Comparative History
Chinese Historiography

All cultures reinvent themselves as they go along. Historiography, meaning the writing up of historical fact so as to convey a certain impression of the past, or to organize it so as to show an overall theme or main development, often has a central role in that process. This can be problematic. The writing of a comprehensive history (like the Histories of Herodotus) may impose a late conception on early events. The recording of an event in which the writer was a participant (like the Anabasis of Xenophon) can reflect the writer's wish to enhance one factor or criticize another. The writing of biography (like the parallel lives of Plutarch) may introduce distortions in order to produce literary symmetry, or to conform to a type. Historiography (the intention of those who write history) is thus often at odds with the past as it actually was. This situation is developed to an extreme in the Chinese case. Chinese history writing under the Empire eventually settled down into a standard pattern, of which the officially recognized Dynastic Histories played the major role, and the more locally or privately compiled local histories were an important complement. A useful overview of this later history is still that of Gardner, Chinese Traditional Historiography (1961). But before that, in the wild period down through the early Empire, which is also the period parallel in time to classical Greece, the construction or interpretation of the Chinese past, real or supposed, had a different shape. It is that period which the rest of this page addresses.

Of the studies cited below, the easier are marked with an asterisk. The more fundamental are identified in the appended comment.

As It Was. Here is a quick history of China in the early period. Writing was introduced in about 01250 (1250 BC), into the small Shang state. We have inscriptional records beginning then, written on bone or shell (the so-called "oracle bones"), which record palace divinations about intended sacrifices, or answers to questions about the future. The Chinese of later periods were unaware of the existence of these oracle bones; they have been available to modern scholars only since the 19th century. They were not historical in intent, and contain only minimal information about events or about Shang social structure. The following Jou Dynasty (also spelled Zhou or Chou) lasted from the 011th through early 08th centuries. From that period we have inscribed bronze vessels, usually presented by the Jou King to some meritorious vassal warrior. These too have their limitations, since there are many forgeries (Jou bronzes have been assiduously collected by connoisseurs since the Sung dynasty), and the genuine ones are difficult to date and contain little historical information.

The genuine sources are few. From the period after the effective end of Jou (the so-called Spring and Autumn period, 08th through 06th centuries), there survives one literal palace chronicle (the Chun/Chyou) covering events of interest to one minor state (Lu). That chronicle does record military and diplomatic events, and gives an invaluable though sketchy picture of that period, as seen from the capital of Lu. It was at this time that the many Chinese domains, which had become effectively independent states since the collapse of their overlord Jou in 0771, began to gear up for unrestricted wars of mutual conquest. Spring and Autumn is an interesting but neglected period of history, and the Chun/Chyou chronicle has been largely overlooked by Chinese historians. Its development as a major historical source has been pioneered by Taeko Brooks. A sampling of her and others' papers on various aspects of the history of those centuries, as read from the Chun/Chyou, is given here. These papers are not historiography, they are examples of digging history directly out of primary sources.

Having taken the old palace state and its old elite chariot force as far as they could go under existing conditions, the states next set out to change the conditions. This change defined the next period, called Warring States (05c through most of the 03c). It led to a greatly expanded palace bureaucracy, the creation of lawcodes and legal procedures to control the population and monopolize resources, and finally to the creation of a new citizen army, led by members of the old warrior elite, but much more powerful than the old elite chariot force. Those familiar with the Napoleonic reforms, both civil and military, will recognize many of the elements in this situation. This is the golden age of Chinese classical thought, the period of the "Hundred Schools" debates about philosophical and practical matters. For the Warring States period, once the Chun/Chyou chronicle ends (in 0481), we have, unfortunately, very little knowledge of persons or dates. Here are two studies of how an event may be recovered, and how an individual may be shown to be imaginary:

One of these warring states, Chin, finally conquered the last of its rivals in 0221, and instituted a large and directly ruled state, which it is customary to call the Empire. Chin quickly fell, and was succeeded by the longer-lasting Han. It is the unification process, its justification and theoretical explanation, which in one way or another dominates nearly all writings of the Chinese classical period, and is the subject of the one Imperial work in this series, the Shr Ji.

