Hostname: page-component-74656f9884-c52kr Total loading time: 0 Render date: 2026-11-07T14:15:58.535Z Has data issue: false hasContentIssue false
  • English
  • Français

NEW LIGHT ON THE LI JI 禮記: THE LI JI AND THE RELATED WARRING STATES PERIOD GUODIAN BAMBOO MANUSCRIPTS

Published online by Cambridge University Press:  02 January 2015

Wen Xing
Wen Xing*
Affiliation:
Xing Wen, 邢文, Dartmouth College; email: wen.xing@dartmouth.edu.
Rights & Permissions [Opens in a new window]

Abstract

This article starts by examining how “Li ji” was used as a title in early China. In addition to the Li, Li jing, Shi li, Li yi and Yi li, etc., “Li ji” was used as an alternative title for the Li canon, and the “Li” as a title was used to indicate what we now refer to as the Li ji or the Xiao Dai Li ji as well. By discussing how the Guodian bamboo slip texts are related to the transmitted version of the Li ji, this article argues that the writing styles of the related excavated bamboo manuscripts are the particular Li gujing- or Li guji-styles, that the “Li gujing” or “Li guji” is more a description of those ancient retrieved texts than an actual title, and that many of the Guodian bamboo-slip texts are the Li guji- or Li gujing-style writings. According to the Li guji-style Zi yi text, it is evident that the material in the Li ji comes from creditable pre-Qin textual sources. Although the Li ji text came to its current shape as a result of later editing work, it does not mean that the received Li ji sections have higher textual value than the excavated manuscripts do. On the contrary, the excavated manuscripts could have many valuable early textual sources that the transmitted Li ji has lost.

提要

本文始於考察 “禮記” 之名在古代中國的用法。《禮》、《禮經》、《士禮》、《禮儀》、《儀禮》等名之外，《禮》經也曾以 “禮記” 為名。同時， “禮” 也被用作我們今天所說的《禮記》或《小戴禮記》之名。通過討論郭店簡與今本《禮記》的關聯，本文認為，相關出土簡文的文體，就是 “禮古經” 或 “禮古記” 的文體，而 “禮古經” 或 “禮古記” 更多地是用於描述某些古代文獻，而非古書的實名；若干郭店簡文即為 “禮古記” 或 “禮古經” 的文體。由 “禮古記” 本《緇衣》可見，《禮記》的材料顯然有著可信的先秦文獻來源。雖然《禮記》經後人編輯成書，但這並不意味著今本《禮記》比出土寫本有著更高的文獻價值。相反，出土寫本或有許多有價值的早期材料，今本《禮記》業已失傳。

Keywords

Li jiLi gujingLi gujiZi yiForgery Bamboo SlipsGuodianShanghai Museum bamboo slips禮記禮古經禮古記緇衣偽簡郭店上博簡

Information

Type
Articles from the Third International Conference on Excavated Manuscripts, Mount Holyoke College, 2004
Information
Early China , Volume 37 , 2014 , pp. 519 - 550
Copyright
Copyright © The Society for the Study of Early China and Cambridge University Press 2014 

Excavated early Chinese manuscripts with counterparts in the transmitted textual tradition provide a better perspective for textual and intellectual examination of the texts than those without such counterparts. As far as early Confucian texts are concerned, no transmitted ancient Chinese classic has as many connections with excavated Warring States period bamboo manuscripts as the Li ji 禮記 does. From the bamboo-slip manuscripts excavated from a Chu 楚 tomb, Guodian 郭店 Tomb No. 1, Jingzhou 荊州, Hubei Province, to the Warring States period Chu bamboo manuscripts collected in the Shanghai Museum,Footnote 1 the textual similarities and connections between these bamboo-slip manuscripts and the Li ji have aroused intense scholarly interest.Footnote 2 Particular attention has been paid to the two bamboo-slip versions of the Zi yi 緇衣Footnote 3 since they can be examined against their received textual counterpart in the Li ji.Footnote 4 However, the Li ji is one of the ancient Chinese canonical texts that has the most confusing early textual history.Footnote 5 The very confused usage of its title “Li ji,” which has not yet been properly clarified, is one such example.Footnote 6 And also, the authenticity of the bamboo-slip Zi yi collected in the Shanghai Museum has become an issue after I developed the calligraphic approach to disprove the authenticity of forged bamboo slips in recent years.Footnote 7 For example, in the Shanghai Museum version of the Zi yi, the right ends of the horizontal strokes of both er 而 of er xin qi suo jian 而信其所賤 on slip 10 and yan 言 of Zi yue wang yan 子曰王言 on slip 15 are blurred, and all the brushstrokes of the first character yun 員 on slip 24 are seriously blurred.Footnote 8 In my system of calligraphic authenticity examination, they are clear indications of forged calligraphy on damaged and/or uneven surface of bamboo slips if the published images are accurate. Evidently, it is a topic for a separate article. This article begins with a concise examination of the usage of the title of the Li ji, and then turns to discuss the connections between the Li ji and the Li ji-related Warring States period bamboo-slip manuscripts excavated from Guodian in order to give the Li ji the minimum treatment that it has long deserved. By discussing how the Guodian bamboo-slip texts are related to the transmitted version of the Li ji, I argue that the writing styles of the related excavated bamboo-slip manuscripts are the particular Li gujing- 禮古經 or Li guji- 禮古記 styles. With various versions in transmitted and excavated textual traditions available, selected textual details of the Zi yi are examined to partially illustrate how and when the Li ji came to its current shape. For the authenticity reason, the Shanghai Museum version of the Zi yi is not treated as a reliable text in the present article.

The Li and the Li ji: The Titles

Today the title “Li ji” 禮記 usually refers to the Li ji in the San Li 三禮 canons, the other two of which are the Li 禮 (or Yi li 儀禮) and the Zhou li 周禮 (also known as the Zhou guan 周官 or Institutes of Zhou), in the Confucian tradition.Footnote 9 However, the use of the term “Li ji” as a title to refer to one of the San Li canons while discussing excavated early manuscripts is confusing without providing an explanation or definition because “Li ji” was also used as the title of the Li or Yi li in early China until at least the fourth century. In other words, if we discuss early Chinese ritual texts, the title “Li ji” could mean either the Li ji or the Li (i.e. Yi li) of the San Li. This confusing situation makes the already baffling textual characteristics of the Li-related texts even more perplexing. The following examination serves two purposes: (1) to clarify what “Li ji” as a title actually refers to in early China, and (2) to provide a context for a better understanding of the Li-related excavated early manuscripts.

Ji 記, “records” or “notes,” is a particular genre of early Chinese writing. Designated as ji writings, the present sections of the Li ji were not supposed to become independent articles or book sections in the way that they currently are treated, but were supposed to be attached to the relevant parts of the Li (or the Yi li) canon. Due to the importance of these ji, which were traditionally attributed, by the “Yiwen zhi” of the Han shu, to the disciples of the qishi zi 七十子 (Seventy Disciples)Footnote 10 of Confucius, they were not only edited into an independent book titled Li ji but also canonized and shared the same importance as that of the Li during the Eastern Han 東漢 (25–220 c.e.) or at latest the Tang 唐 (618–907 c.e.) dynasty.Footnote 11

The title of the Li canon has a number of variations, the principal ones of which are Li 禮, Li jing 禮經, Shi li 士禮, Li yi 禮儀, and Yi li 儀禮. The Li and Li jing have been widely used. For example, the “Rulin zhuan” 儒林傳 of the Han shu 漢書 describes the major teachings on the Li: “the teachings on the Li include those of Da Dai, Xiao Dai, and Master Qing” (《禮》有大戴、 小戴、 慶氏之學).Footnote 12 As a usual practice, the character jing 經 (“canon”) was added to the title of a work in order to show the canonized status of the text. This is why the Li was also called Li jing, the Canon of the Rites. As a title, Li jing can be found in the Lun heng 論衡 of Wang Chong 王充 (c. 27–100 c.e.). The “Xie duan” 謝短 section of the Lun heng states, “Currently the Li jing has sixteen (sections)” 今《禮經》十六.Footnote 13 It talks about the extant sections of the Li. Evidently, both the Li and the Li jing refer to the canon of the Li, and there is no confusion about them at all.

The title of “Shi li” can be found in both the Shi ji 史記 and the Han shu. When writing about Gaotang sheng 高堂生, the “Rulin liezhuan” 儒林列傳 of the Shi ji says, “currently only the Shi li is available; Gaotang sheng is able to explain it” (於今獨有《士禮》，高堂生能言之).Footnote 14 Similarly, the “Yiwen zhi” of the Han shu reads, “when the Han dynasty arose, Gaotang sheng from Lu transmitted the Shi li in seventeen sections” (漢興，魯高堂生傳《士禮》十七篇).Footnote 15 However, the title Shi li implies that the rites specified in the canon are for shi 士 only.Footnote 16 As Liang Yusheng 梁玉繩 (1744–1819 c.e.) argued, none of the rites in the “Yan li” 燕禮 (“Banquet rites”), “Da she” 大射 (“Great archery ceremonies”), “Pin li” 聘禮 (“Guest rites”), “Gong si daifu li” 公食大夫禮 (“Rites for the gong to feast daifu”), and “Jin li” 覲禮 (“Rites for imperial meet”) of the Li classic were for shi 士; they were for zhuhou 諸侯 (“Feudal Lords”) instead.Footnote 17 However, this feature does not change the basic nature of the Li classic. As the “Hun yi” 昏義 section of the Li ji states, the whole system of rites was designed for shi, the lowest member of the early aristocracy. The “Hun yi” reads:

夫禮，始於冠，本於昏，重於喪、祭，尊於朝、聘，和於射、鄉。此禮之大體也。Footnote 18

As for the rites, they start from capping, are based on nuptials, focus on mourning and sacrificing, pay respects at court and guest rituals, and reach harmony through the interstate and local archery meets. These are the general contents of the rites.

