Unearthing the Changes

12. The hexagram name always comes at the beginning of the hexagram statement and is typically separate from the rest of the statement. However, in four cases (Lü 履,

, “Stepping” [hexagram 10 in the traditional sequence]; Pi 否,

, “Negation” [hexagram 12]; Tong Ren 同人,

, “Fellow Men” [hexagram 13]; and Gen 艮,

, “Stilling” [hexagram 52]), the hexagram name is grammatically linked to the words following it. For a suggestion that these four cases are the result of misplaced line statements, see Shaughnessy, “Composition of the Zhouyi,” 120–23.

13. Here and throughout this book, the text of the received Changes is that of Zhou Yi zheng yi 周易正義, in Ruan, Shisan jing zhushu, juan 1–6 (pp. 13–75), with the exception that I omit punctuation in the Chinese text (the Shisan jing zhushu edition provides the traditional small circle [judou 句讀] after each clause). I do this for convenience in comparing the received text with the text of the various manuscripts introduced in this book, none of which includes punctuation. The text of Xian hexagram is found at 4.34–35:46–47.

14. For these reconstructions of the archaic pronunciation, see Schuessler, Minimal Old Chinese and Later Han Chinese, 361.

15. Gao, Zhou Yi gu jing jin zhu, 108. See also Kunst, “Original ‘Yijing,’” 300–301; Rutt, Zhouyi, 254.

16. Kunst, “Original ‘Yijing,’” 38–43, provides a particularly clear discussion of this feature, including tables illustrating the distribution of words with “low” and “high” meanings.

17. Lynn, Classic of Changes, 16.

18. The work of Gao Heng has been particularly influential in this context, especially his Zhou Yi gu jing jin zhu. See also Gao, Zhou Yi gu jing tong shuo. Gao’s influence is readily seen in the translations of Kunst, “Original ‘Yijing,’” and Rutt, Zhouyi.

19. Zhou Yi zheng yi, 9.82–84 (pp. 94–96).

20. For the Mawangdui sequence, see pp. 7–8.

21. In both chapters 1 and 2, I introduce different sorts of archaeological evidence bearing on this question. Most of this evidence seems to support the traditional sequence of hexagrams, but none of it can be regarded as conclusive.

22. The Wangjiatai Gui cang manuscripts have never been published in their entirety, and there is little reason to expect that they will be in the near future; I have translated all passages that have been published to date. For the Fuyang manuscript, I have translated all passages bearing text of the Zhou Yi. I have not translated the 531 fragments that bear only divinatory terms and that cannot be related with individual line statements of the Zhou Yi.

1. Wilhelm, I ging: Erstes und zweites Buch, I ging: Drittes Buch; subsequently reprinted (many times) as I ging: Das Buch der Wandlungen. Wilhelm’s German translation was in turn translated into English as The I Ching; or, Book of Changes. For a recent appraisal of Wilhelm’s translation, see Hon, “Constancy in Change.”

2. For an intellectual biography of Lao Naixuan, see Molino, “Study in Late Ch’ing Conservatism.”

3. Wilhelm, I ging: Das Buch der Wandlungen, 25; Wilhelm, I Ching; or, Book of Changes, 4.

4. For an account of Menzies’s life and his time in China, see Dong, Cross Culture and Faith.

5. Upon Wang’s death in 1900, his collection of oracle bones was purchased by his friend Liu E 劉顎 (1857–1909) and published as Tieyun cang gui 鐵雲藏龜 (N.p., 1903). For an extensive account of the discovery and first studies of the oracle bones, see Lefeuvre, “Les inscriptions des Shang.” Lefeuvre holds (18–19) that the first scholar to recognize the writing on the “dragon bones” was actually Wang Xiang 王襄 (1876–1965), who had bought fragments brought by antiquarians in 1898 to his home in Tianjin 天津.

6. Dong, Cross Culture and Faith, 117–18, 133, notes that early on Menzies had collected 8,080 fragments, 3,668 of them inscribed, that reverted to the collection of Cheeloo University (Qi Lu Daxue 齊魯大學), where he taught, and that later another 5,000 pieces entered the collection of the Royal Ontario Museum in Toronto.

7. Ming, Yinxu buci. Prior to the publication of Menzies’s collection, the still more important collection of Luo Zhenyu 羅振玉 (1866–1940) was published, in several installments: Yinxu shuqi (1913); Yinxu shuqi jinghua (1914); and Yinxu shuqi houbian (1916).

8. For both these reconstructions of archaic pronunciations, see Schuessler, Minimal Old Chinese and Later Han Chinese, 138.

9. Duan, Shuo wen jie zi Duan zhu, 3B.39a. As the Shuo wen goes on to note, the character zhen 貞 was originally written 鼑, in which the 鼎 component served to indicate the pronunciation.

10. The demonstration of this prayerful aspect of early Chinese divination has been one of Western Sinology’s greatest contributions to the study of early Chinese civilization. It was first announced, more or less simultaneously, by David N. Keightley and Paul L-M Serruys (1912–1999); see Keightley, “Shih cheng”; Serruys, “Studies in the Language of the Shang Oracle Inscriptions,” 21–23.

11. Zheng Xuan, Zhou li Zheng zhu, 24.7b. Throughout this book, I attempt to reflect both the divination and moral nuances by translating zhen 貞 as “to determine” (or some variant thereof).

12. For an overview of this early archaeological work, written by the then director of the Institute of History and Philology, see Li Chi, Anyang.

13. Guo and Hu, Jiaguwen heji, vol. 3, no. 6834A.

14. Zheng also conducted a divination performed on renzi and recorded on the same plastron, no. 6834A, whereas on plastron no. 6830 a divination concerning the same topic was conducted by the diviner Ke

. For a study of these and related inscriptions, see Shaughnessy, “Micro-Periodization.”

15. For still the finest Western-language introduction to all aspects of Shang oracle-bone inscriptions, see Keightley, Sources of Shang History.

16. Gu Jiegang, “Zhou Yi gua yao ci zhong de gushi.”

17. Gao, Zhou Yi gu jing jin zhu.

18. Mawangdui Han mu bo shu (yi).

19. Mawangdui Han Mu Boshu Zhengli Xiaozu, “Mawangdui boshu Liushisi gua shiwen”; Mawangdui Han mu wenwu, 416–35; Chen Songchang, “Boshu Xi ci shiwen”; Chen Songchang and Liao Mingchun, “Boshu Ersanzi wen Yi zhi yi”; Chen Songchang, “Mawangdui boshu Mu He Zhao Li shiwen.” Photographs of the entire manuscript, together with a posthumous facsimile edition of the draft transcription done by the editor of the manuscript, Zhang Zhenglang 張政烺 (1912–2005), have now been published; see Zhang, Mawangdui bo shu Zhou Yi zhuan jiao du. For translations of the Mawangdui manuscripts, see Hertzer, Das Mawangdui-Yijing; Shaughnessy, I Ching.

20. Throughout this book, quotations of the received text of the Changes are taken from the Shisan jing zhushu 十三經注疏 edition; see Ruan, Zhou Yi zheng yi. Although the Shisan jing zhushu text is supplied with rudimentary punctuation (the traditional Chinese circle, “。”), because much of this book involves comparisons between the received text and ancient manuscripts, which are generally not punctuated, I present the Chinese text without any punctuation. My own understanding of the punctuation will be clear from the punctuation given in the English translation, which in all cases is my own.

21. For a discussion of the complementarity of Qian and Kun hexagrams, explaining them as an almanac of a full year, see Shaughnessy, “Composition of ‘Qian’ and ‘Kun’ Hexagrams.”

22. It is actually not nearly so simple as deciding between just two different sequences. Even among the canonical commentaries of the Changes regarded as a “classic,” the Za gua 雜卦, or “Mixed Hexagrams,” commentary presents a different sequence from that found in the received text printed above it, and the opening of the Shuo gua 說卦, or “Discussing the Hexagrams,” commentary at least hints at a different sequence. In the Mawangdui manuscript as well, there is evidence of different sequences; for example, the discussion of the hexagrams in the Er san zi wen 二三子問, or “The Two or Three Disciples Asked,” commentary follows, for the most part, the sequence of the received text, and none of the other commentaries discusses the hexagrams in the sequence given by the manuscript itself. In the Han dynasty, several other sequences of hexagrams are also attested: the most influential of these were those of Meng Xi 孟喜 (ca. 90–40 B.C.), which correlates the hexagrams with the twenty-four fortnightly periods of the year and the five noble titles, and Jing Fang 京房 (77–37 B.C.), which organizes the sixty-four hexagrams into “eight palaces” (ba gong 八宫), more or less similar to the arrangement of the Mawangdui manuscript (the sequence is different, but the mechanical arrangement is similar). For a thorough presentation of these different sequences and an analysis of the sequence of the Mawangdui manuscript, with citations of studies published in the 1980s, see Xing Wen, Bo shu Zhou Yi yanjiu, 65–93. See also Smith, Fathoming the Cosmos and Ordering the World, 62–77.

23. As I discuss in chapter 2, there is some evidence, albeit quite tenuous, that the Shanghai Museum manuscript of the Zhou Yi, copied about 300 B.C., had the same sequence of hexagrams as the received text.

24. For two of the most recent discussions of this topic, both arguing in favor of the antiquity of the received sequence, see Liao, Boshu Zhou Yi lunji, 13–14; Li Shangxin, Gua xu yu jie gua lilu, 103–23.

25. For a brief summary of this debate, see Shaughnessy, “First Reading of the Mawangdui Yijing Manuscript.”

26. This text has been given the title Bian nian ji 編年紀, or Annals Stringing Years, by the editors of the Shuihudi manuscripts; see Shuihudi Qin mu zhu jian, 1–7 (plates), 3–10 (transcription). See also Mittag, “Qin Bamboo Annals of Shuihudi.”

27. There are five different sorts of legal texts among the Shuihudi manuscripts; for a study and translation of them, see Hulsewé, Remnants of Ch’in Law.

28. The second of these two texts, which are very similar in content, bears the title Ri shu. The first text, which is more complete, includes 166 bamboo strips with text on both sides; see Shuihudi Qin mu zhu jian, 87–116 (plates), 177–228 (transcription). The second text includes 257 strips, with text on only one side (117–40 [plates], 229–55 [transcription]).

29. See Shuihudi Qin mu zhu jian, Ri shu jia zhong, nos. 2–7, 正貳, 89 (plates), 181 (transcription).

30. The editors of the Shuihudi Qin mu zhu jian indicate (181), almost certainly correctly, that 帥 here is a mistake for shi 師, “troops, army.”

31. Throughout history, Chinese readers and writers have referred to the Changes as either Zhou Yi or Yi jing almost indiscriminately. In current practice in China, there is some tendency to reserve the title Yi jing for just the “classic” portion of the text, i.e., the hexagram and line statements, and to use the title Zhou Yi to refer to the entire text, inclusive of the canonical commentaries, the so-called Ten Wings. In my own writings in English, I have consistently done the reverse, using the title Zhou Yi to refer to the hexagram and line statement portion of the book, especially as understood in its Zhou-dynasty context, and reserving the title Yi jing for the received text in general, especially as understood as one of the Chinese classics. In this book, which deals with manuscripts of the text at the time when it was just in the process of becoming a classic, I try to avoid this distinction by referring to the text simply as the Changes.

32. For the initial report of this discovery, see Shaanxi Zhouyuan Kaogudui, “Shaanxi Qishan Fengchu cun faxian Zhou chu jiaguwen”; for the most recent and most thorough presentation of this and other, similar discoveries in the Zhouyuan area, see Cao, Zhouyuan jiaguwen.

33. For some discussion of this feature, see Shaughnessy, “Western Zhou Oracle-Bone Inscriptions.”

34. These formulaic prayers are introduced with the character si 甶 (i.e., si 思), “to wish for; would that.” Since I first proposed this interpretation of this character, the proposal has generated a great deal of discussion among Chinese scholars, with some agreeing and others disagreeing. For my latest statement on the matter, including references to most of the relevant literature, see Xia, “Zai lun Zhouyuan.”

35. Zhang, “Shishi Zhou chu qingtongqi mingwen zhong de Yi gua,” translated into English by Huber, Yates, et al. as “An Interpretation of the Divinatory Inscriptions on Early Zhou Bronzes.”

36. In quotations of Chinese paleographic materials, an ellipsis (…) indicates an indeterminate number of characters missing, and a box (囗) indicates either that a single character is missing or—as is the case here—cannot be read.

37. See, for instance, Xu Zhongshu, “Shu zhan fa yu Zhou Yi de ba gua,” 383. Xu also reads the lines immediately above ji 既, “having already,” which no one else can decipher, as the character 文, understood as the protograph for lin 吝, “distress,” a divinatory term commonly used in the Changes.

38. Cao, “Zhouyuan xin chu Xi Zhou jiaguwen yanjiu.”

39. Ibid., 45, though noting that the graph in question here,

, has never before been seen, suggests that it should be a verb meaning something like “to divine.” Li Xueqin, “Xin faxian Xi Zhou shi shu de yanjiu,” 6, suggests more plausibly that it should mean “to sacrifice, to make offering.”

