Book announced at https://linguistlist.org/issues/33.2699

EDITOR: Chu-Ren Huang
EDITOR: Yen-Hwei Lin
EDITOR: I-Hsuan Chen
AUTHOR: Yu-Yin Hsu
TITLE: The Cambridge Handbook of Chinese Linguistics
PUBLISHER: Cambridge University Press
YEAR: 2022

REVIEWER: Geoffrey Sampson

SUMMARY

This book contains thirty chapters on aspects of the linguistics of Chinese, by a total of 44 authors or co-authors. All but five of these, to judge from their names, are Chinese (or in one case Korean), and most of the contributors with Western names are affiliated with East Asian universities. Three editors work at the Hong Kong Polytechnic University, the other at Michigan State.

Very unusually for a book of this kind, there is no introduction, explaining what the editors aimed to achieve in putting the book together. Chapter 1 launches straight into a study of phonological awareness among Chinese and how it is affected by differing approaches to initial literacy teaching in different parts of the Chinese-speaking world. I shall not list all thirty chapter titles, but the chapters are grouped into four Parts: 1, “Writing System/Neuro-cognitive Processing of Chinese” (two chapters); 2, “Morpho-lexical issues in Chinese” (eight chapters); 3, “Phonetic-phonological issues in Chinese” (eight chapters); 4, “Syntax–semantics, Pragmatics, and Discourse Issues” (twelve chapters). Another unusual feature of the book is that the lead editor, Chu-Ren Huang, is listed as co-author of as many as eight of the thirty chapters, including at least one from each part.

(There is an issue about Chinese authors’ names. I take it that readers ought to know which of an author’s names is his or her surname, but probably do not need to know e.g. that Qingqing Zhao, one of the contributors here, happens to share the same surname as Yuen Ren Chao, who introduced Western linguistics to China in the early twentieth century. So this review will always give Chinese names in the English order, surname last, irrespective of how they appear in an author’s English-language publications, but will use whatever romanization is used in those publications.)

Lacking a statement by the editors themselves about how they planned this book, I expected that it would be addressed to a target readership of people who were familiar with general linguistics without necessarily knowing much or anything about the Chinese language, or who perhaps had some knowledge of Chinese but wanted to discover what are seen as interesting research questions about this language. I supposed that it would include material introducing matters such as the historical origin and development of Chinese, its relationships with other languages in its family and contacts with geographically adjacent languages, its structure at phonological and grammatical levels, sociolinguistic considerations, and so forth. In other words I imagined the book would turn out to be a rival to the Oxford Handbook of Chinese Linguistics, edited by William Wang and Chaofen Sun (2015), which I reviewed in Linguist List issue 26.4510. That turned out to be wrong. The book under review is considerably narrower but also deeper than that.

It is narrower in the sense that the range of topics covered is limited almost exclusively to the internal structure of the present-day language (and to Standard Chinese, i.e., to the variety of the Mandarin dialect which has been selected as the language to be promoted China-wide). Something is said about one specialized aspect of language history in a chapter about a syntactic issue with an inherently historical dimension, but there is no broader survey of the history of the language – there is no mention of its membership in a language-family, for instance. A chapter about sentence-final particles has interesting material on sex differences in the use of these particles, but there is no wider discussion of the social setting of the language. And the book is deeper than I expected before reading it, in the sense that the chapters are not addressed to readers who are not already knowledgeable about the special subjects they discuss. Chapter 2, for instance, is about the use of neuroimaging techniques such as electroencephalography to investigate how Chinese is processed in the brain; its opening paragraph announces that it “will review a series of neurolinguistics studies that took N400, an event-related potentials (ERPs) component to index the semantic processing, to investigate how the brain processes meaning conveyed by [various aspects of the Chinese language].” A reader who does not already know what an “ERPs component” is will find no explanation here. Readers need background knowledge even in chapters dealing with less obviously technical topics than neuroimaging. P. 76, in a chapter about the vocabulary of Chinese, discusses the Menzerath–Altmann Law, without telling us what it says – I had not heard of it. I understood the point on p. 84 about why Waseda University in Japan is called Sōdai for short only because I know how Japanese script works – a reader ignorant of Japanese would be stumped.

The book is essentially a collection of research articles intended to be read by other researchers in their respective areas, of a kind that might more commonly be found in specialized journals rather than gathered into a book. Linguistic data are introduced and discussed for the contribution they make to theoretical controversies, rather than for their own sake.

