In the summer of 2008, Tsinghua University in Beijing acquired a collection of about 2,000 bamboo slips with brush-written characters in Warring States period (476–222 bc) script.Footnote 1 Radio-carbon testing of a sample slip produced a date of around 305 bc, which corresponds to the approximate dating of the closure of Tomb One at Guodian, near Jingmen, Hubei Province.Footnote 2 The bamboo slips found in the Guodian tomb are in the script of the ancient state of Chu, and the writing on the bamboo slips in the Tsinghua University collection is very similar. However, whereas the texts found at Guodian are primarily philosophical, according to Li Xueqin 李學勤, who leads the editorial team preparing the Tsinghua manuscripts for publication, the Tsinghua manuscripts are primarily historical and many of them are “shu 書” (“Documents”) or “similar to shu”.Footnote 3 This description associates them with the Confucian classic, the Shang shu 尚書 (“Ancient Documents”) which, according to Chinese tradition, is a collection of shu made by Confucius and later canonized as the Shu jing 書經.
But what were shu? How do we recognize a manuscript as such when we see (i.e. read) it? And what distinguishes shu from other ancient texts? The simple answer to such questions would be that shu are what is in the Shang shu, including its lost chapters. We might also add the chapters of the Yi Zhou shu 逸周書 (“Remaining Zhou documents”), since its chapters are also called shu and were supposedly omitted when Confucius selected documents for inclusion in the Shang shu. Straightforward though this answer may seem, it has two problems. First, shu were banned by imperial edict during the Qin dynasty (221–206 bc) and the Shang shu was reconstructed during the Western Han (206–8 ad), so we know very little about its form and content before the Qin; we know even less about the origin and history of the Yi Zhou shu. Second, there appear to have been many more shu circulating in pre-Qin times than were ever included in either of these two corpuses.
The transmitted shu
The title Shu jing is sometimes translated as the “Book of History”. It is not, however, a work of narrative history. Each chapter is a self-contained document, supposedly contemporaneous with the historical period in which it is set and entirely independent of the other chapters; in other words, the chapters are a collection of independent texts that have been arranged in chronological order but not linked to one another by their language or content. Many Han dynasty texts refer to a “100-chapter” work, which was supposedly selected by Confucius from more than 3,000.Footnote 4 This account parallels that in which Confucius is presented as having selected 300 songs from 3,000 and possibly mimics it.Footnote 5 Similarly, the chapters of the Yi Zhou shu are supposedly a compilation of the shu which Confucius discarded.
The history of the Shang shu is too complex to explain in detail here. Briefly, there are two transmitted versions: a “modern script” (jinwen 今文) version with 28 (or 29 where one is divided) chapters, and an “old script” (guwen 古文) version with a further 30 (58 in all) chapters. Historians argue about the date of composition of almost all of the chapters of both versions. According to tradition, the “modern script” version was hidden by Fu Sheng 伏生 in the Qin dynasty and reconstructed after the establishment of the Han. Most modern scholars accept this version as authentic in the sense that the current text was transmitted from the Han dynasty. The “old script” version was said to have been found in the walls of Confucius's house by Kong Anguo 孔安國 (d. c. 100 bc), but modern scholars generally agree that the transmitted version was a construct of the fourth century ad.Footnote 6
The modern script version includes four (or five) sections: “Yu shu 虞書” (sometimes divided as: “Tang shu 唐書” and “Yu shu”), attributed to the time of Yao and Shun; “Xia shu 夏書”; “Shang shu 商書”; and “Zhou shu 周書”. The documents in each section purport to be contemporaneous texts of their respective periods. While most modern scholars agree that the shu attributed to periods before the beginning of the Zhou, with the possible exception of the Pan Geng 盤庚 chapter of the “Shang shu 尚書”, were written later, they also generally agree that at least some of the Western Zhou documents genuinely date from the Western Zhou period.