40. Cao, “Zhouyuan xin chu Xi Zhou jiaguwen yanjiu,” transcribes the graph,

, as shen 神, “spirit,” whereas Li Xueqin, “Xin faxian Xi Zhou shi shu de yanjiu,” suggests dao 禱, “to pray.” In fact, it is very difficult to determine, just from the graphic shape, which of these two transcriptions is more likely correct; however, the context, especially that of the second divination, does seem to require a verb, thus making dao, “to pray,” preferable.

41. For the most thorough surveys of these hexagram symbols, all taking very different approaches, see Cai, “Shang Zhou shi shu Yi gua shili”; Li Zongkun, “Shuzi gua yu yin yang yao”; Xing Wen, “Hexagram Pictures and Early Yi Schools.”

42. Cao, “Tao pai shang de shuzi gua yanjiu.”

43. For my own view on this, see Xia, “Zhou Yi shifa yuan wu zhi gua kao.”

44. Li Xueqin, “Xin faxian Xi Zhou shi shu de yanjiu.”

46. Both Cao Wei, the excavator, and Li Xueqin argue that the only number that would fit in the space available would be the number “1” (━; i.e., yi 一); see Cao, “Tao pai shang de shuzi gua yanjiu,” 68; Li Xueqin, “Xin faxian Xi Zhou shi shu de yanjiu,” 4.

47. For the initial report of this discovery, see Anhui Sheng Wenwu Gongzuodui, Fuyang Diqu Bowuguan, and Fuyang Xian Wenhuaguan, “Fuyang Shuanggudui Xi Han Ruyin Hou mu fajue jianbao”; for the texts in the tomb, see Wenwu Ju Wenxian Shi and Fuyang Diqu Bowuguan Zhenglizu, “Fuyang Han jian jianjie.” For information concerning these various texts, see chapter 6. For a valuable overview of the materials in the tomb related to divination, see Hu, “Fuyang Shuanggudui Han jian shushu shu jianlun.”

48. Han, “Fuyang Han jian Zhou Yi shiwen.” This was subsequently reissued as a separate monograph: Han, Fuyang Han jian “Zhou Yi” yanjiu.

49. Of the 3,119 characters on the strips, 1,110 belong to the Zhou Yi itself, with passages from 170 or more hexagram or line statements in 52 different hexagrams (there are 4,161 characters in the 386 line statements and 64 hexagram statements of the received text). The remaining 2,009 characters belong to divination formulas. Thus, about one-quarter of the Zhou Yi text is represented in the Fuyang manuscript.

50. This call to “rewrite” early Chinese history is most commonly associated with Li Xueqin 李學勤; see, for example, his Chongxie xueshu shi, or Rewriting the History of Scholarship.

51. For one influential discussion of this question, see Petersen, “Which Books Did the First Emperor of Ch’in Burn?”

52. For a study of this question, see Shaughnessy, Rewriting Early Chinese Texts.

53. For these strips, see Baoshan Chu jian. For an overview of this tomb and the deceased buried in it, see Constance Cook, Death in Ancient China.

54. For a study of these court cases in English, see Weld, “Chu Law in Action.”

55. Baoshan Chu jian, nos. 228–29. For more detailed discussion of these divination records, including (in the first case) another translation of this same record, see Harper, “Warring States Natural Philosophy,” 852–56, esp. 855; Kalinowski, “Diviners and Astrologers under the Eastern Zhou,” 381–84.

56. See Shaughnessy, “Composition of the Zhouyi,” 81–97.

57. The first site report was Henan Sheng Wenwu Kaogu Yanjiusuo, “Henan Xincai Pingye jun Cheng mu de fajue.” The formal site report was by the same team: Xincai Geling Chu mu.

58. For this date, see Song Huaqiang, Xincai Geling Chu jian chutan, 135.

59. Ibid., 61–62, 165–85.

60. In the context of the lines and line statements of the Changes, the word is usually written yao 爻, but the two different characters (i.e., 爻 and 繇) certainly refer to the same word.

61. See Henan Sheng Wenwu Kaogu Yanjiusuo, Xincai Geling Chu mu, 187–231 (nos. 甲三 217, 乙四 100 零 532+678, 零 115+22, and 甲三 31), respectively.

62. In the Zhou Yi, this is written 勿恤. Compare the following examples:

63. This is not necessarily to say that the line statements of the Changes were not produced until the fourth century B.C. but only that they must have been produced in similar contexts. The dates of the original composition and final redaction of the Changes are still very much open questions. I see no reason to modify, at least very much, my earlier conclusion that the text of the hexagram and line statements achieved more or less the form it now has toward the end of the Western Zhou period, about 800 B.C.; see Shaughnessy, “Composition of the Zhouyi,” 27–49.

64. Song Huaqiang, Xincai Geling Chu jian chutan, 180–81.

65. The Shuo gua 說卦 commentary says of Dui:

The hexagrams and line statements associated with Dui hexagram in the Zhou Yi (no. 58 in the received sequence) are by far the briefest of all the hexagrams in the text (only thirty characters in total, whereas the next briefest has forty-five, and the median is sixty-five) and give very little indication as to the original meaning of the word dui 兑. Of the canonical commentaries, the Xiang 象, or “Image,” commentary does seem to associate it with speech, saying, Junzi yi peng you jiang xi 君子以朋友講習, “The nobleman with friends speaks about and practices it.” On the other hand, it should also be pointed out that neither of the two hexagram pictures indicated for this divination,

and

, is either the hexagram Dui or even contains the trigram Dui.

66. For an account of the excavation of the Guodian tomb, see Hubei Sheng Jingmen Shi Bowuguan, “Jingmen Guodian yihao Chu mu.” For the bamboo strips, see Guodian Chu mu zhu jian.

67. The proceedings of this conference were published as The Guodian Laozi.

68. For some of my own views on the importance of the Guodian Laozi manuscript, see Shaughnessy, “Guodian Manuscripts.”

69. The interest has been more pronounced in China than in the West. In China there has been a boom in studies of the Si 思 (i.e., Zi Si 子思 [483–402 B.C.]) Meng 孟 (i.e., Mengzi) lineage of Confucian teaching. For some evidence of interest in these topics in the West, see Goldin, “Xunzi in the Light of the Guodian Manuscripts”; Scott Cook, “Debate over Coercive Rulership.”

70. One of the Guodian texts, the Zi yi 緇衣, or Black Jacket, has a received counterpart in the Li ji 禮記, or Record of Ritual. Among the differences between the manuscript version of the text and the received version, a quotation of the Changes in the received text is missing in the manuscript version; it is unclear what significance should be given to this difference.

71. For the Liu de text, see Guodian Chu mu zhu jian, 70–71 (strips 24–25), 188 (transcription); for the Yu cong, see 79–80 (strips 36–37), and 194 (transcription).

72. Shanghai Bowuguan cang Zhanguo Chu zhu shu.

73. Ibid., 3:13–70 (plates), 133–215 (transcription).

74. I have published this view elsewhere: Shaughnessy, “First Reading of the Shanghai Museum Bamboo-Strip Manuscript of the Zhou Yi,” 16–21; Xia, “Shilun Shangbo Zhou Yi de gua xu.” See also, in the present volume, page 48.

75. Shanghai Bowuguan cang Zhanguo Chu zhu shu, 3:251–60.

76. He, “Lun Shangbo Chu Zhu shu.”

77. Jingzhou Diqu Bowuguan, “Jiangling Wangjiatai 15 hao Qin mu.” This discovery is the topic of chapter 3, where full documentation of the scholarship on it can be found.

78. Fang Xuanling et al., Jin shu, 51, “Shu Xi zhuan” 束皙傳, 1432–33; for a detailed account of this discovery and the editing of the texts from it, in English, see Shaughnessy, Rewriting Early Chinese Texts, 131–53.

79. Fang Xuanling et al., Jin shu 51, 1432.

80. For the reconstitution of this Gui cang hexagram statement, see pages 153–54.

81. For a comprehensive comparison of hexagram names from all the different sources of the Gui cang with those of the Changes, see table 4.1.

82. Yu Qiang is said, in the Shan hai jing 山海經, or Classic of Mountains and Seas, to have been the god (or goddess) of the North Sea (Bei Hai); see Shan hai jing jiao zhu, 425.

83. It is interesting to note that whereas a comprehensive Chinese bibliography published in 1989 listed only seven scholarly articles that had discussed the Gui cang in the preceding ninety years (see Lin, Jing xue yanjiu lunzhu mulu, 1:86), in the years since the first notice of the Wangjiatai manuscripts in 1995 there have been scores of articles devoted to the text.

84. Guo Pu, Mu tianzi zhuan zhu, 5.4a–4b.

85. One chapter of the Shi ji 史記, or Records of the Historian, of Sima Qian 司馬遷 (ca. 145–90 B.C.) is devoted almost exclusively to turtle-shell divination; see Sima, Shi ji, 128, “Gui ce lie zhuan” 龜策列傳, 3223–51.

86. The Mawangdui text “Yao” 要, or “The Essentials,” quotes Confucius as claiming to his disciple Zi Gong 子贑 that when he performed divinations, 70 percent of his prognostications were correct; see Chen Songchang and Liao Mingchun, “Boshu Ersanzi wen Yi zhi yi,” 435; Shaughnessy, I Ching, 241.

87. Gao Heng, Zhou Yi gu jing jin zhu, 1.

88. Shaughnessy, “Composition of the Zhou Yi,” 95–97.

89. For a study addressing this issue, see Li Ling, “Formulaic Structure of Chu Divinatory Bamboo Slips,” 77–78.

90. For my most recent study on this point, see Xia, “Zhou Yi ‘Yuan heng li zhen’ xin jie.”

91. A special feature of the Chinese language is that in many cases the give-and-take of communication between two persons or two parties was originally expressed with a single word. Thus, both “to give” and “to take” were written as shou 受 (“to give” was later differentiated in speech with a change of tone and in writing as shou 授); “to buy” and “to sell” were both written as mai 買 (“to sell” was later differentiated as mai 賣); jie 借 is still used for both “to lend” and “to borrow,” and ming 明 can mean either “to explain” or “to understand.” Even the word family deriving from you 有, “to have,” seems to display this feature: the word you 侑 means “to give offering (to the spirits),” whereas the word you 祐 means “to receive blessings (from the spirits)”; in Shang oracle-bone inscriptions, both senses, as well as that of you, “to have,” could be written simply as you 又. The word family that includes both heng 亨 and xiang 饗 reveals this same sort of bidirectional sense of communication.

92. Sun Yirang, Mozi xian gu, 14a–17a. Sun’s emendations to the 1445 Ming-dynasty Dao zang 道藏, or Daoist canon text of the Mozi (this passage is found in that text at 11.9a–b) are based on medieval quotations of the passage, all of which he copiously cites.

93. In his exhaustive commentary, Sun Yirang goes to great lengths to try to rationalize these contradictions, quoting numerous editions to suggest, for instance, that the “three legs” (san zu 三足) of the cauldron should actually read “four legs” (si zu 四足). I suspect that the contradictions were inherent to the prayer made in the divination, pointing toward the supernatural qualities the cauldron was to have. In this respect, it might not be entirely irrelevant to compare the Witches’ Chant from Shakespeare’s Macbeth, which also combines prayer and prediction. Just the first chorus reads:

94. Bi, Mozi zhu, 11.8b. Sun Yirang, Mozi xian gu, 11.16a, specifically differentiates this “command to the turtle” (ming gui zhi ci 命龜之辭) from the “prognostication” (zhan ci 占辭) below.

95. Early quotations of this sentence in both the Yi wen lei ju 藝文類聚 and Yu hai 玉海 give the character as 繇, a variation that Sun Yirang notes is common in early texts; see Sun Yirang, Mozi xian gu, 11.16b.

96. The rhyme of the oracle is again masked by textual variants. Bi Yuan suggests that bei 北, “north,” and guo 國, “kingdom,” rhyme, which is true (*pək vs. *kwək) but also irrelevant since the characters are not in correct rhyme position. Instead, the true rhyme is between dong 東, “east” (*tông), and bang 邦, “country” (*prong), which must have been the original reading of the final character before it was changed to guo 國, “kingdom,” to avoid a Han-dynasty taboo on the name of Liu Bang 劉邦 (r. 202–195 B.C.), the founder and first emperor of the dynasty.

97. Peng peng 逢逢 is evidently to be read as peng peng 蓬蓬, “luxuriant,” as in the Shi jing poem “Cai shu” 采菽 (Mao 222): Qi ye peng peng 其葉蓬蓬, “Its leaves are luxuriant.” In another context (though still with respect to this extended passage, Mozi xian gu, 11.17a), Sun Yirang quotes a passage from the “Jiao si zhi” 郊祀志 chapter of the Han shu 漢書 (Zhonghua shuju ed., 25A.1225) describing how during the reign of Han Wu Di (r. 140–87 B.C.) a bronze cauldron, larger and different from ordinary cauldrons, was discovered in Fenyin 汾陰. When the cauldron was brought to the capital and presented to the emperor, “yellow clouds” (huang yun 黃雲, later described as “yellowish-white clouds” [huang bai yun 黃白雲]) appeared, the emperor’s officials regarding them as an auspicious portent. Of course, this may be sheer coincidence, but it may also indicate some symbolic connection between the cauldrons of Yu 禹 and the appearance of clouds. In this account, the officials recount how Yu cast the nine cauldrons and used them to make offering (xiang xiang 鬺享, 鬺 being yet another variant of xiang 饗 or 享) to Di on High and the ghosts and spirits (shang di gui shen 上帝鬼神), thereby “receiving heaven’s blessings” (xiang cheng tian gu 饗承天祜), well illustrating again the bidirectionality of the word xiang 饗.

