In: Journal of the Chinese Language Teachers Association (1989) XXIV.1:43-61.

CHINESE. By Jerry Norman. Cambridge: Cambridge University Press, 1988. Pp. xi, 292.
$54.50 hardcover. $17.95 paperback.

A Critical Review of Norman's Chinese

Marjorie K.M. Chan and James H.Y. Tai
Ohio State University

Back to Part A (Continued from Part A)

3.2. The scientific classification of Chinese into different regional dialects is very recent. The first such efforts were made by Fang-kuei Li in 1937, which, with only minor modifications, form the basis for the current, conventionally-accepted set of seven dialect groups: (1) Mandarin (or Northern), (2) Wu, (3) Xiang, (4) Gan, (5) Kejia (or Hakka), (6) Yue (or Cantonese), and (7) Min. The classification is based upon two phonological criteria, the first of which is mentioned by N in Chapter 8: the modern reflexes of the Middle Chinese initial voiced stops, and the second, the evolution of the so-called 'entering tone' (that is, the preservation or loss of the entering tone, and the fate of the coda on these syllables, historically closed by *-p, *-t, or *-k).

N proposes a larger subdivision of the dialects into which the seven traditional dialect groups can be placed. Thus, without actually rejecting the traditional scheme, N proposes ten diagnostic features, or criteria, each which a given dialect can be characterized as having positive (+) or negative (-) values. These features include lexical and grammatical consideration in addition to phonological ones. In fact, of N's ten criteria, only two are phonological, namely Criteria 5 and 6. Four of the criteria (1 to 4) are grammatical, involving pronouns, affixation, morphological markers and particles). And the last four (7 to 10) are common, lexical items.

Twelve dialects from the conventional classification of the seven dialect groups are selected by N to undergo the above diagnostic test. Three large groups emerge: (1) a Northern group, co-extensive with the Mandarin dialect group, have all positive values for the ten diagnostic features; (2) a Southern group, consisting of the Kejia, Yue, and Min groups in Southern China, have all negative values for these features; and (3) a Central group, composed of the remaining dialect groups -- Wu, Xiang, and Gan -- which are sandwiched between the Mandarin dialects to the north and west, and the Kejia, Yue and Min to the south, have mixed positive and negative values for the features. N considers the Northern and Southern groups to be subgroups of Chinese, whereas the central group occupies the transitional zone between the north and south, as reflected in their mix of northern and southern features, with a minimum of three or four positive values to a maximum of seven (for Changsha, which should have a '+' for Criterion 2, thereby yielding a total of seven '+' values for the dialect).

N's classification scheme raises a number of issues and questions. The criteria for the traditionally-accepted classification was completely phonologically-based, but the new set of criteria reduces the import of phonological considerations to a mere twenty percent of the total set of diagnostic features. The apparent reduction in importance is illusory, however. The particular choice of the two phonological criteria is very significant. Criterion 5 is essentially designed to isolate the Mandarin dialects, the Northern group (the only ones with positive values), while Criterion 6 separates out the Southern group of Kejia, Yue, Min dialects (the only ones with negative values). This can be seen in the summary of the results below. Criterion 5 states that "there is a register distinction only in the ping tonal category". Criterion 6 states that "velars are palatalized before i" (a historical development).

Criterion 5
Criterion 6

The intention of having two major subgroups of Chinese, a Northern and a Southern one, is immediately apparent from studying these two criteria alone. Additional phonological criteria would, in fact, be superfluous. Furthermore, the other eight lexical and grammatical features can be seen as being selected to conform with the basic division. Thus, despite a varied set of criteria, the heart of N's subgrouping is still phonologically-based, so that the number of other criteria could equally have been reduced or augmented.