The Yi Zhou shu, with 71 chapters in the current anthology, presents even greater problems of dating and authenticity than does the Shang shu. Although, according to tradition, it consists of shu that were omitted when Confucius compiled the Shang shu, the transmitted work had diverse sources with different dates of origin.Footnote 7 Huang Peirong, whose 1976 dissertation remains the most exhaustive study of the text, observes that seven chapters are similar in style to the “Da Gao 大誥” chapter of the Shang shu and therefore probably relatively early.Footnote 8 More recently, Li Xueqin has argued that the “Shi fu 世俘”, “Shang shi 商誓” and “Chang biao 嘗表” are authentic documents of the Western Zhou, using evidence from bronze inscriptions.Footnote 9 However, most chapters of the Yi Zhou shu were probably composed in the Warring States period. Indeed, Huang Peirong argues that some 32 chapters were probably composed by a single hand.Footnote 10
Although Confucius was closely associated with a “one hundred chapter” Shang shu from the Han dynasty on, the early evidence is much less clear. The Lun yu confirms Confucius's interest in the shu, but it tells us almost nothing about the form of the text or texts held by Confucius. It includes only four references to the shu, two of which are very general. One states that Confucius used “formal speech” – perhaps a form of court language – for shi (“songs”), shu, and in the performance of rituals.Footnote 11 In the other, the disciple Zilu 子路 exclaims, “Why must one first read the documents and only then be considered learned! (何必讀書，然後為學)”.Footnote 12 The other two references are more specific. In one, Confucius's disciple Zizhang 子張 remarks: “the shu says, ‘Gaozong went into seclusion and did not speak for three years’”. (書云：高宗諒陰, 三年不言).Footnote 13 This line corresponds roughly to a line from the “Wu Yi 無逸” chapter of the transmitted text.Footnote 14 In the other, Confucius, asked why he does not take part in government, responds: “The Shu says, ‘He who is filial, simply by being filial and acting as a friend to his elder and younger brothers, can extend the effect to government’”. (書云：孝乎惟孝，有于兄弟，施於有政).Footnote 15 This citation is included in a modified form in the forged old script version of the Jun shi 君奭 chapter of the Shang shu, but the original source is no longer extant.Footnote 16 From these references, we cannot even tell whether Confucius had a single text with multiple chapters or an open-ended collection of independently circulating manuscripts like those in the Tsinghua University collection.
Since Confucius and the later Confucian tradition have been so closely associated with the shu, many scholars writing about manuscripts in the Tsinghua University collection assume that shu composed during the Warring States period should be associated with the Confucian school. However, most pre-Han thinkers, including masters who were not followers of Confucius, sometimes cite shu for historical authority. Most significantly, the Confucians’ most vociferous critic, Mozi 墨子, frequently cites the shu. Indeed, the Mozi repeatedly refers to the “documents of the former kings” (xian wang zhi shu 先王之書), rather than simply shu. This suggests that he had a collection of such documents attributed to the ancient kings. He also uses specific terms for documents of different periods, i.e. “Xia shu”, “Shang shu” and “Zhou shu”. That he does not refer to “Tang shu 唐書” or “Yu shu 虞書” even though he often discusses the legend of Yao and Shun may indicate that these chapters, generally thought to be relatively late, were not included in his collection of documents.
Although there is more evidence that Mozi had a defined collection of shu than there is for Confucius, we do not know whether his shu were limited in number or whether they were open-ended. Neither do we know how closely the shu of Mozi resembled those of Confucius. Some of the titles Mozi cites are familiar from the 100 titles given in the “Preface” (xu 序) to the Shang shu, which is traditionally attributed to Confucius, and some of his citations correspond to lines in the transmitted text – but many do not.Footnote 17 Even when the titles of Confucius's and Mozi's shu are the same, we do not know whether the contents of the documents were identical. Confucius and Mozi (and all the philosophers of the Warring States period) used historical legend to present their own political philosophy, and in so doing transformed the legends about ancient kings to accord with their own philosophical stance.Footnote 18 Were there different versions of the same text reflecting the two versions of ancient history? Were there some chapters that were held only by Confucius or by Mozi? These questions cannot be answered with existing evidence.
The stories of Confucius making a selection of 100 shu from 3,000, with the Yi Zhou shu representing documents he had left out, may be apocryphal, but they are based upon an assumption that the category “shu” included many more texts than those found in the transmitted tradition. Indeed, if the Shang shu and Yi Zhou shu are a collection of speeches and accounts of ancient kings, it is unlikely that they were the only ones. Though some of the speeches from the Western Zhou period are probably contemporaneous, the rulers of that period surely made more speeches than have been transmitted in these anthologies, and since scholars agree that many of the shu in the Shang shu and Yi Zhou shu are products of the Warring States period, later people clearly continued to compose texts in the style of ancient shu and we may reasonably assume that not all of these Warring States period imitations were included in either anthology. Thus, although some unearthed manuscripts may be lost chapters of the Shang shu or Yi Zhou shu – or lost versions of those chapters, we should not make this assumption without clear support.
