In: Studies in Chinese Historical Syntax and Morphology: Linguistic Essays in Honor of Mei Tsu-lin, [= Collection des Cahiers de Linguistique d'Asie Orientale], ed. by Alain Peyraube and Chaofen Sun. 1999. Paris: Ecole des Hautes Etudes en Sciences Sociales. Pages 223-239.
SOME REFLECTIONS ON THE PERIODIZATION
OF THE CHINESE LANGUAGE
James H-Y. Tai and Marjorie K.M. Chan
The Ohio State University
0. INTRODUCTION
For understanding major changes in the history of a language, periodization is not only desirable but also necessary. This is particularly true in the case of a language such as Chinese, which has had a long and well-documented history. However, periodization often encounters many unsettled issues, due to different schemas having been proposed for different purposes. Crucial and challenging though the task may be, it can be achieved by addressing some specific issues concerning periodization. In this paper, we examine a few central issues pertaining to the periodization of the Chinese language and propose our views for further discussion.
Section 1 presents some of our reflections on the periodization of Chinese history in order to bring to light certain issues in the task of periodization in general. In section 2, we discuss periodization of the Chinese language, examining periodization of the three major components of the language, namely, phonology, grammar, and vocabulary. In section 3, drawing on some general observations in section 2, we propose our views on the periodization of the Chinese language.
1. PERIODIZATION OF CHINESE HISTORY
The history of a civilization can be simply periodized in terms of centuries, as is often done by scholars for Western history, or in terms of dynasties, following traditional scholars of Chinese history. Alternatively, the history of a civilization can be divided into different major periods in order to find out fundamental patterns of development. The periodization of Chinese history is a good case in point. Thus, Western-educated Chinese historians would rarely just follow the chronology of the dynasties in interpreting Chinese history and civilization. Typically, two or more dynasties are grouped together into a single period. This is the case of the Qin (221-206 B.C.) and Han (206 B.C. - 220 A.D.), representing the first unification of China in which a centralized government was created and consolidated, accompanied by expansion and standardization. Although rarer, it is also sometimes desirable to split a long dynasty, by treating the first portion as the end of one period and the second portion as the beginning of another. A good case in point is the splitting of the Tang dynasty (618-907) into early Tang--which is grouped with the preceding Six Dynasties (222-581) and Sui (581-618)--and late Tang, which is grouped with the next dynasty, Song (960-1127), to form the subsequent Late Tang and Song period, the golden age of Chinese culture (cf. Reischauer and Fairbank 1960). Another case is the subdividing of the Qing dynasty (1644-1911) into two separate periods, with the Opium War of 1839-42 (and the subsequent 1850 Taiping Rebellion) signalling the demise of traditional China and the rise of a new, modern China.
Historians often differ in their conceptualization of the periodization of Chinese history, depending on their foci on significant patterns of development and change. There are at least three factors that need to be taken into consideration. The first concerns the issue of how many major periods to posit. For the period in Chinese history up to the end of the Qing dynasty, Reischauer and Fairbank, for example, posit two major periods, Classic China and premodern China. In contrast, Hucker (1978) posits three periods for that same time frame. They are: the formative age, lasting from prehistory to 202 B.C.; the early empire, from 202 B.C. to A.D. 960; and the later empire, from 960 to 1850. Hucker, in fact, excludes the last few decades of the Qing dynasty in that third period.
The second factor pertains to the issue of where to place a transitional dynasty or a portion thereof that serves both as the end of one stage and the beginning of another. For instance, while Hucker places the Qin dynasty as the end of the formative age, Reischauer and Fairbank treat the Qin dynasty as part of the first phase of early imperial China. Similarly, whereas Hucker considers the entire Tang dynasty as belonging to the end of the early empire, Reischauer and Fairbank treat late Tang as the beginning of premodern China.