One striking feature of the contents list is that successive chapters have sometimes been set up to provide thesis–antithesis tension. So we find Chapter 19, “SVO as the Canonical Word Order in Modern Chinese”, followed by Chapter 20, “SOV as the Canonical Word Order in Modern Chinese”, and Chapter 22, “The Case for Case in Chinese”, followed by Chapter 23, “The Case without Case in Chinese”. These antitheses are not always signalled so obviously by the chapter titles. For instance, several chapters are about wordhood in Chinese. To explain: the psychologically salient units from which Chinese utterances are assembled are what in Chinese are called zì, which coincide with what linguists call morphemes, are pronounced as single syllables, and are written as single graphs or “characters” of Chinese script. (I ignore some marginal special cases for which the equation morpheme = syllable = graph does not hold.) In the Classical Chinese of the first millennium B.C. (a modified form of which continued to be the standard written language of China until only a hundred years ago), there was little temptation to recognize any intermediate units between zì and complete utterances or sentences. But in the vocabulary of the modern language, monosyllabic words have largely been replaced by two-syllable compounds (though the separate morphemes within these remain entirely recognizable, and the compounds are not grouped together as units in Chinese script); hence many linguists see modern Chinese as containing units corresponding to European “words”, distinct from morphemes, while others see “words” as an imposition of Western categories on a non-European language, pointing out that zì are often linked more loosely and variably in a wordlike Chinese compound than is the case for dictionary words of European languages. Chapters 3 and 4 essentially argue that “Chinese has words” and “Chinese doesn’t have words” respectively, though their titles are less straightforwardly opposed.

(Several places in the book compare Chinese with Vietnamese, which – although it originated from a separate language-family – has been so massively influenced by Chinese over a long period that it can be counted today as virtually another Chinese dialect; the overwhelming bulk of entries in a Vietnamese dictionary are Chinese loans. I would have been interested to see what the contributors here who argue that Western-style polymorphemic words are psychologically natural for Chinese-speakers make of recent orthographic developments in Vietnamese. In the French colonial period, the Vietnamese abandoned Chinese script for alphabetic writing, and they used hyphens to link morphemes into Western-style words. But since the end of Western influence on Vietnam in the 1970s, the hyphens have been given up, and morphemes are the only units reflected in written Vietnamese today, as they are in Chinese script. As I understand it, this change in Vietnamese usage occurred spontaneously, whereas pinyin romanization with its orthographic norms – including writing compounds as single words – has been promulgated in China by government decree. It is not clear why polymorphemic words should be more psychologically natural for one language than for the other.)

Because the chapters deal with many different specialized topics and assume prior knowledge, probably no one linguist would be competent to assess all of them adequately. At any rate, I ought to admit that I am not. I was lost when it came to the chapters on syntactic theory. Some of these use grammatical symbols such as “CP” and “DP”, which mean “Complementizer Phrase” and “Determiner Phrase” but were unfamiliar to me, and appear to assume familiarity with Chomsky (1995) – I had lost interest in generative theories of syntax well before that date. Others focus on a controversy between Generative Semantics and Lexicalism, which I do vaguely remember from my youth; but I hope my inability to follow the ins and outs of that is forgivable, because Randy Harris’s “The Linguistics Wars” (2021) suggests that the controversy is dead now (and I notice that these chapters seem to cite many publications from the 1970s and 1980s and few recent items).

One can only guess how the editors briefed their contributors. I wondered whether they were encouraged to focus on similarities rather than differences between Chinese and European languages. As more Chinese academics began to come to grips with Western linguistics at a period when generative linguists were emphasizing concepts such as Universal Grammar, it was noticeable that some of them saw their task as being to develop descriptions of the Chinese language which made it look as much as possible like just another European language. And there are many places in this book where differences between Chinese and European languages seem to be downplayed. The absence of the historical dimension is one of these, because the relationship of modern Chinese to its recorded history, one of the longest of any language in the world, is a distinctive feature of Chinese today. The modern language is continuous with its history in a way that is scarcely possible for languages written alphabetically, where sound-change will tend to make a language opaque to non-experts after only a thousand years or so. Because Chinese script is not based on the language’s sound system, and because Chinese is an isolating language without inflexional grammar, an ordinary educated Chinese person can read plenty of passages written almost three thousand years ago without difficulty (though encountering plenty of other passages which are puzzling), and many phrases and sentences from the Classical language remain alive within modern Chinese.