1. Baoshan Chu jian.

2. There had previously been a few piecemeal discoveries of Warring States bamboo-strip texts, the most important being the 148 strips found in a tomb at Changtaiguan 長臺關, Xinyang 信陽, Henan, for which, see Henan Sheng Wenwu Gongzuodui Di Yi Dui, “Wo guo kaogu shi shang de kongqian faxian”; see also Zhongguo Shehui Kexueyuan Kaogu Yanjiusuo, Xinyang Chu mu, for a detailed report on the discovery. See Shang, Zhanguo Chu zhu jian huibian, for photographs, tracings, and transcriptions of the bamboo strips. Unfortunately, the most important of these strips were very badly preserved and, given the state of the field at the time, also very badly understood.

3. For the formal publication of the manuscripts, see Guodian Chu mu zhu jian. The discovery immediately prompted at least two different international conferences, one held in the United States and the other in China; see Guodian Laozi; Guodian Chu mu guoji xueshu yantaohui lunwenji.

4. See, for example, Henricks, Lao Tzu’s Tao Te Ching.

5. For some discussion of these texts, including especially the Zi yi, see Shaughnessy, Rewriting Early Chinese Texts, 9–61, 63–93.

6. For an account of the purchase of these strips and related issues, see “Ma Chengyuan xiansheng tan Shang bo jian,” 1–8.

7. Shanghai Bowuguan cang Zhanguo Chu zhu shu, vol. 1.

8. For example, Min zhi fumu 民之父母, or The Parents of the People, the title assigned to the text by the editors, which corresponds to the “Kongzi xian ju” 孔子閒居, or “Confucius at Rest,” chapter of the Li ji; Shanghai Bowuguan cang Zhanguo Chu zhu shu, 2:15–30 (plates), 149–80 (transcription). In addition, much of the content of the text Nei li 內豊, or “Internal Ritual,” corresponds to the chapter “Zengzi li xiao” 曾子立孝, or “Zengzi Establishes Filiality,” of the Da Dai Li ji 大戴禮記; see Shanghai Bowuguan cang Zhanguo Chu zhu shu, 4:69–81 (plates), 217–230 (transcription).

9. Shanghai Bowuguan cang Zhanguo Chu zhu shu, 2:91–146 (plates), 249–93 (transcription).

10. Ibid., 3:103–118 (plates), 287–99 (transcription).

11. Ibid., 5:97–123 (plates), 265–83 (transcription).

12. Ibid., 3:11–70 (plates), 133–260 (transcription). For a preliminary study of this manuscript in English, see Shaughnessy, “First Reading of the Shanghai Museum Bamboo-Strip Manuscript of the Zhou Yi.” Much of the information presented in this earlier study is repeated here, though I also introduce a considerable amount of new material.

13. For this reason, I make no attempt here to provide even a complete listing of the texts published to date, much less a thorough description of their contents.

14. Pu Maozuo subsequently also published a two-volume study of the Shanghai Museum Zhou Yi manuscript and its context: Chu zhushu “Zhou Yi” yanjiu. The book is divided into two parts, the first presenting the Shanghai Museum manuscript and the second comprising a conspectus of materials for the development of the Yi jing tradition through the Eastern Han dynasty. The first part reproduces several portions of the Shanghai Bowuguan cang Zhanguo Chu zhu shu volume 3 presentation of the manuscript: the color photographs of the individual strips; the strip-by-strip discussions of the text; and the comparison of the Shanghai Museum manuscript, the Mawangdui manuscript, and the received text. It also includes a discussion of the red and black symbols found on the manuscript, somewhat differently organized from the organization of Shanghai Bowuguan cang Zhanguo Chu zhu shu volume 3 but with no apparent difference in interpretation. In addition to these sections, this first part of the book adds a lengthy general introduction to the manuscript (“Chu zhushu Zhou Yi gaikuang yu yanjiu” 楚竹書《周易》概況與研究) and three new appendices: a bare transcription of the manuscript (“Chu zhushu Zhou Yi yuanwen” 楚竹書《周易》原文), a glossary of phrases in the manuscript (“Chu zhushu Zhou Yi cimu jieshi” 楚竹書《周易》詞目解釋), and a concordance of characters in the manuscript (“Chu zhushu Zhou Yi zhu zi suoyin” 楚竹書《周易》逐字索引). The second part of the book is divided into two chapters. The first surveys archaeological materials related to the Yi jing. These include hexagram symbols found in oracle-bone and bronze inscriptions and on other implements and hexagram symbols and related texts found in bamboo-strip texts, including those from Baoshan 包山, Xincai 新蔡, Tianxingguan 天星觀, Guodian 郭店, and Fuyang 阜陽; and all Yi jing–related texts from Mawangdui, the Xiping 熹平 Stone Classics, as well as the Guicang 歸藏 texts from Wangjiatai 王家台. The second chapter presents quotations and uses of the Yi in transmitted literature, together with a complete text of the received version of the Yi jing. Despite the redundancies with Shanghai Bowuguan cang Zhanguo Chu zhu shu volume 3, the book serves as a convenient source for the early development of the Yi jing tradition in China.

15. For this strip, see Rao, “Zai kaituo zhong de xunguxue,” 1–5; Zeng, “Zhou Yi Kui gua ji liusan yaoci xinquan.”

16. The identification of these six different symbols follows that of Pu Maozuo; Shanghai Bowuguan cang Chu zhu shu (san), 134, 251–60. Li Shangxin, “Chu zhushu Zhou Yi,” 24, suggests that there is a seventh symbol, which he describes as a large solid red square with a smaller solid black square set inside it. The only example of this comes at the end of Yi 頤 hexagram, strip no. 24. Pu treats it as a three-sided hollow red square with an inset smaller solid black square. However, as discussed on p. 40, it is clear that this strip was copied by a different copyist from that of the great majority of the strips, and the analysis of the symbol on it ought to be strictly segregated from that of the other strips. Indeed, the last two symbols identified by Pu, the solid red square with an inset hollow black square and the three-sided hollow black square, also occur exceptionally only on strips copied by this second copyist. For further discussion of these symbols, see pages 40–45.

17. Fang Zhensan, “Zhu shu Zhou Yi caise fuhao chutan,” 22.

18. He, “Lun Shangbo Chu Zhu shu.” Other differences between these two strips include the ways of writing the “heart” signific in hu

(no. 26) vs.

(no. 27), and the “meat” signific in:

(no. 26) vs. she

(no. 27). It is unclear whether these differences indicate a different scribe or rather mere calligraphic flourishes by a single scribe. He Zeheng argues for as many as seven different scribes responsible for the thirteen strips of Group B. However, except in the case of these two strips, I see little or no difference in most of the character forms he indicates.

19. In addition to Li, “Chu zhushu Zhou Yi,” and He, “Lun Shangbo Chu Zhu shu,” see also Jiang, “Shangbo cang Chu zhushu Zhou Yi”; Chen Renren, “Shang bo Yi teshu fuhao.” This last study is an exhaustive critique of the previous attempts by Pu Maozuo, Jiang Guanghui, and Li Shangxin to detect some principle behind the placement and nature of these symbols, concluding not only that all three attempts are seriously flawed but—especially given the fragmentary nature of the Shanghai Museum manuscript—that no consistent explanation is likely to be persuasive.

20. When referring to the hexagrams of the manuscript, I refer to them by the names used for them in the manuscript. Here Ru 孠 corresponds with Xu 需 hexagram of the received text. Conversely, when referring to the hexagrams of the received text, I refer to them by the names used for them in the received text. For a complete listing of the correspondences, see table 2.1.

21. However, he uses only three brief sentences to explain away the third hexagram with mismatched symbols, Yi, which intervenes between these in the received sequence of hexagrams. Since this hexagram has a

symbol at the head and a

symbol at the tail, Pu says that it must follow after a

group and come before a

group.

22. Shanghai Bowuguan cang Zhanguo Chu zhu shu (san), 259.

24. This is suggested, for instance, in He, “Lun Shangbo Chu zhu shu,” 29.

25. This contrast between invertible and convertible hexagrams probably explains the otherwise unbalanced division of the traditional text of the Zhou Yi into two separate scrolls, the first with thirty hexagrams and the second with thirty-four. If one treats the invertible pairs as a single hexagram picture, simply viewed from two different perspectives, then in the first scroll there are eighteen hexagram pictures (twelve invertible pairs and six hexagrams that are only convertible) and also eighteen in the second scroll (sixteen invertible pairs and only two hexagram pictures that are only convertible). For an extended discussion of this distribution, see Richard Cook, Classical Chinese Combinatorics.

26. Table 1.1 compares all sixty-four hexagrams in the Mawangdui sequence with the received sequence; see pp. 7–8.

27. The presentation that follows here is essentially identical with that in Xia, “Shilun Shangbo Zhou Yi gua xu,” and in Shaughnessy, “First Reading of the Shanghai Museum Bamboo-Strip Manuscript of the Zhou Yi,” 17–21.

28. It may be wondered whether the consecutive numbering of these four strips in the publication of the Shanghai Museum manuscript (30, 31, 32, 33) and the more or less similar breakage points of all four strips (between 30.8 cm and 31.5 cm from the top of the strips) are merely fortuitous or if perhaps they suggest that the strips were contiguous or nearly so in the original manuscript, which would suggest a sequence in which hexagrams Dun

(R33) and Kui 楑 (R38) were consecutive (or nearly so). However, given that the reconstruction of strip no. 32 requires a fragment in the possession of the Chinese University of Hong Kong, it may well be that other strips belonging between these two hexagrams were lost in the robbery or transport of the strips to Hong Kong.

29. The top of strip no. 3, about 23 cm long, is missing. Presumably it would have carried the last two characters of the Six in the Fourth line, the eight characters of the Nine in the Fifth line, and the first sixteen characters of the Top Six line.

30. A similar analysis might hold for strips 48 and 50, the first of which is broken at 13.4 cm from the top and the second at 12.6 cm from the top—the beginnings of Gen 艮 (R52) and Jian 漸 (R53) hexagrams (the intervening strip no. 49 is intact and contains the final portion of Gen).

31. Sun Peiyang, “Shanghai Bowuguan cang Zhanguo Chu zhu shu.”

32. For this scroll of seventy-seven strips, written between A.D. 140 and 142, see Lao, Juyan Han jian, 570–75.

33. A stricter transcription than that offered by Pu Maozuo would result in a slightly lower number of exact matches. For instance, where Pu transcribes yi 以, others would transcribe yi

(or simply 厶), and where he transcribes directly as li 利 (i.e., 禾 + 刀), a more exact transcription would be

. In the transcription that I give in the following chapter, I have tried to be as exact as possible, for which reason that transcription differs slightly from that of Pu.

34. Pu Maozuo adduces evidence that

is found in seal inscriptions for cang 藏, “to conceal,” which is written with the same character as zang, the usual reading for the corresponding character in the received text (though cang may well be the proper reading); in any event, the two graphs both share the same phonetic element (qiang 爿) and thus can readily be read for each other.

35. The Mawangdui manuscript, probably copied in the 170s B.C., that is, after the death of Liu Bang and before the accession of Liu Qi, presents this phrase as 啓國承家, whereas the Fuyang 阜陽 Zhou Yi manuscript, buried in 165 B.C., gives 啓邦承家, more or less identical with the Shanghai Museum manuscript; for the Mawangdui manuscript, see Mawangdui Han mu wenwu, 113, and for the Fuyang manuscript, see Han, “Fuyang Han jian Zhou Yi shiwen,” 20, and, in the present volume, p. 222. Whether this means that the copying of the Fuyang manuscript predates Liu Bang’s reign or death or that it simply failed to observe the taboo on his name is unclear.

36. Qin, “Liyong chutu wenxian jiaodu Zhou Yi,” presents a comprehensive discussion of all these variants.

37. For discussion of these variants, see the translation given in chapter 3.

38. Qin, “Liyong chutu wenxian jiaodu Zhou Yi,” 33–35.

39. Shanghai Bowuguan cang Chu zhushu (san), 180. Actually, this second occurrence of the graph

is somewhat more complicated than that in Guai hexagram. Here, the graph is followed by a repetition mark (

, called a chongwen hao 重文號 by Chinese paleographers), indicating either that the entire graph or some portion of it is to be read twice. In the case of this graph, which is composed of the component wang 亡, “to die, not to have,” and another component that Pu Maozuo transcribes more or less literally as 九, it is clear that the wang component is to be read twice, first as part of the technical divination term hui wang

亡 (i.e., 悔亡), “regrets gone,” and then as part of the phrase

馬, which reads in the received text sang ma 喪馬, “losing a horse.”