The third and most important factor concerns what kinds of recurrent patterns in history to capture. For instance, the Sui and Tang dynasties form a period that can be seen as having a recurrent pattern similar to that of the Qin and Han dynasties. In each of these two periods, a short dynasty sets the stage for a long, stable, and prosperous one. Another conceptualization of recurrent patterns juxtaposes dynasties of "barbarian" (non-Han Chinese) rule with those of Han Chinese rule. A third and even more important conceptualization that captures many socio-economic and cultural recurrent patterns is to divide Chinese history politically into cyclic periods of unity and disunity. During periods of unity, China enjoyed peace, prosperity, and population growth; while periods of disunity saw wars, economic hardship, and decrease in population. Nonetheless, it was periods of disunity that witnessed the greatest intellectual and foreign influences that enriched the culture of China.
In relation to language development in general, it is significant to note that urban population grew during periods of unity, paving the way for the formation of a koiné based on the language of the political capital, which is also often the cultural and economic center of the country. In contrast, during periods of disunity, there were massive migrations of Han Chinese from north to south on the one hand, and intermarriage of Han and non-Han Chinese in the north on the other. In both cases, language contact during periods of disunity propelled language borrowing and language change.
2. PERIODIZATION AND THE CHINESE LANGUAGE
2.1. Background and Issues
Ideally, periodization of the history of a language should be based on important changes in all components of the grammar of a language; that is, its phonology, morphology, syntax, and lexicon. This is indeed the spirit that guided the periodization of the history of the English language, which is customarily divided into three major periods: Old English (450 - 1150), Middle English (1150 - 1500), and Modern English (1500 onwards). Scholars of the English language often identify Old English with the period of full inflections, Middle English with the period of levelled inflections, and Modern English with the period of lost inflections (cf. Jespersen 1982, Baugh and Cable 1993). Morphology, however, is not the only criterion for the periodization. Inflectional changes were also accompanied by fundamental phonological and syntactic changes. Phonological changes involve vowel-quality changes from Old English to Middle English, and vowel loss in endings in Modern English. Syntactic changes involve the crucial change from synthetic language in Old English to analytic in Modern English, with the emergence of fixed word order and prepositions, such as 'to' and 'for', to replace lost inflectional endings. As a result of the Norman Conquest of 1066, vocabulary underwent the most dramatic change in the transition from Old English to Middle English. The weakening of vowel endings and the decay of inflections in English were also due to language contact following the Norman Conquest.
There is general consensus on the periodization of the English language with respect not only to the number of major periods, but also to the demarcation points and the major linguistic changes that took place in each period. That does not hold in the case of the Chinese language. Traditional Chinese scholars were generally not concerned about changes in the Chinese language, and were even less concerned about its periodization. This is at least in part because of the long-established tradition of a Classical, literary language (wenyan wen), a tradition that persisted through the centuries until the 1919 May Fourth Movement. The task of subdividing the history of the Chinese language into periods had to await the early twentieth century, with the pioneering work of Bernhard Karlgren (1915-1926). In his periodization, Karlgren was specifically concerned with historical phonology. As that subdiscipline has traditionally been the main focus of historical Chinese linguistics, periodization of the history of the Chinese language has tended to center around phonology, and then to generalize from that to the periodization of the history of the Chinese language in general. Using phonology as the only criterion is obviously inadequate for periodizing the history of the language in general, as aptly observed by a number of scholars (e.g., Wang 1958, Ting 1992), who advocate basing periodization on the three main components of the Chinese language, namely, phonology, grammar (including morphology and syntax), and lexicon.
Indeed, Wang Li (1958:32-35) has proposed a periodization of the history of the Chinese language that takes these three components into consideration. He posits four main periods, with transition periods placed within parentheses:
- Shanggu Period (Early Old Chinese)
Period prior to 3rd c. A.D.
(Transition period: 3rd. and 4th c. A.D.)
- Zhonggu Period (Middle Old Chinese)
Period from 4th c. to 12th c.
(Transition period: 12th and 13c.)
- Jindai Period (Pre-Modern Chinese)
Period from 13th c. to the Opium War in the 19th c.