Given the pivotal role of the phonological criteria, how does one classify a dialect which is considered a Mandarin dialect in the broad, traditional schema, but cannot be assigned positive values for Criteria 5 and 6? A case in point is the Yantai dialect (Qian et al. 1982). It has no register distinction for the Ping tonal category, nor does the dialect have a register distinction for the other two tones, Shang and Qu. Thus, neither '+', nor '-' is an appropriate response to Criterion 5. Moreover, in Yantai, the word for 'to stand' in Criterion 7 is li instead of zhan, as is typical of Mandarin dialects. (Hence, Yantai would receive a negative value for Criterion 7.) What is zhan 'to stand' in Mandarin has undergone a semantic shift in Yantai; the word now has the meaning of 'to play'. Given one minus value for one of the features and no appropriate response possible for Criterion 5, should Yantai be considered a member of the Northern group, or a transitional dialect of the Central group? Or more generally, the question that needs to be posed is: must the feature values be all positive for a dialect to be classified under the Northern group, and all negative to fit into the Southern group? And if so, are ten diagnostic features sufficient, and are those particular features the most suitable for determining membership? And is there any hierarchical ordering of the criteria given by N?

Moreover, note that no syntactic criteria are included among N's diagnostic features. This is due at least in part to the lack of detailed grammatical description for most of the Chinese dialects. The extent to which syntactic criteria will be useful in the future classification of the dialects remains to be investigated. The same holds true for the inclusion of semantic criteria. The question of the relative weight placed on the various types of criteria -- phonological, lexical, morphological, syntactic, and semantic -- also needs to be addressed in the future. In N's current broad classification, there are only two diagnostic features that are phonologically-based, but it is these phonological criteria that prove to be the crucial ones for determining group membership. Clearly, the type of criteria for classification and their relative importance needs to be explored further.

3.3. One of the most important issues in historical Chinese phonology that N raises in Chapter 2 concerns the theory of tonogenesis. In the past, it was assumed that a language which was tonal must always have been so since inception, and that tonal and non-tonal languages could not be genetically related. Such assumptions were challenged by the tonogenetic theory. The literature focuses specifically on the rising tone developing out of a lost final glottal stop, and a falling tone from final *-h (< *-s). The proposal was advanced by Haudricourt (1954a) in his well-known article, "De l'origine des tons en Vietnamien" where he accounts for the origin of tones in Vietnamese based on comparative evidence. (In a less widely read article, "Comment reconstruire le Chinois Archaique," Haudricourt (1954b) acknowledges Maspero's (1916) original contribution in hypothesizing *-s as the source of tones in Vietnamese.)

The impact of Haudricourt's (1954a) article extends beyond Vietnamese, because he boldly proposes that a parallel case holds for Archaic Chinese, Common Miao-Yao and proto-Thai, a proposal which was embraced by Pulleyblank (1962) in his reconstruction of Old Chinese, and subsequently by other scholars for Chinese, and other Asian as well as non-Asian languages.

For the historical phonologists, the issue of whether Chinese was once a non-tonal language remains under debate. Among the evidence for the segmental origin of tones in Chinese in the literature, N cites old Sino-Vietnamese loanwords, Chinese transcriptions of Sanskrit and other foreign words, and modern Chinese dialects, in the case of final glottal stop in ancient ascending tone syllables. Established scholars such as Fang-kuei Li (1971) did not subscribe to the theory for the Old Chinese period, but he was willing to consider the possibility that the four tones in Chinese had evolved from differences in final consonants at an earlier stage of the language. While N finds the theory of tonogenesis very attractive, he deems it overly schematic in its present form (57).

Picking up on the issue, Hombert (1978) reports on an acoustic study that he had conducted on the intrinsic effects of final laryngeals [?] and [h] on the fundamental frequency (Fo) of the preceding vowel. In that study, he used Arabic speakers producing CVC nonsense words, some ending in [?] and others ending in [h]. Leaving aside details of the experimental design and statistics, the results are significant in that Fo goes up a minimum of 9 Hz before [?] and down 25 Hz before [h]. In other words, in CVC syllables ending in a glottal stop, the pitch on the vowel goes up towards the end of the syllable, while in CVC syllables ending in an [h], the pitch goes down. Thus, the results show that [?] has a pitch-raising effect, while [h] has a pitch-lowering effect. A follow-up perceptual study simulating the effects of these two laryngeals on a preceding [i] vowel shows that the perturbations on Fo can be perceived by the auditory system. Hombert (1978:95) then concludes that "the overlap between these two sets of data validate and explain the development of rising and falling tones from the loss of [h] and [?] postvocalically."