Shu as a literary form
We need to define shu in a more comprehensive manner – as a form of literary composition, rather than as chapters of known historical compilations – in order to understand these manuscripts. Thus defined, we can discuss shu without entering into the complex history of the Shang shu. We can also begin to understand the origin and history of shu, as opposed to that of the transmitted text. The most significant features of shu are that: (1) they were – or pretended to be – contemporaneous records; (2) they include formal speeches by model kings and ministers from ancient times (Western Zhou or earlier); and (3) many shu include the expression, wang ruo yue 王若曰, “the king seemingly said”. This expression is not found in all shu, but it provides a key to understanding how shu differed from other texts.
My hypothesis is that shu originated as the pre-prepared, written scripts for royal speeches and that later shu were fictional compositions, written in the style of these ancient documents. That the basic form of shu is that of a formal speech is evident in the names of the six traditional types of shu: “canon” (dian 典), “counsel” (mo 謨), “oath” (shi 誓), “instruction” (xun 訓), “proclamation” (gao 誥) and “decrees” (ming 命). With the exception of dian these are all types of oral utterance. The early graphic forms of dian include bamboo slips tied together, often placed on an altar, and in Zhou bronze inscriptions, the term is used with reference to formal records.Footnote 19 Later, however, dian appear to be exclusively associated with the most ancient period in Chinese history, from which, even in the Warring States and Han periods, direct transmission of the words of the ancient kings probably seemed unlikely. In the transmitted Shang shu, the only dian is the “Yao dian” (sometimes divided into the “Yao dian” and “Shun dian 舜典”). Moreover, the Shuowen jiezi 說文解字 defines dian as “the documents of the five emperors” (wu di zhi shu 五帝之書), which corresponds to the number of “emperors” (di) in the “Basic Annals” of the Shi ji 史記.Footnote 20 This suggests that there may have been dian attributed to Huang Di 黃帝 and Zhuan Xu 顓頊 in the Han dynasty.
Wang ruo yue
As already noted, many shu include the expression, wang ruo yue, “the king seemingly said”, a form of words which is also found in Western Zhou bronze inscriptions. Indeed, to my knowledge, the expression is only found in shu, citations of shu in other texts, and in bronze inscriptions. The context in the inscriptions is similar to that of the Shang shu; that is, in bronze inscriptions, wang ruo yue introduces a formal speech given by the king. As noted above, most scholars agree that at least some of the transmitted shu attributed to the early Western Zhou period are authentic, because of similar linguistic formulae and the overlap between the historical events recorded in the shu and those mentioned in bronze inscriptions. Shu do not, of course, include ancestral dedications, which are a standard part of Zhou bronze inscriptions. Nevertheless the format of many Western Zhou bronze inscriptions is similar to the conventional form of speeches recorded in the Zhou shu section of the Shang shu: that is, a short preface in which the physical setting and/or circumstance is recorded, often with a date, and followed by the words of the king or a minister.
The inscription on the Da Yu ding 大盂鼎, an early Western Zhou vessel (reign of King Kang 康王) is famous because its reference to the inebriation of the last Shang kings tends to support the authenticity of the Jiu Gao 酒誥 chapter of the Shang shu, which also refers to excessive drinking and inebriation by the last Shang kings. Here, it may serve as an example of this format for the introduction of a speech with the expression wang ruo yue:
It was the ninth month when the king was at Zong Zhou 宗周 that he gave Yu 盂 his command. His majesty seemingly said, “Yu, the illustrious King Wen received the great command that was [seen] in the sky. King Wu succeeding King Wen and developing the region, expelled the evil, [his beneficence] spread to the four quadrates, and he truly made the people upright. In carrying out the duties of office, when using ale, they did not dare to sink into drunkenness and in making roasted and steamed offerings, they did not dare to be inebriated… .Footnote 21
唯九月，王在宗周，令(命)盂。王若曰:「盂，丕顯文王，受天有大令(命)，在武王嗣文作邦，闢厥慝，敷有四方，允正厥民，在御事，
，詐酒無敢酖，有祡蒸祀，無敢醻。… .
