Classical Chinese Texts

The first major Chinese effort at writing history. It was designed by Court Astrologer Szma Tan to show the cosmic situation and historic uniqueness of the Han Empire. His son and successor Szma Chyen added to it, and later used parts of it to settle a private grudge against Han Wu-di. The SJ is sometimes exciting, and always interesting as data for certain aspects of the Han mind. It is routinely misused, in contemporary Sinology, as an assured first reference for pre-Han facts.

Title. Given in SJ 130 (6/3319) as Tai-Shr Gung Shu, "The Book of the His Excellency the Grand Astrologer." Consistently with this, the earliest external mention of the work is the HS 30 #83 entry "Tai-Shr Gung 130 pyen" ("The [Book of] the Grand Astrologer: 130 chapters"). It is not known who assigned the present title Shr Ji. In the text itself (eg SJ 121 6/3115), that phrase means "archival records," the sort of thing Confucius is said to have consulted in writing the Chun/Chyou. Whoever first applied it to the Grand Astrologer's work, it is complimentary, implying an objective and principled preservation of the facts of history. As a translation of "Shr Ji," Chavannes' "Mémoires Historiques" seems not far wrong. As a description of the contents of the work, "Shr Ji" itself is highly questionable. "Tai-shr Gung Shu" was better.

Text. The reference text is the Southern Sung woodblock print in the Bwona edition of the 24 Histories, which however is not itself without scribal errors. The Shr Ji has suffered more subsequent textual corruption than the sometimes parallel Han Shu (see Honey Textual), and the Bwona version, and the chief commentaries, should be checked if an argument turns on a single character. The punctuated edition edited by Gu Jye-gang (Junghwa 1960; see Nienhauser Historians) is now the standard text. It is available for online searching at the Academia Sinica site.

Size. No count of the present text is available as of this writing. The size stated in SJ 130 is 526,500 characters, which according to Jang Da-kv seems to be accurate for the then still incomplete Shr Ji.

Form. There are five large divisions, each with its own style and logic. They may well have been originally suggested by the types of material available to Szma Tan in the Han Palace archive. First come SJ 1-12, the primary chronicles of unified dynasties. The next two divisions complement these with other information about the same periods. In the fourth division, we move down a step (to the merely local dynasties), and in the fifth, down another step (to individuals of consequence):

There were probably only 60 Jwan in the original design; see Groundplan below. For original intentions and later changes in all these sections, see the Form pages.

The multi-level arrangement of material is sometimes called the Ji/Jwan form; it has been often and deservedly admired (see Hulsewé Notes).

Content. The time span is from the Yellow Emperor to "His Present Highness," that is, Han Wu-di. Roughly half of the chapters concern Han. Some of the later supplements (like Fvng Shang's contribution to SJ 22) extend the chronological range to 020 for those topics. In content, the Shr Ji ranges beyond political events to astronomy and its alleged kindred arts (the occupational specialty of the Szmas), economics, and Confucian learning. Notably missing from the Treatises are several topics which were introduced into historiography only later, among them Law (first supplied by the Han Shu), Geography (Han Shu), and Officialdom (Hou Han Shu). There was originally a treatise on War, but it was later merged with the Standards chapter (SJ 25); the first separate War chapter appeared in the Syin Tang Shu. For structure differences in the standard histories, and the topics of their Treatises, see Cohen Introduction 355f.

Type: Integral, with a master groundplan. The plan was evidently revised and extended by the main authors during the decades-long course of its realization. Our present text includes the contributions of at least three subsequent writers; it might be called Layered, except that the later layers are intended as completions rather than as extensions.

Authors. Grand Astrologer Szma Tan (from 0137 or later to early 0109), his son and successor Szma Chyen (from 0109 to c090), and several later hands. A vital but neglected task is to distinguish their several contributions. See the Authorship section.

