morphemes, syllables 2 and 3 represent the third morpheme, and syllable 4 representing the fourth morpheme. The relations are therefore more complex. Chinese: Ni de fang ke.  你  的  访  客. English: You-r  visit-or.

Retrospect and prospect In addition to the topics discussed above, there are many others which merit inclusion in an introductory chapter such as this one, but that we must forego due to lack of space. The reader will find them treated in greater depth in various parts of this Encyclopedia, as well as in the recently published Oxford Handbook of Chinese Linguistics (Wang and Sun, ed. 2015). An excellent book-length overview of the field is Norman (1988). For deeper bibliographic coverage, there is Wang and Lyovin (1970) for earlier studies, and Wu (2009) for publications in the Journal of Chinese Linguistics since 1973. Toward presenting a more complete picture on Chinese Linguistics in this chapter, some references are suggested here. For a bird’s eye view of Chinese history, the series of atlases edited by Tan Qixiang 谭其骧 (1982) is a valuable reference. In relating language to folk culture to China, the Manzu 满族 scholar Luo Changpei (1989/1950) was an early pioneer; two recent books in this tradition are Zhou and You 周振鹤、游汝杰 (1986), and Deng Xiaohua 邓晓华 (1993). 176

Chinese linguistics

In the area of language contact, an important discussion was that of Hashimoto Mantaro 桥本万太郎, who published in Japanese in 1977 (Chinese translation 1985). A more recent investigation is the book by Chen Baoya 陈保亚 (1996), which offers a more fine-grained discussion of extended multilingualism in China’s Southwest, and the in-depth analysis of Hanyu and Bai by Wang Feng 汪锋 (2012). Of particular interest is the analysis of extensive mixing of Hanyu and Zangyu (Tibetan) in a community in Sichuan by the Zangzu 藏族 scholar Yeshes Vodgsal Atshogs 意西微萨•阿错 (2005), and of Hanyu and Zhuangyu in Guangdong by the Zhuangzu 壮族 scholar Li Jingzhong 李敬忠 (2006). Minority languages are finally getting the attention they have always deserved.42 In the area of historical linguistics, the textbook by Xu Tongqiang 徐通锵 (1991) has been influential for explaining the subject in Chinese in a wide perspective. The monograph by Pan Wuyun 潘悟云 (2000) presents a systematic account of Chinese phonology as it has developed from 3,000 years ago to the present. With modern phonetics, we are now in a much better position to examine with great precision and with large corpora of Hanyu as it is currently spoken, as illustrated by Peng (2006). With focus specifically on lexical tones, Cheng and Wang (1971) tabulate the categorical changes from Middle Chinese, while Chen (2000) explores the depth of how these tones sandhi from dialect to dialect. The basic units of any language are its words. Masini (1993) explores how the Chinese lexicon was formed, especially over the past two centuries. Cheng (1998) reports the very interesting finding that there is a constant upper bound in the number of sinograms used throughout the dynastic histories, even though the cumulative number inevitably increases over the two millennia. A recent discussion of the full Chinese lexicon is Yip (2000); a detailed analysis of how Chinese words are formed is Packard (2000), especially with respect to derivational morphology. Recently several volumes have been published on aspects of the lexicon which are useful both for research and for language teaching. These include an extensive compilation of synonyms by Zhang (2010), and collections of idioms (2011) and of proverbs (2014) by Jiao and colleagues. Although Chinese scholars have discussed grammatical categories within a traditional perspective for centuries, Sun (1996) offers a modern approach on how these categories may change. The volume by Xu (2006) discusses the major historical changes in Chinese syntax, especially their effects on typology. As for an overall synchronic grammar of the language, an early one in the indigenous framework is Lü Shuxiang 吕叔湘 (1957). The classic treatment, however, is that by Chao (1968), a distillation of many decades of acute observations on the language. It enjoys the singular distinction of having been translated into Chinese twice within a year of each other: in 1979 by Lü, then director of the Institute of Linguistics in Beijing, and in 1980 by Ting Pang-hsin, then director of the Institute of History of Philology43 in Taipei. A series of later grammars followed, each with its own orientation of applications and theories; these include: Li and Thompson (1981), Cheung et al. (1994), Tao (1996), Chen (1999), Chappell (2001), Ross and Ma (2006), Huang et al. (2009), etc. While there is no dearth of treatments on Mandarin, there is a clear lack of grammars of the other dialects, and especially of the minority languages. In recent decades, Chinese linguistics has broadened its horizons considerably. As Edward Sapir wrote cogently nearly a century ago when he discussed the status of linguistics as a science (1929: 214):44 ‘It is peculiarly important that linguists, who are often accused, and accused justly, of failure to look beyond pretty patterns of their subject matter, should become aware of what their science may mean for the interpretation of human conduct in general.’ The broadening is in part reflected in the title of the anthology compiled by Jing-Schmidt (2013): ‘Increased Empiricism’. Linguists have joined up with archeologists and geneticists 177

William S.-Y. Wang

to explore prehistory within a multidisciplinary perspective; see Jin et al. (2001). Increasing attention is being paid to the written language, and the associated issues of literacy, such as in the discussions by Wang et al. (2009) and in the anthology compiled by Feng and Branner (2011). Language is a behavior, ultimately the result of activities of the brain. It is particularly encouraging that scholars from psychology and neurosciences have brought in their expertise and are now collaborating with linguists to investigate the languages and dialects of China. Here is a sample of an expanding literature on both normal language as well as language disorders, including several very useful anthologies: Tzeng et al. (1977), Tzeng and Wang (1983), Chen and Tzeng (1992), Li et al. (2006), Law et al. (2008). A quarter century ago (1988), the linguist Jerry Norman produced a masterly synopsis of Chinese linguistics in a succinct 290 pages; see Coblin’s eulogy (2013). It is still the best single-volume introduction to the essentials of the field. However, so much has happened on many fronts over this quarter century that a comparable synopsis now would certainly require many more pages to do justice to the field. This can be seen in the massive Oxford Handbook of Chinese Linguistics published in 2015, edited by Wang and Sun, which brings together the expertise of 66 scholars in the field.

Notes   1 Chinese words are spelled in Hanyu Pinyin, and will be italicized when they first appear. They are usually written in simplified sinograms jianti zi 简体字; however traditional sinograms zhengti zi 正体字 are used when special needs arise.  2 Confucius, The Analects; translated by D. C. Lau (1992: 120), Chinese University of Hong Kong Press.   3 See: W. S.-Y. Wang (1989) ‘Language in China: a chapter in the history of linguistics’, Journal of Chinese Linguistics 17: 183–222; and Anthony Yu 余国藩 (2008) ‘Cratylus and Xunzi on Names’, Classical, Comparative, and Contemporary: Essays on Literature, Religion, and Other Topics, Columbia University Press.   4 The British named it Mount Everest, apparently not aware that the mountain already had a name which dates back several centuries. A Chinese name for the peak is Shengmufeng 圣母峰, which is an approximation to the Tibetan meaning.   5 Sections of the Changjiang also have other names: Jinshajiang 金沙江 for its initial section and Yangzijiang 扬子江 for its middle section. The spelling Yangtze is an older form of Yangzi.   6 Heather Pringle (2014) ‘Welcome to Beringia’, Science 343(6174): 961–3; John F. Hoffecker, Scott A. Elias, and Dennis H. O’Rourke (2014) ‘Out of Beringia?’ Science 343(6174): 979–80.   7 John D. Bengtson (1999) ‘Wider genetic affiliations of the Chinese language’, Journal of Chinese Linguistics 27: 1–12.   8 However, this view may be complicated by recent findings in molecular genetics; cf. s32 of The HUGO Pan-Asian SNP Consortium et al. (2009) ‘Mapping human genetic diversity in Asia’, Science 326(5959): 1541–5. P. Li (2015) is an up-to-date discussion of Austronesian linguistics in the light of the most recent archeological finds.  9 Dates for dynasties in this chapter follow those given in the Xinhua Zidian 新华字典, 11th ed. (2011). However, it is good to remember that the dynasties often overlap quite a bit, and many dates are still debated among specialists. 10 Ian Johnson (2013) ‘China’s ancient lifeline’, National Geographic, May. 11 Known in Chinese as 利玛窦 Lì Mǎdòu. 12 We thank Professor Paul J. K. Li of Academia Sinica for this map. 13 Elizabeth Pennisi (2013). ‘More genomes from Denisova Cave show mixing of early human groups’, Science 340(6134): 799. 14 Quote is from the US anthropologist F. Clark Howell in his introduction to the 1985 anthology edited by R. Wu and D. Olsen, p. xx. 15 Li, Xueqin, Garman Harbottle, Juzhong Zhang, and Changsui Wang (2003), ‘The earliest writing? Sign use in the seventh millennium BC at Jiahu, Henan Province, China’, Antiquity 77: 31–44.

178

Chinese linguistics 16 Population geneticist Cavalli-Sforza pioneered the multidisciplinary study of human prehistory, and gives a general account of current knowledge in this area in his 2000 book, Genes, Peoples, and Languages, London: Penguin Books. 吴一丰 等译 (2003) 追踪亞当夏娃。台北：远流出版社。 17 Weidenreich, Franz (1943) ‘The skull of Sinanthropus pekinensis: a comparative study on a primitive hominid skull’, Palaeontologia Sinica 10: 1–289. 18 Wu, Xinzhi and Frank E. Poirier (1995) Human Evolution in China, Oxford University Press. 19 Shang, Hong, Haowen Tong, Shuangquan Zhang, Fuyou Chen, and Erik Trinkaus (2007) ‘An early modern human from Tianyuan Cave, Zhoukoudian, China’, PNAS 104: 6,573–78. 20 Wu, Xiaohong, Chi Zhang, Paul Goldberg, David Cohen, Yan Pan, Trina Arpin, and Ofer Bar-Yosef (2012) ‘Early pottery 20,000 years ago in Xianrendong Cave, China’, Science 336: 1,696–70, June 29. 21 See Wu and Olsen 1985 for a fuller listing of these sites. 22 Another Swedish scholar whose work is of great importance was Bernhard Karlgren 高本汉 (1889– 1978), a younger colleague of Andersson. Karlgren introduced linguistics methods developed in Europe to China, and exemplified these methods with dialect studies as well as historical reconstruction. Andersson founded the famous Museum of Far Eastern Antiquities in Stockholm in 1926; Karlgren succeeded him as director of the Museum (1939–1959). 23 W. S.-Y. Wang 1998: Figure 1. 24 Green et al. 2010. 25 The names in Figure 10.2 are written in the Wade-Giles spelling, which was widely used in Western literature until it was replaced by Hanyu Pinyin. Archeology has progressed tremendously over the last several decades since Chang 1986. See Chang 1999. A good overview of the current state of knowledge is the elegant volume edited by Chang and Xu 2005. 26 This date is 1,000 years earlier than the traditional folk view in China that the civilization dates back 5,000 years. 27 This sample is discussed in greater detail in Wang (2013) Love and War in Ancient China. 28 Old Chinese is also called Archaic Chinese; Middle Chinese is also called Ancient Chinese. 29 Translation based on D. C. Lau 2002: 73. 30 See 王士元 (2013) 谁是中国人?《科学中国人》8: 38–43. 31 See Shen, Zhongwei (2011) ‘The origin of Mandarin’, Journal of Chinese Linguistics 39: 1–31. 32 For a comparative discussion of Chinese writing and alphabetic writing, see W. S.-Y. Wang and Tsai 2011. For a historical survey of literacy in China, see F. Wang et al. 2009. 33 Ancient writings often spoke of the offspring knowing the mother but not the father, 知母不知父, which is a natural outcome of primitive biological scenarios. 34 Text starting from this paragraph to the third paragraph below Table 10.5 (including Table 10.5) on p. 172 is originally published in Love and War in Ancient China: Voices from the Shijing, pp. xxix–xxxii, © 2013 City University of Hong Kong. Used by permission of City University of Hong Kong Press. All rights reserved. 35 A highly readable account of Matteo Ricci in China is: Spence, Jonathan D. (1984) The Memory Palace of Matteo Ricci, New York: Viking. 36 Known in Chinese as 金尼阁, Jīn Nígé. His Xiru Ermu Zi 西儒耳目资, contains the first Romanization system for writing Chinese. 37 For the historical background of Chinese tones, see W. S.-Y. Wang and Cheng 1987. It is very likely that Chinese had become a fully tonal language by the Han dynasty, though the tones were not explicitly discussed until several centuries later, presumably under the influence of Sanskrit writings. Text starting from this paragraph to the section end on p. 176 (including Tables 10.7, 10.8 and 10.9) was originally published in Love and War in Ancient China: Voices from the Shijing, pp. xxii–xxvii, © 2013 City University of Hong Kong. Used by permission of City University of Hong Kong Press. All rights reserved. 38 To be more precise, unstressed syllables without tones are also unmarked with diacritics; however they can usually be distinguished by context. 39 Shaanxi is a generally used artificial spelling where the ‘aa’ indicates the third tone, to distinguish it from the name of another province, Shanxi, which has the first tone. 40 In this formula, the V stands for vocalic nucleus, which includes simple vowels as well as diphthongs. 41 There are few exceptions here, such as 葡萄 ‘grape’, 玻璃 ‘glass’, 蟋蟀 ‘cricket’, etc. 42 This is a point I tried to highlight at the formation of the International Association of Chinese Linguistics; see W. S.-Y. Wang 1993.

179

William S.-Y. Wang 43 The Institute of History and Philology was one of the first research institutes established at the Academia Sinica, home of eminent linguists such as Y. R. Chao and F. K. Li. In 2004, some members formed a new Institute of Linguistics for further development of the field. 44 Sapir, Edward (1929) ‘The status of linguistics as a science’, Language 5: 207–14.

Selected references English Abi-Rached, Laurent, et al. (2011) ‘The Shaping of Modern Human Immune Systems by Multiregional Admixture with Archaic Humans’, Science 334: 89–94, 7 October. Baxter, W. H. (1992) A Handbook of Old Chinese Phonology, Berlin: Mouton de Gruyter. Chang, Kwang-chih (1986) The Archeology of Ancient China, 4th ed., New Haven: Yale University Press. Chao, Yuen-Ren (1968) A Grammar of Spoken Chinese, University of California Press, Chinese translations by Lü Shuxiang 吕叔湘 in 1979, and by Ting Pang-hsin 丁邦新 in 1980. Chappell, Hilary (ed) (2001) Sinitic Grammar: Synchronic and Diachronic Perspectives, Oxford: Oxford University Press. Chen, Hsuan-chih and Ovid J. L. Tzeng (eds) (1992) Language Processing in Chinese, Amsterdam: Elsevier. Chen, M. Y. (2000) Tone Sandhi: Patterns across Chinese Dialects, Cambridge: Cambridge University Press. Chen, Ping (1999) Modern Chinese: History and Sociolinguistics, Cambridge: Cambridge University Press. Cheng, Chin-Chuan (1973) ‘A Quantitative Study of Chinese Tones’, Journal of Chinese Linguistics 1: 93–110. Cheng, Chin-Chuan (1968) ‘English Stresses and Chinese Tones in Chinese Sentences’, Phonetica 18: 77–88. Cheng, Chin-Chuan (1994) ‘DOC: lts Birth and Life’, in M. Y. Chen and Ovid J. L. Tzeng (eds) In Honor of William S.-Y. Wang: Interdisciplinary Studies on Language and Language Change, Taipei: Pyramid Press. Cheng, Chin-Chuan (1998) ‘Quantification for Understanding Language Cognition: Quantitative and Computational Studies on the Chinese Language,’ in B. K. T’sou, T. B. Y. Lai, S. W. K. Chan, and W. S.-Y. Wang (eds) Quantitative and Computational Studies on the Chinese Language, Hong Kong: City University of Hong Kong Press, 15–30. Cheng, Chin-Chuan and W. S.-Y. Wang (1971) ‘Phonological Change of Middle Chinese Initials’, Tsing Hua Journal of Chinese Studies 9: 216–70. Cheung, Samuel Hung-nin, Liu Sze-yun, and Shih Li-Lin (1994) A Practical Chinese Grammar, Hong Kong: The Chinese University Press. Coblin, W. South (2013) ‘Jerry Norman: Remembering the Man and His Perspectives on Chinese Linguistic History’, Journal of Chinese Linguistics 41: 219–45. Feng, Li and David Prager Branner (eds) (2011) Writing and Literacy in Early China, Seattle: University of Washington Press. Ho, Dah’an (2015) ‘Such Errors Could be Avoided. Review of “Old Chinese, a New Reconstruction by W.H. Baxter and Laurent Sagart” ’. In press with Journal of Chinese Linguistics. Huang, C-T. James, Y.-H. Audrey Li, and Li Yafei (2009) The Synrax of Chinese, Cambridge: Cambridge University Press. Jiao, Liwei, Cornelius C. Kubler, and Weiguo Zhang (2011) 500 Common Chinese Idioms, London and New York: Routledge. Jiao, Liwei and Benjamin M. Stone (2014) 500 Common Chinese Proverbs and Colloquial Expressions, London and New York: Routledge. Jin, Li, Mark Seielstad, and Chunjie Xiao (eds) (2001) Genetic, Linguistic and Archaeological Perspectives on Human Diversity in Southeast Asia, Singapore: World Scientific. Jing-Schmidt, Zhuo (ed) (2013) Increased Empiricism: Recent Adventures in Chinese Linguistics, Amsterdam and Philadelphia: John Benjamins. Karlgren, Bernhard (1939) Etudes sur la phonologie chinoise, Chao Y. R. 赵元任, Li Fangkuei 李方桂, and Luo Changpei 罗常培 (trans.)《中国音韵学研究》(A Study of Chinese Phonology), Beijing: The Commercial Press.

180

Chinese linguistics Karlgren, Bernhard (1950) The Book of Odes, Stockolm: Museum of Far Eastern Antiquities. Keightley, David N. (1983) The Origins of Chinese Civilization, Berkeley: University of California Press. Law, Sam-Po, Brendan Stuart Weekes, and Anita Mei-Yin Wong (eds) (2008) Language Disorders in Speakers of Chinese, Multilingual Matters / Channel View Publications. Li, Charles N. and Sandra A. Thompson (1981) Mandarin Chinese: A Functional Reference Grammar, Berkeley: University of California Press. Li, Paul Jenkuei (2015) ‘The Discovery of Liangdao Man and its Implications for the Pre-Austronesian Homeland’, Journal of Chinese Linguistics 43: 224–31. Li, Ping, Li Hai Tan, Elizabeth Bates, and Ovid J. L. Tzeng (eds) (2006) The Handbook of East Asian Psycholinguistics: Volume 1, Chinese, Cambridge: Cambridge University Press. Lin, Yen-Hwei (2007) The Sounds of Chinese, Cambridge: Cambridge University Press. Liu, Wu, Chang-Zhu Jin and Ying-Qi Zhang, Yan-Jun Cai, Song Xing, Xiu-Jie Wu, Hai Cheng, R. Lawrence Edwards, Wen-Shi Pan, Da-Gong Qin, Zhi-Sheng An, Erik Trinkaus, and Xin-Zhi Wu (2010) ‘Human Remains from Zhirendong, South China, and Modern Human Emergence in East Asia’, PNAS 107: 19,201–06. Masini, Federico (1993) The Formation of Modern Chinese Lexicon, Journal of Chinese Linguistics Monograph No. 6, Huang Heqing 黄河清 (trans.) (1997) Xiandai Hanyu Cihui de Xingcheng《现代 汉语词汇的形成》, Shanghai: Hanyu dacidian chubanshe 汉语大词典出版社. Mei, Tsulin (1977) ‘Tones and Tone Sandhi in 16th century Mandarin’, Journal of Chinese Linguistics 5: 237–60. Norman, Jerry (1988) Chinese, Cambridge: Cambridge University Press. Norman, Jerry and Tsulin Mei (1976) ‘The Austroasiatics in Ancient South China: Some Lexical Evidence’, Monumenta Serica 32: 274–301. Packard, Jerome L. (2000) The Morphology of Chinese: A Linguistic and Cognitive Approach, Cambridge: Cambridge University Press. Peng, Gang (2006) ‘Temporal and Tonal Aspects of Chinese Syllables: A Corpus-based Comparative Study of Mandarina and Cantonese’, Journal of Chinese Linguistics 34: 134–54. Pulleyblank, E. G. (1993) ‘Old Chinese Phonology: A Review Article’, Journal of Chinese Linguistics 21: 337–80. Ross, Claudia and Jing-Heng Sheng Ma (2006) Modern Mandarin Chinese Grammar, London and New York: Routledge. Shen, Guanjun, Xing Gao, Bin Gao, and Darryl E. Granger (2009) ‘Age of Zhoukoudian Homo erectus Determined with 26Al/10Be Burial Dating’, Nature 458: 198–200. Simmons, Richard VanNess (1999) Chinese Dialect Classification: A Comparative Approach to Harngjou, Old Jintarn, and Common Northern Wu, Amsterdam and Philadelphia: John Benjamins. Sun, Chaofen (1996) Word-order Change and Grammaticalization in the History of Chinese, Stanford: Stanford University Press. Takashima, Ken-ichi (2012) ‘Etymology and Paleography of the Yellow River’, Journal of Chinese Linguistics 40: 269–306. Tao, Hongyin (1996) Units in Mandarin Conversation: Prosody, Discourse and Grammar, Amsterdam and Philadelphia: John Benjamins. T’sou, B. K. Y. (1981) ‘A Sociolinguistic Analysis of the Logographic Writing System of Chinese’, Journal of Chinese Linguistics 9: 1–19. Tzeng, Ovid J. L., D. L. Hung, and W. S.-Y. Wang (1977) ‘Speech Recoding in Reading Chinese Characters’, Journal of Experimental Psychology: Human Learning and Memory 3: 621–30. Tzeng, Ovid J. L. and W. S.-Y. Wang (1983) ‘The First Two R’s’, American Scientist 71: 238–43. Wang, Feng, Yaching Tsai, and W. S.-Y. Wang (2009) ‘Chinese Literacy’, D. Olson and N. Torrance (eds) Cambridge Handbook on Literacy, Cambridge: Cambridge University Press, 386–417. Wang, W. S.-Y. (1973) ‘The Chinese Language’, Scientific American 228: 53–62. February issue. Follow-up discussion in May 1973, 228: 8–9. Chinese translation by You Rujie 游汝杰 1987, in Yuyan yu Renlei Jiaoji《语言与人类交际》, Guangxi Education Press 广西教育出版社, 56–71. Wang, W. S.-Y. (1993) ‘An Association of Our Own’, Newsletter of the International Association of Chinese Linguistics 1(1): 1–2. Wang, W. S.-Y. (ed) (1995) The Ancestry of the Chinese Language, Journal of Chinese Linguistics Monograph No.8. Chinese translation by Li Baojia 李葆嘉 (2005), Hanyu de Zhuxian《汉语的  祖先》, Beijing: Zhonghua shuju 中华书局.

181

William S.-Y. Wang Wang, W. S.-Y. (ed) (1991) Languages and Dialects of China, Journal of Chinese Linguistics, Monograph No. 4. Wang, W. S.-Y. (1998) ‘Three Windows on the Past’, in Victor H. Mair (ed) The Bronze Age and Early Iron Age Peoples of Eastern Central Asia, Pennsylvania: University of Pennsylvania Museum Publications, 508–34. Wang, W. S.-Y. (2013) Love and War in Ancient China: The Voices of Shijing, Hong Kong: City University of Hong Kong Press. Wang, W. S.-Y (forthcoming) A Billion Voices: Languages and Peoples of China, in press with World Scientific Press. Wang, W. S.-Y. and A. Lyovin (1970) CLIBOC: Chinese Linguistics Bibliography on Computer, Cambridge: Cambridge University Press Wang, W. S.-Y. and Cheng Chin-Chuan (1987) ‘Middle Chinese Tones in Modern Dialects’, in Robert Channon and Linda Shockey (eds) In Honor of Ilse Lehiste, Dordrecht: Walter de Gruyter, 513–23. Wang, W. S.-Y. and Yaching Tsai (2011) ‘The Alphabet and the Sinogram: Setting the Stage for a Look across Orthographies’, in P. McCardle, J. R. Lee, B. Miller and Ovid J. L. Tzeng (eds) Dyslexia Across Languages: Orthography and the Brain–Gene–Behavior Link, Brookes Publishing, 1–16. Wang, W. S.-Y. and Chaofen Sun (eds) (2015) Oxford Handbook of Chinese Linguistics, Oxford: Oxford University Press. Wu, Rukang and John W. Olsen (eds) (1985) Palaeoanthropology and Palaeolithic Archaeology in the People’s Republic of China, Orlando: Academic Press. Wu, Yifeng, (ed) (2009) Cumulative Indexes and Abstracts to Journal of Chinese Linguistics, Journal of Chinese Linguistics, Monograph No. 23. Xing, Janet Zhiqun (ed) (2009) Studies of Chinese Linguistics: Functional Approaches, Hong Kong: Hong Kong University Press. Xu, Dan (2006) Typological Change in Chinese Syntax, Oxford: Oxford University Press. Yan, Margaret Mian (2007) Introduction to Chinese Dialectology, LINCOM Studies in Asian Linguistics. Yang, Xianyi (1983) Selection from the Book of Songs. Beijing: Panda Books. Yip, Po-Ching (2000) The Chinese Lexicon: A Comprehensive Survey, London and New York: Routledge. Zhao, Tongmao and Lee Tsungdao (1989) ‘Gm and Km Allotypes in 74 Chinese Populations: A Hypothesis of the Origin of the Chinese Nation’, Human Genetics 83: 101–10. Zhang, Grace Qiao (2010) Using Chinese Synonyms, Cambridge: Cambridge University Press. Zhang, Hongming (1998) ‘Chinese Etyma for “River” ’, Journal of Chinese Linguistics 26: 1–47. Zhengzhang, Shangfang (1991) ‘Decipherment of Yue-Ren-Ge’, Cahiers Linguistiques d’Asie Oriental 20(2): 159–68. Zhou, Minglang (2003) Multilingualism in China: The Politics of Writing Reforms for Minority Languages, 1949–2002, Berlin: Mouton de Gruyter.

Chinese Atshogs, Yeshes Vodgsal 意西微萨•阿错 (2005)〈语言深度接触机制与藏汉语言类型差异问题〉(A Deep Contact Mechanism for Language and the Issue of Typological Differences of Tibetan Languages), Journal of Chinese Linguistics 33: 1–33. Chen, Baoya 陈保亚 (1996)《论语言接触与语言联盟》(On Language Contact and Language Alliance), Beijing: Language and Culture Press 语文出版社. Cheng, Chin-Chuan 郑锦全, Ting Pang-hsin 丁邦新, William S.-Y. Wang 王士元, and Mei Tsulin 梅祖麟 (2001)〈我们对中文音译的看法 – 请以汉语拼音为中文音译的唯一标准〉(Our Views on the Transliteration of Chinese – Please Take Hanyu Pinyin as the Only Standard for Chinese Transliteration), Journal of Chinese Linguistics 29: 165–7. Chou, Fa-Kao 周法高 (1980)《中国语言学》(Chinese Linguistics), Hong Kong: The Chinese University Press 中文大学出版社. Deng, Xiaohua 邓晓华 (1993)《人类文化语言学》(Human Cultural Linguistics), Xiamen: Xiamen University Press 厦门大学出版社. Hashimoto, Mantaro 桥本万太郎 (1977/1985)《语言地理类型学》(Linguistic Geographical Typology), Yu Zhihong 余志鸿 (trans.), Beijing: Peking University Press 北京大学出版社. Li, Jingzhong 李敬忠 (1994/2007)《语言演变论》(Language Change), Guangzhou: Guangzhou Press 广州出版社.

182

Chinese linguistics Lü, Shuxiang 吕叔湘 (1957)《中国文法要略》(Concise Chinese Grammar), Beijing: The Commercial Press 商务印书馆. Luo, Changpai 罗常培 (1950/1989)《语言与文化》(Language and Culture), Beijing: Language and Culture Press 语文出版社. Pan, Wuyun 潘悟云 (2000)《汉语历史音韵学》(Historical Chinese Phonology), Shanghai: Shanghai Education Press 上海教育出版社. Tan, Qixiang 谭其骧 (1982)《中国历史地图集》(Maps of Chinese History), Beijing: SinoMaps Press 地图出版社. Wang, Feng 汪锋 (2012)《语言接触与语言比较 – 以白语为例》(Language Contact and Language Comparisons – The Bai Language as an Example), Beijing: The Commercial Press 商务印书馆. Xu, Tongqiang 徐通锵 (1991)《历史语言学》(Historical Linguistics), Beijing: The Commercial Press 商务印书馆.   and Zhou, Zhenhe 周振鹤 and You Rujie 游汝杰 (1986)《方言与中国文化(修订本)》(Dialects Chinese Culture (Rev.)), Shanghai: Shanghai People’s Publishing House 上海人民出版社.

183

11 Chinese Linguistics: Pragmatics Yan Jiang school of oriental and african studies, University of london

1.  Short definition Pragmatics can be succinctly defined as the study of meaning in context. Although it does not have a long history as an independent academic subject of inquiry, the field has always been vibrant with diverse approaches, cross-disciplinary interactions, and several ingenious and highly sophisticated theoretical frameworks. In Chinese linguistics and applied language studies, pragmatics has also aroused great interest, all the more so because the term ‘pragmatics’ is translated into Chinese as yǔyòngxué 语用学, meaning ‘the study of language use’, which can often be shortened into yǔyòng 语用 (language use). Thus, pragmatics in Chinese has acquired a broader interpretation, encompassing all aspects of language use, making it more liable to be recognized as a useful subject of study by Chinese speakers, whether or not they have been exposed to the more rigorously defined pragmatic studies.

2.  Pragmatic phenomena Although there are several strands of pragmatics, due to page limit, this chapter is only focused on linguistic pragmatics, which claims the longest tradition and has always been engaged with central aspects of pragmatic inquiry. Linguistic pragmatics evolved from the school of Ordinary Language Philosophy as represented by the works of Anglo-analytic philosophers J. L. Austin (who is the founder of Speech Act Theory), P. F. Strawson (who initiated studies on presupposition), and H. P. Grice (founder of the theory of Logic in Conversation). It became a self-contained discipline at the beginning of the 1980s, heralded by the publication of two territory-defining textbooks: Levinson (1983) and Leech (1983). It differs from linguistic semantics in that it goes beyond the analyses of word meaning and sentence meaning, treated respectively by lexical and compositional semantics, and studies how sentence meaning is developed and even twisted in discourse. Its focal object of study is utterance meaning, which is roughly equivalent to propositional meaning read off a sentence in a given use context. The initial meaning of a sentence, composing out of the meaning of the words contained therein, when uttered in a context, may need to be explicated, through disambiguation and enrichment, to reach its full-fledged literal meaning, also called explicature. In addition, it may trigger the perception of some hidden and separate proposition taken for granted by 184

Chinese linguistics: Pragmatics

the speaker, formally called presupposition. Furthermore, it may implicate some extra, inferred propositions that the speaker tries to convey through the use of the original sentence – what is termed implicature. On top of all these layers of pragmatic meaning, a sentence always conveys a higher-order sense: what the speaker wants to do with the proposition or the set of propositions he produces, be it for making a statement, performing a speech act such as declaring, requesting, naming or marrying, making commentaries such as in a live football show, giving directions such as in an iPad manual or in a recipe, expressing irony in that what is stated is to be taken as contrary to what is intended to convey.  .  .  .  These additional tones variously attached to the propositions expressed form kinds of propositional attitude or higher-order explicature. Last but not least, some words and expressions can only be interpreted with reference to the physical properties in the situation of utterance, thereby revealing the very essential referential nature of natural language: that of using language to point to things right on the scene or in the vicinity, with demonstrative or symbolic gestures – what is called the deictic or indexical use of language. The notions introduced so far form a cluster of pragmatic phenomena that have received extensive studies in the literature. Each of them is illustrated below with two examples in Chinese. [1] méi xiǎng dào nǐ hái tǐng néng hē, yě tǐng néng liáo 没想到你还挺能喝,也挺能聊。 ‘I was surprised to find that you can drink a lot, and can talk a lot too’ [Explicature: Explicating the sentence so as to supplement what is left unsaid: what you drink is alcoholic, not just any soft drinks or water.] [2] shìzhě sòng shàng jiān hǎo de niúpái, wǒ fēn fù guò tā, jiān dé lǎo diǎn, qiē kāi shí,   lǐmiàn háishì hóng hóng de xuèsī 侍者送上煎好的牛排,我吩咐过他,煎得老点,切开时,里面还是红红的血丝。 ‘The waiter served our steaks. Although I had told him to cook the steaks well-done, when they were carved up, there were still threads of red blood’ [Explicature: The word lǎo (old) is enriched to give the special interpretation ‘well-done’ in this context.] [3] hěnshǎo yǒu rén zhīdào, fāmíng zìxíngchē de shì déguó de yī gè kànlínrén,   míng jiào déláisī 很少有人知道,发明自行车的是德国的一个看林人,名叫德莱斯。 ‘Not many people know that the man who invented the bicycle was a German forester called Drais’ [Presupposition: The de-construction fāmíng zìxíngchē de (发明自行车的) serves as a presupposition trigger showing that the speaker takes it for granted that the bicycle has been invented. This information is presupposed rather than asserted because nowhere in the sentence is the information ‘the bicycle has been invented’ explicitly stated. What is explicitly stated is only the identity of the inventor of the bicycle.] [4] zhèxiē kòngdì, yàoshì yònglái zhòng zhuāngjià, guójiā jiù bú yòng huā nàme duō   wàihuì jìnkǒu liángshi le 这些空地,要是用来种庄稼,国家就不用花那么多外汇进口粮食了。 ‘If these abandoned lands are used to plant crops, our country will no longer have to spend so much foreign currency to import grains’ 185

Yan Jiang

[Presupposition: ‘bú yòng huā nàme duō wàihuì’ presupposes the fact that a lot has already been spent or has at least been committed to spend on purchasing something.] [5] ‘wǒ xiǎng nǐ yěxǔ bú shì qù sànbù de,’ tā túláo wúyì de pángqiāocèjī shuō ‘我想你也许不是去散步的’, 他徒劳无益地旁敲侧击说。 ‘ “I thought maybe you didn’t [go for a walk],” he said, beating about the bush in the most useless manner.’ (from Sister Carrie, Chapter XXIII, by Theodore Dreiser.) [Implicature: the speaker hinted at some other propositions, e.g. that he suspected that she was seeing someone else. Neither in form nor in meaning does the intended implicature bear any similarities with the original sentence.] [6] gāngcái chīfàn shí jiàn nǐ méi chī shénme dōngxi, xiǎngbì shì wèikǒu bu hǎo 刚才吃饭时见你没吃什么东西,想必是胃口不好。 ‘At the dinner just now, I saw that you didn’t eat much. Thought you must have had a rather bad appetite’ [Implicature: Although ‘méi chī shénme dōngxi’ explicitly asserts that not much was eaten, it implicates that something was eaten, even though the quantity was minimal. This is because, according to the theory of scalar implicature, ‘shénme dōu méi chī’ (什么没都吃) and ‘méi chī shénme’ (没吃什么) can be said to form an ordered set: The item to the left is semantically stronger than the one to the right, conveying information that is higher in degree, larger in number, or more informative than the weaker one. The mentioning of the item to the right implicates the unavailability of the item to the left, as the speaker wants to be informative to the right extent.] [7] zī rènmìng X xiānsheng wéi jiǔdiàn shìchǎng xiāoshòubù zǒngjiān 兹任命X先生为酒店市场销售部总监。 ‘Mr. X is hereby appointed as the director of the Marketing Sales Department of the hotel’ [Propositional attitude: a speech act of appointing.] [8] fēi xiàng guòhé bù shǒu lǐ, shuāngjiǎo wù tà duìmiàn wèi 飞象过河不守礼,双脚勿踏对面位。 ‘(Hong Kong Bus notice) Literally: It is against the rule (in Chinese chess) to march your elephant across the boundary river into the opponent’s half. Paraphrase: Do not put your feet on the opposite seat.’ [Propositional attitude: an indirect speech act, making a request in the guise of a statement.] [9] jīntiān wǒmen zhè’ér xiàyǔ le, bù zhīdào háiyǒu nǎxiē dìfang xiàyǔ le 今天我们这儿下雨了,不知道还有哪些地方下雨了。 ‘Today it rained here at our place. Don’t know which other places also rain’ [Deictic use of jīntiān (today), wǒmen (we) and zhè’ér (here), whose exact referents can only be worked out with reference to the parameters of the utterance act, such as information about the speaker, the place and the time of the utterance, etc.] 186

Chinese linguistics: Pragmatics

One common feature shared by the terms introduced above is that such kinds of pragmatic meaning are inferred rather than completely encoded. That is, pragmatic meaning is never directly gleaned from the encoded meaning of linguistic expressions alone, and can only be obtained by making contextualized inferences, using the linguistic expressions as the starting point of total meaning construction. Hence what matters is not just the identification and analysis of pragmatic phenomena per se, but also the characterization of the general inferential mechanism involved. This calls for the need of an explanatory theory of pragmatics that can give an adequate account of the inferential mechanism, has the potential of giving unified analyses of all the pragmatic phenomena, and is sympathetic and compatible with recent developments in related areas and disciplines, notably cognitive science, psycholinguistics, neurolinguistics, clinical linguistics, natural language logic, artificial intelligence and information technology.

