20 PRIMARY AND SECONDARY SOURCES 479
extant is unknown, but a great many must have been destroyed in the upheavals during the nineteenth and twentieth centuries.
20,4 Historiography
History was written by officials for officials—Etienne Balazs
One of the unique features of Chinese historical studies is that a very large number of the primary sources were works produced in or preserved by a conscious tradition of historical writing and compilation often by officials. The two most famous works of historical criticism (shiping jtif) were:
Liu Zhiji #j£t*$; (661-721), Shitong jfcijjt (Generalities on history), 710.3
Zhang Xuecheng ^^jft (1738-1801), Wenshi tongyi jtjjfcii^, (General meaning of historiography), 1832, Shanghai guji, 1993.4
Note also Wenxin diaolong ii'^J^-fi (30.1).
Because of the long and sophisticated historiographical tradition in China, instead of working from official archives or private documents (as in post-Rankean European historiography), the modern student of Chinese history is liable to be handling historical works compiled by Chinese historians continuously over the last 3,000 years. In order to be able to assess the qualities and the biases of these works, clearly some knowledge of the aims and methods of the historians and compilers who produced them is essential.
Furthermore, in order to be able to start looking for primary sources on the topic of his choice, the student of Chinese
3 Liu's work is discussed by Edwin G. Pulleyblank, "Chinese Historical Criticism, Liu Chih-chi and Ssu-ma Kuang," in Historians of China and Japan, William G. Beasley and Edwin G. Pulleyblank, eds., OUP, 1961, 135-66; David McMullen, State and Scholarship in Tang China, CUP, 1988, "History," 159-205. Guy Gagnon, Concordance Combineedu Shitong et du Shitong xiaofan, 2 vols., Maisonneuve, 1977. For a translation, see Stuart H. Sargent, "Liu Chih-chi: Understanding History: The Narration of Events," in Renditions 15: 27-35 (1981).
4 On Zhang's work, see David S. Nivison, The Life and Thought of Chang Hsueh-ch'eng (1738-1801), SUP, 1966.
480
HISTORICAL GENRES
history will also need to know how traditional historians organized their materials and classified different types of sources. This is a subject discussed in section 9.3.5
The characteristics of traditional Chinese historical writing were as follows: (1) Chinese historians were typically Confucian literati, but more significantly, they were also officials. Their primary focus was on politics defined as the affairs of state, which meant the record of imperial government. In common with officials in other societies, the final record that they compiled was based on official documents and encoded into bureaucratic (Confucian) categories. These were usually far removed from actual transcripts of conversations or events. From this follow the second, third, and fourth characteristics of traditional historical writing: (2) its many close connections with government and the orthodox ideology, as seen in the theory of zhengtong JE.^ (legitimate succession), which official historians traced in an effort to legitimize new dynasties (see Box 8 at the end of this chapter); also as seen in the compilation of compendia of historical precedents as a guide to official action; (3) its strong moral didacticism, with the historian's duty being to bestow baobian J|.$£ (praise and blame), using Confucian moral tenets as the yardstick; (4) its ruthless excision of anything judged in conflict with the above two concerns (thus not only a focus on the elite and its ideology but also a particularly narrow focus on that elite). Additional characteristics of
5 Denis Twitchett, The Writing of Official History under the T'ang, CUP, 1992, analyzes the process of official historical writing and compilation at a formative period. Beasley and Pulleyblank (1961) contains 11 papers on different aspects of Chinese traditional historical writing and still forms the best interpretive introduction to the subject in a Western language. Essays on the Sources for Chinese History, Donald Daniel Leslie, Colin Mackerras and Wang Gungwu, eds., ANU, 1973; Univ. of South Carolina Press, 1975, has some excellent chapters which are cited in the appropriate chapters of the manual. Han Yu-shan, Elements of Chinese Historiography, W. M. Hawley, 1955, contains a lot of basic information. Charles S. Gardner (1900-66), Chinese Traditional Historiography, HUP, 1938; rev. ed., 1961, provides a brief overview.