Proprietors. The SJ was a private work, not an official one. Szma Tan was not hired to write it, and it did not belong to the duties of his office. SJ 130 calls it "yi-jya yen," a private opinion, one version of Chinese history. At some point in its composition, most probably during the stewardship of Szma Chyen, it did come to the notice of the court, and eventually two copies were made, one for the palace, and the other preserved in the family. The work at that time had no material, or only a token beginning, in ten of its intended chapters. The family copy received some last-minute additions by Szma Chyen, and at his death passed to his daughter, and eventually to her younger son Yang Ywn, who probably made some additions with a view to completing it. His friend Chu Shau-sun independently made some additions to the Palace copy. At Yang Ywn's death (he was executed in 054), his copy probably also went to the Palace, and from that point on, the work was an Imperial responsibility. Chvng-di (r 32-07) took steps to arrange its completion by Fvng Shang. The collation and incorporation process was done at some time in Latter Han; the HS 30 Palace Library catalogue still lists Fvng Shang's chapters separately from the incomplete Shr Ji.

Date. The original compilation was made slowly, from some time after Szma Tan's appointment as Grand Astrologer (itself not long after Wu-di's accession in 0140) to the death of his son Szma Chyen (c090, a maximum span of some forty years). In 0134, Han began numbering its years over again, responding to a star phenomenon the previous year (Ywaen Gwang "In the Year of the Star . . ."). It would have been the task of Tan's office to have interpreted such events and recommended such changes, and they may well have influenced the conception of his historical task. One such event which did seem to leave a trace on the Shr Ji was the capture in 0123 of an animal which evoked the "unicorn" of 0481 whose capture supposedly prompted Confucius to cease his labors on the Spring and Autumn chronicle (Chun/Chyou); the resulting "Year of the Unicorn" period (from 0122 on) would have been a convenient precedent for a contemporary historian who needed an objectively justified stopping point. Also potent for historical interpretation was the discovery of an ancient ding in 0117, producing the Ywaen Ding period (0116- ); this was probably interpreted as symbolizing the transfer of ancient sovereignty to the Han. The latest event referred to in what seems to be Szma Chyen's part of the work is the Heir Apparent's uprising of 091. After Chyen's death, some missing sections were written or supplemented by later hands; see above and the Authorship page.

Related Texts. The SJ is unique in its time, but it is also of its time. Our forthcoming work will describe in some detail the interaction of the SJ with earlier and contemporary theories, and the texts in which some of those theories were embodied. That relationship differs markedly as between Tan and Chyen.

Transmission. For a refutation of the widespread notion that the Shr Ji was unknown to scholarship after Han, see the study by Lu Zongli. It remains true that the Shr Ji was less prominent in scholarly usage during the Six Dynasties, and was subject to somewhat more textual corruption in transmission, than the professionally more esteemed Han Shu.

Importance. The Shr Ji is by default the most readily available source of information about all that precedes Han. Unfortunately, much of that information is unverifiable, and some of what is verifiable can be shown to be garbled, fictitious, or agenda-driven. Its botched chronology for the crucial late 04c period is, as Chavannes says, a major embarrassment, one which was only corrected in later years from the Bamboo Annals. The Shr Ji's own agenda as a text, and its place in the Han discourse of power, do not interest modern scholarship, and are still to be fully appreciated. Partly due to the sympathetic image of himself which Szma Chyen has insinuated into the work, the SJ has had a strong personal appeal for later historians and officials. As a model for official history in later times, the Han Shu is more central (see Yang Organization), with the Shr Ji being a developmental cul-de-sac. For events contemporary with it, the Shr Ji is presumptively better than the Han Shu, which at many points is affected by Latter Han Confucian orthodoxy.

Commentaries. The three standard early ones, which are usually included in modern editions, are by PEI Yin (the Ji-jye), Szma Jvn (Swo-yin), and Jang Shou-jye (Jvng-yi). Of more recent works, Takigawa Kametarô is erudite and insightful, Wang Shu-min is highly regarded, and Jang Da-kv is also useful. For some of the major monographic studies, see Suggested Reading, below.

Translations. The annotated version of Chavannes (see Honey Incense 45f), abandoned after 6 volumes, is one of Sinology's monuments. A new complete translation is in progress under the direction of Nienhauser (4v so far issued, of a planned 9v). Watson's translations include much of the Shr Ji. For a list of translations, ordered by SJ chapter, see the separate index to Translations.