3.  Pragmatic theory There have been several sophisticated theories of pragmatics, each competing with the others in trying to achieve explanatory adequacy. Only one of them, relevance theory, is presented here in detail and applied to the analysis of cases in Chinese.1 Aiming at providing a cognitive account of utterance interpretation, relevance theory tries to formulate a model of pragmatic inference with psychological reality, i.e. true to what people do in real-life communication. The theory starts with the underdeterminacy thesis, arguing that every sentence needs to have its encoded meaning developed through inference to reach the more complete meaning. Meaning construction in this sense is carried out in the cognitive context of the hearer, formed by the new assumptions in the utterance he receives and the old assumptions he retrieves from his memory in order to process the new ones. When old and new assumptions interact, cognitive effects can be generated, in the form of deductive conclusions or strengthened assumptions or updated ones that contradict and replace some older ones. But the cognitive context can be a vast pool of assumptions in which old assumptions can freely interact with the new ones, and the resulting cognitive effects will initiate even newer interactions, ad infinitum. The hearer has no time nor resources to process and evaluate every conceivable cognitive effect before deciding on the one he takes to be the intended meaning conveyed by the speaker. What he can do is to pick one cognitive effect he can get without exerting too much cognitive effort and take it to be the one intended by the speaker. This will not sound so arbitrary when several related factors are taken into consideration. First, the multitude of old assumptions in the cognitive context are not scattered in disarray, but are ranked and stacked, according to their different degrees of accessibility in the memory. As old assumptions are retrieved for the purpose of processing the new assumption, the most readily retrieved must be the most accessible, which is most likely to interact with the new assumptions, and most likely to yield enough cognitive effects without costing undue processing effort. Such an interpretation of the utterance is termed an optimally relevant interpretation. Second, the hearer is not working all by himself. He is also aided by the speaker in an unconscious, yet intuitively quasi-cooperative way. The speaker knows that, in order to get her meaning across, there is an optimally relevant way to express her thoughts, one that matches with the somewhat individualistic knowledge state of the hearer to the extent that this is made known to her. That should be the manner of expression she unconsciously chooses to employ. Third, according to relevance theory, communication is by nature a guess work, miscommunication does occur frequently, especially between parties 187

Yan Jiang

who do not make pertinent judgments of each other’s knowledge state. When communication occurs, the hearer can only presume that the speaker has something to inform that will let him derive enough cognitive effects without exerting unjustified extra processing effort. This is called the presumption of optimal relevance. The speaker chooses her wording with the knowledge that the hearer uses the presumption of optimal relevance as an expectation in communication. Hence her wording is likely to help the hearer to infer her meaning, and she can have some degree of confidence that her utterance will be properly comprehended. And the hearer will take his first interpretation as the intended meaning of the speaker, without bothering to entertain alternative interpretations. This intersubjective production and comprehension process makes it possible for communication to succeed with much efficiency, even though it does not guarantee success, as a presumption is the weakest assumption and often turns out not to be failsafe, especially when the communicative parties are not able to make the right judgments of each other’s knowledge state due to unfamiliarity, lapse of memory, or physical or mental exhaustion. Relevance theory summarizes the above considerations into the Principle of Relevance: Every utterance comes with a presumption of its own optimal relevance. Utterance comprehension can now be viewed as a cognitive act in search of relevance. The underdetermined meaning of an utterance is developed to the extent that optimal relevance is attained, which is also the first relevant interpretation that the hearer can read off the processed utterance. Hence comprehension, and communication in general, appear to be instantaneous and effortless. The relevance-theoretic comprehension mechanism can provide a programmatic characterization to the whole range of pragmatic phenomena introduced above. It also has the potential to give accounts to specific cases, with each account being a detailed case-based story. Overall, such a pragmatics starts with the encoded meaning of the utterance, taking the lexical and grammatical meanings as providing conceptual or procedural information. The former initiates other relevant assumptions, the latter instructs the hearer on how to establish relevance: whether to trigger a presupposition or to derive an implicature or to take the proposition with a special attitude or to obtain a specific deictic referent. Assumption introduction and deduction is guided by procedural cues, and is constrained by the principle of relevance, making it possible for pragmatic phenomena to be comprehended in no time, in spite of the underdeterminacy of explicitly encoded meaning.2 Relevance theory is particularly helpful in that it can be used to investigate topics that are not wholly linguistic in nature. In such cross-disciplinary studies, the theory can often provide a novel explanatory perspective. The next section examines a range of pragmatic cases in Chinese, showing how relevance theory can provide original and specific accounts as well as demonstrating how the study of Chinese pragmatics can make its unique contributions to linguistic pragmatics in both empirical and theoretical aspects.

4.  Case Study One: the presupposition of the adverb bái (白) that resembles the principle of relevance Textbook examples of presupposition triggers in English often appear straightforward and obvious, and textbook discussions on the topic often seem to create the impression that all the known types of presupposition triggers can be found in all languages. In reality, each language has very unique ways to encode its triggers. In Chinese, quite a number of them trigger complicated presuppositions and cannot be literally translated into English without 188

Chinese linguistics: Pragmatics

losing their triggering power. Among these, the adverb bái (白) [in vain/for free] provides an interesting case for discussion.3 Predicates modified by ‘bái’ yield two kinds of meaning: (i) doing something gratuitously, as shown in (10)–(12), and (ii) getting something (done) for free, as shown in (13)–(15): [10] wǒ gēge bái gàn yī nián, yī gè zǐér yě méi lào zhào, hái āi le yī dùn chòumà 我哥哥白干一年,一个子儿也没落着,还挨了一顿臭骂。 ‘My elder brother worked hard in vain for a whole year without getting a single penny. Instead, he got a heap of abuse.’ [11] zhēn xiǎng zhīdào wèishénme gàn le yībèizi, què méiyǒu le tuìxiūjīn, sānshí duō nián   dōu bái gàn le? qián dào nálǐ qù le nē? 真想知道为什么干了一辈子,却没有了退休金,三十多年都白干了？钱到哪里 去了呢？ ‘I really want to know why I worked for all my life, but ended up without the pension. Does more than 30 years of work come to nothing? Where did the money go?’ [12] chū gèrén zhuānjí shì hěn róngyì de shìqíng, dàn jīngbànzhě shì fēi yào zuàn nǐ yī   bǐ de, tā jué búhuì gěi nǐ bái gàn 出个人专集是很容易的事情,但经办者是非要赚你一笔的,他决不会给你白干。 ‘It is easy to publish a collection of your own works. But the publishing agent will certainly make you pay a lot for it. There is no way for him to work for you for nothing’ [13] xǐhuān xiàndài tōngsú yīnyuè de qīngniánrén shòubúliǎo xìqǔ de màn jiēzòu yǔ tīng   bú dǒng de chàngcí hé dàobái, qǐng tā bái kàn xì tā yě wèibì zuò dé zhù 喜欢现代通俗音乐的青年人受不了戏曲的慢节奏与听不懂的唱词和道白, 请他白看戏他也未必坐得住。 ‘Those young people who prefer modern pop music to traditional Chinese opera cannot appreciate the latter’s slow tempo and its incomprehensible song-lines and spoken parts. Even if they were treated to the performance for free, they would probably still fail to stay on in the theatre’ [14] miǎndiàn duì xxx guānjiàn shíkè zhǐyào zǒu chū yībù tuì mǎ de qí, shuāngfāng   jiù huì lìkè héqí; dàn xxx cǐkè jǐnzhāng zhī xià wù zǒuchū yībù jìn jiāng,   bèi yy bái dǎ yī pào 缅甸队 xxx关键时刻只要走出一步退马的棋,双方就会立刻和棋; 但xxx 此刻紧张 之下误走出一步进将,被yy白打一炮。 ‘At this critical juncture, Burmese player xxx only needed to make the move of retreating the horse, and the game would have ended with a draw. But xxx was so nervous that he advanced the general by mistake, letting player yy take his cannon for free’ [15] jiǎrú wǒ xiànzài è le, kěyǐ dào diǎnxīnpù bái ná xiē bōbō mā? 假如我现在饿了,可以到点心铺白拿些饽饽吗？ ‘If I were hungry now, could I go to the bakery and get some cakes for free?’ 189

Yan Jiang

This apparent ambiguity of ‘bái’ between the two opposite meanings of ‘loss’ and ‘gain’ is found to be two sides of the same coin: what one side loses, the other side gains, depending on the speaker’s perspectives. When the speaker talks about the subject as the receiver of a transaction, the ‘gain’ sense is conveyed; when the subject is viewed as the giver of a transaction, the ‘loss’ sense is conveyed. The kind of transaction presented by ‘bái’ is special in the sense that the presumed equilibrium between gains and losses is not reached. What ‘bái’ literally contributes as its lexical meaning is a sense of imbalance, not a negation on the transaction itself, but a claim that the transaction is not conducted according to a presupposition it triggers: that normal transactions are expected to be balanced in gains and losses. Negation of ‘bái’ will create a contrary effect, the maintenance or reinstatement of the balance-expecting presupposition. Looking at a broader range of examples involving ‘bái’, it can be concluded that this balance-expecting presupposition does not only apply to transactions involving financial gains and losses measured in terms of exertion of strength and labour. Criteria of evaluation on the presupposed balance may vary a lot, from the very concrete exemplified by [10]–[15] to the very subjective and abstract, as shown in [16]–[23]: [16] ‘bàn tiáo yú,’ tā shuō. ‘nǐ yuánlái shì tiáo wánzhěng de. wǒ hěn bàoqiàn, wǒ chūhǎi   tài yuǎn le. wǒ bǎ nǐ wǒ dōu huǐ le. búguò wǒmen shāsǐ le bù shǎo shāyú, nǐ gēn   wǒ yīqǐ, hái dǎshāng le hǎoduō tiáo. ni shāsǐ guò duōshǎo ā, hǎo yú? ni tóu   shàng zhǎng zhe nà zhī chángzuǐ, kě búshì bái zhǎng de ā.’ ‘半条鱼,’ 他说。‘你原来是条完整的。我很抱歉,我出海太远了。我把你我都 毁了。不过我们杀死了不少鲨鱼,你跟我一起,还打伤了好多条。你杀死过 多少啊,好鱼？你头上长着那只长嘴,可不是白长的啊。’ ‘ “Half fish,” he said. “Fish that you were. I am sorry that I went too far out. I ruined us both. But we have killed many sharks, you and I, and ruined many others. How many did you ever kill, old fish? You do not have that spear on your head for nothing.” ’ (Ernest Hemingway, The Old Man and the Sea) [17] gēn wǒ hǎo zěnme le? chīkuī mā? nándào wǒ jiù bái zhǎng zhème shuài le?   shuài jiùshì běnqián! 跟我好怎么了？吃亏吗？难道我就白长这么帅了？帅就是本钱！ ‘What’s wrong with dating me? Do you think I am not worthy enough? Do I own this handsome face for nothing? Being handsome is my investment capital!’ [18] nà zhǎng dé piàoliàng búshì bái zhǎng dé piàoliàng de, yào zhíqián de. 那长得漂亮不是白长得漂亮的,要值钱的。 ‘Being pretty is not for nothing. It should be rewarding’ [19] sūn jiàndōng de nèixīn yīrán bǐbó yè táo: diēmā xīnkǔ gòng tā niàn le shíwǔ   nián shū, què dé bú zháo tā yī fēn qián de hǎochù, ……bái zhǎng le yīfù   hǎo pínáng. 孙建冬的内心依然鄙薄叶陶: 爹妈辛苦供他念了十五年书,却得不着他一分钱的 好处,……白长了一副好皮囊。 ‘Inside his heart, Sun Jiandong still looked down on Ye Tao: his parents paid for his tuition for 15 years, yet did not get even one cent from him. Although Ye Tao had a handsome face, it was pretty useless’ 190

Chinese linguistics: Pragmatics

[20] wǒmen péiyǎng chūlái de réncái, dōu bú wèi zìjǐ de guójiā fúwù, wǒmen bú shì bái   péiyǎng le mā? wǒmen de tóunǎo nándào bù yīnggāi qīngxǐng yīxiē mā? 我们培养出来的人才,都不为自己的国家服务,我们不是白培养了吗？我们的 头脑难道不应该清醒一些吗？ ‘Of the talents we trained, no one served for our own country. Didn’t we train them gratuitously? Shouldn’t we be more sober on this point?’ [21] huí dào cèzhuāng yǐjīng kāishǐ màishōu le, shuǐkù píruǎn wúlì, zuìhòu de yīdiǎndiǎn   nítāng shuǐ kěliánbābā, bái kàn zhe hēicūn de rénmen yòu fēngshōu le yīhuí 回到册庄已经开始麦收了,水库疲软无力,最后的一点点泥汤水可怜巴巴, 白看着黑村的人们又丰收了一回。 ‘Back to Ce Village, wheat was harvesting. The reservoir did not have enough water. The last bits of muddy water looked miserable. People watched in vain, seeing that the Hei Village farmers had a bumper harvest again, without themselves being able to do the same.’ [22] dàjiā dōu kàn zhèngzhì, bù guǎn xuéshí, jíshǐ yǒu jiùguó de zhēnxīn, érqiě nádào   zhèngquán, yě shì shìdàolíntóu bái dèngyǎn! méiyǒu yīngfù de nénglì yǔ zhīshí 大家都看政治,不管学识,即使有救国的真心,而且拿到政权,也是事到临头 白瞪眼！没有应付的能力与知识。 ‘If we all just study politics and do not care for knowledge, even if we are genuinely dedicated to the salvation of our country and win the power, we still can only stare and do nothing in the end! That is because we will not have the ability and knowledge to run the country.’ [23] zhègè fāxiàn jiājù le wǒ de kǒngjù, yě chèdǐ dǎxiāo le wǒ dúzì yī rén xiāomiè yāoguài   de xióngxīn. shuí dōu zhīdào yīgè rén zhīshēn hé shīzǐ bódòu nà jiào bái gěi 这个发现加剧了我的恐惧,也彻底打消了我独自一人消灭妖怪的雄心。谁都知道 一个人只身和狮子搏斗那叫白给。 ‘This discovery intensified my horror, and completely took away my ambition of dealing with the monster single-handedly. Everyone knows that a single man is no match for a lion, if he wishes to grapple with it.’ In [16], the old man was prescribing his conception of the use of the spear mouth of the big fish, which was to help him kill sharks so that he could keep the big fish as his prey. If this usage was not put into action, the spear was considered to be gratuitously grown. In [17]–[19], a commonly conceived view was presupposed: good-looks should be rewarding, and rewarding in an unspecified sense, since the other end of the balancing scale is not explicated. [20] again presupposes a vague balancing scale: students who received free education are expected to render service to the country as a form of repayment. If not, the teachers will feel that their teaching work is not adequately rewarded, so are the officials who work for the student-aid system or for the education institutes in general. But the real balance is between the students who benefit from the free education and the country that provides the service. [21] presents a case in which people in village A watch people in village B harvesting, but A were not able to do the same. Hence they could only watch in vain. [22] talks about people not being able to rise to the occasion, but can only look around in vain, with no beneficiary side at all. In both [21] and [22], only one end of the balancing scale is clear, the other side is not directly involved. Finally, the ‘bái gěi’ (give in vain) in 191

Yan Jiang

[23] means being no match to the opponent, letting him take the upper hand without being able to put up a resistance. The emphasis is on ‘no match’ rather than on the balancing scale and its two ends. The above considerations lead to some modifications of the content of the presupposition triggered by ‘bái’, that is: events or processes, whether involving humans or not, are expected by humans to preserve a balance of energy flow, which may transform itself into some other forms in the process, so long as the energy is preserved in some way, i.e. labour at one end being compensated by money at the other end, good-looking having the potential of being rewarded by marrying a wealthy spouse, dedicated research being credited with good research indices, etc. On the other hand, since the criteria are wholly subjective, there is no guarantee for proper balance. The presupposition is only to expect that energy flow is preserved to the extent that the speaker does not feel the need to use ‘bái’, or he feels justified to negate ‘bái’ in his utterance. Otherwise, the balance is upset, and ‘bái’ can be used to describe the event. With this understanding, we can enrich the presupposition of ‘bái’ in the following way: Events or processes, whether involving humans or not, are expected by humans to preserve a balance of energy flow, which may transform itself into some other forms in the process, so long as the energy is preserved in some way, to the extent that the balance is still considered to hold. Upsetting the balance at either end will lead to the likelihood of the use of ‘bái’, and awareness of the maintenance of the balance provides the reason for the use of the negation of ‘bái’. The above presupposition, henceforth shortened to ‘bái-presupposition’, is probably a rather rare presupposition because it addresses the general tendency of human cognition. Put in an even stronger way, it is an embodiment of human cognitive principle, which explains most human endeavours throughout the whole human history and even the whole process of human evolution. It then comes as a no surprise that ‘bái-presupposition’ bears some resemblance to the communicative principle of relevance introduced above, which also professes to be a principle of human cognition. Applying ‘bái-presupposition’ to the case of utterance comprehension, the hearer expects that his comprehension effort will not come to nothing. That is tantamount to saying that he expects the utterance to give him some cognitive effects so that his cognitive comprehension effort will not be of no avail. And the speaker is also aware of the hearer’s expectation because her cognitive actions are also guided by the principle of relevance. So she will justify her verbal behavior by producing an utterance that will give the hearer some cognitive effects. Otherwise, even her own cognitive effort will be exerted gratuitously. In verbal communication, therefore, there is an expectation of cognitive effort preservation, to some extent, being a special case of the ‘báipresupposition’. As the concept of ‘relevance’ in relevance theory is a specially defined term, using the notions of cognitive effect and effort, which is different from the usual sense of ‘relevance’ in ordinary English language, we can envisage the possibility of using ‘bái’ to redefine the principle of relevance, making it sound more natural, at least in Chinese. Since the presupposition can be overtly revealed and maintained by using ‘bái’ with negation, we come up with the following definition: [24] The Not-bái Principle of Communication: Speaker: bú bái shuō [不白说] ‘Do not talk with no effects’ Hearer: bú bái fèi jīnglì [不白费精力] ‘Do not exert mental effort in vain’ 192

Chinese linguistics: Pragmatics

[24] follows from the more general [25], which could have been the shorter definition of the ‘bái-presupposition’, yet to do that is to make the definition circular, as it contains ‘bái’, which needs to be defined itself: [25] bú bái fèi lì [不白费力] ‘Do not exert energy in vain’ This coincidental resemblance of ‘bái-presupposition’ to the communicative principle of relevance comes as a happy surprise. But at a more deliberated level, it comes as no surprise at all. Both address one fundamental aspect of human cognition. If there are laws of human cognition to be found, relevance theory discovered the principle of relevance, and we find in ‘bái-presupposition’ a natural language embodiment of the principle in the form of a presupposition. Such a resemblance is not an impossibility.

5.  Case Study Two: the rhetorical figure ‘lièjˇın’ 列锦格 (list gem) as a touchstone in the debate between Contextualism and Semantic Minimalism As a figure of speech christened only for 30-odd years,4 list gem refers to an idiosyncratic structure in Chinese, which consists of several semantically unrelated NPs (single nouns or noun phrases) lined up in a row as one sentence, with no verb predicates nor adjectives or nouns used as predicates. The general understanding is that the figure of list gem should contain three NPs as its norm. A sentence containing only two NPs should best be categorized as couplet, as two single NPs forming a couplet is common in Chinese poetry. Besides, the interpretation of couplet differs a lot from list gem, as a couplet either forms an antithesis or expresses a sequence. On the other hand, in most cases, a sentence containing four or more stand-alone NPs is to be categorized as enumeration or escalation, which means the NPs contained therein are also related in meaning, again being very different from list gem. Classic examples of list gem are found in classical poetry, as shown by [26]–[27]. [26]

kūténg lǎoshù hūnyā xiǎoqiáo liúshuǐ rénjiā gǔdào xīfēng shòumǎ xīyáng xī xià duàncháng rén zài tiānyá



‘Dry-vine old-tree blind-crow, Small-bridge flowing-water human-home Old-road west-wind lean-horse Setting-sun falls west Heart-broken man at end of the world’

[27]

gūcūn luòrì cánxiá qīngyān lǎoshù hányā yī diǎn fēi hóng yǐng xià qīng shān lǜ shuǐ bái cǎo hóng yè huáng huā



‘Lone-village falling-sun patch-twilight, Thin-smoke old-tree jack-daw One spot of shadow cast down by the flying swan-goose Green-mountain olive-water White-grass red-leaf yellow-flower’

枯藤老树昏鸦, 小桥流水人家。 古道西风瘦马。 夕阳西下, 断肠人在天涯。

孤村落日残霞, 轻烟老树寒鸦, 一点飞鸿影下。 青山绿水, 白草红叶黄花。

193

[list gem 1] [list gem 2] [list gem 3] [subject predicate construction] [existential construction]5

[list gem 1] [list gem 2] [complex description] [couplet] [list gem 3]6

Yan Jiang

In [26], the first three lines are three cases of list gem instead of one, as they form three distinct meaning clusters. The remaining two lines do not contain list gems. Likewise in [27], even though only NPs can be found, not all lines are composed of list gems. Each of [28]–[31] are two-line couplets taken from whole poems, but each line in each couplet constitutes a list gem, which introduces further complications to the distinction between couplets and list gems. [28] jī shēng máo diàn yuè, rén jì bǎn qiáo shuāng 鸡声茅店月,人迹板桥霜。[two list gems] ‘Rooster-sound small-inn moon/Foot-print wood-bridge frost’ [29] táolǐ chūnfēng yībēi jiǔ, jiānghú yèyǔ shínián dēng 桃李春风一杯酒,江湖夜雨十年灯。[two list gems] ‘Peach-plum spring-wind a-cup-of-wine/River-lake night-rain ten-year-light’ [30] sān shí gōng míng chén yǔ tǔ, bā qiān lǐ lù yún hé yuè 三十功名尘与土,八千里路云和月。[two list gems] ‘Thirty-feat-fame dust and mud/Eight-thousand-li-road cloud and month’ [31] wèn rǔ píng shēng shì yè, huáng zhōu huì zhōu dān zhōu 问汝平生事业,黄州惠州儋州。[list gem in the second line] ‘Ask you life achievement/Huáng-Town Huì-Town Dān-Town’ The interpretation of list gems could be an interesting topic in cognitive–pragmatic studies of rhetoric, as there do not seem to exist similar structures in languages that have been well exposed to pragmatic studies. Since a list gem forms a sentence but does not have a predicate, it seems necessary to accommodate one through explicating, so as to obtain the complete explicature with a saturated argument–predicate structure. But the three NPs in the list gem are no more than keywords. Each one may be accommodated with a different predicate. And the thorny issue is that different language users may supply different predicates to the same NP, to the extent that they are confident enough in obtaining the explicature of the sentence. Even the poet himself may find it hard to supply the exact predicate, for there simply isn’t one that can capture the rich meaning left open by the non-presence of any predicates, which is the hallmark of implicit meaning. Perhaps the best an interpreter can achieve is some impressions created by the line-up of the three stand-alone NPs. He can try to make inferences to establish the links among the three, but he is always ill at ease in deciding on the one single literal meaning for the sentence. At the same time, he can obtain a rich array of weak implicatures, appreciating fully the poetic effects of list gem. The comprehension process is unavoidably guided by the principle of relevance. The interpreter understands list gem in the optimally relevant way, resulting in a very unclear explicature and a set of weak implicatures. But to the mind of an individual interpreter, some weak implicatures may be even clearer than the explicature. One important theoretical bearing of this case has to do with the recent debate between Contextualism and Semantic Minimalism in philosophy of language.7 Contextualists hold the view that the exact truth value cannot be determined at the sentence level, because meaning directly gleaned from the logical form of any sentence is unavoidably underdetermined and cannot be recovered without the use of free pragmatic inference. Hence truth-conditional semantics cannot be self-sufficient, as pragmatic processes always intrude into the determination 194

Chinese linguistics: Pragmatics

of literal meaning of a sentence, without which the truth value can never be ascertained. Truth can only be established at the level of truth-conditional pragmatics. On the other hand, Semantic Minimalists contend that truth-conditional semantics can still be self-contained and adequate. Although meaning is underdetermined at sentence level, the logical form of the sentence can be reanalyzed so that motivated indexicals can be posited. Thus all pragmatic processes aiming to derive perfect literal meaning are activated by indices, which are repre­ sented as free variables in the logical form and instantiated through pragmatic inference. Truth-conditional semantics is thus salvaged because semantic interpretation refers to all the items in the logical form, including interpretive rules for all the indices. It is the indices that refer outward to contextual information. Pragmatic intrusion is fended off at sentence level. Arguments contributing to this debate come in two kinds. From a theoretical point of view, it is important to propose claims and mechanisms that are strong in explanatory power. From an empirical point of view, it is also important that either side can take into consideration a wide range of structures and cases, including those in different languages that prove to be challenging to existing accounts. In such a testing ground, Contextualism and Semantic minimalism can compete with each other, and the results can best reveal the advantages or inadequacies of the either side. List gem appears to provide a strong case in support of Contextualism, and a knock-down case against Semantic Minimalism, as it is impossible to determine the explicature of list gem sentences: language users don’t bother to flesh out the complete meaning of list gems. If pressed, one may supply an ad hoc explicature, which is most likely to be very different from the one constructed for the second time when he is asked to explicate the very same sentence. When two people are asked to explicate the same list gem, they always differ a lot from each other in what each can produce. Hence, classic list gem examples never have standard explicatures. Contextualists can therefore use list gem to show that the truthconditional content cannot be established at sentence level, as there is no commonly accepted way to explicate a list gem. However, Semantic Minimalists can find an escape hatch by arguing that list gems do not have explicatures at all: the whole point of using list gems is to let them create impressions, which lead to the derivation of weak implicatures. In list gems, we find a rare case of language use which has a lot of implicatures but no agreed explicatures.

6.  Case Study Three: the reach of counterfactual meaning in Chinese – a unique logic? How counterfactual conditionals are expressed in Mandarin Chinese (and in Chinese dialects) has remained a mystery since the advent of modern linguistics.8 Unlike Indo-European languages which almost always use verb inflexion to mark counterfactuals, Chinese lacks explicit morphology to distinguish counterfactual conditionals from indicative ones. Although there are a few lexicalized constructions such as ‘yàobúshì’ [要不是] (if-not-be) and ‘zǎozhīdào’ [早知道] (early-know) whose presence necessarily lead to counterfactual interpretations, there are many others that are interpreted counterfactually in contexts of utterance that do not make any use of these constructions. As native speakers of Chinese do perceive counter­ factual conditionals and produce them as well, the challenge is to provide an adequate mechanism that can characterize the reach of counterfactual meaning in Chinese. Right now, there are two approaches. According to the first approach, as Chinese counterfactuals are largely unmarked, Chinese conditionals are either indicative or counterfactual, depending on the information gathered from the context. It is the contextual information that provides the 195

Yan Jiang

interpretive guidance. If the antecedent of a conditional makes a contrary-to-fact hypothesis in content (though not in form), counterfactual meaning is reached through pragmatic inference.9 This thesis can lead to the claim that every conditional is ambiguous between indicative and counterfactual. But native speakers do not seem to perceive conditionals in this way. At least, more specific claims are needed to explain how context can determine counterfactuality for a conditional. Another approach argues that some words and particles in Chinese serve as markers or weak features of counterfactuality.10 Although each of them may be too weak to make it happen, some such features can work together in some subtle ways to bring about counterfactuality. Such an approach also needs more clarifications. Weak features seem to be all optional in nature. Although sometimes, their presence makes the difference and counterfactuality is reached, some other times, similar counterfactual meaning can be reached without any of the weak features being present. Again, it is not clear how the mechanism works. The account we propose is different from either of the above. We start by examining the semantics of the explicit counterfactuals, i.e. the ‘yàobúshì’ [要不是] (If-not-be) conditionals, followed by a comparison between ‘yàobúshì’ and ‘me51hau3’ [蛮好] (Much-preferred) in Shanghainese, reaching the conclusion that they are by nature falsifying and truthifying counterfactuals. Then we try to generalize the findings to the majority of Chinese counterfactuals, claiming that they can be treated on a par with the ‘yàobúshì’ and ‘me51hau3’ counterfactuals. Some examples involving ‘yàobúshì’ are given below. [32] dāngchū, wǒ cóng diànyǐngyuàn cízhí láidào běijīng de shíhòu, jiālǐrén jí tóngshì,   péngyǒu jiù jílì fǎnduì. yàobúshì wǒ yìng xià yītiáo xīn, gēnběn jiù cí bú liǎo zhí,   gèng lái bú liǎo běijīng. (当初,我从电影院辞职来到北京的时候,家里人及同事、朋友就极力反对。要不 是我硬下一条心,根本就辞不了职,更来不了北京。) ‘Initially, when I quit my cinema job to come to Beijing, my family members, colleagues and friends were all dead against it. Had I not made up my mind, I would not have been able to resign, nor would I have been able to come and make it in Beijing.’ [33] 1:0, zhōngguó zúqiúduì yíng de gòu xuán de, bǐsài jìnxíng de yě gòu xié de.   yàobúshì Hǎo Hǎidōng nà yǒudiǎn yùnqì de jìnqiú, píngjú kěndìng huì   zhāo lái bùmǎn hé fēiyì. suǒyǐ, shuǎng bú qǐlái. (1:0, 中国足球队赢得够悬的,比赛进行得也够邪的。要不是郝海东那有点运气 的进球,平局肯定会招来不满和非议。所以,爽不起来。) ‘One: Nil. The Chinese football team won a very close match with real good luck. Had it not been for Hao Haidong to score that lucky goal, a tie would surely have incurred dissatisfaction and complaints. That is why nobody took it with a light heart.’ [34] xīlà dàibiǎotuán suī shì zuìhòu yīgè rùchǎng, dàn huòdé de zhǎngshēng shì zuì jiǔ,   zuì rèliè de. zhè bú qíguài, yīnwèi tāmen shì dōngdàozhǔ, guānzhòng méiyǒu   lǐyóu lìnxī zìjǐ de zhǎngshēng. zhídé yītí de shì, xīlà dàibiǎotuán cóng zhǔnbèi   rùchǎng dào ràochǎng yīzhōu duō hòu, gòng dédào le 18 fēn 2 miǎo zhōng de   zhǎngshēng, yàobúshì zhǔchírén dǎduàn, hěn nánshuō gǔzhǎng néng chíxù dào   shēnme shíhòu. 196

Chinese linguistics: Pragmatics

(希腊代表团虽是最后一个入场,但获得的掌声是最久、最热烈的。这不奇怪, 因为他们是东道主,观众没有理由吝惜自己的掌声。值得一提的是,希腊代表 团从准备入场到绕场一周多后,共得到了18分2秒钟的掌声,要不是主持人 打断,很难说鼓掌能持续到什么时候。) ‘Although the Greek delegation were the last to enter the stadium, they won the longest and loudest round of applause. This is hardly surprising, as they represented the host country and the spectators had no reason to save on their applauses. It is worth pointing out that the Greek delegation won a round of applause lasting for 18 minutes and 2 seconds, from the time they prepared to enter the stadium to the end of one lap’s march. Had it not been for the ceremony presenter who put the cheers to an end, it would have been hard to tell how long the applause would last.’ In the above examples, ‘yàobúshì’, though internally complex, behaves exactly like a conditional functor. It is used as a single lexical item, different from the compositional meaning of ‘yàoshì  .  .  .  bú’/if  .  .  .  not/‘要是  .  .  .  不’. ‘yàoshì  .  .  .  bú’ is not an explicit marker of counterfactual conditionals, because it can also form indicative conditionals. ‘yàoshì  .  .  .  bú’ resembles ‘if  .  .  .  not’ in English, but ‘yàobúshì’ is similar to the English ‘had it not been for  .  .  .’ in meaning. ‘Yàobúshì’ is internally complex for a special reason. It takes on a proposition P, which should be about a state or event that is both true and real, and returns a protasis P* which is counterfactual. Here, negation obviously plays a vital role, but negation is only a necessary condition, not a sufficient one. In the ‘yàoshì  .  .  .  bú’ construction, which also forms a protasis, negation works as well, but the resulting protasis does not have to be contrary to fact. What is more, the proposition led by ‘yàobúshì’ should not be some proposition about abstract, general, timeless states. This can be established through both introspection and corpus search. Try to create a ‘yàobúshì’ counterfactual with abstract ideas, the resulting conditional is bound to be weird. What also tend to be anomalous are ‘yàobúshì’ counterfactuals containing a proposition with an extra layer of negation. To say the least, they are difficult to process. To sharpen the above characterizations of ‘yàobúshì’, we look for more precise theoretical concepts in linguistic and philosophical studies. One viable notion to adopt is (subjective) veridicality, whose definition can be found in Giannakidou and Mari (2014): ‘truth judgments depending on what epistemic agents know or believe to be true, and other factors in the context relating to the epistemic status of individuals’. Aided by this new definition, ‘yàobúshì’ can be characterized as an operator taking a veridical proposition as its argument: what the agent knows or believes to be true. We can also call ‘yàobúshì’ a veridical operator. A similar operator in Mandarin is ‘zǎozhīdào  .  .  .’ (早知道) /Early know  .  .  .   /, which, however, is not a conditional connective, as it can follow ‘rúguǒ’ or ‘yàoshì’: [35] Tā xiàng xiāoxié tóngzhì kūsù shuō: ‘yàoshì zǎo zhīdào shàngdàng shòupiàn,   wǒ jiù bú gàn le. xiànzài nòng dé yīshēn máfan, xiōngdì jiěmèi dōu bù xiāng rèn,   hái jīyā le yīdàduī méi tuīxiāo chūqù de chǎnpǐn, wǒ qù zhǎo shuí a?’ (她向消协同志哭诉说: ‘要是早知道上当受骗,我就不干了。现在弄得一身麻烦, 兄弟姐妹都不相认,还积压了一大堆没推销出去的产品,我去找谁啊？’) ‘She told staff from Consumer’s Association in tears, “Had I known it was a fraud, I would not have got involved. Now I am all troubles. My brothers and sisters have all turned away from me. And I am stuck with loads of unsold goods. Who can I turn for help?” ’ 197

Yan Jiang

Subjective veridicality constitutes one necessary condition for the use of ‘yàobúshì’, but it still does not capture the intuition that ‘yàobúshì’ does not take on abstract propositions like science or math laws, even though they can be subjectively established as truth. In Armstrong (2002), non-abstract counterfactuals are given the name contingent counterfactuals: ‘Contingent counterfactual claims are often to be found in ordinary discourse, for instance, “If you had not put your foot on the brake so promptly just then, there would have been a nasty accident.” ’ Talks of historical contingencies are also frequently encountered: ‘What if there had been no American War of Independence? What if Ireland had never been divided? What if Britain had stayed out of the First World War? What if Hitler had invaded Britain or had defeated the Soviet Union? What if the Russians had won the Cold War? What if Kennedy had lived? What if there had been no Gorbachev?’ (Ferguson 1999). Another useful source is Rescher (2007), in which he distinguishes between falsifying and truthifying causal counterfactuals. Falsifying counterfactuals hypothesize what actually did happen had not happened, whereas truthifying counterfactuals hypothesize what did not actually happen had happened. According to Rescher, historical counterfactuals of the falsifying type ‘are in general retrospectively cause-determinative in nature’ and ‘generally address the preconditions for an actual occurrence’. Such counterfactuals are less speculative and more situation-bound. This rightly fits the characterization of ‘yàobúshì’ counterfactuals. So what follows ‘yàobúshì’ can now be re-characterized as a proposition which is both veridical and a falsifying contingent. And Armstrong’s term of contingent counterfactuals can now be elucidated as contingency-referenced falsifying counterfactuals. The whole protasis led by ‘yàobúshì’ can therefore be taken as an encoded falsifying counterfactual context. On the other hand, some structures can be encoded as counterfactual, but allowing either falsifying or truthifying possibilities. Jiang and Wang (forthcoming) characterize the behavior of the Shanghainese ‘me51hau3’ (蛮好……) /Much-preferred  .  .  .   / as a counter­ factual desiderative, which can give a truthifying reading: [36]

Mehau ganggang cen ditik qi nao!11 Much-preferred just-now take underground go EM-SMP 蛮好刚刚乘地铁启孬！ (EM-SMP = emotive sentence-final particle) ‘It would have been much better that we took the underground at the time’ 刚才要是坐地铁去就好了。

But the negation of P in such a context turns the whole sentence into a falsifying one: [37]

Mehau nong ganggang vyao gang bak yi tin nao Much-preferred you just-now do-not tell to him hear EM-SMP 蛮好侬刚刚勿要岗摆伊听孬！ ‘It would have been much better that you did not let him/her know (that) at the time’ 你刚才要是不告诉他就好了。

The proposition following ‘me51hau3’ is necessarily antiveridical. That is, the speaker presupposes that the truth-value of the proposition is false. But it can either take the positive form with a truthifying reading, as in [36], or adopt a negated form with a falsifying reading, as in [37]. Thus, ‘yàobúshì + S’ and ‘me51hau3 + S’ share the same feature of being encoded forms of counterfactual meaning, even though the former S is marked by [+veridical], and 198

Chinese linguistics: Pragmatics

the latter, by [+antiveridical]. This comparative study provides a clue to the treatment of implicit counterfactual conditionals in Mandarin. In addition, sentences involving ‘zǎozhīdào  .  .  .’ (早知道) /Early know  .  .  .   / should now be taken as encoded truthifying counterfactual constructions, as shown in [35]. Previous psycholinguistic studies have revealed that native speakers of Chinese are not used to processing unexpected, out of the blue, implicit counterfactual conditionals. But they are ready to weigh up different contingencies if enough contextual information is supplied.12 Relating this finding to our characterization about explicit counterfactuals, we can further hypothesize that even in the case of implicit counterfactuals, native speakers of Chinese still treat them as either falsifying or truthifying counterfactuals. The question then boils down to how a Chinese can tell the difference between an indicative conditional and a falsifying/ truthifying counterfactual conditional. The cues, we suggest, are the weak features, such as the aspect marker ‘le’ 了 (optional marker of perfect aspect), the temporal adverb ‘zǎo’ 早 (early), and the negation words such as ‘bú’ 不 (not) and ‘méi’ 没 (not-in-past). These words have varied meaning and functions, but when used in conditionals, they share the similar function of indicating a change of eventuality, presenting an alternative which falsifies the current case or truthifies what is not there by making things happen. Hence the counterfactual meaning. Related examples are given below: [38] yàoshì nǐ qù le, jiù bú huì bú zhùyì tā nà piàoliàng de tàitài de [truthifying] 要是你去了,就不会不注意他那漂亮的太太的。 ‘If you had been there, you would not have failed to notice his pretty wife’ [39] rúguǒ wǒ zǎo zhīdào tā bú lái, wǒ yě bú huì lái le [truthifying] 如果我早知道他不来,我也不会来了。 ‘If I knew it earlier that he would not come, I would not come either’ [40] yàoshì wǒ zhēnde zhòng le liùhécǎi, nà wǒ mǎshàng jiù huì cízhí [truthifying] 要是我真的中了六合彩,那我马上就会辞职。 ‘If I really won the Mark Six, I would quit my work immediately’ [41] yàoshì gāngcái nàgè rènyìqiú méi fá jìn, jiù huì tī jiāshísài le [falsifying] 要是刚才那个任意球没罚进,就会踢加时赛了。 ‘If that free kick had not been in, the match would have gone into extra-time’ [42] zhè chǎng huǒ, xìngkuī xiāofángduì lái dé zǎo. fǒuzé hòuguǒ búkān shèxiǎng [falsifying] 这场火,幸亏消防队来得早。否则后果不堪设想。 ‘This fire, lucky was it that firemen arrived in good time [to put it out]. Otherwise, it would have caused unimaginable damage.’ Having treated explicit and implicit counterfactual conditionals, the remaining counterfactuals are the conditionals that hypothesize a situation known to be blatantly false according to language users’ knowledge of the world: situations that are absurd, improbable or self-contradictory. Some such examples are given below: [43] yàoshì huàn le wǒ de huà, jiù búhuì duì tā zhème kèqì le 要是换了我的话,就不会对他这么客气了。 ‘If I were to deal with the case, I would not be so nice to him’ 199

Yan Jiang

[44] rúguǒ tàiyáng cóng xībiān chūlái, wǒ yīdìng jià gěi nǐ 如果太阳从西边出来,我一定嫁给你。 ‘If the sun comes out from the west, I will be your wife for sure’ [45] yàoshì tā shuō èr jiā èr děngyú wǔ – nàme hǎo de, èr jiā èr jiù děngyú wǔ 要是他说二加二等于五——那么好的,二加二就等于五。 ‘If he says that two and two are five – well, two and two are five’ [George Orwell: Looking Back on the Spanish War] [46] jiǎrú wǒ yǒu yīshuāng chìbǎng, wǒ xiǎng xiàng xiǎoniǎo yīyàng zìyóu de fēixiáng 假如我有一双翅膀,我想像小鸟一样自由地飞翔。 ‘If I had a pair of wings, I’d wish to fly freely like a little bird’ The above investigation may give the impression that only semantic factors are taken into consideration. But, gradually, we have outlined an inferential path for the comprehension of Chinese counterfactual conditionals, and the overall reasoning is a matter of pragmatic inference – in our present framework of analysis – a comprehension inference conducted under the guidance of the principle of relevance. Whether a sentence is interpreted counterfactually depends on the following factors: [47] The encoded meaning of ‘yàobúshì’ and a cluster of similar expressions [48] The identification of the contingent fact under discussion [49] The pragmatic function of the weak features [50] The identification of the absurd, the impossible, or the self-contradictory content Obviously, [48]–[50] are pragmatic factors. The typology of Chinese counterfactual conditionals is now summarized in the following graph: [51] 

Counterfactual Conditionals Explicit (encoded) Falsifying

Implicit (inferred) Truthifying Counterpossibles

7. Epilogue This is an extended introduction to Chinese theoretical pragmatics, or linguistic pragmatics for Chinese. After introducing some general concepts and phenomena, I present relevancetheoretic pragmatics and apply theoretical notions to three case studies. With these I hope to show how pragmatics can be put to work in linguistic analysis. It is important for students of Chinese grammar to seek insights in pragmatics so as to give novel accounts of some puzzling grammatical issues. It is equally important for workers of Chinese pragmatics to roll up their sleeves and work on actual topics, rather than always remain at the stage of reading, surveying and evaluating works and ideas from the West. 200

Chinese linguistics: Pragmatics

Notes   1 Cf. Blakemore (1992) and Clark (2013) for textbook introductions, and Sperber and Wilson (1995), Wilson and Sperber (2012) for more detailed discussions.   2 More studies can be found in Carston (1988, 2002) and Soria and Romero (2010) on explicature, and in Blakemore (1987) and Escandell-Vidal et al. (2011) on procedural meaning.   3 Substantive studies on this topic can be found in Zhang (2000).   4 Cf. Tan (1996) for its naming history.   5 Composed by Mǎ Zhìyuan 马致远 [c.1250–c.1321].   6 Composed by Bái Pǔ 白朴 [1226–c.1306].   7 Cf. Borg (2004), Carston (2004, 2010), and Stanley (2000).   8 Wang (2013) contains a survey on this topic.   9 Cf. Jiang (2000). 10 Cf. Chen (1988) and Wang (2013). 11 As tone sandhi in Shanghainese varies a lot in different phonetic contexts, it is customary for nonphonetic studies to omit tones in example sentences. 12 Cf. Bloom (1981).