20 PRIMARY AND SECONDARY SOURCES
481
China's traditional historical writing are: (5) its early elevation to the status of an activity differentiated form other branches of writing, to which great importance was attached for two millennia of continuous historical production;6 (6) its development into many well defined genres and subgenres; (7) its scholarly attention to such ancillary disciplines as cataloguing, calendrical sciences, historical geography, ancient scripts and artifacts; and (8) its development of a philosophy of history.7
20.5 Private Historical Writing
In addition to official historical writing, a large amount of historical works were written privately {sishi ^Jl). Often they followed the main forms of the official histories, including annals, annals-biography, topically arranged histories and so forth (Table 12, section 9.3). Indeed, many of the most distinguished works in the official historical canon were credited to or written by private individuals, for example the Zuozhuan or
6 On the overlap between early historical and literary narrative, see David Derwei Wang, "Fictional History/Historical Fiction," in Studies in Language and Literature I (1985), 64-74; Anthony C. Yu, "History, Fiction and the Reading of Chinese Narrative," in Chinese Literature: Essays, Articles, Reviews X (1988), 1-19; Andrew H. Plaks, "Toward a Critical Theory of Chinese Narrative," in Plaks, Chinese Narrative, Critical and Theoretical Essays, PUP, 1977, 309-52; Sheldon Hsiao-peng Lu, From Historicity to Fictionality: The Chinese Poetics of Narrative, SUP, 1994; Henri Maspero, "Historical Romance in History," in China in Antiquity, Frank A. Kierman, Jr., tr., Univ. of Massachusetts Press, 1978, 357-65; and David Johnson, "Epic and History in Early China: The Matter of Wu Tzu-hsu," JAS 40.2: 255-71 (1981). See also studies of early historical works such as the Zuozhuan or the Mu tianzi zhuan.
7 For an overview of Chinese historiography arranged by themes (for example, "talking straight and making a point by indirection," or "the advantages and disadvantages of national, informal and family histories"), see Qu Lindong %§.fa^, Zhongguo gudai shixuepiping zongheng ^ |g] -fc^ $_tf:4tbif$L$? (A-n evaluation of historical criticism in ancient China), Zhonghua, 1994. For a bibliography of Chinese writing on historiography between 1900 and 1985, see Zhongguo shixueshi yanjiu shuyao % 3^ f^^f      Jr> Yang Yixiang et al., eds., Tianjin jiaoyu, 1996.
482
HISTORICAL GENRES
the Sbiji Private historical writing also took many other
forms, including scholarly commentaries, studies of primary sources, and informal jottings (biji 3|Hi*). There were special terms to distinguish privately written histories from official ones, for example, bieshi %\\ ^ zashi   jE^and yeshi Sf
The definition of bieshi (lit. "separate history") changed over time. In general it was applied to nonofficial works which were not in annals or annals-biography form (jizhuanti jfe^lSl)-Sometimes it was used interchangeably with zashiy but normally the distinction between the two was that bieshi were considered more serious works, lying somewhere between zhengshi iE. on the one hand and zashi on the other. Zashi (chap. 24) were often in the form of biji (chap. 31). Baishi ^ was a term sometimes used in book catalogs in the sense of record of folkways, otherwise just as another word for nonofficial history. The general expression yeshi %f$i was not used as a category in book catalogs. In ordinary language it meant nonofficial histories written about the dynasty of the day (private history, as opposed to official history). Yeshi flourished in the Nan-Bei Chao, in the Song and in the Ming. From the Qing onward they typically recount in a semifictional way life at the court. Waishi ^J-^t (informal history) was normally used in the titles of fictional narratives, the most famous example being the eighteenth-century satire Rulin waishi H^Jfa?]-^ {The Scholars).*
Shichao (historical excerpts) began to be published in the Song. They consist of quotations usually culled from the Standard Histories.
Zaiji (contemporary records) are the histories of states not regarded as legitimate. As recorded, for example, in Wu-Yue chunqiu (Table 18, section 19.1), in the Shiliuguo chunqiu (45.2) or in the Manshu (41.5.1).
Bieshi, zaiji or baishi were also often classified as zashi.
8 Harold Kahn, Monarchy in the Emperor's Eyes, Univ. of Massachusetts Press, 1971; rpnt., 1978, contains a perceptive study of official and unofficial historiography of the later empire.
20 PRIMARY AND SECONDARY SOURCES
483
Box 8: Five-Element Theory (Wuxing JL^f) From the Warring States to the Nan-Bet Chao, it was considered vital to correctly identify the pattern of the rise and fall of dynasties and to identify each dynasty with the right one of the five elements or agents.8 At first these were ordered in a mutually overcoming cycle (xiangke in which each element successively overcomes the
next one: water, fire, metal, wood, earth (shui huo jin mu J^,tu±). Then in the Han dynasty, the order was changed to a mutually producing cycle (xiangsheng ;fgjfe_) in which each element produces its successor: wood, fire, earth, metal, water (mu huo tu _±_, jin shui /Jc). The debate on the legitimacy of the Northern Wei dynasty turned on the choice of the right element;9 the same applied to the debate on the legitimacy of the Jin.1D Thereafter, political legitimacy used other symbols, although the five elements continued to be a popular means of ranking people and things (for example, in the choice of the characters for siblings; see 3.2).