For a more detailed but still easy introduction to Chinese history down to the beginning of Warring States, see the Introduction and Chapter 1 of The Emergence of China:

As It Was Written. China's imaginative reshaping of the Chinese past began already in the classical period itself, and extended into the Empire; it is still at work in the present day. There are five texts, or groups of texts, of which even the casual student of Chinese historiography needs to take account. One (the Shu) purports to be from early times, but is not. There is the one genuine chronicle (the Chun/Chyou) from the Spring and Autumn; a primary record and not a work of interpretation. The Dzwo Jwan purports to give a more detailed account of that period, but in fact is a work of Warring States times; it reinterprets Spring and Autumn history in moral terms. The Jan-gwo Tsv (a collection and not a work in the usual sense) purports to give many stories about Warring States persons and the events in which they played a part, but the stories are mostly written in the Empire. Finally, a text dating from the Empire, the Shr Ji, which has been appropriately compared with the work of Herodotus, attempts to give a complete history of China from the beginnings down to its own time, the 02c or early Han period. It interprets all history as culminating in the Empire, specifically the Han Empire.

The Five Principal Texts

Shu or Shang-shu, "Ancient Documents," 04c-03c. These pseudo-archaic texts were composed separately. They were not originally meant as a connected history, and no author or authors are known (most are represented as recorded speeches of ancient rulers or ministers). Whether separate or combined, they do not present a history of China, but they do play an important role in the historiography of China. Some of them purport to date from a time before writing existed in China, and are obviously later compositions. More subtle objections apply to the rest. They cluster around the periods of dynastic transition, which were of paramount theoretical interest to the Warring States thinkers, and there is no external evidence for their existence before the Warring States period. Some of them attempt to insert into antiquity (where they are meant to function as precedents) the recommendations of several 04c political theorists for the state of their own time. That is, they are based on the characteristic Chinese idea that antiquity is known, and that it provides precedents which do, or should, govern the present. They are, so to speak, written on the tabula rasa of the past. Some are particularly interesting as protests about legal abuses, and recommendations for better legal procedures, or sometimes for a different organization of state power. Most were written or begun in the 04c; some were revised, and others were probably written, in the early 03c. The Shu were an open category: Confucians wrote some, and their rivals, the sub-elite Micians, wrote others (the latter have not been preserved). 27 Shu, in the possession of Fu Shvng (who had enjoyed a privileged position under Chin, when the Shu and other Confucian texts were forbidden to the public) were all that was known of this literature in Han. These became canonical, as did a second set forged by Mei Dzv in the 3rd century (AD). The discovery of this forgery is one of the great achievements of premodern Chinese philology; see the account in Elman. The Shu are translated in Legge v3, and later by Karlgren (who confines himself to the 27 Han pieces). The Shu are from different sources, and do not have a single historiographical agenda, though subgroups can easily be distinguished. One obvious subgroup celebrates the achievements of the Prince of Jou (Jou-gung); these are relatively late, and represent at attempt by Lu theorists (Jou-gung was the founding ancestor of the state of Lu) to appropriate an earlier tradition which had a different emphasis. That is, there was competition among the forgers of Shu texts as to what the Shu texts were supposed to prove about antiquity. The selection below gives evidence that at lest one Shu is of late date, and has a discernibly late agenda.

Chun/Chyou (CC) or "Spring and Autumn," 08c-early 05c. This court chronicle of the middling state of Lu, the surviving portion of which runs from 0721 to 0481 (in some versions, to 0479), has given its name to the period from the fall of Jou to the end of the chronicle itself, or 0771-0481. It was kept from month to month by the person at the Lu court who was in charge of ritual matters, including the interpretation of omens such as eclipses. (It is also detailed about covenants and military engagements, both of which would also have required oaths or preparatory sacrifices). This omenistic reading of current events is not exactly historiography, but it does influence what events are recorded. (A known eclipse which did not result in an untoward event was omitted as meaningless). The CC text can be read in the translation of Legge (v5, the large print parts). The Chun/Chyou was unknown outside the intimate circle of the Lu ruler until the late 04c, when it became available along with the Dzwo Jwan commentary (see next). It was then thought to have been written by Confucius, who by subtle variations of phrasing (and by including or omitting certain data) indicated his approval or disapproval. This is the "praise and blame" theory of the text, and to deliver praise and blame has been a major concern of historians since that time. The majority view of Chinese historians is that the purpose of history is precisely to make judgements about historical events.