All these rites listed above are for the shi, thus, Shi li is an appropriate title for the Li classic.Footnote 19

The “Zhong yong” 中庸 mentions “Li yi” 禮儀, which was considered the title of the Li by later scholars such as Zhu Xi 朱熹 (1130–1200 c.e.) and Sun Xidan 孫希旦 (1736–84 c.e.).Footnote 20 However, a much more popular title is “Yi li” 儀禮, which can be found in the Hou Han shu 後漢書.Footnote 21 But Duan Yucai 段玉裁 (1735–1815 c.e.) considers that the “Yi li” in the Hou Han shu is a transcription error. Quoting the Hou Han shu passage, Duan argues,

鄭君本傳曰：“鄭所注《周易》、《尚書》、《毛詩》、《儀禮》、《禮記》、《論語》、《孝經》、《尚書中候》、《乾象曆》。” 按此不應遺《周禮》。疑“ 儀禮禮記” 四字，乃 “《周官》、《禮》、《禮記》” 五字轉寫之誤。蓋《儀禮》本但稱《禮》，無 “儀” 字。漢人無稱《儀禮》者。…… “儀禮” 二字，蓋因《記》云：“威儀三千 ” 者謂《禮經》，故冠 “儀” 于“禮，”使稱說較便。大約梁、陳以後，乃有此稱。Footnote 22

Mr. Zheng's (Zheng Xuan 鄭玄) biography says, “What Zheng annotated includes the Zhou Yi, Shang shu, Mao Shi, Yi li, Li ji, Lun yu, Xiao jing, Shang shu zhonghou, and Qian xiang li.” Note: The Zhou li should not be missed here. I suspect that the four characters “Yi li, Li ji” were transcribing errors of the five characters “Zhou guan, Li, Li ji.” The Yi li was originally only called the Li, without the character “yi.” No Han dynasty individual/person used “Yi li”… The reason why the two characters “yi li” were used was because in the Ji, the Canon of Rites (Li jing) was called “three thousand august ceremonies (yi),” thus “yi” was placed preceding “li” in order to facilitate quotation. This title began to be used approximately after the Liang (502–57 c.e.) and Chen (557–89 c.e.) dynasties.

In other words, Duan believes that “Yi li” was a very late title, the usage of which did not begin until after the Liang and Chen dynasties in late sixth century. Huang Yizhou 黃以周 (1828–99 c.e.) had a similar idea, but traced the usage of the title back to an earlier date in the Eastern Jin 東晉 (317–420 c.e.) dynasty. He suggests:

鄭氏師、弟子并無《儀禮》之名也。《禮》註大題《儀禮》，當是東晉人所加。東晉人盛稱《儀禮》。Footnote 23

Neither Master Zheng nor his disciples used the title Yi li. Yi li, as the general title of the Li, had been added by Eastern Jin dynasty people. It was a very popular practice for the Eastern Jin people to use the title Yi li.

This understanding of the Yi li title has long been accepted in China.Footnote 24 Currently there is no new evidence to challenge it.Footnote 25 “Yi li” as the title of the Li classic has long been widely used.

Other titles for the Li classic also include Jing li 經禮, Qu li 曲禮 and Jin li 今禮, and so on. Both Jing li and Qu li were mentioned in one of the “Li qi” 禮器 sections of the Li ji.Footnote 26 It has been very controversial whether the Jing li refers to the Li canon. According to Zheng Xuan, Jing li 經禮 was the title for the Zhou li 周禮, and Qu li was “shi li” 事禮, which was the Jin li 今禮, i.e., the Li canon. In other words, Zheng Xuan, a second-century expert on the rites canons, did not think that the Jing li was the Li classic as the title itself seems to suggest. However, no evidence shows that Zheng Xuan ever used a single consistent title for the Li canon; neither does any evidence show what title/titles Zheng Xuan himself used for the Li canon.

Later scholars inclined to take the Jing li as the Li canon.Footnote 27 Sun Xidan argued that Zheng Xuan's interpretation of the Qu li was incorrect and that the Qu li that Zheng Xuan discussed actually refers to details of the rites rather than to the Li canon.Footnote 28 Sun offered three interpretations of the term “qu li”: (1) the meticulous details of the rites in the Yi li; (2) the “Qu li” section of the ancient Li canon; and (3) the “Qu li” sections of the Li ji (yi wei Yi li zhong zhi quzhe, yi ze gu Li pian zhi Qu li, yi ze Li ji zhi Qu li ye 一為《儀禮》中之曲折，一則古《禮》篇之《曲禮》，一則《禮記》之《曲禮》也).Footnote 29 This explains why most scholars do not consider Qu li an alternative title for the Li canon.

More confusing is that in addition to the Li, Li jing, Shi li, Li yi and Yi li, and so on, “Li ji” 禮記 was also used as an alternative title for the Li canon in early China. As the ji writings of the Li canon, the original texts of the Li ji could be divided into two groups, Da Dai Li ji 大戴禮記 and Xiao Dai Li ji 小戴禮記, which were named after their attributed editors Da Dai 大戴 (Dai De 戴德, first century b.c.e.) and Xiao Dai 小戴 (Dai Sheng 戴聖, first century b.c.e.).Footnote 30 The earliest textual evidence currently available for the title “Li ji” specifically used for the so-called Xiao Dai Li ji is Zheng Xuan's Liuyi lun 六藝論, which reads,

戴德傳《記》八十五篇，則《大戴禮》是也。戴聖傳《記》四十九篇，則此《禮記》是也。Footnote 31

Dai De transmitted the Ji in eighty-five sections, which is the Da Dai Li. Dai Sheng transmitted the Ji in forty-nine sections, which is this Li ji.

At the latest, after the “Treatise on Bibliography” (“Jingji zhi” 經籍志) in the Sui shu 隋書, recorded the “Da Dai Li ji in thirteen chapters” (Da Dai Li ji shisan juan 《大戴禮記》十三卷) and the “Li ji in twenty chapters” (Li ji ershi juan 《禮記》二十卷), the title Li ji exclusively indicated the Xiao Dai Li ji and there was no further confusion.Footnote 32 Therefore, whenever the title of Li ji is mentioned today, no scholar would take it for a text other than the Xiao Dai Li ji. This is why it is very confusing that the title of Li ji was also used to refer to the Li canon, i.e., the Yi li, in early China.

It is not clear when this confusion started. From the following three examples, we can tell that experts on early texts, such as Sima Qian 司馬遷 (?145–?86 b.c.e.), Zheng Xuan, and Guo Pu 郭璞 (276–324 c.e.), etc., used the “Li ji” to refer to the Li canon.Footnote 33 The “Kongzi shijia” 孔子世家 section of the Shi ji states that “the Shu zhuan and the Li ji came from Confucius” 《書傳》、《禮記》自孔子,Footnote 34 thus identifying Confucius as the compiler of the Li canon.Footnote 35 As we can see here, the Li canon was referred as “the Li ji.” The second example comes from Zheng Xuan's writing. In his annotation on the Shijing 詩經, he quotes the “Shaolao kuishi li” 少牢饋食禮 of the Li canon as follows:

《禮記》：“主婦髲髢。”Footnote 36

The Li ji reads, “The hostess was adorned with wig.”

Just like Sima Qian, Zheng Xuan refers to the Li canon as the Li ji. The third example shows that this practice lasted until the fourth century c.e. In his annotation to the Er ya 爾雅, Guo Pu also quoted the “You si” 有司 section of the Li canon as follows:

《禮記》曰：“厞用席。”Footnote 37

The Li ji says, “…veiling with mats.”

Since the Li and the Li ji are clearly two distinct canons in later history, some scholars simply took the above practice as a mistake. When Xing Bing 邢昺 (931–1010 c.e.) read Guo's note above, he immediately commented: “To speak of the Li ji is a mistake” (Yun Li ji zhe wu ye 云《禮記》者，誤也).Footnote 38

A simple solution to the problem is to take such usage of the title as mistaken. However, this easy solution may cast a veil over some truth concerning the early development of the texts. Both Sima Qian and Guo Pu were experts in ancient Chinese texts, and Zheng Xuan was a distinguished expert on the canons of the rites of his day. If all these experts referred to the Li canon as the Li ji in different cases, the supposition that they all made the same mistake is unconvincing. I argue that at least in and before Zheng Xuan's and Guo Pu's times, i.e. the third and fourth centuries, the usage of the title, “Li ji,” for the Li canon was not yet finalized and constant, thus there were many different titles for the Li canon as discussed above, and the “Li ji” is simply one of them. This may be due to the fact that the Li canon itself was not finalized yet, or that the title “Li ji” was not yet exclusively used for the Xiao Dai Li ji, or that some ji writings had already been treated as the Li canon in the same way that the Commentaries of the Zhou yi 周易 were treated.Footnote 39

Documents other than philosophical or intellectual writings may further prove this point. In the Luoyang ji 洛陽記, the “Li ji” was also used as the title for the Li canon. Recording the status of the Stone Classics of Xiping (Xiping shi jing 熹平石經), the Luoyang ji reads,

南行，《禮記》十五碑悉崩壞。Footnote 40

Further south, the fifteen stelae of the Li ji were all cracked and broken.

As we know, the Li ji, or the Xiao Dai Li ji, was not one of the canons that were engraved on the stelae of the Stone Classics of Xiping. The title “Li ji” in the above line from the Luoyang ji actually indicates the Li canon rather than the Li ji (the Xiao Dai Li ji) known to us today. If a non-intellectual document that has no direct textual connections with those texts by Sima Qian, Zheng Xuan and Guo Pu also has the same practice as those principal intellectual texts have, it is reasonable to consider such practice an existing practical one rather than a random textual error. If the Luoyang ji somehow had indirect or even unidentifiable direct textual connections with those intellectual or historical documents, it only further proves that such practice, i.e., using “Li ji” to refer to the Li canon, was widely accepted.