8 L. S. Yang (Yang Liansheng $M$]§D> "Toward a Study of Dynastic Configurations in Chinese History,'' in his Studies in Chinese Institutional History, HUP, 1963, 1-17; Rao Zongyi $£,;£]S|j> Zhongguo shixue shang zhi zhengtong lun ^ |g] jL^SLfäuüfr (The theory of legitimate succession in Chinese historiography), HKCU Press, 1977; rpnt., Shanghai yuan-dong, 1996, which is an anthology of 170 excerpts on the theme preceded by a lengthy introduction.
9 John Lee, From Five Elements to Five Agents: Wu-hsing in Chinese History," in Sages and Filial Sons, Julia Ching and R. W. L. Guisso, eds., HKCUP, 1991, 163-78.
10 Chan Hok-lam (Chen Xuelin ffci^fa), "Patterns of Legitimation in Imperial China," Part I of Legitimation in Imperial China: Discussions under the Jurchen-Chin Dynasty, Univ. of Washington Press, 1984, 19-48. Chan returns to the same question in an article whose scope is wider than the title implies, "*Ta Chin' (Great Golden): the Origin and Changing Interpretations of the Jurchen State Name," TP 77.4-5: 253-99 (1991).
21
Annals
The earliest and one of the most important methods of arranging historical materials throughout Chinese history began as a bare catalog of court events arranged chronologically: e.g., "on such and such a year, month, day, King X of Y went on a hunt." Much of the writing on the oracle bones (clap. 15), which predates the first extant historical work by many centuries, contains similar records of single events and actions, but the relations between court scapulimancy and the origins of a continuous process of record keeping are obscure. The court chronicler in early China was an important official charged with astronomical as well as archival functions; he played a key part in the arrangement and timing of royal ancestor worship and sacrificial rites and other ceremonies. He may also have had remonstrance functions, in light of his duties of keeping records of models worthy of emulation and of portents heralding disaster.
In later Chinese history the bare catalog of events at court was greatly expanded and elaborate composite chronicles of events throughout the empire began to appear, either written privately or under official sponsorship (although always with the major focus on the court and central government).
21 ANNALS
485
21.1 Chunqiu ^J^.
The earliest extant example of a historical work in the annals style is the Chunqiu (Spring and Autumn Annals), the
court chronicle of the state of Lu.1 It is only 17,000 characters long, but covers 242 years (722-481 BC), which works out at 70 characters per year; so events are tersely recorded. The longest entry is only 47 characters. The average is 10. The shortest is just one character under the year 715 BC: "Ming (pests).
Editorship was traditionally and implausibly attributed to Confucius. The Chunqiu's title was later used to name the period 770-476.
The third of the three commentaries on the Spring and Autumn Annals to have survived, the Zuozhuan jL.i%- {Commentary of Zuo), is much fuller and more lively than the Annals themselves and goes beyond the other two commentaries. It is the
1 In the Western Zhou, Chunqiu meant "year" and was used as one of the generic terms for "annals." There are seven Warring States works with chunqiu in the title. Because of the connection with Confucius who also came from Lu, the Lu chunqiu ^j^k. became one of the Classics and the only one to survive. The most convenient edition of the Chunqiu and its three commentaries is included in ICS Concordances 30-32. See also the older H-Y Index, Supplement 17. Note Bernard Karlgren, "Glosses on the Tso chuan," BMFEA (1969-70); for other references, see Anne Cheng in ECHBG, 67-76.
For translations, see Burton Watson, The Tso chuan: Selections from China's Oldest Narrative History, Col. UP, 1989; James Legge in The Chinese Classics, vol. 4 in 2 parts, HKU Press, 1960; Seraphin Couvreur, Tch'ouen ts'iou et Tso tschouan, 3 vols., Ho-kien fu, 1914; rpnt., Cathasia, 1951.