Again, when chapters about “wordhood” here argue that modern Mandarin has a strongly disyllabic tendency, they seem to treat the issue as merely a matter of statistics (James Myers points out, p. 50, that Chinese translation equivalents for English words are more commonly two zì than one, but this could simply be because the number N of distinct zì is limited and N-squared is much larger than N), or, like Chu-Ren Huang et al. on p. 80, they list types of Chinese compound words which have close parallels in European languages. But when people talk about the modern Chinese trend to disyllabicity, it seems to me that they are usually thinking primarily about kinds of compound words which are rare or absent in European languages: e.g., the very frequent synonym-compounds, such as Mandarin péngyǒu ‘friend’ from péng and yǒu, which in Classical Chinese were separate words each meaning ‘friend’; or reduplications, e.g. jiějiě from jiě, ‘elder sister’; or words like lǎohǔ ‘tiger’, etymologically ‘old-tiger’, where lǎo adds nothing to the meaning of hǔ, but serves to produce a disyllable. None of these types appear in the list on p. 80.

On p. 136 Dingxu Shi and Chu-Ren Huang mention that Mandarin has a few zì which might be seen as inflexional affixes, but these “are not the typical inflectional affixes marking gender, number, or tense.” “Typical” here seems to mean characteristic of European languages. Gender, number, and tense are not grammatical categories applying anywhere in Mandarin (except that nouns and pronouns for human beings can be marked as plural), and I am not sure that all three categories could be described as typical of the languages of the world in general.

In one case languages are assimilated the other way round: English is made to look more like Chinese. On p. 363 San Duanmu discusses whether the /l/ of English ‘villa’ should be analysed as part of the first syllable or the second. This is a false opposition: the /l/ is ambisyllabic, what Charles Hockett called an “interlude consonant”, belonging as much to one syllable as to the other. But that relates to a large phonological difference between Chinese and European languages: in the latter, ambisyllabic consonants are frequent, but in Chinese a consonant is always clearly assignable either to the end of one syllable or to the beginning of another.

EVALUATION

There is good stuff in this collection, particularly – but not only – in the phonetics/phonology section. Karl David Neergard and Chu-Ren Huang’s Chapter 12 presents data on relationships between the density of the phonological neighbourhood of a zì (how many other zì are just one phoneme or phonetic feature away from it) and alternative theories about syllable segmentation. There are several chapters on the phonetics and phonology of tone sandhi patterns in various Chinese dialects (that is, the ways in which the physical realization of a lexical tone is affected by the tone of an adjacent zì); I found Jie Zhang’s Chapter 14 particularly impressive. And an outstanding Chapter 18 by Caicai Zhang and William Wang uses behavioural and neural data on tone perception to shed new light on the general topic of speech perception. (Incidentally, several discussions of tone sandhi here state that Mandarin tone 3 before another tone 3 becomes identical to tone 2. But there is a long research tradition showing that the respective tones are similar but not identical – see e.g. Kratochvil 1984, Tian et al. 2022, and publications cited there.)

Unfortunately, other chapters are by no means written to a comparable standard.

Some contributors make statements which seem incompatible with obvious facts. On p. 366, San Duanmu ascribes to the distinguished phonetician John Wells the view that in English “it is hard to find minimal pairs in which there is a contrast between unstressed /ǝ/ and /ɪ/. Therefore, he proposes that truly unstressed vowels can all be transcribed as /ǝ/.” This looks incredible, since in reality there are abundant minimal pairs (sapper~sappy, cola~coley, etc.); so I checked with Wells, who assures me that Duanmu has misrepresented his position. Discussing the appearance in recent Mandarin of loans from English containing un-Chinese sounds, e.g. ‘TV’ (Mandarin has no /v/), Jun-Ren Lee and Chu-Ren Huang (pp. 9–10) suggest that one might be tempted to infer that “there is no such rule systems as phonology of a language L, nor phonological integrity. Languages have significant freedom in phonological innovation”, but this view must be rejected because it is “contrary to all existing linguistic theories”. Yet this kind of phonological innovation is perfectly common when languages are in contact: think of the un-English sounds in loans from French such as ‘garage’ with [ʒ] or ‘restaurant’ with a nasal vowel. If “existing linguistic theories” really say that such things cannot happen, then too bad for linguistic theories, surely?

A number of contributors belong to the school of linguists who see the subject as having emerged ex nihilo with Noam Chomsky’s early writings, and they represent ideas as novel which are far from so in reality. Chu-Ren Huang et al. begin Chapter 4 by writing “though there have been debates on how to define a word in Chinese” (they cite only publications co-authored by Huang in 2007 and 2012), “there has never been any challenge to the theoretical position that words are the basic lexical unit and the basic unit for grammatical operation.” What about Chih-wei Lu (1960), and other chapters in the book in which that piece appeared, or Yuen Ren Chao (1968: 136ff.)? Chao and Lu were two of the leading exponents of Western linguistics applied to Chinese in the mid-twentieth century. Paul Kratochvíl (1968: 187) called wordhood “the key problem of MSC [Modern Standard Chinese] grammar”.