Dzwo Jwan (DJ) or "Commentary of Dzwo [-chyou Ming, on the Chun/Chyou, also spelled Tso Chuan or Zuozhuan]," 04c. The Chun/Chyou text somehow became available to members of the Kung family (Confucius's clan) at the beginning of the 04th century. They embarked on a commentary, at first with a merely ritual focus; it criticized seeming errors or deficiencies in the rituals described in the CC. At time passed, their interest shifted to the forces which controlled history generally, and these were assumed to be moral. Using an anecdotal format (there is no connected writing in the DJ, which is formatted as comments to CC entries), they showed how this theory was illustrated by events. The idea that Heaven rewards the good and punishes the evil is easily refuted by observation, and so the theory of the DJ writers shifted, and more and longer anecdotes were compiled and added to the text, which by the end of the century had become by far the largest Warring States text. In the end, the moralistic theories of history were replaced by a tougher theory: the idea that the strong are generally victorious over the weak. With that big-state ideological final stratum, the work was presented to the Chi ruler in about 0312. It immediately became known at other major courts (such as that of Ngwei, in central China), where an imitation of the CC (the "Bamboo Annals") was immediately undertaken by courtiers of the Ngwei ruler. The text is translated in Legge v5 (the small print); a new translation of DJ is presently (2013) in preparation by Durrant et al. The reigning theory of the DJ, at the time of its formal appearance and still today, is that it was written by a disciple of Confucius, in order to state more clearly the moral lessons that Confucius had supposedly coded into the Chun/Chyou.

It is claimed by some modern scholars that the information, or even the anecdotes, in DJ were drawn from earlier state chronicles, and thus are contemporary with the events they portray. The easiest refutation of this claim is that the one state chronicle we know the DJ used is the Chun/Chyou of Lu, and the Chun/Chyou of Lu contains nothing resembling an anecdote. As for ancient claims, both the Micians and the Mencians asserted that state chronicles other than that of Lu existed; the Micians even quote long stories from them. But these claims are all made after the DJ (with its historical anecdotes) had become public knowledge in c0312, and are presumably an imaginative expansion based on that exciting event.

Jan-gwo Tsv "Stratagems of the Warring States," 03c-02c. This is not a text but a whole literature, which accumulated during the late 03c. It consists largely of stories of diplomats attempting to form, or defeat, alliances among the Warring States. This kind of story became a major literary genre in early Han. Many such stories were edited under this title by Lyou Syang at the end of Han. There is no one author, and not even a claimed author. Many of the JGT anecdotes are about the great diplomat Su Chin, who turns out to be entirely legendary. The thrust of many of the stories is that if only the diplomats had succeeded in unifying against Chin, the Warring States unification process could have been arrested. That is, these texts are partly an exercise in wishful thinking, and partly a period vogue for intrigue as such (early Han was a time of especially intense palace intrigue). Loyal ministers and generals in JGT tend to come to a bad end, and much blame is placed on imperceptive rulers. Not all the JGT literature was included in Lyou Syang's collection; some material recently recovered from an early Han tomb (Mawangdwei, c0168) overlaps with our JGT, but other parts were previously unknown. JGT was a major source for the Shr Ji, the last text we shall consider. JGT has been translated by Crump, whose study Intrigues reviews the major arguments for inauthenticity. The conventional opinion about the Jan-gwo Tsv material is that is is entirely historical, and that was certainly how the authors of the Shr Ji regarded it.

Shr Ji. "Historical Records," 02c-01c. Of these five texts, only the Shr Ji has an author in the usual sense, and was undertaken on purpose, with a grand overall design (in 120 chapters, later extended to 130) which was gradually filled it. SJ has appropriately been compared with the work of Herodotus, and occupies a comparable position in the historiography of China. It was begun (in c0138) and largely written by the Dauist court astrologer Szma Tan, whose conception and design it still reflects, and whose Dauist theory of life (things change, so quit while you are ahead) it widely exemplifies. It was continued after his death by his Confucian-educated son Szma Chyen (from c0109 to c090), who had a very different view of life and of previous history. It was still incomplete at Chyen's death, and was finished by imperial order at the end of the 01st century. Szma Chyen had offended the Han Emperor and was castrated as a punishment; his resentment appears in several of his additions to the work, and sympathy for Chyen's sufferings dominates all later opinion about the Shr Ji. Szma Tan's conception of the unification of China was that it realized a previous unity of China, symbolized by a claim that all the states (and all the native peoples, some of whom also formed states) are equally descendants of the Yellow Emperor. (The Yellow Emperor is entirely mythical; he was first claimed as an ancestor by the usurping Tyen family of Chi, in the middle 04th century). Szma Tan saw Chin as one phase among several; part of a natural process. For Chyen, on the other hand, Chin was evil and Han was good, and he extensively demonized Chin (as by inventing the tale that scholars who possessed Confucian texts were buried alive; the original proclamation prescribes instead a sentence at hard labor). For one recent account of the Shr Ji seen as the work of Chyen, see Durrant. The Shr Ji is translated in part by Watson; a complete translation is presently (2013) in progress by Nienhauser and others.