The reason why the “Li ji” was used to refer to the Li canon can be traced back to the textual features of the canon. There were different versions of the Li circulating in the Western Han dynasty (206 b.c.e.–25 c.e.): the three major ones were the two transmitted by Dai De and Dai Sheng, and the version recorded in Liu Xiang's 劉向 (79–78 b.c.e) Bie lu 別錄. Twelve of the seventeen sections of the Li canon, as Wang Baoxuan 王葆玹 lists in a table comparing the different arrangements of the seventeen sections in the three editions, have ji 記 writings attached to the end of the sections.Footnote 41 These ji attachments include writings illuminating the meanings of the rites, describing the distinct ritual systems in ancient times, introducing different ceremonies due to the changes in the systems, and recording the words of the rites in practice, and so on.Footnote 42 On the one hand, these ji writings that were attached to the text of the respective sections of the Li classic became part of the canon. On the other, they also are identified as a separate component of the text of the Li canon by being marked with the character ji. However, in the early transmission of the Li canon, the marking character ji does not always appear. For instance, the Western Han wood and bamboo-slip version of the Li canon excavated from Wuwei 武威, Gansu 甘肅 in 1958 does not have the marking character ji between the text of the canon and the ji writings in the sections of “San fu” 喪服, “Tesheng kuishi li” 特牲饋食禮 and “Yan li” 燕禮 where the ji writings were attached.Footnote 43 Although the interpretations of such phenomenon are different as argued by Chen Mengjia 陳夢家 and Shen Wenzhuo, etc., neither a character of “ji” nor a particular mark was used to separate the text of the canon and the ji writings.Footnote 44 Due to such characteristics of the text, it is not always easy for later scholars to distinguish between the primary text of the Li canon and the text of the ji writings that have already merged into the canon portion. Many of such ji writings originally come from the pre-Qin texts of the ji labeled as writings of the students of the Seventy Disciples of Confucius by the “Yiwen zhi” of the Han shu and thus represent teachings of the Sage and worthies.Footnote 45 In other words, they have the same function as the Li canon. Depending on the purpose of quoting, it is not always necessary for scholars, in either Han or post Han dynasties, to make a strict and clear distinction between the Li canon and its ji writings when quoting unless the quoting was for the purpose of jingxue 經學, the Study of Chinese Classics.Footnote 46 Consequently, in many cases, the Li ji material, which is actually the ji of the Li canon, has the same status and importance as that of the Li canon, and naturally, in some cases, the title “Li ji” can be, and actually was used to refer to the Li canon.

As Shen Wenzhuo points out, there are also cases in which the title of the “Li” is used to indicate what we now refer to as the Li ji or the Xiao Dai Li ji.Footnote 47 According to Zheng Xuan, the “Ben sang” 奔喪 and “Tou hu” 投壺 sections of the Li ji are both formal sections, zhengpian 正篇, of the Li canon.Footnote 48 In other words, the usage of the titles of the “Li” and “Li ji” were interchangeable to some degree in early China.

Such unusual usage of the title “Li ji” deserves proper scholarly attention, especially when some scholars simply took it as a textual error. Through the above brief examination, the confusing usage of the titles of the Li and the Li ji not only is clarified, but also the textual background of such usage is brought to our attention. What I would like to further argue is that the textual background of such usage gives us an important perspective for understanding the textual features and style of the Li ji related bamboo-slip manuscripts excavated from Guodian.

The Guodian Texts, the Li gujing and the Li guji

The Warring States (476–221 b.c.e.) period Chu bamboo-slip manuscripts excavated from Guodian include many texts that are related to the Li ji. In addition to the Zi yi, which is basically identical with the “Zi yi” section of the Li ji, other Guodian texts, such as Zun deyi 尊德義, Liu de 六德, and Cheng zhi wen zhi 成之聞之, have quotations from or sentences identical with or similar to those in different sections of the Li ji. Through an examination of how these passages in the Guodian texts connect to the Li ji, I propose that the nature of these textual connections provides a significant context for understanding the nature and style, as well as the intellectual content, of the excavated bamboo manuscripts in question. I argue that the textual characteristics of the excavated material are similar to those of the Li gujing and the Ji 記 listed in the “Yiwen zhi” of the Han shu, as well as in other transmitted texts, and that the “Li gujing” should be better considered a name of a style of writing rather than a title of a text.Footnote 49 I further argue that many excavated Warring States period bamboo texts, in particular, many Guodian texts, could be considered such Li gujing-style, or as I propose below, Li guji-style writings.Footnote 50

The ways in which the Guodian passages connect to the Li ji can be summarized as follows.

1. Basically identical wording

Just like the Guodian bamboo version of the Zi yi, which is essentially identical with the transmitted Li ji version with slight variations of the usage in a limited number of graphs and particles, some Guodian sentences or phrases that relate to the Li ji are basically identical with their counterparts in the relevant sections of the Li ji. The Zun deyi from Guodian has two sentences that are identical in wording with sentences in the “Zi yi” of the Li ji.

In the Li ji version, there is an introductory formula, “The Master says” Zi yue 子曰, that does not appear in the Guodian Zun deyi version; ling 令 is actually the same character as the ming 命 in the Guodian version; the particle er 而 that appears in the Guodian version is omitted; and the final particle yi 矣, which does not change the idea of the sentence, cannot be found in the Guodian passage. Clearly, the textual differences do not change the substantive ideas in any way, and these two quotations are basically identical.

2. Basically identical wording or idea with a different order to the sentences

Sometimes, the wording in the Guodian texts is basically the same as that in the parallel Li ji passages. But it is not unusual that the order of the sentences in the two versions is different. One sentence in the Zun deyi of the Guodian texts has the same idea as and is generally identical in wording with that in the “Qu li” shang 曲禮上 section of the Li ji.

In the Guodian version, the idea is expressed in the sequence of “the punishments” and “the rules of ceremony,” but in the Li ji version the idea is presented in the sequence of “the rules of ceremony” followed by “the punishments.” However, the status of the people is stated in less formal legal terminology in the Guodian version (junzi “exemplary people” and xiaoren “petty people”) than that in the Li ji version (shuren “common people” and daifu “great officers”).

3. Different wording with the identical idea

It is very common to be able to identify the same idea between some sentences in the Guodian texts and those in the Li ji sections although their wording may be distinct. In these cases, usually we may find the same phrases or similar phrases that convey the same ideas. The following two passages from the Guodian text Liu de 六德 and the Li ji section “Jiao te sheng” 郊特牲 represent such an example.

Both texts consider “wisdom” (zhi 智/知) the virtue of a “man” (fu 夫). Although the Li ji line does not correspond to what is in the Guodian version word by word, it is not hard to identify this same understanding from “A man is the person who leads others by his wisdom.” “Leads others by his wisdom” is also the identical phrase that both passages use. The following case discussing the virtue of a “wife” (fu 婦) seems a better example since the two versions basically share the same statement on the virtue of a wife.

In both examples, that “faithfulness is the virtue of a wife” is clearly stated in almost identical words in the two passages while the Guodian version has some particles in the sentence. Although each quote seems to have a different context, the idea about the virtue of a wife is the same. It is possible that they do not share the same textual source, but the ways in which they are presented and the idea that they express are evidently the same.

4. Different wording with a similar or related idea

In the Xing zi ming chu 性自命出 of the Guodian texts, one line talking about “self-cultivation” and “benevolence” has a close relationship with what the “Zhong yong” 中庸 section of the Li ji states. The two related passages read,

In the “Zhong yong” passage, “self-cultivation” (xiu shen) and “being benevolent” (ren) are connected by the “Way” or pursuing the “Way”; pursuing the “Way” or following the “Way” is the highest level of self-cultivation. In the sentence of the Xing zi ming chu text, the “Way” is not mentioned, but “self-cultivation” (xiu shen) is directly connected with the “highest benevolence” (zhi ren); the “highest benevolence” is, of course, close to and reflects the Way.

In another example from the Guodian manuscripts, one sentence in the Zhong xin zhi dao 忠信之道 text is almost identical with one in the “Biao ji” 表記 section of the Li ji. They read,

The only difference is in the first characters in the above quotes. In the Guodian text, it is “heart” (xin), and, in the Li ji text, it is “feeling” (qing). The “heart” is where the emotions (“feeling”) are felt. The connection and similarity between the two words are self-evident in this case.

5. Different wording with related ideas

The Cheng zhi wen zhi 成之聞之 of the Guodian texts has the following statement, which is closely related to a passage in the “Fang ji” 坊記 section of the Li ji. From the comparison below, it is not difficult to see both their differences and similarities in the idea.

The differences between the two passages are clear: in the Cheng zhi wen zhi passage, the negative result of the yielding action is not specified; but in the “Fang ji” passage, the problem of yielding to inferiors or mean people is clearly stated—common people may offend either the noble or the ruler. It is apparent that the above Guodian passage is related to the Li ji passage through both partially identical wording, i.e., “on the sitting mat” (renxi zhi shang), “in seat at court” (chaoting zhi wei) and “yield” (rang), and the related ideas. Although the wording as a whole is distinct, the related ideas are defined by the partially identical texts.

The preceding comparison builds up the textual connection between some Guodian and Li ji texts. According to the “Yiwen zhi,” the Li gujing, just like the Guodian texts, was a group of ancient texts written in a script of archaic-form Chinese characters, i.e., one of the local Six States scripts in the Warring States period before the Chinese writing system was unified by the First Emperor of Qin, Qin Shihuang 秦始皇 (259–210 b.c.e.). Although the “Yiwen zhi” of the Han shu does not make it clear whether or not the Ji was written in an archaic Warring States period script, it does indicate that a Ji in 131 sections was authored by students of the Seventy Disciples of Confucius,Footnote 71 in other words, they belong to the Warring States period. In the Bie lu 別錄, Liu Xiang 劉向 (77–76 b.c.e.) stated, “The archaic script Ji is in 204 sections.”Footnote 72 Since Liu Xiang's bibliographical work was the base of the “Yiwen zhi” of the Han shu, it is reasonable to believe that the Ji in 131 sections recorded in the “Yiwen zhi” was also in an archaic script, guwen 古文,Footnote 73 although 73 sections of the Ji in archaic script had been either lost or deleted by the time of editing the “Yiwen zhi,” since those Ji were specifically noted as in the archaic script by Liu Xiang. Wang Guowei 王國維 simply took them as guwen ji 古文記.Footnote 74 Professor Li Xueqin has the same understanding.Footnote 75 In order to better note the similar textual features between the Ji and the Li gujing, following the convention in the “Yiwen zhi,” I call the Ji the “Li guji.” That is to say, if the Li gujing is a group of ancient texts of the Li canon in archaic script, Li guji is also a group of ancient texts of Ji writings of the Li authored by the students of the Seventy Disciples of Confucius. The above textual examination of the excavated Guodian texts in archaic script and the Li ji counterparts actually presents us with the differences between the guwen Ji and the received Li ji. In other words, the Guodian bamboo texts make the original appearance of the Li guji-style writings, which were originally called “Li gujing” in the “Yi wenzhi” of the Han shu, available to us.