On p. 230 Yen-Hwei Lin accounts for various phonological alternations in Chinese dialects as caused by a general principle of preserving phonemic contrasts: “Contrast preservation has been formalized in theoretical studies”, and there is “a body of research supporting a systemic approach to contrast preservation” (Lin cites various 21st-century publications, though without saying whether any of them discuss the Chinese language); “the relevant theoretical constraints aim to preserve distinctiveness of sound inventories … and maintain contrasts in the underlying representation of words.” Linguists since Jules Gilliéron (1918) and André Martinet (1955) have often suggested that contrast preservation is a constraint on historical sound change. Yet the history of Chinese phonology, and particularly of Mandarin phonology, has been one of repeated massive losses of high-functional-load contrasts. (See e.g. Sampson 2015.) Perhaps Lin sees some way of reconciling that history with the theories cited, but in this book the apparent contradiction is not mentioned.

In some cases, contributors are reckless about details, while writing about topics where detail is everything. Feng-hsi Liu’s Chapter 19 argues against the claim put forward by a number of linguists that Chinese has been shifting over recorded history from SVO towards SOV as its basic word order. To study this, Liu needs to analyse sentence structures in the Classical as well as the modern language. The first two Classical examples she quotes (p. 411) are extracts from the 5th-century-B.C. “Great Learning” (Dà Xué), one of the “Four Books” which are the foundation texts of Confucianism. Liu quotes the examples in order to illustrate how the word yú 於had different meanings in context, a point which is relevant at that stage of her exposition. Each passage is shown in Chinese script, in romanization, in word-for-word English glosses (as throughout the book, the glosses are for some reason separated by underline characters rather than simple spaces), and in English translation, with the word in question printed bold. However, in the first example, the gloss line omits any gloss for yú. And a reader who tries to understand what has gone wrong by matching up successive glosses with successive romanized syllables or Chinese graphs will be foiled by the glosses for the two words following yú being linked by hyphen rather than underline, ‘high-good’, as if they were a single gloss for one Chinese word. (I would suggest the gloss ‘utmost’ rather than ‘high’ for zhì 至.) A minor additional problem with this example is that the word before yú is romanized as ‘zm’ rather than zhǐ. Then with Liu’s second example things get worse. In the line of Chinese script there are two instances of a particle zhě 者, which Classical Chinese uses to mark the end of a topic in a topic–comment structure; and the English translation has no wording corresponding to the words between the two instances of zhě, though these include the word yú, which is the reason for giving the example. The Chinese looks odd – Classical Chinese topic–comment structures normally contain only one topic; so I checked with my copy of the Great Learning and found that the first zhě is a spurious insertion. The original has only the second zhě, and all the wording before it should have been translated as the (sole) topic. Again it is impossible to match up the English glosses with the Chinese words – no gloss is given for the second word, which is a genitive particle, or for the zhě which shouldn’t be there, and there are further cases of hyphens in place of underlines.

But on top of all that, it is quite unclear what difference Liu thinks she has found between the meanings of yú in the two cases. In both examples, English “at” would do very well as a gloss for yú, and Liu says nothing to suggest otherwise.

The chief responsibility for problems like those above must lie with the respective authors, though one might have hoped that the editors would have caught some of them. But the best of us can make simple typing mistakes, and another issue with the book is that much of it appears never to have been proofread. Some chapters by careful authors are free of misprints, but in most chapters they are abundant. Perhaps one could not expect Cambridge’s in-house editors to be capable of proofing a book like this, though in my experience it is routine for university presses to commission outside experts for such tasks. But in any case there are plenty of errors that could easily be spotted by a monoglot English-speaker, and in some cases an English-speaker would have been best placed to spot them. It may be unreasonable to expect a Chinese linguist who knows that ‘revolution’ is the noun from ‘revolve’ to realize that “The people of this country frequently revolve” is a poor translation for a sentence meaning that they have a lot of revolutions (p. 175) – but an English-speaking proof-reader could have queried it, whether or not the proof-reader could read the Chinese.

The caption to a photograph on p. 652 explains that it shows a geological map of ‘Los Angekes’, and Los Angekes is faithfully reproduced in the List of Figures in the prelims. Presumably the author read through his work before submitting it; one of the editors must have prepared the List of Figures from the captions in the individual chapter MSS; and a C.U.P. copy-editor has to have been involved, if only to decide matters such as type style. Yet evidently not one of these people was fazed by ‘Angekes’. (If the wrong letter had been almost anything other than K, I might have supposed that the reference was to some smaller place whose name happened to be similar to Los Angeles; but Spanish does not use the letter K.)