There is, thus, experimental evidence to support the tonogenetic theory per se. It remains then to determine if synchronic and diachronic evidence, supplemented by experimental data, are sufficient to convince scholars of the segmental origin of tones in Chinese. Even given a general acceptance of the theory of tonogenesis for Chinese, there remains yet another issue to be resolved, namely the dating of that process, whether it predates Old Chinese, as Li (1971) wishes to suggest, thereby pushing the phenomenon far back into the remote past, or is a much more recent phenomenon that began in the present era during the initial half of the first millennia, as Pulleyblank (1986) contends.

3.4. As China has no indigenous tradition of grammatical study, the systematic analysis of Chinese grammar did not begin until the end of the nineteenth century. As a result, Chinese has primarily been analyzed in terms of traditional Western grammatical concepts. One of the central issues in the grammatical study in Classical Chinese as well as modern Chinese involves the analysis of word classes. Because of the lack of inflectional and derivational morphology, Chinese is extremely elusive for the analysis in terms of 'parts of speech' as form classes. Thus, Kennedy (1964:323), among others conclude that "in the final analysis word-classes cannot be defined, hence....Chinese grammar must start from different premises." N rejects Kennedy's conclusion and follows many others in defining word classes in Chinese in terms of syntactic functions. Thus, "nouns typically function as subjects or objects" (Norman, 88), and 'verbs are predicatives per [sic] excellence" (Norman, 91). N's definition of nouns as a formal word classes hinges upon the definitions of 'subject' and 'object', both of which are just as resistant to formal analysis in Chinese as 'parts of speech,' since there is neither subject-verb agreement nor subject/object case distinction in this language (except in certain pronouns in Classical Chinese). N's definition of verbs is equally unsatisfactory, for the term 'predicative' is a term of syntactic functions. Modern and Classical Chinese are rich in nominal predicatives, which contain no verbs at all. If nouns and verbs cannot be satisfactorily defined as 'parts of speech' in Chinese, other 'parts of speech' are even more problematic. The controversy over the 'parts of speech' in Chinese in the history of grammatical study of Chinese exposes the inadequacy of 'parts of speech' as universal form classes in natural languages. This is indeed an issue which Kratochvil (1968:106-115) discusses in depth. N evades the issue by applying Western traditional grammatical concepts to Chinese unquestioningly.

Admittedly, one can satisfactorily identify word classes in Chinese by means of 'parts of speech' as notional terms (cf. Lyons 1968) coupled with some morpho-syntactic features available in this language. Nevertheless, it would be more challenging to see how the two Chinese indigenous grammatical terms inspired by the structure of Classical Chinese -- shizi 'full words' and xuzi 'empty words' -- reveal the structure of Classical Chinese. N mentions this native division (88), but turns to Western 'substantives' and 'predicatives' for the first fundamental division of word classes in Classical Chinese. It is puzzling that N avoids using shici 'full words' and xuci 'empty words' to analyze word classes in Classical Chinese, and yet feels it necessary to apply them to Modern Chinese. In his words, "full words fall into six classes: nouns, verbs, adjectives, numerals, measures and pronouns" (157), while "empty word classes are adverbs, prepositions, conjunctions, particles, interjections and onomatopoeia" (158). He gives no justification for the classification. It is hard to understand why measures and pronouns are 'full words', while adverbs are 'empty words'. Kratochvil (1968) reinterprets the distinction between 'full words' and 'empty words' as that between 'words' with lexical meaning and 'markers' with grammatical meaning. His reinterpretation provides us with an insight into one of the fundamental structural principles of Chinese grammar and can further be exploited in different grammatical theories.