As for the record of the Li gujing, two versions of the “Yiwen zhi” passage merit further examination. The more popular Zhonghua shuju 中華書局 edition (1962) reads,

《禮古經》者，出於魯淹中及孔氏，（學七十）[與十七]篇文相似，多三十九篇。Footnote 76

The Li gujing, discovered from Yanzhong of the Lu and the Confucius's family, is similar (to the Seventy Disciples) [to the seventeen] sections,Footnote 77 but has thirty-nine more sections.

The pre-modern Wuying Dian 武英殿 edition (1739) reads,

《禮古經》者，出於魯淹中，及孔氏學七十篇文相似，多三十九篇。Footnote 78

The Li gujing, discovered from Yanzhong of the Lu, is similar to Confucius's Seventy Disciples’ essays, but has thirty-nine essays more.

As we can see, the difference is how we read ji Kongshi xue qishi 及孔氏學七十. Following Liu Chang's 劉敞 (1019–68 c.e.) interpretation, the contemporary Zhonghua shuju edition considers that xue qishi 學七十 (“Seventy Disciples”) is a typographical error for yu shiqi 與十七 (“with/to seventeen”); thus the revised number “seventeen,” rather than the “Seventy Disciples” in its original writing, goes with the number of the seventeen sections of the Li canon.Footnote 79 I argue that such change needs to be reconsidered. As a rule, we contemporary scholars are not supposed to simply change the original writing of an ancient text in order to fit in to a more reasonable or acceptable reading based on our conventional knowledge. We have to read what an early text “wrote” rather than what we think it “intended” to write.Footnote 80 Also, although qishi 七十 (“seventy”) is similar to shiqi 十七 (“seventeen”), xue 學 (disciple/study) is not close to yu 與 (to/with/and) at all. Furthermore, according to my reading, Kongshi xue qishi 孔氏學七十 (“Confucius's Seventy Disciples”) in the original reads much better than Kongshi, yu shiqi 孔氏，與十七 (“Confucius's family, to the seventeen”). I argue that such change was not only based on the numbers of existent sections (seventeen) in the Li canon, but also was due to the ideas of the school of “Doubting Antiquity,” since the original wording mentions Confucius's Seventy Disciples’ writings, which were and may still be considered as unacceptable or impossible in the transmitted textual tradition.

In fact, the Li gujing has been considered the writings that were likely authored by Confucius's Seventy Disciples since the Western Han dynasty when it was retrieved. The “Jingji zhi” of the Sui shu reads,

古經出於淹中，而河間獻王好古愛學，收集余燼，得而獻之，合五十六篇，并威儀之事。Footnote 81

The gujing appeared in Yanzhong. Since King Xian of Hejian was fond of antiquity and loved learning, he had the remains collected from the ashes, acquired (them) and presented them (to the emperor). The total was fifty-six sections, including the material on the august ceremonies.

This “gujing” is the Li gujing, and the so-called “august ceremonies” (weiyi) is an alternative word for the rites, li. The biography of King Xian of Hejian in the Han shu gives more details of these ancient texts. It reads,

獻王所得書皆古文先秦舊書，《周官》、《尚書》、《禮》、《禮記》、《孟子》、《老子》之屬，皆經傳說記，七十子之徒所論。Footnote 82

What King Xian acquired were all pre-Qin ancient texts in archaic scripts, such as the Zhou guan, the Shang shu, the Li, the Li ji, the Mencius, and the Laozi, and so on. All were canons, commentaries, interpretations, and ji writings: what the Seventy Disciples had discussed.Footnote 83

Unlike the Sui shu, the Han shu passage does not specify that the Li gujing was among the ancient books that King Xian of Hejian obtained. However, in the “Liuyi lun” 六藝論, Zheng Xuan clearly states “King Xian of Hejian's Guwen li in fifty-six sections” 河間獻王《古文禮》五十六篇.Footnote 84 The Guwen Li, the Li in the Archaic Script, is the Li gujing, the Li Canon in the Archaic Script. Although the Sui shu is a late document, its statement on the Li gujing in fifty-six sections has supporting evidence from Zheng Xuan.

According to both the Han shu and the Sui shu, what King Xian of Hejian obtained includes both canons and ji writings, the latter of which are commentaries and annotations by Confucius's Seventy Disciples and/or their disciples. Those ancient texts include the Li and the Li ji. In addition to the canons, the form of these early writings was commentary and ji-writing by the disciples or followers of Confucius. In the above Han shu biography of King Xian, it clearly states that the Li ji or the ji writings, etc., that King Xian acquired were “all pre-Qin ancient texts in archaic script.” This provides further evidence for the proposed term “Li guji” to be a reasonable alternative for the Ji in 131 sections listed in the “Yiwen zhi.”

The bamboo-slip texts excavated from Guodian include three bundles of the Laozi 老子, most of which correspond to the transmitted version, as well as a dozen essays dated back to the times of the Seventy Disciples and their disciples. Professor Li Xueqin argues that those texts are closely related to the Li ji.Footnote 85 From the preceding textual comparison between the Guodian texts and the Li ji, we can see how the Guodian texts are close to the Li ji texts and how they are textually related. This implies possible connections between those related passages in textual transmission. As for the themes, the Guodian texts discuss various aspects of the “material concerning the august ceremonies,” including, but not limited to, the connection between the rules of ceremony and the punishments, the virtue of a man and that of a wife, the relationships between male and female, father and son, and righteousness and ceremonies, as well as the issue of yielding to inferiors, and so on. They are all related to the Li and the Li ji. With regard to the textual features, some of the Guodian texts are identical with their Li ji counterparts; some are identical with the related Li ji passages in either the wording or the meanings or concepts, with some slight variations.

In my opinion, such thematic and textual features, which belong to the Li guji, are likely to have been similar to those of the Li gujing although there is no evidence available to us today. Just like the usage of the titles of the “Li” and “Li ji” in early China, the title of the Li gujing, if it is a title, is not a definitive one at all. The “Li gujing,” the “Li Canon in Archaic Script,” is more a description of those ancient retrieved texts than an actual title given to the material by their original author(s) or the copyists. So is the “Li guji.” I suggest that while we continue to use the “Li gujing” as the title for those texts retrieved from Yanzhong in accordance with tradition, the “Li gujing,” as well as the “Li guji,” should more properly be considered names of a style of writing (wenti 文體), for those archaic pre-Qin intellectual writings that are related to either the Li or the Li ji. In other words, I argue that many of the Guodian texts are the Li guji- or Li gujing-style writings. They can be roughly considered to be Li guji or Li gujing-related texts in a general sense; as for whether or not they are the lost Li guji or even Li gujing sections that were mentioned in the “Yiwen zhi” of the Han shu, it awaits further specific evidence.

“Zi yi”: A Comparison between the Li ji and the Guodian Versions

Being aware of the above textual features of many Guodian bamboo-slip texts, i.e., the Li ji related Li guji-style writings, we are better prepared to examine the recently excavated texts in the proper context. Strictly speaking, both the so-called Li guji and the Li gujing only survive as names in the received textual tradition. Even if we were certain that the Chu bamboo-slip texts in question are essays written precisely in the Li guji or Li gujing style, we would only be able to know more about the Li guji and the Li gujing by studying the excavated Chu manuscripts because no extant received Li guji or Li gujing texts are available. However, the availability of both the excavated version and the transmitted version of the Zi yi or “Zi yi” texts changes the situation, and now we are able to know more about the text and textual features of both the excavated manuscript and the received text.

In the following, I compare two passages from pericopes 17 and 14 of the “Zi yi” in the received Li ji version with their counterparts in the Guodian bamboo-slip version.Footnote 86 In the sequence in which they appear in the Guodian version of the Zi yi,Footnote 87 I separate each passage into three groups in order to facilitate the comparison. By examining the textual differences and similarities between the excavated and transmitted versions below, I argue that the received Li ji version of the “Zi yi” is dated later than the excavated bamboo-slip version of the text.

1. Pericope 17

Group 1.1:

Group 1.2:

Group 1.3:

It is readily apparent that the texts of the bamboo-slip version are more concise than the Li ji version in all the quotations. In some cases, as I discussed before, the Li ji version seems to explain and comment on the wording of the bamboo-slip version.Footnote 97 In group 1.1, the bamboo-slip version does not have the second sentence that is present in the Li ji version, “If the heart is sober, the body would be relaxed; if the heart is solemn, the appearance would be respectful.” This second sentence further explains the metaphor prior to it, stating the connection between the “heart” (xin 心) and the “body” (ti 體). In group 1.2, the bamboo-slip version omits the pronoun “it” (zhi 之) after the verb hao 好, “is fond of” and uses a more succinct syntax than the Li ji version does, and thus are much more terse in expression.Footnote 98 In group 1.3, it is apparent that the Li ji version seems to interpret what is in the Guodian bamboo-slip version by revising and enlarging the bamboo version statement. That “the heart decays due to the body” is not explicit enough, in particular the word “decay” (fa 法) in the Guodian version is usually read as “model” or “law.” By stating that “the heart is preserved by the body, and is also wounded by the body” in two separate phrases in the Li ji version, the meaning in the Li ji version is more evident than that in the bamboo-slip version. Similarly, that “the ruler survives due to the people, and also perishes due to the people” in the Li ji version also gives a more complete statement, emphasizing the aspect that “the ruler survives due to the people,” than that in the bamboo-slip version, which only states that “the ruler perishes due to the people.”

