Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  The Oxford Handbook of Chinese Linguistics

Reviewer: Geoffrey Sampson
Book Title: The Oxford Handbook of Chinese Linguistics
Book Author: William Shi-Yuan Wang Chaofen Sun
Publisher: Oxford University Press
Linguistic Field(s): General Linguistics
Historical Linguistics
Cognitive Science
Subject Language(s): Chinese, Mandarin
Language Family(ies): Austronesian
Issue Number: 26.4510

Discuss this Review
Help on Posting
Reviews Editor: Helen Aristar-Dry


This book contains 55 chapters, many quite short, on various aspects of the linguistics of Chinese, written by a total of 66 contributors. (Many chapters are co-authored, and some contributors have written or co-written more than one chapter.) The lead editor, William Wang, worked for many years at the University of California, Berkeley, where he founded the ‘Journal of Chinese Linguistics’; since retiring from Berkeley he has worked at the Chinese University of Hong Kong. Since the death of Chao Yuen-ren (1892–1982), Wang has been the best-known name among those concerned with applying the analytic techniques of modern linguistics to the Chinese language.

The overwhelming majority of contributors here – 57 of the 66 – are themselves Chinese (judging by their names). Furthermore, two-thirds of the Chinese contributors are based in the Chinese-speaking world, whether the People’s Republic, Taiwan, Hong Kong, or Singapore, though the other one-third work in North America, Europe, or Australia. There are also chapters by a Korean and a Japanese about relationships between their languages and Chinese. Only seven contributors have European names (and three of those are based in East Asia).

A first point to make about the volume, then, is that the language to which it is devoted is being discussed far more by scholars drawn from among its own speakers than has been usual in English-language publications about Chinese. I have no figures, but I would think that up to now a large majority of scholarly publications on Chinese in European languages have been by Westerners working in Europe or North America. The considerable quantities of material written and published by Chinese and in Chinese have been little read in the West. So one important virtue of this book is that it provides a window onto the work which the Chinese have been doing on their own language.

(I felt a little less guilty about my own ignorance with respect to that body of work, when I noticed that on the evidence of this book the Chinese scholars themselves in many cases appear to interact with each other less than is usual in the West. There are a number of chapters in which the literature citations are mostly self-citations.)

The 55 chapters other than the editorial introduction are grouped into eight sections:

1: History
2: Languages and Dialects
3: Language Contact
4: Morphology
5: Syntax
6: Phonetics and Phonology
7: Sociocultural Aspects
8: Neuropsychological Aspects

Sections 1 and 2 are largely about the results of the historical processes of language change and dialect splitting, which apply to all languages, and which have converted an unknown remote ancestor of Chinese into modern Chinese with its various regional dialects, together with numerous other languages more or less distantly related to Chinese.

European linguists who are familiar with the history only of the Indo-European family sometimes imagine that other language families must be broadly comparable in nature, but these authors make it clear that this is far from true. The Indo-European family comprises about eight surviving subfamilies (Romance, Germanic, Slavonic, and so on), and it appears that after these separated from one another, until the historical period there was little contact between them, making it fairly easy to distinguish similarities between languages due to common ancestry from similarities resulting from borrowing between separate languages. Furthermore, several languages ancestral to Indo-European subfamilies were major culture languages. (The language which in different circumstances might have been called “Proto-Romance” is no hypothetical reconstruction: it is Latin, which many people study and read today.) Also, whenever Indo-European languages acquired scripts, these were always alphabetic, making it relatively easy to trace the history of sound-changes.

None of these things is true of “Sino-Tibetan”, if that is an appropriate name for the family of languages genetically related to Chinese. It has many members (George van Driem lists about 300 on pp. 141–4 here), but most of these are obscure, poorly documented languages, unwritten (at least until recently) and having small numbers of speakers. Apart from Chinese, a written language since ca 1200 BC, only Tibetan and Burmese have serious cultural significance (having been written since ca AD 900 and AD 1100 respectively). All the proto-languages are hypothetical – unless one counts Chinese itself as a proto-language relative to Mandarin, Cantonese, Hakka, and so on, but from many points of view it makes better sense to count these as dialects of a single language. Languages of the family often remained in contact after they split; and furthermore some languages which were originally unrelated or only distantly related to Chinese have been so heavily influenced by it, in structure and in vocabulary, that they appear now to be sisters rather than distant cousins. Vietnamese, for instance, looks like a slightly eccentric southern dialect of Chinese, alongside Cantonese or Hakka; but, genetically, Vietnamese belongs to the Austroasiatic family, normally taken to be quite separate from Sino-Tibetan. Chinese has never been written phonographically, though Chinese script does give weak clues to earlier pronunciations of words.