(Transition period: from the Opium War in 1840 to 1919 May 4th Movement) - Xiandai Period (Modern Chinese)
Period after the 1919 May 4th Movement
A closer scrutiny of Wang's criteria for his four periods reveals that, in fact, none of the four periods are characterized by all three components of the language. The first period is characterized by phonological and grammatical features; the second period primarily by grammatical features; the third period entirely by phonological features; and the fourth period by lexicon in conjunction with Western-influenced grammar. Thus, at most two of the three components serve to distinguish any given period. Moreover, grammar serves as a criterion in almost all periods, with lexicon serving as a criterion only in the fourth period and never as the sole criterion. While Wang does not intend the criteria listed for each period to be exhaustive, the bases of the periodization do reflect to some extent his view of periodization; that is, Wang (1958:34) considers grammar the most important criterion, and phonology and the lexicon to be of less importance.
The above observations raise two important questions in the periodization of the history of languages in general, and the history of the Chinese language in particular. One, do the three components--phonology, grammar, and lexicon--carry equal weight in the periodization task or is there a hierarchy among them? And two, can the periodization of the history of one component of the language be co-extensive with that of another component? For example, can phonological periodization be co-extensive with either lexical or grammatical periodization, or both? If, indeed, the periodization of the three components can be co-extensive, that should establish the most important divisions in the history of that language.
We will try to address these two questions. To do so, it is not only desirable but also necessary to examine the nature of periodization for each of the three components, namely, phonological periodization, grammatical periodization, and lexical periodization. The following three sections will discuss each case in turn, after which we will return to the two questions posed above.
2.2. Phonological Periodization
From Karlgren onwards, Chinese historical phonologists have posited a number of different schemas for periodization that can be grouped into three main modes. One mode is based on dynasties and the existing documents and sources from that period, as Wang Li (1958) has done, dividing the history of Chinese phonology into nine periods using dynasties (some combined together, as in the cases of Sui-Tang and Ming-Qing) as reference points.
The second mode is source-based and is one most commonly used among historical Chinese phonologists. Delineation of periods is accomplished by identifying key sources and documents, which are then used for proposing historical reconstructions. The pioneer using this mode for phonological periodization is Karlgren (1915-26). He posits five periods: Proto-Chinese (the oldest stage of the Chinese language reconstructible in some form that predates the earliest literary documents), Archaic Chinese (the language of the first Zhou centuries (from 1028 B.C.) based partly on the Shijing rhymes
and partly on xiesheng characters), Ancient Chinese (the language of the Qieyun of 601 A.D. which he treats as reflecting the Changan dialect of Shaanxi province under the Sui dynasty), Middle Chinese (the language of the Song rhyme tables (but never reconstructed by Karlgren)), and Old Mandarin (the language of Nanjing speech of the Ming dynasty, as reflected in the Hongwu Zhengyun (also not reconstructed by
Karlgren)).
Within this mode, a major departure from Karlgren's periodization is that by E.G. Pulleyblank. First, in his various publications (e.g., Pulleyblank 1984, 1991, 1992), he makes no overt reference to a proto-Chinese period. Pulleyblank posits four periods. These are: Old Chinese (as codified in the Shijing, and ignoring xiesheng characters, which are older than the Shijing rhymes but cannot be dated); Early Middle Chinese (as codified in the Qieyun, reflecting the Nanbeichao literary standard, with its northern and southern varieties, and is based ultimately on third century Luoyang; Late Middle Chinese (as codified in the Yunjing, reflecting the Tang dynasty standard based on Changan during the second half of the Tang; and Early Mandarin (as codified in the Zhongyuan Yinyun, reflecting the language of the Yuan dynasty in the capital, Beijing. As can be observed, Pulleyblank's periodization differs from Karlgren in several respects; one concerns different sources used in some cases, as in the case of the exclusion of xiesheng for Old Chinese, and a different rhyme book for Early Mandarin (and the concomitant difference in language base and historical period); another concerns the question of the language base that was codified, as in the case of the Qieyun; and a third concerns the periods delineated, as in the absence of an explicit proto-Chinese period and a separate period for the Song dynasty rhyme tables. For Pulleyblank (p.c.), although a period between LMC and EM to represent a Northern Song standard should be posited, the sources are too sketchy to enable a full reconstruction. Scholars who have generally followed Karlgren's periodization have, nonetheless, tended to omit his Middle Chinese period. Moreover, since his periodization, two of the basic sources for Old (or Early) Mandarin have been the fourteenth century rhyme books, Menggu Ziyun (cf. Luo and Cai 1959), and
Zhongyuan Yinyun (cf. Dong 1954, Stimson 1966, Hsueh 1975), which codified the standard of the Yuan dynasty capital, the ancestral language of present-day Beijing.