Citation Convention. By SJ chapter number, followed if desired by the [HB] volume and page number in the Junghwa punctuated edition, and optionally by the SJ chapter title in parentheses. Thus, for the basic material on Dzou Yen, we would have the reference

Groundplan. The original plan probably featured a division into two halves, each with the cyclical number of 60 chapters, with the first half further divided into two groups of 30 chapters. This feature may have owed something to the structure of the Shr poem repertory; see the separate Shr section.

Interpolations. There are many types. Szma Tan sometimes updated his own previous chapters, and at several points Chyen intruded into Tan's work to change interpretations or add a point of view. After their time, the incomplete Shr Ji was subject to additions from: (a) Szma Chyen's grandson Yang Ywn, (b) Yang Ywn's friend Chu Shau-sun, all of whose contributions are signed, and, by commission from Chvng-di, also (c) addenda to 7 different chapters by Fvng Shang, taking some of them down to the year 020. There are a few odd paragraphs that cannot be accounted for as the work of any of the above, and are probably pious suppletions of later scholars. Tswei Shr developed the view, already aired by the HS commentator Jang Yen, that some missing parts of the SJ were supplied from the corresponding chapters of HS. Lu Zongli (1995) has countered these arguments as they apply to SJ 123 (the Syungnu chapter) versus its HS counterpart, which were intrinsically implausible in any case. The Han Shu was designed to replace the Shr Ji, not to mend it. See the criticisms of the SJ made in the final evaluation statement of HS 62.

Literary Character. Sometimes pedestrian; at other times vivid and compelling. The difference may be seen by reading through the first few of the Bvn-ji or primary chronicle chapters. They begin by being relatively straightforward. Then suddenly, around the end of Chin, the narrative comes alive, offering high drama and poignant tragedy, portrait close-ups and intimate conversations. It is perhaps a defect that the intimate conversations, such as that between Li Sz and the Second Emperor, could not have been overheard by anyone, and that both principals perished too soon, and too suddenly, to have left manuscript autobiographies. That is, the intimate conversations are invented and unreliable, and their advent marks a shift in Shr Ji authorial strategy from documentation to fiction. It may also mark the point at which the contemporary romance Chu/Han Chun/Chyou became available to the compilers. The more lively part of the Shr Ji has won the esteem and affection of all readers for the last several thousand years. Its routine part has as often been criticized and complained of. There may be a source difference behind this appreciation difference.

Chronology. It is widely known, but also widely forgotten, that the Shr Ji chronological framework bears a self-inflicted wound in the vicinity of the latter half of the 04th century, a period which unfortunately is critical for the intellectual history of China. The main damage is to the dates of the Kings of Chi and Ngwei (the most important states in that period, and the courts at which Mencius served as advisor). Later scholars have recovered the correct dates, largely relying on the Bamboo Annals.

The SJ error is not a misprint, or a simple misinterpretation; it was done in order to accommodate, in place of the correct chronology, the self-refuting chronology of the deeds of Su Chin, who never existed. The absurdity of the Su Chin chronology was set forth in great detail by Maspero. The damage to the Shr Ji's chronological framework in adopting that chronology has never been fully assessed (it can readily be seen that some material was relocated, and other material was eliminated), let alone adequately corrected.

Sources. For their identity, see the separate Sources section. On the metaquestion of how carefully the Shr Ji treated its sources, one cautionary example may suffice. It concerns the "Disciple Register" text (Didz Ji), on which the entire SJ 67 (the Disciples of Confucius) is explicitly based. Another version of this now lost source document is included in the Kungdz Jya-yw, as KZJY 38. Close comparison (see Brooks Life, slightly abridged in Brooks Original Appendix 4) shows that the KJY 38 version better preserves the original, which in SJ 67 has been clumsily garbled and pointlessly rearranged. Given this careless or cavalier treatment of the Didz Ji, the SJ reader must wonder how the SJ has handled the documents which cannot now be checked against surviving or reconstructible originals.