It is worth noting that while Kennedy never makes explicit what his 'different premises' are, Tai (1982) recently responded to Kennedy's challenge along the spirit of Categorial Grammar, by defining word classes as functional classes. A cognition-based functional distinction between nouns and verbs in Chinese is also being attempted by Tai (forthcoming).

Defining the unit of 'word' in Chinese is just as elusive as defining 'parts of speech' in terms of form classes in this language. It is no accident that, since the inception of the study of grammar in China, the definition of both 'word' and 'parts of speech' has led to many arguments among Chinese grammarians. Being aware of the controversy, N follows Chinese structuralists in defining a word in Chinese as "a unit which has a specific meaning and can be used freely" (155). If we take 'freely' to be used in a strict sense, then 'measures' in Chinese should not be treated as a word class by N, since their occurrence depends on 'numerals.' More importantly, N does not notice the dual properties of a Chinese morpheme. This can be illustrated in the ambiguity of sentences such as wo kan-le shu le. In the reading of 'I have read', the unit shu 'book' is a morpheme bounded in the word kan-shu 'read-book'; but in the reading 'I have read the book', shu is a word. This dual property of Chinese morphemes has precipitated many nerve-wrecking problems for Chinese grammarians in both traditional structuralism and modern transformationalism, because they are forced to solve phenomena in the Chinese language using Western concepts.

3.5. N's discussion of putonghua 'common language' in the context of China's bidialectalism and bilingualism is misleading in two respects. First, N simply equates putonghua with guoyu 'national language.' Historically, guoyu represents the language policy of the Nationalist Chinese in mainland China before 1949 and has been practiced in Taiwan since 1949. Geographically, it is the standard Chinese in Taiwan but not in mainland China. It differs from putonghua not only with respect to numerous officially codified pronunciations but also in vocabulary and styles, and to a lesser extent in grammar, as reported by Tai (1976,1977,1978) and others. Guoyu and putonghua are indeed two versions of standard Chinese used in two drastically different Chinese societies with minimal contact since 1949. Their mutual intelligibility notwithstanding, they are 'divided' and stand for two different socio-political entities. Modern Standard Chinese as a 'divided' language (à la Kratochvil 1970) is an intriguing sociolinguistic phenomenon which deserves further investigation. Thus, N misleads the reader, when he states that "in the city of Taibei [sic], for example, large numbers of people under forty years of age, speak only the local variety of putonghua" (251). Moreover, while N correctly observes the interesting bidialectal situation in Taipei, the parallel bidialectal situation between local Pekingese and putonghua has somehow escaped N's attention. The sociolinguistic interplay of the local vernacular and putonghua in Peking today is just as entangled as the bidialectal situation is in Taipei. In Taipei, guoyu and the local Southern Min dialect have mutually influenced each other, forming the new norm of Taiwan Mandarin (Cheng 1985). Similarly, in Peking putonghua and local Peking dialect have been shaping each other and changing the norm of putonghua in China (Hu 1987).

The second misleading aspect of N's discussion of putonghua as well as guoyu resides in his mistaking 'codified' for 'standard'. If 'standard' has any clear reference, it would be to the speech of educated speakers of Mandarin in mainland China and Taiwan. In mainland China, the broadcast Chinese by Central People's Broadcast Station is perceived as 'standard putonghua', while in Taiwan, the pronunciation of announcers in the Central Broadcast Station in Taipei is perceived as 'standard guoyu'. As reported in Tai's (1976,1977,1978) comparison of the two versions of broadcast Chinese, they are very different with respect to many aspects of pronunciation, vocabulary and styles. This shows that broadcast Chinese will play a central role in the further development of 'standard Chinese' in both mainland China and Taiwan. Yet, N has totally forgotten the role of broadcasting and other forms of mass media in shaping language development in a developing country, and in one which is undergoing rapid changes in industry and technology.