The same four source-based periods as Pulleyblank have been posited by Baxter (1992). He also recognizes a separate Early and Late Middle Chinese period, differing from Pulleyblank in extending the sources for Old Chinese by using not only the early Chinese classics but also Zhou bronze inscriptions.
The third and final mode is based on major phonological changes, including changes in syllable structure. Ting (1993) posits six major periods: Proto-Chinese, Old Chinese (ca. 1700 to 200 B.C.), Ancient Chinese (ca. A.D. 200 to 420), Middle Chinese (ca. A.D. 420 to 900), Medieval Chinese (ca A.D. 900 to 1650), and Modern Chinese (ca. 1650 to present). Of these six major periods, three are further subdivided: Old Chinese into Early, Middle, and Late Old Chinese; Middle Chinese into Early and Late Middle Chinese; and Medieval Chinese into Early, Middle, and Late Medieval Chinese. Each subperiod is generally indexed by important phonological changes. Not clear, however, is the question of what phonological changes motivate the division into the major periods. For Ting, if syllable structure serves as the main phonological criterion for periodization, then he has only posited four major periods, reflecting gradual simplification of the structure of the syllable (i.e., reduction of medials and gradual loss of consonant clusters in initial and coda positions). On the basis of syllable structure simplification, He (1994), in fact, suggests positing only three major periods: Old Chinese, Middle Chinese, and Modern Chinese.
In principle, one might consider the third mode of periodization, namely, periodization based on changes in sound structure to be one for the history of the phonology of a language. Besides the obvious appeal of this mode (since the main concern, after all, is sound changes from one period to another), it would enable us to compare more easily across the three linguistic areas of periodization in the language. However, given the nature of the Chinese writing system, the actual sound changes can be understood only through interpretations via reconstructions. Most scholars accept the reconstruction of Old Chinese initial consonant clusters, whereas more controversial, for example, are consonant clusters at syllable-final position in Old Chinese, the number of medials, and other unresolved issues pertaining to finer points in reconstruction efforts. Compounding the problem are questions of whether the reconstruction is phonemically-based, or more phonetically-based, affecting both issues of syllable structure and specific sound changes.
2.3. Grammatical Periodization
Compared with historical phonology and its long philological tradition, the study of historical grammar is still in its infancy. This is because there are many uncertainties concerning the dating of various vernacular innovations in existent texts. The problem is then further complicated by the heavy influence of classical and literary models in even the written vernacular language. Given these and other considerations, grammatical periodization has not drawn as much attention from historical linguists as has phonological periodization.
Furthermore, Mair (1994) has forcefully argued that literary Chinese (wenyan wen) and vernacular Chinese (baihua wen) have been separate and parallel systems ever since the earliest stage in oracle inscriptions on shells and bones (around 1200 B.C.). On this view, vernacular Chinese does not evolve out of Classical, literary Chinese; rather, it has its own separate evolution and, hence, its own periodization. Such a view, however, has not been embraced by scholars sketching out the periodization of Chinese grammar. The following periodization schema are given by scholars who assume that Classical Chinese largely reflects the spoken language of that time, and that the written vernacular language evolved from Classical, literary Chinese.