For stylistic analysis, see David Schaberg, "Foundations of Chinese Historiography: Literary Representation in Zuo zhuan and Guoyu" Ph.D., Harvard University, 1996; John C. Y. Wang, "Early Chinese Narrative: the Tso-chuan as Example," in Andrew,H. Plaks, Chinese Narrative, Critical and Theoretical Essays, PUP, 1977. For a dictionary of the language, as well as a place name and personal-name index, see Chunqiu zuozhuan cidian JM^£#^]$-> Yang;Bojun ^tjJ^ and Xu Ti jfefe,. eds., Zhonghua, 1985.
486
HISTORICAL GENRES
prime source on the history of the period 722-468 BC.2 Another important early chronicle (mainly of the state of Wei) is the Zhushujinian 4t1§"&1^F- (Table 18, 19.1).
21.2 Annalistic Sources and Veritable Records
Annalistic writing was one of the main methods of arrangement adopted by Sima Qian in the Shiji (in the Basic Annals section) and it remained an integral part of the Standard Histories thereafter (see chap. 22). Xun Yue (148-209), Hanji :<^Zi was based on the Hamhu. It is the first annals of a single dynasty. From the Tang, the official writing of annalistic history had become standardized into the following types:
1. Qijuzbu (Court Diaries), sometimes translated as diaries of activity and repose, imperial diaries or audience records. They record the decisions and actions of the emperor in the conduct of government business, normally as this occurred in formal sessions of the imperial court each morning. In periods when the emperor conducted the government informally or secretly, the qijuzbu were weak or nonexistent; the qijuzbu began in the Later Han and continued on and off until 1911. They were in no sense intended for publication. Only a tiny fraction survive up to the Qing, from which time the greater part, both in Chinese and Manchu, are extant in 12,000 ce?
2. Sbizhengji ff^jfjcfi (Records of Current Government), were in some periods confidential records compiled under the authority of ministers.
3. Rili 0 M (Daily Calendars) were a condensation, arranged day by day, of the first two.
2 Note that the first recorded title (in the Hansbu "Yiwenzhi," Zhonghua, 1,713) is Zuoshizbuan £|?v^f-. It is usually referred to simply as the Zuozhuan.
3 The Chinese versions of the following reigns have been printed: Kangxi, Yongzheng, Daoguang, Xianfeng, Tongzhi, and Guangxu (see 50.5 for details). The Neige Manwen qijuzbu ^^;$x43.y§F>i (Manchu versions of the qijuzbu) are available on microfilm at the Yishiguan in Beijing (50.2.1).
21 ANNALS
487
Only fragments of the Court Diaries, the Records of Current Government and the Daily Calendars have survived from before the Qing.
Given that the Court Diaries were composed almost immediately after the event, and were usually secret, they tend to be more reliable than other officially compiled sources, for example the Veritable Records, or even more so, the Standard Histories, which were compiled only after the downfall of a dynasty by its successor and therefore often came several hundred years after the event.
All three of the above were used as the chief source for the annalistic Veritable Records {Shilu -Jfijl), of which there was one for each emperor, as well as for the less detailed National Histories (Guosbi H)jt)> which were compiled in some periods for each reign. The keeping of Shilu began in the Nan-Bei Cbao and continued until 1911. Less than 1 percent of the Shilu survive from before the Ming (a small part of the year 805 from the Tang and portions of the years 983 and 996 from the Song).4
The Shilu of the Ming and Qing covering the years 1368 to 1911 are extant in over 7,500 juan. They form an extremely important source for the Ming and for the Qing, but they were compiled only after the end of each reign, so they could come at some distance from what happened at the beginning of a long reign, and always with the risk of strong influence from participants (or their successors) in the events recorded.
Despite these drawbacks, the Shilu contain an extraordinarily detailed record, unequaled for any comparable period in any country. Many edicts, memorials and other documents, as well as day-to-day events, are preserved in them.5 They constitute
4 Denis Twitchett, The Writing of Official History under the Tang, CUP, 1992.
5 On the process of the compilation of the Veritable Records, see L. S. Yang (Yang Liansheng #^££), "The Organization of Chinese Official Historiography: Principles and Methods of the Standard Histories from the T'ang through the Ming Dynasty," in Beasley and Pulleyblank, 44-59, rpnt. in Yang's Excursions in Sinology, HUP, 1969, 96-111. Twitchett (1992), 119-59 examines the compilation of the Tang Veritable Records
Footnote continued on nextpage
488
HISTORICAL GENRES
the single most important source for the last 500 years of Chinese history.