In consequence, the membership of the language-family which includes Chinese, the internal structure of that family, and the position of Chinese within the family, are all highly debatable. The default view has been that the initial split was between Chinese on one hand, and the hypothetical ancestor of all the other languages on the other; but some think that Chinese separated out several branches down from the root of the family tree. Others do not believe it possible to assign much internal structure to the tree: van Driem (p. 140 here) advocates an “agnostic” position according to which the several hundred individual member languages of the family fall into 42 groups of one or a few languages each (Chinese being one of these), but no intermediate subfamilies are recognized between the 42 groups and the family as a whole. Van Driem argues that no real evidence for an initial “Chinese v. the rest” split has ever been offered; because the term “Sino-Tibetan” is strongly associated with that hypothesis, he believes that the term should be given up. He suggests replacing it with “Trans-Himalayan family”: most of the individual languages in the family are spoken south of the Himalaya ridge, but (since Chinese is the language with by far the most speakers in the world) the majority of _speakers_ are located north and east of that range.

Evidently, Indo-European offered not a typical but a specially favourable case for the 19th-century development of historical linguistics.

Common ancestry is not the only way in which languages can be related. Many European languages are heavily influenced by Latin in ways that have nothing to do with their ultimate genetic relationship (which is distant, except in the case of the Romance languages), but stem from the role Latin played as the lingua franca for educated communication during the Middle Ages and later. Within East Asia, Chinese has played a similar role. Chapters in section 3 explore how this has affected Korean and Japanese (languages which no-one takes to be genetically related to Chinese – they may be distantly related to each other). There are also chapters on the much less well-known issue of how some non-Chinese languages have affected Chinese. Particularly interesting, to me, was Xiangdong Shi’s chapter 18 on “The Influence of Buddhist Sanskrit on Chinese”. I have often wondered, for instance, how it came about that the word ‘shi4’, which in Old Chinese was a demonstrative, “this”, turned in modern Mandarin into a copular verb, “is”. (Old Chinese had no equivalent word.) Shi believes that this development, and others like it, were largely caused by translations of Buddhist literature during the first millennium of our era which tended to imitate the grammar of their Sanskrit originals. (It seems strange, though, that religious writings could have been capable of radically reshaping such fundamental aspects of spoken language in a largely illiterate society.)

One topic which, perhaps surprisingly, is not covered in section 3 is the heavy influence of European languages on Chinese from the 19th century on. In the first place, Chinese acquired a huge number of new vocabulary items to translate Western concepts: some were coined within China, many were coined in Japan using Chinese roots and then “re-imported” into China. (Most non-Western languages have simply borrowed words for modern technology and the like from European languages directly, but Chinese has scarcely ever borrowed words in modern times, preferring to generate neologisms by forming novel compounds from native roots.) Also, linguistically perhaps a more significant development, under European influence the grammar of the written Chinese language became complex and logically-explicit in ways that were fairly alien both to the classical written language, and to modern speech. (This issue is alluded to briefly on p. 534.)

The subjects of the first three sections have been well studied over a long period, so the chapter-topics will have been largely self-defining and the editors’ main task will have been to select suitable authors to cover the respective topics. The situation changes when the book moves on to sections 4 and 5, on grammar. There is far less of an established tradition of describing Chinese grammar, which is not surprising: the grammatical features of Chinese, as a language near the isolating extreme, are much less explicit and clearcut than even those of English, let alone of more heavily inflected European languages. In consequence, chapter-topics in these sections were probably determined partly at least by what individual linguists happen to have been working on recently. For instance, Jingxia Lin did a 2011 Stanford PhD on “the encoding of motion events in Mandarin”, so she covers that topic in chapter 24 here.