References Armstrong, David. M. (2002) ‘Truths and Truthmakers’, in Richard Schantz (ed) What is Truth? Current Issues in Theoretical Philosophy vol. 1, Berlin and New York: Walter de Gruyter, 27–37. Blakemore, Diane (1987) Semantic Constraints on Relevance, Oxford: Blackwell. Blakemore, Diane (1992) Understanding Utterances, Oxford: Blackwell. Bloom, Alfred (1981) The Linguistic Shaping of Thought, Hillsdale, NJ: Lawrence Erlbaum. Borg, Emma (2004) Minimal Semantics, Oxford: Clarendon. Carston, Robyn (1988) ‘Implicature, Explicature, and Truth-theoretic Semantics’, in Ruth Kempson (ed) Mental Representations: The Interface between Language and Reality, Cambridge: Cambridge University Press, 155–81. Carston, Robyn (2002) Thoughts and Utterances: The Pragmatics of Explicit Communication, Oxford: Blackwell. Carston, Robyn (2004) ‘Explicature and Semantics’, in Steven Davis and Brendan Gillon (eds) Semantics: A Reader, Oxford: Oxford University Press, 817–45. Carston, Robyn (2010) ‘Explicit Communication and “Free” Pragmatic Enrichment’, in Belén Soria and Esther Romero (eds) Explicit Communication: Robyn Carston’s Pragmatics, Basingstoke: Palgrave Macmillan, 217–85. Chen, Guohua 陈国华 (1988)〈英汉假设条件句比较〉(‘A Comparison between English and Chinese Hypothetical Conditionals’),《外语教学与研究》(Foreign Language Teaching and Research) 73: 10–19. Clark, Billy (2013) Relevance Theory, Cambridge: Cambridge University Press. Escandell-Vidal, Victoria, Manuel Leonetti, and Aoife Ahern (eds) (2011) Procedural Meaning: Problems and Perspectives, Bingley: Emerald Group Publishing. Ferguson, Niall (1999) ‘Virtual History: Towards a “Chaotic” Theory of the Past’, in Niall Ferguson (ed) Virtual History: Alternatives and Counterfactuals, New York: Basic Books, 1–90. Giannakidou, Anastasia and Alda Mari (2014) ‘Future and Universal Epistemic Modals: Reasoning with Non-veridicality and Partial Knowledge’, Manuscript. Jiang, Yan 蒋严 (2000)〈汉语条件句的违实解释〉(‘On Counterfactual Interpretation of Chinese Conditionals’), in《语法研究和探索》(Grammatical Analysis and Investigation), vol. 10, Beijing: The Commercial Press, 257–79. Jiang, Yan and Wang Yuying 王宇婴 (forthcoming) ‘Counterfactual Subjunctive Assertions in Shanghai Dialect.’ To appear in Commemorative Essays on the 120 th Birthday of Professor Y. R. Chao. Levinson, Stephen. C. (1983) Pragmatics, Cambridge: Cambridge University Press. Leech, Geoffrey (1983) Principles of Pragmatics, London: Longman. Rescher, Nicholas (2007) Conditionals, Cambridge, MA: The MIT Press. Soria, Belén and Esther Romero (eds) (2010) Explicit Communication: Robyn Carston’s Pragmatics, Basingstoke: Palgrave Macmillan.

201

Yan Jiang Sperber, Dan and Deirdre Wilson (1986/1995) Relevance: Communication and Cognition, Oxford: Blackwell. Stanley, Jason (2000) ‘Context and Logical Form’, Linguistics and Philosophy 23(4): 391–434. Tan, Yongxiang 谭永祥 (1996)《修辞新格》(New Figures of Speech), Guangzhou: Jinan University Press. Wang, Yuying 王宇婴 (2013) The Ingredients of Counterfactuality in Mandarin Chinese, Beijing: China Social Science Press. Wilson, Deirdre and Dan Sperber (2012) Meaning and Relevance, Cambridge: Cambridge University Press. Zhang, Yisheng 张谊生 (2000)《现代汉语副词研究》(Studies on Adverbs in Modern Chinese), Shanghai: Xuelin Press.

202

12 Chinese Linguistics: Semantics Hsieh Shu-Kai National Taiwan University, Taiwan

1. Introduction Semantics talks about the meaning of language, and the role it plays in actual communication. In this subfield of linguistics, there have been studies focusing on different levels of units (morphemes, words, sentences, discourse) with different approaches (sense: lexical semantics; denotation: formal semantics; mental image: cognitive semantics, etc.). This chapter does not aim to achieve a comprehensive introduction of related works, but will try to convey more specialty in the semantic studies of Chinese.1 In singling out the relevant topics for inclusion in this chapter, the criteria I have used are firstly, giving priority to the particular properties of Chinese language, and secondly, skipping technical details for a smoother reading of the non-specialists who are curious about Chinese semantics. The intricate relation between language and writing system in Chinese has arguably led the studies of Chinese semantics much more lexically driven, necessitating much deeper focus on the issues of lexical meanings. This has motivated the organization and approaches of this chapter. In the following, Section 2 starts with addressing the basic ingredients of a semantic theory of Chinese, with the main focus on the interface of morphology and semantics, Section 3 moves to the topics of lexical semantics, and Section 4 summarizes the chapter.

2.  Chinese morpho-semantics Most (Chinese) linguistic researches assume the notion of wordhood. In the developmental history of linguistic theory, words have been supposed to be the core linguistic units for the sake of scientific investigation of languages.2 Among different versions of the definition of ‘word’, the most popular one, ‘the minimum (meaningful) free form’ (Bloomfield 1926: 155), claimed to be universally applicable to all languages. Morphologically, new words can be created via different ways of word formation, and result in different types of complex words. For example, you could coin the word payable, deriving it from the verb pay by adding the suffix –able, or you could form the compound fast food by combining two words to form a composite one. However, the semantics of word formation in Chinese is a field that is much less clear than its counterparts in other languages. This is mainly due to the controversial status of 203

Hsieh Shu-Kai

word, which has resulted from the intricate relationship between ideographic writing system and language. In the Chinese writing system, in principle, each character 字 (zi4) represents one syllable from the aspect of sound, and one morpheme which carries strong semantic weight. Word inventory in contemporary Chinese thus consists of: a single morpheme, or combination of two (and arguably three) morphemes. It was reported and can be empirically justified that in many cases, whether a single morpheme or the combination of morphemes constitute a word or not is open to debate, let alone the unstable notion of compound which is built on top of word. This confusion regarding what a word is in Chinese among lay speakers3 and disagreement among linguists has worsened due to the fact that the relative predominance of the monosyllabic word in Ancient Chinese has shifted to bisyllabic words in Modern Chinese. The huge amount of semantic information encoded by the characters has deeply ingrained the idea of character-centered lexicon in Chinese minds. It is noted that although it is astonishingly difficult to give a definition of ‘word’, it does not necessarily follow that it is impossible. Taking an understanding of the above-mentioned background as the point of departure, in the following section, we will take a closer look at the morpho-semantics of Chinese by way of its quantitative behavior and conceptual representation.

2.1.  Chinese morphemes, words and compounds Linguistically, words are made up of at least one meaningful piece, which is defined as a morpheme, the minimal meaningful unit that is used to form words. Morphemes can be further divided into free morphemes, those that can stand alone; and bound morphemes, those cannot stand alone and must be attached to other morphemes to form words. Bound morphemes come in different varieties: those that come before the base of the word are prefixes, and suffixes are bound morphemes that come after the base of the word (Lieber 2009). As remarked previously, we can take the characters for the orthographic unit representing morphemes in Chinese, but still encounter problems. For instance, 健康 (jian4-kang1, ‘health’) is a word consisting of two bound morphemes, while 财富 (cai2-fu4, ‘wealth’), though recognized as a word, consists of two morphemes where the boundedness (i.e. whether they can be used alone in Modern Chinese) is a question mark. The indeterminacy of morphological status then affects the decision of whether 财富 is one word composed of two bound morphemes (财 and 富), or a compound word composed of two words that are both free morphemes, respectively.4 To tackle the issue, Chinese linguists have proposed different kinds of wordhood test, such as insertion criteria (Lu 1957), exocentric structure (Chao 1968), conjunction reduction (J. Huang 1984), etc. However, none of the criteria seem to be always reliable to license a true word in every case, leaving the problem still unsolved. In addition, orthographically, the lack of inflectional markers and word delimiters such as space in texts, also worsens the achievement of consensus regarding the distinction between words, compounds and phrases. Interplayed with structural ambiguity, the automatic word segmentation has become a long-standing heated topic in Chinese Natural Language Processing (NLP). For instance, the phrase ‘已结婚的和尚未结婚的青年’ can be grammatically segmented into two structural readings: ‘已/结婚/的/和/尚未/结婚/的/青年’ and ‘已/结婚/的/ 和尚/未/结婚/的/青年’. This has led to diverse segmentation principles in the field of Chinese NLP, where a set of operational rules is provided to guide word segmentation for engineering purpose.5 In addition to the linguistic discussion, we can also observe the intricate morpheme behavior in a quantitative way, which falls under the rubric of morphological productivity. 204

Chinese linguistics: Semantics

The notion of morphological productivity in the field of Morphology refers to the process of lexeme formation that can be used by native speakers to form potential words, in particular the process of derivational affixation that involves a base word and an affix. In English for example, though the suffix –th might be recognized as a suffix such as in warmth, it is rarely used to make new words, compared to other suffixes such as –ity and –ness (Lieber 2009). That is, different morphemes have different potentials to form words. A number of ideas have been proposed to measure the degree of productivity. Baayen (1992) considered the hapax legomena (i.e. words that appear only once in a given corpus), and proposed a wellrecognized measure P = n1/N, where N is the number of tokens (e.g. of all nouns ending in –ness), n1 represents the hapax; while Nishimoto (2003) proposes a type-based measure instead, and reports the yielding results of productivity ranking of five Chinese suffixes ‘–men, –hua, –r, –zi, –tou’ where –men is the most productive and –tou is the least productive. For the sake of illustrative purposes, the author calculated the Baayen’s P values for four major suffixes in Chinese based on a sample of the Academia Sinica Corpus,6 with the results of ((noun plural –men 们: 0.53), (nominal derivational suffix –zi 子: 0.35), (noun-forming suffix –lu 率: 0.45), (occupational suffix –jia 家: 0.32), (experiential suffix –guo 过: 0.33), (experiential suffix –jian 见: 0.36)). Comparing with some sample P values from English suffixes (Sproat 2002) , such as ((–ness: 0.0044), (–ish: 0.0034), (–action: 0.0006), (–ity: 0.0007)), we can find a significant difference showing that Chinese suffixes might have higher morphological productivity in terms of Baayen’s P value. The experiment can give us a quick idea of how Chinese morphology, due to its ideographic characters (morphemes), differs essentially from other languages. In fact, the notion of suffix does not fit well with Chinese morphology either. A suffix is defined as a bound morpheme in linguistics, which is an issue in dispute in Chinese. To highlight the specialty, Packard (2000: 77–8) proposes that unlike ‘typical’ affixing languages, Chinese has a large class of morphemes – which he coined as bound roots – that possess certain affixal properties (i.e. they are bound and productive in forming words), but encode lexical rather than grammatical information.7 These morphemes may occur as either the left- or right-hand component of a word. For example, the morpheme 家 (jia1) is treated as an occupational suffix (-家) as mentioned in previous experiments, or prefix (as in 家-, ‘anything related to home’). However, as remarked repeatedly, the difficulty is exacerbated by the fact that the boundary between bound and free morpheme is mostly fuzzy. A native speaker can easily come up with a sentence where the morpheme 家 itself is a word used alone, as in 我家在那儿 (‘My house is over there’). The phenomenon leads many Chinese linguists to emphasize the role of the morphemic component (i.e. character) itself as the building block in the semantic compositional process of di- or trisyllabic words, regardless of its morphological status.8 To get a deeper understanding of the semantic compositional process, researchers have proposed various rules for deriving the meaning of a morpheme combination from the meaning of the morphemic components. Löbner (2013: 13) noted that these compositional rules are complicated by the fact that there are several interpretation rules for the same composition rule. Compounds are particularly intricate in how the meaning of the components are semantically related: a 书架 (shu1-jia4, ‘book shelf’) is a shelf on which books can be stored; a 书名 (shu1-ming2, ‘book title’) is a title that names a book; while a 书券 (shu1 quan4, ‘book coupon’) is a voucher entitling the holder to a discount of a particular book. Nevertheless, the inter­ pretation of compounds (morpheme combinations) is not just mere arbitrary guessing, for the native speaker obviously agrees on how to interpret regular compounds. Based on empirical corpus data, Sproat (2002) suggested some semantic relations such as ‘used for’ 205

Hsieh Shu-Kai

in 饭碗 (fan2wan3, ‘rice bowl’); ‘powered by’ in 电灯 (dian4deng, ‘electric light’), etc. All the suggestions indicate that there should be a conceptual knowledge system underlying the characters. The following section will lead the reader to the studies of how Chinese character-morphemes encode semantic-conceptual knowledge and how the structures can be explicitly specified.

2.2.  Ontology with and within Chinese characters Notwithstanding the long-standing belief that core conceptual information is implicitly encoded in Chinese characters, the writing system in general has marginal relevance in formal linguistic theories, as it is a conventionalized secondary representation, and the issue of how writing systems encode linguistic elements has thus rarely been studied in semantics. In this section, I will walk readers through how the current studies of ontology can reveal the missing part of Chinese linguistics. First of all, we start with three assumptions: (i) Concepts and lexical meanings (i.e. word sense) are different, (ii) Character-morphemes are relatively objective cues to reflect conceptual organization, and (iii) Character-morpheme networks constitute the interface of concepts and word senses. Figure 12.1 schematizes a three-layered representation of our proposal (C.-R. Huang et al. 2008). At the middle layer (the character level), we take an ‘atomic globule’ network view (Aitchison 2003), where the characters – realized as core concept units – which share similar conceptual information, cluster together. Down to the word level, we take the ‘cobweb’ view, as words – built up from a pool of characters – are connected to each other through different semantic relations. It has been further argued that radicals 部首 (bu4 shou3) – the composite units of characters – represent conventionalized knowledge system, and offer an opportunity to explore cross-language ontologies (Wong and Pala 2002). The radicals, a.k.a. 意符 (yi4 fu2, ‘semantic symbols’) in the Chinese lexicological tradition have been considered as essential semantic components within characters. The most classic lexicographical work – 说文解字 (ShuoWenJieZi) compiled by Xu Shen (121 ce) – is organized according to the semantics of the 540 radical forms. Recent studies (Chou and Huang 2007) show that the class of characters sharing the same radicals can be linked to the qualia structure proposed by Pustejovsky (1998). The qualia structure, whose original idea can be traced back to Aristotle’s aitiae (‘modes of explanation’) (Physics 194 b17–20), can be viewed as a set of semantic constraints by which a lexical item is conceptually understood. Aristotle stated that there are four modes of explanation that drive our basic understanding of an object or relation in the world: (1) Material causality: what constitutes the object in question, (2) Formal causality: what distinguishes the object from others, (3) Efficient causality: how the object in question came about, and (4) Final causality: what purpose the object has. Pustejovsky (1998) transforms the ‘modes of explanation’ into one of the lexical interpretive levels called qualia structure with four qualia: (1) Formal quale: what distinguishes it from others, (2) Constitutive quale: what constitutes it, (3) Telic quale: what purpose it has, and (4) Agentive quale: how it comes about. In applying the theory to Chinese radical semantic knowledge, C.-R. Huang et al. (2007) analyzed the Bovid Domain Ontology Conventionalized by Radical ‘羊’ and found that the deriving concepts, i.e. characters shared with the radical, can be explained by the qualia, e.g. formal quale: ‘羳’ (fan2, ‘bovid which has yellow belly’) involves both the constitutive part of a bovid’s belly and its visual attribute; constitutive quale: ‘群’ (qun2, ‘a group of’), and agentive quale: ‘羜’ (zhu4, ‘lamb of five months old’), etc.9 206

Chinese linguistics: Semantics

CONCEPT

Character level (HanziNet)

Word level (WordNet)

Figure 12.1  The pyramid structure model

3.  Chinese lexical semantics As we have seen so far, the lexical meanings are complex. This complexity can be explained by tracing back to its ontological status at the morpheme (character) level, it has also been tackled largely at the lexeme (word) level. Lexical semantics is the study of word meaning, by representing lexical knowledge in terms of phenomena such as argument linking, selectional restrictions, lexical relations, etc. As introduced in Section 2, the linguistic status of word in Chinese has not been commonly recognized, it is worth noting that despite such controversy, the establishment of the concept of word as a relatively stable linguistic unit has its advantages in mirroring the conceptualization of language on the one hand, and helping determine the course of rigorous compositional linguistic analysis on the other hand. In this section, I will sketch some discussions of lexical semantics in Chinese, as well as an implementation of a lexicon database called WordNet, which can serve as a bridge for cross-language comparison. 207

Hsieh Shu-Kai

3.1.  Word meaning (semantic properties of words) The meaning of words, a.k.a. word sense, can be defined as the instantiation of lexicalized concepts. In the context of lexical semantics, the sense of a word can be scrutinized through the lens of its inherent features and meaning relations with other words, rather than by its denotational relation to things in the world. Many intricate lexical phenomena have been distinguished and discussed. Polysemy and Homonymy are two typical examples for the variants of word meanings. Polysemy refers to the phenomenon where a lexeme has at least two interrelated meanings. Considering the verb example 开 (kai1), which means ‘move something (door or windows) so as to leave a space allowing access’ (e.g. in 开门 ‘kai1 men2’); ‘make available’ (e.g. in 开放 ‘kai1 fang4’) or unfold or be unfolded (e.g. in 开展 ‘kai1 chang3’), which are closely related.10 In contrast, Homonymy is concerned with the phenomenon where two lexemes have unrelated meanings, but happen to share the same form. For example, though 打 (da3) can have multiple meanings, at least two of them are totally unrelated – ‘punch’ and ‘classifier for counting objects’ – thus the lemma 打 represents two homonymous lexemes which share the same lemma form. The polysemous behavior of words results in the difficulties of computational semantic processing of human languages. A subfield of computational semantics called Word Sense Disambiguation (WSD) has emerged to handle this issue. Based on a sense inventory (e.g. from WordNet11 or HowNet12), a few machine-learning techniques have been employed to automatically pick out the most likely sense for a certain word/all words in a text. However, the assumption in WSD tasks, that is, one-sense-per-word in a certain context, will encounter meaning shifts in various contexts. Asher (2011) describes co-predication as one of the sorts of predication in which two or more predicators select the same argument with distinct yet systematically related senses. So in (a), the modifier delicious selects the [content] sense, while took forever selects [event] sense of lunch. Similarly, in (b), 种 (‘zhong4’) and 很好喝 (‘hen2 hao3 he1’) both predicate the [plant] and [liquid] sense of 茶 (‘cha2’), respectively.13 (a) The lunch was delicious but took forever. (b)

埔里种的茶很好喝。 Pu4li3 zhong4 de cha2 hen2 hao3 he1. Puli plant DE tea very good drink ‘The tea grown in Puli tastes delicious’

It is worth noting that the co-predication, where lexical meanings allow co-existing multiple readings in a given sentence, is different from widely recognized lexical ambiguity, where multiple readings are allowed, but can and need to be resolved given more contextual information. As can be seen in (c), where the ambiguous word 看病 (‘kan4 bing4’ ) can be disambiguated given that fact that the reference status of 他 (‘ta1’, he), i.e. either as ‘doctor’ or ‘patient’, is determined. (c)

他正在看病。 ta1 zheng4 zai4 kan4 bing4 he PROG seeing sickness ‘He is seeing the doctor/He is examining the patient’ 208

Chinese linguistics: Semantics

3.2.  Semantic relations Semantic relations also constitute the core part of our lexical knowledge. Saussure ([1916] 1983) emphasized that these differences are of two kinds: paradigmatic (concerning substitution) and syntagmatic (concerning positioning). Paradigmatic relations hold between words of which the meaning is related in some systematic ways. Often they belong to the same syntactic category, as for example 便笺, 便条, 条子 (note, post-it note, notelet). A few paradigmatic relations that receive the most attention in lexical semantics are semantic paradigmatic ones between word senses, where they share many semantic properties, but fail to share others. For example, synonymy constitutes the relation among words having the same meaning (e.g. 从不 cong2 bu4 / 未曾 wei4 ceng2; ‘never’); hypernymy-hyponymy defines a ‘type-of ’ meaning (歌 ge1 / 民谣 min2 yao2; ‘song/folk song’); meronymy-holonymy for the ‘part-of ’ meaning (手 shou3 / 手掌 shou3 zhang3; ‘hand/palm’), etc. A set of substitution tests as well as logical conditions for each relation is also defined in Tsai et al. (2002) and Huang et al. (2004), which are meant to be working criteria that facilitate identification of relation instances and promote consistency of decisions among linguists. In contrast, syntagmatic relations are based on the co-occurrence of words within a sentence, like collocations. Typically, they hold between words of different syntactic categories such as nouns and verbs, e.g. the relations between [吃]–[水果] (chi1–shui2guo3; ‘eat–fruit’) in the sentence 他吃了一盘水果 (He ate a dish of fruit). There has been no available lexical resource for explicit specification of syntagmatic relations in Chinese; however, interested readers can refer to a corpus-based query system called Chinese Word Sketch Engine14 developed at Academia Sinica. Figure 12.2 shows the sketch results of words syntagmatically related to the target word 吃.

3.3.  Chinese WordNet In addition to the description of word senses, semantic relations are at the core of WordNetlike architecture, and constitute the essential and integral part of linguistic and conceptual knowledge formalization. A lexical knowledge base is a general repository of knowledge about lexicalized concepts and their relationships. WordNet (Fellbaum 1998) and FrameNet (Baker et al. 1998) represent the advances of these two kinds of semantic relations respectively. In

Home Concordance Word Sketch Thesaurus Sketch-Diff sinica freq = 4660 SentQbject of365 6.1 Object 2616 3.0 Modifier 1375 2.8 Subject 2080 2.4 pp 在 96 37.92 7 1 36.94 46 39.92 36 26.06 M 36.56

75 36.32

20 23.01

7 21.47

10 27.61

25 33.89

20 22.84

12 20.69

18 24.92

7 1 32.98

108 22.39

29 20.53

6 19.99

143 32.74

13 21.94

7 18.87

16 19.16

39 32.56

50 21.1

37 18.07

8 18.91

27 30.06

70 20.41

48 17.58

Figure 12.2  Syntagmatic relations from Chinese Word Sketch Engine

209

9

0.5

5

22.06

183 0.2 30 28.07 7 15.13 5 13.96 5 8.09

Hsieh Shu-Kai

particular, the WordNet-like framework pioneered by the Princeton WordNet, and greatly enriched by EuroWordNet (Vossen 1998) and further extended to Global WordNet Grid, has thus become the de-facto standard for a lexical knowledge base enriched with lexical semantic relations (LSRs).15,16,17 Modeled on English WordNet, the Chinese WordNet18 project aims to provide a bilingual mirror for lexical knowledge representation. Following Princeton’s English WordNet, Chinese WordNet (CWN) is structured as a hierarchy of synsets (sets of lexical units that are synonyms), and each synset is intended to represent lexicalized concept. These lexicalized concepts are connected with different kinds of lexical semantic relations (such as hypernymy, meronymy, antonymy, etc.) Each synset is also described by a lexicographic definition and examples. For instance, 好 has 15 senses distinguished in CWN, which are assigned to 15 different synsets. One of these synsets is the set containing the overlapping sense of {好, 不错}, etc. Figure 12.3 shows a snapshot of the system. As defined in English WordNet, different semantic and lexical relations have also been implemented as a core part in the CWN. It is noted that the CWN developer team has posted many interesting issues reflecting the specific properties of Chinese language. For example, concerning the superordinate relations (a.k.a. sister terms) in WordNet, which are used to define those coordinate words that have the same hypernym, Huang et al. (2007, 2015) observed that not all coordinate terms are equal when detailed lexical analysis is done for a set of coordinate terms sharing the same hypernym. For instance, the first intuition for the concept of 季节 (ji4 jie2, ‘season in a year’) will be four seasons – spring, summer, autumn, and winter. Therefore, the term spring is a hyponym with respect to season, while spring, summer, autumn and winter are co-hyponyms, i.e. hyponyms of the same superordinate term. However, other terms for seasons, such as 干季 (dry season) and 湿季 (rainy season) are not thought of intuitively as parallel as the four seasons, although all of them share the same superordinate concept, ‘seasons in a year’. So the default relation of co-hyponym in WordNet

C h in e s e w o r d n e t

(ABOUT)

(DOCS)

(DOWNLOADS)

(STATS)

(Enter a word, e.g. (Query)

15 l.

Figure 12.3  Chinese WordNet

210

(QUERY)

(LICENCE)

Chinese linguistics: Semantics

is too general to cover the detailed relations for its set of hyponyms. These coordinate terms could be re-classified into conceptually salient groups, and the conventional ways used to define the co-hyponyms, that is, antonymy and near-synonymy, seem inadequate to account for such conceptual classification. In addition to varied lexical semantic relations connecting different word senses and lemma, it has been also found that interesting relations also exist among their morphological behavior. This can be pointed out in three aspects in general: (i) inflectional variations of words (e.g. work, working and worked, which share everything except their tense form), (ii) derivational variations (e.g. work and worker), and (iii) regular compounds, which comprise the modifier and the head part (e.g. flower shop) (Murphy 2003; Löbner 2013). As mentioned, the vast majority of Chinese characters represent the morpheme, and the intricate ways of combining morphemes to form a word or compound in Chinese has raised much discussion in previous literature. In order to enrich the lexical semantic relations in CWN, Hsieh and Chang (2014) proposed a simple computational method to automatically acquire semantic relations relying on the in situ morpho-semantic structure in Chinese, and the system has achieved great results. This method assumes a model called Morpho-Semantic Linkage (MSL) in Chinese. Because of its strong semantic weight, in most cases of characters (i.e. a word-to-be morpheme or word-used-to-be morpheme), there must be some relation existing among them (A-AB, B-AB, AB-BC, etc), and this ‘to-be labeled’ morpho-semantic relation is defined as Morpho-Semantic Linkage in Chinese. LSRs among concepts (in terms of CWN) is thus only a subset of MSLs (recall the qualia relations mentioned). Figure 12.4 shows the scheme. So the proposed method enriches the CWN with relations leveraged by operationalizing MSL, i.e. based on the position and semantic role of (word-to-be) morphemes in modification construction. It is noted that it is not concerned with semantic relations between nominals, as often studied in the field of Information Extraction, i.e. the relations between N1 and N2 in N1N2 compounds. It aims to operationalize linguistic theories for the automatic labeling task of lexical relations on word-pairs. For N1N2 compounds, pairs are labeled with hypernymy–hyponymy, and pairs are labeled with meronymy–holonymy. In some ways the MSL space parallels the lexical fields theory (Löbner 2013: 215). The notion of lexical field comes from the observations that most lexical items form groups with other lexemes, the senses of these lexemes have something in common, and they are interrelated by definable meaning relations. For example, we can define lexemes sharing the same morpheme (thus the same radical component) as forming a lexical field. Written in

MSL space LSRs

Figure 12.4  Morpho-semantic linkage space

211

Hsieh Shu-Kai

set notation, it can be a set including a list of lexemes in brackets {.  .  .}, such as {车, 车辆, 车门, 车窗, 车厢, 汽车, 火车, 机车,  .  .  .}. The lexical semantic relations among the lexemes can be seen in its subsets: (1) hypernymy–hyponymy: {车–汽车}; {车–火车};  .  .  . (2) meronymy–holonymy: {车–车门}; {车–车窗};  .  .  . (3) (near-) synonymy: {车–车辆};  .  .  . (4) co-hyponymy: {车辆–汽车};  .  .  . Further exploration can be done using radical information as clues to determine the finegrained classification of the retrieved relations.

4. Summary The study of meaning has attracted a number of disciplines, including linguistics, psychology, cognitive science, philosophy, as well as many others. As the subfield of linguistics, contemporary semantics is characterized by various formal modeling, abstract representation, and empirical evidence that can be verified or falsified in a scientific manner. This chapter sketches core issues surrounding lexical meanings in Chinese, ranging from morpho-semantics to lexical semantics. As has been remarked, there are clearly several other interesting topics such as formal semantics and cognitive semantics that could each deserve a chapter-length full treatment, but I have had to leave them out and focus on the parts bearing on lexical aspects. Hopefully, it still provides a well-rounded picture of the endeavor in the study of Chinese semantics.

Notes   1 Interested readers are referred to Löbner (2013) for general introduction.   2 There have been few alternative views proposed. For example, Davis (2001) aims to reorient linguistics by reminding that the concept of ‘word’ does not pre-exist for ‘ordinary’ speakers.   3 Sproat and Shih (1996) only measured 76% agreement among human judges.   4 For instance, in real sentences like ‘有土斯有[財]’ and ‘國家[富]起來是件好事’, 財 and 富 are used as free morphemes.   5 There have been at least three ‘word segmentation standards’ proposed by different groups which reflect different views of Chinese wordhood: (1) ROCLING standard (Huang et al. 1997; CNS14366, 1999), (2) U-Penn standard (Xia 2000), (3) Peking University (GB/T13715, 1992).   6 Calculation is based on Sproat’s data available at http://rws.xoba.com/exercises/.   7 An example of a bound root from English could be the –ceive (in receive, conceive, etc.).   8 For a detailed review on this topic, please refer to Zhu (2004).   9 The qualia structure also enables the compositional interpretation within the compounds. See Lee et al. (2010). 10 More sense distinctions can be found at http://lope.linguistics.ntu.edu.tw/cwn2/query/ 11 http://wordnet.princeton.edu/. 12 http://www.keenage.com/. 13 Note that the pairings like [content]–[event] or [information]–[physical object] are only some instances of what is coined as systematic polysemy. Refer to Pustejovsky (1998) for more discussion. 14 http://wordsketch.ling.sinica.edu.tw/. 15 http://wordnet.princeton.edu/. 16 http://framenet.icsi.berkeley.edu/fndrupal/. 17 http://globalwordnet.org/global-wordnet-grid/. 18 http://lope.linguistics.ntu.edu.tw/cwn2. This is different from HowNet, a Chinese–English bilingual online lexicon where the inter-conceptual relations and inter-attribute relations of concepts are assumed. See http://www.keenage.com.

212

Chinese linguistics: Semantics

Bibliography Aitchison, Jean (2003) Words in the Mind: An Introduction to the Mental Lexicon, Oxford: Blackwell Publishing. Asher, Nicolas (2011) Lexical Meaning in Context: A Web of Words, Cambridge: Cambridge University Press. Baayen, Harald (1992) ‘On Frequency, Transparency and Productivity’n Yearbook of Morphology, 181ook. Baker, Collin, Charles J. Fillmore, and John B. Lowe (1998) ‘The Berkeley FrameNet Project’, in Proceedings of the COLING-ACL 1998, 86–90. Bierwish, Manfred (1970) ‘Semantics’, in J. Lyons (ed) New Horizons in Linguistics, Harmondsworth: Penguin, 166–84. Bloomfield, Leonard (1926) ‘A Set of Postulates for the Science of Language’, Language 2(3): 153–64. Chao, Yuen-Ren (1968) A Grammar of Spoken Chinese, Berkeley: University of California Press. Chou, Ya-Min and Huang Chu-Ren (2007) ‘Hantology: An Ontology based on Conventionalized Conceptualization’, in Proceedings of the Fourth Onto Lex Workshop, Jeju, Korea. CNS14366 (1999)《中文信息处理分词规范》(Word-breaking Standards in Chinese Information Processing), Taipei: The Central Standardization Office of the Ministry of Economy of Taiwan. Cruse, Alan (1986) Lexical Semantics, Cambridge: Cambridge University Press. Davis, Hayley (2001) Words: An Integrational Approach, Richmond: Curzon. Dong, Zhendong and Dong Qiang (2006) HowNet and the Computation of Meaning, Hackensack, NJ: World Scientific. Fellbaum, Christina (1998) WordNet: An Electronic Lexical Database, Cambridge MA: MIT Press. GB/T13715 (1992)《信息处理用现代汉语分词规范》(Modern Chinese Word-breaking Standards in Information Processing), Beijing: State Standardization Management Committee. Huang, Chu-Ren, Chang, Ru-Yng, and Lee, Shiang-Bin (2004) ‘Sinica BOW (Bilingual Ontological Wordnet): Integration of Bilingual WordNet and SUMO’, in Proceedings of The 4th International Conference on Language Resources and Evaluation (LREC2004), Lisbon, Portugal. Huang, Chu-Ren, K. J. Chen, and L. L. Chang (2007) ‘Segmentation Standard for Chinese Natural Language Processing’. CKIP Technical Report 96-01. Taipei: Academia Sinica. Huang, Chu-Ren and Hsieh Shu-Kai (2015) ‘Chinese Lexical Semantics’, in W. Wang and C. Sun (eds) The Oxford Handbook of Chinese Linguistics, Oxford: Oxford University Press. Huang, Chu-Ren, Su I-Li, Hsiao Pei-Yi, and Ke Xiu-Ling (2007) ‘Paranyms, Co-hyponyms and Antonyms: Representing Semantic Fields with Lexical Semantic Relations’, in Proceedings of Chinese Lexical Semantics Workshop, 20–23 May, Hong Kong Polytechnic University, Hong Kong. Huang, Chu-Ren, Elanna I. J. Tseng, Dylan B. S. Tsai, and Brian Murphy (2003) ‘Cross-lingual Portability of Semantic Relations: Bootstrapping Chinese WordNet with English WordNet Relations’, Language and Linguistics 4(3): 509–32. Huang, Chu-Ren, Yang Y.-J., and Chen S.-Y. (2008) ‘An Ontology of Chinese Radicals: Concept Derivation and Knowledge Representation Based on the Semantic Symbols of Four Hoofedmammals’, in Proceedings of the 22nd Pacific Asia Conference on Language, Information and Computation, Cebu City, Philippines. Huang, Chu-Ren 黄居仁, Zhou Yamin 周亚民, and Hsieh Shu-Kai 谢舒凯 (2008)《语言、文字与 知识架构: 由汉字出发的知识本体研究》(Language, Words, and Knowledge Structure: Studies on Knowledge Ontology Based on Chinese Characters), Taipei: Academia Sinica. Huang, James (1984) ‘Phrase structure, Lexical Integrity, and Chinese Compounds’, Journal of the Chinese Language Teachers Association 19: 53–78. Hsieh, Shu-Kai (2011) ‘Sense Structure in Cube: Lexical Semantic Representation in Chinese WordNet’, International Journal of Computer Processing of Languages 23(3): 243–53. Hsieh, Shu-Kai and Chang Yu-Yun (2014) ‘Leveraging Morpho-semantics for the Discovery of Relations in Chinese WordNet’, in Proceedings of 7th International Conference on Global WordNet, Tartu, Estonia. Jian, Y. and Pan H. (2005) Introduction to Formal Semantics, Beijing: Chinese Social Sciences Publisher (in Chinese). Lee, Chi-Yao, Chang Chia-Hao, Hsu Wei-Chieh, and Hsieh Shu-Kai (2010) ‘Qualia Modification in Noun-Noun Compounds: A Cross-Language Survey’, in Proceedings of the 22nd Conference on Computational Linguistics and Speech Processing (ROCLING 2010), Chi-Nan University, Taiwan, 379–90.