21.3 Zizhi tongjian jf>&igJS£
Although many annals and chronicles (not to mention Veritable Records) were written during and after the Han, it was not until the Song that a major step forward took place in this genre with the compilation of the Zizhi tongjian ')f >£i|J|i (Comprehensive mirror for aid in government) by Sima Guang *i$)JL (1019-86). This magisterially carries the history of China from 403 BC in continuous chronicle form over the following 1,362 years down to AD 959. In the catholicity of sources consulted, many since lost, in its discussion of disputed points where there was a divergence of evidence (kaoyi as well as in its huge table of contents (running to 74 pages in the Zhonghua edition), it marked an important new level for the chronicle form as well as for general historical methodology. It had an enormous influence on later Chinese historical writing, either directly or through its many abbreviations, continuations, and adaptations.6 It remains an extraordinarily useful first reference for a quick and reliable coverage of events at a particular time.
In writing it, Sima Guang was assisted by three outstanding scholars: Fan Zuyu ftHS.^ who prepared the first draft on the Tang; Liu Shu |fij-*£, who prepared the draft on the Wei, Jin
(no longer extant) and ISMH, 29-33, does the same for the Ming Shilu, which are extant. For editions of the Ming and Qing Shilu, see 49.1 and 50.5, respectively.
6 On the compilation of Sima Guang's great work, see Edwin G. Pul-leyblank, "Chinese Historical Criticism: Liu Chih-chi and Ssu-ma Kuan'g," in Beasley and Pulleyblank, 135-66; also Zhang Xu Tongjian xue iHHt^ (Studies on the Comprehensive Mirror), Shanghai: Kaim-ing shudian, 1948; rev., 1957; rev. and exp., Anhui renmin, 1981.
Chaps. 54 to 78 of the Zizhi tongjian, covering the last years of the Han and the Three Kingdoms, have been translated into English (45.2).
21 ANNALS
489
and Nan-Bet Chao; and Liu Ban who worked on the Han period. Sima Guang's son did the proofreading.
The best modern edition of the Tongjian is the punctuated, movable-type one prepared by a team of 12 editors led by Wang Chongwu Nie Chongqi Jf-and Gu Jiegang
The text used was a Qing reprint of the Yuan dynasty edition of Zizhi tongjian by Hu Sanxing $fi3LJ$ (1230-1302). Hu incorporated the kaoyi ^ into the text as notes (they had originally been printed separately). Many of the sources quoted in them have since been lost. Hu also provided his own comments and corrections separately (appendix to vol. 20 of the Zhonghua edition, 1-190). The Zhonghua editors have incorporated the textual notes of Zhang Yu (1865-1937) which were taken from the most important Song, Yuan and Ming editions and added these and other Song commentaries in bracketed notes in the main text.7
Many adaptations and summaries of this massive 9,612 page chronicle were made. The most popular of all was a highly moralistic one entitled Zizhi tongjian gangmu jff >&xj§.|gi#^ gj. It was devised by Zhu Xi (1130-1200) and written by his pupils.8 The Siku editors classified it under historiography {shi-ping £_if), nevertheless, because of Zhu Xi's name it was the Gangmu which was read in the Ming and Qing rather than the much longer and more rigorous Tongjian.9 But it is the Tong-
7 Sima Guang £J gjfc, Zizhi tongjian ^>&ili|g£, Guji, 1956; 9th rpnt., 20 vols., Zhonghua, 1995; convenient and readable, two-volume, reduced-size reprint (four pages to the page with original page numbers indicated), 1997.
8 Conrad Schirokauer, "Cbu Hsi's Sense of History," in Ordering the World- Approaches to State and Society in Sung Dynasty China, Robert P. Hyams and Conrad Schirokauer, eds., UCP, 1993, 193-220.
9 J. A. M. de Moyriac de Mailla's Histoire generate de la Chine, ou an-nales de cet empire; traduites du Tong-Kien Kang-Mou, 13 vols., Paris: Pier-res et Clousier, 1777-85; Taibei rpnt., 1968 is an abridged translation from the Manchu version of Zhu Xi's Zizhi tongjian gangmu (Summary of the comprehensive mirror for aid in government) and its later continuations. It was the largest general history of China available in a West-
Footnote continued on next page
490
HISTORICAL GENRES
jian itself and the full continuations of it that are of interest to the modern historian.10 The most important of these cover the Song and are listed in 47.1 and 47.2.