40. Mawangdui Han Mu Boshu Zhengli Xiaoxu, “Mawangdui boshu Liushisi gua shiwen.”

41. Fan, “Jian bo Zhou Yi Guai gua ‘sang’ zi bu shuo.” It is clear that he is correct in this reading. Whereas in two other occurrences of the graph qian 牽 in Mawangdui manuscripts, the graph features an “ox” (niu 牛) component below, in this case there is a “tree” (mu 木) component; see Chen Songchang, Mawangdui jian bo wenzi bian, 604, 43.

42. Park, “Shanghai Museum Zhouyi Manuscript,” 119, notes that in Mawangdui manuscripts, the number of “berries” in the graph for sang, “mulberry,” is variable, from a low of two to a high of four. It would seem that the “two berry” form of the graph led Xu Shen 許慎 to analyze the character incorrectly in his Shuo wen jie zi 說文解字, treating it as a “combined meaning” (hui yi 會意) character combining ku 哭, “to cry,” and wang 亡.

43. One is tempted to suggest that just as the editors of the Mawangdui Zhou Yi manuscript originally mistook the character sang 桑, “mulberry,” for the character qian 牽, “to lead,” so too might editors and copyists of the Zhou Yi in antiquity have made the same mistake. However, it is clear that the editors of the Mawangdui manuscript were led to their mistake by the reading of the received text, so that at least the process of transcription cannot be said to be entirely comparable.

44. For a complete listing of these variants, see Shaughnessy, “First Reading of the Shanghai Museum Bamboo-Strip Manuscript of the Zhou Yi,” 11, table 1.

45. For this manuscript and its additional divination terms, see chapters 6 and 7.

46. Indeed, Pu Maozuo transcribes

simply as 利.

47. The bamboo strip that would have carried the hexagram picture and hexagram name of this hexagram is missing from the Shanghai Museum corpus, but it is clear that it would have been written 尨 as in these line statements.

48. For example, the Xu gua 序卦 commentary, basing itself on the traditional sequence of the hexagrams, moving from Zhun 屯 hexagram (R3), which is explained as “a plant first sprouting,” explains the name of the hexagram as follows: wu sheng bi meng 物生必蒙, “when things are born they are necessarily meng.” The commentary of Wang Bi 王弼 (226–249), which serves as the basis of the orthodox exegetical tradition, explains the hexagram statement as “a youth wishing to resolve that which confuses him” (yu jue suo huo ye 欲决所惑也); see Zhou Yi zheng yi, 1.19a. This is reflected in the English translation of Richard Wilhelm’s German translation of the hexagram name: “Youthful Folly”; see Wilhelm, I Ching; or, Book of Changes, 20. It is also worth noting that already in the Mawangdui manuscript Mu He 缪和, Confucius is quoted as saying, Fu meng zhe, ran shao wei you zhi ye 夫蒙者, 然少未又知也 “As for meng, thus the young do not yet have knowledge”; see Chen Songchang, “Mawangdui bo shu Mu He Zhao Li shiwen,” 370. In my own translation of the name of the hexagram, “Shrouded,” I have opted to emphasize the basic meaning of the word, “covered,” though hoping also to suggest something of the meaning of “benighted, unenlightened.”

49. Shanghai Bowuguan cang Chu zhu shu (san), 137.

50. According to the phonetic reconstructions of Axel Schuessler, mang 尨 had an Old Chinese pronunciation of mrôŋ and a Middle Chinese pronunciation of måŋ, whereas meng 蒙 had Old Chinese and Middle Chinese pronunciations of môŋ and muŋ, respectively; see Schuessler, Minimal Old Chinese and Later Han Chinese, 169.

51. There would be no difference so long as one does not automatically assume that one spelling is “correct” and the other “incorrect”; notions of correct orthography may well be anachronistic for the period of the Shanghai Museum manuscript, written several hundred years before the first known Chinese dictionary.

52. Ōno, “Shu Eki Mō ka shinkai.”

53. This sense is hinted at in the commentary of Wang Bi, made more explicit in the subcommentary of Kong Yingda 孔穎達 (574–648), and fully elaborated in the later commentary of Cheng Yi 程頤 (1033–1107); for the first two, see Zhou Yi zheng yi, 1.19b; for the latter, see Cheng, Zhou Yi Cheng shi zhuan, 42 (1.11b). See also Wilhelm, I Ching; or, Book of Changes, 22; Wilhelm begins his comment on the line by saying, “Law is the beginning of education.”

54. Lu, Zhou Yi yin yi, 907 (1.4a).

55. Although the

of the manuscript seems not to be otherwise attested, it differs from the lei 羸 of the received text only in the substitution of a “horn” signific (jiao 角) for a “sheep” signific (yang 羊), which of course also features prominent horns; this would seem to be a natural transformation. As for 缾 and ping 瓶, according to the Shuo wen, 缾 is in fact the standard form of ping, “jar,” with the 瓶 of the received text given as a variant form, reasonably so since both 缶 and 瓦 are significs for earthenware implements.

56. Wu 毋 (archaic *mə) is the prescriptive negative “don’t,” whereas wei 未 (archaic *məs), though phonetically similar, is quite different in meaning, typically meaning “not yet.” In a lengthy review of my Rewriting Early Chinese Texts, William G. Boltz criticized an observation I made in that work concerning the apparent fluidity of negatives in early Chinese manuscripts versus received versions of the same texts; see Boltz, “Reading Early Chinese Manuscripts.” Although I appreciate Professor Boltz’s care in comparing and contrasting the various negatives of classical Chinese, I believe that there is considerable evidence in support of my observation, including, for instance, this case here.

57. Li Ling, “Du Shang bo Chu jian Zhou Yi” (the punctuation is Li’s).

58. Ibid., 62. The other character that he transcribes directly is 缾, mentioned above as doubtless identical with the received text’s ping 瓶.

59. Zhou Yi zheng yi, 60 (5.48) (p. 60).

60. Wang Niansun and Wang Yinzhi, Jing yi shu wen, 1.29b–30a.

61. Duan, Shuo wen jie zi Duan zhu, 5B.2a.

62. Although no commentaries that I have seen have questioned the meaning of jing 井, every other character in the clause has elicited various explanations. For instance, Lu, Zhou Yi yin yi, 941 (21a), reports that Zheng Xuan and Wang Su 王肅 (195–256) both explicitly read 射 as yi, meaning “to press upon, to pour into,” and this seems to be the sense given to it by both Wang Bi and Kong Yingda (see Zhou Yi zheng yi, 5.9b). As will be seen below, the character fu 鮒 has also been interpreted variously, either as “frog” or “small fish” (i.e., sardine).

63. Hanyu dacidian, 12:1215.

64. Lu Deming, Zhou Yi yin yi, 941 (21a); Zhou Yi zheng yi, 5.9b.

65. The earliest commentator on the Zhou Yi to have stated this explicitly seems to be Yu Fan 虞翻 (164–233), cited in Li Dingzuo, Zhou Yi jijie, 490 (10.4b). For more evidence, see Gao, Zhou Yi gu jing jin zhu, 165.

66. Xiao, Wen xuan, 1:228; see also Knechtges, Wen Xuan, 417 (Knechtges translates fu as “goldfish”).

67. Shanghai Bowuguan cang Chu zhu shu (san), 197.

68. Li Ling, “Du Shang bo Chu jian Zhou Yi,” 63.

69. By the same token, neither should we disregard the received reading, trying at all costs to read the manuscript differently.

70. Shanghai Bowuguan cang Zhanguo Chu zhu shu (san), 197.

71. Li Ling, “Du Shang bo Chu jian Zhou Yi,” 63.

72. Park, “Shanghai Museum Zhouyi Manuscript,” 238, suggests that words written in Chu script with the “rat” signific (shu 鼠) are typically written in Qin script (and thus in the received script) with the 豸 signific, usually indicative of some sort of wild animal. Elsewhere in the Shanghai Museum Zhou Yi manuscript, the 鼠 signific occurs in the graph

(strip no. 37, the Nine in the Second line statement of Jie 繲 [R40]), which corresponds with the character hu 狐, “fox,” in the received text, suggesting that 鼠 and quan 犬, “dog,” are also to some extent interchangeable as significs.

73. Li Falin, Zhanguo Qin Han kaogu, 9. I am grateful to Zhang Lidong 張立東 for this reference.

74. Some might suggest that this demonstrates that the Zhou Yi was not composed in the first place until the Warring States period, but this seems to me to confuse the senses of “writing” in terms of “composition” and “copying.” Indeed, the reading of the Shanghai Museum Zhou Yi manuscript in this case demonstrates that a sense of the line very different from that of the received text was available already in the fourth century B.C.

75. See He Linyi, Zhanguo guwen zidian, 65–66.

76. Shanghai Bowuguan can Zhanguo Chu zhu shu (san), 196.

77. These are final sounds that share the same main vowel but have different endings, either open, stop, or nasal.

78. Li Ling, “Du Shang bo Chu jian Zhou Yi,” 62–63.

79. Han jian 汗簡, Guwen sisheng yun 古文四聲韵 (Beijing: Zhonghua shuju, 1983), H58b (p. 29), and G4–13b (p. 56).

80. Chen Wei, “Shangbo jian Cong zheng Zhou Yi jiaodu.” For the Zhuangzi and Cheng Xuanying passages (from the “Ze yang” 則陽 chapter [25]), see Zhuangzi jishi 莊子集釋 (Taipei: He Luo tushu chubanshe, 1974), 885, 886.

81. Kunst, “Original ‘Yijing,’” 58.

82. Some recent textual criticism, beginning with the work of Joseph Bédier (1864–1938) in the 1920s, has emphasized the legitimacy of variant readings; see Bédier, “La tradition manuscrite”; see also Cerquiglini, Éloge de la variante. This marks a distinct break from the quest for the urtext that figures so prominently in the figures of Karl Lachmann (1793–1851) and his many followers.

83. Unfortunately, we must admit that it is almost always impossible to translate in this way. Just as scribes in the later Chinese tradition had to use one character—at the expense of other related characters—to write the text, so too when we translate—whether into English or any other language—must we usually choose just one from among any number of different words. The alternative would probably be to make the Changes even more difficult to understand than it already is.

1. The initial report of the discovery was Jingzhou Diqu Bowuguan, “Jiangling Wangjiatai 15 hao Qin mu.” Wang Mingqin 王明欽, the lead excavator of the tomb, has given a more detailed report, focusing especially on the bamboo strips in the tomb: “Wangjiatai Qin mu zhu jian gaishu.” This latter report still does not constitute a formal publication of the complete contents of either the tomb or the bamboo strips, and there are some indications that such a report may not be forthcoming. It seems that we may have to make do with what information is presently available.

2. Most texts were written in the Qin “clerical script” (Qin li 秦隸). However, Wang Mingqin reports that the two texts that have been identified as two versions of the Gui cang 歸藏 were written in the Chu script current earlier in the mid–Warring States period (see Wang Mingqin, “Wangjiatai Qin mu zhu jian gaishu,” 28), a characterization that is apparent also in the one photograph published in the initial report; see Jingzhou Diqu Bowuguan, “Jiangling Wangjiatai 15 hao Qin mu,” 41.

3. It is unclear to me just how many fragments there are. Wang Mingqin’s report mentions 813 numbered fragments, but he notes that a number of other fragments were not given numbers, and from the materials published to date, it would seem that these unnumbered fragments are fairly numerous and, in at least some cases, of some considerable importance. For further remarks on this, see n. 42 below.

4. For some description of these daybooks, see Wang Mingqin, “Wangjiatai Qin mu zhu jian gaishu,” 42–47. The first and still best known of the daybooks to be found in ancient tombs were those at Shuihudi 睡虎地, also in Hubei, unearthed in 1975 and introduced briefly in chapter 1. At Wangjiatai, the daybooks constitute the greatest number of bamboo strips, but since their contents are generally similar to those at Shuihudi, Wang Mingqin has not given a detailed presentation of them. However, he does note that there is at least one type of text among the Wangjiatai daybooks not seen elsewhere, which is of particular interest for its similarity to the divination texts added to the Fuyang 阜陽 Zhou Yi manuscript (for which, see chapters 6 and 7). Wang Mingqin terms it “daily taboos” (ri ji 日忌) and says that it includes an itemized listing of the auspices for various events on each of the thirty days of a month. Here I give just one example, contained on two strips.

十五日曰载是胃望以作百事大凶風雷畾日月宜飤邦君更歲不朝邦多廷獄作民多寡陽疾亡人得戰

The fifteenth day is called “Carrying”; this pertains to the full moon. In undertaking them, the hundred affairs are greatly ominous. The wind and thunder boom, and it is proper that the sun and moon be eclipsed. The leader of the country changes the year and does not come to court, and the country has much court litigation arising. Many of the people are orphaned, there are illnesses of excess yang, and people who abscond end up in war.