213

Hsieh Shu-Kai Lieber, Rochelle (2009) Introducing Morphology, Cambridge: Cambridge University Press. Löbner, Sebastian (2013) Understanding Semantics, London and New York: Routledge. Lu, Zhiwei 陆志韦 (1957)《汉语的构词法》(Word Formation in Chinese), Beijing: Science Press. Maienborn, C., K. von Heusinger, and P. Portner (eds) (2011) Semantics: An International Handbook of Natural Language Meaning, Vol. 1:3, Amsterdam: De Gruyter Mouton. Miller, George and Christina Fellbaum (2003) ‘Morphosemantic Links in WordNet’, Journal of Traitement automatique de langue 44(2): 69–80. Murphy, Lynn (2003) Semantic Relations and the Lexicon, Cambridge: Cambridge University Press. Murphy, M. L. (2010) Lexical Meaning, Cambridge: Cambridge University Press. Nishimoto, Eiji (2003) ‘Measuring and Comparing the Productivity of Mandarin Chinese Suffixes’, Computational Linguistics and Chinese Language Processing 8(1): 49–76. Packard, Jerome L. (2000) Morphology of Chinese: A Linguistic and Cognitive Approach, Cambridge: Cambridge University Press. Pustejovsky, James (1998) The Generative Lexicon, Cambridge, MA: MIT Press. Ruppenhofer, J., M. Ellsworth, M. R. L. Petruck, C. R. Johnson, and J. Scheffczyk (2010) ‘FrameNet II: Extended Theory and Practice’, (https://framenet2.icsi.berkeley.edu/docs/r1.5/book.pdf.) Saussure, Ferdinand (1916/1983) Course in General Linguistics, trans. Roy Harris, London: Duckworth. Sproat, Richard (2002) ‘Corpus-based Methods in Chinese Morphology’, Tutorial slide at COLING 2002, Taipei, Taiwan (http://www.cslu.ogi.edu/~sproatr/newindex/coling.pdf ). Sproat, Richard and Shih Chilin (1996) ‘A Corpus-based Analysis of Mandarin Nominal Root Compounds’, Journal of East Asian Linguistics 4(1): 1–23. Su, Xinchun 苏新春 (2008)《汉语词义学》(Chinese Semantics), Beijing: Foreign Language Teaching and Research Press. Tsai, Dylan Bo-Sheng, Huang Chu-Ren, Tseng Shu-Chuan, Lin Jen-Yi, Chen Keh-Jiann, Chuang Yuan-Hsun (2002)〈中文词义关系的定义与判定原则〉(‘The Definition and Defining Principles of the Meanings of Chinese Terms’), Journal of Chinese Information Processing 16(4): 21–31. Vossen, Piek (1998) EuroWordNet: A Multilingual Database with Lexical Semantic Networks, Dordrecht: Kluwer Academic Publishers. Wong, K. F., Li W., Xu R., and Zhang Z. (2009) Introduction to Chinese Natural Language Processing, San Rafael, CA: Morgan and Claypool. Wong, Shun-Hua and Karel Pala (2002) ‘Chinese Characters and Top Ontology in EuroWordnet’, in Singh, U. N. (ed) Proceedings of the First Global WordNet Conference, Mysore, India. Xia, Fei (2000) ‘The Segmentation Guidelines for the Penn Chinese Treebank (3.0)’, (http://www.cis. upenn.edu/~chinese/segguide.3rd.ch.pdf ). Yip, P. C. (2000) The Chinese Lexicon. A Comprehensive Survey, London and New York: Routledge. Zhu, Yan 朱彦 (2004)《汉语复合词语义构词法研究》(Morpho-semantic Studies of Chinese Compounds), Beijing: Peking University Press.

214

13 Chinese Morphology Jerome L. Packard University of Illinois, USA

Introduction The Chinese language has significant morphological processes, forming complex words from word components and using affixation processes as one way of forming those complex words.1 Complex word formation dominates in a discussion of Chinese morphology however, because the number of complex, multimorphemic words in the language is far greater than the number of non-complex single-morpheme words. While Chinese does have affixation processes, those properties play a role that is less frequent and visible than other ways of forming complex words. As a first step in discussing morphology in Chinese, it is worthwhile to consider where the language fits when compared to the rest of the world’s languages from the perspective of morphological typology, i.e. whether Chinese is an isolating (or analytical ), agglutinative or inflecting language (cf. Sapir 1921/1949).

Morphological typology In terms of morphological typology, Chinese is usually cited as being a language of the isolating or analytical variety. But the extent to which Chinese is considered a language of the isolating type depends upon how ‘isolating’ is defined (Packard 2009). If the term is used to mean whether the words of the language are mostly monomorphemic – then Chinese cannot really be considered very isolating, because most words in Chinese are multimorphemic rather than monomorphemic (when counted in terms of word ‘types’ versus ‘tokens’2). However, if by ‘isolating’ we mean whether the morphemes of Chinese are easy to isolate and identify, then Chinese must be considered quite isolating indeed. The second definition of isolating given above – whether the morphemes of a language are clearly identifiable – is defined by three criteria: (i) whether morpheme boundaries in the language are sharply defined, (ii) whether there is a single distinct morphemic identity represented within a defined morphemic space, and (iii) whether allomorphy is a significant feature of the language, i.e. whether morphemes in the language have multiple phonological forms. Addressing the first criterion, morpheme boundaries in Chinese are extremely well defined, perhaps more clearly defined than in any of the world’s languages. In reading a Chinese text, 215

Jerome L. Packard

it is a simple task to identify the morphemes because the orthographic unit in Chinese – the character – lines up with Chinese morphemes on a virtually perfect one-to-one basis. English, in contrast, tends to have a one-to-one relation between the word and the orthographic unit. The second criterion for deciding whether Chinese morphemes are easy to identify – i.e. whether Chinese usually has only a single morphemic identity represented within a defined morphemic space – is answered resoundingly in the affirmative, as Chinese has virtually no morphemic forms that contain more than one morphemic element. To offer a contrasting example, in Latin the accusative singular form of the word ‘girl’ puella (nom. sing.) is puellam, with puell– serving as the root meaning ‘girl’ and the suffix –am representing both accusative case and singular number. The Latin morpheme –am has two morphemic identities – accusative and singular – and there is no way to isolate the part of –am that stands for singular number and the part that stands for accusative case. The term ‘overlapping exponence’ is used in linguistics to describe morphemic forms that contain more than one morphemic element, and in Chinese, there is no overlapping exponence whatsoever. In Chinese, there are no forms that combine two types of morphemic information into one defined morphemic space. The third criterion used to rate whether the morphemes of a language are easy to identify is the presence of allomorphy. In Chinese, there is very little allomorphy, i.e. there are very few instances of one morpheme having multiple phonological identities. Even the nominal suffix 儿 –er, which may appear to have multiple phonological identities when it suffixes to different nouns, in fact has only one, i.e. [er]. The apparent phonological changes of –er that occur when it suffixes to different nouns in fact are completely predictable phonologically, and actually cause changes to the phonological form of the noun to which it suffixes, and not changes to the –er suffix itself. In fact, what may be the sole example in Mandarin of what could arguably be termed allomorphy is the alternation of the negative morphemes 不 bu and 没 mei in different contexts. The distributional facts are clear: if negation occurs either in the context of completed aspect or in conjunction with the verb 有 you ‘be/have’ that marks existence or possession, the alternate mei is used, and in all other contexts negation is marked by the alternate bu. Aside from this very clear example of mei/bu alternation, allomorphy does not appear to exist in Mandarin. Returning now to the discussion of the morphemic structure of Chinese words, Chinese words are mostly complex, i.e. multimorphemic, entities. If the words in a selected portion of Chinese text are categorized according to whether they consist of one morpheme or more than one morpheme, the majority will invariably be found to be multimorphemic. Having discussed the identity of Chinese in terms of morphological typology, let us now examine the structure and formation of Chinese words. But before considering the structure of words and how they are formed, criteria must first be provided to identify and isolate the entity that we are calling ‘word’, that is, we must define what a word is in Chinese.

Defining the word Despite much discussion and speculation to the contrary, the definition of word in Chinese is relatively straightforward: a word in Chinese is an entity that may independently occupy a syntactic part-of-speech slot. This syntactic definition of word is driven by syntactic analysis, and is only one of several possible definitions (semantic, morphological, phonological, etc.; see Packard 2000) that could be used. It is, however, an ideal choice for defining ‘word’ because but it has eminent theoretical justification, given that words may be defined using the syntactic criterion in virtually all of the world’s languages. 216

Chinese morphology

This definition of ‘word’ based on syntactic form class criteria has the advantage of being independent of morphological structure, i.e. there are no morphological criteria per se that influence our decision regarding what is and is not to be considered a word. The arbitrariness that sometimes arises is in defining form class in a way that has sharp boundaries – a problem that occurs in virtually all languages. So even though part of speech identities such as Noun, Verb and Adjective (Stative Verb) sometimes overlap, they nonetheless generally lend themselves to clear definition, as do Prepositions (Coverbs), Adverbs, and Conjunctions. Some instances of potential arbitrariness do arise, as happens in any language. So, for example, the sentence-final aspect marker le that indicates change-of-state may seem difficult to assign to a form class category, until we consider the fact that there does exist in Chinese a sentence-final slot that functions to provide its occupant with a modification scope over the entire sentence. So such sentence-final particles as 了 le ‘aspect’, 啊 a ‘exclamation’, 吧 ba ‘suggestion’, 吗 ma ‘question’, 哇 wa ‘incredulity’, 呢 ne ‘nominal question’, 诶 ei ‘suspicion’ and 呀 ya ‘surprise’, each carry specific meanings, and those meanings have scope over the entire utterance by virtue of the fact that they occupy the sentence-final functional slot. In viewing the sentence-final position in Chinese defined in this way, all of these particles may be considered words. This syntactic definition of ‘word’ safely accounts for most of the data. Even the components of idiomatic four-morpheme phrases are accounted for in a relatively straightforward manner. For example, in modern usage 子 zi ‘son, child’ is a bound morpheme, because it cannot stand alone as a noun. But in the four-morpheme idiomatic phrase 望子成龙 wang-zi-cheng-long hope-child-become-dragon ‘have great expectations for one’s child’, zi appears free because it is has the identity of a noun in that phrase, and wang appears free because it is has the identity of a verb. In fact, zi and wang must still be considered bound because while they function as free elements in the limited, classical syntax of a fourmorpheme idiomatic phrase, according to the rules of modern syntax they are still unable to serve as a free words. In order to function as free words in modern syntax wang and zi must be further augmented with morphemic material. In modern syntax, in the case of zi, the proper free word is 子女 zinü son-daughter ‘children’ and in the case of wang the proper free word is the two-syllable 希望 xiwang hope-hope ‘hope’. Up to this point we have mostly discussed only simple Chinese words – those that are composed of single free morphemes. But in fact most words in Chinese are complex – that is, formed by combining morphemes – and those complex words may be conveniently categorized according to the nature of the components that compose them. So let us move on to a discussion of word component morphemes in Chinese.

Word component morphemes Having defined word in Chinese, it is important to consider the properties of Chinese word components before we consider how these components fit together to form the different word types.

Free content morphemes A free content morpheme is a morpheme that contains content rather than functional (grammatical) information, and is free, i.e. it may stand alone in a syntactic slot as a X0-level lexical item. Single-morpheme words are legion and non-mysterious in Chinese: as examples 217

Jerome L. Packard

we have the nouns 火 huo ‘fire’, 水 shui ‘water’ and 树 shu ‘tree’; the main verbs 跑 pao ‘run’, 开 kai ‘open’ and 打 da ‘hit’; the auxiliary verbs ken 肯 ‘willing to’, xiang 想 ‘think about’, 会 hui ‘can, will’, 能 neng ‘can’ and 要 yao ‘want to’; the adjectives (also known as stative verbs) 快 kuai ‘fast’, 高 gao ‘high’ and 瘦 shou ‘thin’; and the prepositions (also known as coverbs) 跟 gen ‘with’, 替 ti ‘for’ and 用 yong ‘with’.

Free function morphemes Free function morphemes are what are commonly known as function words. Some examples are 的 de ‘nominalization marker’, 才 cai ‘then’, 就 jiu ‘then’, 都 dou ‘all’, 还 hai ‘still’, 只 zhi ‘only’ and 也 ye ‘also’.

Bound content morphemes Bound roots Bound roots are morphemes that, like free content morphemes, contain content rather than functional (grammatical) information, but unlike free content morphemes are bound, and so must be augmented with additional morphemic information before they can occupy a syntactic slot. Some examples of bound roots in Chinese are 脑 nao ‘brain’, 房 fang ‘house’, 橡 xiang ‘rubber’, 镜 jing ‘mirror’, 裤 ku ‘pants’, 森 sen ‘forest’, 足 zu ‘foot’, 虎 hu ‘tiger’, 羽 yu ‘feather’ and 衣 yi ‘clothing’. Note that, unlike in some other languages, most bound roots in Chinese are quite flexible regarding which position they may occupy in a two-syllable word. In English, for example, bound roots are Greek and Latinate word stems that tend to be either the first or second morpheme in a word, as seen in examples such as pseudo– ‘false’, anti– ‘against’, neuro– ‘nerve’, oculo– ‘eye’, bio– ‘life’, –itis ‘inflammation’, –cide ‘killing’, –cyte ‘cell’, –ectomy ‘removal of’, and –gnosis ‘knowledge’. In Chinese, in contrast, a given bound root often can be either the first morpheme or the second morpheme in a word. So, for example, the bound root 脑 nao ‘brain’ is the first morpheme in the word 脑袋 naodai brain-bag ‘brain’, and it is the second morpheme in the word 电脑 diannao electric-brain ‘computer’, and the bound root 衣 yi ‘clothing’ is the first morpheme in the word 衣服 yifu clothingclothing ‘clothing’, and it is the second morpheme in the word 上衣 shangyi top-clothing ‘overcoat’.

Bound function morphemes A bound function morpheme is a morpheme that, like a function word, contains functional rather than content information but is bound, and so must be attached to another morpheme before it can appear in a syntactic form class slot. Bound function morphemes are also known as affixes, and come in two varieties: those that are not related to grammatical processes are derivational (or word-forming) affixes, and those that tend to signal grammatical processes are grammatical (or inflectional) affixes. This derivational–inflectional affix category distinction will be familiar to practitioners of linguistic morphology because of its broad traditional application to word formation processes across languages. The characteristic of being related to grammatical processes is only one of a cluster of properties that serves to distinguish derivational and inflectional affixes in Chinese. These are presented in the following discussion of the two affix types. 218

Chinese morphology

Derivational (word-forming) affixes Derivational affixes are bound function morphemes that have the following properties: ● ● ● ● ●

They can change the form class of morphemes or words to which they apply They apply selectively to only certain members of a category They have a relatively variable and unpredictable meaning across contexts They may attach either to free words or to bound roots They may attach to the right or left of a word or word component, i.e. they may be either prefixes or suffixes. ● They are unrelated to processes traditionally thought of as being ‘grammatical’, such as the marking of number, aspect or mood. The following are examples of Chinese derivational prefixes: 无 wu– ‘not’, 未 wei– ‘not’, 非 fei– ‘not’, 第 di– ‘ordinalizer’, 复 fu– ‘again’, 可 ke– ‘may’ and 再 zai– ‘again’. The following examples of derivational suffixes are the nominalizers 子 –zi, 儿 –er, 头 –tou, 性 –xing, and 度 –du; the verbalizing suffix 化 –hua, the adverbial suffix 然 –ran, and the agentive suffix 者 –zhe. While the difference in Chinese between derivational affixes and inflectional affixes (to be discussed below) is relatively straightforward and clear, the difference in the language between bound roots and derivational affixes is less clear. The difference hinges upon the distinction between content and function morphemes, which in the field of linguistic morphology is traditionally considered to involve a continuous rather than discrete difference, in other words a difference in degree rather than a difference in kind. For example, let us contrast several bound roots that generally entail the meaning ‘one who______’ (i.e. 客 –ke, 奴 –nu, 员 –yuan, 手 –shou, 民 –min, 霸 –ba, 精 –jing), with the derivational affix 者 –zhe which also means ‘one who______’, but is considered a derivational affix rather than a bound root because it is a productive suffix that marks grammatical case, i.e. it indicates agency. First, –zhe is more productive in the way seen in functional rather than content elements, able to suffix to virtually any verb (and some non-verbs as well: 第三 disan ordinal-three ‘the third’ > 第三者 disanzhe ordinal-three-AFF ‘the third party (in a relationship)’; 前 qian front ‘front’ > 前者 qianzhe front-AFF ‘the one in front’). The cited bound roots on the other hand generally co-occur with a more limited number of morphemes and so are more restricted in their productivity. Second, –zhe is usually restricted to simply mean ‘one who______’ (读者 duzhe read-one ‘reader’, 患者 huanzhe sick-one ‘patient’, 老者 laozhe old-one ‘one who is old’, and 编者 bianzhe edit-one ‘editor’). However, the bound roots (客 –ke, 奴 –nu, 员 –yuan, 手 –shou, 民 –min, 霸 –ba and 精 –jing) entail the meaning ‘one who______’, but also include an additional element of meaning, such as one who is adept at something, one who does something as a profession, or one who has achieved notoriety for doing something. As examples we have ‘one who (is adept at) ______’ (手 –shou: 射手 sheshou shoot-one ‘marksman’, 能手 nengshou able-one ‘expert’, 高手 gaoshou high-one ‘expert’, 水手 shuishou water-one ‘sailor’, and 鼓手 gushou drum-one ‘drummer’); ‘one who ______ (as a profession)’ (员 –yuan: 演员 yanyuan act-one ‘actor’, 教员 jiaoyuan teach-one ‘teacher’; 民 –min: 农民 nongmin farm-one ‘farmer’, 股民 gumin stock-one ‘investor’); ‘one who (is notorious for) ______’ (客 –ke: 政客 zhengke politics-one ‘political hack’, 刺客 cike stab-one ‘assassin’, 说客 shuoke speak-one ‘persuader’); ‘one who (is stuck doing, or is a “slave” to) ______’ (奴 –nu: 车奴 chenu car-one ‘one who is stuck making car payments’, 房奴 fangnu 219

Jerome L. Packard

house-one ‘one who is stuck making mortgage payments’); ‘one who (constantly does or is very good at) ______’ (霸 –ba: 学霸 xueba study-one ‘one who is very good at studying’, 面霸 mianba face-one ‘one who frequently goes on job interviews’, 投霸 touba submit-one ‘one who is a frequent job applicant’); and ‘one who (is infamous for) ______’ (精 –jing: 害人精 hairenjing harm-person-one ‘one who always harms others’, 马屁精 maipijing horserear-one ‘flatterer’ and 白骨精 baigujing white-bone-one ‘female who is successful, educated and wealthy’).

Inflectional (grammatical) affixes Inflectional affixes are bound function morphemes, like derivational affixes, but they have the following properties that distinguish them from the latter: ● They do not change the form class of morphemes or words to which they affix ● They do not apply selectively; that is, they apply more broadly to most members of a given category ● They have a relatively constant meaning across contexts ● They affix only to free words ● They affix only to the right edge of a word (or word component), i.e. in Chinese they may only be suffixes. ● They are related to processes considered to be ‘grammatical’, such as the marking of grammatical number, aspect or mood. The following are examples of inflectional suffixes in Chinese: 了 –le ‘completed aspect’, 过 –guo ‘experienced aspect’, 着 –zhe ‘continuing aspect’, 得/不 –de/–bu ‘irrealis potential’, 们 –men ‘human plural’, 得了/不了–deliao/–buliao ‘irrealis (in)ability’, and 得 –de ‘extent complementation’. These suffixes are general because they apply to virtually all members of their designated categories, they attach as suffixes to free words only, and they mark grammatical operations that contribute a relatively constant meaning.

Word types and morphological processes There are several word types in Chinese, composed using a variety of morphological processes. But as it turns out, listing the different word types in Chinese ends up being largely an exercise in taxonomy, because in Chinese a word’s type has little effect on its status as a word and on how it functions within the grammar. The fact that a word is what minimally occupies a form class slot is true whether it is simple or complex and regardless of the elements from which it is composed. So let us first examine the two categories of word, simple and complex, in Chinese, and the different morphological processes used to compose them.

Simple words A simple word is a word composed of a single morpheme. Since morphemes and syllables – as well as the orthographic units characters – are virtually coextensive in Chinese, simple words are usually single-syllable, monomorphemic one-character elements. Simple words can be nouns, main verbs, auxiliary verbs, adjectives, prepositions, or the various function words. 220

Chinese morphology

Complex words A complex word is a word composed of more than one morpheme. Complex words come in various forms, depending on the process used to form them and what the nature of their components is.

Compounding Compounding is the joining together of two preexisting words to make a new word, which is termed a compound. Here are some examples: 火山 huoshan fire-mountain ‘volcano’, 打破 dapo hit-break ‘break’, 冰河 binghe ice-river ‘glacier’, 马熊 maxiong horse-bear ‘brown bear’, 猫熊 maoxiong cat-bear ‘panda’, 冰山 bingshan ice-mountain ‘iceberg’, 马路 malu horse-road ‘street’, 走进 zoujin walk-in ‘walk in’, 进去 jinqu in-go ‘go in’, 动作 dongzuo move-do ‘activity’, 买办 manban buy-manage ‘comprador’, 买卖 maimai buy-sell, ‘business’, 操守 caoshou operate-preserve ‘integrity’, 传动 chuandong transmit-move ‘a (vehicle’s) transmission’. Chinese is erroneously thought to be a language in which the majority of complex words are compounds – in fact that distinction belongs to bound root words created via composition (see below).

Composition Composition is the joining of two word components, at least one of which is a bound root, to make a word. Therefore the type of word formed from the process of composition is called a bound root word. Bound root is the most common morpheme type in Chinese, and so it is not surprising that composition is the most common word-formation process in Chinese, and that the most common word type in Chinese is the bound root word.

Affixation Affixation involves the attachment of an affix. The two types of affix that exist in Chinese (see section 4.4 above) are derivational (or word-forming) affix and inflectional (or grammatical ) affix. Therefore, the two types of affixation that exist in Chinese are derivation and inflection, which yield the two types of affixed words derived words and grammatical words.

A.  Derived words Derived words are words formed by the addition of a word-forming affix to the left (prefix) or right boundary (suffix) of a word. The following examples include both prefixation and suffixation. 阿混 ahun AFF-muddle ‘muddle-head’, 插头 chatou insert-AFF ‘plug’ 电化 dianhua electricity-AFF ‘electrify’, 调子 diaozi tune-AFF ‘tune’, 房子 fangzi house-AFF ‘house’ 孵化 fuhua hatch-AFF ‘incubate’, 复查 fucha AFF-investigate ‘reinvestigate’, 复发 fufa AFF-occur ‘relapse’, 复活 fuhuo AFF-live ‘resurrect’, 复习 fuxi AFF-study ‘review’, 复现 fuxian AFF-appear ‘reappear’, 钢化 ganghua steel-AFF ‘temper’, 感化 ganhua feelAFF ‘sensitize’, 骨化 guhua bone-AFF ‘ossify’, 归化 guihua return-AFF ‘naturalize’, 开化 kaihua open-AFF ‘civilize’, 可体 keti AFF-body ‘fit’, 欧化 ouhua Europe-AFF ‘Europeanize’, 皮儿 pir skin-AFF ‘skin’, 汽化 qihua vapor-AFF ‘vaporize’, 劝化 quanhua persuade-AFF ‘convert’, 烧化 shaohua burn-AFF ‘burn’, 石化 shihua stone-AFF ‘petrify’, 退化 tuihua 221

Jerome L. Packard

retreat-AFF ‘degenerate’, 无视 wushi AFF-see ‘disregard’, 氧化 yanghua oxygen-AFF ‘oxidize’, 烟子 yanzi smoke-AFF ‘soot’, 羽化 yuhua feather-AFF ‘(of an adult insect) emerge’.

B. Grammatical words Grammatical (or inflected) words are words that are formed by the affixation of an inflectional suffix. Most grammatical suffixes in Chinese are considered word-external inflection because they attach to the right boundary of a simple or complex word. The 得/不 –de/–bu marker of ‘irrealis potential’ on the other hand is considered word-internal inflection because it attaches inside a word, on the right side of the first syllable (i.e. the head) of a compound result verb. Word-external inflection  Word-external inflection suffixes an inflectional affix to the right boundary of a simple or complex word. Some word-external inflectional affixes are: 了 –le ‘completed aspect’, 过 –guo ‘experienced aspect’, 着 –zhe ‘continuing aspect’, 们 –men ‘human plural’ and 得 –de ‘extent complementizer’. What these inflections have in common is that they all attach to the right edge of a word. To provide examples, the aspect marker 了 –le suffixes to the verb 吃 chi ‘eat’ to produce 吃了 chile ‘have eaten’; the marker of experience aspect 过 –guo suffixes to the verb 看 kan ‘see’ to produce 看过 kanguo ‘have seen before’; the marker of continuing aspect 着 –zhe suffixes to the verb 走 zou ‘walk’ to produce 走着 zouzhe ‘walking’. The grammatical marker of human plural 们 –men suffixes to the second-person pronoun 你 ni to produce 你们 nimen ‘you (pl)’. The 得 –de affix marking extent complementation suffixes to a verb, and serves to introduce a phrasal complement that indicates a measure of the extent of application of the verb to which it suffixes. So, for example, the active verb 走 zou ‘walk’ that takes the suffix –de yielding 走得 zoude may then be followed by an extent complement phrase such as 很快 hen kuai ‘very fast’ to give 走得很快 zoude hen kuai ‘walk very fast’. There is similarity between these ‘phrasal’ resultatives marked by a verb inflection and ‘lexical’ resultatives formed by word-internal inflection that is more apparent than real. This is discussed in the following section. Word-internal inflection In word-internal inflection, the ‘irrealis potential’ inflection markers 得 –de and 不 –bu suffix (some would say infix) internal to the word, to the right side of the first syllable (head) of a compound result verb, signaling the ability or inability of the potential result indicated by the verb to actually take place. So, for example, the compound result verb 找到 zhaodao seek-attain ‘find’ may take the –de potential inflection to yield 找得到 zhaodedao seek-AFF-attain ‘able to find’, or it may take the –bu potential inflection to yield 找不到 zhaobudao seek-AFF-attain ‘unable to find’. As mentioned above, ‘phrasal’ resultatives formed by a word-external verb inflection and ‘lexical’ resultatives formed by word-internal inflection may appear similar and in some cases may be phonologically and orthographically identical. They are, however, entirely distinct in terms of both their semantics and the process of their formation. For example, the simple verb 做 zuo means ‘do’, and may be suffixed with an extent complement (such as 很快 hen kuai ‘very fast’) as can any verb, as in 做得很快 zuode hen kuai do-AFF:extent-very fast ‘do quickly’. The verb zuo may also be suffixed with the extent complementizer 得 –de and the simple extent complement 好 hao ‘good, well’ to yield 做得好 zuodehao ‘do well’. 222

Chinese morphology

There also independently exists the compound result verb 做好 zuohao do-good ‘do well’ which, as a compound result verb, may affix the ‘irrealis potential’ inflection marker 得 – de to the right side of its first syllable to yield 做得好 zuodehao ‘able to do well’. Now, even though these two three-morpheme strings (one an internally-inflected word, and one the simple verb zuo inflected with the extent inflection –de followed by a complement of extent hao) that have the different meanings ‘do well’ and ‘able to do well’ are phonologically and orthographically identical (i.e. 做得好 zuodehao), they are differently derived and semantically distinct. This is best demonstrated by how the two different forms accommodate negation. The negative form of the lexical potential resultative 做得好 zuodehao is 做不好 zuobuhao, while the negative form of the phrasal extent resultative 做得好 is 做得不好 zuode bu hao. The meanings of the two forms are clearly different, with the negated phrasal extent resultative meaning ‘do poorly’ and the negated lexical resultative meaning ‘unable to do well’.

Reduplication Reduplication is the creation of a new word by copying all or part of another word. The word type formed from this process is called a reduplication, and the resulting meanings include diminution, superlative, intensification, enhanced description or adverbial modification. Verbs, adjectives, nouns and classifiers all may be reduplicated to form new words. Both simple and complex words may be reduplicated, and when it is complex words that are reduplicated, different parts of the word are duplicated depending on the part of speech and the meaning of the duplicated word.

A. Single-syllable reduplication When it is a simple verb that is reduplicated, the meaning is generally one of limitation or diminution of verb action, as in 走 zou walk ‘walk’ > 走走 zouzou walk-walk ‘walk a bit’, 看 kan look ‘look’ > 看看 kankan look-look ‘take a look’ and 吃 chi eat ‘eat’ > 吃吃 chichi eat-eat ‘eat a little’. Such simple verb reduplication often inserts the morpheme 一 yi ‘one’ between the reduplicated elements, as in e.g. 走 zou walk ‘walk’ > 走一走 zouyizou walkone-walk ‘walk a bit’. When it is a simple noun or classifier that is reduplicated, the meaning is one of pro­ liferation or enumeration, for example 人 ren person ‘person’ > 人人 renren person-person ‘every person’, 天 tian day ‘day’ > 天天 tiantian day-day ‘every day’ or 本 –ben one:volume ‘a volume’ > 本本 benben volume-volume ‘every volume’ or 个 –ge one:piece ‘a piece’ > 个个 gege piece-piece ‘every piece’. When it is an adjective (stative verb) that is reduplicated, it has the effect of intensification or manner adverbialization. When the reduplicated adjective indicates intensification, the nominal modifier 的 de usually occurs before the modified element, for example, 红 hong red ‘red’ > 红红的 honghongde red-red-MOD ‘really red’ and 高 gao high ‘high’ > 高高的 gaogaode high-high-MOD ‘really high’. When the reduplicated adjective indicates manner adverbialization, the adverbial suffix 地 –de is usually added to the reduplicated element, as in 快 kuai fast ‘fast’ > 快快地 kuaikuaide fast-fast-AFF ‘quickly’ or 好 hao good ‘good’ > 好好地 haohaode good-good-AFF ‘well’. In manner adverbial reduplication (especially in the Beijing dialect), the tone on the reduplicated syllable changes to high-level and a phonological [r] sound is added to the reduplicated syllable, as in 慢 màn slow ‘slow’ > 慢慢儿地 mànmānrde fast-fast-r-AFF ‘slowly’. 223

Jerome L. Packard

B. Bi-syllabic reduplication Stative verbs  Two-syllable stative verbs accomplish reduplication by first duplicating the first, then duplicating the second syllable in the form AB > AABB. The function of reduplication in these words is for the purpose of adverbial modification and intensification. For example, 高兴 gaoxing ‘happy’ > 高高兴兴 gaogaoxingxing ‘very happily’, 快活 kuaihuo ‘cheerful’ > 快快活活 kuaikuaihuohuo ‘very cheerfully’ and 无聊 wuliao ‘bored, boring’ > 无无聊聊 wuwuliaoliao ‘in a very boring manner’. Active verbs  Two-syllable active verbs accomplish reduplication by duplicating the entire two-syllable form as one unit, in the form AB > ABAB. The function of reduplication in these words is for diminution. For example, 研究 yanjiu ‘research’ > 研究研究 yanjiuyanjiu ‘research a bit’, 讨论 taolun ‘discuss’ 讨论讨论 > taoluntaolun ‘discuss a bit’, 商量 shangliang ‘consult’ > 商量商量 shangliangshangliang ‘consult a bit’.

Loan words Chinese has many loan words that are either phonetic loans or semantic loans (calques), or a combination of the two.

A. Phonetic loans The following are examples of phonetic loan words: 三明治 sanmingzhi ‘sandwich’, 保龄 baoling ‘bowling’, 克隆 kelong ‘clone’, 博客 boke ‘blog’, 卡 ka ‘card’, 卡通 katong ‘cartoon’, 可乐 kele ‘cola’, 吉他 jita ‘guitar’, 咖啡 kafei ‘coffee’, 土司 tusi ‘toast’, 坦克 tanke ‘tank’, 基因 jiyin ‘gene’, 尼龙 nilong ‘nylon’, 巧克力 qiaokeli ‘chocolate’, 布丁 buding ‘pudding’, 幽默 youmo ‘humor, funny’, 引擎 yinjing ‘engine’, 拷贝 kaobei ‘copy’, 朋克 pengke ‘punk’, 柠檬 ningmeng ‘lemon’, 沙发 shafa ‘sofa, upholstered chair’, 派 pai ‘pie’, 浪漫 langman ‘romantic’, 汉堡 hanbao ‘hamburger’, 爵士 jueshi ‘jazz’, 白兰地 bailandi ‘brandy’, 的士 dishi ‘taxi’, 维他命 weitamin ‘vitamin’, 荷尔蒙 heermeng ‘hormone’, 苏打 suda ‘soda’, 酷 ku ‘cool’, 雪茄 xuejia ‘cigar’, 雷达 leida ‘radar’, 马达 mada ‘motor’, 高尔夫 gaoerfu ‘golf ’, 麦克风 maikefeng ‘microphone’.

B. Semantic loans (calques) The following are examples of semantic loan words (calques): 黑板 heiban black-board ‘blackboard’, 电话 dianhua electric-talk ‘telephone’, 电视 dianshi electric-view ‘television’, 电脑 dianhua electric-brain ‘computer’, 键盘 jianpan key-plate ‘keyboard’, 软体 ruanti soft-body ‘software’, 蜜月 miyue honey-moon ‘honeymoon’, 硬体 yingti hard-body ‘(computer) hardware’, 硬盘 yingdie hard-saucer ‘hard drive’, 牛仔裤 niuzaiku cow-boy-pants ‘jeans’, 热狗 regou hot-dog ‘hotdog’, 摩天楼 motianlou brush-sky-building ‘skyscraper’, 忙音 mangyin busy-sound ‘busy signal’.

C. Phonetic-semantic loans (phonetic calques) The following are examples of words borrowed from other languages that are phonetically similar to the borrowed word and also bear a semantic resemblance to the borrowed word in its translated form: 播客 boke broadcast-guest ‘podcast’, 黑客 heike black-guest ‘hacker’, 224

Chinese morphology

拖拉机 tuolaji pull-pull-machine ‘tractor’, 博蒂 bodi abundant-fruit ‘birdie (golf )’, 脱口秀 tuokouxiu pull-mouth-display ‘talk show’, 利基 liji advantage-base ‘niche (market)’, 冰淇淋 bingqilin ice-cream-shower ‘ice cream’, 俱乐部 julebu utensil-happy-place ‘club’, 信用卡 xinyongka trust-use-card ‘credit card’.

Abbreviation An abbreviation is a shortened version of a longer word or phrase, with acronyms being a sub-type of abbreviation. The following are examples of abbreviation in Mandarin: 北大 beida north-big < 北京大学 Beijing Daxue north-capital big-study ‘Peking University’, 劳保 laobao work-protect < 劳动保险 laodong baoxian work-move protect-risk ‘labor insurance’, 文革 wenge culture-change < 文化大革命 wenhua da geming culture-script big change-mandate ‘Cultural Revolution’, 环保 huanbao surround-protect < 环境保护 huanjing baohu surround-border protect-protect ‘environmental protection’, 校车 xiaoche schoolvehicle < 学校公车 xuexiao gongche study-school public-vehicle ‘school bus’, 车险 chexian vehicle-risk < 汽车保险 qiche baoxian steam-vehicle protect-risk ‘car insurance’, 电大 dianda electric-big < 广播电视大学 guangbo dianshi daxue wide-broadcast electric-view big-study ‘TV University’. Abbreviations are not to be confused with contractions, which are phonologically shortened versions of spoken forms of words created by omission (and/or modification) of the individual sounds of words or phrases. For example, 别 bie ‘do not’ is a contraction of 不要 buyao not-want ‘do not’, 甭 beng ‘need not’ is a contraction of 不用 buyong not-use ‘need not’, and 孬 nao ‘bad’ is a contraction of 不好 buhao not-good ‘bad’. An acronym is an abbreviation that is formed by retaining only the initial morphemes of a word or phrase, for example 工农兵 gong-nong-bing work-farm-soldier < 工人, 农民, 兵员 gongren, nongmin, bingyuan work-person farm-person soldier-person ‘workers, peasants and soldiers’, 桌椅凳 zhuo-yi-deng table-chair-stool < 桌子, 椅子, 凳子 zhuozi, yizi, dengzi table-AFF chair-AFF stool-AFF ‘tables, chairs and stools’, 欧非美 Ou-Fei-Mei EuropeAfrica-America < 欧洲, 非洲, 美洲 Ouzhou, Feizhou, Meizhou Europe-continent Africacontinent, America-continent ‘Europe, Africa and America’.