4. N's book is an excellent introduction to the Chinese language and its history. While comments on insufficient coverage of various topics have been made, it is always necessary to appreciate the difficulty of being selective in order to keep a book such as this general in nature, manageable in size, and economical in cost. The book has succeeded admirably on all three grounds. N is a highly respected scholar whose book will undoubtedly be one of the most important reference sources in English on the Chinese language for many years to come. The authors have adopted N's book as a required textbook for two courses in the Department of Asian Languages and Literatures at the Ohio State University. It was the main textbook for the Introduction to Chinese Linguistics (Chinese 680, Winter 1988), and a supplementary textbook for Studies in Chinese Historical Phonology (Chinese 882, Spring 1988). We have found N's book extremely useful and challenging to instructors and students alike. We also highly recommend N's book for general linguists who are interested in various theoretical implications of the Chinese language.


1. Precisely because of this bias, N's book is particularly useful as a textbook for a Chinese linguistics course oriented towards graduate students majoring in the Chinese language.

2. At the Ohio State University, we are in the process of compiling a glossary list for N's book.


Bloomfield, Leonard. 1933. Language. New York: Holt, Rinehart and Winston.

Chao, Yuen Ren. 1948. Mandarin Primer. Cambridge: Harvard University Press.

Chao, Yuen Ren. 1968. A Grammar of Spoken Chinese. Berkeley: University of Berkeley Press.

Chen, Matthew and William Wang. 1973. 'Sound change: actuation and implementation.' Foundations of Language 8.4:457-498.

Cheng, Chin-Chuan. 1973. A Synchronic Phonology of Mandarin Chinese. The Hague: Mouton.

Cheng, Robert. 1985. 'A Comparison of Taiwanese, Taiwan Mandarin, and Peking Mandarin.' Language 61.2:352-377.

Chomsky, Noam. 1957. Syntactic Structures. The Hague: Mouton.

Chomsky, Noam and Morris Halle. 1968. The Sound Pattern of English. New York: Harper and Row.

Ferguson, Charles A. 1964. 'Diglossia.' In Dell Hymes (ed.), Language in Culture and Society: A Reader in Linguistics and Anthropology. New York: Harper and Row.

Hartman, Lawton M. 1944. 'Segmental Phonemes of the Peiping Dialect.' Language 20:28-42.

Hashimoto, Anne Yue. 1971. 'Mandarin Syntactic Structures.' Unicorn 8. Princeton: Chinese Linguistics Project.

Haudricourt, Andre G. 1954a. 'De l'Origine des Tons en Vietnamien.' Journal Asiatique 242:68-82.

Haudricourt, Andre G. 1954b. 'Comment Reconstruire le Chinois Archaique.' Word 10:351-364.

Hockett, Charles F. 1947. 'Peiping Phonology.' Journal of the American Oriental Society 67:253-267.

Hombert, Jean-Marie. 1978. 'Consonant Types, Vowel Quality, and Tone.' In Victoria A. Fromkin, ed. Tone: A Linguistic Survey. New York: Academic Press. Pp. 77-111.

Hsueh, F.S. 1985. 'The Theoretical Basis of the Rhyme-Tables and the Meaning of their Inner/Outer Distinction.' Yuyan Yanjiu 1985.1:38-56.

Hu, Mingyang. 1987. Beijing Hua Chu Tan. Beijing: Commercial Press.

Huang, James C.-T. 1982. Logical Relation in Chinese and the Theory of Grammar. MIT dissertation.

Kahn, Daniel. 1976. Syllable-Based Generalizations in English Phonology. Ph.D. dissertation, University of Massachusetts. Distributed by the Indiana University Linguistics Club.

Kennedy, George. 1964. Selected Works of George A. Kennedy. New Haven: Far Eastern Publications.