It is interesting to compare these texts with that in the Shanghai Museum version. The last sentence in the Shanghai Museum version, i.e., “thus the ruler perishes” (jun yi wang 君以亡), likely accidentally misses the character min “people” during the copying process as Shaughnessy suggested. In this case, yi 以 of jun yi wang 君以亡 should still mean “due to” as it does in the Guodian version. Thus the translation of the sentence jun yi wang 君以亡 actually should be “while the ruler perishes due to…” The problem is, however, if the character min had actually been missed during the inscribing process, why the missing character was not made up during the proofreading process. It is evident that the Shanghai Museum bamboo-slip version of the Zi yi was proofread. On slip 11, qi qi 丌= of ru qi qi fuke jian 女丌丌弗克見 had been missed during the copying process, but was added later by squeezing in between ru and fu.Footnote 99 Similar proofreading and correction can also be identified on slip 13, where yi of zi yi ai 慈 was also a later addition during proofreading.

2. Pericope 14

Group 2.1:

Group 2.2:

Group 2.3:

In the above examples, we see further evidence proving that the Chu bamboo-slip version is more concise than the Li ji version, and that the Li ji version text seems to provide commentaries on and explanations of the Chu bamboo version. We also see evidence suggesting that the Li ji version of the text must be dated later than the excavated Chu bamboo version. In groups 2.2 and 2.3, it is self-evident that the passages of the bamboo-slip version are concise and without extended expositions, as the Li ji version has more phrases and explanations. As seen in group 2.3, the passage of the Li ji version seems to be offering more detailed explanations.

In group 2.2, the Li ji version reads “the courtiers cannot be treated incautiously; they are the guides for the people,” which parallels “the ministers cannot be disrespected; they are the models for the people.” No matter whether this second half of the parallelism is a later addition to the Li ji version or comes from a source different from the Guodian version, it is evidence showing the late date of the Li ji version. Earlier versions usually do not need such rhetorical additions; neither do they need further explanations of a similar point. In group 2.3, the Li ji version goes further beyond yi xiao mou da 以小謀大 (“plan big (issues) with petty (people)”), which was expanded to or supported by yi yuan yan jin 以遠言近 (“does he discuss (issues) close at hand with those who are at a distance”) and yi nei tu wai 以內圖外 (“does he plot the external (affairs) with those inside (i.e., eunuchs, wives and court favorites”), etc. Such commentarial additions and rhetorical flourishes in the Li ji version present further explanations and support the earlier argument yi xiao mou da 以小謀大, thus suggesting a later date for the Li ji version.

In group 2.1, the phrase “the common people are not at peace” 百姓不寧 in the Li ji version does not help with the following consequential point of “insufficient sincerity and respect” 忠敬不足. That is to say, it does not make much sense for baixing bu ning 百姓不寧 (“the common people are not at peace”) to be placed in front of zhongjing bu zu 忠敬不足 (“insufficient sincerity and respect”). I argue that baixing bu ning 百姓不寧 in the Li ji version in group 2.1 is an error in transcribing bangjia zhi bu ning 邦家之不寧 (“the state is not at peace”) from a source that is close to the bamboo-slip version, due to either a mistake in written transmission or a memory error in oral transmission or a compound error of similar nature. These two phrases share the same word bu ning 不寧 (“not at peace”) that could be easily confused during textual transmission without sufficient care.

Comparing with the Guodian version, we can see that the Li ji version is short of a portion in the second half of the sentence. This is not the result of misarranged bamboo slips (cuojian 錯簡) because it is a problem of only four characters. It is more likely to be an eye-skip error during the copying process. Clearly, both the meaning and the syntax of the sentences in the Guodian version read smoothly. On the other hand, both the logic and the syntax of the sentences in the Li ji version have obvious problems—not only does the clause “the common people are not at peace” 百姓不寧 fit better in the second half with “the ministers are not well managed” 大臣不治 and “the courtiers will collude with each other” 邇臣比矣, but also a balanced 3-by-3 clause structure is distorted into a less felicitous 4-by-2 clause structure for the whole sentence. I consider it evidence to date the Li ji version to a date later than that of the Chu bamboo-slip version. If the Li ji version were an earlier version, it is difficult to explain why the author(s) or editor(s) of the Li ji, who must have had been well trained scholars in Confucian texts, did not revise such elementary stylistic problems in writing. However, if the Li ji version is a received later version and the logical and syntactic problems were generated during the process of textual transmission, it would be quite reasonable for the Li ji editor(s) or transmitter(s) to stick to what they had received and not to make any changes or revisions based on their own understanding. Compared to the Li ji version, the Guodian version with its strict logic and parallelism reads more like a version that was carefully written or edited rather than one that was transmitted with concomitant transmission problems. Therefore, the Guodian version with fewer textual problems is an earlier version than the Li ji one. Coincidentally, an example in group 2.1 provides further interesting evidence for this interpretation.

In group 2.1, the passage in the Shanghai Museum version is incomplete due to, likely, the break of the bamboo slip. According to the available inscriptions, we may reasonably conclude that this passage is identical with the one in the Guodian version. As an expert in the “Zi yi” text, the Shanghai Museum editor would surely have been able to fill the lacuna easily based on both the received Li ji and the excavated Guodian versions, but chose not to do so. What the editor did was to punctuate the text based on the wording of the extant text of the Li ji, rather than relying on the words that probably were there originally, as seen from the Guodian version. Thus we see a period after bangjia zhi bu ning ye 邦家之不寧也 (“the state is not at peace”) rather than a comma as it ought to have been.Footnote 107 Just like the editor(s) of the Li ji who stuck to the wording of baixing bu ning 百姓不寧 in group 2.1, the Shanghai Museum editor presumed that the wording in the lacuna in the missing portion of the bamboo slip was the same as the wording in the extant Li ji, rather than that in the Warring States period bamboo-slip version excavated from Guodian. This fact reflects one common truth: later editors are usually inclined to stick to what is available in writing and prefer not to make even seems to be obviously correct changes. That is to say, the version that retains textual problems and does not make even basic corrections is usually a later version. In group 2.1, the excavated bamboo-slip version of the Zi yi is an earlier version, and the Li ji version that retains evident textual problems must be dated later.

Rethinking the Li ji

From the perspective of the examination of both the excavated and the transmitted texts, I would like to make the following observations on the Li ji.

The Li ji is a canon. It is a canon in the form of ji writings that were supposed to have been attached to the Li canon, but eventually it actually became an independent classic. That its title, “Li ji,” was previously used as the title for the Li canon reflects the canonical nature and status of the Li ji as well as the textual traits of the Li canon—the Li canon includes the ji writings and the canon itself does not set up its exclusive title in its early stage.

The Li ji related excavated texts, such as those excavated from Guodian, are the Li guji-style writings. They are related to the Li and the Li ji as well as to Confucius's Seventy Disciples and their disciples, and bear the characteristics of the Li gujing. Some of those Li guji-style writings in the Warring States period could have become part of the Li gujing in the Han dynasty, no matter whether or not they survive today in the form of one line or one passage in the excavated texts. The opposite textual transformation could also be true: some of those Li gujing category writings in the Warring States period, i.e., early Li canons, could have become part of the Li guji thus part of the Li ji later. Such textual transformation is perhaps also partially Zheng Xuan's understanding or reasoning behind his judgment that some sections of the Li ji were originally the formal sections of the Li canon. Like the “Ji in 131 sections” listed in the “Yiwen zhi” of the Han shu, those Li ji related Li guji-style Guodian bamboo-slip texts should be the sources, either directly (like, perhaps, the Zi yi) or indirectly (like those ideas, lines or passages), of the Li ji or Xiao Dai Li ji.

The Li ji has extensive sources dating back to at least the Warring States period. According to the Li guji-style Zi yi text, it is evident that the material in the Li ji comes from creditable pre-Qin textual sources rather than being textual forgeries made by post-Qin scholars. Some source texts, such as the Zi yi, had clearly been finalized and widely circulated in the Warring States period as we can see from the excavated bamboo-slip manuscripts. Such textual material, including many of those that are not discussed in this article, reflects the ideas and scholarship of the Seventy Disciples and their disciples. Although more specific textual evidence is required to prove the connection between the attributed authorship or editorship of the Li ji sections and the disciples or disciples of the disciples of Confucius, more and more support for the possibility of such authorship or editorship is now available in the excavated bamboo-slip manuscripts.

Based on the evidence of the variations between the excavated texts and the transmitted Li ji text, there is no doubt that the Li ji text came to its current shape as a result of later edited work. Such edited work polished the wording, expanded the meaning, modified quotations of earlier canonical writings, and corrected or made adjustments that preserved copying errors of individual graphs or the mis-arrangement of the bamboo slips. Based on the comparisons between different versions of the Zi yi, I am convinced that many excavated Warring States period bamboo texts that relate to the Li ji are dated earlier than their counterparts in the Li ji text. This does not mean that the received Li ji sections have higher textual value than the excavated manuscripts do. On the contrary, the excavated manuscripts could have many valuable early textual sources that the transmitted Li ji has lost.

References

1. bowuguan, Jingmen shi, ed., Guodian Chu mu zhujian 郭店楚墓竹簡 (Beijing: Wenwu, 1998)Google Scholar; Chengyuan, Ma 馬承源, ed., Shanghai bowuguan cang Zhanguo Chu zhushu (yi) 上海博物館藏戰國楚竹書（一） (Shanghai: Shanghai guji, 2001)Google Scholar, as well as the following volumes of the series published in the following years. However, the authenticity of the relevant bamboo-slip manuscripts in the Shanghai Museum collection is to be further examined and determined.