When it comes to passages dealing with Chinese examples, careless errors are innumerable. Again and again, in chapter after chapter, a word appearing in a line of Chinese script will be missing from the line of romanized Chinese below, or vice versa; or the words in lines of Chinese script, romanization, and English will fail to match. In a passage about noun classifiers on p. 118, the phrase 三部电话, “three telephones”, is romanized as ‘san_ju_dianhua’, with ‘ju’ in bold because it is the point of the example; but 部is pronounced ‘bù’, not ‘ju’. P. 124 has an example romanized ‘Zhangsan_reng-le_yanhe’, with ‘yanhe’ corresponding to Chinese 烟盒, and Englished as “Zhangsan threw the inkpot”. But yānhé 烟盒means ‘cigarette case’. In that instance the inconsistency does not affect the point being made (either cigarette case or inkpot would do for throwing, though the latter doubtless gives a more satisfying effect). But pp. 181–2 give four example sentences for which native speakers were asked to make acceptability judgements. The example judged worse than the others by some informants is said to mean “You may eat fruits and vegetables raw”, and in Chinese script it contains 可以 kěyǐ ‘may’, but in place of that verb the romanization line gives ‘jingchang’ and the gloss line gives ‘often’. The word 經常 jīngcháng could be translated ‘often’ or ‘regularly’, but without knowing whether the informants were judging the “may eat” version of the example or the “regularly eat” version we cannot tell what to make of their responses.

A particularly frequent error in the Chinese examples is confusion between -n and -ng, which are the only two consonants that can end a Mandarin syllable. I noticed ‘min’ for ‘ming’ 名 and ‘ming’ for ‘min’ 民, p. 78; ‘shen’ for ‘sheng’ 生, p. 82; ‘jiang’ for ‘jian’ 見, p. 140; ‘man’ for ‘mang’ 芒, p. 143. I know that some Chinese speakers are apt to confuse these sounds, but the Standard Chinese pronunciations are not in doubt.

It is difficult to take seriously the abstruse (and often confusingly expressed) theorizing which dominates many of these chapters, when the treatment of basic data is so slapdash. The overall impression given by this book is that the editors asked academics of very different levels of competence and scholarly standards whether they had anything that might do as a contribution, and when the MSS came in the editors added chapter numbers and shipped them off to Cambridge to be turned into a printed book, with minimal or no quality control. The minority of valuable contributions resemble a party of respectable out-of-towners looking for dinner who have wandered by mistake into a dive.

REFERENCES

Chao, Yuen Ren (1968). A Grammar of Spoken Chinese. University of California Press.

Chomsky, A. N. (1995). The Minimalist Program. MIT Press.

Gilliéron, J. (1918). Généalogie des mots qui designent l’abeille d’après l’ALF. Champion (Paris).

Harris, R. A. (2021). The Linguistics Wars: Chomsky, Lakoff, and the battle over deep structure (2nd edition). Oxford University Press.

Kratochvíl, P. (1968). The Chinese Language Today. Hutchinson.

Kratochvil, P. (1984). “Phonetic tone sandhi in Beijing dialect stage speech”. Cahiers de linguistique – Asie orientale 13.135–74.

Lu, Chih-wei (1960). “The status of the word in Chinese linguistics”. In P. Ratschnevsky, ed., Beiträge zum Problem des Wortes im Chinesischen. Akademie-Verlag (Berlin).

Martinet, A. (1955). Economie des changements phonétiques. Francke (Bern).

Sampson, G.R. (2015). “Article for discussion: A Chinese phonological enigma”, and “Reply to the comments”. Journal of Chinese Linguistics 43.679–91 and 740–53.

Tian, Zuoyu, et al. (2022). “Mandarin tone sandhi realization: evidence from large speech corpora”. In Proceedings of Interspeech 2022, 18–22 Sep 2022, Incheon, pp. 5273–7.

Wang, William S.-Y. and Chaofen Sun, eds (2015). The Oxford Handbook of Chinese Linguistics. Oxford University Press.

ABOUT THE REVIEWER

Geoffrey Sampson graduated in Chinese Studies from Cambridge University, and his academic career was spent partly in Linguistics and partly in Informatics, with intervals in industrial research. After retiring as professor emeritus from Sussex University in 2009, he spent several years as Research Fellow at the University of South Africa. He has published contributions to most areas of Linguistics, as well as to other subjects. His latest books are ''Voices from Early China'' (2020), and ''God Proofs'' (2022).