Kratochvil, Paul. 1968. The Chinese Language Today: Features of an Emerging Standard. London: Hutchinson & Co.

Kratochvil, Paul. 1970. 'The norm and divided Chinese,' Journal of the Chinese Language Teachers' Association 8.2:62-69.

Li, Fang-kuei. 1937. 'Languages and dialects.' The Chinese Yearbook. Shanghai: Commercial Press. (Reprinted in Journal of Chinese Linguistics (1973) 1.1:1-13.)

Li, Fang-kuei. 1971. 'Shangguyin Yanjiu' (Studies on Archaic Chinese). Tsing Hua Journal of Chinese Studies, New Series 9:1-61. (English version in Monumenta Serica 31 (1971) 31:219-287, translated by G.L. Mattos.)

Li Rong. 1963. Hanyu Fangyan Diaocha Shouce (Chinese dialect investigation handbook). Peking: Kexue Chubanshe.

Li, Charles N. and Sandra A. Thompson. 1981. Mandarin Chinese: A Functional Reference Grammar. Berkeley: University of California Press.

Lyons, John. 1968. Introduction to Theoretical Linguistics. Cambridge: Cambridge University Press.

Martin, Samuel E. 1957. 'Problems of Hierarchy and Indeterminacy in Mandarin Phonology.' Bulletin of the Institute of History and Philology 29:209-229.

Maspero, Henri. 1916. 'Etudes sur la Phonetique Historique de la Langue Annamite: les Initiales.' Bulletin de l'Ecole Francaise d'Extreme-Orient 12:1-127.

McCarthy, John J. 1979. Formal Problems in Semitic Phonology and Morphology. Ph. D. dissertation, M.I.T. Distributed by the Indiana University Linguistics Club.

Pulleyblank, Edwin G. 1962. 'The Consonantal System of Old Chinese.' Asia Major, New Series 9. Part I: 58-144; Part II: 206-265.

Pulleyblank, Edwin G. 1984. Middle Chinese: A Study in Historical Phonology. Vancouver: University of British Columbia Press.

Pulleyblank, Edwin G. 1986. 'Tonogenesis as an Index of Areal Relationships in East Asia.' Linguistics of the Tibeto-Burman Area 9.1:65-81.

Qian, Zengyi et al. 1982. Yantai Fangyan Baogao (Report on the Yantai dialect). Jinan: Qilu Shushe.

Tai, James H-Y. 1976. Lexical Changes in Modern Standard Chinese in the People's Republic of China since 1949. Washington, D.C.: USIA.

Tai, James H-Y. 1977. Syntactic and Stylistic Changes in Modern Standard Chinese in the People's Republic of China since 1949. Washington, D.C.: USIA.

Tai, James H-Y. 1978. Phonological Changes in Modern Standard Chinese in the People's Republic of China since 1949. Washington, D.C.: USIA.

Tai, James H-Y. 1982. 'Relevant Categorial Distinctions in Chinese.' CLS 18:495-506. Chicago Linguistic Society.

Tai, James H-Y. Forthcoming. A Cognition-Based Functional Grammar of Chinese, Volume 1.

Wang, William S-Y. 1963. 'Some Syntactic Rules for Mandarin.' Proceedings of the Ninth International Congress of Linguistics, pp. 192-202.

Wang, William S-Y. 1965. 'Two Aspect Markers in Mandarin.' Language 41:457-70.

Wang, William S-Y. 1967. 'Phonological Features of Tone.' International Journal of American Linguistics 33.2:93-105.

Wang, William S-Y. 1969. 'Competing Changes as a Cause of Residue.' Language 45.1:9-25.

Yip, Moira. 1980. The Tonal Phonology of Chinese. Ph.D. dissertation, M.I.T. Distributed by the Indiana University Linguistics Club.

Back to Part A

Return to Publications Page.
Copyright (c) 1997 Marjorie K.M. Chan. All rights reserved.
Last update: 4 April 2004 (to convert to UTF8-encoding).