2. Professor Li Xueqin 李學勤 argues that some Guodian bamboo texts, such as Zi yi 緇衣, Wu xing 五行, Cheng zhi wen zhi 成之聞之, Zun de yi 尊德義, Xing zi ming chu 性自命出, and Liu de 六德, belong to the Zisi 子思 (483–402 b.c.e.) School and quite possibly constitute, or are at least related to, the work known as the Zisi 子思 or Zisizi 子思子, listed in the “Yiwen zhi” 藝文志 of the Han shu 漢書, ed. Gu, Ban 班固 (Beijing: Zhonghua, 1962), 30.1724Google Scholar. See Xueqin, Li, “The Confucian Texts from Guodian Tomb Number One: Their Date and Significance,”in The Guodian Laozi: Proceedings of the International Conference, Dartmouth College, May 1998, ed. Allan, Sarah and Williams, Crispin (Berkeley: The Society for the Study of Early China and The Institute of East Asian Studies, University of California, Berkeley, 2000), 107–11Google Scholar; see also Xueqin, Li, “Jingmen Guodian Chu jian zhong de Zisizi” 荊門郭店楚簡中的《子思子》, Wenwu tiandi 文物天地 1998.2, 28–30Google Scholar; Xueqin, Li, “Guodian jian yu Rujia jingji” 郭店簡與儒家經籍, Renmin zhengxie bao 人民政協報, August 3, 1998Google Scholar. However, a number of scholars have questioned Li's argument; see Sarah Allan and Crispin Williams, eds., The Guodian Laozi, 179–82. Liu Lexian 劉樂賢 strongly argues for a connection between the Xing zi ming chu and the Zisizi; see his “Xing zi ming chu de xuepai xingzhi” 《性自命出》的學派性質, International Research on Bamboo and Silk Documents: Newsletter 國際簡帛研究通訊 3 (2000), 1–3Google Scholar.

3. In this article, the titles of the texts are italicized if they are considered independent works, and are in quotation marks if they are considered one chapter or section of an independent work. Before I conclude whether or not a title belongs to a particular book, I treat it as an independent work. For example, the bamboo text Zi yi is italicized because I am unable to prove that it is one section of the Zisizi, and the transmitted text “Zi yi” is in quotation marks because it is a section of the Li ji. I use “pericope” for jie 節, “sub-section” for zhang 章, “section” for pian 篇, “chapter” for juan 卷, “volume” for ce 冊, and “book” for shu 書 in this article. See also McDermott, Joseph P., “The Ascendance of the Imprint in China,” in his A Social History of the Chinese Book: Books and Literati Culture in Late Imperial China (Hong Kong: Hong Kong University, 2006), 43–81Google Scholar.

4. Representative works include, but are not limited to, Zongyi, Rao 饒宗頤, “Zi yi ling jian” 緇衣零簡, Xueshu jilin 學術集林 9 (1996), 66–68Google Scholar; Hao, Peng 彭浩, “Guodian Chu jian Zi yi de fenzhang ji xiangguan wenti” 郭店楚簡《緇衣》的分章及相關問題, Jianbo yanjiu 簡帛研究 3 (1998), 44–49Google Scholar; Wen, Xing 邢文, “Chu jian Zi yi yu Xian-Qin lixue” 楚簡《緇衣》與先秦禮學, in Guodian Chu jian guoji xueshu yantaohui lunwen ji 郭店楚簡國際學術研討會論文集, ed. Wuhan daxue Zhongguo wenhua yanjiuyuan (Wuhan: Hubei renmin, 2000), 155–64Google Scholar; Liu Xinfang 劉信芳, “Guodian jian Zi yi jiegu” 郭店簡《緇衣》解詁, in Guodian Chu jian guoji xueshu yantaohui lunwen ji, 165–81; Boltz, William G., “Liijih ‘Tzy I’ and the Guodiann Manuscript Matches,” in Und folge nun dem, was mein Herz begehrt: Festschrift fur Ulrich Unger zum 70. Geburtstag, ed. Nagel-Angermann, Monique, Hong, Yu, and Giele, Enno (Hamburg: Hamburg Sinologische Gesellschaft e.V., 2002), 209–21Google Scholar; Wei, Chen 陳偉, “Shangbo Guodian erben Zi yi duidu” 上博郭店二本《緇衣》對讀, in Shangbo guancang Zhanguo Chu zhushu yanjiu 上博館藏戰國楚竹書研究, ed. Mingchun, Liao 廖名春 and Yuanqing, Zhu 朱淵清 (Shanghai: Shanghai shudian, 2002), 417–25Google Scholar; Yu Wanli 虞萬里, “Shangbo jian Guodian jian Zi yi chuanben hejiao shiyi” 上博簡郭店簡《緇衣》傳本合校拾遺, in Shangbo guancang Zhanguo Chu zhushu yanjiu, 426–39; Feng Shengjun 馮勝君, “Du Shangbo jian Zi yi zhaji erze” 讀上博簡《緇衣》劄記二則, in Shangbo guancang Zhanguo Chu zhushu yanjiu, 448–55; Hanyi, Xia 夏含夷 (Shaughnessy, Edward L.), “Shilun Zi yi cuojian zhengju jiqi zai Li ji ben Zi yi bianzhuan guocheng zhong de yuanyin he houguo” 試論《緇衣》錯簡證據及其在《禮記》本《緇衣》編纂過程中的原因和後果, in Xin chutu wenxian yu gudai wenming yanjiu 新出土文獻與古代文明研究, ed. Weiyang, Xie 謝維揚 and Yuanqing, Zhu (Shanghai: Shanghai daxue, 2004), 287–96Google Scholar; Xigui, Qiu 裘錫圭, “Zhongguo gudian xue chongjian zhong yinggai zhuyi de wenti” 中國古典學重建中應該注意的問題, in Zhongguo chutu guwenxian shijiang 中國出土古文獻十講 (Shanghai: Fudan daxue, 2004), 1–16Google Scholar; Kern, Martin, “Quotation and the Confucian Canon in Early Chinese Manuscripts: The Case of ‘Zi Yi’ (Black Robes),” Asiatische Studien LIX.1 (2005), 293–332Google Scholar; Shaughnessy, Edward L., “Rewriting the Zi Yi: How One Chinese Classic Came to Read as It Does,” in his Rewriting Early Chinese Texts (Albany: State University of New York Press, 2006), 63–130Google Scholar; Wanli, Yu, Shangbo guancang Chu zhushu Zi yi zonghe yanjiu 上博館藏楚竹書《緇衣》綜合研究 (Wuhan: Wuhan daxue, 2009)Google Scholar; Chan, Shirley, “The Ruler/Ruled Relationship in the Ziyi (Black Robe) Contained in the Newly Excavated Guodian Chu Slip-Texts,” Journal of Asian History 43.1 (2009), 19–30Google Scholar; Jiepeng, Shi 史傑鵬, “Cong Guodian he Shangbo jian Zi yi de jitiao jianwen tan jinbeng Zi yi de xingcheng” 從郭店和上博簡《緇衣》的幾條簡文談今本《緇衣》的形成, Chuantong Zhongguo yanjiu jikan 傳統中國研究集刊 9–10 (2012), 160–72Google Scholar; Cook, Scott, The Bamboo Texts of Guodian: A Study and Complete Translation, Volume 1 (Ithaca, NY: Cornell University, 2012), 355–418Google Scholar.

5. Ye, Hong 洪業 (Hung, William), “Li ji yinde xu” 禮記引得序, in Index to Li Chi 禮記引得, Harvard-Yenching Institute Sinological Index Series 27 (Taipei: Chinese Materials and Research Aids Service Center, 1966), i–xlGoogle Scholar.

6. William G. Boltz concisely describes the situation as follows: “Even the name Li chi [Li ji] 禮記 in Han texts refers to the Li ching [Li jing], i.e., the I li [Yi li], together with the incorporated chi ‘records’; the term does not refer to what is known today as the Li chi.” In Early Chinese Texts: A Bibliographical Guide, ed. Loewe, Michael (Berkeley: Society for the Study of Early China & The Institute of East Asian Studies, University of California, Berkeley, 1993), 234Google Scholar.

7. With regret, I would like to take this opportunity to announce that my previous publications and scholarship on the bamboo-slip version of the Zi yi collected at Shanghai Museum are no longer valid unless evidence for its authenticity is available. As for my calligraphic approach to disprove the authenticity of forged excavated texts, please refer to my articles such as “Chinese Paleography, Calligraphy, and Pattern Recognition: Styles and Scripts in Excavated Ancient Chinese Documents,” Proceedings: 11th International Conference on Document Analysis and Recognition (Los Alamitos, California, Washington, and Tokyo: IEEE Computer Society, 2011), 951–56Google Scholar; “Zheda cangjian bianwei (shang): Chu jian Zuozhuan” 浙大藏簡辨偽（上）——楚簡《左傳》, Guangming Daily 光明日報 May 28, 2012, 15; “Zheda cangjian bianwei (xia): Zhanguo shufa” 浙大藏簡辨偽（下）——戰國書法, Guangming Daily 光明日報 June 4, 2012, 15; “Zheda cangjian zai bianwei – Wenben fuyuan de guanlianxing yu Zheda weijian zai pipan” 浙大藏簡再辨偽——文本復原的關聯性與浙大偽簡再批判, Guangming Daily 光明日報 June 25, 2012, 15; “Chujian shufa de bifa yu tishi – Da Liu Shaogang xiansheng” 楚簡書法的筆法與體勢——答劉紹剛先生, Guangming Daily 光明日報 July 2, 2012, 15; “Qin jiandu shufa de bifa: Qin jiandu shuxie jishu zhenshixing fuyuan” 秦簡牘書法的筆法：秦簡牘書寫技術真實性復原, Jianbo 簡帛 8 (2013), 439–50Google Scholar; “Calligraphic Styles between Authentic and Forgery Bamboo-slip Manuscripts,” Manuscript Cultures, “Lun lishi: Qin Han jianbo lishi de zai kaocha” 论隶势：秦汉简帛隶势的再考察, conference volume of the International Conference on Excavated Manuscripts and Early Chinese Civilization (Shanghai: Zhongxi, 2014), forthcoming.

8. Ma Chengyuan, Shanghai bowuguan cang Zhanguo Chu zhushu (yi), 54, 59, and 68.

9. The term san Li was coined by Zheng Xuan 鄭玄 (127–200 c.e.). See Xuan, Qian 錢玄, San Li tonglun 三禮通論 (Nanjing: Nanjing shifan daxue, 1996), 55Google Scholar.

10. Qishi zi or the “Seventy Disciples” was used so often in the Chinese textual tradition that it even generated some textual confusion and argument as we will discuss below. In this article, Seventy Disciples is used as a proper noun.