Section 6, on Phonetics and Phonology, contains not just impressionistic studies of articulatory phonetics but a good deal of objective instrumental data on acoustic phonetics. As one would expect, special attention is given to tone phenomena; and a particularly interesting chapter, Yi Xu’s chapter 36, discusses interaction between lexical tones and sentential intonation patterns. (The teaching of Chinese in the West commonly ignores that issue completely.)

Section 7 deals with topics such as language reform, minority languages in China, codeswitching, and linguistic differences between the sexes. Inevitably, this “Sociocultural Aspects” section sometimes reflects political pressures. When Hongkai Sun writes (p. 543) that following adoption of the 1954 Chinese constitution “The historical vestiges of discrimination toward minority languages and writings were thoroughly erased”, I wonder how some of the many Tibetans who have been burning themselves to death in protest against the Chinese occupation might respond, if they were in a position to do so. But clearly it would be foolish to expect a member of the Chinese Academy of Social Sciences not to hew to the government line.

The final section is something of a mixed bag, including chapters on topics that were new to me, such as Hua Zhu’s chapter 53 on developmental speech and language disorders among Chinese children. Chapter 52 by Catherine McBride and others on “Developmental Dyslexia in Chinese” is particularly significant in view of claims (e.g. by Taylor and Taylor 1983: 404) that this condition is extremely rare among users of Chinese-type script. The last two chapters are about the (different) sign languages used by the deaf and dumb on Taiwan and in Hong Kong.

The coverage of the book is certainly extensive. My overall impression is that it devotes rather more attention to aspects of Chinese which parallel the properties of European languages, than to aspects which make Chinese different. With so many Chinese among the contributors, that is not unexpected. Whereas Western linguists who study Chinese tend to be most interested in things that distinguish it from European languages, many Chinese who apply linguistic analysis to their language seem to envisage their task as one of fitting Chinese in as smoothly as possible to the analytic framework developed by Western linguists largely on the basis of European languages.

In order to check whether my impression of the overall balance was more than an uncorrected prior assumption, as I began reading I listed a few leading respects which, to me, make the Chinese language “special”, and while I worked through the book I noted how far it covered these. The points were:

(i) the remarkably high incidence of homophones, which is not just unusual but seems to contradict an otherwise well-confirmed universal law of sound change (cf. Sampson 2015b);

(ii) the evidence provided by Mandarin Chinese against the idea that the sounds of a language can be uniquely resolved into a set of phonemes (Chao Yuen-ren’s classic analysis of this evidence (1934) has never been matched by work on any other language, to my knowledge);

(iii) the “classifier” system, in view of which it could be argued that all Chinese nouns are mass nouns rather than count nouns;

(iv) the difficulty of applying the notion of “word” as a unit of speech or writing to Chinese.

(On a world scale, Chinese is not unusual in using classifiers, but they are unknown in European languages. Before it became fashionable among linguists to see all human languages as manifestations of a single Universal Grammar, the question whether or not the “word” concept can usefully be applied to Chinese was a standard topic for the linguistics of Chinese – see e.g. Ratchnevsky 1960.)

Of these four issues, point (ii) is referred to in Lian-Hee Wee and Mingxing Li’s chapter 35 on “Modern Chinese Phonology” (pp. 474 and 478) – only briefly, but since phonemic analysis is so much less central to linguistics in the 21st century than it was in the mid-20th century, that is reasonable. Several contributors mention that the word unit in Chinese (point (iv)) is “a fuzzy concept” (Benjamin Tsou and Oi Yee Kwong, p. 603); but the one extended discussion of this issue, namely Chu-Ren Huang and Nianwen Xue’s chapter 26 on “Modeling Word Concepts Without Convention”, approaches it from an unexpected direction. They tell us that the International Organization for Standardization has published a standard definition of “word” (ISO 24613 of 2008) intended to apply to any and every human language, and their chapter is mainly about the problems of devising software to implement that standard for Chinese. With due respect to the ISO organization, it cannot claim (and probably would not claim) to be qualified to resolve scientific questions about linguistic structure. The ISO standard may have commercial relevance for the natural-language engineering industry, but it can hardly tell us anything new about the nature of the Chinese language.