There is a dictionary to help find one's way around the
Tongjian:
Zizhi tongjian da cidian ^}&Ifii§Ut^:& (Great dictionary of the Zizhi tongjian), Shi Ding and Shen Zhihua >^.^^, eds. in chief, Jilin renmin, 1994.
Also indexes:
Shiji tsugan sakuin 'jf j&iil^^^!, Saeki Tomi -füölj'j ed., Toyoshi ken-kyukai, 1961.
Shiji tsugan kochü chimei sakuin '^>^i3.M#Iii^^'^?|) Araki Toshi-kazu — and Yoneda Kenjiro iftsj^fvj^ß comps., Jimbun,
1967. An index to the place names in Hu Sanxing's notes.
ern language for two centuries (until the publication of The Cambridge History of China) and it was used by many later textbook writers.
10 For example, Bi Yuan ^jl-j/L (1730-97) et al., Xu Zizhi tongjian Hf 12 vols., Beijing, 1957; 6th rpnt„ 1988. Replaces the other con-
tinuations compiled in the Qing and covers the years 960-1370.
22
Standard Histories
The first great innovation in historical writing and departure from the early annals style was the work of Si-ma Tan (d. 110 BC) and his son, Sima Qian $] £t (? 145-86 BC), court astronomers and librarians during the second and at the beginning of the first century BC (44.2). They wrote what many centuries later was chosen as the first of the zhengshi iE.j£_ (Standard Histories or Histories, for short). There was eventually one for each legitimate dynasty. In all, 24 Histories written over 1,832 years. They number just under 40,000,000 characters (Hanzi i^tf)> of which 13,966 are different ones.
22.1 Structure, Contents and Titles
The Shiji is a history of China from the Yellow Emperor down to Han Wudi arranged in a manner which, with certain adaptations, was to set the form for a new way of writing history that came to be known as Jizbuanti ,&£^ff (annals-biography) after two of its most important parts. It was to be used in all of what later became known in the eighteenth century as the Ershisishi —"f'WiL (The 24 histories).
The Shiji contains 130 chapters (juan J^) divided into 12 Benji ^jfe (basic annals), 10 Biao ^ (tables), 8 Shu (monographs), 30 Shijia (hereditary houses), and 70 Liezhuan f\] i^- (collected biographies or memoirs). The two main innovations of Sima Qian were the monographs (Shu and the memoirs (Liezhuan f■J'ff). The monographs (in later Standard
492
HISTORICAL GENRES
Histories usually called Zhi and sometimes translated as treatises) cover the historical evolution of selected institutions such as rituals, the calendar, astronomy or political economy (with the emphasis on taxation and coinage). The memoirs are groups of biographies or profiles of both famous and (some) less famous people (as well as foreign peoples) of each age. These two sections of the Sbiji, together with the older-style basic annals (Benji ^$&) which carried on the court annals tradition, became the three major elements of the jizhuan genre of history writing. In the Chinese historiographical tradition the annals were considered the core and the biographies as illustrations of the core. The monographs were entirely left out of eight of the earlier Histories. They were considered belonging to a separate tradition of institutional history writing. The other sections of the Sbiji, the biao and the shijia (which cover the history of the major pre-Han states) were incorporated only sporadically or not at all in the later Histories.
The most important difference between all the Standard Histories from the Hanshu i$L^~ onward and the Shiji was that they covered only one dynasty and made no attempt to cover the vast sweep of history (3,000 years) embraced by the Shiji. The generic term is duandaisbi WsiK£j> as distinct from works such as the Shiji, which are called tongshi Only the
monographs continued to cover periods of time extending earlier than the dynasty in question (see Table 23 for a list of the different topics covered in the monographs in each of the Histories). Another important difference between the earlier and later Standard Histories is that prior to the Tang, eight of what later were recognized as Standard Histories were written by private individuals and another seven were commissioned from one or two individuals. After the Tang reorganization of the Bureau of Historiography, the Histories (with some notable exceptions) tended to become more and more standardized as they became the final step in a cumulative process of compilation by committees of official historians (which History fell into which category of authorial composition is indicated in Table 21).
22 STANDARD HISTORIES
493
Explanation of Table 21
One asterisk indicates that the work was written by a private individual. Two asterisks indicate that the work was commissioned from one or two individuals.
Those Histories which were officially compiled (marked with three asterisks) had a nominal editor-in-chief (whose name is listed in the Table), but they were the work of many hands at the Historiography Bureau. They based themselves on materials compiled by the Bureau of the previous dynasty.