Neologism A neologism is a new word in the language that is not a loan and not an abbreviation. The concept of ‘new word’ is always relative to a given time period – generally the neologisms cited below have been coined within the past 30 years: 八卦 bagua eight-symbol ‘rumored/ gossipy’, 富二代 fuerdai rich-two-generation ‘second-generation rich kids’, 飙车 biaoche storm-car ‘drag racing’, 爆料 baoliao explode-material ‘tip (news)’, 粗口 cukou crude-mouth ‘curse’, 发飚 fabiao emit-storm ‘have a fit, lose one’s temper’, 狗仔队 gouzaidui dogson-team ‘paparazzi’, 过山车 guoshanche pass-mountain-vehicle ‘roller coaster’, 挤提 jiti squeeze-take ‘run (on a bank)’, 闭卷 bijuan close-scroll ‘closed-book exam’, 冰毒 bingdu ice-poison ‘crystal meth’, 吊球 diaoqiu drop-ball ‘drop shot’, 火锅 huoguo fire-pot ‘slam-dunk’, 假唱 jiachang fake-sing ‘lip-synch’, 街舞 jiewu street-dance ‘hip-hop’.

Notes 1 I thank Yun Yao for help with several of the examples; I alone, however, am responsible for any errors of interpretation.

225

Jerome L. Packard 2 In terms of word ‘types’ versus ‘tokens’, meaning that if words are counted in a given page or paragraph, the total number of single-morpheme words may come close to the number of multi­ morphemic words, because many monomorphemic words (e.g. the modification marker de) have a very high frequency, a fact which serves to inflate the number of monomorphemic words if counted in terms of token frequency.

References Packard, Jerome L. (2000) The Morphology of Chinese, New York: Cambridge University Press. Packard, Jerome L. (2009) ‘Chinese as an Isolating Language’, in K. Brown and S. Ogilvey (eds) Concise Encyclopedia of Languages of the World, Oxford: Elsevier, 220–3. Sapir, Edward (1921/1949) Language, New York: Harcourt Brace Jovanovich (copyright renewed 1949).

226

14 Chinese neologisms: Word-formation strategies in Chinese Antonella Ceccagno University of Bologna, Italy

Introduction This paper presents some central word-formation strategies in Chinese by mainly focusing on compound words and word-formation patterns emerging as a result of the central role played by compounding in Chinese. In fact, while in some languages compounding is a peripheral word-formation strategy, in Chinese it is the most productive means of word formation. This emerges clearly from the corpus of neologisms developed over the last 30 years included in The Contemporary Chinese Dictionary (CCD) (2002) where out of 709 new words (those with no more than two syllables) almost 95% (672) are compounds, while only a little more than 2% (16) are derived words. Packard (1998) points out that a central change in the Chinese language has been the shift from monosyllabic to disyllabic words that started during the Zhou dynasty (1122–256 bc). The growing disyllabic nature of Chinese has engendered new processes of word formation, among which compounding stands as the most salient one. It has been shown that 70–80% of Chinese words are compounds (Zhou et al. 1999; Xing 2006). Compounding is a much discussed topic in linguistics as reflected in the publication of new works focusing on the questions posed by compounds in different languages (see for example the books edited by Lieber and Stekauer 2009; and by Scalise and Vogel 2010). Compounds can be described as ‘words that are composed of more than one meaningful element’ (Plag 2003). If the variety of constituents is taken into consideration, a compound can be described as a word formed by at least two elements – including phrases, roots, words, and stems. The formation of compound words in Chinese has been discussed recently by Huang (1998), Packard (2000), Ceccagno and Scalise (2006, 2007), Ceccagno (2007), and Ceccagno and Basciano (2007, 2009a, 2009b). After discussing the existing definitions for Chinese compounds, Ceccagno and Basciano (2009a: 67) argue that a compound word in Chinese is formed by at least two morphological words. With this definition, also words including bound roots – which are not syntactic words but are morphological words – are included as constituents of compound words. In fact, in Chinese the only difference between syntactic 227

Antonella Ceccagno

words and bound roots lies in the fact that bound roots cannot be used independently in a syntactic structure. The word-formation processes in Chinese discussed in this paper are (i) metacompounding, or compound words whose constituents are truncated forms of underlying compounds; (ii) neologisms coined from previous constituents of metacompounds; (iii) re-analysis of syllables from polysyllabic words, loanwords, calques and hybrids as morphemes, and the resulting new morphemes; (iv) monosyllabic neologisms coined as loanwords; (v) neologisms formed through the re-analysis of the semantics of the constituents of existing compound words, and (vi) neologisms coined on the Internet that intentionally aim at breaking the prevailing word-formation strategies.

Methodology Traditional linguistics used to analyze established words, which however often display various kinds of idiosyncrasies and therefore are not indicative of the speakers’ knowledge of the structure of possible words. This approach has also been adopted in the analysis of word formation in Chinese, where traditional forms were analyzed along with new ones (Huang 1998; Packard 2000; Ceccagno and Scalise 2006). Recently, however, linguists have highlighted the importance of focusing on neologisms in the analysis of the morphological competence of today’s speakers. Therefore neologisms are now more widely used as crucial data (Plag 2003 among others) because they also help to single out patterns in word formation in a given historical period. By focusing their research on Chinese neologisms, Ceccagno and Basciano (2007, 2009a, 2009b) have been able to unveil some outstanding word-formation patterns and to develop a new theoretical approach to the study of Chinese compound words. Accordingly, this chapter addresses some central word-formation patterns by focusing on neologisms. To do so it makes use of two different sets of data. The first source consists of 709 disyllabic new words presented in the CCD (2002) new words section. The dictionary does not provide information on when the new words were formed. However, some of the listed neologisms were already formed in the early 1980s, while most of them seem to have been coined afterwards. The second source is a list of 2,255 new words found on the website Shanghai Daily’s Well-Acclaimed Buzzwords (WABs) in 2013. Language change takes place through innovation and propagation. The new words from WABs (2013) are in the stage of innovation, that is their usage is still recent and therefore still subject to abandonment by the speaking community. In fact, new words or loans do not necessarily last for long. They can be abandoned and replaced by other neologisms. The success of the semantic innovation in fact largely depends on the social dignity of the newly coined word (for a sociolinguistic analysis, see Labov 1972 and Croft 2000). Instead, the new words listed in CCD (2002), exactly because they are listed in a dictionary, are in the stage of propagation – they are used by different groups of the linguistic community (on innovation and propagation, see Croft 2000).

Abbreviations The basic phonological unit – or ‘the standard foot’ in Feng’s terminology (1998) – in Modern Chinese is made up of two syllables. Given that in Classical Chinese lexical morphemes are mainly monosyllabic, the prosodic requirements are mainly met through the creation of disyllabic compound words1 (for a discussion see Feng 1998; Arcodia 2007). 228

Chinese neologisms: Word-formation strategies in Chinese

Abbreviation is quite a widespread pattern for coining new lexical items in Chinese. This is a neologism-forming strategy linked to the prosodic requirements of the language that also has morphologic and semantic implications (Ling 2000; Wang 2001; Yuan 2002; Ceccagno and Basciano 2009a; Basciano and Ceccagno 2009). The most common process of abbreviation of nouns occurs when two disyllabic words are abbreviated into one, with one syllable selected from the first word and one syllable from the second. When possible, abbreviations seek to maintain semantic transparency. To this end, in the abbreviation process, the selected lexical morphemes are those with the same meaning of the compound word of which they are the truncated form, or that are semantically close to it. Wang (2001) argues that the morphemes chosen for the abbreviated neologisms are those that express the core meaning, while Ceccagno and Basciano (2009a) define them as those that are ‘semantically relevant’. An example that illustrates the choice of semantically relevant constituents for the abbreviated compounds is 影评 yǐngpíng, the abbreviation for 电影评论 diànyǐng pínglùn ‘cinema + comment = film review’, (see Yuan 2002; XHSC 2002). The compound 评论 pínglùn and the morpheme 评 píng chosen as one of the two constituents for the abbreviated compound word share the meaning of ‘comment’, ‘to comment on’, while 论 lùn, which has been eliminated, means ‘to discuss’, ‘opinion’, ‘theory’. These meanings are less relevant from the semantic point of view. The same mechanism is evident in the choice of the morpheme 影 yǐng as the first constituent of the abbreviation. In fact, 影 yǐng shares with 电影 diànyǐng the meaning of ‘cinema’, while 电 diàn, meaning ‘electric’, ‘telegram’, ‘to send a telegram’ is not semantically relevant and therefore is not included in the abbreviation. Abbreviation according to semantic relevance is a word-forming strategy that, to a certain extent, taps into the diachronic evolution of the Chinese language and its constant tension between recursion and abbreviation (for a discussion, see Ceccagno and Basciano 2009a). In fact, the diachronic shift of Chinese towards a preference for disyllabic words instead of monosyllabic words has brought about the creation of new compounds with the same meaning of pre-existing monosyllabic words. This has created a dynamic relationship of cross-reference between monosyllabic lexical morphemes and compound words that makes it possible to abbreviate existing words by going back to the morphemes with the same meaning. Coordinate compounds2 whose constituents are in a relation of synonymy, such as, for example, 美丽 měilì ‘beautiful + beautiful = beautiful’ are the best examples of the crossreference between monosyllabic words and compounds. Therefore it could be argued that this subcategory, more than others, lends itself to the formation of new abbreviated compounds according to semantic relevance.

Metacompounds Only in a limited number of abbreviations by means of semantic relevance, however, is semantics entirely transparent. In many other cases, one or all the constituents of the new compound word, despite being selected as the most semantically relevant, do not have the exact meaning of the compound word of which they are an abbreviated form. This is the case of the example in (1). (1)  报纸摘要 bàozhǐ zhāiyào ‘newspaper + to make a summary = newspaper clippings’, abbreviation 报摘 bàozhāi 229

Antonella Ceccagno

In the abbreviated word 报摘 bàozhāi both the original form 报纸 bàozhǐ ‘report/newspaper + paper = newspaper’ and the constituent of the abbreviated compound 报 bào mean ‘newspaper’. Thus, this constituent of the new compound word is semantically transparent. However, 摘 zhāi means ‘to select’, ‘to pick (flowers)’ and therefore does not coincide with the meaning of the original form 摘要 zhāiyào ‘to summarize + main points = summary/make an abstract’. Here full semantic transparency is not achieved even though the two meanings are close. Besides, while 摘 zhāi is only a verb, 摘要 zhāiyào may also be a noun with the meaning of ‘abstract/summary’. Thus, this type of abbreviated compound can be fully understood only by referring back to the semantic and formal elements of the underlying compound. Ceccagno and Basciano (2007, 2009a) and Ceccagno (2009) argue that in the case of compounds in which at least one of the surface constituents is a truncated form of an underlying compound and for which the analysis of the surface form is problematic, the features of the underlying form are to be taken into consideration. They call these compounds ‘metacompounds’ and suggest analyzing them through a method called ‘metacompounding’. In examples (2) and (3), I present some compounds formed through abbreviation in which the semantics of the surface form can be clarified only by analyzing them as metacompounds. (2)

内退 nèituì ‘retire under the legal age (but be entitled to all privileges as a normal retired person)’ 退休 tuìxiū

内 nèi

In (2) the constituent 退 tuì of the metacompound has the meanings ‘to retreat’, ‘to move back’, ‘to withdraw’, ‘to cancel’, but does not have the explicit meaning of ‘retirement (from work)’, which is instead provided by the underlying compound 退休 tuìxiū. In (3) another example is presented in which the underlying compound clarifies the semantics of the metacompound. (3)

话网 huàwăng ‘telephone network’ [电话] [diànhuà] ‘telephone’

网 wăng ‘net’

The compound 话网 huàwǎng ‘words + net’ in (3) should be considered a metacompound because the semantics becomes transparent only if the constituent 话 huà is analyzed as a truncated form of the underlying compound 电话 diànhuà ‘electric + words = telephone’. In fact the CCD (2002) does not list ‘telephone’ as one of the meanings of 话 huà. Therefore, only after the constituent 话 huà is considered the truncated form of the underlying compound 电话 diànhuà, can the semantics of the entire metacompound be fully grasped (similar compounds which ‘inherit’ their semantics from the underlying forms are discussed in Myers 2006). 230

Chinese neologisms: Word-formation strategies in Chinese Table 14.1  期 qī as a compound constituent and as a metacompound constituent 期房

qīfáng

期价 [期货]价 期市 [期货]市

qījià [qīhuò] jià qīshì [qīhuò] shì

‘date/term + house = forward house; house to be completed and delivered for use at an agreed date on real estate market’ ‘date + price = futures price’ ‘futures + price’ ‘date + market = futures market’ ‘futures + market’

One morpheme can be just a constituent of a compound in some cases and the truncated form of an underlying compound in other cases, as shown in Table 14.1. In Table 14.1, 期房 qīfáng ‘date/term + house = forward house’ is a regular compound whose meaning is obtained compositionally by combining the two constituents 期 qī ‘date/ term’ and 房 fáng ‘house’. Instead, the other two neologisms are metacompounds. In fact, only by considering 期 qī as the truncated form of the underlying compound 期货 qīhuò ‘futures’, does the semantics of the entire compound become transparent. Therefore, a compound word is considered a metacompound when its constituents do not have among their meanings one meaning that compositionally contributes to the meaning of the surface compound and therefore it is only by analyzing the underlying form that the compound can be correctly interpreted. Ceccagno and Basciano (2007, 2009a) and Ceccagno (2009) also discuss examples in which the compound words present not only semantic opacity but also categorical opacity if they are not analyzed as truncated forms of underlying compounds. Both the constituents of the compounds in (4) are truncated forms of underlying compounds: (4)

[卫视]N wèishì ‘satellite TV’ [卫星]N wèixīng ‘satellite’ [卫]V wèi ‘defend’

[星]N xīng ‘star’

[电视]N diànshì ‘television’ [电]N [视]V diàn shì ‘electricity’ ‘to look at’

The surface compound 卫视 wèishì presents both semantic and categorial opacity, and it is not possible to identify the head. However, if the underlying compounds 卫星 wèixīng and 电视 diànshì are taken into consideration, the structure of the metacompound and its meaning become clear and the head can be identified. The input lexical categories are not [V+V] but rather [N+N]; the compound is not coordinate but attributive; the compositional meaning is ‘satellite + TV’ and not ‘defend + to look at’; and the head is on the right 231

Antonella Ceccagno

(on headedness, see Scalise and Fabregas 2010; on the positions of the head in Chinese compounds, see Ceccagno 2007, Ceccagno and Basciano 2007, 2009a). By adopting metacompounding as a method of analysis, all compound properties – semantics, lexical category and the head – can be identified. Therefore, a compound is considered a metacompound not only in case of semantic opacity but also when the compound structure and/or the position of the head are opaque, and only by analyzing the underlying form/s can they be disambiguated. Thus, a metacompound can be thought of as an iceberg with only its upper portion exposed, and whose morphology can only be explained by referring to the submerged parts. Out of 709 disyllabic neologisms listed in CCD (2002), 49 or almost 7% of the total are metacompounds. They are presented and analyzed in Ceccagno and Basciano (2009a). Metacompounding is only intended as a method for the synchronic analysis of opaque compounds that are the truncated forms of underlying compounds. Therefore, this method only focuses on the analyzed compounds and does not try to retrace the original forms. For instance, in the analysis of the compound 彩电 cǎidiàn ‘color television’ it will be enough to analyze the constituent 电 diàn as the truncated form of the underlying compound 电视 diànshì ‘electric + to look at = television’, without going back to the previous form 电视机 diànshìjī or to 电视接收机 diànshì jiēshōujī, which is the original one. As for the constituent 彩 cǎi, even though it is the abbreviated form of 彩色 cǎisè ‘color’, it should not be considered the truncated form of an underlying compound because one of the meanings of 彩 cǎi is ‘color’. Abbreviation is also used for another word-forming strategy. This is a neologism-forming process that involves loanwords, calques, and hybrids and it will be addressed.

Traditional word-formation patterns for loanwords and hybrids One much discussed feature of the Chinese lexicon is the very high correspondence between syllables, morphemes and written characters (Chao 1968, to name only one work on this topic). In Chinese the large majority of morphemes correspond to a single syllable. And conversely, the majority of syllables can be analyzed as morphemes, for example, 花 huā ‘flower’, 马 mǎ ‘horse’, 书 shū ‘book’. This peculiarity has an impact on what is perceived as a word by native speakers. Chung (2006) contends that both the morpheme and the syllable should be considered as the relevant morphological units playing a central role in the processes of word formation. DeFrancis (1990: 187) has coined the word morphosyllabic for describing the close correspondence between the phonologic syllabic unit and the morpheme in Chinese. Given that the large majority of Chinese morphemes have a lexical nature, Chinese polysyllabic words are mainly built starting from available monosyllabic units and in many cases they can be analyzed according to the meaning of each monosyllable, as in (5): (5) 生 shēng ‘to be born’, ‘to give birth’, ‘life’, ‘to grow’ 生物 shēng-wù ‘to be born + matter = living creature’ 生物层 shēngwù-céng ‘living creature + layer = biosphere’ Loanwords instead are often rendered by creating new polysyllabic but monomorphemic words (on loanwords in Chinese, see, for example, Shi 2000).

232

Chinese neologisms: Word-formation strategies in Chinese

Many loanwords have entered into the Chinese lexicon starting from the second half of the nineteenth century, when new concepts for which there was no Chinese equivalent were introduced in China (Masini 1993). However, some disyllabic morphemes, such as 葡萄 pútáo ‘grape’, 蝴蝶 húdié ‘butterfly’, and 玻璃 bōli ‘glass’ had been introduced in Chinese as loanwords much earlier. Loanwords are formed by dividing foreign words into syllables and associating the syllables to Chinese morphemes/characters that are phonologically close to the foreign word. For example ‘sofa’ is rendered in Chinese as 沙发 shāfā; the name ‘Alexander’ is rendered as 亚历山大 Yālìshāndà. In the specific context of loanwords – and only there – the morphemes/characters lose their original meaning. For instance, the morphemes/characters 亚历山大 yā + lì + shān + dà (‘Asia + to experience + mountain + large’) are used for rendering the word ‘Alexander’ only because they are phonologically close to the foreign word (but for an emerging different pattern see below). Even though the phonological form plays the central role in the formation of neologisms, the best loanwords are those where both sound and meaning remind of the foreign word. For instance, in the loanword 黑客 hēikè ‘hacker’ the two syllables provide the Chinese sound most similar to the sound in English, while the two morphemes/characters mean ‘black guest’, thus also contributing semantically to the neologism. In particular, the morpheme/ character 黑 hēi ‘black’ has been preferred to other homophones in order to strengthen the negative meaning of the loanword. A particular case of loanword formation can be observed in the word 咖啡 kāfēi ‘coffee’, where the strategy of creating new characters has been adopted. The radical 口 kǒu ‘mouth’, which provides vague semantic information, has been added to the pre-existing characters 加 jiā ‘add’ and 非 fēi ‘mistake’. The resulting new characters 咖 kā – which has a pronunciation similar to 加 jiā but not exactly the same – and 啡 fēi have being coined to form the new monomorphemic disyllabic word 咖啡 kāfēi ‘coffee’. Hybrids are coined by combining a loanword with a native element that provides semantic information on its taxonomical category. In Chinese the phonological part of the hybrid is tied to morphemes, such as 车 chē ‘vehicle’ and 酒 jiǔ ‘wine’, as shown in the examples in (6): (6) 摩托车 mótuō-chē ‘motor (loanword) + vehicle = motorcycle’ 酒吧 jiǔ-bā ‘wine + bar (loanword) = wine bar’

Re-analysis of syllables of loanwords and hybrids as morphemes As seen above, Chinese loanwords are rendered by creating polysyllabic and monomorphemic words. Ceccagno and Basciano (2009a, 2009b) highlight a neologism-formation strategy that consists of the re-analysis of syllables of loanwords as morphemes with the same meaning of the original polysyllabic morpheme. As a rule, these syllables should not be used independently, given the fact that they are not morphemes but syllables that in the context of the polysyllabic word are to be intended as devoid of meaning. In actual fact, syllables of loanwords are re-analyzed as morphemes – which often are bound – and become constituents of newly formed compound words. This takes place through a process of truncation. In (7), examples of compound words formed using this strategy are presented:

233

Antonella Ceccagno

 (7) 奶咖 nǎikā ‘milk + kā = caffèlatte’ 清咖 qīngkā ‘pure + kā = black coffee’ In the examples in (7), the syllable 咖 kā, of the morpheme 咖啡 kāfēi, is re-analyzed as a morpheme that encompasses the meaning of the polysyllabic morpheme. As such, 咖 kā is used as the constituent of new compound words with the meaning of ‘coffee’. Another example of a loanword syllable re-analyzed as a morpheme that takes on the meaning of the entire loanword is presented in (8):  (8) 红客  hóngkè ‘red + kè = red hacker’ Red hackers are: ‘Chinese hackers motivated by patriotism. They use their expertise to defend the security of domestic networks and fend off attacks’ (WABs). In this case the second constituent of the new word is the second syllable of the monomorphemic word 黑客 hēikè ‘hacker’, a loanword from English. This is therefore another example of a syllable from a polysyllabic loanword re-analyzed as a morpheme that represents the entire loanword and as such is used as a constituent in a new compound word. Here, as in the previous examples, the truncated form absorbs the meaning of the polysyllabic word and becomes a bound morpheme. In this case, the neologism plays with colors. In the newly formed word 红客 hóngkè ‘red hacker’, red as a color for ‘red China’ replaces the black color. This suggests that in this word-formation strategy also the syllable 黑 hēi of 黑客 hēikè ‘hacker’ is implicitly intended as a morpheme. In (9) another example of a syllable of a polysyllabic loanword analyzed as a morpheme is presented. 麦克风 màikèfēng is the Chinese loanword for the English word ‘microphone’. Among the clients of karaoke parties, the new compound word ‘microphone monopolizer’ has been coined in which 麦 mài, the first syllable of 麦克风 màikèfēng, is used as the first constituent of the compound with the meaning of ‘microphone’ (Shanghai Daily, 21/8/2006):  (9) 麦霸  màibà ‘mài + tyrant = microphone monopolizer’ The same pattern of re-analysis of syllables as bound morphemes presented above is used starting from hybrids, as in the example presented in (10): (10) 扎啤 zhāpí ‘to prick + pí = draft beer’ 淡啤 dànpí ‘light + pí = light beer’ In the neologisms 扎啤 zhāpí ‘draft beer’ and 淡啤 dànpí ‘light beer’, the syllable 啤 pí of the hybrid 啤酒 píjiǔ ‘beer’ is used as a morpheme that takes on the meaning of ‘beer’. The hybrid 酒吧 jiǔbā ‘alcohol + bā = bar’ has undergone a similar process. Starting from the 1990s, neologisms containing ‘bar’ only use 吧 bā (Cheng 2004), which in the original loanword provided the phonological part of the hybrid, as shown in (11): (11) 网吧 wǎngbā ‘net + bā = Internet bar’ 吧台 bātái ‘bā + platform = bar/pub counter’ He (2001, cited in Cheng 2004) lists 24 neologisms formed with 吧 bā. The strategy of forming neologisms through the re-analysis of syllables of loanwords and hybrids as morphemes and using them as constituents of new compounds seems to be 234

Chinese neologisms: Word-formation strategies in Chinese

linked to two main features of Chinese. The first is the strong tendency of Chinese towards disyllabism and the trend to abbreviate words in order to meet this requirement. The second is the central role played by the syllable in Chinese and the strong correspondence between the syllable/character and the morpheme, to the point that speakers of Chinese language tend to associate each syllable/character with a meaning. It could be argued that the re-analysis of syllables as morphemes is a word-forming strategy that tends to bring back the traditional correspondence between syllable, character and morpheme. In fact, previously existing morphemes that had been transformed into loanword syllables are given back the status of morphemes.

Breaking the established rules in neologism formation In some cases newly formed words and phrases seem to intentionally aim at breaking pre-existing word-formation rules. This is the case of the neologism in (12) coined as an homophone – but not a homograph – of the loanword 亚历山大 Yālìshāndà with the meaning of ‘Alexander’ presented above: (12) 压力山大  yālìshāndà ‘pressure + mountain + large = pressure heavy as a mountain’ 压力山大 yālìshāndà is ‘a new way for Chinese people to express the feeling of too much stress or pressure’ (Chinasmack 2013). A similar neologism forming process is shown in (13) where starting from the Chinese word for the famous Spanish painter Picasso, 毕加索 Bìjiāsuǒ, a new word is formed: (13) 闭家锁  bìjiāsuǒ ‘close + house + locked’ This neologism has been translated as ‘stay-at-home person’. It refers to young persons ‘staying at home to pursue obsessive interests, particularly Japanese video games, comic books and animation movies’ (WABs). In (12) and (13) the syllables are interpreted as morphemes/characters. What is more, the syllables are not connected with the corresponding characters/morphemes for ‘Alexander’ and ‘Picasso’ but with characters/morphemes that are homophones but allographs. The tones are the same. This new word-formation pattern has emerged in the written language and taps into one central feature of the Chinese language, i.e. the high number of homophones. The new words presented in (12) and (13) are by no means an isolated phenomenon. The neologism 谐音成语 xiéyīn chéngyǔ ‘homophonic idiom’ has been coined to describe the widespread use on the web and in advertising of characters that are homophones but written differently. Online it is trendy to write 女猪 nǚzhū ‘female + pig’ instead of and with the meaning of 女主 nǚzhǔ ‘female protagonist’ (WABs); 打铁 dǎtiě ‘hit + iron = to forge iron’ is often used instead of and with the meaning of 打贴 dǎtiē ‘writing a post’. In these last two examples, the character that replaces the correct one is not a homotone but only a homophone. This phenomenon may have a lasting impact on Chinese writing.

New morphemes and increased lexical ambiguity As shown above, Chinese exhibits a strong tendency towards abbreviation. As a result, some constituents of metacompounds that are the truncated form of an underlying compound may 235

Antonella Ceccagno

over time acquire the meaning of the underlying compound. 警 jǐng is one such word. Previously, it was an abbreviation for 警察 jǐngchá ‘police’ but now ‘police’ is one of its meanings. Therefore the compound in (14) should not be analyzed as a metacompound but as a regular compound given that there is no need to go back to underlying compounds when analyzing it. (14) 特警  tèjǐng ‘special + police = special police’ The same holds true for the morpheme 影 yǐng. Its original meanings were ‘shadow’, ‘image of an ancestor’, ‘to hide’, but after being used in the formation of compounds as the truncated form of the underlying compound 电影 diànyǐng ‘film’, ‘cinema’, it has absorbed the meaning of ‘cinema’.3 This is the reason why in the analysis of the compound 影评 yǐngpíng (above) the constituent 影 yǐng has not been considered the truncated form of the underlying compound 电影 diànyǐng. A diachronic analysis is out of the scope of this chapter. Here I only intend to highlight the fact that the process of abbreviation in the formation of metacompounds can diachronically assign to a constituent the same meaning of the underlying compounds of which it was the abbreviated form. This seems to be a widespread trend in Chinese, even though it is impossible to foresee which truncated forms will become lexical morphemes. As seen above, a word-forming strategy consists of re-analyzing syllables of polysyllabic but monomorphemic words as morphemes. This has more than one implication. First, syllables of polysyllabic words that till recently had only been used as syllables, as a result of the re-analysis strategy, become new bound morphemes. Compound neologisms such as those in (15) create previously nonexistent morphemes that can be used as constituents of new compound words: (15) 奶咖 nǎikā ‘milk + kā [咖啡 kāfēi ‘coffee’] = caffèlatte’ new morpheme: 咖 kā 葡糖 pútáng ‘pú [葡萄 pútáo ‘grape’] + sugar = glucose’ new morpheme: 葡 pú 淡啤 dànpí ‘light + pí [啤酒 píjiǔ ‘beer’] = light beer’ new morpheme: 啤 pí Second, in the case of loanwords whose syllables already existed as morphemes before the loanword was coined, the re-analysis strategy contributes to increase the lexical ambiguity of existing morphemes. The neologism 麦工 Màigōng ‘mài + job = McJob’ (WABs, 9/4/2007) is a calque from English, where the word ‘McJob’ has existed since the 1980s with the meaning of a job, such as the one offered in McDonald’s restaurants, that is low paying, often temporary, and offers minimal or no benefits and opportunity for promotion (NOAD 2005). In Chinese, ‘McDonald’s’ is rendered as 麦当劳 Màidāngláo, which is a polysyllabic and monomorphemic word. In the 麦工 Màigōng neologism, 麦 mài, the first syllable of the loanword, is re-analyzed as a morpheme with the meaning of ‘McDonald’s style’, with the negative connotation that ‘Mc’ for ‘McDonald’ has taken in English. Another example is the neologism 麦时尚 Màishíshàng ‘McFashion’, a calque from English that indicates those brands which offer designer-looking clothes for a small budget: ‘as fashion has begun to resemble fast food: fast and cheap’ (Shanghai Daily 9/4/2007). Therefore 麦 mài is a good example of the growing lexical ambiguity resulting from the word-forming strategies discussed in this chapter. The original meaning of the morpheme 麦 mài is ‘wheat’. With the re-analysis of the first syllable of a loanword as a new morpheme, presented in (9), 麦 mài also acquires the meaning of ‘microphone’; and the calque 236

Chinese neologisms: Word-formation strategies in Chinese

from English described above adds to 麦 mài the new meaning of ‘in the same style as McDonald’s’. Another word-forming pattern consists of coining monosyllabic loanwords. Given that existing morphemes are used for the new loanwords, this strategy also contributes to increased lexical ambiguity in Chinese. Examples of recently coined monosyllabic loanwords are presented in (16): (16) 秀 xiù ‘show’ (original meaning: ‘beautiful’, ‘elegant’) 酷 kù ‘cool’ (original meanings: ‘cruel’; ‘very’, ‘extremely’) 猫 māo ‘modem’ (original meaning: ‘cat’) This pattern also involves morphemes newly formed as a result of the re-analysis of syllables as morphemes. In (17) the newly formed bound morpheme 咖 kā has taken on the meaning of ‘cast’: (17) A咖  A kā ‘A + cast = leading actor’ According to WABs, the neologism A咖 refers to ‘a leading actor in a movie or drama’ and ‘frontline showbiz pop stars’. Therefore, as it happens with metacompounding, also the word-formation strategy that re-analyzes syllables as morphemes contributes to increased lexical ambiguity in Chinese. Lexical ambiguity is also the result of the creation of new monosyllabic loanwords.

Lexical ambiguity as a fishpond for new compound neologisms Another word formation pattern in Chinese consists of coining neologisms by selecting a different meaning for one or more constituents of existing compound words. As a general rule, in Chinese compounding helps reduce lexical ambiguity. In fact, the restrictions of meaning that the compound’s constituents impose on each other helps disambiguate the morphemes from the semantic point of view, exactly because only one of the possible meanings is chosen. Ceccagno and Basciano (2009a) have singled out a word-formation pattern in Chinese that taps into the lexical ambiguity of the constituents of existing compound words. Different meanings are chosen for the constituents and as a result a new compound word is created that is homonymous with the existing one, as shown in (18): (18) 透析  tòuxī  ‘dialysis’  new meaning: ‘penetrating analysis’ In the compound word 透析 tòuxī ‘dialysis’, the meaning ‘to pass through’ is chosen among the possible meanings of the first constituent 透 tòu, and the meaning ‘to separate, to divide’ is chosen for the second constituent 析 xī. But given that 透 tòu also has the meaning of ‘penetrating’ and 析 xī also means ‘to analyze’, a homonymous compound word has been coined, with the meaning of ‘penetrating analysis’. In both cases, 透析 tòuxī can be a noun and a verb. Another example of a compound neologism coined by tapping into the different meanings of the constituents is presented in (19): 超生  chāoshēng  (i) ‘reincarnation’; (ii) (fig.) ‘be lenient’  new meaning: (19)  ‘have more children than the family planning policy allows’, ‘have unplanned births’ 237

Antonella Ceccagno

In the previous Buddhist compound word 超生 chāoshēng, the compositional meaning of ‘reincarnation’ was obtained by selecting the meaning ‘to transcend’ for the first constituent 超 chāo4 and the meaning ‘life’ for the second, 生 shēng. In the newly formed compound, the meaning ‘to exceed’ is instead chosen for the first constituent 超 chāo and the meanings ‘to give birth/to be born’ for the second, 生 shēng. With these new meanings selected for the constituents, a new compound is formed with the meaning of ‘exceeding the number of births (allowed by the family planning policy)’. A recent neologism formed through the reinterpretation of the constituents of an existing compound is presented in (20): 月光  yuèguāng ‘moonlight’ new meaning: 月光 (族) yuèguāng(zú) (20)  ‘(people with) empty pockets at the end of the month’ In the previously existing compound 月光 yuèguāng the meaning ‘moonlight’ was formed compositionally with the first constituent 月 yuè meaning ‘moon’ and the second 光 guāng meaning ‘light/ray’. In the new compound word, instead, the meaning of ‘month’ is chosen for 月 yuè and the meaning of ‘empty’ is chosen for 光 guāng. 族 zú denotes a group of people who share common characteristics.5 Thus, 月光族 yuèguāngzú refers to people who have spent all their money by the end of the month (WABs). In (21) a case of semantic loan from English that engenders the reinterpretation of an existing compound word is presented: (21) 溜冰  liūbīng ‘slide + ice = skating’  new meaning: ‘to take drugs’ In English ‘ice’ is also the street name for ‘crystal methamphetamine hydrochloride’, which is a synthetic drug. The new meaning has emerged as a metonymy linked to the crystal form of the molecular structure of both ice and the drug. The Chinese language also adopts the English metonymy and thus adds a new meaning to the word 冰 bīng ‘ice’. With the reinterpretation of the second constituent of the compound according to the new meaning of ‘ice’ as a street name for a drug, the compound 溜冰 liūbīng takes on the new meaning of ‘to take ice’. This word-forming strategy seems to stem from and to build both on lexical ambiguity and on compounding as an ambiguity reducing word-forming strategy. In fact, on the one hand, morphemes’ lexical ambiguity facilitates the choice of new meanings for the constituents; on the other hand, the restrictions of meaning that the compound’s constituents impose on each other help disambiguate the morphemes from the semantic point of view. This makes it possible to coin new compound words with the same constituents of previously existing compounds. This word-forming strategy is different from the word-forming pattern through semantic change. Semantic change is used extensively as a word-forming strategy. For instance, in the new words listed in CCD (2002) about 50 new words or 7% of the total are formed through semantic change. In (22) a new word coined as a metaphor is presented: (22) 牛皮癣  niúpíxuǎn ‘psoriasis’ new meaning: ‘nagging problem, eyesore ads’ In this case, the previous compound word 牛皮癣 niúpíxuǎn ‘psoriasis’, a disease that affects the skin, has taken on the new meaning of ‘prolonged nagging problem’ and according to WABs (2013) this new meaning is by now also used to depict ‘the eyesore ads, such as 238

Chinese neologisms: Word-formation strategies in Chinese

illicit trashy ads posted or printed on walls, telecomm poles, door steps or even pavements, which are very difficult to get rid of ’. In the examples presented in (18), (19), (20), and (21), instead, the new meaning is chosen at the constituents level and, as argued above, it taps into the lexical ambiguity of the constituents and relies on compounding as a means for disambiguating the meaning of the constituents.

Conclusions This chapter has presented some crucial word-forming strategies in Chinese; namely, metacompounding, the re-analysis of syllables as morphemes, neologisms formed by selecting a new meaning for the constituents of existing compound words, and the creation of new morphemes or new meanings for existing morphemes. Most of these word-forming strategies have been singled out only recently. Moreover, neologisms coined on the Internet that intentionally aim at breaking the prevailing word-formation strategies have also been discussed. The examples that illustrate these strategies have been drawn from a rich corpus that includes (i) neologisms formed in the last decades, found in the neologism section of CCD (2002); (ii) recent and very recent neologisms found on the WABs website (2013). With the exception of monosyllabic loanwords, the strategies presented in this chapter tend to meet the strong preference of Chinese for disyllabism. This chapter has shown that both metacompounding and the re-analysis of syllables as morphemes are word-forming strategies that create new compound words through truncation. In metacompounds at least one constituent is the truncated form of an underlying compound word; in the re-analysis of syllables as morphemes a constituent of a new compound word is a loanword’s syllable transformed into a bound morpheme. In both cases, the prosodic requirements of the language cause constituents of compounds to acquire a new meaning from an underlying item. Most of the word-forming patterns discussed in this paper – i.e. metacompounding, the re-analysis of syllables as morphemes, the selection of new meanings for the constituents of existing compounds, and the coining of monosyllabic loanwords – seem to contribute to increased lexical ambiguity. This suggests that prosodic trends play a central role in word formation in Chinese, to the point that they tend to override semantic–pragmatic considerations (see also Arcodia and Montermini 2012; and Duanmu 2012). Instead, recent word-forming patterns for words coined on the Internet – and therefore presented in the written form – seem to intentionally violate each and every pre-existing word-forming strategy.