As for points (i) and (iii): classifiers are frequently mentioned in passing (e.g. on p. 406), as they could scarcely fail to be, but I found no passage focusing on this aspect of Chinese grammar. And the homophone issue is not discussed. (Neither “homophone” nor “classifier” are entries in the index of the book.)

I conclude that my initial impression of the balance of coverage was broadly justified. But then, any reference book inevitably has its biases. The one identified here is far from extreme. For instance, some chapters which stress ways in which the Chinese language is “different” would include:

Fuxiang Wu and Yancheng He’s chapter 28, on various ways in which Chinese seems to be exceptional with respect to current theories about Greenbergian word-order universals;

Liejiong Xu’s chapter 29, on the greater frequency and variety of topic–comment structuring in Chinese syntax than in English;

Ping Chen’s chapter 30, on the idea that definiteness is not a category encoded in Chinese grammar.

I found these and other chapters very successful in promoting novel insights.


It is safe to say that the various chapters of this book amount to a more wide-ranging account of the Chinese language than has ever before been assembled in a single volume, certainly in a European language and probably at all.

A more detailed evaluation of the book requires a judgement about intended readership. Before I read the book, I took it that a typical target reader for a “Handbook” of this kind would be someone who is reasonably knowledgeable about general linguistics, but who knows little about the Chinese language in particular, or who perhaps has some ability to speak and/or read the language but has not seriously attempted to think analytically about its structure. Linguists of Chinese would also be expected to find the book useful, but they are comparatively few in numbers, so I would not have expected the book to be addressed principally to them.

If that is right, I would have to say that many readers of the former, more numerous class are likely to find the book fairly hard going.

Consider, for instance, the many Chinese names referred to in the book. In a country as vast and with such a long history as China, which has never used an alphabetic script, the student is confronted with a bewildering variety of alternative names, transliteration systems, and related cultural and linguistic references. Take one of the earliest mentions in this book of a non-Mandarin Chinese dialect, the editor William Wang’s reference on p. 24 to “Xiamen”. Xiamen is the port city traditionally called “Amoy” in English. These are alternative romanizations of the same Chinese name (it means “mansion-gate”): Amoy is its pronunciation in its own dialect, Xiamen is the Mandarin pronunciation of the same pair of roots, spelled according to the modern ‘pinyin’ scheme. Xiamen lies in Fujian province (traditionally “Fukien” in English), so Xiamen dialect would be (a subvariety of) Fujian Chinese, one of the six or seven major Chinese dialect divisions. (Fujianese is specially important linguistically, because it is believed to have split off from the ancestor of all the other modern dialects several centuries before that ancestor began to fission into Mandarin, Cantonese, and so forth around AD 600.) However, in “cultural” contexts, the two-syllable names of Chinese provinces are commonly replaced by unrelated single-syllable alternatives, often the names of the independent states which occupied roughly the same territories before the unification of the Chinese Empire in the first millennium BC; Fujian province is called “Min”, the name of the river which flows through it. Other contributions to this book, for instance Chinfa Lien’s chapter 12, refer to “Min” dialect, which is the term commonly used.

Scholars well versed in Chinese history and linguistics perforce have to take this sort of thing in their stride. But for a newcomer they are confusing. Nothing in this book, so far as I have noticed, tells the reader that “Xiamen dialect” is a synonym or hyponym of “Min dialect”, or that “Yue dialect” (the topic of Anne Yue’s chapter 13) is the same as what English-speakers (and the editors and other contributors here) call “Cantonese”, spoken in an area which includes Hong Kong and the city which is called “Canton” in English and “Guangzhou” in Chinese. In general, the many contributors to the book have been left to forge their separate tracks through the tangled jungle of Chinese nomenclature, and the reader follows as best he can.

(Things get worse in the chapters on contacts between Chinese and neighbouring languages. The reader must work out for himself, for instance, that the ‘Go’ of Japanese ‘Go-on’, “pronunciation of Go”, is the Japanese form of Chinese ‘Wu’, the “cultural” name for the province of Jiangsu, traditionally written Kiangsu in English.)