The inscription ends with a list of gifts and a solemn injunction that Yu not fail in the duties of his appointment, in commemoration of which the vessel was cast.
The other Western Zhou bronze inscriptions that include the expression wang ruo yue also record ceremonial appointments in which the king gives a speech charging the appointee with the duties of office and presenting him with various gifts. As Chen Mengjia 陳夢家 observed in his “Wang ruo yue kao 王若曰考”, although the speeches were attributed to the king, they were delivered by someone else. Ruo in the expression wang ruo yue serves to indicate this.Footnote 22 These appointment inscriptions often describe the ceremony in detail, including the arrival of the king, the entrance of the appointee, and the king's verbal order that a certain scribe should make the appointment with a text written on bamboo slips (ce ming 冊命). The speech that is then read out is preceded by the expression, wang ruo yue. For example, in the inscription on the Mu gui 牧簋, we find: “His majesty gave the order to the Scribe of the Interior, Wu, to appoint Mu with a text written on bamboo slips. His majesty seemingly said…”. The speech is followed by an expression of gratitude by the appointee. Then, the bamboo slip manuscript of the speech of appointment (ming ce 命冊) is presented to the appointee, who hangs it from his waist and goes out. From this we know that the term ce refers to bamboo slips as a material object, and that they were given to the appointee.Footnote 23
In a small number of bronze inscriptions, a speech attributed to an official rather than the king is preceded by the expression ruo yue.Footnote 24 Thus the function of ruo yue is not to indicate royalty per se. Rather, it marks the oral performance of a speech by someone other than the person to whom the words are attributed. In the Shang shu, the context is similar to that of bronze inscriptions – formal speech attributed to a king or high official. Thus, we may reasonably assume that, at least in those shu which are authentic texts of the Western Zhou, the expression had the same meaning as it did in the bronze inscriptions; that is, it denoted that the speeches attributed to the ruler were actually delivered by someone else.
This usage explains the puzzling appearance of the term wang ruo yue in some chapters of the Shang shu in which the Duke of Zhou rather than the king is the protagonist. For example, the Duo fang 多方 chapter includes the line, “the Duke of Zhou said, ‘the king seemingly said’” (周公曰王若曰), and in the Duo shi 多士 chapter, we find:
It was the third month, when the Duke of Zhou first dwelled at the new settlement, Luo. He thereby declared to the remaining many officers of the Shang king, “His majesty seemingly said, ‘Oh, [you] remaining officers of Yin…
惟三月，周公初于新邑洛，用告商王士。王若曰：“爾殷遺多士
Some scholars have interpreted these lines as meaning that the Duke of Zhou was calling himself king and thus attempting to usurp the throne, but, as Chen Mengjia pointed out, a more apt interpretation is that the Duke of Zhou delivered the speech but the words were attributed to the king. In the Li zheng 立正 and Jun shi 君奭 chapters, on the other hand, we have Zhou Gong ruo yue. In those cases, Zhou Gong would have been speaking on his own behalf, with someone else reading the speech, but the term would not imply that he was claiming kingship.
“Documents” (shu) and “bamboo slip manuscripts” (ce)
The term ming shu 命書, “document of appointment”, also occurs in bronze inscriptions, with a similar meaning to ming ce. Here, Li Feng has had an important insight that helps us to understand the difference and, I believe, the meaning of shu more generally. Observing a pattern of distinction in the usage of these two expressions, Li suggests that ming ce refers to the physical document, i.e. the bamboo slip manuscript which was physically given to the official appointed after he expressed gratitude to the ruler, whereas ming shu refers to its text.Footnote 25 A particularly revealing example is this inscription on the Mian gui 免簋, a late Western Zhou vessel:
It was the first auspiciousness period of the twelfth month; the king was at Zhou. At daybreak, his majesty approached the Great Temple, Xing Shu assisted Mian. When the appointment was made, his majesty bestowed the document (shu) on the Bamboo-slip-manuscript maker Yin to use in appointing Mian. [It] said, I command you… .