Notes 1 As pointed out by Arcodia (2007), disyllabification and compounding are two distinct phenomena, albeit related and interacting with each other. 2 Coordinate compounds are those whose constituents exhibit a semantic relation of synonymy, as in the case of 美丽 měilì, or a logical coordination (‘and’), as in 新锐 xīnruì ‘new + sharp = new and sharp’, or antonymy, as in 呼吸 hūxī ‘exhale + inhale = breath’, redundancy as in 苹果 píngguǒ ‘apple + fruit = apple’, or reduplication, as in 天天 tiāntiān ‘day + day = every day’. A classification scheme that identifies three macro-types in compounding (subordinate, attributive, and coordinate) has been argued for by Bisetto and Scalise (2005). On coordinate compounds, see also Arcodia et al. (2010). For the classification of Chinese compounds, see Ceccagno and Scalise (2007) and Ceccagno and Basciano (2007).

239

Antonella Ceccagno 3 The Mathews’ Chinese–English Dictionary (MCED 1972) does not list the meaning of ‘cinema’ for 影 yǐng. However, Le Dictionnaire Français de la Langue Chinoise (DFLC 1976) does list it with the meaning of ‘cinema’, as an abbreviation. 4 Recently 超 chāo has also been used with the meaning of ‘ultra’, as can be seen in the word 超声 波 chāoshēngbō ‘ultra + sound + wave = ultrasound’. 5 族 zú is an example of semantic change. It is a bound morpheme originally meaning ‘clan’, ‘race’, ‘nationality’, ‘ethnicity’ that through semantic change has taken on the meaning of ‘a group of people who share common characteristics’. Shen (1995) and Ceccagno and Basciano (2009a) consider 族 zú as a lexeme undergoing grammaticalization that is in the process of becoming an affix.

References Arcodia, Giorgio Francesco (2007) ‘Chinese: A Language of Compound Words?’, in Fabio Montermini, Gilles Boyé, and Nabil Hathout (eds) Selected Proceedings of the 5th Décembrettes: Morphology in Toulouse, Somerville, MA: Cascadilla Proceedings Project, 79–90. Available at: http://www. lingref.com/ (document #1617). Arcodia, Giorgio Francesco, Nicola Grandi, and Bernhard Wälchli (2010) ‘Coordination in Compounding’, in Sergio Scalise and Irene Vogel (eds) Cross-disciplinary Issues in Compounding, Amsterdam and Philadelphia: John Benjamins, 177–98. Arcodia, Giorgio Francesco and Fabio Montermini (2012) ‘Are Reduced Compounds Compounds? Morphological and Prosodic Properties of Reduced Compounds in Russian and Mandarin Chinese’, in Vincent Renner, François Maniez, and Pierre J. L. Arnaud (eds) Cross-disciplinary Perspectives on Lexical Blending, Berlin-New York: Mouton de Gruyter, 93–113. Basciano, Bianca and Antonella Ceccagno (2009) ‘The Chinese Language and Some Notions From Western Linguistics’, Lingue e Linguaggio 8(1): 105–35. Bisetto, Antonietta and Sergio Scalise (2005) ‘Classification of Compounds’, Lingue e Linguaggio 2: 319–32. CCD (2002)《现代汉语词典》(The Contemporary Chinese Dictionary), Beijing: Foreign Language Teaching and Research Press. Ceccagno, Antonella (2007) ‘Regolarità e peculiarità dei composti del cinese’, in Laura De Giorgi and Guido Samarani (eds) Percorsi della civiltà cinese tra passato e presente, Venezia: Cafoscarina, 149–66. Ceccagno, Antonella (2009) ‘Metacompounds in Chinese’, Lingue e Linguaggio 2: 195–212. Ceccagno, Antonella and Bianca Basciano (2007) ‘Compound Headedness in Chinese: An Analysis of Neologisms’, Morphology 2: 207–31. Ceccagno, Antonella and Bianca Basciano (2009a) Shuobuchulai. La formazione delle parole in cinese, Bologna: Serendipità. Ceccagno, Antonella and Bianca Basciano (2009b) ‘Sino-Tibetan: Mandarin Chinese’, in Rochelle Lieber and Pavol Stekauer (eds) The Oxford Handbook of Compounding, Oxford: Oxford University Press, 478–90. Ceccagno, Antonella and Sergio Scalise (2006) ‘Classification, Structure and Headedness of Chinese Compounds’, Lingue e Linguaggio 2: 233–60. Ceccagno, Antonella and Sergio Scalise (2007) ‘Composti del cinese: analisi delle strutture e identificazione della testa’, in Annamaria Palermo (ed) La Cina e l’altro, Napoli: Università degli studi di Napoli ‘L’Orientale’, 503–43. Chao, Yuen-Ren (1968) A Grammar of Spoken Chinese, Berkeley-Los Angeles: University of California Press. Cheng, Lixia (2004)〈语言接触,类推与形态化〉(‘Linguistic Contact, Analogy and Morphologization’), Waiyu yu waiyu jiaoxue 4: 53–6. Chinasmack, available at: http://advertising.chinasmack.com/2013/top-10-chinese-buzzwords-of-thelast-year.html. Chung, Karen (2006) Mandarin Compound Verbs, Taiwan Journal of Linguistics: Book Series in Chinese Linguistics, Taipei: Crane Publishing. Croft, William (2000) Explaining Language Change: An Evolutionary Approach, Harlow, Essex: Longman. DeFrancis, John (1984/1990) The Chinese Language: Facts and Fantasy, Honolulu: University of Hawaii Press; Taipei: Crane Publishing (reprinted).

240

Chinese neologisms: Word-formation strategies in Chinese DFLC (1976) Dictionnaire Français de la Langue Chinoise, Paris-Taipei: Insitut Ricci-Kuangchi Press. Duanmu, San (2012) ‘Word-length Preference in Chinese: A Corpus Study’, Journal of East Asian Linguists 21: 89–114. Feng, Shengli (1998) ‘Prosodic Structure and Compound Words in Classical Chinese’ in Jerome L. Packard (ed) New Approaches to Chinese Word Formation, Berlin-New York: Mouton de Gruyter, 196–259. Huang, Shuanfan (1998) ‘Chinese as a Headless Language’, in Jerome L. Packard (ed) New Approaches to Chinese Word Formation, Berlin-New York: Mouton de Gruyter, 261–83. Labov, William (1972) Sociolinguistic Patterns, Philadelphia, PA: University of Pennsylvania Press. Lieber, Rochelle and Pavol Stekauer (eds) (2009) The Oxford Handbook of Compounding, Oxford: Oxford University Press, 478–90. Ling, Yuanzheng (2000)《现代汉语缩略语》(Abbreviations in Contemporary Chinese), Beijing: Yuwen Chubanshe. Masini, Federico (1993) ‘The Formation of Modern Chinese Lexicon and its Evolution Towards a National Language: The Period from 1840 to 1898’ (English translation), Journal of Chinese Linguistics Monograph Series 6. MCED (1972) Mathews’ Chinese–English Dictionary, Cambridge, MA: Harvard University Press. Myers, James (2006) ‘Processing Chinese Compounds: A Survey of the Literature’, in Gary Libben and Gonia Jarema (eds) The Representation and Processing of Compound Words, Oxford: Oxford University Press, 169–96. NOAD (2005) The New Oxford American Dictionary (2nd ed.), Oxford: Oxford University Press. Packard, Jerome L. (1998) ‘Introduction’, in Jerome L. Packard (ed) New Approaches to Chinese Word Formation, Berlin-New York: Mouton de Gruyter, 3–34. Packard, Jerome L. (2000) The Morphology of Chinese, Cambridge: Cambridge University Press. Plag, Ingo (2003) Word-Formation in English, Cambridge: Cambridge University Press. Scalise, Sergio and Antonio Fabregas (2010) ‘The Head in Compounding’, in Sergio Scalise and Irene Vogel (eds) Cross-disciplinary Issues in Compounding, Amsterdam and Philadelphia: John Benjamins, 109–26. Scalise, Sergio and Irene Vogel (eds) (2010) Cross-disciplinary Issues in Compounding, Amsterdam and Philadelphia: John Benjamins. Shen, Mengying (1995)〈试论新词缀化的汉民族性〉(‘On the Chineseness of the New Affixation Trend’),《南京师大学报》(Journal of the Nanjing Normal University) 1: 35–41. Shi, Youwei (2000)《汉语外来词》(Loanwords in Chinese), Beijing: The Commercial Press. WABs, Shanghai Daily’s Well-acclaimed buzzwords, available at: http://buzzword.shanghaidaily.com/ Wang, Jihui (2001)《现代汉语缩略语词研究》(Research on Abbreviations in Contemporary Chinese), Tianjin: Tianjin Renmin Chubanshe. XHSC (2002)《现代汉语缩略语词典》(Dictionary of Abbreviations in Contemporary Chinese), Beijing: Yuwen chubanshe. Xing, Janet Zhiqun (2006) Teaching and Learning Chinese As a Foreign Language: A Pedagogical Grammar, Hong Kong: Hong Kong University Press. Yuan, Hui (2002) ‘Introduction’, in Hui Yuan and Ruan Xianzhong (eds)《汉语缩略语词典》(Dictionary of Abbreviations in Contemporary Chinese), Beijing: Yuwen Chubanshe. Zhou, Xiaolin, William Marslen-Wilson, Marcus Taft, and Shu Hua (1999) ‘Morphology, Ortography and Phonology in Reading Chinese Compound Words’, Language and Cognitive Processes 14(5–6): 525–65.

241

15 Chinese: Parts of Speech Candice Chi-Hang Cheung The Hong Kong Polytechnic University, Hong Kong, China

1. Introduction Whether Chinese has the same range of parts of speech (or categories) as the Indo-European languages has been of much discussion in the literature. In particular, while it is generally recognized that Chinese makes a distinction between nouns and verbs, scholars hold different opinions regarding the rest of the categories (see Chao 1968; C. N. Li and Thompson 1981; D. Zhu 1982; and Xing and Ma 1992, inter alia). The differences in opinions are partly due to the different theoretical backgrounds adopted by linguists, and partly due to the use of different terminological conventions. As a result, different criteria are employed for classifying words, and different terminological conventions are used to label the categories. In order to address the question of whether Chinese possesses the same range of categories as the Indo-European languages, I shall make use of the familiar categories of the IndoEuropean languages whenever possible. Before we proceed, it is important to mention that this chapter is intended to offer a comprehensive survey of the major categories in Chinese, aiming to establish the set of categories that is commonly found in Chinese and the Indo-European languages (section 2) and those that are available in Chinese only (section 3). In particular, in the discussion of the major categories shared by Chinese and the Indo-European langugages, I shall examine their characteristic features and discuss in what ways they are similar to and different from those in the Indo-European languages. Furthermore, I shall review the factors that contribute to the long-standing debate on the categorial status of adjectives, prepositions and localizers.

2.  Categories commonly found in Chinese and the Indo-European languages This section introduces the categories that are commonly found in Chinese and the IndoEuropean languages, including nouns, verbs, adjectives, adverbs, prepositions and conjunctions.

2.1. Nouns The existence of nouns as an independent category in Chinese is uncontroversial (Chao 1968, D. Zhu 1982, C.-T. J. Huang, Li and Li 2009: Ch. 1). One of the characteristics of nouns is that they can be preceded by a [Numeral + Classifier] sequence (Chao 1968, D. Zhu 1982), 242

Chinese: Parts of speech

as shown in the following examples.1 In this regard, Chinese nouns behave like mass nouns in the Indo-European languages, as they always require the presence of classifiers in enumeration (see Chierchia 1998 and Cheng and Sybesma 1999 for discussion of the availability of count-mass distinction in Chinese).2,3 (1) (a)

一个 人 yī-ge rén one-Cl person ‘one person’

(b) 三本 书 sān-běn shū three-Cl book ‘three books’

 

两杯 啤酒 liǎng-bēi píjiǔ two-glass beer ‘two glasses of beer’

(d) 四磅 猪肉 sì-bàng zhūròu four-pound pork ‘four pounds of pork’

(c)

Another characteristic of nouns is that they cannot be negated by bù 不 ‘not’ (Chao 1968, D. Zhu 1982 and C.-T. J. Huang, Li and Li 2009: Ch. 1) nor can they be modified by the intensifier hěn 很 ‘very’ (Chao 1968, D. Zhu 1982), as evidenced by the ill-formedness (indicated by ‘*’ below and throughout) of (2) and (3), respectively.4 (2) (a)

*不 *bù not

人 rén person

(b) *不 *bù not

书 shū book

 

*不 *bù not

啤酒 píjiǔ beer

(d) *不 *bù not

猪肉 zhūròu pork

(c)

(3) (a)

*很 *hěn very

人 rén person

(b) *很 *hěn very

书 shū book

 

*很 *hén very

啤酒 píjiǔ beer

(d) *很 *hěn very

猪肉 zhūròu pork

(c)

Apart from the fact that nouns in Chinese behave like mass nouns in the Indo-European languages, Chinese nouns differ from those in the Indo-European languages in that they are not inflected for number, gender and case. This is due to the fact that Chinese is an isolating language and is morphologically impoverished. Furthermore, unlike Indo-European languages, Chinese lacks definite and indefinite articles. Thus, the same form of a noun such as xiǎohái 小孩 ‘child’ may appear in the subject or object position. In terms of interpretational properties, it can be interpreted as singular or plural and generic, definite or indefinite, depending on the contexts (Chierchia 1998, Cheng and Sybesma 1999, Y.-H. A. Li 1999):5 (4) (a)

小孩 喜欢 猫。 xiǎohái xǐhuān māo child like cat (i)   ‘Children like cats’  (generic) (ii)  ‘The child/children like(s) cats’  (definite) 243

Candice Chi-Hang Cheung

(b) 他 会 去 找 小孩。 tā huì qù zháo xiǎohái he will go find child (i)  ‘He will go find a/some child/children’  (indefinite) (ii)  ‘He will go find the child/children’  (definite)

2.1.1.  Classification of nouns Based on the semantic properties of nouns and the types of classifiers being used, we can divide the class of nouns into four subclasses, namely, count nouns, mass nouns, collective nouns and abstract nouns (D. Zhu 1982; see also Chao 1968).

2.1.1.1.  Count nouns Count nouns refer to the type of noun that denotes tangible and discrete entities that are countable. Typical examples of count nouns include kèrén 客人 ‘guest’, shū 书 ‘book’, shé 蛇 ‘snake’, huā 花 ‘flower’, among many others. Unlike count nouns in the Indo-European languages, which can be directly combined with a numeral (e.g. one guest, three books, etc.), count nouns in Chinese require the use of sortal classifiers, such as wèi 位, běn 本, tiáo 条, etc., which encode some distinctive features like shape, as illustrated below (D. Zhu 1982; see also Chao 1968): (5) (a)

一位 客人 yī-wèi kèrén one-Cl guest ‘one guest’

(b) 三本 书 sān-běn shū three-Cl book ‘three books’

 

两条 蛇 liǎng-tiáo shé two-Cl snake ‘two snakes’

(d) 十朵 花 shí-duō huā ten-Cl flower ‘ten flowers’

(c)

For instance, the sortal classifier tiáo 条 in (5c) classifies long, cylindrical and flexible objects (H.-Y. J. Tai and Wang 1990), and thus it can serve as the classifier of shé 蛇 ‘snake’ and entities with similar characteristics: (6) (a)

一条 绳子 yī-tiáo shénzi one-Cl rope ‘one rope’

(b) 三条 鱼 sān-tiáo yú three-Cl fish ‘three fish’

Furthermore, a subset of count nouns designating human beings can be suffixed by the collective marker -men 们 to refer to a group of people that are known to the speakers and hearers (Iljic 1994, 2001, Y.-H. A. Li 1998, among others). This can be witnessed by the fact that a count noun suffixed with -men 们 is interpreted as definite, as in (7a), in contrast with the bare count noun, which can be interpreted as indefinite or definite, as in (7b):

244

Chinese: Parts of speech

(7) (a)

他 tā he ‘He

(b) 他 tā he (i)   (ii) 

会 去 找 孩子们。 huì qù zhǎo háizi-men. will go find child-MEN will go find the children’  (definite) 会 去 找 孩子。 huì qù zhǎo háizi. will go find child ‘He will go find a/some child/children’  (indefinite) ‘He will go find the child/children’  (definite)

However, unlike genuine plural markers in the Indo-European languages, count nouns suffixed with -men 们 cannot be preceded by a [Numeral + Classifier] sequence even when the numeral is greater than ‘one’ (Iljic 1994, 2001; Y.-H. A. Li 1998, among others), whereas a bare count noun can, as illustrated by the contrast between (8a) and (8b). (8) (a)

*他 *tā he

会 huì will

(b) 他 会 tā huì he will ‘He will go

去 qù go

找 zhǎo find

三个 sān-ge three-Cl

去 qù go find

找 三个 zhǎo sān-ge find three-Cl three children’

孩子们。 háizi-men child-MEN 孩子。 háizi. child

2.1.1.2.  Mass nouns In contrast with count nouns, mass nouns refer to the type of noun that denotes entities that are not countable. Typical examples of mass nouns include shuǐ 水 ‘water’, jiǔ 酒 ‘wine’, táng 糖 ‘sugar’, qìyóu 汽油 ‘petrol’, bù 布 ‘cloth’, among many others. As noted by Chao (1968), there are four ways of classifying mass nouns. One way is to make use of standard measure words such as chǐ 尺 ‘foot’, bàng 磅 ‘pound’, jiālún 加仑 ‘gallon’, gōngjīn 公斤 ‘kilo’, as illustrated below: (9) (a)

一两 yī-liǎng one-ounce ‘one ounce

 

两加仑 liǎng-jiālún two-gallon ‘two gallons

(c)

银子 yínzi silver of silver’ 汽油 qìyóu petrol of petrol’

(b) 三磅 sān-bàng three-pound ‘three pounds (d) 十公斤 shí-gōngjīn ten-kilo ‘ten kilos of

糖 táng sugar of sugar’

米 mǐ rice rice’

Another way is to make use of container measure words such as bēi 杯 ‘glass’, píng 瓶 ‘bottle’, etc. or temporary measure words such as shēn 身 ‘body’, wūzi 屋子 ‘room’, etc. that indicate the temporary location of the mass:

245

Candice Chi-Hang Cheung

(10)

(a)

一杯 yī-bēi one-glass ‘one glass

水 shuǐ water of water’

(b) 三瓶 sān-píng three-bottle ‘three bottles

(c)

一身 yī-shēn one-body ‘a body of

雪 xuě snow snow’

(d) 一屋子 yī-wūzi one-house ‘a houseful

酒 jiǔ wine of wine’

烟 yān smoke of smoke’

A third way is to employ indeterminate number/amount measure words such as yīdiǎn 一点 ‘a little’ and yīxiē 一些 ‘some’: (11)

(a)

一点 yīdiǎn a little ‘a little

(b) 一些 yīxiē some ‘some

水 shuǐ water water’

酒 jiǔ wine wine’

A fourth way is to employ measure words that describe the shape in which the mass can be gathered. Examples of this type of measure word include duī 堆 ‘heap’, tán 滩 ‘pool’, etc., as shown below: (12)

(a)

两堆 liǎng-duī two-heap ‘two heaps

泥 ní earth of earth’

(b) 一滩 yī-tán one-pool ‘one pool

水 shuǐ water of water’

2.1.1.3.  Collective nouns Collective nouns are inherently plural. Some illustrative examples are given below: (13)

(14)

(a)

父母 fù-mǔ father-mother ‘parents’

(b) 饭菜 fàn-cài rice-vegetable ‘food’

(c)

子女 zí-nǚ son-daughter ‘children’

(d) 亲友 qīn-yǒu relative-friend ‘friends and relatives’

(a)

纸张 zhǐzhāng paper ‘paper’

(b) 花朵 huāduǒ flower ‘flowers’

(c)

马匹 mápǐ horse ‘horses’

(d) 书本 shūběn book ‘books’ 246

Chinese: Parts of speech

The examples in (13) involve juxtaposition of two subordinate terms, whereas the examples in (14) are seemingly composed of a noun and a sortal classifier, as evidenced by the fact that the second element in the collective nouns in (14a–d) can be used as a sortal classifier, as shown below. The fact that the second member can be used as a sortal classifier in Modern Chinese is not surprising, as many sortal classifiers have their historical origins as nouns (Wang 1955, Liu Shiru 1965, among others): (15)

(a)

一张 纸 yī-zhāng zhǐ one-Cl paper ‘a piece of paper’

(b) 三朵 花 sān-duǒ huā three-Cl flower ‘three flowers’

(c)

两匹 马 liáng-pí mǎ two-Cl horse ‘two horses’

(d) 四本 书 sì-běn shū four-Cl book ‘four books’

Due to the fact that collective nouns are inherently plural, they cannot be preceded by a [Numeral + Classifier] sequence, as evidenced by the ill-formedness of the following examples (Chao 1968, D. Zhu 1982): (16)

(a)

*两个 *liǎng-ge two-Cl Intended:

(c)

*一张 纸张 *yī-zhāng zhǐzhāng one-Cl paper Intended: ‘one piece of paper’

父母 fù-mǔ father-mother ‘two parents’

(b) *三碟 *sān-dié three-Cl Intended:

饭菜 fàn-cài rice-vegetable ‘three plates of food’

(d) *四朵 *sì-duǒ four-Cl Intended:

花朵 huāduǒ flower ‘four flowers’

Instead, collective nouns can only take measure words concerned with grouping (e.g. duì 对 ‘pair’ and pī 批 ‘batch’), temporal measure words concerned with location (e.g. zhuō 桌 ‘table’) or indeterminate measure words such as yīxiē 一些 ‘some’, as demonstrated below: (17)

(a)

两对 liǎng-duì two-pair ‘two pairs

(c)

一批 yī-pī one-batch ‘a batch of

父母 fù-mǔ father-mother of parents’ 纸张 zhǐzhāng paper papers’

(b) 一桌 yī-zhuō one-table ‘a table of (d) 一些 yīxiē some ‘some

饭菜 fàn-cài rice-vegetable food’

花朵 huāduǒ flower flowers’

2.1.1.4. Abstract nouns Abstract nouns, as its name implies, do not refer to concrete entities. Rather, they are often used to denote abstract notions, such as complex or sophisticated situations, experiences, processes, qualities or phenomena in diverse areas. Some illustrative examples of abstract nouns are given below: 247

Candice Chi-Hang Cheung

(18)

(a)

真理 zhēnlǐ truth ‘truth’

(b) 病 bìng sickness ‘sickness’

(c)

矛盾 máodùn contradiction ‘contradiction’

(d) 效果 xiàoguǒ effect ‘effect’

Abstract nouns may be preceded by the general classifier ge 个, measure words indicating type, such as zhǒng 种 ‘type, kind, category’ or indeterminate measure words such as yīxiē 一些 ‘some’ and yīdiǎn 一点 ‘a little’ (Chao 1968, D. Zhu 1982), as demonstrated below: (19)

(a)

一个 真理 yī-ge zhēnlǐ one-Cl truth ‘a truth’

(b) 一种 yī-zhǒng one-kind ‘a kind of

(c)

一些 yīxiē some ‘some

(d) 一点 yīdiǎn a.little ‘a little

矛盾 máodùn contradiction contradictions’

病 bìng sickness sickness’

效果 xiàoguǒ effect effect’

2.2. Verbs Unlike nouns, verbs can be negated by bù 不 ‘not’ in Chinese (Chao 1968; C.-T. J. Huang, Li and Li 2009: Ch. 1): (20)

(a)

我 不 休息。 wǒ bù xiūxī. I not rest ‘I do not rest’

(b)

他 不 喜欢 数学。 tā bù xǐhuān shùxué. he not like mathematics ‘He does not like mathematics’

(c)

他 不 批评 别人。 tā bù pīpíng biérén. he not criticize other.person ‘He does not criticize others’

(d)

我 不 送 wǒ bú sòng I not give ‘I do not give

他 tā him him

礼物。 lǐwù. present (a) present(s)’

2.2.1.  Classification of verbs: action verbs vs. stative verbs Among the verbs in Chinese, they can be divided into two classes: action verbs and stative verbs. The former depict actions, whereas the latter express the states of being. A well-known characteristic of action verbs is that they are compatible with aspect markers, such as the progressive aspect marker zài 在, as in (21a-d) (see C. N. Li and Thompson 1981: Ch. 6 and Smith 1991 for discussion of the aspect markers in Chinese), while stative verbs can be modified by the intensifier hěn 很 although its presence is not mandatory, as shown in (22a–d) (Chao 1968, D. Zhu 1982, inter alia).6 248

Chinese: Parts of speech

(21)

(22)

(a)

他 在 休息。 tā zài xiūxī. he Prog rest-Perf ‘He is resting’

(b)

我 在 批评 他。 wǒ zài pīpíng tā. I Prog criticize he ‘I am criticizing him’

(c)

我 在 研究 这个 问题。 wǒ zài yánjiū zhè-ge wèntí. I Prog study this-Cl problem ‘I am studying this problem’

(d)

他 在 教 我 法语。 tā zài jiāo wǒ Fáyǔ. he Prog teach me French ‘He is teaching me French’

(a)

他 (很) tā (hén) he very ‘He likes

(b) 我 (很) wǒ (hěn) I very ‘I respect

(c)

他 (很) 羡慕 我。 tā (hěn) xiànmù wǒ. he very envy me ‘He envies me (a lot)’

喜欢 数学。 xǐhuān shùxué. like mathematics mathematics (a lot)’

尊敬 他。 zūnjìng tā. respect him him (a lot)’

(d) 我 (很) 想念 他。 wǒ (hén) xiǎngniàn tā. I very miss him ‘I miss him (a lot)’

2.2.2.  Classification of verbs: subcategorization properties Another way of classifying verbs is based on the number of arguments for which they subcategorize. Similar to the verbs in the Indo-European languages, verbs in Chinese can be classified into intransitive, transitive and ditransitive verbs.

2.2.2.1. Intransitive verbs Intransitive verbs are those that subcategorize for a single argument. Typical examples of intransitive verbs are lái 来 ‘come’, sǐ 死 ‘die’, shuì 睡 ‘sleep’, fāshēng 发生 ‘happen’, pǎo 跑 ‘run’, xiūxī 休息 ‘rest’, líkāi 离开 ‘leave’, etc. Some illustrative examples are given below:7 (23)

(a)

他 tā he ‘He

来了。 lái-le. come-Perf came’

(b) 他  死了。 tā  sǐ-le. he die-Perf ‘He died’

(c)

他 tā he ‘He

睡了。 shuì-le. sleep-Perf slept’

(d) 刚刚 发生了 gānggāng fāshēng-le just.now happen-Perf ‘A car accident happened

(e)

他 tā he ‘He

跑了。 pǎo-le. run-Perf ran’

(f)

车祸。 chē-huò. car-accident just now’

他  休息了。 tā  xiūxī-le. he rest-Perf ‘He rested’

Interestingly, many intransitive verbs in Chinese can be used transitively (see Chappell 1999, Xu 1999, 2001, J. Han 2000, Wen and Chen 2001, Shen et al. 2002, Sun and Wu 2003, S.-W. Tang 2004, Hole 2005, X. Zhu 2005, Pan and Han 2006, C.-T. J. Huang 2007, inter alia), as shown below: 249

Candice Chi-Hang Cheung

(24)

(a)

你 来 碗 牛肉 面 吧？ nǐ lái wǎn niúròu miàn ba? you come bowl beef noodle Prt ‘Have a bowl of beef noodles, will you?’

(b) 他 tā he ‘His (c)

死了 父亲。 sǐ-le fùqīn. die-Perf father father died’

他们 tāmen they ‘A car

(d) 他 tā he ‘He

昨天 发生了 zuótiān fāshēng-le yesterday happen-Perf accident happened to them

车祸。 chē-huò. car-accident yesterday’

跑了 两个 犯人。 pǎo-le liǎng-ge fànrén. run-Perf two-Cl criminal had two criminals ran away’

2.2.2.2. Transitive verbs As for transitive verbs, they subcategorize for two arguments: an external argument that corresponds to the subject and an internal argument that corresponds to the object. Both action and stative verbs can be used as transitive verbs. Typical examples of action verbs that are transitive include zuò 做 ‘do’, mǎi 买 ‘buy’, qù 去 ‘go’, pīpíng 批评 ‘criticize’, etc. and those of stative verbs that are transitive include xǐhuān 喜欢 ‘like’, zūnjìng 尊敬 ‘respect’, xiǎngniàn 想念 ‘miss’, xiànmù 羡慕 ‘envy’, etc. As mentioned earlier, action verbs can be followed by aspect markers, such as the perfective aspect marker -le 了, as shown in (25a–d), whereas stative verbs can be modified by the intensifier hěn 很, as in (26a–d). (25)

(26)

(a)

他 做了 很多 坏 事。 tā zuò-le hěnduō huài shì. he do-Perf many bad thing ‘He did many bad things’

(b) 他 买了 一本 书。 tā mǎi-le yī-běn shū. he buy-Perf one-Cl book ‘He bought a book’

(c)

他 去过 北京。 tā qù-guo Běijīng. he go-Exp Beijing ‘He has been to Beijing’

(d) 他 批评了 别人。 tā pīpíng-le biérén. he criticize-Perf other.people ‘He criticized others’

(a)

他 很 喜欢 数学。 tā hěn xǐhuān shùxué. he very like mathematics ‘He likes mathematics a lot’

(b) 我 很 尊敬 他。 wǒ hěn zūnjìng tā. I very respect him ‘I respect him a lot’

(c)

他 很 羡慕 我。 tā hěn xiànmù wǒ. he very envy me ‘He envies me a lot’

(d) 我 很 想念 他。 wǒ hén xiǎngniàn tā. I very miss him ‘I miss him a lot’ 250

Chinese: Parts of speech

2.2.2.3.  Ditransitive verbs Ditransitive verbs subcategorize for three arguments: an external argument that corresponds to the subject and two internal arguments, one corresponding to an indirect object and the other corresponding to a direct object. Typical examples of ditransitive verbs include sòng 送 ‘give’, jiāo 教 ‘teach’, dì 递 ‘hand’, jì 寄 ‘send’, huán 还 ‘return’, mài 卖 ‘sell’, jièshào 介绍 ‘introduce’, etc. Unlike English whose ditransitive verbs may appear in double object constructions (e.g. I gave him a book) and dative constructions (e.g. I gave a book to him), ditransitive verbs in Chinese may appear in three types of constructions: double object constructions in the form of [Verb + Noun Phrase + Noun Phrase], dative constructions in the form of [Verb + Noun Phrase + gěi 给 + Noun Phrase]8 and verb-gěi 给 double object constructions in the form of [Verb-gěi 给 + Noun Phrase + Noun Phrase], with the verb and gěi 给 forming a verbal complex (see D. Zhu 1979, Y.-H. A. Li 1990, and F.-H. Liu 2006 for more detailed discussion). As noted by F.-H. Liu (2006), not all ditransitive verbs can appear in the three constructions in Chinese. As shown below, while sòng 送 ‘give’ can appear in all three constructions (27a–c), jiāo 教 ‘teach’ can appear in double object constructions and verb-gěi double object constructions but not dative constructions (28a–c) and dì 递 ‘hand’ can appear in dative constructions and verb-gěi 给 double object constructions but not double object constructions (29a–c) (the examples in (28)–(29) are adapted from Liu 2006). (27)

(28)

(a)

他 tā he ‘He

送了 sòng-le give-Perf gave me a

我 一本 wǒ yī-běn me one-Cl book’

(b) 他 tā he ‘He

送了 一本 sòng-le yī-běn give-Perf one -Cl gave a book to me’

(c)

他 tā he ‘He

送给 sòng-géi give-to gave me a

(a)

他 tā he ‘He

教了 jiāo-le teach-Perf taught me a

书。 shū. book

书 shū book

我 一本 wǒ yī-běn me one -Cl book’ 我 一个 wǒ yī-ge me one-Cl method’

给 géi to

我。 wǒ. me

书。 shū. book

方法。 fāngfǎ. method

(b) *他 教了 一个 方法 给 *tā jiāo-le yī-ge fāngfǎ gěi he teach-Perf one-Cl method to Intended: ‘He taught a method to me’ (c)

他 tā he ‘He

教给 jiāo-gěi teach-to taught me

我 一个 wǒ yī-ge me one-Cl a method’

方法。 fāngfǎ. method

251

我。 wǒ. me

Candice Chi-Hang Cheung

(29)

(a)

*他 递了 我 *tā dì-le wǒ he hand-Perf me Intended: ‘He handed

一杯 啤酒。 yī-bēi píjiǔ. one-glass beer me a glass of beer’

(b) 他 tā he ‘He

递了 一杯 dì-le yī-bēi hand-Perf one-glass handed a glass of beer

(c)

递给 我 一杯 啤酒。 dì-gěi wǒ yī-bēi píjiǔ. hand-to me one-glass beer handed me a glass of beer’

他 tā he ‘He

啤酒 给 píjiǔ géi beer to to me’

我。 wǒ. me

2.3. Adjectives Adjectives and stative verbs share many common properties in Chinese. For this reason, there are scholars who hold the view that adjectives should be subsumed under stative verbs and that adjectives should not be regarded as an independent category (McCawley 1992; see also Chao 1968). Contrary to this view, many scholars argue that adjectives should be regarded as an independent category distinct from stative verbs (D. Zhu 1982; Paul 2006, 2010). Adopting the latter view, I begin by reviewing the common properties shared by adjectives and verbs (section 2.3.1), followed by a discussion of the arguments for treating adjectives as an independent category (section 2.3.2).