5. This code is essentially identical with a code of the same name also found in 1975 at Shuihudi. There are ninety-six strips belonging to this text, seventy-two of them numbered and twenty-four not numbered. Although the Wangjiatai text is quite fragmentary as compared with the same text at Shuihudi, it does offer the potential to correct the textual sequence presented by the editors of the Shuihudi strips. Whereas each statute of the Shuihudi code was begun on a new bamboo strip, and thus there was no way to determine their sequence, in the case of the Wangjiatai code the statutes are written continuously, with a 乙-shaped mark separating them. Unfortunately, Wang Mingqin has presented only two strips of this text. For the same text from Shuihudi, see Shuihudi Qin mu zhu jian, 33–40 (plates), 67–75 (transcription).

6. The text is intended to be read from the inside out, with the title in the middle (written twice, in reverse directions). The innermost square of this diagram is very similar to the text termed Wei li zhi dao 為吏之道, or The Way of Being an Officer, also discovered in 1975 at Shuihudi (for which, see Shuihudi Qin mu zhu jian, 79–85 [plates], 165–75 [transcription]), though there are variations in individual characters and also in the sequence of some phrases. The middle ring of the text serves as a commentary on the innermost ring, while the outer ring indicates both positive and negative consequences for the explanations of the middle ring. In the case of this text, which comprises sixty-five bamboo strips (including two strips without any writing that the archaeologists did not number), Wang Mingqin has given a rather complete description, noting the considerable significance of its layout; “Wangjiatai Qin mu zhu jian gaishu,” 39–42.

7. Ibid., 47–48, does not mention the total number of strips belonging to this text, except to say that, exceptionally, they were originally numbered, with numbers from 1 (yi 一) to 101 (bai yi 百一) appearing at the bottom of the bamboo strips. The contents record anomalous events in nature and society—“males of the six domestic animals became pregnant” (mu liu chu yun qi shen 牡六畜孕其身), “the moon fell from the heavens” (yue jiang zi tian 月降自天), “the earth shook” (di dong 地動), “a horse gave birth to a cow” (ma sheng niu 馬生牛), and so on—and then relate these to the fortunes of the state and the ruler: “the country will have a great loss” (bang you da sang 邦有大喪), “within three years someone will attack its country” (bu chu san nian huo fa qi bang 不出三年或伐其邦), “there will be a woman to die” (you nü sang 有女喪), and so forth. Unfortunately, these strips were very badly preserved, and it has not been possible to piece them together.

8. The initial site report, published in 1995, referred to the text as “a previously unseen type of Yi prognostication [Yi zhan 易占]”; Jingzhou Diqu Bowuguan, “Jiangling Wangjiatai 15 hao Qin mu,” 40–41. This attribution was adopted in Constance Cook, “Myth and Fragments of a Qin Yi Text.” However, by the year after the release of the preliminary site report, Wang Mingqin, the lead archaeologist of the find, published a study in which he demonstrated conclusively the manuscripts’ connection with quotations said to come from the ancient text Gui cang; see Wang Mingqin, “Shi lun Gui cang de jige wenti.” Other scholars making the same suggestion at just about the same time include Lian Shaoming (“Jiangling Wangjiatai Qin jian yu Gui cang”) and Li Jiahao (“Wangjiatai Qin jian Yi zhan wei Gui cang kao”).

9. The numbering is that given in the various publications of Wang Mingqin. Since all these publications, including the transcriptions of the Gui cang texts, have been presented in simplified Chinese characters, I perforce give the transcriptions here also in simplified characters. The symbol □ indicates that a single character is missing or cannot be transcribed, whereas the symbol

indicates an indeterminate number of missing characters, usually caused by a break in the bamboo strip. In my translations, these are rendered with .. and …, respectively. In addition, the symbol

, included on the original bamboo strips, indicates that the preceding character is to be read twice.

10. As seen below and in chapter 5, the Wangjiatai fragments consistently refer to the divinations as bu 卜, which originally had a specialized sense of “turtle-shell divination,” though it later came to be used for any type of divination. Medieval quotations of the text, on the other hand, consistently refer to “divining by milfoil” (shi 筮). Both sources refer to the result of the divination as “having the stalks prognosticated” (mei zhan 攴占 or 枚占). This shows that the bu of the Wangjiatai fragments should be understood as “to divine” in the general sense.

11. Zhang Hua, Bo wu zhi, 9, “Za shuo shang” 雜說上, 2b. The Lu shi 路史, or Revealed History, compiled at the very end of the Southern Song dynasty by the father-son team of Luo Ping 羅萍 and Luo Bi 羅泌 (completed in 1170), contains an almost, but not quite, identical quotation:

12. Guo Pu, Shan hai jing zhu, 7, “Hai wai xi jing” 海外西經, 1a. Li Fang et al., Taiping yulan, 929.1a:4128, contains a fuller, if somewhat garbled, version of this quotation:

13. At least three separate recensions of Gui cang quotations were produced in the first half of the nineteenth century: Yan, Quan shang gu San dai Qin Han San guo Liu chao wen (1836), 104–5; Hong, Gui cang (1926), 1.1a–4a; and Ma, Yuhan shanfang jiyi shu (1871). For a convenient comparison of all these quotations and the Wangjiatai texts, see Kondō, “Ōkatai Shin bo chikuken Ki sō no kenkyū,” 317–21.

14. Kondō, “Ōkatai Shin bo chikuken Ki sō no shotan,” 73, makes the point that there is no evidence to show that the Wangjiatai text is really the ancient Gui cang. There is also some question as to how the name 歸藏 should even be read and understood. The graph 藏 has two different readings: cang, a verb meaning “to store,” and zang, a noun meaning “storehouse, repository.” As far as I can tell, the only statement attempting to explain the title is attributed to Zheng Xuan 鄭玄 (127–200): “As for the title Gui cang, of the ten thousand things there is none that does not return to be stored in its midst” (Gui cang zhe, wan wu mo bu gui cang yu qi zhong ye 歸藏者, 萬物莫不歸藏于其中也; quoted in Li ji zheng yi, 21.187 [p. 1415]). Even this explanation is ambiguous. Although it clearly uses cang as a verb, it seems to portray the text as a whole as the place where the ten thousand things are stored, and thus apparently as a “storehouse.” However, since the preponderance of Chinese scholars now read the title as Gui cang, I follow their practice.

15. For an earlier study of my own of this text, see Shaughnessy, “Wangjiatai Gui Cang.” Some of the information that follows is taken from that earlier study, but most is new.

16. Zhou li zhushu, 24.164–65 (pp. 802–3).

17. In his commentary to the “Li yun” 禮運 chapter of the Li ji 禮記, where Confucius is quoted as saying, “We wanted to view the Way of Yin and for this reason went to Song, but it was insufficient to show it. I got the Kun Qian there” (Wo yu guan Yin dao, shi gu zhi Song, er bu zu zheng. Wu de Kun Qian yan 我欲觀殷道是故之宋而不足征吾得坤乾焉), Zheng Xuan states: “He got a yin-yang text of the Yin dynasty. Existing versions of that text include the Gui cang” (de Yin yin yang zhi shu, qi shu cun zhe you Gui cang 得殷陰陽之書其書存者有歸藏); Li ji zheng yi, 21.187:1415.

18. See Wang Chong, Lun heng, 28, “Zheng shuo” 正說, 4b–5a. Zheng Xuan is quoted similarly in the preface (“Xu” 序) of the Zhou Yi zheng yi, 5a. This preface also quotes Du Zichun 杜子春 (1st c. A.D.) as attributing the Lian shan to Fuxi 伏羲 and the Gui cang to Huang Di 黄帝.

19. Wang Chong, Lun heng, 12, “Xie duan” 謝短, 11b.

20. This portion of the Xin lun was lost by the end of the Northern Song dynasty, but this remark is preserved in Li Fang et al., Taiping yulan, 608.5a:2737.

21. Zhang Heng, Ling xian quotes without attribution a passage that can now surely be identified as coming from the Gui cang:

羿請不死之藥于西王母姮娥竊之以奔月將往枚筮之于有黄有黄占之曰吉翩翩歸妹獨將西行逢天晦芒毋惊毋恐後且大昌恒娥遂托身于月是為蟾蠩

Yi requested the medicine of immortality from the Western Queen Mother. Heng E stole it to flee to the moon. When she was about to go, she had the stalks divined by milfoil by You Huang. You Huang prognosticated them and said: Auspicious. So soaring the returning maiden, alone about to travel westward. Meeting heaven’s dark void; do not tremble, do not fear. Afterwards there will be great prosperity. Heng E subsequently consigned her body to the moon, and this became the frog.

22. See his comments on possible early divination texts other than the Zhou Yi, at Xi 熹 15 and Cheng 成 16, in Chunqiu Zuo shi zhuan Du shi jijie, 5.24a–b, 13.18a.

23. Wei, Sui shu, 27.909, 913. Prior to the listing in the Sui shu, the Gui cang was also included in the Qi lu 七錄 of Ruan Xiaoxu 阮孝緖 (479–536), which said of it that it was “a book of miscellaneous divinations”; cited in Ma, Gui cang, “Xu” 序, 1a, which provides a good overview of the textual history of the Gui cang.

24. The Sui shu states that the Zhong jing contained 29,945 scrolls, but this is surely based only on the original table of contents; in his Jin Yuan Di Si bu shumu 晉元帝四部書目, or Bibliography in Four Parts from the Reign of Jin Yuan Di, Li Chong 李充 (fl. 345–357) of the Eastern Jin reports that only 3,014 scrolls of the Zhong jing were still extant; see Yang and Gao, Zhongguo lishi wenxianxue, 76–77.

25. Wei, Sui shu, 27.909.

26. Fang Xuanling et al., Jin shu, 51.1432.

27. This biography was included in Wang Yin’s Jin shu 晉書, which was extant into the Tang dynasty and was thus an important source for Fang Xuanling’s Jin shu, but it was lost thereafter. This passage of its biography of Shu Xi is quoted in Yiwen leiju, 40, 732.

28. For evidence that Wang Yin wrote his Jin shu in the decades of the 320s and 330s, see Song Zhiying, “Wang Yin Jin shu chu tan.”

29. Guo Pu, Shan hai jing zhu, 7, “Hai wai xi jing” 海外西經, 1a.

30. See Zhang Hua, Bo wu zhi, 9, “Za shuo shang,” 2b.

31. For evidence of these quotations, see Shaughnessy, Rewriting Early Chinese Texts, 157.

32. Wang Mingqin, “Wangjiatai Qin mu zhu jian gaishu,” 36. Wang notes an even more persuasive example of the faulty nature of the transcription available to Zhang Hua:

Not only does this quotation include the same mistranscriptions of mei and shi but also, and more important, the reading Si Hua Tao Tao 四華陶陶 can now be seen as a mistranscription of 皋

陶

, the

being a repetition mark (chongwen hao 重文號) indicating that the two characters are to be read twice in succession (i.e., Gao Yao, Gao Yao 皋陶, 皋陶). Whoever was responsible for the transcription used by Zhang Hua obviously understood the first mark as a combined-character mark (hewen hao 合文號) indicating that two separate characters had been written together (presumably understanding gao 皋 as si hua 四華), and only the second as a repetition mark. A mistake such as this could derive only from someone working with bamboo-strip manuscripts.

33. This suggestion had actually been made almost fifty years before the discovery of the Wangjiatai bamboo strips; see Guo Moruo, Qingtong shidai, 2. For the suggestion in connection with the Wangjiatai discovery, see Wang Mingqin, “Gui cang yu Xia Qi de chuanshuo”; Wang Ning, “Qin mu Yi zhan yu Gui cang zhi guanxi”; Zhu Yuanqing, “Wangjiatai Gui cang yu Mu tianzi zhuan”; Ren and Liang, “Gui cang Kun Qian yuanliu kao.”

34. Li Fang et al., Taiping yulan, 82.7a–b:383. This chapter of the Taiping yulan, which collects quotations concerning Qi, the first ruler of the Xia dynasty, includes the following quotation, attributed to the Shi ji 史記:

35. Xue, Jiu Tang shu, 46.1966.

36. Chen Kui, Zhongxing guange shumu, 1.1a.

37. For discussion of this division of the text, see Xing Wen, “Hexagram Pictures and Early Yi Schools,” 584–87.

38. Ouyang Xiu, Ouyang Wen Zhong gong wen ji, 124.1b.

39. For the description of the Gui cang, see Luo and Luo, Lu shi, “Hou Ji” 後紀, 5.5b. As will be seen below, one common type of divination reads, “Would that his country not have any trouble” (qi bang shang wu you jiu 其邦尚毋有咎), or “distress” (lin 吝). In Jia Gongyan’s subcommentary to the Zhou li (at “Chun guan Tai bu” 春官太卜), one example of this type of statement (attributed to Jie 節 hexagram) is given, essentially unintelligibly, as qi guo chang wu ruo gu 其國常毋若谷 (guo 國 is obviously a substitute for the tabooed bang 邦, whereas chang 常, ruo 若, and gu 谷 are graphic errors for shang 尚, you 有, and jiu 咎); Zhou li zhu shu, 24.6b. In Luo and Luo, Lu shi, “Fa hui” 發挥, 1.15b, this is further garbled into chang wu gu yue 常毋谷月; it is unclear whether this sort of corruption derives from a misquotation of Jia Gongyan’s quotation, or if it might have come from a direct quotation of some Gui cang text.