唯 十又二月初吉， 王在周， 昧爽 ， 王各于大廟， 邢叔佑免， 即令， 王授作冊尹書 ， 俾冊命免， 曰： 令 汝… .Footnote 26
Thus, the term shu refers to the document in the abstract – the text of the speech which is a literary composition, whereas ce is its physical form as writing.
The origin of shu
We may hypothesize, then, that shu were originally the scripts of speeches composed for the purpose of delivery by officials on behalf of the ruler or high minister in a formal ceremony. Since they were the scripts of speeches delivered by someone other than the person to whom the words were attributed, they were necessarily written down in advance of the performance. The expression ruo yue, “seemingly said”, marked the fact of their performance by someone other than the purported author; the king or minister “seemed” to say them, but did not actually voice them. Presumably a copy would have been kept in the royal archives, with a record as to the date, place and circumstances of delivery, and the bronze inscriptions indicate that a bamboo slip manuscript of the speech was given to the person to whom the speech was addressed.
Western Zhou bronzes were commonly cast for the particular purpose of recording official appointments or other benefices, and so they record the speeches made by the king at the time of appointment. However, the kings would also have made other formal speeches that were not cast on bronze. Presumably, these were also pre-recorded on bamboo slips, read out by an official in a formal ceremony, and archived with a record of the date, place and circumstances of delivery. Another copy (or copies) may also have been presented to those to whom the speech was addressed. This practice of writing scripts in advance of formal utterances that solemnized government appointments on bamboo slips suggests a context for the origin of the documents found in the Shang shu and helps us to understand how shu were unlike other literary compositions.
Since the earliest shu were the archived scripts of ceremonial utterances made at specific times and places, we may reasonably suppose that they would be accompanied by a notation concerning the time, place and circumstances of their delivery. Similarly, the documents of the Shang shu often have a preface setting the scene, usually with a date, followed by a speech. Some of those shu attributed to the early Western Zhou in the Shang shu may even have originated as actual scripts, although they will necessarily have undergone a certain amount of alteration in the process of transmission, which included rewriting in different stages of the development of the script, including the “new” Han script. Later, texts that were written in the style of such scripts were also regarded as shu.Footnote 27 Significantly in terms of their development, the early Western Zhou shu tend to provide only sparse information concerning the date, place and circumstances of delivery, whereas shu attributed to earlier periods, but presumably written later, tend to have more elaborate narrative information in addition to the speeches.
How these scripts of speeches came to be circulated is not clear. However, the presentation of the bamboo slip manuscripts after royal speeches were delivered to the person(s) addressed in the speech, mentioned in the bronze inscriptions, suggests that there was a mechanism. The traditional supposition that Confucius had access to the archives of the state of Lu is also worth noting, since the core chapters of the Shang shu are documents associated with the founder of that state, the Duke of Zhou. We may also suppose that once such documents began to be circulated they inspired imitations. Thus, while the original shu were actually scripts of speeches, other works were written in the style of such scripts. While these might be considered forgeries, it is possible that at least some of them were written in the spirit of fictional reconstructions.
An important aspect of this literary form – a contemporaneous record of direct speech – is that the form itself demands an acceptance of historical authenticity: this is not a historical record or an interpretation. There is no intermediary: it is what kings and ministers actually said. Because the shu originated as scripts of speeches, they have an authenticity as contemporaneous documents that is absent from narrative history. This point is essential because it meant that by studying them, one could commune directly with the kings of ancient times without the intervention of interpretation. To read or recite such words and perform the ritual actions was thus to model oneself upon them. It is why only the speeches of greatest kings and ministers were included and why the shu were so important to Confucius, who claimed that he “transmitted, but did not create” (述而不作).Footnote 28
In sum, if we define shu as a literary form, a shu is any text which claims to be a contemporaneous record of a speech of an ancient king. Some shu will be authentic scripts of speeches prepared for royal delivery, some will be based upon such speeches, and others will be fictional reconstructions of what an ancient ruler or minister might have said. Like the Songs (shi), for which – as we have seen above – Confucius used a type of formal speech, the shu represent actual words of the ancients. However, whereas the Songs originated with oral performance and were written down later, shu were literary compositions from the outset, though they were intended for oral performance. In this sense, they can be regarded as the first Chinese literary compositions.