2.3.1.  Common properties of adjectives and verbs Unlike adjectives in the Indo-European languages, those in Chinese can function as predicates without the presence of the copular verb shì 是 ‘be’. Furthermore, they are typically accom­ panied by the intensifier hěn 很 in the predicate position, as demonstrated below.9 In this regard, adjectives behave like stative verbs in Chinese (D. Zhu 1982). (30)

(a)

他 很 高。 tā hěn gāo. he very tall ‘He is tall’

(b) 他 很 聪明。 tā hěn cōngmíng. he very smart ‘He is smart’

(c)

这本 书 很 贵。 zhè-běn shū hěn guì. this-Cl book very expensive ‘This book is expensive’

(d) 他 很 伤心。 tā hěn shāngxīn. he very sad ‘He is sad’

In addition, similar to verbs, adjectives in the predicate position can be negated by bù 不 ‘not’: (31)

(a)

他 不 高。 tā bù gāo. he not tall ‘He is not tall’

(b)

252

他 不 聪明。 tā bù cōngmíng. he not smart ‘He is not smart’

Chinese: Parts of speech

(c)

这本 书 不 贵。 zhè-běn shū bú guì. this-Cl book not expensive ‘This book is not expensive’

(d)

他 不 伤心。 tā bù shāngxīn. he not sad ‘He is not sad’

2.3.2.  Distinguishing adjectives from verbs In spite of these similarities, there is a need for distinguishing adjectives from verbs and treating adjectives as an independent category (D. Zhu 1982; C.-T. J. Huang, Li and Li 2009: Ch. 1; Paul 2010). Specifically, D. Zhu (1982) notes that adjectives crucially differ from verbs in that they cannot subcategorize for an internal argument or an object, as evidenced by the ill-formedness of the following examples. In this regard, adjectives behave on a par with intransitive verbs in being one-place predicates. (32)

(a)

*他 *tā he

高 gāo tall

(b)

*他 *tā he

聪明 cōngmíng smart

(c)

*这本 *zhè-běn this-Cl

(d)

*他 *tā he

那个 nà-ge that-Cl

书 shū book

伤心 shāngxīn sad

男孩。 nánhái. boy

那个 nà-ge that-Cl

学生。 xuésheng. student

贵 guì expensive 这件 zhè-jiàn this-Cl

那本 nà-běn that-Cl

书。 shū. book

事。 shì. matter

Despite the fact that adjectives are akin to intransitive verbs in not being able to take an object, adjectives should not be subsumed under intransitive verbs, since adjectives show different reduplication patterns from verbs in Chinese (D. Zhu 1956, 1980, 1982; C. N. Li and Thompson 1981: Ch. 3; Lü 1984; C.-T. J. Huang, Li and Li 2009: Ch. 1; Paul 2010). Specifically, while disyllabic adjectives of the AB form can undergo reduplication to give rise to the AABB pattern, as in (33a–d), disyllabic verbs of the AB form typically give rise to the ABAB pattern after reduplication, as demonstrated in (34a–d). (33) (a) gānjìng 干净 ‘clean’ → gāngānjìngjìng 干干净净 ‘thoroughly clean’ (b) lǎoshi 老实 ‘honest’ → láolǎoshishi 老老实实 ‘honest and frank’ (c) jiǎndān 简单 ‘simple’ → jiánjiǎndāndān 简简单单 ‘rather simple’ (d) hútu 胡涂 ‘muddleheaded’ → húhútutu 糊胡涂涂 ‘rather muddleheaded’ (34) (a) qǐngjiào 请教 ‘inquire’ → qǐngjiàoqǐngjiào 请教请教 ‘inquire a little’ (b) pīpíng 批评 ‘criticize’ → pīpíngpīpíng 批评批评 ‘criticize a little’ (c) yánjiū 研究 ‘research’ → yánjiūyánjiū 研究研究 ‘research a little’ (d) tǎolùn 讨论 ‘discuss’ → tǎolùntǎolùn 讨论讨论 ‘discuss a little’ Apart from the fact that adjectives and verbs show different reduplication patterns, they also differ in terms of interpretational properties (Paul 2010). Specifically, reduplication of adjectives often yields a higher degree of liveliness or intensity (D. Zhu 1956; Chao 1968: 209; 253

Candice Chi-Hang Cheung

Tang T.-C. 1988), while reduplication of verbs often gives rise to the so-called ‘tentative aspect’ according to Chao (1968: 204). The different reduplication patterns and interpretational properties thus constitute another piece of evidence for distinguishing adjectives from verbs. Additional evidence for analyzing adjectives as an independent category comes from the fact that adjectives are composed of predicative and non-predicative adjectives (Y. Li 1996; Aoun and Li 2003; Paul 2005, 2010) just like those in the Indo-European languages.10 If adjectives were subsumed under intransitive verbs, one would expect that all adjectives can be used predicatively on a par with intransitive verbs, contrary to fact. Some illustrative examples of non-predicative adjectives are given below (see Aoun and Li 2003; Paul 2005, 2010 for more examples of non-predicative adjectives): (35)

(a)

*这条 *zhè-tiáo this-Cl Intended:

(b)

*他们的 目标 很 共同。 *tāmen-de mùbiāo hěn gòngtóng. they-Poss goal very common Intended: ‘Their goals are common’

道路 很 主要。 dàolù hén zhǔyào. road very main ‘This road is main’

Finally, it is worth mentioning that both predicative and non-predicative adjectives can function as attributive adjectives when they precede de 的 and a noun, as illustrated below (Aoun and Li 2003; Paul 2005, 2010; see also C. C.-H. Cheung 2012): (36)

(a)

聪明 cōngmíng smart ‘intelligent

(37)

(a)

主要 的 zhǔyào de main DE ‘main roads’

的 学生 de xuésheng DE student students’ 道路 dàolù road

(b)

贵 guì expensive ‘expensive

(b)

共同 gòngtóng common ‘common

的 书 de shū DE book books’

的 目标 de mùbiāo DE goal goals’

2.4. Adverbs Adverbs typically occur between the subject and a verb or an adjective, as shown in the following examples. The presence of the verb or adjective is mandatory, as adverbs, unlike verbs and adjectives, cannot serve as predicates:11 (38)

(a)

他 tā he ‘He

刚 *(来)。 gāng *(lái). just arrive just arrived’

(b)

他 tā he ‘He

常常 *(跳舞)。 chángcháng *(tiàowǔ). often dance dances often’

(c)

他 tā he ‘He

非常 *(高)。 fēicháng *(gāo). extremely tall is extremely tall’

(d)

他 tā he ‘He

真 *(聪明)。 zhēn *(cōngmíng). truly smart is really smart’

254

Chinese: Parts of speech

2.4.1.  Classification of adverbs: sentence-level adverbs vs. VP-level adverbs Adverbs in Chinese can be roughly divided into two types: sentence-level adverbs and VP-level adverbs (C. N. Li and Thompson 1981: Ch. 8; Paul, forthcoming). The former may appear before or after the subject, whereas the latter can only appear after the subject.12

2.4.1.1. Sentence-level adverbs Among the sentence-level adverbs, they can be further divided into time adverbs and attitude adverbs, both of which can immediately precede or follow the subject (C. N. Li and Thompson 1981: Ch. 8), with only a few exceptions (Paul, forthcoming). 2.4.1.1.1. Time adverbs  Typical examples of time adverbs include jīntiān 今天 ‘today’, qùnián 去年 ‘last year’, zànshí 暂时 ‘temporarily’, jiānglái 将来 ‘(in the) future’, xiànzài 现在 ‘now’, xiàwǔ 下午 ‘(in the) afternoon’, among many others. Time adverbs can either occur in the sentence-initial position or immediately after the subject, as demonstrated below: (39)

(a)

今天 他 很 高兴。 jīntiān tā hěn gāoxìng. today he very happy ‘He is very happy today’

(b) 他 今天 很 高兴。 tā jīntiān hěn gāoxìng. he today very happy ‘He is very happy today’

(40)

(a)

去年 他 没有 回来。 qùnián tā méiyǒu huílai. last.year he not.have return ‘He did not return last year’

(b) 他 去年 没有 回来。 tā qùnián méiyǒu huílai. he last.year not.have return ‘He did not return last year’

(41)

(a)

暂时 他 住 在 这里。 zànshí tā zhù zài zhèli. temporarily he live at here ‘He lives here temporarily’

(b) 他 暂时 住 在 这里。 tā zànshí zhù zài zhèli. he temporarily live at here ‘He lives here temporarily’

The time adverbs in the above examples clearly function as sentential adverbs, as they signal the time at which or during which the event described by the sentence occurs. 2.4.1.1.2. Attitude adverbs  Attitude adverbs denote the speakers’ attitude toward or evaluation of the event described by the sentence. Typical examples of attitude adverbs include xiǎnrán 显然 ‘obviously’, yěxǔ 也许 ‘perhaps’, dàgài 大概 ‘probably’, xìngkuī 幸亏 ‘fortunately’, dāngrán 当然 ‘of course’, tūrán 突然 ‘suddenly’, qíshí 其实 ‘in fact’, among many others. The following examples show that they can occur in the sentential-initial position or immediately after the subject on a par with time adverbs: (42)

(a)

显然 他 不 喜欢 你。 xiǎnrán tā bù xǐhuān nǐ. obviously he not like you ‘Obviously, he does not like you’ 255

(b) 他 显然 不 喜欢 你。 tā xiǎnrán bù xǐhuān nǐ. he obviously not like you ‘Obviously, he does not like you’

Candice Chi-Hang Cheung

(43)

(a)

也许 他 yéxǔ tā perhaps he ‘Perhaps he

(44)

(a)

大概 他 dàgài tā probably he ‘Probably he

(b) 他 也许 tā yéxǔ he perhaps ‘Perhaps he

会 来。 huì lái. will come will come’ 是 shì be has

感冒 gǎnmào cold caught a

会 来。 huì lái. will come will come’

了。 (b) 他 大概 是 le. tā dàgài shì Prt he probably be cold’ ‘Probably he has

感冒 gǎnmào cold caught a

了。 le. Prt cold’

However, as noted by Paul (forthcoming), a few attitude adverbs, especially those containing shuō 说 ‘speak’, such as lǎoshí shuō 老实说 ‘frankly speaking’, tǎnbái shuō 坦白说 ‘honestly speaking’, huàn jù huà shuō 换句话说 ‘in other words’, among others, are confined to the sentence-initial position, as evidenced by the contrast between (45a) and (45b). (45)

(a)

老实说， 我 lǎoshí-shuō, wǒ frankly-speaking I ‘Frankly speaking, I do

(b)

*我 老实说 不 wǒ lǎoshí-shuō bù I frankly-speaking not Intended: ‘Frankly speaking,

不 喜欢 他。 bù xǐhuān tā. not like him not like him’ 喜欢 他。 xǐhuān tā. like him I do not like him’

2.4.1.2. VP-level adverbs As for VP-level adverbs, they can be subdivided into manner and non-manner adverbs, both of which occur between the subject and the verb (C. N. Li and Thompson 1981: Ch. 8). 2.4.1.2.1.  Manner adverbs  Manner adverbs serve to modify the verb phrase by describing the manner in which the action depicted by the verb phrase is carried out. The majority of the manner adverbs is derived from adjectives through the addition of the suffix -de 地, such as xìngfèn-de 兴奋地 ‘excitedly’, yánlì-de 严厉地 ‘sternly’, kuàilè-de 快乐地 ‘happily’, zǐxì-de 仔细地 ‘meticulously’, jiǎndān-de 简单地 ‘simply’, zhèndìng-de 鎭定地 ‘calmly’, etc. As mentioned earlier, they are confined to the position between the subject and the verb, as illustrated in (46a–c). (46)

(a)

他 tā he ‘He

兴奋地 跑 xīngfèn-de pǎo excitedly run ran in excitedly’

(b)

他 tā he ‘He

严厉地 责备了 yánlì-de zébèi-le sternly reproach-Perf sternly reproached me’

(c)

他们 快乐地 跳舞。 tāmen kuàilè-de tiàowǔ. they happily dance ‘They danced happily’

进来。 jìnlái. enter 我。 wǒ. me

256

Chinese: Parts of speech

For many monosyllabic and disyllabic adjectives, reduplication applies before they are suffixed by -de 地, such as mànmàn-de 慢慢地 ‘slowly’, jìngjìng-de 静静地 ‘quietly’, kuàikuài-de 快快地 ‘quickly’, qīngqīng-de 轻轻地 ‘softly’, rènrènzhēnzhēn-de 认认真真地 ‘seriously’, shūshūfufu-de 舒舒服服地 ‘comfortably’, gāogāoxìngxìng-de 高高兴兴地 ‘happily’, qīngqīngchǔchǔ-de 清清楚楚地 ‘clearly’, etc., as shown in (47a–c). In this respect, the Chinese suffix -de 地 behaves like the -ly suffix in English, which can turn an adjective into an adverb (e.g. slow → slowly, quiet → quietly, etc.). (47)

(a)

他 tā he ‘He

慢慢地 回答 我的 问题。 mànmàn-de huídá wǒ-de wèntí. slowly answer I-Poss question slowly answered my question(s)’

(b)

他 tā he ‘He

静静地 躺在 草地 jìngjìng-de tǎng-zài cǎodì quietly lie- at grass quietly lay on the grass’

(c)

他 tā he ‘He

会 认认真真地 huì rènrènzhēnzhēn-de will seriously will consider this problem

上。 shang. on

考虑 这个 kǎolǜ zhè-ge consider this-Cl seriously’

问题。 wèntí. problem

Furthermore, there is a subset of disyllabic manner adverbs, which can be optionally followed by -de 地 (D. Zhu 1982): (48)

(a)

他们 赶紧(地) 往 屋里 tāmen gánjǐn(-de) wǎng wū-li they rapidly toward house-inside ‘They rapidly ran inside the house’

(b)

树叶 在 shùyè zài leave at ‘The leaves

(c)

他们 认真(地) 讨论了 这个 tāmen rènzhēn(-de) tǎolùn-le zhè-ge they seriously consider-Perf this-Cl ‘They seriously considered this problem’

微 wēi light slowly

跑。 pǎo. run

风中 悠悠(地) 落 fēng-zhōng yōuyōu(-de) luò wind-in slowly drop fluttered down in the light wind’

下。 xia. down

问题。 wèntí. problem

2.4.1.2.2.  Non-manner adverbs  As for non-manner adverbs, they are mainly composed of VP-level temporal adverbs, such as yǐjīng 已经 ‘already’, zǎo 早 ‘long ago’, cháng 常 ‘often’, gāng 刚 ‘just’, mǎshàng 马上 ‘immediately’, among many others (see C. N. Li and Thompson 1981: Ch. 8; D. Zhu 1982; Yip and Rimmington 1997: Ch. 14, 2004: Ch. 9.1.2).13 Unlike sentence-level time adverbs, VP-level temporal adverbs are restricted to the position between the subject and the verb, as illustrated below:

257

Candice Chi-Hang Cheung

(49)

(a)

他 tā he ‘He

已经 离开了。 yǐjīng líkāi-le. already leave-Perf has already left’

(b)

他 tā he ‘He

早 zǎo long.ago knew that

(c)

他 tā he ‘He

常 来 这里。 cháng lái zhèli. often come here often comes here’

(d)

他 tā he ‘He

刚 回来。 gāng huílai. just return just returned’

知道了。 zhīdào-le. know-Perf long ago’

2.5. Prepositions Prepositions are one of the most poorly defined categories in Chinese partly due to the fact that many of the prepositions have their historical origins as verbs (C. N. Li and Thompson 1974a; D. Zhu 1982; Norman 1988; Heine et al. 1991; and Hopper and Traugott 1993) and partly due to the fact that Chinese is a morphologically impoverished language and there is no inflectional morphology to mark prepositions or verbs. Furthermore, some prepositions have characteristics of verbs (Chao 1968; C. N. Li and Thompson 1974a, 1974b, 1981; D. Zhu 1982). For these reasons, some scholars consider prepositions as a ‘mixed’ category containing verbs and prepositions (McCawley 1992), while others maintain that prepositions form an independent category distinct from verbs (C. N. Li and Thompson 1974a, 1974b; see also C.-T. J. Huang, Li and Li 2009). This section is not intended to resolve the debate as to whether prepositions are best regarded as a mixed or independent category. Instead, it will focus on the basic properties of prepositions and their categorization (section 2.5.1), followed by a discussion of the verbal properties of a subset of prepositions (section 2.5.2), seeking to provide a better understanding of the characteristic features of prepositions and the underlying factors that give rise the debate on the categorial status of prepositions in Chinese.

2.5.1.  Basic properties of prepositions and categorization In Chinese, prepositions (also referred to as ‘coverbs’ in C. N. Li and Thompson 1974a, b, 1981) are typically followed by a noun phrase and most of the prepositions and the following noun phrase appear between a subject and a verb phrase, as visualized in the generalized schema below (C. N. Li and Thompson 1974a, 1974b, 1981): (50) Subject + Preposition + Noun Phrase + Verb (+ Noun Phrase) According to C.-T. J. Huang, Li and Li (2009: Ch. 1), there are mainly four classes of words that are considered as prepositions in the literature: (51) (a) zhìyú 至于 ‘as regards’, guānyú 关于 ‘concerning’, etc. (b) cóng 从 ‘from’, duì 对 ‘toward’, etc. (c) gěi 给 ‘to/for’, zài 在 ‘at’, etc. (d) bǎ 把, bèi 被, jiào 叫, ràng 让, etc.14 All four classes of prepositions can be followed by a noun phrase, as shown in (52)–(55). However, as noted by C.-T. J. Huang, Li and Li (2009), the class of prepositions in (51a) requires the preposition and the following noun phrase to precede the subject (see also 258

Chinese: Parts of speech

C. N. Li and Thompson 1981), as demonstrated in (52a–b), unlike those in (51b–d), which can appear after the subject, as illustrated in (53)–(55). (52)

(53)

(54)

(55)

(a)

至于 zhìyú as.regards ‘As regards

(b)

关于 guānyú concerning ‘Concerning

(a)

他 tā he ‘He

从 北京 带 回来 很多 礼物。 cóng Běijīng dài huílai hěnduō lǐwù. from Beijing bring back many present brought back many presents from Beijing’

(b)

他 tā he ‘He

对 duì toward does not

(a)

他 tā he ‘He

给 学生 写过 不少 推荐 gěi xuésheng xiě-guo bùshǎo tuījiàn for student write-Exp not.few recommendation wrote quite a few recommendation letters for students’

(b)

他 tā he ‘He

在 zài at put a

(a)

他 tā he ‘He

把 花瓶 打破 bǎ huāpíng dǎ-pò BA vase hit-break broke the vase’

(b)

那个 警察 neì-ge jǐngchá that-Cl policeman ‘That policeman was

其他 问题, 我们 以后 qítā wèntí, wǒmen yǐhòu other question we later other questions, we will talk about

再 说。 zài shuō. again speak them later’

这件 事, 我 没有 意见。 zhè-jiàn shì, wǒ méiyǒu yìjiàn. this-Cl matter I not.have opinion this matter, I do not have an opinion’

这件 事 没有 意见。 zhè-jiàn shì méiyǒu yìjiàn. this-Cl matter not.have opinion have an opinion on this matter’

桌子 zhuōzi table book on

上 放了 shang fàng-le top put-Perf the table’

一本 yī-běn one-Cl

信。 xìn. letter

书。 shū. book

了。 le. Prt

被 流氓 打伤 了。 bèi liúmáng dǎ-shāng le. BEI hooligan hit-wound Prt wounded by some hooligans’

Interestingly, as noted by C.-T. J. Huang, Li and Li (2009), the classes of prepositions in (51b–c) and the following noun phrase can also appear in the pre-subject position like those in (51a), as in (56)–(57), while the class of prepositions in (51d) cannot, as evidenced by the ill-formedness of (58). (56)

(a)

从 cóng from ‘From

北京, 他 带 回来 很多 礼物。 Běijīng, tā dài huílai hěnduō lǐwù. Beijing he bring back many present Beijing, he brought back many presents’ 259

Candice Chi-Hang Cheung

(57)

(58)

(b)

对 duì toward ‘On this

(a)

给 gěi for ‘For

学生, xuésheng, student students, he

(b)

在 zài at ‘On

桌子 上, zhuōzi shang, table top the table, he put

(a)

*把 *bǎ BA

花瓶, huāpíng, vase

(b)

*被 *bèi BEI

流氓, liúmáng, hooligan

这件 zhè-jiàn this-Cl matter, he

事, 他 没有 意见。 shì, tā méiyǒu yìjiàn. matter he not.have opinion does not have an opinion’

他 写过 不少 推荐 tā xiě-guo bùshǎo tuījiàn he write-Exp not.few recommendation wrote quite a few recommendation letters’

他 tā he

他 放了 tā fàng-le he put-Perf a book’ 打破 dǎ-pò hit-break

那个 neì-ge that-Cl

一本 yī-běn one-Cl

信。 xìn. letter

书。 shū. book

了。 le. Prt

警察 jǐngchá policeman

打伤 dǎ-shāng hit-wound

了。 le. Prt

While the two classes of prepositions in (51b–c) behave alike in that the prepositions and the following noun phrase can either appear in the pre-subject or post-subject position, the two classes differ in that a subset of the prepositions in (51c) can form complex verbal predicates such as sòng-gěi 送给 ‘give-to’ and fàng-zài 放在 ‘put-at’, as in (59), whereas those in (51b) cannot (C.-T. J. Huang, Li and Li 2009). (59)

(a)

他 tā he ‘He

送给了 我 一件 sòng-gěi-le wǒ yī-jiàn give-to-Perf me one-Cl gave me a sweater’

(b)

他 tā he ‘He

把 bǎ BA put a

毛衣。 máoyī. sweater

书 放在 桌子 shū fàng-zài zhuōzi book put-at table book on the table’

上。 shang. top

In sum, we have seen that all four classes of prepositions must be followed by a noun phrase. In this respect, they behave like prepositions in the Indo-European languages. In addition, the preposition and the following noun phrase form a syntactic unit that serves to modify the verb phrase.

2.5.2.  Verbal nature of prepositions Despite the fact that prepositions in Chinese can take a noun phrase as its object on a par with those in the Indo-European languages, many prepositions share some properties of verbs due to their historical transition from verbs to prepositions. In the following, we shall consider two properties that show the verbal nature of prepositions. 260

Chinese: Parts of speech Table 15.1  The list of prepositions that can be followed by -zhe 着 àn 按 āi 挨 cháo 朝 chòng 冲 duì 对 kào 靠

píng 凭 shùn 顺 xiàng 向 wǎng 往 wèi 为 yán 沿

‘according to’ ‘adjacent to’ ‘facing’ ‘facing’ ‘toward’ ‘on the strength of ’

‘depend on, according to’ ‘along’ ‘facing’ ‘toward’ ‘for’ ‘along’

2.5.2.1.  Compatibility with aspect markers One of the properties shared by verbs and many prepositions in Chinese is that they are compatible with aspect markers. For instance, it has been noted that a subset of prepositions can be followed by the durative aspect marker -zhe 着 (see C. N. Li and Thompson 1981: Ch. 6.2 for detailed discussion of the durative aspect marker -zhe 着) similar to verbs (Chao 1968, and C. N. Li and Thompson 1974a, 1974b, 1981). The representative prepositions that can be followed by -zhe 着 are given in Table 15.1 (adapted from C. N. Li and Thompson 1981: 361). Some illustrative examples showing that the prepositions in Table 15.1 can be followed by -zhe 着 are given below. As noted by C. N. Li and Thompson (1981: Ch. 9), the presence of -zhe 着 does not contribute to the meaning of the sentence, as witnessed from the fact that the following examples have the same meanings with or without -zhe 着: (60)

(a)

你们 得 按(着) nǐmen déi àn(-zhe) you must according.to -Dur ‘You must speak in order’

(b)

他 tā he ‘He

朝(着) cháo(-zhe) toward-Dur walked over

(c)

他 tā he ‘He

对(着) 我 duì(-zhe) wǒ toward-Dur me nodded to me’

我 走 wó zǒu me walk to me’

次序 cìxù order

发言。 fāyán. speak

过来。 guòlái. come

点了 diǎn-le nod-Perf

点 diǎn nod

头。 tóu. head

The fact that -zhe 着 is devoid of durative meaning when it occurs with the above prepositions suggest that they have undergone grammaticalization into prepositions and are on the verge of losing their verbal properties. Additional support for this view comes from the fact that the list of prepositions that can take -zhe 着 is fairly restricted and must be memorized (C. N. Li and Thompson 1981: Ch. 9), as evidenced by the fact -zhe 着 cannot be added freely to any prepositions: (61)

(a)

*他 *tā he

从着 cóng-zhe from-Dur

北京 Běijīng Beijing

带 dài bring

回来 huílai back

261

很多 hěnduō many

礼物。 lǐwù. present

Candice Chi-Hang Cheung

(b)

*他 *tā he

给着 gěi-zhe for-Dur

学生 xuésheng student

写 xiě write

推荐 tuījiàn recommendation

(c)

*他 *tā he

在着 zài-zhe at-Dur

图书馆 túshūguǎn library

看 kàn read

书。 shū. book

信。 xìn. letter

In addition to -zhe 着, a few prepositions can be followed by -le 了 akin to verbs (Chao 1968, and C. N. Li and Thompson 1981): (62)

他 tā he ‘For

为了 你的 事 一 wèi-le nǐ-de shì yī for-Perf you-Poss matter one your matter, he didn’t sleep for the

整 晚 zhéng wǎn whole night whole night’

都 dōu all

没 méi not.have

睡。 shuì. sleep

Despite their apparent similarities, whether -le 了 still marks the perfective aspect when it occurs with a preposition is far from clear when compared with true verbs (cf. (25)). The fact that a limited set of prepositions is compatible with -le 了 should thus be regarded as evidence for their historical transition from verbs to prepositions.

2.5.2.2.  Prepositions functioning as verbs The verbal nature of prepositions can also be reflected from their dual categorial status: many prepositions in Chinese can be used as prepositions as well as verbs (Chao 1968, and C. N. Li and Thompson 1974a, b, 1981), as shown in Table 15.2 (adapted from C. N. Li and Thompson 1981: 368–9). The following examples illustrate that the prepositions in Table 15.2 can be used as prepositions (63a, 64a, 65a) as well as verbs (63b, 64b, 65b). (63)

(64)

(a)

我们 wǒmen we ‘We will

(b)

是不是 有 人 在 按 shì-bu-shì yǒu rén zài àn be-not-be have person Prog press ‘Is someone ringing the doorbell?’

(a)

他 tā he ‘He

(b)

他的 房子 tā-de fàngzi he-Poss house ‘His house faces

会 按 你的 意思 办 huì àn nǐ-de yìsī bàn will according.to you-Poss idea do do this matter according to your ideas’

朝 我 笑了 cháo wǒ xiào-le facing me smile-Perf smiled at me’

笑。 xiào. smile

朝 海。 cháo hái. face sea the sea’ 262

门铃？ mén-líng? door-bell

这件 zhè-jiàn this-Cl

事。 shì. matter

Chinese: Parts of speech Table 15.2  Prepositions functioning as prepositions and as verbs Prepositions

As prepositions

As verbs

āi 挨 àn 按 bǐ 比 chǎo 朝 chéng 乘 chòng 冲 chúle 除了 dàitì 代替 dāng 当 dào 到 duì 对 gěi 给 gēn 跟 jiě 解 kào 靠 nì 逆 píng 凭 shùn 顺 tì 替 wǎng 往 xiàng 向

‘next to’ ‘according to’ ‘compare’ ‘facing’ ‘take advantage of ’ ‘facing’ ‘except, besides’ ‘in place of ’ ‘in front of ’ ‘to’ ‘to’ ‘to, for’ ‘with’ ‘from’ ‘on the strength of ’ ‘against’ ‘depend on, according to’ ‘along’ ‘in place of ’ ‘facing’ ‘facing’

‘be next to’ ‘press’ ‘compare’ ‘face’ ‘ride on’ ‘face’ ‘remove’ ‘take the place of ’ ‘serve as’ ‘arrive’ ‘face’ ‘give’ ‘follow’ ‘untie, relieve’ ‘lean against, depend on’ ‘be opposed to’ ‘depend on’ ‘follow’ ‘substitute for’ ‘face’ ‘face’

(65)

(a)

他 tā he ‘He

(b)

春天 到了。 chūntiān dào-le. spring arrive-Perf ‘Spring has come’

常 到 cháng dào often to often goes to

北京 Běijīng Beijing Beijing’

去。 qù. go

While many prepositions have verbal counterparts, there are quite a few prepositions that lack verbal counterparts, such as zhìyú 至于 ‘as regards’, guānyú 关于 ‘concerning’, cóng 从 ‘from’, bǎ 把, bèi 被, among others. Given the fact that many prepositions show verbal properties and have dual categorial status, a natural question that arises is whether prepositions in Chinese are best analyzed as a ‘mixed’ category consisting of verbs and prepositions. This is a plausible view that has been explored in the literature (McCawley 1992). An alternative is to maintain that prepositions form an independent category distinct from verbs and attribute the verbal properties of a subset of prepositions to their historical development from verbs (C. N. Li and Thompson 1974a, 1974b). On this view, prepositions should be distinguished from their verbal counterparts, which are best analyzed as homophonous verbs. An advantage of this view is that it can provide a natural explanation for the verbal properties of a subset of prepositions while maintaining a clear distinction between prepositions and verbs. 263

Candice Chi-Hang Cheung

2.6. Conjunctions Similar to conjunctions in the Indo-European languages, conjunctions in Chinese are function words in that they form a closed class and have grammatical rather than content meaning. In Chinese, conjunctions can be divided into two major types: those that function to conjoin phrases (section 2.6.1) and those that serve to conjoin clauses (section 2.6.2).

2.6.1.  Phrasal conjunctions For conjunctions that serve to conjoin phrases, they can be further divided into three types, depending on the types of phrases they conjoin. They include (i) conjunctions that serve to conjoin nouns or noun phrases, (ii) conjunctions that serve to conjoin adjectives or adjectival expressions and (iii) conjunctions that serve to conjoin verbs or verb phrases.

2.6.1.1.  Conjunctions conjoining nouns In Chinese, there are four conjunctions that serve to conjoin nouns or noun phrases. They are hé 和 ‘and’ (Cheung et al. 1994: 46), gēn 跟 ‘and, with’, tóng 同 ‘and, with’ and yǔ 与 ‘and, with’ (Yip and Rimmington 2004: Ch. 22.1), as illustrated below: (66)

(a)

我的 学生 和 他的 学生 都 wǒ-de xuésheng hé tā-de xuésheng dōu I-Poss student and he-Poss student all ‘My students and his students are Shanghaiese’

(b) 北京 Běijīng Beijing ‘Beijing

跟 香港 都 gēn Xiānggǎng dōu and Hong.Kong all and Hong Kong are both

上海 Shànghǎi Shanghai

是 很 值得 去 看 shì hěn zhìdé qù kàn be very worth go see places worth seeing’

(c)

他 同 妹妹的 关系 很 tā tōng mèimei-de guānxi hěn he and sister-Poss relationship very ‘The relationship between him and his sister

(d)

他 tā he ‘He

对 法国的 duì Fǎguó-de to France-Poss is interested in the

是 shì be

人。 rén. person

的 地方。 de dìfang. DE place

不 错。 bú cuò. not bad is quite good’

文化 与 历史 都 wénhuà yǔ lìshǐ dōu culture and history all culture and history of France’

很 hén very

感 gǎn feel

兴趣。 xìngqù. interest

Despite the fact that all four conjunctions have the meaning of ‘and’, they differ in terms of register, i.e. some are more formal than the others (Lü 1980).

2.6.1.2.  Conjunctions conjoining adjectives As for conjunctions that function to conjoin adjectives or adjectival expressions, they include simplex conjunctions that involve a single word like ér 而 ‘also’ (Yip and Rimmington 2004: Ch. 22.1) and complex ones that involve more than one word, such as yòu 又  .  .  .  yòu 又 ‘both  .  .  .  and  .  .  .’ (Cheung et al. 1994: 427), etc. Both types of conjunctions can conjoin adjectives or adjectival expressions in the predicative (67a, 68a) as well as attributive positions (67b, 68b). 264

Chinese: Parts of speech

(67)

(68)

(a)

他 tā he ‘He

坚定 而 jiāndìng ér steadfast and is steadfast and

(b)

他 tā he ‘He

是个 认真 而 shì-ge rènzhēn ér be-Cl serious and is a serious and rigorous

(a)

这个 西瓜 又 大 又 zhè-ge xīguā yòu dà yòu this-Cl watermelon also large also ‘This watermelon is large and sweet’

(b)

他 tā he ‘He

是 shì be is a

个 ge Cl smart

勇敢。 yónggǎn. brave brave’ 严谨 的 yánjǐn de rigorous DE philosopher’

又 聪明 又 yòu cōngmíng yòu also smart also and diligent student’

哲学家。 zhéxuéjiā. philosopher

甜。 tián. sweet

勤奋 qínfèn diligent

的 de DE

学生。 xuésheng. student

2.6.1.3.  Conjunctions conjoining verbs We have seen that yòu 又  .  .  .  yòu 又 ‘both  .  .  .  and  .  .  .’ can be used to conjoin adjectives. In addition to adjectives, it can serve to conjoin verbs or verb phrases to emphasize the coexistence of the two situations depicted by the verb phrases (Cheung et al. 1994: 427), as demonstrated below: (69)

(a)

他们 对 他 又 打 tāmen duì tā yòu dǎ they to him also hit ‘They hit and scolded him’

(b)

他们 又 唱 又 tāmen yòu chàng yòu they also sing also ‘They sang and danced, and

又 yòu also

骂。 mà. scold

跳, 非常 高兴。 tiào, fēicháng gāoxìng. dance extremely happy were extremely happy’

Another conjunction used to conjoin verbs is bìng 并 ‘also’. Unlike yòu 又  .  .  .  yòu 又 ‘both  .  .  .  and’, which is often used to conjoin intransitive verbs, bìng 并 is used to conjoin transitive verbs that share the same object (Yip and Rimmington 2004: Ch. 22.1): (70)

(a)

他们 讨论 tāmen tǎolùn they discuss ‘They discussed

(b)

大家 dàjiā everyone ‘Everyone

并 通过了 这个 bìng tōngguò-le zhè-ge also pass-Perf this-Cl and passed this project’

都 同意 并 dōu tóngyì bìng all agree also agreed and supported

计划。 jìhuà. project

拥护 我的 yōnghù wǒ-de support I-Poss my proposal’

265

提议。 tíyì. proposal

Candice Chi-Hang Cheung

In (70a), zhè-ge jìhuà 这个计划 ‘this project’ is the shared object of tǎolùn 讨论 ‘discuss’ and tōngguò 通过 ‘pass’, both of which are transitive verbs. In (70b), the shared object is wǒ de tíyì 我的提议 ‘my proposal’.

2.6.2.  Clausal conjunctions Most of the conjunctions in Chinese are clausal conjunctions. They serve to conjoin two clauses and display three characteristic features. First, many of the clausal conjunctions occur in pairs and in some cases, one of the elements in the pair can be optionally omitted (C. N. Li and Thompson 1981: Ch. 23; Yip and Rimmington 2004: Ch. 22.2), as shown in (71). (71) (a) búdàn 不但  .  .  .  érqiě 而且 (b) jíshǐ 即使  .  .  .  yě 也  .  .  . (c) níngkě 宁可  .  .  .  yěbù 也不 (d) (rúguǒ) (如果)  .  .  .  jiù 就 (e) suīrán 虽然  .  .  .  dànshì 但是  .  .  . (f) wúlùn 无论  .  .  .  dōu 都  .  .  . (g) (yàoshì) (要是)  .  .  .  jiù 就  .  .  . (h) ( yīnwéi) (因为)  .  .  .  suǒyǐ 所以  .  .  . (i) zhǐyǒu 只有  .  .  .  cái 才  .  .  .

‘not only  .  .  .  but also  .  .  .’ ‘even though  .  .  .  still  .  .  .’ ‘would rather  .  .  .  and not  .  .  .’ ‘if  .  .  .  then  .  .  .’ ‘although  .  .  .  but  .  .  .’ ‘no matter what  .  .  .  still  .  .  .’ ‘if  .  .  .  then  .  .  .’ ‘because  .  .  .  therefore  .  .  .’ ‘only if  .  .  .  then  .  .  .’

There are also some clausal conjunctions that are used individually (C. N. Li and Thompson 1981: Ch. 23; Yip and Rimmington 2004: Ch. 22.2), as in (72). (72) (a) búguò 不过  .  .  . (b) .  .  .  de shíhòu 的时候 (c) kěshì 可是  .  .  . (d) yǐqián 以前  .  .  .

‘but  .  .  .’ ‘when  .  .  .’ ‘but  .  .  .’ ‘before  .  .  .’

Some illustrative examples of the two types of clausal conjunction are given below: (73)

(a)

他 tā he ‘He

不但 骂 búdàn mà not.only scold not only swore at

(b) 我 宁可 不 wǒ níngkě bù I rather not ‘I would rather not (c)

(如果) (rúguǒ) if ‘If he is

(d)

大家 dàjiā everyone ‘Although

人, rén, person people,

而且 还 érqiě hái but.also in.addition but also hit them’

打 dǎ hit

人 rén person

呢。 ne. Prt

喝 酒， 也 不 能 酒 后 驾 车。 hē jiǔ, yě bù néng jiǔ hòu jià chē. drink wine and not can wine after drive car drink than drive after drinking’

他 喝醉了, tā hē-zuì-le, he drink-drunk-Perf drunk, we will take him

我们 wǒmen we home’

虽然 很 累, suīrán hěn lèi, although very tired everyone was tired, their 266

就 jiù then

送 sòng take

他 tā he

但是 心情 都 dànshì xīnqíng dōu but mood all mood was cheerful’

回 huí return 很 hěn very

家。 jiā. home 愉快。 yúkuài. cheerful

Chinese: Parts of speech

(e)

无论 wúlùn no.matter ‘No matter

(f)

(要是) 明天 不 下雨, 我 (yàoshì) míngtiān bú xiàyǔ, wǒ if tomorrow not rain I ‘If it doesn’t rain tomorrow, I’ll definitely

(g) (因为) (yīnwéi) because ‘Because

(74)

你 ní you how

走得 多 快, zǒu-de duō kuài, go-Res how fast fast you go, he can still

他 腿部 受了 tā tuǐbù shòu-le he leg bear-Perf he had a leg injury, he

他 都 能 tā dōu néng he still can catch up’ 就 jiù then go’

一定 yídìng definitely

伤, 所以 shāng, suóyǐ wound therefore did not take part in

赶得 gǎn-de catch-Res

上。 shàng. up

去。 qù. go

没 参加 比赛。 méi cānjiā bǐsài. not.have join match the match’

(h)

只有 乐观, 你的 zhíyǒu lèguān, nǐ-de only.if optimistic you-Poss ‘Only by being optimistic could

(a)

他 tā he ‘He

(b)

我 回来 的 wǒ huílai de I return DE ‘When I returned

(c)

来, 可是 没 赶上 飞机。 我 本来 想 早   点 wǒ běnlái xiáng záo  diǎn lái, kěshì méi gǎn-shàng fēijī. I originally think early a.little come but not.have chase-ascend plane ‘I had originally intended to come earlier, but I didn’t catch the plane’

(d)

我 离开 以前, wǒ líkāi yǐqián, I leave before ‘Before I left home,

不 是 很 bú shì hěn not be very is not smart, but

病 才 能 恢复得 快。 bìng cái néng huīfù-de kuài. sickness then can recover-Res quick you be able to have a speedy recovery’

聪明, 不过 考上 大学 cōngmíng, búguò kǎo-shàng dàxué smart but exam-ascend university he passed the university entrance exam’

时候, shíhòu, time home, my

了。 le. Prt

妈妈 已经 睡着了。 māma yǐjīng shuì-zháo-le. mother already sleep-fall-Perf mother was already asleep’

弟弟 还 dìdì hái younger.brother still my younger brother still

没 回来。 méi huílai. not.have return had not returned’

Second, some of the clausal conjunctions have monosyllabic and disyllabic variants depending on the rhythmic requirement (Yip and Rimmington 2004: Ch. 22.2): (75) (a) dànshì 但是 ‘but’ (b) suīrán 虽然 ‘although’ (c) rúguǒ 如果 ‘if’ (d) yīnwéi 因为 ‘because’

→  →  →  → 

dàn 但 ‘but’ suī 虽 ‘although’ rú 如 ‘if’ yīn 因 ‘because’

267

Candice Chi-Hang Cheung

The following examples illustrate the use of the monosyllabic variants: (76)

(a)

你的 建议 很 nǐ-de jiànyì hén you-Poss suggestion very ‘Your suggestion is good, but

好, 但 时机 还 hǎo, dàn shíjī hái good but time still the time is still not ripe’

(b)

文章 wénzhāng article ‘Although

却 很 有 què hěn yǒu yet very have it is very forceful’

(c)

你 如 有 困难, 我 可以 帮助 nǐ rú yǒu kùnnán, wǒ kěyǐ bāngzhù you if have problem I can help ‘If you have a problem, I can help you’

(d) 因 yīn because ‘Because

虽 短, suī duǎn, although short the article is short,

年代 niándài age it was in

不 bù not

成熟。 chéngshú. ripe

力。 lì. force

久远, 这件 事 jiúyuǎn, zhè-jiàn shì remote.past this-Cl matter the remote past, there is no way

你。 nǐ. you 已经 yǐjīng already to check

无法 考察。 wúfǎ kǎochá. no.way check this matter’

Third, for clausal conjunctions that occur in pairs, the first member of the pair may appear before or after the subject (Yip and Rimmington 2004: Ch. 22.2; see also Y. Shi 1986: section 3.7; McCawley 1988; Gasde and Paul 1996; Tsao 1996; Zhou 2002: Ch. 5; N. N. Zhang 2009), as demonstrated below: (77)

(78)

(a)

要是 yàoshì if ‘If you

你 不 舒服, 就 别 nǐ bù shūfu, jiù bié you not well then not are not well, then don’t come’

来 lái come

了。 le. Prt

(b)

你 要是 不 舒服, 就 别 nǐ yàoshì bù shūfu, jiù bié you if not well then not ‘If you are not well, then don’t come’

来 lái come

了。 le. Prt

(a)

虽然 suīrán although ‘Although

他 身軆 不 tā shēntǐ bù he body not he wasn’t well, he

好, hǎo, good rarely

但是 很少 dànshì hěnshǎo but rarely requested leave’

请假。 qǐng-jià. request-leave

(b)

他 虽然 身軆 不 tā suīrán shēntǐ bù he although body not ‘Although he wasn’t well, he

好, hǎo, good rarely

但是 很少 dànshì hěnshǎo but rarely requested leave’

请假。 qǐng-jià. request-leave

Finally, it is worth mentioning that clauses can be conjoined without the presence of any explicit clausal conjunctions in Chinese (C. N. Li and Thompson 1981: Ch. 23; Yip and Rimmington 2004: Ch. 22.2), as shown below: 268

Chinese: Parts of speech

(79)

(a)

(如果) 你 (rúguǒ) nǐ if you ‘If you go, I’ll

去, 我 (就) qù, wǒ (jiù) go I then go with you’

(b)

他 刚 想 出去 (的时候), 忽然 下起 大雪 来 了。 tā gāng xiǎng chūqù (de shíhòu), hūrán xiàqǐ dà-xué lái le. he just want go.out DE time suddenly fall heavy-snow come Prt ‘When he was just thinking of going out, it suddenly began to snow heavily’

跟 gēn with

你 nǐ you

去。 qù. go

3.  Categories uniquely available in Chinese In this section, I turn to discuss the three categories that are available in Chinese, but not in the Indo-European languages, namely, classifiers, localizers and sentence-final particles.