The numbering from 1-24 in the left-hand column is the same order as that in the Siku quanshu zongmu. With the exception of the Nanshi and the Beisbi, it follows the chronological order of the dynasties. The numbering in brackets (1-26) is somewhat different because it shows the order according to the dates of compilation.
Table 21: The Standard Histories
Title
Author, editor (dates)       Compiled Period
(presented or covered printed)
*l{l).Sbiji$ji (The Records of the Grand Historian)
SimaTan sj&sl (? 180-110 BC)and Sima Qian sj (?145-86BC)
* * 2 (2) .[Qian] Hanshu   Ban Gu ^ gj (AD 32-
{History of the 92) Former Han)
*3 (4). Hou Hanshu $L:t% IST (History of the Later Han)
Fan Ye (398-445)
:f"4 (3). Sanguozhi Chen Shou
Z- SI ,t- (The Record of (233-97) the History of the Three Kingdoms)
***5 (lS).Jinshu% (History of the fin)
Fang Xuanling
(578-648)
104-87 (91) BC Earliest times to 95 BC
AD 58-76 (92)     206 BC -AD 24
398-445 (445) 25-220
285-97 (297) Wei 221-65;
Shu 221-64; Wu 222-80
644 (646) 265-419
Table continues
494
HISTORICAL GENRES
Table 21—continued
Title
Author, editor (dates)
Compiled (completed or printed)
Period covered
**6 (5). Songshu %r% {History of the Song)
*7 (6). Nan Qishu iW* {History of Southern Qi)
**8 (8). Liangshu ^ (History of the Liang)
**9 (9). Chensbu {History of the Chen)
**10 (7). Weishu ^ {History of the Wei)
**11 (10). Bei Qishu itVr ~$j {History of the Northern Qi)
***12 (11). Zhoushu JI| {History of the Zhou)
***13 (12). Suishu {History of Sui)
*U{U).Nanshifo$l {History of the Southern Dynasties)
*15 (15). Beishi (History of the North-em Dynasties)
**+16 (\€).JiuTangshu £A* (OldHistory of the Tang)
Shen Yue ft^ (441-513)
Xiao Zixian ^"-f-^ (489-537)
Yao Cha
(533-606) and Yao Silian 4* ^ A (i 637)
Yao Cha ^l^. (533-606) and Yao Silian -jfefc&Jfe
Wei Shou Ifcifc (506-72)
LiDelin (530-590) and Li Boyao (565-648)
Linghu Defen 4^|J& (583-661)
Wei Zheng (580-643)
Li Yanshou ^Ji-^-(£1. 618-76)
492-493
628-35 (636) 622-29 (636)
551-54 (554) 627-36 (636)
420-78
489-537 (537) 479-502
Liu Xu fijHij (887-946)
502-56
557-89
386-550
550-77
ca. 629 (636) 557-81
629- 36 (636) 581-617
630- 50 (659) 420-589
Li Yanshou 4?é_# 630-50 (659) 368-618 (fl. 618-76)
940-945 (945) 618-906
Table continues
22 STANDARD HISTORIES 495
Table 21—continued
Title Author, editor Compiled Period
(dates) (completed or covered
printed)
**17 (18). Xin Tangshu Ouyang Xm gfcflH^   1043-60 (1060) 618-906
(Mw History (1007-72) and Song
of the Tang) Qi (998-1061)
***18 (17). Jiu Wh- Xue Juzheng If&SE.    973-74 (974) 907-60
f A flit (°^ (912-81) History of the Five Dynasties)
*19 (19). Xin Wudaishi Ouyang Xiu %U%if   1044-60(1072) 907-60
^£.^$1 (New His- (1007-72) tory of the Five Dynasties). Original title: Wudai shiji if^jfc
***20 (22). Songsbi Tuotuo (Toghto        1343-45 (1345) 960-1279
5fc & [History of the
Song) (1313-55)
***21 (20). Iwosfo" Same as SongsW 1343-44(1344) 916-1125
i^5^_ (History of the
Liao)
***22 (21). Jinshi Same as Songshi 1343-44 (1344) 1115-1234
(History of the Jin)
***23 (23). Yuansbi SongLian 1369-70(1370) 1206-1369
>t jt (#wtory (1310-81)
y«d«)
***24 (24). Mingsbi Zhang Tingyu 1678-1735 1368-1644
Bfi 3t (Pistory of the flL 3*i (1739)
jtfmg) (1672-1755)
*25 (25).      KwnaW Ke Shaomin 1890-1920 1206-1367
$T ytJL (NewHistory. (1850-1933) (1920) of the Yuan)
***26 (26). Qingshgao Zhao Erxun 1914-27 (1927) 1644-1911
TJS-jSUft (DraftHistory (1844-1927) o/Y&e Qmg)
496
HISTORICAL GENRES
The concept of a single Standard History for each successive dynasty emerged only very gradually. The term zhengshi jE itself was used for the first time in the Liang dynasty in ad 523 and incorporated into the Suishu "Jingjizhi Pt^MH^" {Standard History oftheSui, monograph on dynastic bibliography) as a category distinct from bashi (histories of dynasties not recognized as legitimate) or bieshi %i] (not officially recognized history). As such it was applied to important historical works mainly in the jizhuan (and to some extent in the bian-nian) genres and it continued to be used in this sense until the Qing. The present usage to refer to the 24 (or 25, or 26) Histories became only the norm after the compilation of the Siku quanshu    J^-iMf in the eighteenth century (9.5).