The Shr Ji is the first and only comprehensive account of Chinese history, including the Warring States. That grand conception was not continued by later Chinese histories, which confined themselves to the affairs of one dynasty. The Shr Ji is thus a major source for persons investigating Warring States matters, but it is a treacherous source. Tan was an uncritical user of documents; he drew much of his material from the largely legendary Jan-gwo Tsv, and even spoiled his fairly accurate received chronology so as to accommodate the entirely mythical career of the JGT figure Su Chin. These chronological wounds in the text have never been healed, and they continue to greatly hamper serious research on the Warring States period.

For a brief background overview of the early Imperial period, see Chapter 8 of Emergence. It ends with an extract from the Shr Ji, which vividly shows how the heroism of a vanished age was ineffectual against the bureaucratic savvy of the new and coming age. That remains a deep and correct insight about the old China and the New, and goes far to vindicate Szma Tan as a historian. Not that he wrote that passage, but he had the good sense to include it from an earlier history (now lost), the Chu/Han Chun/Chyou "The Saga of Chu and Han").

A hundred years after the completion of the Shr Ji, there appeared the Han Shu, a more complete remake of the Han portions of the Shr Ji. It confined itself to Han, and eliminated the pre-Imperial period altogether. The Han Shu was the model for the later series of 24 dynastic histories, which are the main thread of later history writing. There is no Greek parallel to this later historical tradition. The Han Shu inaugurates the age of imperial historiography, and with it the classical period and its doings, as something to be either recorded or interpreted, come to a close.

Some Early Concepts

These five works do not really form a continuous tradition, but one or another of them do include some ideas which are further developed in standard Imperial historiography (to which, as such, there is no Greek parallel, since Greece did not have a future on the scale of Imperial China). They are not mutually exclusive. Among them are:

Rulers. An assumption lying behind many of these texts is that the ruler of the state is the determinative factor, for better or worse, of the success of the state. This undoubtedly reflects reality; in China, the ruler was relatively more powerful than in Greece, and the major Chinese states were larger than those of classical Greece.

Mandate. Another notion reflected in some of these texts is that the right to rule is conferred by Heaven, and may be forfeited by misrule. Conquest narratives (the Jou conquest of Shang is a major focus) exist to elaborate on this theme; there is a Bad Last Ruler topos, according to which heinous crimes and misdeeds of a late ruler (typically personal rather than, say, economic) cause the loss of the Mandate, and the acquisition of the Mandate by some power which behaves better, and thus is enabled to conquer the forfeiting power. The Conqueror must previously have demonstrated conspicuous virtue, often (but not always) conceived as concern for the people. This is the populistic tendency in early Chinese thought: the idea that the state is for the people, and not vice versa. Like so many themes expressed in these texts, it is largely wishful thinking, adopted at most symbolically by this or that ruler.

Gods. This belief in Heaven as interested in the doings of states and men was not the dominant religious idea in China. One curious aspect of the political unification process is that it was not facilitated by the presence of a common religion. There was a god called Di (functionally equivalent to, and probably cognate with, the Indo-European Zeus/Deus, primarily a weather god), who appears in the Shang oracle bones and also in some of the later philosophical and literary texts, but who had little importance. The chief objects of sacrifice in China were ancestors, and ritual propriety forbade anyone but a descendant from offering sacrifices to the dead. This sort of religion was thus divisive. There was nothing in China to correspond to the oracle at Delphi, which was available to inquirers from any part of Greece, and which was part of a uniform Greek culture.