3.1. Classifiers Classifiers (also referred to as ‘numeral classifiers’) are a salient feature of classifier languages like Chinese. They are regarded as a semi-lexical category due to the fact that many classifiers are historically derived from nouns. In our discussion of count and mass nouns in sections 2.1.1.1–2.1.1.2, I have implicitly assumed that sortal classifiers should be differentiated from measure words. However, whether sortal classifiers ought to be distinguished from measure words remains a point of contention in the literature. In fact, two traditional views can be identified in the literature: one tradition does not differentiate between sortal classifiers and measure words. For instance, Chao (1968: 584–620) regards classifiers as individual measures and subsumes them under the rubric of ‘measure words’. A similar view is held by C. N. Li and Thompson (1981: 106), who state that ‘any measure word can be a classifier’. One piece of evidence in support of this view comes from the fact that both sortal classifiers (e.g. wèi 位, běn 本, tiáo 条 and duǒ 朵) and measure words (e.g. chǐ 尺 ‘foot’, bàng 磅 ‘pound’, jiālún 加仑 ‘gallon’ and gōngjīn 公斤 ‘kilo’) can appear between a numeral and a noun: (80)

(81)

(a)

一位 客人 yī-wèi kèrén one-Cl guest ‘a guest’

(b)

三本 书 sān-běn shū three-Cl book ‘three books’

(c)

两条 蛇 liáng-tiáo shé two-Cl snake ‘two snakes’

(d)

十朵 花 shí-duǒ huā ten-Cl flower ‘ten flowers’

(a)

一尺 yī-chǐ one-foot ‘one foot

(b)

三磅 sān-bàng three-pound ‘three pounds

(c)

两加仑 liǎng-jiālún two-gallon ‘two gallons

(d)

十公斤 shí-gōngjīn ten-kilo ‘ten kilos of

布 bù cloth of cloth’ 汽油 qìyóu petrol of petrol’

269

糖 táng sugar of sugar’

米 mǐ rice rice’

Candice Chi-Hang Cheung

The other tradition holds the view that measure words are distinguishable from sortal classifiers. An observation that is indicative of such a distinction comes from the fact that sortal classifiers tend to resist the insertion of de 的, while measure words can always allow the insertion of de 的 (Tai and Chao 1994; Kuo 1998; Cheng and Sybesma 1999), as shown below:15 (82)

(83)

(a)

*一位 *yī-wèi one-Cl

(c)

*两条 *liǎng-tiáo two-Cl

(a)

一尺 yī-chǐ one-foot ‘one foot

(c)

两加仑 liǎng-jiālún two-gallon ‘two gallons

的 de DE

客人 kèrén guest

的 de DE

蛇 shé snake

的 布 de bù DE cloth of cloth’ 的 汽油 de qìyóu DE petrol of petrol’

(b)

*三本 *sān-běn three-Cl

的 de DE

书 shū book

(d)

*十朵 *shí-duǒ ten-Cl

的 de DE

花 huā flower

(b)

三磅 sān-bàng three-pound ‘three pounds

(d)

十公斤 shí-gōngjīn ten-kilo ‘ten kilos of

的 糖 de táng DE sugar of sugar’

的 米 de mǐ DE rice rice’

Additional support for the distinction between sortal classifiers and measure words comes from their different semantic functions. According to Tai (1990: 312), ‘[a] classifier categorizes a class of nouns by picking out some salient perceptual properties, whether physically or functionally based, which are permanently associated with the entities named by the class of nouns; a measure word does not categorize but denotes the quantity of the entity named by a noun’. The semantic difference becomes more transparent if we compare the sortal classifier tiáo 条 in yī-tiáo shé 一条蛇 ‘a snake’ with the measure word chǐ 尺 ‘foot’, as in yī-chǐ bù 一尺布 ‘a foot of cloth’: the former picks out the salient properties of shé 蛇 ‘snake’, i.e. it has a long and cylindrical body, whereas chǐ 尺 ‘foot’ does not pick out any salient properties of ‘cloth’ nor does it categorize a particular type of cloth. Rather, it denotes the quantity. Their different semantic functions can also be witnessed from their selectional restrictions. In particular, while sortal classifiers can only be used to classify a limited and specific set of nouns, e.g. count nouns (see section 2.1.1.1), measure words can be used with a wide variety of nouns (Tai 1990). For instance, container measure words such as xiāng 箱 ‘box’ and dài 袋 ‘bag’ can be used with count nouns as well as mass nouns: (84)

(a)

一箱 yī-xiāng one-box ‘one box

书 shū book of books’

(b)

三袋 sān-dài three-bag ‘three bags

苹果 píngguǒ apple of apples’

(85)

(a)

一箱 yī-xiāng one-box ‘one box

酒 jiǔ wine of wine’

(b)

三袋 sān-dài three-bag ‘three bags

米 mǐ rice of rice’

270

Chinese: Parts of speech

In light of the aforementioned arguments, I follow the second view and maintain that sortal classifiers should be distinguished from measure words. I shall discuss sortal classifiers and measure words in the following sections.

3.1.1.  Sortal classifiers Sortal classifiers often reflect the intrinsic features of the nouns they classify. Somewhat similar to gender in the Indo-European languages, the choice of classifiers is to some extent predictable from the meaning of the noun, especially when the classifiers are encoded with specific semantic information, such as shape, animacy, etc. Let us consider some common sortal classifiers that classify nouns in terms of shape. As shown below, sortal classifies denoting shape can be divided into three subclasses, namely, those denoting long and narrow entities (86), those denoting round and oval entities (87) and those denoting entities with a flat surface (88). (86) Sortal classifiers denoting long and narrow entities: (a) zhī 支: yī-zhī xuějiā 一支雪茄 ‘a cigar’, yī-zhī làzhú 一支蜡烛 ‘a candle’, yī-zhī qiāng 一支枪 ‘a gun’, yī-zhī jiàn 一支箭 ‘an arrow’, etc. (b) gēn 根: yī-gēn zhēn 一根针 ‘a needle’, yī-gēn xiàn 一根缐 ‘a thread’, yī-gēn tóufà 一根头发 ‘a hair’, yī-gēn huǒchái 一根火柴 ‘a match’, yī-gēn xiāngcháng 一根香肠 ‘a sausage’, yī-gēn zhúzi 一根竹子 ‘a bamboo’, etc. (c) tiáo 条: yī-tiáo chóng 一条虫 ‘a worm’, yī-tiáo shé 一条蛇 ‘a snake’, yī-tiáo yú 一条鱼 ‘a fish’, yī-tiáo shéngzi 一条绳子 ‘a rope’, yī-tiáo hé 一条河 ‘a river’, yī-tiáo tuǐ 一条腿 ‘a leg’, yī-tiáo huángguā 一条黄瓜 ‘a cucumber’, yī-tiáo chuán 一条船 ‘a boat’, yī-tiáo gǒu 一条狗 ‘a dog’, etc. (87) Sortal classifiers denoting round and oval entities: (a) kē 颗: yī-kē zhēnzhū 一颗珍珠 ‘a pearl’, yī-kē zhǒngzi 一颗种子 ‘a seed’, yī-kē zhàdàn 一颗炸弹 ‘a bomb’, yī-kē zǐdàn 一颗子弹 ‘a bullet’, yī-kē shǒuliúdàn 一颗手榴弹 ‘a hand grenade’, yī-kē xīn 一颗心 ‘a heart’, yī-kē yáchǐ 一颗牙齿 ‘a tooth’, yī-kē xīng 一颗星 ‘a star’, yī-kē tang 一颗糖 ‘a candy’, etc. (b) lì 粒: yī-lì mǐ 一粒米 ‘a grain of rice’, yī-lì dòuzi 一粒豆子 ‘a bean’, yī-lì huāshēng 一粒花生 ‘a peanut’, yī-lì pútáo 一粒葡萄 ‘a grape’, yī-lì shāzi 一粒沙子 ‘a grain of sand’, etc. (c) tuán 团: yī-tuán máoxiàn 一团毛缐 ‘a ball of wool’, yī-tuán miàn 一团面 ‘a ball of dough’, yī-tuán miánhuā 一团棉花 ‘a ball of cotton wool’, yī-tuán huǒ 一团火 ‘a fire’, yī-tuán yún 一团云 ‘a dense patch of cloud’, etc. (88) Sortal classifiers denoting entities with a flat surface: (a) zhāng 张: yī-zhāng zhǐ 一张纸 ‘a piece of paper’, yī-zhāng biǎo 一张表 ‘a chart’, yī-zhāng piào 一张票 ‘a ticket’, yī-zhāng yóupiào 一张邮票 ‘a stamp’, yī-zhāng chuáng 一张床 ‘a bed’, yī-zhāng zhuōzi 一张桌子 ‘a table’, etc. (b) piàn 片: yī-piàn yèzi 一片叶子 ‘a leaf’, yī-piàn miànbāo 一片面包 ‘a slice of bread’, yī-piàn ānmiányào 一片安眠药 ‘a sleeping tablet’, yī-piàn tǔdì 一片土地 ‘a stretch of land’, yī-piàn wāngyáng 一片汪洋 ‘a vast sheet of water’, etc. (c) miàn 面: yī-miàn jìngzi 一面镜子 ‘a mirror’, yī-miàn luó 一面锣 ‘a gong’, yī-miàn qí 一面旗 ‘a flag’, etc. 271

Candice Chi-Hang Cheung

Animacy also plays a role in the choice of classifier. For instance, animate beings, including mammals, birds and insects are commonly classified by the sortal classifier zhī 只, as in (89)–(91), whereas plants are generally classified by kē 棵, as shown in (92). (89)

(90)

(91)

(92)

(a)

一只 羊 yī-zhī yáng one-Cl sheep ‘a sheep’

(b)

两只 狼 liǎng-zhī láng two-Cl wolf ‘two wolves’

(c)

三只 老虎 sān-zhī láohǔ three-Cl tiger ‘three tigers’

(d)

六只 老鼠 liù-zhī láoshǔ six-Cl mouse ‘six mice’

(a)

一只 鸟 yī-zhī niǎo one-Cl bird ‘a bird’

(b)

四只 燕子 sì-zhī yànzi four-Cl swallow ‘four swallows’

(c)

三只 麻雀 sān-zhī máquè three-Cl sparrow ‘three sparrows’

(d)

十只 鸡 shí-zhī jī ten-Cl chicken ‘ten chickens’

(a)

一只 甲虫 yī-zhī jiǎchóng one-Cl beetle ‘a beetle’

(b)

两只 蜻蜓 liǎng-zhī qīngtíng two-Cl dragonfly ‘two dragonflies’

(c)

三只 蝴蝶 sān-zhī húdié three-Cl butterfly ‘three butterflies’

(d)

五只 毛虫 wǔ-zhī máochóng five-Cl caterpillar ‘five caterpillars’

(a)

一棵 树 yī-kē shù one-Cl tree ‘a tree’

(b)

两棵 葱 liǎng-kē cōng two-Cl spring.onion ‘two spring onions’

(c)

三棵 麦子 sān-kē màizi three-Cl wheat ‘three stalks of wheat’

(d)

四棵 草 sì-kē cǎo four-Cl grass ‘four tufts of grass’

The most versatile sortal classifier is ge 个, which can be used to classify a wide range of nouns, including nouns denoting humans (93), inanimate objects (94) and abstract concepts (95). (93)

(a)

一个 人 yī-ge rén one-Cl person ‘a person’

(b)

272

两个 朋友 liǎng-ge péngyǒu two-Cl friend ‘two friends’

Chinese: Parts of speech

(94)

(95)

(c)

三个 老师 sān-ge lǎoshī three-Cl teacher ‘three teachers’

(d)

六个 医生 liù-ge yīshēng six-Cl doctor ‘six doctors’

(a)

一个 手表 yī-ge shóubiǎo one-Cl watch ‘a watch’

(b)

三个 西瓜 sān-ge xīguā three-Cl watermelon ‘three watermelons’

(c)

两个 岛 liǎng-ge dǎo two-Cl island ‘two islands’

(d)

十个 城市 shí-ge chéngshì ten-Cl city ‘ten cities’

(a)

一个 主意 yī-ge zhǔyì one-Cl idea ‘an idea’

(b)

三个 目的 sān-ge mùdī three-Cl purpose ‘three purposes’

(c)

两个 愿望 liǎng-ge yuànwàng two-Cl wish ‘two wishes’

(d)

一个 机会 yī-ge jīhuí one-Cl opportunity ‘an opportunity’

In the literature, ge 个 has been viewed as a ‘general classifier’ (also referred to as a ‘neutral classifier’), since it can be used with a wide range of nouns of different semantic properties (C.-R. Huang and Ahrens 2003). More often, however, particular sets of nouns with common characteristics or belonging to the same type are classified by more specific sortal classifiers rather than ge 个 (see Ahrens 1994). Furthermore, many nouns may be classified by two or more alternative sortal classifiers, as shown below: (96)

(a)

一个 客人 yī-ge kèrén one-Cl guest ‘a guest’

(b)

一位 客人 yī-wèi kèrén one-Cl guest ‘a guest’

(97)

(a)

一个 窗 yī-ge chuāng one-Cl window ‘a window’

(b)

一扇 窗 yī-shàn chuāng one-Cl window ‘a window’

(98)

(a)

一条 金鱼 yī-tiáo jīnyú one-Cl goldfish ‘a goldfish’

(b)

一尾 金鱼 yī-wěi jīnyú one-Cl goldfish ‘a goldfish’

273

Candice Chi-Hang Cheung

 (99)

(a)

一支 步枪 yī-zhī bùqiāng one-Cl rifle ‘a rifle’

(b)

一杆 步枪 yī-gān bùqiāng one-Cl rifle ‘a rifle’

(100)

(a)

一支 毛笔 yī-zhī máobǐ one-Cl writing.brush ‘a writing brush’

(b)

一管 毛笔 yī-guǎn máobǐ one-Cl writing.brush ‘a writing brush’

In many cases, the choice of the classifiers is determined by the register, discourse type, and the age and dialectal background of the speakers.

3.1.2.  Measure words Measure words are employed to denote the quantity or amount of the entity named by a noun. In Chinese, they can be subdivided into four types, namely, standard measures, collective measures, container measures and generic measures (Chao 1968, C. N. Li and Thompson 1981: Ch. 4.2.1, D. Zhu 1982, Yip and Rimmington 2004: Ch. 2.8, among others).

3.1.2.1. Standard measures Standard measure words express nationally or internationally accepted standards of measurement, indicating standards for weight, quantity, volume or size. Typical examples include jīn 斤 ‘catty’, dǎ 打 ‘dozen’, liǎng 两 ‘ounce’, bàng 磅 ‘pound’, gōngjīn 公斤 ‘kilo’, gōngshēng 公升 ‘liter’, chǐ 尺 ‘foot’, jiālún 加仑 ‘gallon’, etc. Some illustrative examples are given below: (101)

(a)

一斤 yī-jīn one-catty ‘a catty of

(c)

三两 sān-liǎng three-ounce ‘three ounces

(e)

四公斤 sì-gōngjīn four-kilo ‘four kilos

白菜 báicài Chinese.cabbage Chinese cabbage’ 银子 yínzi silver of silver’

苹果 píngguǒ apple of apples’

(b)

两打 liáng-dǎ two-dozen ‘two dozens

鸡蛋 jīdàn egg of eggs’

(d)

五磅 wǔ-bàng five-pound ‘five pounds

茶叶 chá-yè tea-leaf of tea’

(f )

六公升 水 liù-gōngshēng shuǐ six-liter water ‘six liters of water’

3.1.2.2.  Collective measures Collective measure words resemble collective nouns in English. In most cases, specific collective measures are used for nouns denoting humans and inanimate beings, whereas nouns denoting animate beings such as mammals and insects typically occur with qún 群 ‘a group’, as illustrated below: 274

Chinese: Parts of speech

(102)

(103)

(104)

(a)

一班 yī-bān one-bunch ‘a bunch of

年轻 人 niánqīng rén young person young people’

(b)

一队 yī-duì one-file ‘a file of

(c)

一帮 yī-bāng one-group ‘a group of

孩子 háizi child children’

(d)

一股 yī-gǔ one-gang ‘a gang of

(a)

一束 yī-shù one-bunch ‘a bunch of

花 huā flower flowers’

(b)

一串 yī-chuàn one-bunch ‘a bunch of

(c)

一挂 yī-guà one-string ‘a string of

鞭炮 biānpào firecracker firecrackers’

(d)

一笔 yī-bǐ one-sum ‘a sum of

钱 qián money money’

(a)

一群 yī-qún one-flock ‘a flock of

羊 yáng sheep sheep’

(b)

一群 yī-qún one-herd ‘a herd of

牛 niú cow cows’

(c)

一群 yī-qún one-swarm ‘a swarm of

(d)

一群 yī-qún one-swarm ‘a swarm of

蜜蜂 mìfēng bee bees’

战士 zhànshì soldier soldiers’ 土匪 túfěi bandit bandits’ 钥匙 yàoshí key keys’

蚂蚁 máyǐ ant ants’

However, in some cases, nouns denoting humans, animate beings or inanimate beings may occur with the same collective measure words. Two typical examples are qún 群 and pī 批: the former denotes ‘a crowd’ or ‘a group’, which is compatible with nouns denoting humans or animate beings such as mammals, fish and insects, as shown in (105a–f ), whereas the latter denotes ‘a group’, ‘a batch’ or ‘a pile’ and is compatible with nouns denoting humans or inanimate beings, as in (106a–d). (105)

(a)

一群 yī-qún one-crowd ‘a crowd of

(c)

一群 yī-qún one-pack ‘a pack of

(e)

一群 yī-qún one-swarm ‘a swarm of

人 rén person people’

狼 láng wolf wolves’ 蜜蜂 mìfēng bee bees’

(b)

一群 yī-qún one-crowd ‘a crowd of

(d)

一群 yī-qún one-shoal ‘a shoal of

(f )

一群 yī-qún one-swarm ‘a swarm of

275

小孩 xiǎohái child children’

鱼 yú fish fish’ 蚂蚁 máyǐ ant ants’

Candice Chi-Hang Cheung

(106)

(a)

一批 yī-pī one-group ‘a group of

人 rén person people’

(b)

一批 yī-pī one-group ‘a group of

(c)

一批 yī-pī one-batch ‘a batch of

货 huò goods goods’

(d)

一批 yī-pī one-pile ‘a pile of

学生 xuésheng student students’

文件 wénjiàn document documents’

In addition, there are specific collective measures such as duì 对 and shuāng 双, both meaning ‘a pair’, that are used with nouns, which denote humans, animate or inanimate entities that come in pairs: (107)

(108)

(a)

一对 夫妇 yī-duì fū-fù one-pair husband-wife ‘a couple’

(b)

一对 yī-duì one-pair ‘a pair of

天鹅 tiān’é swan swans’

(c)

一对 yī-duì one-pair ‘a pair of

(d)

一对 yī-duì one-pair ‘a pair of

耳环 ěr-huán ear-ring ear-rings’

(a)

一双 手 yī-shuāng shǒu one-pair hand ‘a pair of hands’

(b)

一双 眼睛 yī-shuāng yǎnjing one-pair eye ‘a pair of eyes’

(c)

两双 liǎng-shuāng two-pair ‘two pairs of

(d)

三双 sān-shuāng three-pair ‘three pairs

枕头 zhěntóu pillow pillows’

鞋 xié shoe shoes’

袜子 wàzi sock of socks’

Note that even though there are nouns that are treated in pairs in English (e.g. a pair of trousers, a pair of scissors, a pair of spectacles, etc.), they are incompatible with duì 对 or shuāng 双 in Chinese, as evidenced by the ill-formedness of (109a, 110a, 111a). Instead, they are classified by sortal classifiers, as shown in (109b, 110b, 111b). (109)

(a)

*一对/*一双 *yī-duì/*yī-shuāng one-pair/one-pair Intended: ‘a pair of

裤子 kùzi trousers trousers’

(b)

一条 裤子 yī-tiáo kùzi one-Cl trousers ‘a pair of trousers’

(110)

(a)

*一对/*一双 *yī-duì/*yī-shuāng one-pair/one-pair Intended: ‘a pair of

剪刀 jiǎndāo scissors scissors’

(b)

一把 剪刀 yī-bǎ jiǎndāo one-Cl scissors ‘a pair of scissors’

276

Chinese: Parts of speech

(111)

(a)

*一对/*一双 *yī-duì/*yī-shuāng one-pair/one-pair Intended: ‘a pair of

眼镜 (b) yǎnjìng spectacles spectacles’

一副 眼镜 yī-fù yǎnjìng one-Cl spectacles ‘a pair of spectacles’

Collective measures may also express indeterminate numbers or amounts through the use of yīxiē 一些 ‘some’ and yīdiǎn 一点 ‘a little’: the former is applicable to count and mass nouns (112a–b), whereas the latter is applicable to mass nouns (113a–b) (see section 2.1.1.2). (112)

(a)

一些 yīxiē some ‘some

(113)

(a)

一点 yīdiǎn a.little ‘a little

朋友 péngyǒu friend friends’ 酒 jiǔ wine wine’

(b)

一些 yīxiē some ‘some

(b)

一点 yīdiǎn a.little ‘a little

酒 jiǔ wine wine’ 牛奶 niúnǎi milk milk’

3.1.2.3.  Container measures Container measure words are typically used with nouns denoting food and drinks. Typical examples include bēi 杯 ‘cup’, píng 瓶 ‘bottle’, hé 盒 ‘box’, guō 锅 ‘pot’, wǎn 碗 ‘bowl’, xiāng 箱 ‘box’, guàn 罐 ‘can’, hú 壶 ‘(tea) pot’, dài 袋 ‘bag’, bāo 包 ‘packet’, among many others (Chao 1968, C. N. Li and Thompson 1981: 109, D. Zhu 1982). Some illustrative examples are given below: (114)

(a)

一杯 yī-bēi one-cup ‘a cup of

(b)

两瓶 liǎng-píng two-bottle ‘two bottles

(c)

三盒 巧克力 sān-hé qiǎokèlì three-box chocolate ‘three boxes of chocolate’

(d)

一锅 yī-guō one-pot ‘a pot of

(e)

四碗 饭 sì-wǎn fàn four-bowl rice ‘four bowls of rice’

(f )

六箱 liù-xiāng six-box ‘six boxes

茶 chá tea tea’

酒 jiǔ wine of wine’

炖肉 dùnròu stew stew’ 书 shū book of books’

Note that these container measure words can be used as nouns in their own right, in which case they are classified by sortal classifiers and some of them must be followed by the suffix -zi 子 (see C. N. Li and Thompson 1981: 42–3 for detailed discussion of -zi 子; see also Cheng 2009), as shown below: 277

Candice Chi-Hang Cheung

(115)

(a)

一个 杯子 yī-ge bēizi one-Cl cup ‘a cup’

(b)

两个 瓶子 liǎng-ge píngzi two-Cl bottle ‘two bottles’

(c)

三个 盒子 sān-ge hézi three-Cl box ‘three boxes’

(d)

一个 yī-ge one-Cl ‘a pot’

(e)

四个 碗 sì-ge wǎn four-Cl bowl ‘four bowls’

(f )

六个 箱子 liù-ge xiāngzi six-Cl box ‘six boxes’

锅 guō pot

Another type of container measure may employ body parts or enclosed areas as temporary ‘containers’. They are typically preceded by the numeral yī 一 ‘one’ and can be optionally followed by de 的, as demonstrated below (Chao 1968, C. N. Li and Thompson 1981: 111–12): (116)

(a)

一脸 yī-liǎn one-face ‘a faceful

(的) 灰 (de) huī DE dust of dust’

(b)

一屋子 yī-wūzi one-house ‘a houseful

 

(c)

一头 yī-tóu one-head ‘a headful

发 (的) 白 (de) bái fà DE white hair of grey hair’

(d)

一地 yī-dì one-floor ‘a floorful

(的) 面粉 (de) miànfěn DE flour of flour’

 

(e)

一肚子 yī-dùzi one-stomach ‘a stomachful

(f )

一桌 yī-zhuō one-table ‘a tableful

(的) 客人 (de) kèrén DE guest of guests’

(的) 气 (de) qì DE anger of grievance’

(的) 烟 (de) yān DE smoke of smoke’

Since any nouns denoting body parts or enclosed areas can serve as temporary container measures, this type of container measure forms an open class. This derivation is akin to the suffixation of -ful in English, which is applicable to nouns to derive measure words (e.g. house → houseful, head → headful, etc.)

3.1.2.4. Generic measures Generic measure words, denoting types and kinds, can be used with count nouns, mass nouns and abstract nouns. Typical examples are zhǒng 种 ‘kind’, lèi 类 ‘category’ and yàng 样 ‘type’, as demonstrated below: (117)

(a)

一种 yī-zhǒng one-kind ‘a kind of

水果 shuíguǒ fruit fruit’

(b)

278

两种 liáng-zhǒng two-kind ‘two kinds of

蔬菜 shūcài vegetable vegetables’

Chinese: Parts of speech

 

(c)

四种 酒 sì-zhóng jiǔ four-kind wine ‘four kinds of wine’

(d)

一种 yī-zhǒng one-kind ‘a kind of

病 bìng illness illness’

(118)

(a)

这类 zhè-lèi this-species ‘this species

动物 dòngwù animal of animal’

(b)

这类 zhè-lèi this-kind ‘this kind

红酒 hóng-jiǔ red-wine of red wine’

 

(c)

这类 zhè-lèi this-genre ‘this genre

问题 wèntí problem of problems’

(d)

这类 zhè-lèi this-kind ‘this kind

角色 juésè role of role’

(119)

(a)

几样 jǐ-yàng several-kind ‘several kinds

东西 dōngxi thing of things’

(b)

各样 gè-yàng different-kind ‘several kinds

 

(c)

几样 jǐ-yàng several-kind ‘several kinds

饭菜 fàn-cài rice-vegetable of food’

(d)

四样 食物 sì-yàng shíwù four-kind food ‘four kinds of food’

商品 shāngpǐn commodity of commodities’

3.2. Localizers Localizers (also referred to as fāngwèicí 方位词 in Zhu 1982: 40) express the spatial location of the entities denoted by nouns. Most of the localizers are monosyllabic and they include shàng 上 ‘up’, xià 下 ‘down’, qián 前 ‘front’, hòu 后 ‘back’, lǐ 里 ‘inside’, wài 外 ‘outside’, páng 旁 ‘side’, nèi 内 ‘inside’, etc. (Chao 1968, D. Zhu 1982: Ch. 4.4, C. N. Li and Thompson 1981: Ch. 11, Chappell and Peyraube 2008). Most of the disyllabic localizers are derived from the monosyllabic ones through the addition of suffixes such as -bian 边, -mian 面 and -tou 头 (Chao 1968, Zhu 1982: Ch. 4.4, C. N. Li and Thompson 1981: Ch. 11, Chappell and Peyraube 2008). The representative localizers are listed in Table 15.3 (adapted from D. Zhu 1982: 40–1 and C. N. Li and Thompson 1981: 391). In terms of distribution, localizers are preceded by a noun phrase, as shown in the following generalized schema (Chao 1968; see also C. N. Li and Thompson 1981: Ch. 11): (120) Noun Phrase + Localizer Some illustrative examples are given below: (121)

(a)

他 tā he ‘He

 

(b)

车子 停 在 chēzi tíng zài car stop at ‘The car is parked

在 zài at fixed

房子 里 修理 fángzi li xiūlǐ room inside repair the television inside the

电视机。 diànshìjī. television room’

学校 外边。 xuéxiào wàibian. school outside outside the school’ 279

Candice Chi-Hang Cheung Table 15.3  Localizers in Chinese Localizers

Gloss

shàng 上 xià 下 lǐ 里 wài 外 qián 前 hòu 后 páng 旁 zhōngjiān 中间 dōngbù 东部 nánbù 南部 xībù 西部 běibù 北部 nèi 内

shàngbian 上边 xiàbian 下边 lǐbian 里边 wàibian 外边 qiánbian 前边 hòubian 后边 pángbian 旁边

shàngmian 上面 xiàmian 下面 lǐmian 里面 wàimian 外面 qiánmian 前面 hòumian 后面

zuǒbian 左边 yòubian 右边 dōngbian 东边 nánbian 南边 xībian 西边 běibian 北边

zuǒmian 左面 yòumian 右面 dōngmian 东面 nánmian 南面 xīmian 西面 běimian 背面

shàngtou 上头 xiàtou 下头 lǐtou 里头 wàitou 外头 qiántou 前头 hòutou 后头

‘on top of, above’ ‘under, below’ ‘in, inside’ ‘outside’ ‘in front of ’ ‘in back of, behind’ ‘beside’ ‘in the center of ’ ‘left of ’ ‘right of ’ ‘east of ’ ‘south of ’ ‘west of ’ ‘north of ’ ‘inside’

One of the important debates on localizers is their categorial status. As noted by Chappell and Peyraube (2008), in regard to the categorial status of localizers, almost every possibility has been entertained in the literature. For instance, they have been analyzed as adjectives (Ma 1898: Ch. 3), adverbs (Lü 1947, Jingxi Li and Liu 1955), nominal suffixes (Cartier 1972), postpositions forming a discontinuous constituent with the prepositions they frequently co-occur with (Hagège 1975: 220 ff., Peyraube 1980: 53 ff.; see also Tai 1973, Ernst 1988), spatial clitics (F.-H. Liu 1998, C. Sun 2008) and even pronouns (Rygaloff 1973: 143). However, one of the dominant views is that localizers are best analyzed as a subclass of nouns (D. Zhu 1982, Y.-H. A. Li 1990, C.-T. J. Huang, Li and Li 2009: Ch. 1). One piece of evidence comes from the fact that the noun phrase and the following localizer, similar to noun phrases, can function as the subject (122a) or object (122b) (C.-T. J. Huang, Li and Li 2009: Ch. 1; see also Y. Li 1983, 2003; Y.-H. A. Li 1990). (122)

(a)

城 外 很 chéng wài hén city outside very ‘The outside of the city

 

(b)

他 tā he ‘He

去过 qù-guo go-Exp has been

美丽。 měilì. beautiful is beautiful’

城 外。 chéng wài. city outside to the outside of the city’

Another piece of evidence is that the noun phrase plus the localizer, similar to noun phrases, can be preceded by the preposition zài 在 ‘at’ (C.-T. J. Huang, Li and Li 2009: Ch. 1), as demonstrated below: 280

Chinese: Parts of speech

(123)

(a)

他 tā he ‘He

在 zài at fixed

房 里 修理 电视机。 fáng li xiūlǐ diànshìjī. room inside repair television the television inside the room’

 

(b)

他 tā he ‘He

在 zài at fixed

这个 房子 zhè-ge fángzi this-Cl room the television in

修理 电视机。 xiūlǐ diànshìjī. repair television this room’

Despite these similiarities, there are reasons for distinguishing localizers from nouns. For one thing, unlike nouns that can be preceded by a [Numeral + Classifier] sequence (see section 2.1), localizers cannot. For another, if localizers were nouns, it would wrongly predict that they can always function as the subject or object on their own without the preceding noun phrase:16 (124)

(a)

*外 很 *wài hén outside very Intended: ‘The

美丽。 měilì. beautiful outside is beautiful’

(b)

*他 去过 *tā qù-guo he go-Exp Intended: ‘He

外。 wài. outside has been outside’

Based on these reasons, I adopt the view that localizers should be regarded as an autonomous category in line with Chongxing Li (1992) and Z. Chu (1997, 2006), among others. The fact that localizers, in particular the disyllabic ones, share similar properties with nouns may be attributed to the fact that they are historically derived from nouns (Chappell and Peyraube 2008).

3.3.  Sentence-final particles Sentence-final particles are one of the most distinctive features of Chinese. As its name implies, they are used in the sentence-final position, serving various semantic and pragmatic functions. There are roughly six sentence-final particles in Chinese, including le 了, ne 呢, ma 吗, ba 吧, ou 呕 and a/ya 啊/呀 (C. N. Li and Thompson 1981: Ch. 7; see also Chao 1968, D. Zhu 1982), which I shall discuss in detail in the following sections.

3.3.1. le 了 As noted by C. N. Li and Thompson (1981: Ch. 7), the sentence-final le 了 is different from other sentence-final particles in that it can be followed by other sentence-final particles, such as a 啊, ou 呕 and the yes–no question particle ma 吗, as illustrated below: (125)

Q:

他 tā he ‘Is

 

A:

他 tā he ‘Of

是 学生 shì xuésheng be student he a student?’ 当然 是 dāngrán shì of.course be course he is!’

吗？ ma? Q 了 le Prt

啊。 a. Prt

281

Candice Chi-Hang Cheung

(126)

我 告诉 他 那件 事 wǒ gàosù tā nà-jiàn shì I tell him that-Cl matter ‘I told him about that matter’

(127)

他 去 tā qù he go ‘Did he

买 东西 mǎi dōngxi buy thing go shopping?’

了 le Prt

了 le Prt

呕。 ou. Prt

吗？ ma? Q

In terms of its semantic and pragmatic functions, it has been noted that the sentence-final le 了 can mark a change of state or indicate what is expressed is contrary to expectation (Chao 1968, C. N. Li and Thompson 1981, Ross 1995, Sybesma 1999, Soh 2009). The change of state reading becomes more transparent when we compare sentences with and without the sentence-final le 了:17 (128)

(a)

他 tā he ‘He

当了 父亲。 dāng-le fùqīn. become-Perf father became a father’

 

(b)

他 tā he ‘He

当了 父亲 了。 dāng-le fùqīn le. become-Perf father Prt is now a father. (He wasn’t one before and now he has a child)’

(129)

(a)

他 tā he ‘He

不 怕 鬼。 bú pà guǐ. not fear ghost is not afraid of ghosts’

 

(b)

他 tā he ‘He

不 bú not is no

怕 鬼 pà guǐ fear ghost longer afraid

了。 le. Prt of ghosts (and he was before)’

The examples in (128a, 129a) are simple declarative sentences, while the addition of the sentence-final le 了 in (128b, 129b) conveys a change of state. More precisely, the addition of the sentence-final le 了 in (128b) signals that the person referred to as tā 他 ‘he’ was not a father before and has become a father. In (129b), the sentence-final le 了 signals a change of state from being afraid of ghosts to no longer being afraid of ghosts. As for the contrary to expectation reading, consider the following pair of sentences with and without the sentence-final le 了 (taken from Soh 2009: 632). (130)

(a)

这片 西瓜 很 甜。 不必 zhè-piàn xīguā hěn tián. Bú bì this-Cl watermelon very sweet not need ‘This watermelon is sweet. It is not necessary to 282

加 糖。 jiā táng. add sugar add sugar’

Chinese: Parts of speech

 

(b) 这片 西瓜 很 甜 了。 不必 加 糖 了。 zhè-piàn xīguā hěn tián le. Bú bì jiā táng le. this-Cl watermelon very sweet Prt not need add sugar Prt ‘This watermelon is sweet (contrary to what one may assume/expect). It is not necessary to add sugar (contrary to what one may assume/expect)’

(130a) is a simple declarative sentence asserting that the watermelon is sweet. In contrast, the speaker uttering (130b) not only makes the same assertion, but also implies that someone in the discourse context may think otherwise and the sentence is intended to correct this wrong assumption (see C. N. Li and Thompson 1981: 263–70).

3.3.2. ne 呢 The sentence-final ne 呢 can be used in declarative and interrogative sentences. In declarative sentences, it is mainly used to indicate that the sentence is the speaker’s response to some previous claim, expectation or belief on the part of the hearer. Pragmatically, it has the effect of drawing the hearer’s attention to the information conveyed by the sentence (C. N. Li and Thompson 1981: Ch. 7; see also Chao 1968, M. Hu 1981, C.-H. C. Chu 1984, 1985a, 1985b, 1998, C.-J. W. Lin 1984, King 1986, Shao 1989, G. Wu 2005). To highlight the function of ne 呢, let us consider the following pair of sentences: (131)

(a)

他们 有 五条 tāmen yóu wǔ-tiáo they have five-Cl ‘They have five cows’

牛。 niú. cow

 

(b)

他们 有 tāmen yóu they have ‘(Listen,) they

牛 呢。 niú ne. cow Prt cows’

五条 wǔ-tiáo five-Cl have five

(131a) is a simple declarative sentence asserting that the people referred to as ‘they’ have five cows. In contrast, the speaker uttering (131b) not only makes the same assertion, but also draws the hearer’s attention to this piece