A passage apparently quoted uniquely by the Lu shi may suggest once again that the Gui cang quoted in medieval sources came from the Jizhong tomb find of A.D. 279. After the Yi yao yin yang gua, which, as suggested above, probably corresponds to the Gui cang, the description in the Jin shu of the texts found in that tomb, next mentions a Gua xia Yi jing 卦下易經, or Classic of Changes Beneath the Hexagrams, in one bundle, which it says “was similar to the Shuo gua [說卦, or Discussion of the Hexagrams] but different” (Jin shu, 51.1432). The Lu shi, “Fa hui” 發挥, 1.15b, quotes the following passage as coming from the Gui cang:

40. For citations of recensions of these quotations, see n. 13 above.

41. See, for example, Sun Yirang, Zhou li zheng yi, 47.7a; Yu Yongliang, “Yi gua yao ci de shidai ji qi zuozhe,” 167; Rong, “Zhanbu de yuanliu,” 276–77.

42. Of these, 80 discrete strips can be identified in Wang Mingqin’s two major studies, “Shi lun Gui cang de jige wenti” and “Wangjiatai Qin mu zhu jian gaishu”; this constitutes just about exactly 20 percent of the total number of strips. Of these 80 strips, 54 are numbered and 26 unnumbered. In his general description of all the bamboo strips from the tomb, Wang states that the unnumbered strips were those that were “severely broken” (p. 27), but in his transcription of the Gui cang strips it is hard to see a pronounced difference between the numbered and unnumbered strips; true, most of the unnumbered strips have fewer than ten characters, but at least one of them (quoted in “Shi lun Gui cang de jige wenti,” 110) has twenty-five characters. Another measure of the percentage of the corpus published to date may be the total number of characters. In “Wangjiatai Qin mu zhu jian gaishu” (p. 29), Wang states that the 394 strips include somewhat more than 4,000 characters. By my count (and counts can vary depending on whether one includes the hexagram pictures [of which there are 70, not included in my count], or the duplication marks [also not included in my count], or even the symbols apparently indicating places where characters should be present but cannot be read [indicated with □ in the transcription]), the 80 strips published to date include just about 1,000 characters, or 25 percent of the total, more or less proportional with the percentage of total strips that have been published. On the one hand, this perhaps suggests that the transcriptions published to date are more or less representative of the entire corpus. On the other hand, it might also suggest that the strips that have not been published are not appreciably more fragmentary than those published and would potentially be of great interest at least in terms of comparing the two versions of the text.

43. Wang Mingqin notes complete strips among both the Ri shu and Zheng shi zhi chang texts, the Ri shu measuring between 22.6 cm and 22.9 cm in length, approximately the length of one “foot” in Qin-dynasty measures, and the Zheng shi zhi chang measuring 34.5 cm, about one and a half Qin feet. He further notes that the Gui cang strips “possibly exceed the length of those of the Zheng shi zhi chang”; see Wang Mingqin, “Wangjiatai Qin mu zhu jian gaishu,” 28.

44. Even though this is never clearly stated in any of Wang Mingqin’s descriptions of the manuscript, another indication that this is the case is that his transcriptions indicate a break in the bamboo strip with the symbol

; when a hexagram picture is given, this symbol never appears before it.

45. Since the hexagram pictures seem to have been written at the top of the bamboo strips, they were therefore most liable to being broken off from the rest of the bamboo strip. This probably accounts for their relatively fewer number vis-à-vis the hexagram names.

46. See table 4.1 for a listing of these hexagram names, together with a comparison of the names as they appear in medieval sources and also in the Zhou Yi.

47. I indicate the entry number of the strip when available; strips without an entry number are apparently among the unnumbered fragments. In addition to the examples listed here, see also note 71 below for another example, for which, however, there is no complete transcription of the second text.

48. The name of this hexagram, Bi 比, corresponds with Bi 比, “Alliance,” hexagram in the Zhou Yi tradition. Wang Hui suggests that the name should here be read as bi 芘, “hibiscus”; see Wang Hui, “Wangjiatai Qin jian Gui cang jiaoshi (28 ze),” 76. Although this seems to make better sense of the context here, the evidence is not sufficiently compelling to reject the traditional understanding.

49. This hexagram, the name of which is written yi 亦 in one text and ye 夜 in the other, corresponds with Gu 蠱, “Parasites,” hexagram in the Zhou Yi tradition. Since yi 亦 is the phonetic element in the character ye, 夜, “night,” it is likely that the reading of both these texts indicates the same word, but it is unclear how this might be related to the Zhou Yi hexagram name.

50. Wang Mingqin’s transcription gives a box (□) after the character da 大, usually indicative of a missing character. However, the da is explicitly indicated.

51. For example, I find it striking that strips 212, 213, and 214 are all nearly complete, bearing 26, 35, and 30 characters, respectively. As I have suggested with respect to the Shanghai Museum manuscript of the Zhou Yi, such similarity of preservation may indicate contiguous placement in the manuscript.

52. For a study of these statements and their place in the context of Chinese legends concerning Qi, see Wang Mingqin, “Gui cang yu Xia Hou Qi de chuanshuo.”

53. Guo Pu, Shan hai jing zhu, 9, “Haiwai dong jing” 海外東經, 2a.

54. Liao, “Wangjiatai Qin jian Gui cang guankui,” 15.

55. Other hexagrams that involve Xia Hou Qi include Gua 寡, “Orphan,” Song 讼, “Lawsuit,” Qun 囷, “Bundled,” Jing 井, “Well,” Guan 灌, “Libation,” Jin

, “Jin,” as well as a quotation that does not preserve the name of the hexagram:

56. There are also two other statements that would seem to be complete, except for the hexagram, in medieval quotations and for which there may also be fragments among the Wangjiatai manuscripts.

57. Li Fang, Taiping yulan, 85.2a:401.

58. Zhuangzi, 3.6b.

59. Xiao, Wen xuan, 13.600 (at “Yue fu” 月賦 by Xie Zhuang 謝莊).

60. Ibid., 60.2609 (at “Ji Yan Guanglu wen” 祭顏光祿文 by Wang Sengda 王僧達).

61. Quoted in Liu, Hou Han shu buzhu, at Hou Han shu, “Zhi” 志, 10.3216.

62. Gan, Xin jiao Sou shen ji, 108, juan 14.

63. As I demonstrate below, jing 惊, “to tremble” (archaic *krang), fits the rhyme scheme of the oracle perfectly. Although kong 恐, “to fear” (archaic *khong?), is also a near rhyme, it fits less well, presumably suggesting the preferability of the Sou shen ji quotation.

64. Li Fang, Taiping yulan, 929.1a:4128.

65. Ibid., 82.7a–b:383. I suspect that the final word, qing 卿 (archaic *khrang), “minister,” of the final phrase, yi wang si qing 以王四卿, is almost surely a mistake for the graphically and phonetically similar xiang 鄉 (*hang), “direction,” si xiang 四鄉, “four directions,” being a relatively common locution in pre-Qin texts. Indeed, in early inscriptions and manuscripts, there seems to have been no graphic distinction between the two words.

66. This statement perhaps corresponds with the following unnumbered fragment from Wangjiatai:

67. Ma Guohan’s edition of Gui cang quotations includes this quotation; see Ma, Gui cang, 13a:29.

68. The three fragments too incomplete to provide much information are those for Shao Du 少督, “Lesser Examination” (i.e., Xiao Chu 小畜) hexagram, no. 206; for Qun 囷, “Bundled (i.e., Kun), no. 208; and Ju

, “Frightened” (i.e., Kui 睽), which is unnumbered. The statement for Jian 渐, “Gradual Progression” (no. 335) is nearly as complete as the examples examined here but does not add materially to their understanding.

69. I might mention as well that Li Jiahao, “Wangjiatai Qin jian Yi zhan wei Gui cang kao,” 46, suggests transcribing the last character here as hu 狐 (*gwâ), “fox,” rather than as

, and says that this would rhyme with both xi 席 (*s-lak), “mat,” and xi

(*kêh), “creek.” However, it is hard to follow his suggestion here; not only do the words not seem to rhyme but also the final character is not in rhyme position in any event.

70. It is an unnumbered fragment mentioned on p. 110 of his 1996 article “Shi lun Gui cang de jige wenti.”

71. It is perhaps worth noting that the entry numbers of the statements of this qi bang shang wu you jiu format are relatively close: 206, 208, 213, 302, 335, plus two unnumbered fragments. All but one of these numbers belong to what the archaeologists identified as the B group of bamboo strips (strips 181–304), and the one exception (no. 335) is in the contiguous group C (strips 305–42); for this stratigraphy of the tomb’s bamboo strips, see pp. 154–57. This might suggest that statements of this format were found in only one of the two manuscripts. However, Wang Mingqin, “Wangjiatai Qin mu zhu jian gaishu,” 35, mentions in an offhand manner that Kun 困 hexagram (i.e., what he elsewhere transcribes as Qun 囷) has two hexagram statements, one of which reads “Xia Hou Qi bu qi bang shang wu you lin 夏后启卜亓邦尚毋有吝 (this must refer to strip no. 208) and the other of which gives the final lin 吝, “distress,” as jiu 咎, “trouble.” This latter statement, not included in any of the transcriptions published to date, would seem to suggest that essentially the same statement was included in both versions of the manuscripts.

72. Other examples are Bi 比, “Alliance” (no. 216), Wu Wang 毋亡, “Not Lost” (no. 471), Jie 介, “Strengthened” (no. 207), Dun

, “Withdrawing” (no. 463), and possibly Dui 兑, “Removal” (no. 334). In the case of Dui, Wang Mingqin lists two fragments of bamboo strips as follows:

73. Compare these lines with the ultimate stanza of the Shi jing 詩經 poem “Xin tai” 新台, or “New Terrace” (Mao 43):

74. Cai, “Qin jian Gua Tian Juan zhu gua jiegu: Jianlun Gui cang Yi de ruogan wenti” proposes transcribing it as gua 寡, “orphan,” but this is by no means certain.

75. Liao, “Wangjiatai Qin jian Gui cang guankui,” 17. Liao suggests transcribing this graph as yin 寅 but does not suggest how he understands it.

76. Shaughnessy, “Wangjiatai Gui Cang.”

77. Because the yin lines of the hexagrams are written ^, similar to the archaic form of the graph for the number “six,” and given the early writing of hexagrams with numbers (for which see pp. 142–43), there has been a tendency on the part of some scholars to interpret the Wangjiatai Gui cang hexagrams as “numerical symbols,” made up of the numbers “6” and “1.” This seems to me to be an overly literal reading of these symbols. Moreover, as Liao Mingchun has argued (“Wangjiatai Qin jian Gui cang guankui,” 18–19), traditions surrounding the Gui cang have always privileged the numbers “7” and “8,” whereas “6” has been especially associated with the Zhou Yi tradition.

78. The earliest listing of Gui cang hexagram names, though without the associated hexagram pictures, goes back to the Song dynasty: Li Guo, Xixi Yi shuo, 1.19b–20a. Rao, “Yin dai Yi gua ji qi youguan zhanbu zhu wenti” was the first to draw a systematic connection between the Mawangdui hexagram names and those of the Gui cang.

79. Liao Mingchun has argued that the name of the hexagram, Jie 介, can help to explain the corresponding hexagram name in the Zhou Yi: Yu 豫. This hexagram name has traditionally been understood to mean “happy” or “compliant,” but it is hard to relate this meaning either with the hexagram with which this forms a pair (Qian 謙, “Modesty”) or with the corresponding Gui cang hexagram name. However, among their various meanings, both jie and yu share the sense of “big,” which moreover can serve as an antonym to the paired qian 謙, “Modesty.” See Liao, “Wangjiatai Qin jian Gui cang guankui,” 18.

80. In the fourth case, that of Ming Yi 明夷 hexagram, the Gui cang hexagram statement concerns a divination by Xia Hou Qi about riding a dragon to ascend into the heavens. In the Zhou Yi, the hexagram of the same name has traditionally been understood to mean something like “Brightness Obscured,” but the first line statement clearly concerns the flight of some sort of bird (probably of the pheasant family) and the top line statement includes the phrase “first ascending into the heavens” (chu deng yu tian 初登于天). It might not be too impressionistic to see a relationship between these different divination topics, though the question of which might have influenced which is unclear.

81. For the initial demonstration of this symbolism, see Gu Jiegang, “Zhou Yi gua yao ci zhong de gushi,” 14. See also Shaughnessy, “Marriage, Divorce, and Revolution.”

83. Although this sort of three-phrase structure probably marks the normative Zhou Yi line statement, it is even more common to find just the introductory phrase describing the omen, often joined with technical divination terms, as in the following example, also from Ding hexagram:

84. In my earlier study “Wangjiatai Gui Cang,” I examined two examples of milfoil divination recorded in the Zuo zhuan that include divination results that do not match the Zhou Yi and suggested that they may have stemmed from divination using the Gui cang. Now, with more of the text of the Gui cang available for comparison, it would seem that the divination results in these two cases (for the fifteenth year of Duke Xi 僖 [645 B.C.] and the sixteenth year of Duke Cheng 成 [575 B.C.]) are of a different form from those of the Gui cang and thus are probably not from that text. This would suggest, as we should probably expect, that there were more than just two different divination texts and systems in use at the time.