11. According to Lu Deming 陸德明, the Zheng Xuan annotated version of the Li ji was imperially established as one of the officially taught canons in the Eastern Han dynasty. See his “Jingdian shiwen xulu” 經典釋文序錄, in Jingdian shiwen huijiao 經典釋文匯校, ed. Zhuo, Huang 黃焯, Yanzu, Huang 黃延祖 (Beijing: Zhonghua, 2006)Google Scholar, 1.19a. However, I do not see any other supporting evidence for this. But in any case, the Li ji was canonized at the latest in 642 c.e. when Emperor Taizong of Tang 唐太宗 (599–649 c.e.) ordered Kong Yingda 孔穎達 (574–648 c.e.) and Yan Shigu 顏師古 (581–645 c.e.), etc., to edit and publish an official imperial edition of the Li ji with commentaries as a canonized text, even though the annotated canon was published in 653 c.e. after both Taizong of Tang and Kong Yingda passed away. See “Ruxue zhuan” 儒學傳 of the Xin Tang shu 新唐書, ed. Xiu, Ouyang 歐陽修 and Qi, Song 宋祁, etc. (Shanghai: Shanghai guji and Shanghai shudian, 1986)Google Scholar, 198.602a and “Gaozong ji” 高宗紀 of Jiu Tang shu 舊唐書, ed. Xu, Liu 劉昫 (Shanghai: Shanghai guji and Shanghai shudian, 1986)Google Scholar, 4.17b. Together with the Zuo zhuan 左傳, the Li ji was ranked as “Large Canon,” dajing 大經. See the “Xuanju zhi” 選舉志 of the Xin Tang shu, 44.128c. As for canonicity or canonization, it has various forms in China depending on the nature of the texts. In the field of the Study of Chinese Classics, if a text either was imperially established as an officially taught canon (li yu xueguan 立於學官), or achieved wide acclaim and recognition in the field, I consider it canonized. Other ideas of canon or canonicity can be found in Davies, Philip R., Scribes and Schools: the Canonization of the Hebrew Scriptures (Louisville: Westminster John Knox, 1998)Google Scholar; Harris, Wendell, “Canonicity,” PMLA 106.1 (1991), 110–21Google Scholar; Yu, Pauline, “Poems in Their Place: Collections and Canon in Early Chinese Literature,” Harvard Journal of Asiatic Studies 50.1 (1990), 163–96CrossRefGoogle Scholar and “Canon Formation in Later Imperial China,” in Culture and State in Chinese History: Conventions, Accommodations, and Critiques, ed. Huters, Theodore, Wong, R. Bin, and Yu, Pauline (Stanford: Stanford University, 1997), 83–104Google Scholar, etc. See also Wen, Xing, ed., Rethinking the Study of Chinese Classics, a special issue of Contemporary Chinese Thought 36.4 (summer 2005)Google Scholar. As for the counter-canonical vector of literary collections, see the insightful discussion by Fong, Grace S., “Gender and the Failure of Canonization: Anthologizing Women's Poetry in the Late Ming,” Chinese Literature: Essays, Articles, Reviews 26 (2004), 129–49Google Scholar.

12. Han shu, 88.3615.

13. Chong, Wang 王充, Lun heng 論衡 (Shanghai: Shanghai guji, 1990), 127bGoogle Scholar.

14. Qian, Sima 司馬遷, Shi ji 史記, (Beijing: Zhonghua, 1982), 121.3126Google Scholar.

15. Han shu, 30.1710.

16. Boltz translates “shi” as “a common officer,” who was a low level member of the aristocracy. Early Chinese Texts, 234–35.

17. Liang Yusheng 梁玉繩, Shi ji zhi yi 史記志疑, 35.

18. Li ji zhengyi 禮記正義 (zhushu, Shisanjing 十三經注疏 ed., Beijing: Zhonghua, 1980), 61.453aGoogle Scholar.

19. Shen Wenzhuo 沈文倬 argues that the “Shi li” in the Shi ji text was originally not a title for the Li classic. Wenzhuo, Shen, Zongzhou liyue wenming kaolun 宗周禮樂文明考論 (Hangzhou: Zhejiang daxue, 1999), 214–19Google Scholar.

20. Xidan, Sun 孫希旦, Li ji jijie 禮記集解 (Beijing: Zhonghua, 1989), 1Google Scholar.

21. See Xuan, Zheng's biography, etc. Hou Han shu, ed. Ye, Fan 范曄 (Beijing: Zhonghua, 1965), 35.1212Google Scholar.

22. Duan Yucai 段玉裁, “Li shiqi pian biaoti Han wu yi zi shuo” 《禮》十七篇標題漢無 “儀” 字說, Jing yun lou ji 經韻樓集, 2. See Qian Xuan, San Li tonglun, 6–7.

23. Huang Yizhou, Li shu tonggu 禮書通故, see Qian Xuan, San Li tonglun, 7.

24. Qian Xuan, San Li tonglun, 3–7. Shen Wenzhuo, Zongzhou liyue wenming kaolun, 1.

25. Some scholars consider the term “yi li” in the “Xie duan” 謝短 section of the Lun heng the earliest appearance of the title for the Li classic. This does not seem very convincing. The “Xie duan” reads, “Gaozu (Liu Bang 劉邦, r. 202–195 b.c.e.) asked Shusun Tong to create the Yi pin in sixteen sections, (but) where are they? And furthermore to determine the ceremonies and rites (yi li); these can be seen in the sixteen sections.” (高祖詔叔孫通制作《儀品》十六篇，何在？而復定儀禮，見在十六篇。) See Hui, Huang's 黃暉 argument, in his Lun heng jiaoshi 論衡校釋 (Beijing: Zhonghua, 1990), 561–62Google Scholar.

26. Sun Xidan, Li ji jijie, 651.

27. See Sun Xidan's references in Li ji jijie, 1.

28. Sun Xidan, Li ji jijie, 2–3. Qian Xuan has the same idea, see San Li tonglun, 5–6.

29. Sun Xidan, Li ji jijie, 3.

30. However, it has been widely accepted that Dai De and Dai Sheng did not edit the two groups of the Li ji at all. See William Hung (Hong Ye), “Li ji yinde xu,” xxv–xxviii. See also Pinzhen, Wang 王聘珍, Da Dai Li ji jiegu 大戴禮記解詁, ed. Wenjin, Wang 王文錦 (Beijing: Zhonghua, 1983), 2–6Google Scholar. This issue merits further examination.

31. Quoted by Kong Yingda in his “Preface to the Li ji zhengyi” (Li ji zhengyi xu 禮記正義敘).

32. Sui shu 隋書, ed. Zheng, Wei 魏徵 (Shanghai: Shanghai guji and Shanghai shudian, 1986), 32.3365bGoogle Scholar. In this article, I use the title Xiao Dai Li ji to identify the Li ji wherever there might be a possible confusion.

33. It is not difficult to find other examples. For instance, in Lu Zhi's 盧植 (d. 192 c.e.) biography in the Hou Han shu, the work entitled Li ji that he examined in order to correct the stele version of the Li canon is actually the Li canon, i.e., the Yi li, rather than the Xiao Dai Li ji. See Hou Han shu, 64.2116.

34. Shi ji, 47.1935–36.

35. Confucius was identified as the editor of the Li canon in a number of major texts. The “Rulin zhuan” of the Han shu says that Confucius “pieced together the rites of the Zhou” (zhui Zhou zhi li 綴周之禮) (Han shu, 88.3589). The “Za ji” 雜記 section of the Li ji also clearly attributes the “Shi sang li” 士喪禮 to Confucius. Li ji zhengyi, 43.339b.

36. Mao Shi zhengyi 毛詩正義 (Shisanjing zhushu ed., Beijing: Zhonghua, 1980), 1:3.16bGoogle Scholar.

37. Er ya zhushu 爾雅注疏 (Shisanjing zhushu ed., Beijing: Zhonghua, 1980), 2, 3.15bGoogle Scholar.

38. Er ya zhushu, 2, 3.15b.

39. It was a somewhat popular practice for ancient Chinese to quote the Commentaries of the Yi 易 as the Yi canon itself. For example, in the “Yiwen zhi” of the Han shu, in many cases where the Commentaries of the Yi were quoted, they were simply presented after “The Yi reads” (Yi yue, 易曰). See Han shu 30.1704, 1706, 1710, 1711, and 1720, etc.

40. Quoted by Li Xian 李賢 (654–84 c.e.), see Hou Han shu, 60b.1990. See also Guowei, Wang 王國維, “Wei shijing kao yi” 魏石經考一, in Guantang jilin 觀堂集林 (Beijing: Zhonghua, 1959), 955–59Google Scholar. As for the Luoyang ji, Li Xian mentioned Lu Ji's 陸機 (261–303 c.e.) Luoyang ji in the “Guangwu di ji” 光武帝紀 of the Hou Han shu. See Hou Han shu 1a.40. However, Wang Guowei did not think the Luoyang ji with the “Li ji” stela quote was the Luoyang ji by Lu Ji. According to Wang Guowei's examination, the Luoyang ji could be one of the four or five books with the same title. See Guantang jilin, 956.

41. Baoxuan, Wang 王葆玹, Jin gu wen jingxue xinlun 今古文經學新論 (Beijing: Zhongguo shehui kexue, 1997), 301Google Scholar.

42. See Shen Wenzhuo, Zongzhou liyue wenming kaolun, 32, for a detailed discussion.

43. bowuguan, Gansu sheng and Zhongguo kexuyuan kaogu yanjiusuo, eds., Wuwei Han jian 武威漢簡 (Beijing: Zhonghua, 2005), 96–100, 116–20, and 133–35Google Scholar.

44. Shen also clarifies that the section marks of □ and ○ were not used to distinguish the text of the canon from the ji writings, either, but were used just to divide sections only. For a summary of this issue, see Huangjun, Zhang 張煥君 and Xiaolong, Diao 刁小龍, Wuwei Han jian Yi li zhengli yu yanjiu 武威漢簡《儀禮》整理與研究 (Wuhan: Wuhan University, 2009), 9–10Google Scholar. Chen Mengjia has a different observation and interpretation; see Gansu sheng bowuguan, Wuwei Han jian, 29.

45. Han shu, 30.1710.

46. For the translation of jingxue, see my “Guest Editor's Introduction,” Rethinking the Study of Chinese Classics, 3–10.