Interpretation. Each SJ chapter ends (or in the Byau or "Tables" section, begins) with an authorial comment introduced by the formula "Tai-shr Gung ywe." This device is probably a more systematic version of the occasional "Jywndz ywe" judgements in the Dzwo Jwan (there is nothing like them in the Gwo Yw).

Traditional View. The esteem in which the Shr Ji is held has led to a sort of epic convention, by which its supposed author is invariably referred to as "the great historian Szma Chyen," and the work itself, without argument and as a matter of first principles, is assumed to provide the definitive account of the Warring States facts and persons which it includes. Even the inventions of the SJ are sometimes defended as presumably resting on some now lost source. All this shows modern Sinology at its least impressive. Like any other book or tract from ancient times, the Shr Ji is useful if it is carefully used.

If the position here expounded is correct, the Shr Ji is a philosophical panorama rather than a collection of documents, and is best read as a work of Dauist statecraft (compromised by Confucian additions) or appreciated as literature (especially in the parts which rest on the Chu/Han Chun/Chyou). The standard scholarly picture of the Warring States period, which has been largely built on the Shr Ji, in preference to earlier sources even where those exist, will need to be rebuilt out of sounder materials.

Reading. Chavannes has a section on the Tan/Chyen authorship question. Later attempts also exist, none so far definitive, and none even affecting their authors' treatment of the Shr Ji itself. See Pokora on Chu Shau-sun's contribution; his intended book on the subject (which would have given to Shau-sun a much larger role in the Shr Ji than his signed contributions imply) was never completed. The classic papers on the flawed 04c chronology of the Shr Ji are by Maspero; see also Chavannes 1/33. Accounts of the personality and agenda of Szma Chyen are legion, but also problematic, since they do not rest on a prior determination of Szma Chyen's share in Shr Ji authorship, but for the unambiguously Chyen segments of the Shr Ji, there are useful suggestions in Durrant. Dorothee Schaab-Hanke has clarified the astrological agenda of Szma Tan in laying out the groundplan of the Shr Ji, and implicitly shown the force and accuracy of Chavannes' translation of Szma Tan's official title: "Le Grand Astrologue." For HS / SJ relations in the test case of SJ 123 (the Account of Ferghana), Lu Zongli has convincingly shown that SJ 123 is primary, and the HS counterpart derivative (the attempted refutation in Honey Textual is methodologically flawed).

Research. The most vital currently open questions, for an understanding of what the Shr Ji is (and thus, how modern scholarship can best make use of it) are (1) the assignation of major authorship, as between Szma Tan and Szma Chyen, but not forgetting their successors, and (2) the modification of the formal plan of the work during its compilation. The first question tells us who we are dealing with, and the second question gives us one aspect of what that series of persons had in mind. A third fundamental question is (3) the sources of the Shr Ji, in particular its relations with the Jan-Gwo Tsv type of material (which involves the larger question of the survival of Warring States chronicles or traditions) and with the lost Chu/Han Chun/Chyou. More generally, (4) it would be useful to reconstruct the Palace Library of the early Wu-di period from indications in the Shr Ji, which would help to clarify the complex history of the Jan-Gwo Tsv material (about a quarter of which is utilized in the SJ), and would contribute generally to scientific bibliography. Also requiring close study is (5) the damage done by substituting the Su Chin chronology for the more accurate received chronology, in the intellectually vital period of the late 04th century, including the secondary effects of the suppression and shifting of material within the SJ. More generally, there is (6) the place of the Shr Ji effort in the context of Han thought and the dangerous court politics at the end of Jing-di's reign and most of that of Wu-di. (7) It is evident that Lyou Syang saw himself in some sense, not only as a successor in the official custody of the Palace Library holdings, but as taking a different path, one sometimes critical of the Szma team, in the use he made of those holdings. He is in a sense the earliest Shr Ji commentator, and an exploration of his role, and that of Lyou Syin after him, would be useful. We note that (8) the Rvn An letter question has been solved. As suspected by Pelliot, it is a clever forgery of the later part of the Han dynasty.

Research is underway at the Project on some of these issues, including the central authorship question, and monographic publication is planned for the near future.