1. Jingzhou Diqu Bowuguan, “Jiangling Wangjiatai 15 hao Qin mu”; Wang Mingqin, “Shi lun Gui cang de jige wenti”; Wang Mingqin, “Wangjiatai Qin mu zhu jian gaishu.” The transcriptions given in these publications are all in modern, simplified characters. Without access to the original bamboo strips, or to photographs or even drawings of them, it is necessary to rely on Wang’s work. Fortunately, I have been able to check his transcriptions against a set of transcriptions done independently by Peng Hao 彭浩, former director of the Jingzhou Museum and one of the most experienced paleographers in China. The two transcriptions agree in most respects, suggesting that the transcriptions published to date can be used with some confidence.

2. Yan Kejun, Quan shang gu San dai Qin Han San guo Liu chao wen, 104–5; Hong, Gui cang, 1.1a–4a; Ma, Gui cang. For a convenient comparison of all these quotations and the Wangjiatai texts, see Kondō, “Ōkatai Shin bo chikuken Ki sō no kenkyū.” For a less-convenient but still more complete listing of quotations, see Wang Ning 王寧, “Chuanben Gui cang jijiao,” 傳本歸藏輯校, http://www.gwz.fudan.edu.cn/SrcShow.asp?Src_ID=1003 (accessed December 13, 2011).

3. Li Guo, Xi xi Yi shuo, 1.19b–20a.

4. Jia, Zhou li zhu shu, 24.6b. Jia’s subcommentary to the Zhou li (at “Chun guan Tai bu” 春官太卜) attributes this quotation to a Kun Kai shi 坤開筮 of the “current” (jin 今) Gui cang; the kai 開 of this title is the standard replacement for the graph qi 啟, the name of the Han emperor Liu Qi 劉啟 (i.e., Han Jingdi, r. 157–141 B.C.), which was tabooed after his reign. Thus, this quotation doubtless derives from the Qi shi jing 啟筮經 section of the Gui cang.

5. Yu Shinan, Beitang shu chao, 158.1a.

6. Zhuangzi, 3.6b, quoted by Lu Deming 陸德明 in Yin yi shiwen 音義釋文.

7. Li Fang et al., Taiping yulan, 85.2a:401.

8. It is not clear that these two pieces correspond to the same hexagram, but there would seem to be no other correlation. The proximity of the two pieces in situ (nos. 461, 470) would perhaps argue against their belonging to two different versions of the text.

9. Guo Pu, Shan hai jing zhu, 9, “Haiwai dong jing” 海外東經, 2a.

10. Du, Yu zhu bao dian, 1.21b. This quotation is lacking in the Qing-dynasty recensions of Gui cang quotations but is supplied in Wang Ning, “Chuanben Gui cang jijiao.” It is unclear whether it refers to this same Lü hexagram or not, but I place it here based on the similarity of its content with Wangjiatai fragment no. 470. Note that the quotation includes at least two obvious mistakes: jie 借 for xi 昔 and qi 起 for qi 啟.

11. Li Fang et al., Taiping yulan, 79.2b–3a:367–68.

12. Ouyang Xun, Yi wen lei ju, 99.19.

13. The hexagram picture given here is, in the Zhou Yi tradition, that of Xiao Guo 小過 hexagram. There is an unnumbered fragment, entered above (p. 177), that includes both the hexagram picture

(i.e., the hexagram picture for Da Guo in the Zhou Yi tradition) and also the hexagram name Da Guo 大过 (i.e., Da Guo 大過). Therefore, it would seem that this hexagram name here is simply miscopied and should read Xiao Guo 小过 (i.e., 過) or some variation thereof.

14. The earliest source to quote this passage is Yu Shinan, Beitang shu chao, 82.3b. It is not certain that this quotation corresponds to the fragment of Guan hexagram in the Wangjiatai manuscripts. However, since Jin 晉 hexagram records a different offering made by Xia Hou Qi 夏后啟 at a different place (at “the mound of Jin” [Jin zhi xu 晋之墟]), it seems likely that these passages correspond.

15. Li Fang et al., Taiping yulan, 840.1b:3753.

16. Guo Pu, Er ya Guo zhu, 11.16a, quotes the Gui cang as saying “two vases and two pots” (liang hu liang shu 兩壺兩羭); Xing Bing’s 邢昺 subcommentary provides the complete passage, stating that it comes from the “Qi mu jing” 齊母經 section of the Gui cang; see Xing Bing, Erya zhu shu, 10.14a.

17. Zhang Hua, Bo wu zhi, 2b.

18. Jia, Zhou li zhu shu, 24.6b.

19. Apparently this strip fragment contained only the hexagram picture, which can be identified with that of Jian 蹇 hexagram in the Zhou Yi tradition.

20. Yu Shinan, Beitang shu chao, 132.2a.

21. Zhang Heng, Ling xian; see also Liu, Hou Han shu buzhu, Zhi 志 10A, “Tianwen zhi shang” 天文志上, 3216.

22. Ouyang Xun, Yi wen lei ju, 62.12b.

23. Zhang Hua, Bo wu zhi, 9.2b.

24. Guo Pu, Shan hai jing zhu, 7, “Hai wai xi jing” 海外西經, 1a.

25. Guo Pu, Mu tianzi zhuan zhu, at Mu Tianzi zhuan 穆天子傳, 5.4a.

26. Li Fang et al., Taiping yulan, 82.7a–b:383, attributes this quotation to the Shi ji 史記, but it is clear that this attribution is mistaken and should actually be to the Gui cang.

27. Guo Pu, Mu tianzi zhuan zhu, at Mu tianzi zhuan, 2.1b.

28. Yu Shinan, Beitang shu chao, 150.13a.

29. Zhang Hua, Bo wu zhi, 9.2b.

30. Li Fang et al., Taiping yulan, 82.12a:385.

31. Ibid., 912.2a:4040.

32. Zhang Hua, Bo wu zhi, 9.2b.

33. Xu Jian et al., Chu xue ji, 488, juan 20.

34. Zhang Hua, Bo wu zhi, 9.2b.

36. Guo Pu, Shan hai jing zhu, 15, “Da huang nan jing” 大荒南經, 4b.

38. Ibid., 16, “Da huang xi jing” 大荒西經, 1a.

39. Ibid., 18, “Hai nei jing” 海內經, 5b.

41. Ibid., 2, “Xi shan jing” 西山經, 10a.

42. Ibid., 6, “Hai wai nan jing” 海外南經, 1b.

43. Li Shan, Wen xuan zhu, 612, juan 13, at Mi Zhengping 禰正平, “Yingwu fu” 鸚鵡賦.

44. Xu Jian, Chu xue ji, 205, juan 9.

45. Ibid., 16, juan 1.

46. Ouyang Xun, Yiwen lei ju, 92.22a.

47. Li Shan, Wenxuan zhu, 1004, juan 20, at Yan Yannian 顏延年, “Qiu hu shi” 秋胡詩.

48. Li Fang et al., Taiping yulan, 472.7a:2169.

49. Ouyang Xun, Yiwen lei ju, 84.12a.

50. Li Fang et al., Taiping yulan, 835.1b:3727.

1. This chapter is revised and expanded from Shaughnessy, “Fuyang Zhou Yi.”

2. For the initial report of the tomb’s excavation, see Anhui Sheng Wenwu Gongzuodui, Fuyang Diqu Bowuguan, and Fuyang Xian Wenhuaguan, “Fuyang Shuanggudui Xi Han Ruyin Hou mu fajue jianbao.” To date, no formal report of the excavation has ever been issued.

3. The initial report of the texts found in the tomb is Wenwu Ju Guwenxian Yanjiu Shi and Anhui Sheng Fuyang Diqu Bowuguan Fuyang Han Jian Zhenglizu, “Fuyang Han jian jianjie.” For a concise overview of the tomb’s discovery and contents, see Pian and Duan, Ben shiji yilai chutu jianbo gaishu, 60–64. For a more extensive survey, including information not seen in other published sources, see Hu and Li, Chang jiang liuyu chutu jiandu yu yanjiu, 507–43.

4. Hu Pingsheng provided the first detailed, published account of this text: “Fuyang Han jian Zhou Yi gaishu.” The first complete transcription, together with drawings of 138 strips identified as belonging to the Zhou Yi manuscript, is to be found in Han, “Fuyang Han jian Zhou Yi shiwen.” Also included in the same issue of Daojia wenhua yanjiu is a lengthy study of the text by Han: “Fuyang Han jian Zhou Yi yanjiu.” These two studies were subsequently reprinted in a somewhat different format together with photographs and drawings of all 752 strips identified as belonging to the Zhou Yi manuscript (as well as an additional 48 fragments that are possibly related) in Han, Fuyang Han jian “Zhou Yi” yanjiu.

5. For an initial transcription of the Shi jing manuscript, see Fuyang Han Jian Zhenglizu, “Fuyang Han jian Shi jing,” as well as a study of it by Hu and Han, “Fuyang Han jian Shi jing jianlun.” These were later superseded by the same authors’ Fuyang Han jian “Shi jing” yanjiu.

6. There are only eight strips, bearing a total of fifty-six characters, that have been identified as belonging to the Zhuangzi, one from the “Zeyang” 則陽 chapter (25), six from the “Wai wu” 外物, or “External Things” chapter (26), and one from the “Rang wang” 讓王, or “Yielding Kingship,” chapter (28). For a brief study, including photographs of all eight strips and full transcriptions, see Han and Han, “Fuyang chutude Zhuangzi Zapian Han jian.”

7. According to Hu and Li, Chang jiang liuyu chutu jiandu yu yanjiu, 537–58, there are passages from thirty-four chapters of eleven of the twelve monthly annals (ji 紀). However, as far as I know, no transcription has been published to date.

8. According to Wenwu Ju Guwenxian Yanjiu Shi and Fuyang Diqu Bowuguan Zhenglizu, “Fuyang Han jian jianjie,” 23, there are “several” (ruogan 若干) fragments of poetry, including one with four characters from the “Li sao” 離騷 and one with five characters from the “She jiang” 涉江 poems of the Chu ci.

9. For an initial transcription of the Cangjie pian, see Fuyang Han Jian Zhenglizu, “Fuyang Han jian Cangjie pian” and, in the same issue, a study of it by Hu and Han, “Cangjie pian de chubu yanjiu.”

10. For an initial transcription of the Wan wu text, see Fuyang Han Jian Zhenglizu, “Fuyang Han jian Wan wu,” as well as a study of it by Hu and Han, “Wan wu lüe shuo.”

11. See Hu, “Some Notes.” As far as I know, the only more or less complete description of this text in Chinese is Hu and Li, Chang jiang liuyu chutu jiandu yu yanjiu, 519–23, 540.

12. For an overview of the materials in the tomb related to divination, see Hu, “Fuyang Shuanggudui Han jian shushu shu jianlun.”

13. For photographs, drawings, transcriptions, and a study of ninety-six strips identified as belonging to this Chunqiu shi yu text, as well as a board containing its table of contents, see Han, Fuyang Han jian “Zhou Yi” yanjiu, 167–205.

14. For photographs, drawings, transcriptions, and a study of the board containing the table of contents of the Ru jia zhe yan, see ibid., 151–63.

15. For these diviner’s boards and their use, see Li Ling, Zongguo fangshu kao, 82–102. See also Harper, “Han Cosmic Board.”

16. Loewe, Biographical Dictionary of the Qin, 597, gives Zao’s year of death as 164 B.C.; however, I can find no evidence that Zao died late in the year, which given the imperfect correlation between the Chinese calendar and the Western calendar might otherwise account for such a discrepancy. All sources agree that 164 B.C. (the sixteenth year of Han Wendi) was the first year of the reign of Zao’s son, Xia Hou Si 夏侯賜.

17. The initial site report does not mention that the tomb was filled with water. For information concerning the excavation (e.g., the waterlogged nature of the tomb) and particularly the efforts to “organize” (zhengli 整理) the bamboo strips and their texts, I am grateful to Hu Pingsheng, who eventually joined the editorial team. The information credited to him here and below is based on Hu and Li, Chang jiang liuyu chutu jiandu yu yanjiu; a public lecture he gave at the University of Chicago in November 2001; and numerous conversations in the years since.

18. Some of this process is described in the preface by Lü Jimin 吕濟民 to Han, Fuyang Han jian “Zhou Yi” yanjiu, 1–2.

19. There are another forty-eight fragments, many of them split in half vertically, that it has proved not possible to read; for photographs and drawings of them, see Han, Fuyang Han jian “Zhou Yi” yanjiu, 42–44.

20. According to Han, Fuyang Han jian “Zhou Yi” yanjiu, 45, five hexagram pictures are visible. However, the published photographs of the relevant strips for Ben 賁 and Da Guo 大過 hexagrams (strips 106 and 138) show little or no trace of these, and on p. 88, where Han compares hexagram pictures with those of other manuscripts, he illustrates only the hexagram pictures of Da You 大有, “Great Offering” (strip no. 64), Lin 林, “Forest” (i.e., 臨, “Looking Down,” strip no. 86), and Li 離, “Fastening” (strip no. 151).