47. Shen Wenzhuo further argues that the Li ji or the Xiao Dai Li ji was also referred to as the “Li” with the evidence from the “Wen sang” 問喪 section of the Li ji and from the Shiqu yi zou 石渠議奏, or Shiqu lun 石渠論, a collection of memorials debating the differences between the Five Classics, i.e., the Yi 易, Shi 詩, Shu 書, Li 禮, and Chunqiu 春秋, at the Shiqu ge 石渠閣 Meeting chaired by the Emperor Xuandi 宣帝 of Han in 51 b.c.e. See Shen Wenzhuo, Zongzhou liyue wenming kaolun, 33.

48. Zheng Xuan, San Li mulu 三禮目錄, see Li ji zhengyi, 56.425b and 58.437a. When Zheng Xuan said that the two sections were zhengpian (formal sections) of the Li, he perhaps tried to exclude the possibility of considering the “Ben sang” and “Tou hu” sections somehow appendix sections.

49. See also Wen, Xing, “Li guji yu Zisi zhi xue” 《禮》古記與子思之學, Hunan daxue xuebao (Shehui kexue ban) 湖南大學學報（社會科學版） 23.3 (2009), 5–11Google Scholar.

50. Although in Chinese I prefer “Li gujing” (《禮》古經) to “Li gujing” (《禮古經》), in English I use “Li gujing” and “Li guji” to refer to “Li gujing” and “Li guji” (《禮》古記) because it is less confusing when the whole terms are italicized.

51. Guodian Chu mu zhujian, 174.

52. Based on James Legge's 1885 translation, available on line at Internet Sacred Text Archive, with adaptations. http://www.sacred-texts.com/cfu/liki/index.htm and http://www.sacred-texts.com/cfu/liki2/index.htm. For a printed paper version, see the volume with The Lî Kî (The Book of Rites) in Legge, James, The Chinese Classics (Taipei: SMC Publishing, 2000, rpt.)Google Scholar, # 4 of “Book XXX. SZE Δ or “The Black Robes [1],” http://www.sacred-texts.com/cfu/liki2/liki230.htm.

53. Li ji zhengyi, 55.419c–420a.

54. Based on James Legge's translation, with adaptations. See # 4 of “Book XXX. SZE Δ or “The Black Robes [1],” http://www.sacred-texts.com/cfu/liki2/liki230.htm.

55. Guodian Chu mu zhujian, 174.

56. Li ji zhengyi, 3.21b.

57. Based on James Legge's translation, with adaptations. # 10. 49 of Part IV, Section I, “Book 1. KHÜ LΔ or “Summary of the Rules of Propriety,” http://www.sacred-texts.com/cfu/liki/liki01.htm.

58. Guodian Chu mu zhujian, 187.

59. Li ji zhengyi, 26.228c.

60. Guodian Chu mu zhujian, 187.

61. Li ji zhengyi, 26.228b.

62. Guodian Chu mu zhujian, 181.

63. Li ji zhengyi, 52.401b.

64. Guodian Chu mu zhujian, 163. This line was reconstructed by Qiu Xigui, see Guodian Chu mu zhujian, 164 n13.

65. Based on James Legge's translation in # 50 of “BOOK XXIX. PIÂO KΔ or “The Record on Example [1],” http://www.sacred-texts.com/cfu/liki2/liki229.htm.

66. Li ji zhengyi, 54.416a.

67. James Legge's translation in # 50 of “BOOK XXIX. PIÂO KΔ or “The Record on Example [1],” http://www.sacred-texts.com/cfu/liki2/liki229.htm.

68. Lexian, Liu, “Du Guodian Chu jian zhaji san ze” 讀郭店楚簡札記三則, Zhongguo zhexue 中國哲學 20 (1999), 362Google Scholar.

69. Guodian Chu mu zhujian, 168.

70. Li ji zhengyi, 51.391b.

71. Han shu, 30.1710.

72. Quoted by Lu Deming, “Jingdian shiwen xulu,” 18a.

73. The “Yiwen zhi” of the Han shu came from Liu Xin's 劉歆 (?53 b.c.e.–23 c.e.) Qi lue 七略, which was based on his father Liu Xiang's Bie lu.

74. Guowei, Wang, Guantang jilin 觀堂集林, 7.324 (Beijing: Zhonghua, 1959)Google Scholar.

75. Xueqin, Li, “Guodian jian yu Li ji,” in Li Xueqin wenji 李學勤文集 (Shanghai: Shanghai cishu, 2005), 434Google Scholar.

76. Han shu, 30.1710.

77. “To the Seventy Disciples” was the original phrase while the editors of this edition took it as “to the seventeen (sections).”

78. Qian Han shu 前漢書 (Dian, Wuying 武英殿 ed., 1739; rpt. ed.; Shanghai: Shanghai guji and Shanghai shudian, 1986), 30.164bGoogle Scholar.

79. Qian Han shu (Wuying Dian ed.), 30.164b, notes.

80. William G. Boltz, “The Study of Early Chinese Manuscripts: Methodological Preliminaries,” in The Guodian Laozi, eds., Sarah Allan and Crispin Williams, 39–51. The Chinese version is titled, “Gudai wenxian zhengli de ruogan jiben yuanze” 古代文獻整理的若干基本原則, in Wen, Xing, ed. and trans., Guodian Laozi: Dong xi fang xuezhe de duihua 郭店《老子》：東西方學者的對話 (Beijing: Xueyuan, 2002), 44–58Google Scholar.

81. Sui shu, 32.117b.

82. Han shu, 53.2410.

83. Qishizi zhi tu 七十子之徒 can be read as either the Seventy Disciples or the Seventy Disciples and their followers, depending on how tu 徒 is read.

84. Quoted in Lu Deming's “Jingdian shiwen xulu,” 17b–18a.

85. Xueqin, Li, “Guodian jian yu Li ji” 郭店簡與《禮記》, in his essay collection, Chongxie xueshushi 重寫學術史 (Shijiazhuang: Hebei jiaoyu, 2002), 170–76Google Scholar. Chen Lai 陳來 is convinced that the 14 texts excavated from Guodian are extremely close to the Li ji, thus suggests calling the Guodian texts “Jingmen Li ji” 荊門禮記. Chen Lai, “Guodian jian ke cheng ‘Jingmen Li ji’” 郭店簡可稱“荊門禮記,” Renmin zhengxie bao 人民政協報, August 3, 1998.

86. These passages in question are selected from Xing Wen, “Chu jian Zi yi yu Xian-Qin lixue,” 155–64. However, not all the arguments and discussions are included here. As I announced earlier in this article, the portions related to the Shanghai Museum bamboo-slip version are no longer valid unless the evidence for its authenticity is provided, and I also delete the relevant portions in this article accordingly.

87. For a discussion on the sequences of the Li ji and Guodian versions of the “Zi yi” or Zi yi and their connections with the composition of early texts, see Boltz, William G., “The Composite Nature of Early Chinese Texts,” in Text and Ritual in Early China, ed. Kern, Martin (Seattle and London: University of Washington Press, 2005), 56–58Google Scholar.

88. Li ji zhengyi, 55.1650.

89. Guodian Chu mu zhujian, 129.

90. Li ji zhengyi, 55.1650.

91. As shown below, shen 身 is written as ti 體 in the Guodian version. Traditionally, shen indicates the upper body below the neck or the body between the neck and the feet, while ti indicates the whole body, i.e., shen is only part of ti. In Group 1.2, in both the Li ji and Guodian cases, both shen and ti are contrasted with xin 心 (heart), thus they can be considered and translated the same, which is body.

92. Guodian Chu mu zhujian, 129.

93. Li ji zhengyi, 55.1650.

94. Guodian Chu mu zhujian, 129.

95. I follow Qiu Xigui's 裘錫圭 interpretation on fa 法 here. According to Qiu, this character is read fei 廢 and the fa and the fei were interchangeable in early China (er zi gu tong 二字古通). See Guodian Chu mu zhujian, 132n27.

96. Ma Chengyuan, Shanghai bowuguan cang Zhanguo Chu zhushu (yi), 179–80. In his transcription and translation, Shaughnessy adds the character min 民 “people” before the character wang 亡 “perish” to the Shanghai Museum version, but he provides no justification for this emendation. See Shaughnessy, Rewriting Early Chinese Texts, 100–101. However, this example could provide evidence for the authenticity examination, thus I keep it here.

97. Xing Wen, “Chu jian Zi yi yu Xian-Qin lixue,” 156.

98. Shaughnessy correctly pointed out the ambiguity that lies in the character hao 好, which can be read as either an adjective or a verb according to the modern grammatical category. Shaughnessy, Rewriting Early Chinese Texts, 89. However, I am not convinced by his argument that “a total of something like twelve extra characters” was inserted into the Li ji text in order to resolve the ambiguity. He does not specify which twelve characters should be inserted, but it seems to me they most likely would have been xin zhuang ze ti shu, xin su ze rong jing 心莊則體舒，心肅則容敬 (“If the heart is sober, the body would be relaxed; if the heart is solemn, the appearance would be respectful.”). As I said in “Chu jian Zi yi yu Xian-Qin lixue,” 156, these two phrases clearly explain the preceding metaphor of “heart” and “body,” i.e., “the people take the ruler as their heart, while the ruler takes the people as his body.” In other words, they should be grouped into group 1.1 rather than group 1.2 as he suggests. See Shaughnessy, Rewriting Early Chinese Texts, 89–90.

99. Ma Chengyuan, Shanghai bowuguan cang Zhanguo Chu zhushu (yi), 55.

100. Li ji zhengyi, 55.421a

101. Guodian Chu mu zhujian, 130.

102. Note that 囗 indicates a missing graph. Ma Chengyuan, Shanghai bowuguan cang Zhanguo Chu zhushu (yi), 186–87.

103. Li ji zhengyi, 55.1649.

104. Guodian Chu mu zhujian, 130.

105. Li ji zhengyi, 55.1649.

106. Guodian Chu mu zhujian, 130.

107. Chen Peifeng 陳佩芬, “Zi yi” 緇衣, in Ma Chengyuan, Shanghai bowuguan cang Zhan'guo Chu zhushu (yi), 186.