In terms of the written language, the history of Chinese grammar can be conveniently divided into the following four major periods, based on Norman's (1988) terminology: the Preclassical Period (before the 5th c. B.C.), Classical Period (5th c. B.C. - 200 A.D.), Postclassical Period (200 A.D. until 1920), and Modern Period (1920 to present). Preclassical Chinese is based on works written on bronzes, oracle bones, and texts before Confucius' time. Classical Chinese is based on texts of the period from the end of the Spring and Autumn period down to the end of the Han dynasty, thus encompassing major works that range from Lunyu (Confucian Analects) to Sima Qian's Shiji (Records of the Historian). Postclassical Chinese is essentially that of wenyan wen, or literary Chinese, that lasted until the 1919 May Fourth Movement. Modern Chinese is then baihua wen, or the written vernacular language that emerged after the May Fourth Movement. This kind of periodization, albeit convenient, does not reflect the actual grammatical changes that were taking place during that long, Postclassical Chinese period. The grammar of the official written language of that period was modelled after Classical Chinese, even as the grammar of the colloquial language was undergoing numerous changes, as can be detected in vernacular texts that extend from Wei-Jin Nanbeichao to the Qing dynasty.
As noted earlier, Wang Li periodizes the history of the Chinese language into four major periods. However, in terms of the important grammatical changes, there are actually only three periods, namely, Shanggu (Early Old Chinese, the period prior to 3rd c. A.D.), Zhonggu (Middle Old Chinese, the period from 4th c. to 12 c.), and Xiandai (Modern Chinese, from 1919 to the present). His Jindai (Pre-Modern Chinese) is periodized purely for phonological changes. A comparison with the grammatical periodization given in Norman reveals that the Preclassical and Classical Chinese periods correspond to a single, Shanggu period in Wang. Based strictly on grammatical criteria, his Zhonggu (Middle Old Chinese) period in fact would have extended from the post-Han to the May Fourth Movement; that is, corresponding to the Post-Classical period in Norman.
Pan (1982) posits four major periods. The first is Shanggu, which is further subdivided into Qianqi (covering the Shang dynasty through Western Zhou), Zhongqi (Spring & Autumn Periods, Warring States period), and Houqi (W. Han)). The second major period is Zhonggu, which is further subdivided in Qianqi (E. Han), Zhongqi (Wei-Jin Nanbeichao), and Houqi (Tang). The third period is Jindai, from the Song dynasty to the Opium War. The fourth and last period is Xiandai, from post-May 4th movement to the present. It can be seen that Pan's periodization does not differ significantly from that of Wang Li.
A somewhat different periodization is proposed by Peyraube (1988), who posits five major periods. These are: Archaic Chinese (from 14th c. to 3rd c. B.C.), Pre-Medieval Chinese (or, Han Chinese, from 2nd c. B.C. to 3rd c. A.D.), Medieval Chinese (from 3rd c. to 13th c.), Pre-Modern Chinese (from 1250 to 1400), and Modern Chinese (from 15th c. to 18th c.). He further divides Archaic Chinese into three subperiods, namely, Jiaguwen (oracle bones) period (from 14th c. 11th c. B.C.), Early Archaic Chinese (bronze inscription subperiod, from 11th c. to 6th c. B.C.), and Late Archaic Chinese (from 5th to 3rd c. B.C.). Pre-Medieval Chinese is further divided into the historical Former Han and Later Han subperiods. Medieval Chinese has two subperiods: Early Medieval Chinese (Wei-Jin Nanbeichao) and Late Medieval Chinese (6th c. - 1250). The last two major periods are not further subdivided. Although Peyraube mentions contemporary Chinese (as distinct from modern Chinese), he does not posit a contemporary Chinese period per se.
Ohta (1988) divides the history of Chinese grammar into five major periods. These are: Shanggu (Early Old Chinese), covering the period from Shang through the Han dynasty; Zhonggu (Middle Old Chinese), covering Wei-Jin Nanbeichao; Jingu (Late Old Chinese), covering Tang through Ming dynasty; Jindai (Pre-Modern Chinese), covering the Qing dynasty; and Xiandai (Modern Chinese), beginning with the Republic of China. He further subdivides Shanggu into three subperiods, and Jingu into two subperiods. The result is a total of eight smaller periods, and it is within these smaller periods rather than the major periods that Ohta documents the important grammatical changes. The motivation for his five major periods is not explicitly stated. Nevertheless, upon more careful examination, it can be surmised that Ohta recognizes Wei-Jin Nanbeichao as one important major period (his Zhonggu period), serving as a crucial transition period from Classical Chinese to the development of written vernacular Chinese beginning with the Tang dynasty. The positing of a separate period for the Qing dynasty (his Jindai period) appears to be due to the proliferation of written vernacular fiction during this period. And, similar to Wang, Ohta posits a Modern Chinese period, except that Ohta's begins with the Republic of China, based on the standardization of the modern Mandarin and Western influence in modern Chinese grammar.