Until the Tang, the usual reference was to the Sanshi (The three Histories): that is the Shiji, the Hanshu and the Hou Hanshu t$j$L^j (replacing the Dongguan Hanji JtHlvJlli)-From the Tang, the Sanguozhi X K] & was added to make the Sishi (The four Histories), or Qian sisbi iftfWjt. (The first four Histories), terms still in use. During the Tang, six new Histories of different kingdoms of the preceding period were compiled by the Bureau of Historiography. Together with three privately written works this made a total of 13 works recognized as the main histories of China up to the end of the dynasty. They were referred to as the Shisandai shi -hX^jtL (The Histories of thirteen dynasties). During the Northern Song, the Nansbi and Beishi lt$L> which had been privately compiled in the Tang, were added to the Histories, as were the newly compiled Xin Tangshu WxM^ (New History of the Tang) and Xin Wudaisbi iff jLf^ $1 (New History of the Five Dynasties). This brought the total to 17. Thereafter new ones were added as they were officially compiled. At the end of the fourteenth century there were 21. The completion of the Ming-shi 9^5^_ (Standard History of the Ming) in 1735 raised the number to 22, to which were added at the time of the compilation of the Siku Quanshu the recovered editions of the Jiu Tangshu M (Old History of the Tang) and the Jiu Wudaisbi |f (Old History of the Five Dynasties). The number now stood at 24
11 STANDARD HISTORIES
497
and the phrase Ershisishi (The 24 Histories) was coined
on the completion of the palace edition in 1775 (22.2). The number rose to 25 after the official addition of the Xin Yuanshi $f it $L (New History of the Yuan) in 1921. Although the Qing-shi gao (Draft History of the Qing) has not been officially-
included in the Histories, it is in the old tradition and is therefore included in Table 21 to bring the total to 26.
Although varying greatly in quality and length, the Standard Histories constitute a monumental oeuvre, the importance of which can hardly be exaggerated. They provide remarkably accurate coverage of over 2,000 years of Chinese history (from the official, Confucian standpoint, as seen from the imperial court), and they include historical profiles of the rulers, events, leading personalities, major institutions and administrative boundaries of each dynasty, as well as a considerable quantity of detailed information on the peoples of East, Central, and Southeast Asia. The value of the earlier Standard Histories is greatly enhanced by the fact that many of the sources upon which they were based have since been lost and alternative sources are lacking. For these reasons the scissors-and-paste methods of some of their editors should be regarded as an asset.
A huge literature exists on the Histories, especially on the first four, which are regarded as monuments of literary and historical excellence. A fraction of this literature is listed in the appropriate chapters of Part V. The following section notes various reference tools for the Histories as a whole.1
1 Zhongguo jizhuanti wenxian yanjiu ilfe#^3C^Jt-^t^L, (Research on Chinese annals-biography literature), Wang Jingui JL^-ft, ed., Beijing daxue, 1996. Chügoku seishi kenkyü bunken mokuroku    gJiE.jt^f%x.
3 i£ (Bibliography of research on the Chinese Standard Histories), Kokusho kankokai ■ Sl^-f'J'-ff'^) eds., Kokusho kankökai, 1977. Covers books and articles in Japanese, Chinese and Korean from 1868 to 1977. Arranged by Standard History.