The problem relates not just to names but to the substance of the Chinese language, in connexion with cited forms from the Middle Chinese of the centuries around AD 600 or the Old Chinese of the centuries around 1000 BC. For most of my lifetime, people who had occasion to mention early spoken Chinese forms agreed in quoting the reconstructions proposed by Bernhard Karlgren (1957); for many years he had the field largely to himself (the alternative system of Li Fang-kuei (1974–75) was much less widely circulated). Karlgren’s system was certainly not the last word, and since the 1990s there has been a flourishing of new research. There is little doubt that many of the new ideas about particular aspects of Chinese historical phonology are superior to Karlgren’s, but the newer researchers disagree with one another and change their minds. For instance, it has been known for a long time that Old Chinese had a contrast between two types of syllable, sometimes called Type A and Type B, without it being at all clear how this contrast was realized phonetically. Karlgren thought the distinction lay in Type-B syllables having a medial [-j-] semivowel after their initial consonant, but there are many alternative theories. When he wrote his ‘Handbook of Old Chinese Phonology’ (1992), William Baxter argued at some length that Karlgren’s theory was correct, and Baxter transcribed words accordingly. By the time of Baxter and Sagart (2014), though, he had changed his mind; Baxter now believes that the distinction lay in A-types having pharyngealized initials, which he spells, in line with IPA norms, using a superscript symbol for the voiced pharyngeal approximant (typographically awkward – though not so awkward, admittedly, as Karlgren’s spellings).

Scholars are obviously entitled to disagree and to change their minds about debatable issues, but highly debatable matters should probably have little place in a survey book like this one. For most purposes it would surely be better for scholars who refer to early Chinese spoken forms, without themselves engaging in controversy about alternative hypotheses, to settle on some agreed scheme of transcription that identifies those contrasts on which there is consensus. (Of the reconstruction schemes currently in print, the most suitable would probably be that of Axel Schuessler (2007): his book is organized as a dictionary, making it easy to check how he spells any given word in Middle and in Old Chinese; the book is “the first linguistically sophisticated etymological dictionary of Chinese”, as remarked here by Zev Handel (p. 77); and Schuessler does not pretend to have identified more phonetic detail than is really plausible – for instance he distinguishes A-type syllables with a simple circumflex accent, on which different scholars are free to impose their own phonetic interpretations.) But, rather than following some common approach in this way, the many contributors who refer to early Chinese word-forms in this book have made independent choices of transcription scheme, and in many cases they do not identify the schemes they use. As a result, one finds the same word at the same historical period spelled in very different ways on different pages; non-experts would find it hard to know whether such differences reflect real differences, e.g. between dialects, or merely differences of scholarly opinion.

Furthermore, while some features of early Chinese are reconstructed quite reliably (thus there is no doubt at all that many Middle Chinese words ended in stops, -p -t or -k, though modern Mandarin contains no trace of these), other features proposed by individual researchers are idiosyncratic and seem quite implausible to other researchers. Some contributors here discuss forms containing highly idiosyncratic features as if they were uncontroversial. Baoya Chen and Zihe Li’s Table 8.1 quotes Old Chinese forms ending in numerals from 1 to 8 which appear to represent elements of an unusually rich tone system, yet most scholars believe that Old Chinese was not a tone language at all. (Before reading Chen and Li I had not realized that anyone thought it had tones.) Readers gain little feeling for where knowledge ends and guesswork begins.

With matters like these, contributors and editors may simply have failed to realize how challenging their material is for linguists who are not expert Sinologists. But there are also cases where it must be obvious that content is entirely opaque to anyone not deeply versed in the discipline. For instance, a leading source of information about Middle Chinese pronunciation is the various tables of rhyming words prepared for the use of poets at that period, and consequently it is traditional to refer to Middle Chinese syllable-finals (nuclear vowel or diphthong, plus final consonant if any) by the Chinese graph conventionally used in the rhyme tables to stand for words sharing that rhyme. Thus the /-â/ final is represented by the graph meaning “song” (pronounced /kâ/ in Middle Chinese). Mitsuaki Endo’s chapter 16 assumes that readers are familiar with this system. Yet even people who can read Chinese could not infer the system from modern pronunciations; the words grouped under the “song” graph do not all rhyme with one another today, and the word for “song” itself has a different vowel in modern Mandarin.