A very different periodization is presented by Lü Shuxiang (1985), who recognizes only two major periods in the history of Chinese grammar. These two periods are: Gudai Hanyu (Old Chinese), the period prior to late Tang and the Five Dynasties; and Jindai Hanyu (Pre-Modern Chinese), the period after the Five Dynasties to the present. Unlike Wang (and Pan), Lü does not posit Modern Chinese as a separate major period. Instead, Lü argues for treating Modern Chinese as a subperiod within Pre-Modern Chinese. He does not regard the post-Opium War and post-1919 grammatical changes to be significant enough to merit the status of a major period. In discussing Pre-Modern Chinese (Jindai Hanyu), Hu (1991) agrees with Lü that, in terms of grammatical changes, there is no need to posit Modern Chinese as a major period separate from that of Pre-Modern Chinese. Hu further treats Pre-modern Chinese as a major period that extends from early Tang to early Qing, observing that the grammar of colloquial, northern Mandarin, as spoken by the average person today, is basically no different from that found in the novel, Hong Lou Meng (Dream of the Red Chamber). Based on this observation, with respect to the spoken, colloquial language (kouyu), for Hu there is no justification for positing Modern Chinese as a separate, major period. It is only in the written language (shumian yu) that a separate Modern Chinese period can be justified.
Hu's distinction of written versus spoken language raises a recalcitrant problem for grammatical periodization. From recent studies of natural discourse, conversational as well as narrative, there is enough indication that written language and spoken language do not have the same grammar. Nonetheless, in the study of the history of Chinese grammar, we assume that Classical Chinese basically reflects the vernacular language spoken at that time. Similarly, we assume that such works as the Zu Tang Ji (A.D. 952) and the Dunhuang Bianwen (9th-10th c.) are based on the vernacular language of the late Tang and early Song period, and that novels such as Hong Lou Meng more or less reflect early Qing dynasty spoken Mandarin. In sum, the history of spoken Chinese grammar relies on vernacular written Chinese materials despite their limitations.
If we are to periodize the history of Chinese grammar in terms of fundamental changes in grammatical structures in the language, two major periods and one transition period are posited. The first major period is the Classical Chinese period, from the Spring and Autumn to the end of the Han dynasty. The second period, Pre-Modern Chinese, stretches from late Tang and early Song to the present. The transition period, which we will call the Post-Classical Chinese period, is from Wei-Jin Nanbeichao to early Tang. With respect to important grammatical features, Classical Chinese has measure words but very few classifiers (cf. Wang 1994). It was during the Wei-Jin Nanbeichao that classifiers began to appear in large numbers (cf. Liu 1965). In that same period, nominal prefixes and suffixes came into existence (Norman 1988), as did the locative zai verb phrase, in its distribution before and after the main verb, as in modern Chinese (cf. Peyraube 1994). Thus, the Wei-Jin Nanbeichao marks a crucial period in the development of Chinese grammar, namely, the beginning of the development of modern Chinese grammar. In the same vein, late Tang and Song constitutes another crucial period of grammatical changes. First, as Mei (1994) observes, during this period the disposal construction, verb-compounding, potential complement constructions, and verb-suffixation came into existence. Second, as pointed out by Peyraube (1994), grammaticalization took place involving the comparative construction, disposal construction, dative construction, as well as the zai locative. It is this grammaticalization of verbs into prepositions that yielded the word order change from post-verbal prepositional phrases in Classical Chinese to pre-verbal position in modern Chinese. In essence, it is during late Tang and Song that the core of modern Chinese grammar reached its maturity.
Return to Publications Page
Copyright (c) 1996-1999 Marjorie K.M. Chan. All rights reserved.
Last update: 4 April 2004 (to convert to UTF8-encoding).
URL: http://people.cohums.ohio-state.edu/chan9/articles/period-a.htm