21. A few fragments that would be appropriate at two or more points in the Zhou Yi text (such as those containing only the formulaic divination terms ji 吉, “auspicious,” or xiong 兇 [i.e., 凶], “ominous,” that end many line statements) have been placed at their first possible occurrence. Other fragments that contain only numerical tags for lines (e.g., chu liu 初六, “First Six,” or jiu er 九二, “Nine in the Second”) or only divination statements have been arbitrarily listed at the end of the transcription.

22. On p. 63 of “Fuyang Han jian Zhou Yi yanjiu,” Han Ziqiang gives the number of characters of the divination statements as 2,009, whereas on p. 74, he gives the number as 2,169; these numbers are repeated in Fuyang Han jian “Zhou Yi” yanjiu (pp. 87, 95). Simple arithmetic suggests that the former number is correct, but it is apparent that there may be different counts of fragments and characters.

23. The following discussion of the physical properties of the Fuyang Zhou Yi manuscript will not be of interest to all readers. Some readers may wish to skip to the following section. Readers seeking more detail concerning the physical properties of the text may wish to consult Xia, “Fuyang Han jian Zhou Yi jiance xingzhi ji shuxie geshi zhi lice,” from which the following discussion is drawn.

24. Han, “Fuyang Han jian Zhou Yi shiwen,” 16 (or p. 46 in Fuyang Han jian “Zhou Yi” yanjiu). In fact, as I note below, there is another strip, no. 58, that is 19 cm long.

25. Hu, “Fuyang Shuanggudui Han jian shushu jian lun,” 22.

26. Hu, “Fuyang Han jian Shi jing jiance xingzhi ji shuxie geshi zhi lice.”

27. Here and throughout this chapter (and in chapter 7), in my English translations I differentiate the text of the Zhou Yi proper from that of the divination statements by displaying the Zhou Yi text in roman letters and the divination statements in italics. It goes without saying that no such orthographic distinction is apparent in the original manuscript.

28. The only occurrence of the graph xing 興, “to arise,” in the received text of the Zhou Yi comes in this Nine in the Third line of Tong Ren. It is of course possible that xing here marks the end of a divination statement followed by another divination statement initiated with the word bu 卜, “divining,” but this is not very likely.

29. As noted in the next section, the standard format of these divination statements is to indicate first a topic of divination, followed by the result. Thus, here the divination statement bu yu bu yu 卜雨不雨 is to be interpreted as I have done above, and not, for instance, as “divining about whether it will rain or not rain.”

30. The transcription below accords with the photograph and hand copy of the fragments given on p. 13 of Han, Fuyang Han jian “Zhou Yi” yanjiu. The editor’s transcription given on p. 62 of that book indicates a different point of breakage between the two fragments (after the ji 吉, “auspicious,” that ends the hexagram statement of the Zhou Yi). This might suggest that fragment no. 151 is made up of two separate fragments that the editors rejoined before photographing them (the hand copy of the fragment seems to indicate this, though the photograph does not), but it seems preferable here to rely on the published photographs.

31. In the Lun heng 論衡, Wang Chong 王充 (27–ca. 100) reports that “two feet four inches is [the length of] the writings and sayings of the sages” (Huang, Lun heng jiao shi, 557, “Xie duan” 謝短). For an exhaustive examination of early textual evidence concerning the physical nature of bamboo-strip texts, see Wang Guowei, Jian du jian shu kao. Among the bamboo strips discovered at Guodian 郭店, several texts, including the Zi yi 緇衣, which was subsequently included in the classic Li ji 禮記, were written on bamboo strips 32.3 cm long, whereas the Heng xian 恆先 text included among the Shanghai Museum manuscripts is 39.4 cm.

32. I had hoped to be able to examine points of breakage on individual fragments of bamboo to determine which strips may have been adjacent to which, and thus subject to similar stresses, and on the basis of this perhaps to gain some insight into the sequence of hexagrams in the manuscript—a topic of great interest in Yi jing studies ever since the discovery of the Mawangdui manuscript. Unfortunately, because of the extremely fragmentary nature of the Fuyang manuscript, I have not succeeded in this sort of reconstruction.

33. Thirty-two hexagram names are present in the fragments. Of these, the following differ to at least some extent from the names in the received text (giving first the name in the Fuyang manuscript, in italic transcription and characters, followed by a colon and then the name in the received text, again in italics and characters, and then, in parentheses, the number of the hexagram in the received sequence): Zhun 肫: Zhun 屯 (3), Sui 隋: Sui 隨 (17), Lin 林: Lin 臨 (19), Shi Zha 筮閘: Shi Ke 噬嗑 (21), Pu 僕: Bo 剝 (23), Wu Wang 无亡: Wu Wang 无妄 (25), Chuan 椽: Dun 遯 (33), Deng 登: Sheng 升 (46).

34. According to Han, Fuyang Han jian “Zhou Yi” yanjiu, 100, the Fuyang manuscript contains 63 characters that differ from both the received text and the Mawangdui manuscript, 51 characters that are the same as the received text but differ from the Mawangdui manuscript, and 26 characters that differ from the received text but are the same as the Mawangdui manuscript. In addition, there are eighteen places in the text where the characters are the same as both but the sequence is different. For convenient tables illustrating these differences, see 143–48 of his book.

35. Although the word bu 卜 originally referred specifically to the kind of pyromantic divination that produced cracks in bone or shell (and especially in turtle shell), by the Han period it had come to have a broader sense of any sort of divination.

36. Only a portion of the top line of the final character yu 雨, “rain,” is visible in the photograph, but that Han Ziqiang is correct in supplying it in his transcription can be seen by comparison with such other fragments as nos. 631, 632, and 633. Also, as noted above, it is worth recalling that the bu 不, “not,” following the first yu, “rain,” is not to be read together with it as a question marker but rather should be read with the following yu as part of the prognostication.

37. For qi fei zheng you sheng 其匪

有眚, “his not going to correct has curses,” the received text reads qi fei zheng you sheng 其匪正有眚, “his not being correct has curses”; this and other variations between the Fuyang manuscript and the received text of the Zhou Yi are displayed—but not otherwise commented upon—in the translation presented in the next chapter.

38. For my current understanding of this most common of formulas in hexagram statements, see Xia, “Zhou Yi ‘Yuan heng li zhen.’s”

39. Of the variants found in this hexagram vis-à-vis the received text, 壄 for ye 野, gao 高, “high,” for qi 其, “its,” tang 唐, “platform,” for yong 墉, “wall,” hao or jiao 鄗, “Hao” or “suburb,” for jiao 郊, “suburb,” and

for hui 悔, “regret,” all except gao, “high,” and tang, “platform,” are well-attested allographs for the characters in the received text, though 鄗 also has attested uses as the name of the Zhou capital city in addition to its more common use as an allograph for jiao 郊, “suburb,” and could reasonably be understood in that sense here. In short, the manuscript text of the Zhou Yi is very similar to the received text.

40. Han, “Fuyang Han jian Zhou Yi yanjiu,” 74 (or p. 96 in Fuyang Han jian “Zhou Yi” yanjiu), cites three examples (the Six in the Third line of Guan 觀, “Looking Up,” strip no. 94; the Six in the Second line of Yi 頤, “Jaws,” strip no. 133; and the hexagram statement of Li 離, “Fastening,” strip no. 151) in which the divination statement follows immediately after the line or hexagram statement without an introductory bu.

41. It is possible that these two strips should not be rejoined in this way. Although the first character of strip no. 19, “family” (jia 家), neatly completes the Nine in the Second line of Meng hexagram, not only is it also the last word of the Top Nine line of Sun 損, “Decrease,” hexagram (no. 41) but it is also a word found with some frequency among the Fuyang divination statements themselves (e.g., in fragments 566–91, the word occurs twenty-two times) and thus might instead be the last word of a preceding divination phrase. Despite these two possibilities, the example here seems sufficiently plausible (especially considering the precedents noted above in n. 28) to warrant consideration.

42. The received text of the Zhou Yi reads na 納, “to take in,” rather than the lao 老, “old,” of the Fuyang text (the Mawangdui manuscript reads ru 入, “enter,” cognate with na).

43. Li Jingchi, “Zhou Yi shi ci xu kao.”

44. For example, there seems to be no reason to differentiate such similar formulas as “beneficial to see the great man” and “beneficial to divine” on the basis of subject matter. In my own study of the Zhou Yi, I proposed four different components of a line statement: a Topic, by which I meant more or less the same as Li Jingchi’s “image prognostication”; an Injunction, by which I meant the various statements of advice usually beginning with the word li 利, “beneficial”; a Prognostication, which I restricted to just the words ji, xiong, li, and lin, often following immediately after the word zhen 貞, “to determine”; and a Verification, a different group of divination terms including wu you li, wu bu li, wu jiu, hui, and hui wang (terms that do not ordinarily occur together with the verb zhen); see Shaughnessy, “Composition of the Zhouyi,” 136–58.

45. It should be noted that there is nothing necessarily indicating that no. 143 should in fact follow no. 142, its content being exclusively made up of (portions of) two divination statements. By placing it immediately after no. 142, Han Ziqiang obviously assumes that the first partial divination statement completes the final partial divination statement included on no. 142. This seems reasonable, and thus I follow it here. Whether this placement is correct or not does not affect the overall analysis presented here.

46. These archaic reconstructions are those in Schuessler, Minimal Old Chinese and Later Han Chinese, 278 (series 26–15), 281 (26–26), and 280 (26–24).

47. This line is not among the Fuyang fragments.

48. Schuessler, Minimal Old Chinese and Later Han Chinese, 50 (1–27), 60 (1–66), and 55 (1–45).

1. The text is taken from Han, Fuyang Han jian “Zhou Yi” yanjiu, 3–44 (photographs and drawings), 45–86 (transcription).

1. Because of the nature of bamboo-strip texts, there is no conclusive evidence concerning the sequence of the hexagrams in the manuscript. However, as discussed in chapter 2, there is a certain amount of physical evidence and also considerable circumstantial evidence suggesting that the manuscript was either in the same sequence as the received text or in a sequence very similar to it.

2. The Zuo zhuan 左傳 (2nd year of Duke Zhao 昭, 540 B.C.) mentions that the Jin 晉 emissary Han Xuanzi 韓宣子 saw the Images of the Changes (Yi xiang 易象) and the “Lu Springs and Autumns” (Lu chunqiu 魯春秋) during an embassy to Lu; this is sometimes understood to show that the Changes, if indeed “Yi xiang” refers to the Changes proper, was not available in the state of Jin. However, even in the Zuo zhuan, there is considerable evidence to suggest that the Changes was known and used throughout the northern states. In addition to other evidence for the availability of the Changes in Lu, anecdotes mention figures from at least the following states quoting the Changes or using it in divination: Zhou 周 (in Chen 陳; 22nd year of Duke Zhuang 莊, 672 B.C.), Jin 晉 (multiple times; 1st year of Duke Min 閔, 661 B.C.; 15th year of Duke Xi 僖, 635 B.C.; etc.), Zheng 鄭 (6th year of Duke Xuan 宣, 603 B.C.), Qi 齊 (25th year of Duke Xiang 襄, 548 B.C.), and Wei 衛 (7th year of Duke Zhao, 535 B.C.).

3. Preliminary reports of the major corpus of Warring States bamboo-strip texts donated to Qinghua University in 2008 indicate that one of the lengthiest texts in the corpus provides a detailed description of milfoil divination methods. Scholars at the university’s Center for the Study and Preservation of Unearthed Texts (Chutu Wenxian Yanjiu Yu Baohu Zhongxin 出土文獻研究與保護中心) hope to publish this manuscript in volume 4 of their series Qinghua Daxue cang Zhanguo zhu jian 清華大學藏戰國竹簡; publication is tentatively scheduled for late 2013 or 2014.

4. It is doubtless the case that all members of society, including the very lowest levels, shared the same view of divination and engaged in some of the practices that we term divination. However, as is so often the case, archaeological evidence is lacking for the lowest levels of society.

5. The Mawangdui manuscript Yao 要 contains the following response by Confucius when asked by his disciple Zi Gong 子貢 whether he believed in divination:

If the commendations do not reach to the numbers, then one merely acts as a magician, but if the numbers do not lead to virtue, then one merely acts as a scribe. The divinations of scribes and magicians tend toward it but are not yet there, delight in it but are faulty. Perhaps it will be because of the Changes that sires of later generations will doubt me. I seek its virtue and nothing more. I am on the same road as the scribes and magicians, but return to a different place. The nobleman’s virtue and action seek blessings in it; therefore, he does sacrifice, but does so little. His humaneness and propriety seek auspiciousness in it; therefore, he does divine, but does so rarely. Do not the divinations of the priests and magicians come last!

魚网之設鴻則離之	The fish’s net being set out, A wild goose then got caught in it.
燕婉之求得此戚施	One pretty and shapely was sought, And yet got this ugly old frog.