Power. The Chun/Chyou, being a record of selected events and not an interpretation of them, in effect documents the workings of a power politics situation. At its end (it was compiled over almost the whole of the 04c), the Dzwo Jwan turns cynical, and embraces the idea that power is more important than sacrificial or other piety. The classical military texts of course take this view; the Sundz "Art of War" states at the outset that it is military force on which the survival or extinction of states depends.

Virtue. By contrast, in all but its final phase, the Dzwo Jwan reinterprets Spring and Autumn history in order to show that it is lack of virtue (whether appearing as violations of liturgical propriety or as atrocities committed against the people) which cause state failure, up to and including loss of the Mandate. This view is also strongly held, and not abandoned, by Confucian theorists such as the later Mencian group (first half of the 03c).

Unity. The overriding historical concept in the classical period was the collective memory, at some points elaborated into a myth, of political unity under the Jou. The goal of all the states was to replace Jou as the hegemonic power, or (toward the end) to make a new unity under their own ancestral line. The first ruler to take the Jou ruler's title of King (wang) was the ruler of Chi, in 0342. Others soon followed. The image of Jou was itself militarized in the late 04c and 03c. Medium-sized Lu, which also aspired to hegemony, based its hopes on the fact that the Lu rulers were descended from the Prince of Jou (Jou-gung), supposedly a decisively influential figure in preserving the Jou hegemony in its early and insecure years. It turns out that the position of Jou-gung was itself mythologized and exaggerated. (Some of the Shu were written to support just this claim). The destruction of Lu and Jou in 0249 (by Chu and Chin, respectively) more or less put a stop to this line of thought,

Cycles. Beginning in the late 04c there arose a theory of cyclic change, one variety of which was the Five Element theory. In that theory, the successive dominance of five astral powers (identified with planets, but always spoken of as the elements earth, water, fire, wood, metal), was reflected in changes of dynasty on earth. This became important in Han, when such things as the color of sacrificial robes was changed to match the supposed dominant element. The usual view is that Chin was dominated by Water (and the color black, and the direction West), whereas Han was dominated by Fire (and the color Red), though this was later changed to Earth and Yellow, reflecting the direction Center). A variant of this cyclic idea was the Mencian expectation that a Sage would arise every 500 years, in order to set the world again on the path of virtue, and thus inaugurate another period of peace and good government.

Devolution. We may be reminded of Greek theories of Golden and later Ages. Many Chinese theorists considered that life had been better in the past, and had become corrupt in the present. In this idea, the Confucians (who looked back to the supposedly perfect days of early Jou) and the Micians (who preferred pre-Jou antiquity) and the Dauist primitivists (who regarded the simple life as best, and deplored the rise of law and indeed of language as corrupting the pristine state of society) tended, for once, to agree. This idea of an exemplary antiquity is behind some of the Shu as well. There is very little sense (save with the Legalists, who were busy constructing a more efficient and thus more successful state, and with the Micians, who wanted to determine a somewhat different set of public values for the new state) that things were going to be better tomorrow.

That is to say, those who were making history had a notably different set of ideas than most of those who were writing or commenting on history.

All this, we realize, is unlikely to make a favorable impression on those coming to the subject from a classical Greek background. The major works either have no authors or were not written by their claimed authors. There are no consecutive accounts; the basic unit is the chronicle entry (average: 8 words) or the personal anecdote. Even the monumental Shr Ji is a mosaic of shorter narratives about single persons or single reigns (which is why 120 chapters). Some of the major works (Shu, JGT) are really assemblages and not, in origin, consecutive works in any sense. These facts are just part of how the ancient Chinese behaved textually. Most utterances are gnomic: pithy and memorable, rather than expansive. Most texts are short, and grow long by accretion rather than by the planned effort of an individual (the Shr Ji, with its groundplan, is the only exception). That is part of the classical style. Long narratives, long poems, and long plays do exist in China, but they are characteristic of Imperial and not of classical times. One result is that the modern reader of classical texts must be prepared to savor the meaning of a cryptic remark, or to grasp the emblematic meaning of a short story. The rhythm, for both performer and reader, is shorter.

That ends our overview. Questions or corrections are welcome, and may be sent via the "Project" mail link given at the bottom of this page.

 Back to Comparative History Page

6 Feb 2013 / Contact The Project / Project Home Page