Another kind of problem is that the many non-Western contributors differ in their degrees of mastery of English. In a few cases their English is not really adequate for publication without heavy editing, which it has not received. In the case of Endo’s chapter about relationships between Japanese and Chinese languages, this is a topic which I have written about myself at an elementary level (Sampson 2015a: 208–232), yet even so I sometimes could not understand what Endo was saying. (I did understand that when he glossed Japanese ‘kurage’ as “jerry fish” (p. 219), he meant “jellyfish”.)

I learned a great deal from this book. For anyone who has enough background in Chinese linguistics not to be confused by the problems I have discussed, it will probably become one of the first volumes to consult on many of the topics covered. But it seems a pity that it has not been written so as to be accessible to a wider and less expert readership.

What would have been needed for that was a policy of heavier editorial intervention, requiring the diverse contributors to conform to consistent standards for items like names and reconstructed word-forms, and to spell out matters that are well-known to insiders but opaque to outsiders. This need not have made the book longer. Another valuable kind of editorial intervention would have been to eliminate the many repetitions of material discussed independently by different contributors. To give just one example, the traditional ‘fanqie’ system for indicating the pronunciation of a graph, in the absence of a phonographic script, by giving two other graphs of which the first shares the same initial consonant as the target and the second shares its final and tone, is described in at least three places, always as if for the first time.

However, one must recognize that the kind of editorial policy needed might have been less practical than it may seem to 21st-century Westerners, accustomed to editors who override authors’ wording preferences fairly ruthlessly. I have never forgotten my amazement, as an 18-year-old undergraduate, at the verbal contortions that my (distinguished and much older) Chinese teacher would put himself through in order to avoid telling me bluntly, and correctly, that I had made a mistake. The level of editorial intervention which would have made this a more widely-accessible book could well, in East Asian circumstances, be socially impossible.


Baxter, William H. 1992. A handbook of Old Chinese phonology. Berlin: Mouton de Gruyter.

Baxter, William H. and Laurent Sagart. 2014. Old Chinese: a new reconstruction. Oxford: Oxford University Press.

Chao Yuen-ren. 1934. The non-uniqueness of phonemic solutions of phonetic systems. Bulletin of the Institute of History and Philology, Academia Sinica, vol. iv, part 4, pp. 363–397. Reprinted in Martin Joos (ed.), Readings in Linguistics, New York: American Council of Learned Societies, 1957.

Karlgren, Bernhard. 1957. Grammatica Serica recensa. Stockholm: Museum of Far Eastern Antiquities.

Li Fang-kuei. 1974–75. Studies on Archaic Chinese [translation of 1971 Chinese original]. Monumenta Serica 31.219–287.

Ratchnevsky, Paul (ed.) 1960. Beiträge zum Problem des Wortes im Chinesischen. Berlin: Akademie-Verlag.

Sampson, Geoffrey. 2015a. Writing systems (2nd edn). Sheffield: Equinox.

Sampson, Geoffrey. 2015b. A Chinese phonological enigma. Journal of Chinese Linguistics 43.679–691.

Schuessler, Alex. 2007. ABC etymological dictionary of Old Chinese. Honolulu: University of Hawai’i Press.

Taylor, Insup and M. Martin Taylor. 1983. The psychology of reading. New York: Academic Press.
After studying classical and modern Chinese at Cambridge and Yale Universities, Geoffrey Sampson taught linguistics in the first half of his career, and computer science in the later half; he has published on many areas of both subjects. Since retiring from his British university chair, he has worked as research fellow in linguistics at the University of South Africa. A new edition of his book 'Writing Systems' was published in Britain and the USA by Equinox in 2015.

Format: Hardback
ISBN-13: 9780199856336
Pages: 792
Prices: U.K. £ 115.00